SACHE Problem Set Volume 1

Preface

The American Institute of Chemical Engineers (AIChE) has a 30-year history of involvement with process safety and loss control for chemical and petrochemical plants. Through its strong ties with process designers, builders, and operators, safety professionals and academia, the AIChE has enhanced communication and fostered improvement in the high safety standards of the industry. Its publications and symposia have become an information resource for the chemical engineering profession on the causes of incidents and means of prevention.

The Center for Chemical Process Safety (CCPS), a directorate of AIChE, was established in 1985 to intensify development and dissemination of the latest scientific and engineering practices for prevention and mitigation of catastrophic incidents involving hazardous materials; advance the state of the art of engineering practices through research; and develop and encourage the use of undergraduate education curricula that will improve the safety knowledge and consciousness of engineers.

Over 55 corporations from all segments of the chemical process industries (CPI) support the Center in its work. They help fund the Center; they select CCPS projects relevant to improved process safety; and they furnish the professionals who give the Center's works technical direction and substance.

Since its founding, CCPS has cosponsored several international, technical symposia and has published seven volumes in its Guidelines series, a technical workbook, and the proceedings of three meetings. Some of these publications have become the core of new courses in AIChE's Continuing Education series for engineers. CCPS research projects now in progress will yield new data for improved process safety.

One of the first projects undertaken by CCPS was directed to undergraduate engineering education. Its primary objective was to increase the student's awareness of, interest in, and knowledge of safety, health, and loss prevention concepts in chemical engineering, and to increase recognition of the engineer's responsibilities in these areas. This project was also intended to help college and university engineering schools meet accreditation requirements published in Criteria for Accrediting Programs in Engineering in the United States, Effective for Evaluations during the 1989- 1990 Academic Year by the Accreditation Board for Engineering and Technology (ABET).

Studies done by the Safety and Health Division of AIChE, by the CCPS Undergraduate Education Subcommittee, and by ABET revealed that the most effective way to introduce safety, health, and loss prevention concepts to undergraduate engineering students was through their integration into existing courses. It was found that adding a safety and health course as a core requirement to an already crowded curriculum was next to impossible. As an elective course, ABET criteria

could not be satisfied. Consequently, the Subcommittee designed this material for use in many existing courses leading to chemical and other engineering degrees.

The subcommittee identified 34 areas that contained important safety, health, and loss prevention concepts. The concepts were divided into problematic and nonproblematic categories. The former were developed into problems to illustrate fundamentals of health and safety and that could be used in teaching traditional engineering subjects, such as thermodynamics, fluid mechanics, or heat transfer. Wherever possible, nonproblematic concepts were introduced in the problem description or background material provided the student and instructor.

Ninety problems have been developed, covering many of the identified conceptual areas for several basic engineering subjects. The material is presented in two books: the book for students contains background material for each problem and the problem itself; the Instructor's Guide has the student's material as well as descriptive material for each problem about the concepts involved and the problem's solution.

Many of the problems were used and critiqued by engineering faculty in 20 colleges and universities nationwide. As a result of this testing and problem preparation by some of the Chemical Engineering faculty at the University of Arkansas, the instructional aids presented in these books can be conveniently and easily incorporated into current engineering courses at all levels of the undergraduate engineering curriculum.

The problems teach safety, health, and loss prevention'as an integral part of many engineering solutions and can provide new engineers with insights to industrial situations they are likely to encounter. This material may also be used as a reference for graduates as they begin industrial careers and by industrial in-house courses for new engineering employees. They further demonstrate that safety and health issues can be handled by basic engineering principles and logic, and are not foreign to basic engineering practices. Finally, through use of this material, we hope to instill increased recognition and acceptance of the professional and ethical responsibilities which engineers must have to provide safe chemical plants, processes and products.

Acknowledgments

The American Institute of Chemical Engineers (AIChE) wishes to thank the Center for Chemical Process Safety (CCPS) and those involved in its operation, including its many sponsors, whose funding made this project possible; the members of its Technical Steering Committee, who conceived of and supported this project; and the members of its Undergraduate Education Subcommittee for their dedicated efforts, technical contributions, and the guidance necessary for the preparation of this work.

The Chairman of the Undergraduate Education Subcommittee was F. Owen Kubias, The Glidden Company. The Subcommittee members were Roger W. Bohl, Dow Chemical U.S.A.; Guy Colonna, National Fire Protection Association; Daniel A. Crowl, Wayne State University; John Davenport, Industrial Risk Insurers; Dr. James A. Gideon, National Institute for Occupational Safety and Health; Stanley S. Grossel, Hoffmann-LaRoche, Inc.; Dr. Walter B. Howard, consultant; Dr. Joseph J. Levitzky, Yale University; J. F. Louvar, BASF Corporation; Gene I. Matsumoto, S. C. Johnson & Son, Inc.; Robert Nelson, Industrial Risk Insurers; John Noronha, Eastman Kodak Company; Gary A. Page, American Cyanamid Company; Robert M. Rosen, BASF Corporation; Jerry Schroy, Monsanto Chemical Company; Richard F. Schwab, Allied-Signal, Inc.; Dr. Klaus Timmerhaus, University of Colorado. Thomas W. Carmody, Russell G. Hill, Lester H. Wittenberg, and Ray E. Witter of the Center for Chemical Process Safety were responsible for the overall administration and coordination of this project.

AIChE thanks Dr. J. Reed Welker and Dr. Charles Springer, faculty in the Chemical Engineering Department, University of Arkansas, Fayetteville, Arkansas, for using their expertise to provide advice and to prepare the problems, solutions, and background information for these books. The University of Arkansas helped

' support this project from its inception to its completion, and this support is deeply appreciated.

AIChE also gratefully acknowledges the contributions of Murray Underwood, Washington University, St. Louis, and J. Arnold Glass, retired, in helping to develop the basic safety concepts and associated topics; Prenticti-Hall for permission to use problems from their book, Process Safety: Fundamentals with Applications, written by D. A. Crowl and J. F. Louvar; and Eastman Kodak Company for permission to use some problems from their in-house training program.

The assistance and philosophic support of the project's goals, given by the National Institute for Occupational Safety and Health and the U.S. Environmental Protection Agency, were of considerable help and are gratefully acknowledged.

The members of the CCPS Undergraduate Education Subcommittee wish to thank the faculty members of many colleges and universities for testing problems in their classrooms and providing critiques to improve the material for future students. Finally, the subcommittee members thank their employers for providing the time to participate in this project.

U N D E R G R A D U A T E E D U C A T I O N P R O B L E M M A T R I X

ENGINEERING COURSE

Heat Transfer

Mass rransfer -

68

- 77

Problematic Health and

Safety Concept

I . Properties of Materials

Introl Fund.

lomentun transfer

2. Process Design

3. Explosions

4. Toxic Exposure Control. Personal Protective Equipment

5. Process Control Interlocks. A l m s

6 . Toxicology and Industrial Hygiene

7. Vapor Releases

8. lnening and Purging

9. Storing, Handling, and Transpon

10. Fire Protection Systems

11. Haz. Waste Generation and Disposal

12. Rupture Discs and Relief Valves

13. Process Hazard Reviews

14. Static Electricity

IS. Physical Hazards

Glossary

This glossary defines many of the terms used on Material Safety Data Sheets (MSDS). It also explains some of the significance of the terms related to safety, health, and loss prevention. The glossary can provide substantial assistance in understanding terms commonly used by safety and health professionals.

ACUTE EFFECT An adverse effect on a human or animal body, with severe symptoms developing rapidly and coming quickly to a crisis. See also, "Chronic. " Importance: How much and how long one is exposed to a chemical is the critical factor to how adverse the health effects will be.

ACUTE TOXICITY The adverse (acute) effects resulting from a single dose or exposure to a substance. Importance: Ordinarily used to denote effects in experimental animals.

ACGIH American Conference of Governmental In- dustrial Hygienists; an organization of professional personnel in governmental agencies or educational institutions engaged in occupational safety and health programs. Importance: ACGIH develops and publishes recommended occupational exposure limits (see TLV) for hundreds of chemical substances and physical agents.

APPEARANCE AND ODOR The physical properties of a chemical, such as color and smell. Importance: Knowing what chemicals look and smell l i e allows an employee to recognized unsafe working conditions.

ASPHYXIANT A vapor or gas which can cause unconsciousness or death by suffocation (lack of oxygen). Most simple asphyxiants are harmful to the body only when they become so concentrated that they reduce oxygen in the air (normally about 21 percent) to dangerous levels (19.5 percent or lower). Importance: Asphyxiation is one of the principal potential hazards of working in confined spaces.

BOILING POINT The temperature at which a liquid changes to a vapor state, at a given pressure; usually expressed in degrees fahrenheit at sea level pressure (760 mmHg, or one atmosphere). For mixtures, the initial boiling point or the boiling range may be given. Importance: The lower the degree for the boiling point, the faster a liquid evaporates, in-

creasing the amount of vapor present at room temperature for both health and fire exposures.

"c:'OR CEILING The letter " C or the word "ceiling" on the TLV or PEL shows the highest airborne concentration of a specijic chemical that is allowed in the workplace. This concentration should never be exceeded, even for short periods of time. See also, "PEL" and "TLV. " Importance: Chemicals that react rapidly in the body, causing ill health effects carry this value.

CARCINOGEN A cancer-causing material. Importance: If a substance is known to be cancer causing, a potential health hazard exists and special protection and precaution sections should be checked on the MSDS.

C.A.S. NUMBER Chemical Abstracts Service Num- ber. Importance: C.A.S. Numbers are used on MSDS's to identify specific chemicals.

cc Cubic centimeter; a volume measurement in the metric system, equal in capacity to 1 milliliter (ml). One quart is about 946 cubic centimeters.

CHEMICAL FAMILY A group of single elements or compounds with a common general name. Ex- ample: acetone, methyl ethyl ketone, and methyl isobutyl ketone are of the "ketone" family; acrolein, furfural, and acetaldehyde are of the "aldehyde" family. Importance: Elements or compounds within a chemical family generally have similar physical and chemical characteristics.

CHEMTREC Chemical Transportation Emergency Center; a national center established by the Chemical Manufacturers Association in Wash- ington, D.C. in 1970, to relay pertinent emergency information concerning specijic chemicals. Importance: Chemtrec has an emergency 24- hour toll free telephone number (800-424- 9300).

CHRONIC EFFECT An adverse effect on a human or animal body, with symptoms which develop

slowly over a long period of time. Also, see "Acute." Importance: The length of time that a worker is exposed is the critical factor. Long periods of time pass, with repeated exposure to a chemical, before any ill effects are detected in a worker.

CHRONIC TOXlClTY Adverse (chronic) effects resulting from repeated doses of or exposures to a substance over a relatively prolongedperiod of time. Importance: Ordinarily used to denote effects in experimental animals.

co Carbon monoxide, a colorless, odorless, flammable and very toxic gas produced by the incomplete combustion of carbon; also a by- product of many chemical processes.

CO, Carbon dioxide, a heavy, colorless gas, produced by the combustion and decomposition of organic substances and as a by-product of many chemical processes. C02 will not burn and is relatively non-toxic (although high concentrations, especially in conjined spaces, can create hazardous oxygen-dejicient environments.) Importance: CO and C02 are often listed on MSDS's as hazardous decomposition products.

COMBUSTIBLE A term usedto classify certain liquids that will burn, on the basis offlashpoints. Both the National Fire Protection Association (NFPAJ and the Department of Transportatton (DOT) define "combustible liquids" as having a flash point of 100°F (37'8°C) or higher. See also, "Flammable." Importance: Combustible liquid vapors do not ignite as easily as flammable liquids; however, combustible vapors can be ignited when heated, and must be handled with caution. Class I1 liquids have flash points at or above 10O0F, but below 140°F. Class 111 liquids are subdivided into two subclasses:

Class IIIA: Those have flash points at or above 140°F but below 200°F.

Class IIIB: Those having flash points at or above 200°F.

CONCENTRATION The relative amount of a substance when combined or mixed with other substances. Examples: 2 pprn Xylene in air, or a 50 percent caustic solution. Importance: The effects of overexposure depend on the concentration or dose of a hazardous substance.

CORROSIVE AS dejined by DOT, a corrosive material is a liquid or solid that causes visible destruction or irreversible changes in human tissue at the site of contact on-in the case of leakage from itspackaging-a liquid that has a severe corrosion rate on sreel. Importance: A corrosive material requires different personal protective equipment to prevent adverse health effects.

DECOMPOSITION Breakdown of a material or substance (by heat, chemical reaction, electrolysis, decay, or other processes) into parts or elements or simpler compounds. Importance: Decomposition products often present different hazards than the original material.

DERMAL Used on or applied to the skin. Importance: Dermal exposure, as well as inhalation exposure, must be considered to prevent adverse health effects.

DERMAL TOXICITY Adverse effects resulting from skin exposure to a substance. Also referred to as "Cutaneous toxicity. " Importance: Ordinarily used to denote effects in experimental animals.

EMERGENCY AND FIRST AID PROCEDURES Actions that should be taken at the time of a chemical exposure before trained medical personnel arrive. Importance: These procedures may lessen the severity of an injury or save a person's life if done immediately following a chemical exposure.

EPA U.S. Environmental Protection Agency; Fed- eral agency with environmental protection regulatory and enforcement authority. Importance: EPA regulations must be met for the disposal of hazardous materials, as well as in spill situations.

EVAPORATION RATE A number showing how fast a liquid will evaporate. Importance: The higher the evaporation rate, the greater the risk of vapors collecting in the workplace. The evaporation rate can be useful in evaluating the health and fire hazards of a material.

FLAMMABILITY LIMITS The range of gas or vapor amounts in air that will burn or explode f a flame or other ignition source is present. Importance: The range represents an unsafe gas or vapor mixture with air that may ignite or explode. Generally, the wider the range the greater the fire potential. Also, see LEL, LFL, UEL, UFL.

FLAMMABLE A "Flammable Liquid" is defined by NFPA and DOT as a liquid with a jlash point below 100°F (37.8OC). Importance: Flammable liquids provide ig- nitable vapor at room temperatures and must be handled with caution. Precautions such as bonding and grounding must be taken. Flam- mable liquids are: Class I liquids and may be subdivided as follows: Class IA: Those having flash points below 73°F

and having a boiling point below 100°F.

Class TS: Those having flash points below 73°F and having a boiling point at or above 100°F.

Class IC: Those having flash points at or above 73°F and below 100°F.

FLASH POINT The lowest temperature at which vapors above a liquid will ignite. There are several flash point test methods, and flash points may vary for the same material depending on the method used. Consequently, so the test method is indicated when thejash point is given (150' PMCC, 200" TCC, etc.) A closed cup type test is used most frequently for regulatory puvoses. Flash point test methods:

Cleveland Open Cup (CC) Pensky Martens Closed Cup (PMCC) Setaflash Closed Tester (SETA) Tag Closed Cup (TCC) Tag Open Cup (TOC)

Importance: The lower the flash point temperature of a liquid, the greater the chance of a fire hazard.

FORMULA The conventional scientijic designation for a material (water is H20, sulfuric acid is H2S04. Sulfur dioxide is SO2, etc.) Importance: Chemical formulas identify specific materials.

GENERAL EXHAUST A system for exhausting air containing contaminants from a general work area. See also, "Local Exhaust. " Importance: Adequate ventilation is necessary to prevent adverse health effects from exposures to hazardous materials and vapor accumulations that can be a fire hazard.

g Gram; a metric unit of weight. One U.S. ounce is about 28.4 grams.

g/kg Grams per kilogram; an expression of dose used in oral and dermal toxicology testing to indicate the grams of substance dosed per kilogram of animal body weight. See also, "kg." Importance: A measure of the toxicity of a substance.

HAZARDOUS MATERIAL In a broad sense, a hazardous material is any substance or mixture of substances having properties capable of producing adverse effects on the health or safety of a human being. Importance: Knowing what a hazardous material is and what materials are hazardous is important in preventing adverse health or safety effects.

INCOMPATIBLE Materials which could cause dangerous reactions from direct contact with one another are described as incompatible. Importance: On a MSDS, incompatible materials are listed to prevent dangerous reactions in the handling and storage of the material.

INGESTION The taking of a substance through the mouth. Importance: A route of exposure to a hazardous material.

INGREDIENTS A listing of chemicals that are in a mixture.

Importance: Knowing exactly what chemicals and how much of each is in a mixture helps you to understand the potential hazard a mixture presents.

INHALATION The breathing in of a substance in the form of a gas, vapor, fume, mist, or dust. Importance: A route of exposure to a hazardous material.

INWITOR A chemical which is added to another substance to prevent an unwanted chemical change from occurring. Importance: Inhibitors are sometimes listed on a MSDS, along with the expected time period before the inhibitor is used up and will no longer prevent unwanted chemical reactions.

IRRITANT A substance which, by contact in suf- jicient concentration for a sufficient period of time, will cause an inflammatory response or reaction of the eye, skin, or respiratory system. The contact may be a single exposure or multiple explosures. Some primary irritants: chromic acid, nitric acid, sodium hydroxide, calcium chloride, amines, chlorinated hydrocarbons, ketones, alcohols. Importance: Knowing that a substance is an irritant allows you to be aware of the signs and symptoms of overexposure.

kg Kilogram; a metric unit of weight, about 2.2 U.S. pounds, See also, "glkg," "g," and "mg."

L Liter; a metric unit of capacity. A U.S. quart is about 9/10 of a liter.

LC Lethal Concentration: A concentration of a substance being tested which will kill a test animal.

LC50 Lethal Concentration 50; The concentration of a material in air which, on the basis of laboratory tests, is expected to kill 50 percent of a group of test animals when administered as a single exposure (usually 1 or 4 hours). The LC50 is expressed as parts of material per million parts of air, by volume (ppm) for gases and vapors, or as micrograms of material per liter of air (pg/L) or milligrams of material per cubic meter of air ( m g l d ) for dusts and mists, as well as for gases and vapors. Importance: Both are measures of the toxicity of a substance.

LD Lethal Dose; A concentration of a substance being tested which will kill a test animal.

LDSO Lethal Dose 50; A single dose of a material which on the basis of laboratory tests is expected to kill 50% of a group of test animals. The LD50 dose is usually expressed as milligrams or grams of material per kilogram of animal body weight (mglkg or glkg). Importance: Both are measures of the toxicity of a substance.

LEL OR LFL Lower Explosive Limit or Lower Flammable Limit of a vapor or gas; the lowest concentration (lowest percentage of the sub-

stance in air) that will produce a flash of j re when an ignition source (heat, arc, or&me) is present. See also, "UEL." Importance: At concentrations lower than the LELILFL, the mixture is too "lean" to burn.

LOCAL EXHAUST A system for capturing and ex-, hausting contaminantsfrom the air at the point where the contaminants are produced (welding, grinding, sanding, dispersion operations). See also, "General Exhaust." Importance: Adequate ventilation is necessary to prevent adverse health effects from exposures to hazardous materials and prevent vapor ac- cumuIations that can be a fire hazard.

MATERIAL IDENTIFICATION The name of a chemical. It may be a trade name, chemical name or any other name a chemical is known by. On a MSDS this section also includes the name, address, and emergency telephone number of the distributing chemical company. Importance: Proper identification of a chemical allows an employee to get additional health hazard and safety information.

m3 Cubic meter; a metric measure of volume, about 35.3 cubic feet or 1.3 cubic yards.

MELTING POINT The temperature at which a solid substance changes to a liquid state. For mixtures, the melting range may be given. Importance: The physical state of a substance is critical in assessing its hazard potential, route of exposure and method of control.

mg Milligram; a metric unit of weight. There are 1,000 milligrams in I gram (gl of a substance.

mgkg Milligrams per kilogram; an expression of toxicological dose. See also, "$/kg." Importance: A measure of the toxicity of a substance.

mg/m3 Milligrams per cubic meter; a unit of measuring concentrations of dusts, gases, or mists in air. Importance: The effects of overexposure depend on the concentration or dose of a hazardous substance.

mL Milliliter; a metric unit of capacity, equal in volume to I cubic centimeter (cc), or about 11 16 of a cubic inch. There are 1,000 milliliters in 1 liter (L).

mm ~g Millimeters (mm) of Mercury (Hg); a unit of measurement for low pressures or partial vacuums. Importance: Vapor pressures are expressed in mm Hg.

MUTAGEN A substance or agent capable of altering the genetic material in a living cell. Importance: If a substance is known to be a mutagen, a potential health hazard exists, and special protection and precaution sections should be checked on the MSDS.

NOSH National Institute for Occupational Safety and Health of the Public Health Service, U.S. Department of Health and Human Services

(DHHS). Importance: Federal agency which-among other activities-tests and certifies respiratory protective devices, recommends occupational exposure limits for various substances and assists in occupational safety and health investigations and research.

N* Oxides of Nitrogen; undesirable air pollutants. Importance: Often listed on a MSDS as a hazardous decomposition product.

OLFACTORY Relating to the sense of smell. Importance: The olfactory organ in the nasal cavity is the sensing element that detects odors and transmits information to the brain through the olfactory nerves. This sense of smell is a "built in" vapor detector.

ORAL Used in or taken into the body through the mouth. Importance: A route of exposure to a hazardous material.

ORAL TOXICITY Adverse effects resulting from taking a substance into the body via the mouth. Importance: Ordinarily used to denote effects in experimental animals.

OSHA Occupational Safety and Health Adminis- tration of the U.S. Department of Labor. Importance: Federal agency with safety and health regulatory and enforcement authorities for most U.S. industry and business.

OXIDIZING AGENT, OXIDIZER A chemical or substance which brings about an oxidation reaction. The agent may (1) provide the oxygen to the substance being oxidized (in which case the agent has to be oxygen, or contain oxygen), or (2) it may receive electrons being transferred from the substance undergoing oxidation. DOT defines an oxidizer or oxidizing material as a substance which yields oxygen readily to stim- ulate combustion (oxidation) of organic matter. Importance: If a substance is listed as an oxidizer on the MSDS, precautions must be taken in the handling and storage of the substance. Keep away from flammables and combustibles.

PEL Permissible Exposure Limit; an exposure established by OSHA regulatory authority. May be a Time Weighted Average (TWA) limit or a maximum concentration exposure limit. See also, "Skin." Importance: If a PEL is exceeded, a potential health hazard exists, and corrective action is necessary.

POISON, CLASS A A DOT term for atremedy dangerous poisons, that is, poisonous gases or liquids of such nature that a very small amount of the gas, or vapor of the liquid, mixed with air is dangerous to life. Some examples: phos- gene, cyanogen, hydrocyanic acid, nitrogen peroxide.

POISON, CLASS B A DOT term for liquid, solid, paste, or semisolid substances-other than Class A poisons or irritating materials- which

are known (or presumed on the basis of animal tests) to be so toxic to man as to afford a hazard to health during transporation. Importance: If a substance is known to be a poison, health and safety hazards exist and special protection and precaution sections should be checked on the MSDS.

POLYMERIZATION A chemical reaction in which one or more small molecules combine to form larger molecules. A hazardous polymerization is such a reaction which takes place at a rate which releases large amounts of energy. Importance: If hazardous polymerization can occur with a given material, the MSDS usually will list conditions which could start the reaction and the time period before any contained the inhibitor is used up.

ppm Parts per million; a unit for measuring the concentration of a gas or vapor in air-parts (by volume) of the gas or vapor in a million parts of air. Also used to indicate the concentration of a particular substance in a liquid or solid. Importance: The effects of overexposure depend on the concentration or dose of a hazardous substance.

ppb Parts per billion; a unit for measuring the concentration of a gas or vapor in air-parts (by volume) of the gas or vapor in a billion parts of air. Usually used to express measurement of extremely low concentrations of unusually toxic gases or vapors. Also used to indicate the concentration of a particular substance in a liquid or solid. Importance: The effects of overexposure depend on the concentration or dose of a hazardous substance.

REAC~ION A chemical transformation or change; the interaction of two or more substances to form new substances. Importance: Knowledge of reactions can prevent unsafe chemical changes.

REACTIVITY A description of the tendency of a substance to undergo chemical reaction with itself or other materials with the release of energy. Undesirable effects-such as pressure buildup, temperature increase, formation of noxious, toxic, or corrosive by-product-may occur because of the reactivity of a substance to heating, burning, direct contact with other materials or other conditions in use or in storage. Importance: Knowledge of what conditions to avoid can prevent unsafe chemical reactions.

REDUCING AGENT In a reduction reaction (which always occurs simultaneously with an oxidation reaction) the reducing agent is the chemical or substance which ( I ) combines with oxygen, (2) loses electrons in the reaction. See also, "Ox- idizing Agent. " Importance: If a material is listed as a reducing agent on the MSDS, precautions must be taken

in the handling and storage of the substance. Keep separate from oxidizing agents.

RESPIRATORY SYSTEM The breathing system; includes the lungs and air passages (trachea or "windpipe," larynx, mouth, and nose) to the air outside the body, plus the associated nervous and circulatory supply. Importance: Inhalation is the most common route of exposure in the occupational workplace.

SENSITIZER A substance which on first exposure causes little or no reaction in man or test animals, bur which on repeated exposure may cause a marked response not necessarily limited to the contact site. Skin sensitization is the most common form of sensitization in the industrial setting, although respiratory sensitization to a few chemicals is also known to occur. Importance: Knowing that a substance is a sensitizer allows you to be aware of the signs and symptoms of overexposure.

%KIN" A notation, sometimes used with PEL or TLV exposure data; indicates that the stated substance may be absorbed by the skin, mucous membranes, and eyes-either by airborne or by direct contact-and that this additional exposure must be considered part of the total exposure to avoid exceeding the PEL or TLV for that substance. Importance: Even if workplace concentrations of a chemical do n9t exceed the TLV or PEL, the risk to health may be severe because breathing and skin contact are combined. Skin protection is advised.

SKIN SENSITIZER See "Sensitizer. " SKIN TOXIC~TY See "Dermal Toxiciry." SOLUBILITY IN WATER A term expressing the per-

centage of a material (by weight) that will dissolve in water at ambient temperature. Importance: Solubility information can be useful in determining spill cleanup methods and fire-extinguishing agents and methods for a material.

SO. Oxides of Sulfur; undesirable air pollutants. Importance: Often listed on a MSDS as a hazardous decomposition product.

SPECIAL PRECAUTIONS Instructions that describe proper handling and storage procedures specijic to that material. Importance: Following these procedures would prevent excessive employee exposure. These procedures tell you additional information needed to handle the material safely.

SPECIAL PROTECTION INFORMATION A description of engineering precautions and personal protection that should be provided when working with a chemical in order to reduce an employee's exposure. Importance: Reducing the potential for exposure reduces the risk to health and safety.

SPECIFIC GRAVITY The weight of a material compared to the weight of an equal volume of water;

an expression of the density (or heaviness) of the material. Example: If a volume of a material weighs 8 pounds, andan equal volume of water weighs 10 pounds, the material is said to have a specific gravity of 0.8.

Importance: Insoluble materials with specific gravity of less than 1.0 will float in (or on) water. Insoluble materials with specific gravity greater than 1.0 will sink (or go to the bottom) in water. Most flammable liquids have specific gravity less than 1.0 and, if not soluble, will float on water-an important consideration for fire suppression and spill clean-up.

SPILL OR LEAK PROCEDURES Steps that should be taken if a chemical spill or leak occurs. Importance: Proper removal of a chemical spill or leak from the work area eliminates the potential accumulation of hazardous concentrations of the chemical, reduces the risk of creating an environmental pollution problem and con- forms with local, state and federal regulations.

S T A B U ~ ~ Y An expression of the ability of a material to remain unchanged. Importance: For MSDS purposes, a material is stable if it remains in the same form under expected and reasonable conditions of storage or use. Conditions which may cause instability (dangerous change) are stated-for example, temperatures above 150"F, shock from dropping.

STEL Short Term Exposure Limit; ACGIH ter- minology. See also, "TLV-STEL. "

SYNONYM Another name or names by which a material is known. Methyl alcohol, for example, is also known as methanol, or wood alcohol. Importance: A MSDS will list common name(s) to help identify specific materials.

TERATOCEN A substance or agent to which exposure of a pregnant female can result in mal- formations in the fetus. Importance: If a substance is known to be a teratogen, a potential health hazard exists and special protection and precaution sections should be checked on a MSDS.

TLV Threshold Limit Value; a term used by ACGIH to express the airborne concentration of a material to which nearly allpersons can be exposed day after day without adverse effects. ACGIH expresses TLVs in three ways: TLV-TWA: The allowable Time Weighted Av- erage concentration for a normal &hour work- day or 40-hour work week. TLVSTEL: The Short-Term Exposure Limit, or maximum concentration for a continuous 15- minute exposure period (maximum offour such periodsper day, with at least 60 minutes between exposure periods, and provided that the daily TLV-TWA is not exceeded). n v - c : The Ceiling exposure limit-the con-

instantaneously. TLV'S are reviewed and revised annually where necessary by the ACGIH Importance: If a TLV is exceeded, a potential health hazard exists and corrective action is necessary. Also see "Skin" relative to TLV's.

~oxrcrrv The sum of adverse effects resultingfrom exposure to a material, generally by the mouth, skin, or respiratory tract. Importance: Knowledge of the toxicity of a material helps prevent adverse health effects from exposure.

TRADE NAME The trademark name or commercial trade name for a material. Importance: A MSDS will list trade name(s) to help identify specific materials.

WA Time Weighted Average exposure; the airborne concentration of a material to which a person is exposed, averaged over the total exposure time-generally the total work-day (8 to 12 hours). See also, "TLV."

UEL OR UFL Upper Explosive Limit or Upper Flammable Limit of a vapor or gas; the highest concentration (highest percentage of the substance in air) that will produce a Jash o f f r e when an ignition source (heat, arc, orjame) is present. Importance: At higher concentrations, the mixture is too "rich" to bum. See also, "LEL."

UNSTABLE Tending toward decomposition or other unwanted chemical change during normal handling or storage. Importance: A MSDS will list materials that are unstable and conditions to avoid to prevent decomposition or unwanted chemical changes.

V A ~ R D E N S ~ ~ Y The weight of a vapor or gas compared to the weight of an equal volume of air; an expression of the density of the vapor or gas. Materials lighter than air have vapor densities less than 1.0 (example: acetylene, methane, hydrogen). Materials heavier than air (examples: propane, hydrogen sulJide, ethane, butane, chlorine, sulfur dioxide) have vapor densities greater than 1.0. Importance: All vapors and gases will mix with air, but the lighter materials will tend to rise and dissipate (unless confined). Heavier vapors and gases are likely to concentrate in low places-along or under floors, in sumps, sewers and manholes, in trenches and ditches -and can travel great distances undetected where they may create fire or health hazards.

VAWR PRESSURE The pressure exerted by a saturated vapor above its own liquid in a closed container. Importance: The higher the vapor pressure, the easier it is for a liquid to evaporate and fill the work area with vapors which can cause health or fire hazards.

VENTILA~ON See "General Exhaust," and "Local 1. t hould not be exceeded even Exhaust."

Problem No. 01

CHEMICAL ENGINEERING TOPIC: Fundamentals

SAFETY AND HEALTH CONCEPT: Toxic Exposure Control and Personal Protective Equipment; Toxicology and Industrial Hygiene

BACKGROUND: Ventilation is an extremely important method of reducing the level of toxic airborne contaminants in the workplace. Since it is impossible to eliminate absolutely all leakage, some method will always be needed to remove toxic materials from the air when they are present in the process streams.

Ventilation includes "general ventilation" (which is also sometimes referred as "dilution ventilation") and "local exhaust ventilation," which is a method of removing contaminants before they enter the workplace air. Local exhaust ventilation is much preferred as a method of contaminant control as the following problem should illustrate.

The Occupational Safety and Health Administration (OSHA) has set the permissible exposure limit (PEL) of vinyl chloride (VC) at 1.0 parts per million (ppm) as a maximum time-weighted average (TWA) for an eight hour work day, because VC is believed to be a human carcinogen. (A carcinogen is an agent that causes or promotes the initiation of cancer, therefore, exposure to a carcinogenic substance may increase the likelihood of the subject developing cancer in the future.) Exposure to VC on a long term basis (chronic exposure) may result in liver damage as well as some other symptoms. Acute exposure (one time exposure to relatively high concentrations) may cause central nervous system depression.

If VC escapes into the air, its concentration must be maintained at or below the PEL. If dilution ventilation were to be used, we might estimate the required air flow rate by assuming complete mixing in the workplace air, and then assume that the volume of air flow through the room will carry VC out with it at the concentration of 1.0 ppm.

PROBLEM: We have an operation where VC will evaporate at a rate of 10 glmin into the air. What flow rate of air will be necessary to maintain the PEL of 1.0 ppm by dilution ventilation? That is, what volume rate of air carrying VC at 1.0 ppm will be required to remove the VC at a rate of 10 glmin?

We must also correct for the fact that complete mixing will not be realized. A recommended way to do this is to multiply the air flow rate by a safety factor. In this case, use a factor of 10.

An alternative is to partially enclose the operation and use local exhaust ventilation. Assume that this operation can be carried out in a hood with an opening of 30 in. wide by 25 in. high. Imagine that this hood looks like the ones you see in your

2 SAFETY, HEALTH, AND LOSS PREVENTON IN CHEMICAL PROCESSES

chemistry laboratory. If the "face velocity," that is the average velocity of air entering the hood opening must be 100 ftlmin to effectively capture the VC vapor generated inside the hood, what air flow rate will be required?

Which way seems best to you? Explain why dilution ventilation is not recommended for maintaining air quality.

Problem No. 02

CHEMICAL ENGINEERING TOPIC: Fundamentals; Design


BACKGROUND: Ventilation is an extremely important method of reducing the level of toxic airborne contaminants in the workplace. Ventilation includes "general ventilation" (which is also sometimes referred as "dilution ventilation") and "local exhaust ventilation," which is a method of removing contaminants before they enter the workplace air. The local exhaust ventilation is much preferred as a method of contaminant control because it removes the contaminant before it can enter the workplace air. It also will require much less air flow if properly designed, which means less equipment and energy required for the job.

Many industrial operations involve the exposure of solvents to the air, and thus a problem of evaporation will occur. Most common solvents will display some sort of toxic effect, some of them more severe than others. Trichloroethylene is an excellent solvent for a number of applications, and is especially useful in degreas- ing. Unfortunately, tricloroethylene can lead to a number of harmful health effects. It has been shown to be carcinogenic in animal tests. (Carcinogenic means that exposure to the agent might increase the likelihood of the subject getting cancer at some time in the future.) It is also an irritant to the eyes and respiratory tract. Acute exposure causes depression of the central nervous system, producing symptoms of dizziness, tremors, and irregular heartbeat, plus others.

PROBLEM: Trichloroethylene has a molecular weight of approximately 131.5, so the vapors are much more dense than air. As a first thought, one would not expect to find a high concentration of this material above an open tank because we would assume that the vapor, being dense, would sink to the floor. If this were so, then we would place the inlet of a local exhaust hood for such a tank near the floor. However, Industrial Ventilation* points out that toxic concentrations of many materials are not much more dense than the air itself, so where there can be mixing with the air we may not assume that all the vapors will go to the floor. For the case of trichloroethylene OSHA has established a time-weighted average 8 hr PEL of 100 ppm; a 15-min ceiling of 200 ppm; and a 5-min peak of 300 ppm.

*Industrial Venti1ation:A Manual ofRecommended Practice, 19th ed. Cincinnati: ACGIH, 1986.

S A I E l Y , HEALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

Determine the density of a mixture of trichloroethylene in air at each of these limiting concentrations, as well as that of a saturated vapor at 2YC, and compare the values with that of pure air at the same temperature. That is, determine the specific gravity (relative to air) for each mixture. Which, if any, of these concentrations would you feel might readily sink to the floor, and which might circulate with the normal air currents which we would find in a room?

Problem No. 03

CHEMICAL ENGINEERING TOPIC: Fundamentals: Gas Mixture Composition and Amagat's Law of Partial Volumes

SAFETY AND HEALTH CONCEPT: Toxic Exposure Control and Personal Protective Equipment

BACKGROUND: It may not always be obvious to the workers when a breathing apparatus is required. Whenever workmen must enter a vessel for maintenance, entry should not be made until the air within the vessel is tested for flammable agents, oxygen content, and, ifneeded, for toxic agents. Lack of oxygen is the most common cause of death in vessel entry. As an example of what can happen, consider that in order to prevent fire or explosion, storage tanks or other vessels that have contained flammable materials are frequently purged with nitrogen prior to required maintenance. Purging with nitrogen will prevent fires from welding or spark-producing activities but if a worker should enter this vessel without a breathing apparatus, he will be quickly overcome from lack of oxygen.

Normal air contains about 21% oxygen, by volume. A typical adult's total lung '

volume is about 5.5 L. During normal breathing, each inspiration and expiration involves about 500 ml of air. Of this 500 ml, about 150 ml occupies the tracheobronchial tree, where no interchange of oxygen can take place with the blood. Thus, only 350 ml of air is actually exchanged in each inhalation.

The alveolar air, that is, the air fromdeepwithin the lungs that is exhaled contains only about 12% oxygen, but combined with that which remained in the tracheobronchial tree, the net exhaled composition is about 16%.

When the concentration of air being inhaled drops below 16% oxygen, symptoms of distress will appear. Because the lack of oxygen affects the central nervous system first, loss of consciousness is an early consequence, and will occur at concentrations below about 11 to 12% oxygen. Breathing will cease if the oxygen content drops below about 6%.

PROBLEM: If it is assumed that loss of consciousness occurs when the average concentration of oxygen in the lungs and tracheobronchial tract drops below 11%, estimate how many breaths a worker will be able to take when he enters a vessel that contains 100% nitrogen before he loses consciousness. If help comes in time, he may recover if he gets into fresh air before the average concentration drops below about 6%. How much time is there to help him? (A person who is breathing normally will inhale about 30 L/min at 500 mllinhalation.) What might you conclude about using air-purifying respirators rather than air-supplying respirators for such a case?

Problem No. 04

CHEMICAL ENGINEERING TOPIC: Fundamentals: Ideal Gas Law; Mass Balance

SAFETY AND HEALTH CONCEPT: Toxicology and Industrial Hygiene: Chronic Toxicity

BACKGROUND: Most of the common solvents used in the laboratory and industry are either toxic, flammable, or both. They may be acutely toxic, in which case care must be taken for even short exposures. There are other chemicals that may be tolerated for short times with no apparent immediate health effects, but which may cause serious health problems such as organ degeneration or cancer if people are exposed to them for a long time at toxic concentrations that are relatively low. These chemicals are classified as exhibiting chronic toxicity.

Federal standards, based on the toxicity of various chemicals, have been set for the "Permissible Exposure Limit," or PEL. The PEL is the maximum level of exposure permitted in the work place based on a time-weighted average (TWA) exposure. The TWA exposure is the average concentration permitted for exposure day after day without causing adverse effects. It is based on exposure for 8 hr per day for the worker's lifetime. Of course, the normal concentration of a toxic material is usually essentially zero, so that the worker is not exposed to the PEL concentration for more than short periods under unusual conditions. A "Short Term Exposure Limit" (STEL) is specified in the new standards and is based on a 15-min exposure. Both the STEL and the PEL are lower than the concentrations that are expected to cause injury. PELS and other TWA criteria may reflect either acute or chronic toxicity effect and should not be used as a comparative measure of toxicity except in a very broad way. A comparative end-point used in toxicology for acute inhalation exposure is called the LCso (lethal concentration, 50%). This is a statistically derived concentration that will kill 50% of a group of test animals following a short term exposure, and it is used to estimate the acute toxicity to man. Acute toxicity values are usually expressed for some animal species, and (not surprisingly) there is very little human exposure data.

PROBLEM: Some decades ago benzene was thought to be a relatively innocuous chemical with a somewhat pleasant odor and was widely used. It has been found that benzene can cause chronic adverse blood effects such as anemia and possibly leukemia with chronic exposure. Benzene has a PEL for 8-hr exposure of 1.0 ppm. If liquid benzene is evaporating into the air at a rate of 2.5 mllmin, what must the ventilation rate be to keep the concentration below the PEL? The ambient temperature is 6S°F and the pressure is 740 mm Hg.

Problem No. 05


HEALTH AND SAFETY CONCEPT: Vapor Releases

BACKGROUND: No matter how carefully workers do their jobs, the possibility of accidents remains. The more planning and preparation that has gone into accident anticipation and contingency planning, the better the chance of avoiding complications, injury, or property damage if and when an accident occurs. In planning for possible accidents, one of the more likely occurrences might be a release of a large quantity of toxic or flammable vapors or gases. Methods are available for estimating the resulting concentrations from such releases; and from such estimates it is sometimes possible to predict what areas of a plant or of the area surrounding a plant might have to be evacuated.

Computational methods used for such emission sources as power plant stacks can also be used for accidental releases. Such computations estimate the effect of dilution as a plume leaves the source location. A derivation of any of the methods is beyond the scope of this problem statement.

One of the simpler models to predict dispersion is called the "Gaussian Plume Model," and expresses the average concentration at a location downwind of a continuous source.

where m = y2/[2(uy)2] n = (z -~>~/ [2 (uz)~ l n' = (Z + H ) ~ / [ ~ ( O ~ ) ~ ] C = the concentration at a selected point downwind, mg/m3. u = wind velocity, m/s x = distance downwind from the source to the point of interest, m or km y = distance cross-wind away from the centerline to the point of interest, m z = height above ground level to the point of interest, m

H = the height of the source above ground, m uz = diffusion coefficient in thez direction (vertical), m uy = diffusion coefficient in they direction (cross-wind), m Q = source strength (emission rate), mgls

S A l T T Y , HEALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

This equation is valid for windblown plumes across fairly level ground. It is based on an assumption of how the components of the plume would be dispersed. The diffusion coefficients depend upon the stability of the atmosphere and the distance downwind from the source. The diffusion coefficients may be estimated for most purposes from the following:

a, =arb and az=cxd +f

Values for the constants, a, c, d, and f are given in the following table. The value of b is always 0.894 andx is expressed in kilometers:

Values of Constants to Compute Diffusion Coefficients

S tabiity x c lkm x > lkm class a c d f c d f

Note: The value of a is independent of the downwind distance, x.

The stability categories may be estimated from the following:

DAYLIGHT NIGHT Wind speed Sunlight intensity

( 4 s ) strong moderate weak Cloudy Clear

<2 A A-B B E F 2-3 A-B B C E F 3-5 B B-C C D E 5-6 C C-D D D D > 6 C D D D D

Suggested stability classes are from Turner, D.B., "Workbook for Atmospheric Dispersion Estimates," HEW, Washington, D.C., 1969.

PROBLEM NO. 5 9

PROBLEM: Emergency plans are being formulated so that rapid action can be taken in the event of an equipment failure. It is predicted that if a particular pipeline were to rupture it would release ammonia at a rate of 100 lblsec. Persons exposed to 500 parts per million (ppm) of ammonia will be endangered and anywhere that the concentration might be that high should be evacuated until repairs are made. What recommendation would you make as to how far from the rupture people should be evacuated iE

A. The wind is 6 miles per hour and the sun is shining brightly. B. The night is overcast and the wind is 10 miles per hour.

Constants for the d i i s ion coefficient estimates are from Martin, D.O., J. Air Pollution ControlAssoc., Vol. 26, No. 2 (1976).

Problem No. 06

CHEMICAL ENGINEERING TOPIC: Fundamentals: Gas Laws, Partial Pres- sures and Partial Volumes

HEALTH AND SAFETY CONCEPT: Toxicology and Industrial Hygiene

BACKGROUND: When toxic materials are used and/or produced by chemical processes, it is necessary to ascertain that the workers are not exposed to the material(s) to such an extent that they receive a harmful dose. Since the most frequent route of entry for toxic materials is by inhalation, ascertaining the extent of exposure or exposure potential often takes the form of determining the concentration of toxic material in the air that the workers breathe.

The methods of sampling, capturing, and analyzing the air vary a great deal- depending upon the nature of the contaminant, the method of analysis that is to be used, and the time period over which the information is desired. Also, sometimes we will want to know the concentration variations within a room- perhaps to study the efficiency of contaminant control practices - whereas, at other times, it may be desirable for the workers to carry "personal samplers," which are fastened to their clothing, as close as can conveniently be to their faces.

Particulate samples may be collected on a filter, with air being aspirated through the filter. Some method will be used to determine the sample volume, either a flow meter or a pump that moves a known volume of air per unit of time. The particles can be collected and analyzed as needed, although sometimes only the mass is required.

Many of the potentially harmful agents are vapors that may be in the air. The methods used for vapor concentration are quite varied. There are direct reading instruments, which do not actually accumulate samples, but only need to have sampled air pass through a detector. Usually these devices will output to arecorder or a computer so that both the time-weighted average (TWA) as well as the instantaneous maximum may be determined, and sometimes it may be helpful to know the time cycling of the concentration. All direct reading devices will have threshold concentrations below which the contaminant in question is not detectable. In these cases, and some others, it will be necessary to sample a large volume of air and concentrate the contaminant.

For any of various reasons it may not be appropriate to employ direct reading instruments, and sampling will have to be performed in some other way. Especially, it is not feasible to use direct reading instruments as personal samplers.

A sample taken over a very short time is called a "grab sample." A grab sample, or instantaneous sample is frequently collected in an evacuated container, and the contents subsequently analyzed by an instrument such as a gas chromatograph. In

PROBLEM NO. 06 11

such a procedure, the contaminant will have to be at such a concentration that direct analysis is possible.

Sampling may be carried out over an extended time, and the results of the subsequent analysis then represent an average, actually a time-weighted average, concentration. Such a procedure is called "integrated sampling." Methods of obtaining large samples over extended times involve passing the air sample through a collector device which might be an absorption train similar to that used for stack gas sampling, or might be an adsorption bed (e.g., a bed of granular material such as activated charcoal or silica gel). In this way, a time-weighted average can be obtained, and a contaminant in a large volume of air can be concentrated, in order to increase the sensitivity of the determination. Some care must be exercised to prevent exceeding the capacity of a concentrating collector, and some means must be available to determine that the capacitywas not exceeded. One common method is to use two sampl~ collectors in series. If the first collector is not overloaded, then there will be no contaminant collected in the second. If, however, a quantity of the contaminant is found on the second collector, the results of the analysis will be rejected.

PROBLEM: A 2-L grab sample of air (33"C, 99 kPa, and 70% relative humidity) was collected in a stainless steel container which had been evacuated to a hard vacuum. The sample was admitted to the container by opening a valve and allowing the air to enter until the pressures were equalized, whereupon, clean dry helium was admitted until the pressure was 500 kPa.

The sample was taken to a gas chromatography laboratory where the temperature was 23°C. The next day, a sample from the container was released to the chromatograph until the pressure in the container was reduced to 400 kPa. On analysis, the portion of the sample admitted to the instrument was found to contain 1.65 ng of benzene. What is the concentration of benzene in the original workroom air (in mgfm3), and is it in excess of the permissible exposure limit (PEL) of 1 part per million (ppm) on a mole basis?

NOTE: A grab sample as considered in this problem would not usually be used for determining compliance with the PEL, which is an 8-hr average limit.

Problem No. 07

CHEMICAL ENGINEERING TOPIC: Fundamentals: Mass Balance

SAFETY AND HEALTH CONCEPT: Inerting and Purging

BACKGROUND : Many chemical plant operations require that vessels and piping be inerted or purged. There are several reasons for this requirement: For example, when a plant is first constructed, the piping and vessels will be filled with air. The air may have to be purged because it interferes with the process; because it can lead to flammable mixtures with the chemicals that flow through the piping; and because flammable mixtures may be formed in vessels. If piping or vessels that contain a flammable or toxic material must be taken out of service and repaired or inspected, they must first be purged to remove the flammable or toxic material. Otherwise, workers will be exposed to the hazard of working under unsafe conditions.

If piping or vessel is being purged "into service," the air that is in the system at the start can be purged out with an inert material. Nitrogen is frequently used for purging into service because it is relatively cheap, it can be easily obtained, and it does not pollute the atmosphere during the purging process. If piping or vessel is being purged "out of service," the process is usually a little more complicated. First, an inert gas must be used to purge the flammable or toxic gas from the system. Then, the inert gas itself must be purged from the system using air if people are to enter the vessel. During the purging process, care must be taken that the exit stream, which contains varying concentrations of flammable or toxic material, is not discharged to the atmosphere. The hazardous material must be removed before discharge to prevent possible danger to workers and neighbors.

Liquids in piping and vessels present special problems. Sometimes the liquid is soluble in water and can simply be rinsed out. Sometimes a detergent solution can be used to wash the equipment. Sometimes a solvent must be used to wash the equipment and the solvent must then be rinsed or washed out with water. Other methods can also be devised. In all cases, the material washed or rinsed from the equipment must be disposed of properly. Elaborate procedures may have to be devised for some systems.

Whatever procedure is used to purge equipment, the final step should be to check the atmosphere in the equipment to make certain that the concentration of flammable or toxic material has been reduced to safe levels. The measurement of residual concentrations is sometimes required by law; it is always needed for good practice. Detailed procedures must be prepared and followed if workers are to enter a vessel. These include completion of a tank entry permit that lists the detailed procedures to be followed.

PROBLEM NO. 07 13

PROBLEM: A tank used for storing liquefied natural gas must be taken out of service and inspectedinternally. All the liquidnatural gas that canbe pumped from the tank is removed. The tank is then allowed to warm from its service temperature of about -260°F to ambient temperature. The tank then contains only natural gas (assumed to be pure methane) gas at ambient temperature and atmospheric pressure. Purging is accomplished in two steps: first, liquid nitrogen is sprayed gently onto the tank floor, where it vaporizes and displaces the methane. The cold nitrogen vapor displaces the warm natural gas in a piston-like flow as the nitrogen fills the tank. Once all the methane has been displaced, the nitrogen is allowed to warm to ambient temperature. Air is then blown into the tank. It mixes with the nitrogen rapidly and completely, so that the concentration of oxygen in the air-nitrogen mixture leaving the tank is equal to that in the tank.

a. How many gallons of liquid nitrogen will be required to displace all the methane from a tank with a total volume of 175,000 barrels (bbl)?

b. How many cubic feet of air will be required to increase the oxygen concentration to 20% by volume?

Problem No. 08


SAFETY AND HEALTH CONCEPT: Properties of Materials

BACKGROUND: Three things are commonly recognized as being required for a fire to occur: (1) a fuel, (2) an oxidizing agent, and (3) an ignition source. The combustion reaction normally occurs in the gas phase, and, in most cases, the oxidizer is air. If a flammable gas is mixed with air, there is a minimum gas concentration below which ignition will not occur. That concentration is called the lower flammable limit (LFL), and it is usually expressed in terms of the mole percent or volume percent of the flammable gas in air. If the gas concentration is less than the LFL, the gas mixture will not ignite. The LFL depends on the temperature of the gas-air mixture, and at temperatures as high as those in flames, essentially all the gas will burn. The LFL is usually measured at ambient temperature (25°C) and 1.0 atm pressure. There is usually an upper limit gas concentration above which ignition will not occur. It is called the upper flammable limit (UFL), and it is also measured at ambient temperature and atmospheric pressure. The range of concentrations between the LFL and the UFL is called the flammability region. Some gases have very wide flammability regions, and others are much narrower. The following data are examples of LFL and UFL for a few common flammable materials:

MATERIAL LFL UFL MoIe Percent

Methane 5.0 15.0 Propane 2.2 9.5 n-pentane 1.5 7.8 Hydrogen 4.0 75 Ammonia 16 25

These data were taken from the NFPAStandard325M, Properties of Flammable Liquids published by the National Fire Protection Association.

The LFL has been measured for many common materials and is available in the literature. Occasionally, it will be necessary to estimate the lower flammable limit of a gas. One simple method of doing so is based on the observation that for many hydrocarbons the LFL is about half the concentration required for stoichiometric combustion of the gas in air.

PROBLEM NO. 08 15

PROBLEM. Estimate the LFL for methane, propane, n-pentane, hydrogen, and ammonia. Compare your results to the data given above. In making the calculation, you may assume that air is 21 mole percent oxygen and the balance nitrogen. Assume that if hydrogen and carbon appear in the fuel molecule they will burn to water and carbon dioxide. If nitrogen is in the fuel molecule, the reaction products may contain a variety of oxidation products of nitrogen. Generically, these are labeled as NO,, and most of them are highly toxic. However, most of the nitrogen oxidation products are formed at very low concentrations at flame temperatures. Thus, for a fust approximation, it may be assumed that the combustion products are pure nitrogen if nitrogen is present in the fuel molecule. A small part of the nitrogen in the air is also oxidized to NO, in the fue, but again, the concentrations are small enough that the effect can be neglected for these calculations.

Problem No. 09


SAFETY AND HEALTH CONCEPT: Vapor Releases

BACKGROUND: Many industrial chemicals are toxic or flammable; frequently a chemical is both toxic and flammable. Regardless of whether the chemical is toxic or flammable, it can present a danger to plant operators and the public if it is released from its containers. Substantial effort is taken to assure that toxic or flammable materials are not spilled or released from containment. However, there is always a chance that such materials might be released, and if they are, provisions have to be made for protection of the plant operators and anyone who lives or works nearby.

One method of providing protection for the public is to locate plants well away from housing areas, shopping areas, or other public places. Sometimes, there is not sufficient land available to provide such separation distances as are desired, and sometimes people build houses and businesses near plants that are already operating. In any case, there may be times when a release of flammable or toxic material may occur under conditions that might endanger the public. If such an event is possible, the plant management should provide (1) methods for warning the public, and (2) a community emergency plan for the most likely situations so emergency personnel will be prepared to take immediate action. As part of the emergency response plan, there must be an estimate of the time available for the emergency response actions, such as evacuation, to take place.

Alternative methods can be used to mitigate the effects of a hazardous material release. They include rapid shutdown of process equipment to limit the amount of hazardous material released, backup instrumentation to reduce the probability of losing process control, and measures to reduce the impact of a release that has occurred. If a liquid is released, one method of mitigation is to limit its spread by providing impounding dikes. Chemicals can sometimes be neutralized by chemical reactions. Mitigation methods usually depend on the properties of the chemical that is released. If the chemical is released as a gas, mitigation is harder because the movement of the gas cannot be controlled.

If a toxic material is dispersed in the air, the engineer must know how high its concentration can rise without danger to people. The Occupational Safety and Health Administration has set a concentration level for many chemicals called the permissible exposure limit (PEL). The PEL is the maximum time-weighted average concentration that a worker may be exposed to for an 8-hr work day for his lifetime. Another concentration that is frequently referred to is the concentra-

PROBLEM NO. 09 17

tion that is immediately dangerous to life or health (IDLH). This concentration is the level at or below which a person exposed for 30 min will not lose control and will be able to put on protective equipment or take other protective action. The IDLH concentration is higher than the PEL concentration, and people exposed at the IDLH concentration must take action to protect themselves. An even higher concentration frequently used in the LCso concentration. The LCso concentration is a concentration at which 50% of a group of test animals would die if exposed to the hazardous chemical during a standard test. This concentration is frequently used as an estimate of the degree of danger to humans as well. (Most toxicity studies are performed using test animals. Humans obviously cannot be exposed to lethal concentrations of toxic chemicals for determining toxicity. There are some differences in tolerance for chemicals for man and animals caused by differences in metabolism and other factors, so the animal toxicity tests are not always easy to interpret. However, the results of animal toxicity tests are used to guide the selection of acceptable exposure limits for humans.)

If the hazardous material spill is of short duration, there is a choice to be made concerning evacuation. If people are in their houses, it will take some time before the toxic material can enter the house. If the release is small, it may be more dangerous to try to evacuate people than to have them remain inside (with the doors and windows closed and air circulation stopped) until the danger is past. This and other options should be considered as part of the emergency action plan.

PROBLEM: A chemical plant uses acrolein (acrylaldehyde) as an intermediate in its chemical process. The nearest residences to the plant are 3000 ft from the point where a spill is most likely to occur. It has been estimated that under adverse atmospheric conditions, if a release occurs at the plant, the concentration of acrolein in air at the nearest residence can reach a maximum of 10 ppm. The release will be liquid, which will be contained in an impounding area where it will slowly evaporate. The concentration at the nearest residence is well above the IDLH concentration, and plans are made for the evacuation of the residents in the event of a release. The wind speed under the adverse conditions is 2.2 mph. It can be assumed that the houses admit air from the outside at a rate of three air changes per hour. The air in the houses can be assumed to be well mixed, so any gas that enters will be at a uniform concentration throughout the house.

1. Assuming the vapor from the acrolein spill moves at the same speed as the wind, how much time is available for evacuating people from the residences before the toxic vapor arrives?

2. How much additional time will be required for the concentration of acrolein to increase to the IDLH concentration inside the residences? Begin by writing an unsteady state mass balance for the acrolein in the house. Account for the acrolein in the house as well as that entering from the outside and that leaving from the inside. Assume the density of air in the house is constant.

18 SAFETY, HEALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

3. Compare the results of Part 2 with those of Part 1. How might the difference influence your judgment of the time operators have to complete mitigation procedures?

4. Can you suggest simple hazard mitigation methods that might be used to reduce the acrolein concentration at the residences?

Problem No. 10

CHEMICAL ENGINEERING TOPIC: Fundamentals: Stoichiometry


BACKGROUND: Three criteria must be met if a fue is to occur: (1) there must be fuel present; (2) there must be an oxidizer present; and (3) there must be an ignition source. In many industrial processes, the materials that are stored, transported, or reacted to manufacture new materials are flammable. Other chemicals are used as fuels, so they must be flammable if they are to have any value. While using or handling fuels can be done safely, care is required to assure that ignition takes place only where and when it is wanted.

For most fuels, combustion takes place only in the gas phase. For example, gasoline does not burn as a liquid. However, when gasoline is vaporized, it burns readily. Controlled burning, such as occurs in an automobile engine, is beneficial and can be used to generate work. Uncontrolled burning, such as might occur if gasoline is spilled and ignited, is wasteful in all cases, and it is dangerous in many cases. Other flammable liquids are similar. If a flammable gas is released into air, it will also form flammable mixtures.

There is a minimum concentration of fuel in air that can be ignited. If the fuel concentration is less than this lower flammable limit (LFL) concentration, ignition will not occur. Above the LFL, the amount of energy required for ignition is quite small. For example, a spark can easily ignite most flammable mixtures. There is also a fuel concentration called the upper flammable limit (UFL) above which the fuel-air mixture cannot be ignited. Fuel-air mixtures in the flammable concentration region between the LFL and the UFL can be ignited. Both the LFL and the UFL have been measured for most of the common flammable gases and volatile liquids. The LFL is usually the more important of the flammability concentrations because if a fuel is present in the atmosphere in concentrations above the UFL, it will certainly be present within the flammable concentration region at some location. LFL concentrations for many materials can be found in the NFPA Standard 325M, "Properties of Flammable Liquids," published by the National Fire Protection Association.

The LFL represents the minimum concentration of fuel that must be present in air for the mixture to be flammable. It is usually expressed as a volume percent, which is equal to the mole percent under conditions at which the LFL is measured (atmospheric pressure and 25°C). There is also a minimum oxygen concentration required for ignition of any fuel. It is closely related to the LFL and can be calculated from the LFL. The theory on which the calculation is based is that the

20 SAFET"Y, HEALTH, AND LOSS PREVENTlONIN CHEMICALPROCESSES

minimum amount of oxygen required for ignition is the stoichiometric quantity required for complete combustion of the fuel at the LFL concentration. This minimum oxygen concentration is frequently used as the maximum permissible oxygen concentration in storage tanks or other places where fuel vapors may be present under ordinary circumstances. In most industrial applications, an experimentally measured minimum oxygen concentration is used.

The minimum oxygen concentration required for ignition can be estimated by multiplying the LFL concentration by the ratio of the number of moles of oxygen required for complete combustion to the number of moles of fuel being burned. The minimum oxygen concentration estimated by this method may not be accurate enough for all purposes, particularly for some especially reactive substances. For example, gases such as acetylene can decompose by an exothermic reaction that proceeds very much like a combustion reaction even though there is no oxygen present. In such cases there is no limiting oxygen concentration, because the reaction can occur without the presence of oxygen. The important thing to remember is that the properties of the particular substance being considered must be known and accounted for in any system design.

The oxygen concentration in a flammable mixture can be reduced by adding fuel vapor to the air or by adding an inert material such as nitrogen or carbon dioxide. If nitrogen is the diluent, estimating the maximum permissible oxygen concentration from the stoichiometry of the reaction works quite well. However, if carbon dioxide is the inerting gas, slightly higher maximum permissible oxygen concentrations are measured. The higher maximum permissible oxygen concentrations measured when carbon dioxide is present are due to the higher specific heat of carbon dioxide. Some other inerting agents, such as the halons (halogenated hydrocarbons) are effective at inhibiting ignition at even higher oxygen concentrations. Their effect is due to chemical effects rather than physical effects.

PROBLEM: Estimate the maximum permissible oxygen concentration for n- butane. The LFL concentration for n-butane is 1.9 mole percent.

(This problem is based on a problem in the text Chemical Pmess Safety: Fundamentals with Applications, by D. A. Crow1 and J. F. Louvar, published by Prentice Hall, Englewood Cliffs, NJ.)

Problem No. 11

CHEMICAL ENGINEERING TOPIC: Fundamentals: Mass Balance


BACKGROUND: Many chemical plant operations require that vessels and piping be inerted or purged. If a vessel is to be opened for maintenance or repair, for example, and if the vessel has contained either toxic or flammable materials, purging is required before workers can enter the vessel. For a vessel entry, piping leading to or from the vessel will have to be blanked off and at least the portions of the vessel that are open to the pipe will have to be purged as well. Purging must be continued until the atmosphere in the vessel is safe for entry.

If the chemical in the vessel is flammable, purging must be accomplished in two steps: first, the flammable material is purged from the vessel with an inert gas such as nitrogen, and the nitrogen is then purged with air. When a vessel that is to be used for storing or processing flammable chemicals is initially put into service, it must be purged with an inert gas before flammable chemicals are put in the vessel. This step is required to assure that a flammable mixture of the chemical and the air in the tank does not form.

If the vessel or piping contains liquids, purging is sometimes accomplished by washing. Either water or a detergent solution is used to wash the vessel or piping. Usually, the vessel must then be dried before it is returned to service to prevent contamination of its contents.

Regardless of the method used to purge equipment, the final step should be to check the atmosphere in the equipment to make certain that the concentration of flammable or toxic material has been reduced to safe levels. For tank entry, the oxygen concentration must also be checked before the tank is entered. The measurement of residual concentrations is sometimes required by law; it is always needed for good practice. Detailed procedures must be prepared and followed if workers are to enter a vessel. These procedures include completion of a tank entry permit that lists the detailed procedures to be followed.

PROBLEM: A 150-ft3 tank containing air is to be inerted to 1% oxygen concentration. Pure nitrogen is available for the job. The tank has a maximum allowable working pressure of 150 psia, so either of two methods is possible. In the first method, air is purged by a continuous sweep of nitrogen. The nitrogen is simply allowed to flow into the tank at essentially atmospheric pressure. It is assumed that the nitrogen mixes rapidly and completely with the air in the tank, so the gas leaving the tank has the same concentration of oxygen as the gas in the tank.


In the second technique, the tank is pressurized, the pure nitrogen inlet stream is turned off, and the gas mixture in the tank is then exhausted, lowering the pressure in the tank to atmospheric pressure. If the pressurization technique is used, multiple pressurization cycles may be required, with the tank returned to atmospheric pressure at the end of each cycle. Complete mixing is assumed for each cycle.

In this problem, you may assume that both nitrogen and air behave as ideal gases and that the temperature remains constant at 80°F throughout the process. Deter- mine the volume of nitrogen (measured at standard conditions of 1.0 atrn and 0°C) required to purge the tank using each purging technique. For the pressurization technique, assume the pressure in the tank is raised to 140 psig (a little below its maximum working pressure) with nitrogen and then vented to 0 psig.

(This problem is based on a problem in the text Chemical Process Safety: Fundamentals with Applications, by D. A. Crow1 and J. F. Louvar, published by Prentice Hall, Englewood Cliffs, NJ.)

Problem No. 12


SAFETY AND HEALTH CONCEPT: Explosions

BACKGROUND: There are many cases when it is good practice to vent an enclosure or building so structural and mechanical damage is limited in the event of a deflagration within the enclosure. The National Fire Protection Association Standard 68, "Guide for Venting of Deflagrations," provides guidance on how the enclosure should be built to provide proper venting. The discussion following is based very closely on NFPA 68. The most recent version of NFPA 68 is the 1988 edition. The standard is updated periodically, as are all NFPA Standards.

The American Heritage Dictionaly defines an explosion as "The sudden rapid violent release of mechanical, chemical, or nuclear energy from a confined region; especially, such a release that generated a radially propagating shock wave accom- panied by a loud, sharp report, flying debris, heat, light, and fire." NFPA 68 provides a much more restricted definition: "The bursting or rupture of an enclosure or a container due to the development of internal pressure from a deflagration." A deflagration is "Propagation of a combustion zone at a velocity that is less than the speed of sound in the unreacted medium." A deflagration is differentiated from a detonation, which is, in NFPA parlance, "Propagation of a combustion zone at a velocity that is greater then the speed of sound in the unreacted medium." Obviously, the technical definitions used by NFPA are intended to be more precise than those used in ordinary conversation.

We will be interested in three kinds of deflagrations: those involving gases, those involving dusts, and those involving mists. Deflagrations can occur in any flammable gas if the concentration of the gas in air is within the flammable concentration range. The flammable concentration range of gases is a unique property of the gas at a given temperature and pressure. Deflagrations in dusts are more difficult to quantify. First, since the dust is in the form of fine particles (NFPA 68 defines dust to be composed of particles 420 microns or less in diameter), the rate of deflagration will depend on the size of the dust particles. The concentration of dust required to sustain a deflagration will also depend on the size of the particles. Dusts may adsorb or absorb moisture from the atmosphere, and the amount of moisture may also affect the deflagration rate. Mists are fine liquid droplets dispersed in the atmosphere. Their combustion properties also depend on the droplet size and concentration. Many mists will ignite easily, even though the same liquid would not ignite if in a pool. Combinations of gases, dusts, and mists also occur frequently, leading to even more complex situations.

24 S A F E T Y , HEALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

Three things must be present for a deflagration to occur. There must be a fuel in the proper concentration, an oxidant in great enough quantity to support the combustion, and an ignition source strong enough to ignite the fuel-oxidant mixture. The most frequent oxidant is oxygen in air, and we will limit our attention to mixtures of fuel in air; but it should be kept in mind that other oxidants may be present in special situations. The ignition source ordinarily need not be very strong. A spark or a small open flame is usually all that is necessary to ignite a flammable mixture.

NFPA 68 provides a simple venting equation for determining the vent area required to prevent unusually large damage to a low pressure structure (less than 1.5 psig), for example a building. The equation is

where 2 2 Av = vent area (ft or m )

C = venting equation constant As = internal surface area of enclosure (ft2 or rn2)

Pred = maximum internal pressure that can bewithstood by the weakest structural element (psi or kPa)

The venting equation constant, C, is given in NFPA 68 for various classes of gases and dusts, and is based on experiments run with the actual dust or gas involved. It is important that the dust be tested to establish the value of C to ensure reliable results from the venting equation. The constant C has units of (pressure)05. One side of the enclosure is always assumed to be at atmospheric pressure, so the pressure used in the equation is the gauge pressure. The internal surface area of the enclosure includes the floor, the. ceiling, and the walls (i.e., all the area of the inside of the enclosure). There are other cautions that must be taken for the final design of the vents, including the design of the vent covering, the provision of sufficient strength for reaction forces, and the location of structural members within the enclosure. A detailed design will require close reference to NFPA 68 and careful attention to all its provisions. However, we can make a quick estimate of the area required from the venting equation for a simple situation.

PROBLEM: A building is 200 ft long, 50 ft wide, and has a 15-ft-high ceiling. It is connected to a tee at one end that is 80 ft long and 30 ft wide, with a 15-ft-high ceiling. The plan view is shown in the sketch on page 25. Corn starch and powdered nonfat dry milk are being packaged in the building. Corn starch is in Dust Hazard Class 2, which has a value of C of 0.12 (psi)05 and owdered nonfat dry milk is in

0s Dust Hazard Class 1, with a C value of 0.10 (psi) .

PROBLEM N0.12

Estimate the venting area required for the building for each of the products. A structural analysis of the building has determined that it can withstand a maximum internal overpressure of 0.25 psi.

Problem No. 13


SAFETY AND HEALTH CONCEPT: Process Control, Interlocks, and Alarms

BACKGROUND: Almost all chemical engineering processes require some form of process control in order to be operated safely and efficiently. In many cases, the instrument that is used for detection or control is separated from the process area because the analytical equipment must be located indoors or away from the environment of the process area. In such cases, samples are frequently taken at a process area and then transmitted to the analytical equipment through relatively long runs of small diameter tubing. Thus, there is a delay between the time a process variable changes and the time the change is detected by the analytical equipment. The analytical equipment may be used to raise process alarms as well as for process control, and it may be a critical element in process control for emergency shutdown. There may be a significant difference between the time a process goes outside of its normal control limits and the time emergency operations begin. Delay in normal control can lead to delay in process changes, which may, in turn, lead to product deterioration and safety problems. Process design should enable samples to be taken and analyzed quickly enough for proper process control, as well as for proper alarms and emergency shutdown.

PROBLEM: In a chemical plant, air samples from a process area are continuously drawn through a %in. diameter tube to an analytical instrument located 125 ft from the process area. The Win. tubing has an outside diameter of 0.25 in. 6.35 I mm) and a wall thickness of 0.030 in. (0.762 mm). The sampling rate is 10 cm /sec under ambient conditions of 22°C and 1.0 atm. The pressure drop in the transfer line can be considered negligible. Chlorine gas is used in the process, and if it leaks from the process, it can poison workers who might be in the area of the leak. Determine the time required to detect a leak of chlorine in the process area with the equipment currently installed. You may assume the analytical equipment takes 5 sec to respond once the gas reachesthe instrument. You may also assume that samples travel through the instrument sample tubing without dilution by mixing with the air ahead of the sample. Suggest methods of reducing the sampling time if the current detection time seems too long to be acceptable.

How would your answer be affected if the sampling system is connected to a sequential sampler which must sample five different streams, each of which is drawn separately and has the same total sample delay time? If the samples are

PROBLEM NO. 13 27

drawn to the detector continuously and the instrument response continues to be 5 sec, what is the maximum sampling delay time?

Would you fmd the system acceptable if the process contained a less toxic material, such as ammonia?

Both ammonia and chlorine have good warning properties (they can be smelled at very low concentrations). Would a material such as carbon monoxide, which cannot be detected by human senses, require different considerations? Would a diierent detection system be required?

Problem No. 14


SAFETY AND HEALTH CONCEPT. Explosions

BACKGROUND: National Fire Protection Association Standard No. 68 (1988 edition) defines an explosion as "The bursting or rupture of an enclosure or container due to the development of internal pressure from a deflagration." A deflagration is "Propagation of a combustion zone at a velocity that is less than the speed of sound in the unreacted medium." Detonation, which is "Propagation of a combustion zone at a velocity that is greater than the speed of sound in the unreacted medium," may also occur, and if the total energy release is the same, a detonation will be more damaging.

Deflagrations may occurwhen three types of fuel-air mixtures are present: those with fuels in the form of gases, mists, and dusts. The fuel must be within the flammable range for the deflagration to be initiated. The oxygen in the air serves as the oxidant, and a small spark or flame is all that is needed to ignite the mixture. If a deflagration occurs in a closed space, such as a building, an explosion may occur in which the building is destroyed. There are several methods available for preventing an explosion, including prevention of formation of flammable mixtures in the building. If, despite efforts at prevention, flammable mixtures do form and they are ignited, the damage to the building may be reduced by venting the deflagration. NFPA 68 provides a simple equation for determining the vent area required to prevent unusually large damage to a building if a deflagration occurs in the building. The equation is

where 2 2 Av = vent area (ft or m )

C = venting equation constant 2 2 As = internal surface area of enclosure (ft or m )

Pred = maximum internal pressure that can be withstood by the weakest structural element (psi or kPa)

The venting equation constant is given in NFPA 68 for various classes of gases and dusts. It is based on an extensive set of experimental tests using the gas or dust in question. The tests are difficult to perform, and the results vary somewhat from test to test, so the NFPA 68 values are meant to be used as a guide to establish

PROBLEM NO. 14 29

conservative design bases. The units of the constant are and it as always assumed that the outside of the enclosure is at atmospheric pressure, so the pressure used in the equation is the gauge pressure. The internal surface area includes the floor, the walls, and the ceiling. While a complete design will require close reference to NFPA 68, we can make a simple estimate of the vent area required from the venting equation for simple building layouts.

PROBLEM: A building housing natural gas compressors is 300 ft long, 75 ft wide, and has walls 20 ft high. The roof is gabled and has a pitch of 3/12 (i.e., the roof rises 3 ft for each 12 ft of horizontal run). Natural gas is primarily methane, for which Cis 0.37 (kpa)O5 and the building is designed for a maximum overpressure of 0.3 psi. Estimate the venting area required for the building.

Problem No. 15


HEALTH AND SAFETY CONCEPT: Storing, Handling, and Transport; Static Electricity

BACKGROUND: A number of accidents have occurred in the past due to discharge of electrical charges accumulated on the surface of objects or materials that are not properly grounded. The accumulated charge of this nature is called "static electricity." Static electricity is generated whenever objects of a different conductivity are brought together and then separated, or whenever the objects are rubbed together. Flowing fluids could generate such static charges.

If accumulated electrical charge is not dissipated to the ground through low- resistance conductors, then there exists a possibility that the potential will build up to such an extent that sparking will be produced. A common example is lightning, which is the discharge of static charges from clouds.

Since it is frequently impractical to prevent the generation of static charge, it is appropriate to provide means by which the charges can be dissipated without the accumulation of large potential differences required for arcing.

Charge accumulation is usually prevented by a combination of techniques, which include providing readily available grounding, the bonding of conducting equipment together (with electrical conductors), and the minimizing of generation by use of appropriate flow patterns.

Another obvious method of preventing accidents is to avoid the presence of flammable or explosive mixtures. While this may frequently be impractical, it might sometimes be a safety practice that should be considered. In the problem below, you are asked to determine what temperatures might be required in order to prevent the formation of flammable mixtures in air for a number of flammable solvents.

PROBLEM: In order to provide an added measure of protection against explosion or fire due to static electricity discharging, you are asked to determine what temperatures should be employed in drum-filling operations so that flammable mixtures, that is concentrations equal to or greater than the lower flammable limit (LFL), will not be produced by vapor air mixtures in equilibrium with the liquids.

The liquids being considered are the following:

Acetone Toluene Methyl ethyl ketone Ethyl benzene Benzene m-Xylene

PROBLEM NO. 15 31

Your instructor will provide you with literature references for the LFL values, or else will supply the data. To determine the maximum temperature of handling so as not to create a concentration greater than the LFL, you should determine the temperature where the vapor pressure is such as to just produce such a mixture.

As an additional precaution, it is desirable to prevent the generation of concentrations greater than 25% of the LFL. Determine also what temperatures this would require.

Please make a judgment regarding which, if any, of these temperatures would be practical for drum-filling operations. If you feel that the temperature control method is impractical, try to suggest an alternative method for the prevention of the formation of flammable mixtures for the drum-filling operation.

Problem No. 16

CHEMICAL ENGINEERING TOPIC: Fundamentals; Design

HEALTH AND SAFETY CONCEPT. Toxicology and Industrial Hygiene; Protec- tive Equipment

BACKGROUND: A potential hazard of the workplace that is sometimes over- looked is excessive noise. Excessive noise has been found to cause several types of physiological damage. Damage to hearing is perhaps the most important type of damage, but there are others, including heart disease, ulcers, and other stress-related diseases. Excessive noise can also interfere with communication, interfere with sleep or relaxation, and could be the cause of failure to hear an alarm.

Noise, of course, is sound. Sound is caused by rapid fluctuations of air pressure on the eardrum of the listener. The sound pressure means the root mean square value of the pressure changes above and below atmospheric.

The usual value cited for the threshold of audibility, that is, the minimum sound level that can typically be heard is a sound pressure of about 0.00002 Pa. At the upper limit, where there is physical pain associated with the sound, the sound pressure is about 20 Pa. The range of interest then, is a million fold or more. To accommodate such a wide range, one could not readily use a linear scale, so a logarithmic scale is used. The scale used is called a decibel scale, and the sound level units are called decibels.

The decibel scale is arranged so that the lower limit of audibility, that is, 0.00002 Pa sound pressure, corresponds to 0 decibels (dB). Then, the sound level, or intensity, is defined by

where I is the sound pressure and l o is the reference level (i.e., 0.00002 Pa). The 20-Pa sound pressure corresponds to 120 dB.

Following are some typical sound levels for your comparison:

Approximate Activity sound level (dB)

Turbojet Engine 160 Compressor 120 Rock and Roll Band 112 Power Lawn Mower 95 Conversation 70 Quiet Room 40

PROBLEM NO. 16 33

Due to the possibility of health effects as well as for comfort, the American Conference of Governmental Industrial Hygienists (ACGIH) ,has established acceptable "doses" of noise. The time of exposure is analogous to a dose of toxic material. The acceptable time is related to the intensity according to the following table:

Duration (hrlday)

16 8 4 2 1

112 114 118

Sound level (dB)

80 85 90 95

100 105 110 115

There is to be no exposure to either continuous or intermittent levels above 115 dB without hearing protection. If workers are exposed to more than one sound source, then the cumulative effect is to be considered. The cumulative effect is evaluated by dividing the actual time of exposure by the permissible exposure time for each source. These quantities are then added. If the sum is greater than unity, the exposure limit is exceeded.

This can be expressed in equation form as follows:

where T is the actual time of exposure and P is the permissible time of exposure. If E is greater than or equal to unity then hearing protection would be required, or the noise level must be reduced.

PROBLEM: A worker is required to spend his or her working day in four different locations, each of which has different noise levels. The various locations are described as follows:

Location A, where a compressor is operating with a noise level of 100 dB. Location B, where a blower and a pump combine to produce a noise level of 95 dB. Location C, where a packaging operation causes a noise level of 110 dB. Location D, which is a computer room, where the noise level is 85 dB.

Walking in between locations, the noise level is usually about 80 dB. Approximately 20 min is spent in Location A and 45 min in Location B. The worker will spend 5 min in Location C, and 4.5 hr in Location D. The rest of the 8-hr work day is spent


walking between locations. There is also the half-hour lunch break, where the noise level is about 75 dB.

Would hearing protection be required for this worker in order to satisfy the ACGIH guidelines?

If hearing protection were supplied in locations A, B, and C, so that the worker would be exposed to a constant level of 80 dB while in these locations, would that be adequate?

What would you say about this person spending 2 hr in the evening at a rock concert?

Problem No. 17

CHEMICAL ENGINEERING TOPIC: Fluid Mechanics


BACKGROUND: When designing a ventilation system, the engineer must bear in mind that an adequate duct system must be provided. If the hoods pick up the various contaminants, there will be nothing gained if the duct system is inadequate to remove the material collected. Although a complete ventilation system may grow by modification over time, it must still be an integrated unit capable of performing its function.

A complete ventilation system consists of the hoods, where the contaminants are picked up with the air; a duct system, which serves as a path for removing the contaminants; air-moving equipment (blowers), which supplies the energy to move the air; air inlets, which must be supplied with clean air, either fresh air or cleaned recirculated air; and a discharge system. Often it is also appropriate to have an air cleaner included before the blower or fan prior to discharge, and most certainly prior to recirculation.

Particulate matter may settle out in the ducts if the velocity is not adequate to transport the particles. For smokes, fumes, gases and vapors, any economicvelocity is adequate. However, for larger or heavier particles, if the duct velocity is too low, the particles will settle out and perhaps block the flow, or the load may be heavy enough to cause part of the ductwork to collapse. The book, Industrial Ventila- tion -A Manual of Recommended Practice (19th ed., ACGIH, 1986), provides a great deal of both practical and theoretical information on the design of ventilation systems. Included in the book are recommended duct velocities, on which the table on the next page is based.

Iron foundries, where cast iron products are manufactured, have particularly difficult problems with the ventilation scheme because there is a great deal of dust generated from the grinding of castings. Transport of the grinding dust through the ductwork requires a fairly high velocity. Also, silica, an especially undesirable dust, is a large component of the grindings because the casting molds are made of sand. Silica inhalation can lead to the disease known as "silicosis," an impairment of lung function, frequently incapacitating. Thus, it is important that a ventilation system function so as to minimize the exposure of workers to airborne silica.

PROBLEM: Foundry grindings will be transported through a single duct from five grinding work stations. Each work station will have a hood that requires 3000 CFM


of air flow. Determine the required duct diameter to assure adequate transport of the dust. Find the power required for a combined motorblower efficiency of 40%. The duct equivalent length is 400 ft and there is an entrance loss of 0.4 in. H20 at the hood entrances. There is a cyclone air cleaner with a pressure drop of 4.1 in. H20.

Motors are available in increments of 1 hp up to 5 hp, and in increments of 5 hp above that. Recommend a motor size. Refer to the duct velocity table, which follows.

Recommended Duct Velocities (Minimum Velocities Recommended for the Transport of Various Types of Contaminant)

Type of Contaminant Examples Recommended Velocity (ftlmin)

Vapors, gases, smokes, fumes

Very fine light dust

Dry dusts and powders

Average industrial dust general foundry dust

Heavy Dusts

Heavy or moist dust (very heavy dust)

All vapors, gases, Any economic metal oxide fumes velocity

Cotton lint, wood flour 2000-2500

Fine rubber dust, cotton 2500-3000 dust, light shavings

Sawdust, grinding dust, 3500-4000 limestone dust

Metal turnings, sand 4000-4500 blast dust, lead dust

Lead dust with small 4500 and up chips, moist cement dust quick lime dust

Adapted from Industrial Ventilation -A Manual of Recommended Practice (19th ed. Cincinnati: ACGIH, 1986), which should be consulted for additional detail.

Problem No. 18



BACKGROUND: Respirators are devices that are worn over the face to prevent inhaling harmful material. Normal practice is to provide workplace air which is suitable for breathing without any such protection, but there may be times when the systems providing for clean air fail to do so for any of various reasons. At these times, respirators may be the only method of protection available. They are therefore very important safety equipment items, and it is vital that they be used appropriately, in recognition of their limitations. Most of the other devices that maybe used for worker protection have some kind of backup in the event of failure, but for respirators, there is none.

Chemical cartridge respirators provide protection against vapors and gases being inhaled. One type of device uses an adsorbent, such as charcoal to adsorb organic vapors and thus to purify the air that the wearer inhales. The bed of charcoal will remove essentially all of the contaminant until breakthrough occurs, after which the concentration will rise very rapidly.

Respirators are frequently used in dusty conditions to purify the breathing air of workers who must be there. If the only contaminant is a particulate material, then chemical adsorption or reaction will not be required and only mechanical filtration will be needed. A respirator that serves only to remove particulate from the breathing air will continue to serve adequately until the pressure drop across the filter element and the accumulated cake becomes excessive. The limitation on pressure drop may be the worker and his respiratory capabilities, or it may be that higher pressure drops (lower pressure under the facepiece) promote leakage, or both.

Individuals diier a great deal in their ability to tolerate pressure drop through a breathing device. A worker who feels that he or she is "gasping for air" will be tempted to remove the respirator and inhale deeply. Thus, the pressure drop must be kept low enough to prevent the sensation of being short of breath.

OSHA has established criteria for the pressure drop across filtering respirators. Specifically, at a flow rate of 85 Llmin, the initial (clean filter) pressure drop may be no more than 20 mm H20, and after a specified test, no more than 50 mm H20, for dusts, fumes, and mists with a single use filter. (There are different requirements for some other situations.)

38 SAFETY, HEALTH, AND LOSS PREVENTION IN C H E M I C A

PROBLEM: A particular filter was tested for compliance by passing a test dust through the filter medium at 32Llminfor 90 min. Before the test, the pressure drop through the filter was 17 mm H20 at a flow rate of 85 Llmin and after the test was 43 mm Hz0 at a flow rate of 85 Llmin. The test dust was at a concentration of 54 mg/m3.

If this filter element were to be used by a worker for an extended time, breathing at a rate of 40 Llmin (average), how often will the filter element have to be changed if the worker's pressure drop tolerance is 35 mm Hz0 and if he is breathing a dust which is at an average concentration of 16 mg/m3?

Note that if the worker's respiration rate is 40 Wmin, remember that he spends only about half of the time inhaling, and the other half exhaling. Thus the flow rate through the filter during inhalation is more than 40 Llmin.

Problem No. 19


HEALTH AND SAFETY CONCEPT: Process Control, Interlocks and Alarms

BACKGROUND: Instrumentation is widely used in the modern chemical processing plant to maintain process variables within acceptable ranges. Keeping the process variables within range is also an absolute necessity when there are potential hazards in the process.

So many of our modern practices involve high temperature and pressure that instrumentation must be heavily depended on to provide the controls to prevent accidents. However, instruments sometimes fail to function, and sometimes the utilities to an instrument may fail. Electronic instruments will provide no protection when the power supply fails, and pneumatic instruments will not function if the air supply fails. Thus, when disastrous consequences will result from instrument utility failure, it is necessary to have backup.

Pneumatic instruments require compressed air as their utility. In a large or even a modest-sized chemical complex, the compressed air will come from a centrally located compressor, which will supply air for a large number of instruments. A backup supply is often provided through the use of portable cylinders of compressed air, and occasionally nitrogen instead of air. In operation, pneumatic instruments bleed air at a somewhat variable rate through an orifice. The orifice is slightly restricted by a flapper, so that the position of the flapper relative to the outlet of the orifice varies the pressure behind the orifice. It is this pressure variation that is used to control the valves or other control elements.

PROBLEM: A small control room has 12 pneumatic controllers. During a plant emergency, the instrument air compressor that supplies air for these controllers suffered damage and no longer operates. Automatic tripping devices enabled the emergency supply for this control room. The emergency supply consists of a cylinder of compressed air. The cylinder volume is 1.6 ft3, and has air at 2200 psig. We wish to estimate how long the supply will last.

Assume that each of the instruments has a 0.75-mm diameter orifice, and that due to the presence of the flapper, the average effective orifice coefficient is 0.45.

The pressure behiid the orifice ranges from 5 to 20 psig, usually. Assume that it is at 14 psig, as an average, during the emergency.

When the available pressure drops to no more than 25 psig, the devices will no longer function.


Now assume that a maintenance crew had replaced the air with nitrogen at the same temperature and pressure because the air was needed somewhere else. Furthermore, since nitrogen will enter the control room, the oxygen content of the air in the room will start to decrease. If a worker enters a room where nitrogen has dispIaced air so that the oxygen content is below about 16%, he or she will be overcome and lose consciousness, probably before realizing that anything is wrong. If no outside air enters the room, but the air in the room bleeds out, can the oxygen content be reduced to below 16% by the nitrogen? The room is 8 ft by 10 ft by 8 ft high.

NOTE: OSHA regulations stipulate that the oxygen content must be at least 19% for room entry. However, people have variable tolerances due to individual differences and differences in physical condition.

Problem No. 20

CHEMICAL ENGINEERING TOPIC: Momentum Transfer: muid Mechanics

SAFETY AND HEALTH CONCEPT: Fire Protection

BACKGROUND: Although fires occur very infrequently in chemical operations, when they do occur, they can be very destructive. The damage caused by a fire can be kept to a minimum if a properly designed fire protection system is installed. There are many facets to a well-designed fire protection system, including the detection systems, the fire protection hardware, and the fire fighters. The fire protection systems may rely on several kinds of fire extinguishing and control agents, including dry chemicals, foams, carbon dioxide or halons, and water. Water is included as the major fire protection agent in almost all situations (there are a few cases where water may be incompatible with the equipment or materials being protected). Water is generally cheaper than other agents, and it is generally available in large quantities. The high heat capacity and high heat of vaporization of water also contribute to its suitability as a fire control agent. Water is also inert (with respect to most other materials and chemicals) and can be stored and delivered relatively easily. Thus, water is supplied to most plant locations for fire control.

There are some problems to be considered when using water as a fire control agent. One of the first things to consider is that the quantity of water used to fight a fire is usually many times the theoretical minimum needed for extinguishment or control. Thus, there is a large amount of runoff water that must be disposed of. At first thought, it would seem that the runoff water could just be allowed to flow through drains and be disposed of as though it were rain water. In some cases, that can be done, but in many cases the runoff water will be contaminated with chemicals that are either process chemicals spilled during the fire or products of pyrolysis or combustion during the fire. In those cases, the water may be so contaminated that it cannot simply be discharged to a storm drain or sewage treatment system. It must be collected and treated before discharge. In almost all cases, the rate at which water is applied for fire protection is much greater than the rate at which water will fall during a rainstorm. Thus, special drainage systems must be provided to assure that the runoff water can be disposed of properly.

The design of fire water distribution systems is performed in the same way as any set of hydraulic calculations, although variations in methodology are sometimes used to reduce the amount of work. In addition, the fire water system will usually have several branches and loops so that the flow calculations are more complicated than they would be for a single pipeline leading from one point to

42 SAFETY, HEALTH, AND LOSS PREVENITON IN CHEMICAL PROCESSES

another. The design calculations may also have to account for the fact that the flow may differ from time to time because a single fire water supply system will be used for an entire plant, and it is not likely that the entire system will be in operation at once.

PROBLEM: Water is supplied to a pump for distribution to a fire water spraying system. The water is taken from an outdoor reservoir 100 m from the pump inlet, and the water surface is 5 m above the inlet to the pump. The pump is to deliver water to a water spray system on top of several liquefied gas storage tanks 500 m from the pump, at a discharge elevation of 35 m. The pressure at the nozzle manifold must be at least 700 kPa and water is to be delivered to the nozzle manifold for the spraying system at a rate of 900 m3/hr. The net positive suction head of the pump is 10 m. The pipeline from the pump to the nozzle manifold is 10-in. schedule 40 steel pipe. Pressure losses caused by fittings and valves upstream of the pump are equivaIent to 15 m of piping, and pressure losses caused by fittings and valves downstream of the pump are equivalent to 50 m of piping.

a. What is the minimum diameter for the supply piping? b. What is the power required to drive the pump? Assume the pump has 80%

efficiency and give your answer in kilowatts. c. For what outlet pressure should the pump be designed?

Problem No. 21

CHEMICAL, ENGINEERING TOPIC: Momentum Transfer: Fluid Mechanics

SAFETY AND HEALTH CONCEPT. Fire Protection

BACKGROUND: While very large fires occur at chemical plants occasionally, most of the fires that occur are smaller and present less danger to the operators. In most cases, the plant will have either a fire brigade or the operators will receive some training in fire protection. Part of that training will usually include the use of manually applied water sprays. Although there are other fire extinguishing and control agents that can be used in chemical operations, water should always be available. The water can be used either for direct extinguishment of the fxe or for exposure protection. Exposure protection refers to the practice of spraying equipment near the fire so it will not be damaged, leading to failure, and thus spreading the fire. Sometimes exposure protection will be required for long periods, especially if a fire is large, in which case, it may not be possible to extinguish it, and it must simply be allowed to burn out as the fuel is exhausted. During the burning period, the surroundings must be protected to minimize fire losses.

For some fires, particularly those where the fuel for the fire is an ordinary solid material, water sprays are directed at the fuel, cooling it and extinguishing the fire. However, in many cases, such as those where the fuel will react with water, or for liquid fuels that do not mix with water, spraying the fuel with water will do more harm than good. Then, it may not be possible to extinguish the fire. If the fire cannot be extinguished or reduced in size through the direct application of water, the water can still be used to minimize the damage that might otherwise occur. Equipment and buildings near the fire can be sprayed with water to cool them and prevent damage by the fire. In addition, water sprays are frequently used to shield firefighters or emergency operations crews during fire fighting and emergency operations.

Water may be applied either through fmed systems or through portable systems. Fixed systems are made of permanently mounted piping and nozzles and do not require that a firefighter remain in the immediate area. They are frequently provided in critical areas where access is limited or where firefighters cannot gain access. Manual fire fighting using water from hoselines is a traditional practice for locations where smaller quantities of water are needed, where there are enough trained personnel to man the hoses, and by fire departments or mutual aid groups that assist in fire protection. Hoselines are made from fabric reinforced material that can withstand high pressures. Many plants use standard 12/2-in. or 2b5-in.


hoses with nozzles. These serve as the first line of defense because they can be put into operation relatively quickly if personnel are available and well trained.

There are two problems associated with the design and use of fire water spray systems. One is to design the nozzle and piping system so that the required rate of water can be delivered. Such calculations can be made through the use of standard orifice and nozzle equations. The second is to determine the forces that must be resisted during discharge. The fire fighting nozzle will have a substantial reaction force, and the systemmust be designed to withstand the force. Permanent supports and reinforcing structures must be provided for fixed systems, and portable systems must be designed with the idea in mind that they will have to be held by fire fighters or be provided with temporary supports.

PROBLEM: The nozzle on a fire hose is designed to minimize friction losses. It can be considered to be an orifice with a discharge coefficient of 0.97 if the nozzle is designed to deliver a solid stream of water. Nozzle flow rates are frequently calculated by

where q is the flow rate through the n o d e in gallons per minute; d is the nozzle bore diameter in inches; and P is the pressure at the nozzle, in psig.

a. Starting with the basic equations for flow through an orifice, show that the nozzle flow rate equation is applicable and the dimensional constant 28.95 is correct. What assumptions are inherent in the equation? What are the dimensions of the constant? b. Find the bore diameter required for delivery of 100 gaVmin at a nozzle pressure of 100 psig. c. Calculate the reaction force on the nozzle. Would you judge it to be tolerable for a single, unaided firefighter?

Problem No. 22

CHEMICAL ENGINEERING TOPIC: Momentum Transfer: Fluid Mechanics

SAFETY AND HEALTH CONCEPT: Storage and Handling: Fluids

BACKGROUND: Large quantities of liquefied petroleum gas (LPG) are used each year, primarily as a fuel. The use as a fuel is seasonal, with much more being used during the winter than during the summer. However, the rate of production is more uniform, so supplies are increased during the summer and reduced during the winter. One method of storing the LPG is to inject it into a cavern in a salt dome. In some locations, primarily in Texas and Louisiana, there are large natural salt deposits. These deposits are broad and deep, and if a well is drilled into them and water is injected, some of the salt will be dissolved. If injection of fresh water is continued, a large cavern will be formed in the salt dome. (The salt "mined" by this method is usually used for the production of chlorine.) The caverns may have a volume of a million or more barrels (a barrel, as used in the petroleum industry, is equal to 42 gallons). Because of the shape of the salt domes and the method of solution mining used to form the caverns, the caverns are irregular in shape, but tend to have a much smaller diameter than length. Figure 1 shows an idealized sketch of a cavern in a salt dome. It appears, ideally, as a long vertical cylinder.

The tubing entering the cavern is arranged so that the cavern can be kept full at all times. If the cavern is not kept full of liquid or gas under pressure, it may collapse. In the case of propane storage, the pressure in the cavern is high enough for the propane to be kept as a liquid under the ambient temperature of the cavern. At least two streams must be able to enter and/or leave the cavern. That is usually accomplished by providing a concentric pipe system, with a large outer pipe and a smaller inner pipe. Normal operation for storage of LPG begins with the cavern full of saturated brine following the formation of the cavern. LPG is then pumped into the cavern through the annular space between the inner and outer pipes. The brine in the cavern is forced out the inner pipe. The inner pipe extends nearly to the bottom of the cavern and the outer pipe ends near the top. Thus, when the LPG is pumped into the cavern, it enters at the top. The LPG has a lower density than the brine, so it floats on top of the brine in the cavern. The liquid levels are monitored carefully to assure that LPG is not forced out of the cavern during filling. The brine forced from the cavern is transferred to a large storage pit, where it remains until the LPG is needed.

When LPG is needed, it is forced from the cavern by pumping brine back into the cavern. The brine enters through the inner pipe near the bottom of the cavern, and the LPG is forced out of the cavern through the annulus between the two pipes.

46 SAFEI17, HEALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

PROPANE ??m

Figure 1. Schematic of salt dome storage well.

Again, the process is carefully monitored to keep brine from being displaced through the LPG piping.

There are a number of potential hazards associated with salt dome storage. They are similar to those in other industries that handle flammable liquids and gases. AII involve the flammability of the LPG (or other products) that are stored in the

PROBLEM NO. 22 47

cavern. Special precautions should be taken to assure that none of the product is spilled from the piping. Care must also be taken to assure there is no leakage from the cavern, either through the cavern walls or the piping that penetrates the salt dome to the cavern. Automatic shutdown systems should be installed to isolate the cavern and its contents in the event of a failure of the surface piping. Special rules govern the period of operation before the well must be taken out of service and tested. The piping is subject to high salt concentrations, and corrosion is always a problem. Gas detectors and fire detection systems should be provided to monitor for hazardous conditions. Personnel training must be kept current, and emergency response training should be practiced periodically. Other potential safety problems are encountered as well, but accessibility and low cost make salt domes an excellent location for storage of hydrocarbons. In fact, the Strategic Petroleum Reserve is designed to utilize underground storage in caverns mined in salt domes. Safety considerations play a large part in its design and operation.

PROBLEM: A cavern in a salt dome is used for storage of LPG. The LPG has the properties of propane. The top of the cavern is 780 ft below the ground surface, and the outer pipe extends 800 ft below ground level. The end of the inner pipe is 2300 ft below ground level. The cavern has an average diameter of 50 ft. All piping is Schedule 160 steel; the inner pipe is nominal 8-in. diameter, and the outer pipe is nominal 14-in. diameter. Corrosion has increased the surface roughness of the pipe to 0.002 ft. LPG is to be injected and produced from the cavern at rates consistent with production and demand, respectively. The brine may be considered to be saturated with salt. At the conditions of the cavern, the brine has a specific gravity of 1.194 and a viscosity of 2 cp. The LPG has a specific gravity of 0.49 and a viscosity of 0.1 cp. The temperature rnay be assumed to be constant at 80°F

1. If the cavern is full, with the propane level just at the end of the inner pipe, and no transfer is taking place, what will the pressure in the propane line be at the well head (ground level)?

2. If propane is being injected into the well at a rate of 70,000 bbl/day, what well-head pressure will be required at the time the propane just reaches the level of the inner pipe?

3. If the propane pump is 85% efficient, how many horsepower will be required for the pump used to fill the cavern?

Problem No. 23


SAFETY AND HEALTH CONCEPT: Storage, Handling, and Transport

BACKGROUND: Many different chemicals are used in modern chemical processing operations. Some of the chemicals normally do not present a hazard to humans in their chemical effects (air and water, for example). Others are flammable, or toxic, or both, and can create hazards for humans and the environment if they are not kept under careful control. Any material used within the plant must be moved to the point where it is needed, used for a chemical process or unit operation, and the product or used material moved away for storage, continued processing, ultimate use, or disposal. In most plants, the most convenient way to transport materials from one place to another is through piping. Even solids may be transported through piping as fluidized gas-solid mixtures or as liquid-solid slurries. Fluid transport is preferred because the materials can be prevented from release to the atmosphere, they can be handled by machinery that is readily controlled, and the pumps (or compressors) used for moving the fluid are relatively simple in design and economical in operation. If fluids are moved through piping, the pressure in the piping must be high enough to keep liquids in the liquid phase, to provide sufficient pressure drop to move the fluid through the piping, and to keep the fluid at the pressure required for the process. Thus, there is always a chance that piping may rupture and a leak may occur.

Piping is designed and constructed to various codes that specify the maximum allowable pressure for the piping. In larger piping or in cases where extremely hazardous materials are transported, the codes may require that all the piping be welded. For less hazardous materials, the piping may be be assembled with flanged connections or screwed connections. The piping must also be inspected and maintained to assure that there is no gradual deterioration caused by corrosion, by weathering and slumping of supports, or by damage from an accident. If piping is damaged, the portion of the plant involved must be shut down and the piping repaired.

It is important to be able to estimate how much material may be lost if there is a leak in the piping in order to plan for proper emergency operations. Both the leak rate and the total amount leaked are important in forecasting the consequences of the leak.

PROBLEM: Benzene is being transferred through long, small-diameter plant piping under a pressure of 100 psig. At 1:00 PM the operator notices a drop in

PROBLEM NO. 23 49

pressure in the piping. He immediately restores the pressure to 100 psig and sends another operator to trace the piping to see if the cause of the pressure drop can be determined. At 230 PM a 0.25-in. diameter hole is discovered in the piping. The flow of benzene is stopped; the line is depressured; the section of piping containing the hole is isolated; and the piping is drained, purged, repaired, and returned to service. Estimate how many gallons of benzene are spilled before the flow is stopped.

(This problem is based on a problem in the text Chemical Process Safety: Fundamentals and Applications, by D. A. Crow1 and J. F. Louvar, published by Prentice Hall, Englewood Cliffs, NJ.)

Problem No. 24


SAFETY AND HEALTH CONCEPT: Storing, Handling, and Transport

BACKGROUND: All chemical manufacturing operations require the storage and movement of process materials. Sometimes the quantity of material stored is small, but sometimes it is quite large. Refineries may store millions of gallons of hydrocarbons, for example, whiie a company that manufactures semiconductor devices will have relatively little inventory of process materials. In some cases the materials stored normally do not present a hazard to humans, at least in the sense that they are neither flammable nor toxic. (Sometimes, seemingly innocuous materials can create a hazard. Many years ago, a large molasses tank failed in Boston, drowning a number of people who had been walking in the street.) In many cases the materials stored inprocessingplants are either toxic, or flammable, or both. Special precautions are then required to assure there is no release of such materials so that damage to the environment or injury to people will be prevented.

Large leaks from storage tanks are very infrequent. Smaller leaks occur at higher frequency; they may be caused by corrosion, in which case timely inspection and regular maintenance of the tanks can prevent them in most cases. Larger leaks are more likely to be caused by some outside event such as puncturing the tank with maintenance equipment or damage from severe storms. Some of these larger leaks can be prevented by proper procedures and by better design.

Tanks are constructed to codes of standards that experience has shown to provide safe operation. The tanks are tested using a variety of methods such as dye penetrants, x-raying, vacuum boxes, and magnetic techniques. Before the tank is filled with a toxic or flammable process fluid, it is usually hydrostatically tested at a pressure equal to or greater than its design pressure. All of these testing methods have the same goal: to assure that the tank will not fail in service. In addition, relief devices are provided on the tank to prevent hazardous overpressure or underpressure.

In many cases in which a tank is to contain a flammable material, if air is allowed to enter the tank while a flammable liquid is stored in the tank, a flammable concentration of vapor can form in the air above the liquid. In such cases, an inert gas, usually nitrogen, is used to fill the vapor space. This practice, called padding the tank, is used to prevent flammable mixtures from occurring. Thus, the possibility of ignition inside the tank is eliminated, and the volatile liquid can be stored more safely.

PROBLEM NO. 24 51

By following strict design procedures, the relevant codes, and good maintenance practices, most tank leaks can be prevented. The following problem is an illustra- tion of what can happen if a tank is involved in an accident. Proper procedures and training can prevent such accidents, and the engineer has the responsibility not only of designing the tank for its intended service, but also of making sure the tank is tested and maintained.

Written permits must be issued if work is to be done either on the tank or in the immediate vicinity so safe practices will be followed.

PROBLEM: A cylindrical pressure vessel 20 ft high and 8 ft in diameter is used to store benzene. The tank is padded with nitrogen to a constant, regulated pressure of 1 atm gauge to prevent air entering and a subsequent explosion. The liquid level within the tank is presently at 17 ft. A 1-in. puncture occurs in the tank 5 ft off the ground due to an accident.

Estimate: (a) the number of gallons of benzene spilled from the tank, (b) the time required for the benzene to leak out, and (c) the maximum flow rate of benzene through the leak. Assume that the nitrogen pressure remains at 1 atm gauge as long as benzene flows from the tank.


Problem No. 25

CHEMICAL ENGINEERING TOPIC: Fluid Mechanics, Choked Flow


BACKGROUND: One of the very critical areas of protecting workers from dangerous conditions is that of assuring that the air the workers breathe is in fact suitable for breathing. Much attention must be given to minimizing the exposure to harmful concentrations of toxic vapors, but sometimes it is a matter of assuring an adequate level of oxygen in the breathing air.

The U.S. Occupational Safety and Health Administration (OSHA) is an organization charged with responsibility for assuring safety in the workplace. One of the concerns is that of providing that no one will enter a room, or other enclosure where the oxygen concentration is too low. Normal air is about 21% oxygen (mole or volume percent), with the remainder being primarily nitrogen. OSHA regulations require that the air be at least 19.5% oxygen for entry into an enclosure. Although people have a variable tolerance for lower than normal oxygen, this level is believed to be safe for most people in reasonably good health, and some would be able to tolerate a lower level, perhaps.

If the oxygen content of the breathing air were to drop to 16% or below, many people would lose consciousness and eventually suffer great injury or death because they would become incapacitated and not be able to escape without assistance. A complication in this situation would be that the person would not realize the difficulty until it became too late.

Although nitrogen is not toxic, it nevertheless can displace oxygen from the air in a room or other enclosure, so that the air is no longer suitable for breathing. The following problem addresses such a concern.

PROBLEM: A large tank of nitrogen at a pressure of 200 psig and 80°F has developed a leak which is equivalent to a 0.1-in.-diameter hole. Estimate the rate that nitrogen will leak from this hole, and at this rate, how long would it take to reduce the level of oxygen in a room to an average concentration less than 19.5%. The room is a small control room, 14 ft by 10 ft by 8 ft high. Note, to provide a result that is conservatively safe, assume that the nitrogen enters the room and displaces the air by plug flow, that is, the displaced air will be 21% oxygen. The barometric pressure may be taken as 1 atm, and the temperature of the room is 80°F. Assume the discharge coefficient is 1.0.


Problem No. 26


SAFETY AND HEALTH CONCEPT. Rupture Disks and Relief Valves

BACKGROUND: Some method of pressure relief is required on all pressure vessels and for other process equipment where increasing pressure might rupture the vessel. Much of the piping used in modern chemical operations also requires overpressure protection. Either relief valves or rupture disks may be used for pressure relief. In many cases, either a rupture disk or a relief valve can be used; usually, one or the other is preferred. Relief valves are usually used for process protection, and rupture disks are used for vessel protection. The relief valve or rupture disk must be designed so it will operate at a known pressure and prevent the pressure within the system from increasing. Thus, the flow rate the valve can handle is a major concern in its design.

Sometimes the maximum flow rate the relief system must handle is based on process flow conditions, such as the maximum flow rate a pump can deliver at the relief pressure or the maximum flow rate for a compressor at the relief pressure. In other cases, the relief system maybe designed to relieve the pressure in a reactor if the reaction gets out of control. Another important consideration is the protection of vessels from overpressure during a fire. In most vessels, the contents will be either gases or liquids. In either case, a rise in temperature will cause a rise in pressure in the vessel, either through expansion of the gas or through increase of vapor pressure of the liquid. In such cases, the relief system is designed to relieve the pressure at a rate determined by the heat transfer rate to the vessel.

Relief valves and rupture disks have similar purposes, but their designs are different. A relief valve has a spring-loaded valve stem. Rather than turning a handle or using a control system to open or close the valve, the force exerted by the pressure inside the valve is resisted by the force of a spring. When the pressure increases to the set point of the valve, the spring can no longer resist the force caused by the pressure and the valve begins to open. At a slightly higher pressure, the relief valve will be fully open and allow maximum flow. As the contents of the vessel are released, the pressure in the vessel will begin to decrease. Once the pressure decreases to a level a little below the set point, the spring will be strong enough to reset the valve; the pressure will no longer decrease. A relief valve may open and close many times during a prolonged incident in which the pressure rises, and the vessel is then vented to relieve the pressure. Relief valves must be maintained carefully to ensure their proper function. Their calibration must be checked periodically, and they must be kept clean. When improper maintenance is

54 SAFETY, HEALTH, AND LOSS PREVENTION IN CHEMICAL. PROCESSES

used, a relief valve may not fully reseat when the pressure is reduced. The valve will continue to leak until it is repaired. Since the seat of the relief valve may be in contact with the material in the vessel, some corrosion may occur. Leaking and corrosion may cause failure: either the valve may open at lower pressures or it may fail to open when the pressure increases. Either result can be dangerous and should be avoided by careful inspection and maintenance. If properly sized, installed, and maintained, relief valves have been shown to be reliable and to reduce the prob- abiity of damage caused by overpressure.

Relief valves may also be designed to fail at a given temperature. Sometimes a combination relief valve will be used that will operate at either a given temperature or a given pressure. Home water heater relief valves are a combiiation temperature-pressure relief valve, for example. The design of a relief valve depends on the material to be vented. If a gas or liquid alone is to be vented, the design is relatively simple, and the relief system can be designed on the basis of single phase fluid flow. However, in many systems containing liquids under their own vapor pressure, venting can be a combination of liquid and vapor. In such cases, the two-phase fluid that is vented must be accounted for. The methods used to design such relief systems are substantially different than those for single phase fluids.

PROBLEM: A tank containing benzene is to be relieved if the pressure in the tank reaches 15 psig. The tank is cylindrical with hemispherical heads. The overall tank length is 60 ft and it is 12 ft in diameter. The relief valve is to be designed so that the tank can be vented if a fire occurs. In order to keep the tank from being overfilled, procedures are put in force to limit the liquid depth in the tank to 11 ft. The venting rate is to be consistent with the heat transfer rate given by

as given in National Fire Protection Association Standard 30, "Flammable Liquids Code." In the heating rate equation, q is the net heating rate to the contents of the tank, in Btuthr, and A is the area of the tank wetted by the liquid portion of the tank contents, in ft2. Only the given units may be used. Calculate the nominal pipe size of the relief valve, assuming (a) only liquid flows through the valve and (b) only vapor flows through the valve. The friction loss coefficient K for the relief valve is 3.0. You may assume the vapor to be an ideal gas for flow calculations.

Problem No. 27

CHEMICAL ENGINEERING TOPIC: fluid Mechanics

SAFETY AND HEALTH CONCEPT: Rupture Disks and Relief Valves

BACKGROUND: Some method of pressure relief is required on all pressure vessels and for other process equipment where increasing pressure might rupture the vessel. Much of the piping used in modern chemical operations also requires overpressure protection. Either relief valves or rupture disks may be used for pressure relief. In many cases, either a rupture disk or a relief valve can be used; usually one or the other is preferred. Relief valves are more frequently used for process protection, and rupture disks are more frequently used for vessel protection. The relief valve or rupture disk must be designed so it will operate at a known pressure and prevent the pressure within the system from increasing. Thus, the flow rate the valve can handle is a major concern in its design.

Sometimes the maximum flow rate the relief system must handle is based on process flow conditions, such as the maximum flow rate a pump can deliver at the relief pressure or the maximum flow rate for a compressor at the relief pressure. In other cases, the relief system maybe designed to relieve the pressure in a reactor if the reaction gets out of control. Another important considei-ation is the protection of vessels from overpressure during a fire. In most vessels, the contents will be either gases or liquids. In either case, a rise in temperature will cause a rise in pressure in the vessel, either through expansion of the gas or through increase of vapor pressure of the liquid. In such cases, the relief system is designed to relieve the pressure at a rate determined by the heat transfer rate to the vessel.

Relief valves and rupture disks have similar purposes, but their designs are quite different. A rupture disk is a simple device that consists essentially of a ihin disk of material held in place between two flanges. The disk is usually made of metal, although it may be made of other materials. The choice of material is important because the rupture disk must be designed to close tolerances in order to operate properly. In use, the disk ruptures when the pressure level rises to a chosen level. The vessel is then vented and the pressure in the vessel eventually drops to atmospheric pressure. The rupture disk is chosen to be large enough to vent the vessel at the maximum rate required.

Relief devices may also be designed to fail at a given temperature. Sometimes a combination relief device will be used that will fail at either a given temperature or a given pressure. Home water heater relief valves are a combination temperature-pressure relief valve, for example.


The design of a rupture disk depends on the material to be vented. If a gas or liquid alone is to be vented, the design is relatively simple, and the relief system can be designed on the basis of single phase fluid flow. However, in many systems containing liquids under their own vapor pressure, venting can be a combination of liquid and vapor. In such cases, the two-phase fluid that is vented must be accounted for. The methods used to design such relief systems are substantially different than those for single phase fluids.

PROBLEM: A tank containing benzene is to be relieved if the pressure in the tank reaches 15 psig. The tank is cylindrical with hemispherical heads. The overall tank length is 60 ft and it is 12 ft in diameter. The rupture disk is to be designed so that the tank can be vented if a fire occurs. In order to keep the tank from being overfilled, procedures are put in force to limit the liquid depth in the tank to 11 ft. The venting rate is to be consistent with the heat transfer rate given by

as given in National Fire Protection Association Standard 30, "Flammable Liquids Code." In the heating rate equation, q is the net heating rate to the contents of the tank, in Btuthr andA is the area of the tank wetted by the liquid portion of the tank contents, in ft2. Only the given units may be used. Calculate the diameter of the rupture disk required, assuming (a) only liquid flows through the orifice and (b) only vapor flows through the orifice. You may assume the orifice coefficient to be that of a sharp-edged orifice, and the vapor to behave as an ideal gas.

Problem No. 28

CHEMICAL ENGINEERING TOPIC: Fluid Mechanics, Design

SAFETY AND HEALTH CONCEPT. Rupture Disks and Relief Valves

BACKGROUND: Runaway chemical reactions may occur for a variety of exothermic reactions. Figure 1 shows a typical temperature-time curve for an exothermic reaction. The reactor-reaction combination has a stable range where process controls keep the reaction under control; an unstable range where the reaction may be brought back under control by cooling, inhibiting, and quenching; and a runaway range where the reaction is out of control and the reactor must be vented. Normally, operation is typically kept in the stable range where the controls and the process operate as designed. However, the reaction could get out of control for reasons such as too much initiator, loss of cooling, or loss of mixing. Typical behavior would then be for the reactor temperature to gradually rise at first, then accelerate to the point that a runaway occurs. During the unstable range the reaction might be restabilized by methods such as emergency cooling or inhibiting.

I Condition: Stable I Unstable I Runaway

I 1

Deaignt I I TIME I

Process I Restabilize I Venting Controls Emergency (determination of

I Cooling I vent size & settings, ! Inhibitor ! mechanical forces,

Quenching emission for specific (add water) reaction/reactor

systems)

Figure 1. Venting and restabilization concepts for chemical exothermic runaways.

57


However, at the point of runaway, the onlyrecourse is to vent thereactor to prevent reactor failure. This emergency venting is frequently confused with explosion venting which is used for vapor-air and dust-air deflagrations. They differ because emergency venting is for venting of a reactor whereas explosion venting is for venting an explosion that occurs inside a building.

Even though we take a number of measures to reduce the probability of a runaway, we almost always have to provide some form of protection for the reaction vessel in the event of a runaway; the only way to protect the vessel is emergency venting. Emergency venting design relates to several process and equipment considerations. For a given vessel and its vent design, we must limit the maximum quantity of a "runaway" chemical, so that the maximum venting pressure is limited to a certain acceptable level. Many users limit it to 110% of the vessel's maximum allowable working pressure (MAWP).

The Design Institute for Emergency Relief Systems (DIERS) Users Group, which is an affiliate of the American Institute of Chemical Engineers, has developed some methodology to design emergency relief systems.' The DIERS study was very extensive and complicated. It involved significant developments and applications of complex theories and experiments. Some aspects were reaction kinetics under runaway conditions and multiphase critical flashing flow for viscous and nonviscous systems. A number of DIERS users have attempted to simplify the DIERS technology.273P The following two venting analyses represent the two extreme cases, tempered aad gassy reactions. Tempered reactions are reactions that have energy removed due to significant vaporization of the liquids. Here the heat of vaporization cooling during vapor or two-phase flow venting is sufficient to temper the reaction. An equation representing the relief behavior for a vent length LID < 400 is3

where Mo = allowable mass of the reactor mixture charge (kg) to limit the venting

overpressure to Pp (psig) Dp = rupture disk diameter, inches Ps = the allowable venting overpressure (psi), i.e., the maximum venting

pressure minus the relief device set pressure Pp = maximum venting pressure (psig) Ps = the relief device set pressure (psig). Note that the relief device set

pressure can range from the vessel's MAWP to significantly below the MAWP.

Ts = the equilibrium temperature corresponding to the vapor pressure where the vapor pressure is the relief device set pressure (K)

PROBLEM NO. 28 59

dTldts = the reactor mixture self-heat rate ("Clmin) at temperature Ts (K) as determined by a DIERS or equivalent test

C p = specific heat of the reactor mixture (cdg-K or Btu/lb-"F)

Note that the equation given above is a dimensional equation and the dimensions given in the nomenclature must be used.

Gassy reactions are reactions that are not tempered because significant noncondensible gas is formed, and the heat of vaporization during vapor or two-phase flow venting is insufficient to temper the reaction at any point. Hence, to limit the overpressure at a specific maximum pressure, the venting rate must equal the peak volumetric generation rate. An equation representing the relief behavior for a length LID < 400 is3

where Vp = vessel total volume (gal) Pp = maximum allowable venting pressure (psia) Ms = sample mass used in a DIERS test or equivalent test, g

dPldt = pressure rise in test (psilsec) TT = maximum temperature in test (K)

up = Pp - Pamb, psi Pamb = ambient pressure at the end of the vent line, psia This is also a dimensional equation and the dimensions given in the nomencla-

ture must be used.

PROBLEM

Part 1 -Tempered Reaction: A 750-gal reactor containing a styrene mixture has an 8-in. rupture disk and a vent line with equivalent length LID = 400. The vessel MAWP is 100 psig and the rupture disk set pressure is 15 psig. The styrene mixture had a self heat rate of 50°C/min at 160°C as it tempered in a DIERS venting test. What is the allowable reactor mixture charge to limit the overpressure to 10% over the set pressure?

Part 2 -Gassy Reaction: A nominal 750-gal reactor with a net volume of 880 gal containing tetrazole mixture has an 8-in. rupture disk and a vent line with an equivalent length LID = 400. The vessel MAWP = 100 psig and the rupture disk set pressure is 15 psig. The DIERS venting test showed that the reaction was "gassy". The test mass was 25 g, the peak rate of pressure rise was 500 psilmin and the maximum test temperature was 250°C. What is the allowable reactor mixture charge to limit the overpressure to 10% of the MAWP?


REFERENCES

1. H. G. Fisher. DIERS Research Program on Emergency Relief Systems. Chem Engr Pro& 81(8), 33-36 (August 198.5).

2. H. K Fauske, G. H. Clare, and M. J. Creed. Laboratory Tool for Characterizing Chemical Systems. Proceedings of the International Symposium on Runaway Reactions, Cambridge, MA, March 7-9,1989. Center for Chemical Process SafetyIAIChE, New York, 1989, pp. 372-394.

3. J. A. Noronha, R J. Seyler, and A. J. Torres. Simplified Chemical and Equipment Screening for Emergency Venting Safety Reviews Based on the DIERS Technology. Proceedings of the Interna- tional Symposium on Runaway Reactions, Cambridge, MA, March 7-9,1989. Center for Chemical Process SafetylAIChE, New York, 1989, pp. 660-680.

4. D. P. Mason. Highlights of FM Inspection Guidelines on Emergency Relief Systems. Proceedings of the International Symposium on Runaway Reactions, Cambridge, MA, March 7-9,1989. Center for Chemical Process SafetyIAIChE, New York, 1989, pp. 722-750.

5. H. K Fauske. A Quick Approach to Reactor Vent Sizing. PlanVOperations Progress, 3(3), 145-146 (1984).

6. Chemical Engineering Progress, 81(8), 33-62 (198.5).

(This problem was provided by Mr. John Noronha of Eastman Kodak Company.)

Problem No. 29

CHEMICAL ENGINEERING TOPIC: Fluid Mechanics; Fundamentals


BACKGROUND: Many of the common materials used in chemical processing are toxic or flammable; frequently they are both toxic and flammable. Special care must be taken when such materials are used. Consider flammable materials for example. If a flammable liquid vaporizes, or if the material is normally a gas, fuel and air may mix and result in a flammable mixture. Two concentrations of fuel in air are important: the upper flammable limit (UFL) and the lower flammable limit (LFL). The LFL is the smallest concentration of fuel in air that can be ignited. At lower concentrations, there is too little fuel to ignite. If the concentration is above the UFL, there is insufficient oxygen in the air for the fuel to ignite. Only concentrations of fuel in air between the LFL and the UFL will ignite. The flammability limits depend on the temperature, so if the fuel-air mixture is heated, the flammability limits widen. At temperatures reachedin flames, virtually all the fuel will be burned if enough oxygen is present. Pressure also affects the flammability limits for fuel-air mixtures, but the effect is not important in most ignition problems because most fires burn near atmospheric pressure. Data for the LFL and UFL of various chemicals can be found in Sax's Dangerous Properties of Industrial Materials; NFPA 325M, "Properties of Flammable Liquids"; and the NIOSH "Pocket Guide to Chemical Hazards."

One method of preventing damage or injury caused by ignition of fuel-air mixtures is to keep the mixtures from forming. That may be accomplished by keeping all fuels from escaping into the air or by ventilating the location with enough air to keep the concentration below the LFL. Usually, prevention is preferable to control, so the goal is to prevent escape of flammable materials. If accidental escape occurs, control must be used to reduce the concentration or to keep the concentration from exceeding the LFL. The most frequent method of control is ventilation. Ventilation is important enough that National Fire Protec- tion Association Standard 69, "Explosion Prevention Systems," describes veniila- tion systems as a method of preventing explosions. NFPA 69 (1986 Edition) describes equations for calculating the quantity of ventilating air required for reducing the concentration of a fuel-air mixture or for preventing a flammable mixture from forming. The methods used in the calculations are essentially material balances, and they are limited in application to the conditions discussed. Other material balances can be written and solved under other conditions. One such different set of conditions is considered in the following problem.


PROBLEM: A natural gas compressor station has several large compressors used to boost the pressure in a cross country pipeline. The capacity of the pipeline is 500 million standard cubic feet (measured at 1.0 atm and 60°F) per day. The compressors are housed in a building 200 ft long, 80 ft wide, and 30 ft high at the eaves. The building has a sloped roof, and if the volume of the compressors and ancillary equipment is considered, thenet free air volume of the building is 510,000 ft3. The temperature in the building averages about 90°F, and the pressure about 1.0 atm. The building has a ventilation system designed to operate at two speeds; the low speed provides a ventilation rate equal to 6 air changes per hour and the high speed provides 20 air changes per hour. The ventilation rate is based on entering air, measured at building temperature and pressure. The building is normally ventilated at the low rate, but a gas-sensing system monitors the air in the -

building, and if gas is released in the building and reaches a concentration of 25% of the LFL, the ventilation system automatically increases to the high rate.

1. Assume there is aleak of 10,000 standard cubic feet of natural gas (considered to be pure methane) per minute in the building. How long will it take for an alarm to sound if the alarm sounds when the ventilation system increases to the higher speed at 25% of the LFL concentration?

2. What is the maximum concentration reached if the leak continues indefinitely? Ventilation will be at the higher speed.

3. If the gas leak is shut off when the concentration reaches 25% of the LFL, how long will it be until the concentration has dropped to 5% of the LFL if (a) ventilation is at the low rate and (b) ventilation is at the high rate?

4. If the leak detection alarm is given at 25% of the LFL, how long will it be before the concentration reaches the LFL? Ventilation will be at the higher speed as long as the gas concentration is more than 25% of the LFL. Is that sufficient time for the source of gas to be found and shut down in time to prevent the concentration from reaching the LFL?

5. Calculate the energy released if the natural gas, which has a heat of combustion of 1044 Btdstandard cubic ft, is ignited and completely burned when the concentration of gas is at the LFL. Compare that energy to the equivalent mass of TNT, if the explosive energy of TNT is 2000 Btullb.

In your calculations, you may assume the air and gas in the building to be completely mixed and the air-gas mixture leaving the building to be the same composition as the mixture in the building. Ideal gas behavior may be assumed to apply because the pressure is quite low. Note that there is always a flammable mixture somewhere because there must be time for the pure gas to mix with air. The assumption of complete mixing is made to simplify the problem. The results can be used for designing the ventilation system, but it must be kept in mind that mixing is not truly instantaneous.

Problem No. 30

CHEMICAL ENGINEERING TOPIC: Fluid Mechanics, Fundamentals, Ther- modynamics

SAFETY AND HEALTH CONCEPT: Toxic Exposure Control

BACKGROUND: Many of the materials used in industry are either toxic or flammable, and some are both toxic and flammable. Toxic materials may cause immediate health problems such as poisoning, and others may cause delayed health problems such as cancer. The term "acute exposure" is used to describe a condition where a single exposure to a material, usually at relatively high concentration, causes immediate health effects. Exposures that occur over a long period and are usually repeated periodically, eventually causing health effects at relatively low concentrations, are called "chronic exposures." It is difficult to measure the effects of either acute or chronic exposure because we cannot deliberately expose humans to the effects of toxic materials. Thus, our knowledge comes from either animal experiments or from accidental exposures of humans. In either case, the ap- plicability of the information can be questioned for use in describing human health effects in many cases.

Even though our information may be imperfect, we must strive for a safe workplace for workers and a safe environment for neighbors. Therefore, various groups have instituted exposure standards for most of the materials used in industry. The Occupational Safety and Health Administration (OSHA) has established a permissible exposure limit (PEL) for many materials. The PEL is a concentration to which a worker may be exposed on a continuing basis. The PEL is a time-weighted average and is based on an 8-hour daily (40 hour weekly) exposure over the worker's lifetime. Another higher limit is the immediately dangerous to life and health (IDLH) concentration. The IDLH concentration represents the concentration level from which one could escape within 30 minutes without experiencing any escape-impairing or irreversible health effects. The PEL concentration should be listed on the material safety data sheet (MSDS), which should be reviewed by all workers who might be in the vicinity of the material. There is also a concentration called the Emergency Response Planning Guide (ERPG) concentration, provided by the American Industrial Hygienists Association. It can be useful in planning for emergency operations in a plant.

Some of the most serious chances of exposure occur if a hazardous material escapes into the atmosphere. When the toxic material mixes with the atmosphere, it may expose workers to its toxic effect. In most cases where the materials are extremely toxic, the acceptable exposure level can only be attained by preventing

64 SAFTXY, HEALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

the release. However, if an accidental release occurs, ventilation is usually the easiest method for preventing a buildup of the hazardous material inside a building.

Several references may be consulted to find the allowable exposure levels of materials. The National Institute for Occupational Safety and Health (NIOSH) has published a substantial body of information, and OSHA is the source for legally mandated exposure levels. The NIOSH "Pocket Guide to Chemical Hazards" is a compact listing of PEL and IDLH values as well as a substantial amount of other information. It is a condensed summary of information taken in part from the NIOSHIOSHA Occupational Health Guidelines for Chemical Hazards. Sax's Dangerous Properties of Industrial Materials is another source of information. The OSHA PEL values are published in Title 29 of the Code of Federal Regulations, Part 1910, Subpart 2, "General Industry Standards for Toxic and Hazardous Substances."

PROBLEM: Ammonia is used as a refrigerant. It is compressed outdoors, and the ammonia is then liquefied and circulated through the refrigerator's evaporator heat exchanger. The evaporator is indoors. The room containing the evaporator has a net open volume of 1000 m3 and a ventilation rate equal to two air changes per hour. A leak develops in an ammonia line that results in 50 g of ammonia being released per minute.

1. How long will it take for the ammonia concentration to increase to the PEL concentration?

2. How long will it take to reach the IDLH concentration? 3. If the flow of ammonia is stopped at the time the IDLH concentration is

reached, how long will it be before the concentration is reduced to the PEL concentration?

You may assume that the mixtures of gases are ideal at the low pressures involved. Also assume that the ammonia is mixed with the air in the room completely and instantaneously and that the air-ammonia mixture leaving the room has the same concentration as the mixture in the room.

Problem No. 31


HEALTH AND SAFETY CONCEPT: Storing, Handling, and Transport

BACKGROUND: There are many occasions when the thrust forces caused by high velocity, high rate flow will cause excessive forces on piping and associated equipment. Sometimes the design of relatively simple devices can be complicated by the possibiity of severe mechanical loads on piping equipment. If proper precautions are not taken to prevent failure due to these forces, then very serious accidents can occur.

Some examples of such situations might be in the design of safety relief systems wherein there exists the possibiity of suddenly initiated, very high velocity flow, with the consequent possibility of the discharge piping reacting with significant movement, as for example, after the manner of a garden hose that is not being held. Our experience tells us that the hose will move erratically about, discharging water in many directions. One should be aware that even heavy steel piping can behave similarly if it is not suitably constrained.

Another serious situation can develop from misuse of a rather common item that exists in laboratories, perhaps in laboratories where you have worked. This is the compressed gas cylinder. A typical situation might exist at your gas chromatograph, for example, where air is being used in conjunction with hydrogen, in the flame ionization detector. The air is usually supplied at a pressure in excess of 2000 lb/in2. If the cylinder is not properly restrained and held, it can be easily knocked over with possibly disastrous results if the valve is broken off in the fall.

The result of unexpectedly high thrust forces from flow may frequently be disastrous because of the rapid, violent, and unpredictable motion of a pipe, or as in the example above, a rather heavy gas cylinder. There is also the distinct possibiity of equipment failure from the forces which would result in the discharge of dangerous materials to the air.

In the problem below, you are asked to estimate some forces, the magnitude of which might easily cause equipment failure with the consequent loss of large quantities of a highly flammable material to the air. The result would almost certainly be a serious gas cloud explosion and fire.

PROBLEM: A tank ship hauling Liquefied Natural Gas (LNG) is being unloaded into a 600,000 bbl storage tank at a rate of 55,000 gaVmin through a 30-in. diameter, schedule 10 stainless steel pipe. The schematic diagram of the pipe inside the tank is shown on the next page. Determine the total upward force on the tank roof and the horizontal and vertical forces on the splash plate.

Problem No. 32

CHEMICAL ENGINEERING TOPIC: Fluid Mechanics; Thermodynamics


BACKGROUND: There are many occasions when the thrust forces caused by high velocity, high rate flow will cause excessive forces on piping and associated equipment. Sometimes the design of relatively simple devices can be complicated by the possibility of severe mechanical loads on piping equipment. If proper precautions are not taken to prevent failure due to these forces, then very serious accidents can occur.

Some examples of such situations might be in the design of safety relief systems wherein there exists the possibility of suddenly initiated, very high velocity flow, with the consequent possibility of the discharge piping reacting with significant movement, as for example, after the manner of agarden hose that is not being held. Our experience tells us that the hose will move erratically about, discharging water in many directions. One should be aware that even heavy steel piping can behave similarly if it is not suitably constrained.

The result of unexpectedly high thrust forces from flow may frequently be disastrous because of not only the rapid, violent, and unpredictable motion of a pipe, as in the example above, but there is also the distinct possibility of equipment failure from the forces which would result in the discharge of dangerous materials to the air.

Another serious situation can develop from misuse of a rather common item that exists in laboratories, perhaps in laboratories where you have worked. This is the compressed gas cylinder. A typical situation might exist at your gas chromatograph, for example, where air is being used in conjunction with hydrogen, in the flame ionization detector. The air is usually supplied at apressure in excess of 2000 lb~in.~. If the cylinder is not properly restrained and held, it can be easily knocked over with possibly disastrous results if the valve is broken off in the fall. In the problem below, you will have an opportunity to estimate some of the consequences of such an accident.

PROBLEM: A compressedgas cylinder contains air that is intended as the oxidizer for a hydrogen flame detector on your gas chromatograph. During the hook-up procedure, the assistant (not you!) removes the safety cap from the cylinder and begins to attach the pressure regulator when the cylinder slips from his grasp and the bottom slides across the newly waxed floor. As the cylinder falls, the valve strikes the table edge, and the valve is broken off, exposing an opening in the top

68 SAFEEY, HEALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

of the cylinder that is 5% in. in diameter. The air, which was originally at room temperature (75°F) and 2250 1b/ia2 will escape.

What initial flow rate will occur through the break? As a consequence of the flow, what force will be exerted on the cylinder? As a consequence of the force, what acceleration will be imparted to the cylinder? What velocity and what kinetic energy will the cylinder have when it gets to the laboratory wall which is 20 ft away?

The cylinder weighs 140 lb and has an internal volume of 1.5 ft3. To simplify the problem, you may assume that the air behaves as an ideal gas, and that the thrust remains constant as long as the critical pressure ratio is exceeded. The shape of the opening will not be like a sharp-edged orifice, but rather will be a converging nozzle. Assume that the discharge coefficient is 0.9.

Problem No. 33



BACKGROUND: There are many occasions when the thrust forces caused by high velocity, high rate flow will cause excessive forces on piping and associated equipment. Sometimes the design of relatively simple devices can be complicated by the possibility of severe mechanical loads on piping equipment. If proper precautions are not taken to prevent failure due to these forces, then very serious accidents can occur.

The result of unexpectedly high thrust forces from flow may frequently be disastrous because of the rapid, violent, and unpredictable motion of a pipe. A possible instance of the uncontrolled high velocity flow might be that which would occur with the rupture of a pipe carrying a high pressure fluid. There is also the distinct possibility of equipment failure from the forces which would result in the discharge of dangerous materials to the air.

A very serious situation can develop from misuse of a rather common item that exists in laboratories, perhaps in laboratories where you have worked. This is the compressedgas cylinder. A typical situation might exist at your gas chromatograph, for example, where air is being used in conjunction with hydrogen, in the flame ionization detector. The air is usually supplied at a pressure in excess of 2000 lb/in2. If the cylinder is not held upright, it can be easily knocked over with possibly disastrous results if the valve is broken off in the fall.

Another example of such situations might be in the operation of safety relief systems wherein there is the possibility of suddenly initiated, very high velocity flow, with the consequent possibility of the discharge piping reacting with significant movement, as for example, after the manner of a garden hose that is not being held. Our experience tells us that the hose will move erratically about, discharging water in many directions. One should be aware that even heavy steel piping can perform similarly if it is not suitably constrained. In the problem below, you will be given an opportunity to calculate the thrust force that might occur as a consequence of a release by a pressure relief system.

PROBLEM: A 4-in. schedule 40 steel pipe rises vertically from an elbow off of a horizontal pipe. The pipe is the discharge end of a pressure relief device from a reactor vessel. If the system vents, it will require a discharge velocity of 120 ft/sec of a flashing liquid-vapor mixture which has a mean density of 17 lb/ft3.

How much lateral force will be exerted on the end of the pipe at the elbow?

Problem No. 34

CHEMICAL ENGINEERING TOPIC: Design


BACKGROUND: In order for combustion to occur, there is a need for the simultaneous presence of oxygen (or air), an ignition source, and a combustible or flammable vapor or dust. The combinations often occur in industry. Flammable or combustible liquids do not cause combustion by themselves, but they can form vapors that can cause combustion. Similarly, large-size solids or wet dusts do not often support combustion, but fine, dry dusts do.

Deflagrations, unlike fires, are combustion phenomena associated with sudden pressure rises where the pressure wave moves at a speed less than the speed of sound. Detonations, also unlike fires, are combustion phenomena associated with sudden pressure rise where the pressure wave moves at a speed more than the speed of sound. It is often impossible to protect buildings or even a strong vessel against a detonation.

There are three methods of protecting avessel, once a deflagration occurs. They are explosion suppression, explosion venting, and deflagration pressure containment (DPC). In DPC, a vessel is built strong enough to withstand the pressures generated due to vapor-air or dust-air deflagrations. The vessel can be designed to allow deformation but to prevent rupture, or a stronger design can be provided that will prevent deformation. The DPC method has been used widely in Europe but has been used on a more l i i t e d basis in the United States. Muchgreater usage is expected in the United States following the adoption of the 1986 edition of National Fire Protection Association (NFPA) Standard 69 on Explosion Preven- tion Systems.

DPC can have several advantages over other explosion prevention and protection systems. Most important, it is a passive system. Hence, it is more reliable and thus reduces the risk of personal injury, vessel damage, and subsequent business loss. It may also lower overall capital, maintenance, and operating costs than other alternatives. Therefore, it can make both good safety sense and good business sense to use DPC. There are also some limitations of DPC. Chapter 5 of NFPA Standard 69 should be consulted for details.

Section 5.3 of NFPA 69 provides the design bases for vessels required to withstand a deflagration without having deformation or without rupturing. The follow summary gives the methodology presented in NFPA 69. That methodology is based on the thermodynamics of the deflagration and the mechanical design of the vessel.

PROBLEM NO. 34 71

The design pressure of the vessel is to be high enough to prevent rupture (PI) or to prevent deformation (Pd). The design pressure required to prevent deformation is higher than that to prevent rupture, because deformation of the vessel occurs before rupture. Deformation occurs when the yield strength is exceeded in the vessel walls, and rupture occurs when the ultimate strength is exceeded. Since a vacuum can follow a deflagration, vessels designed for DPC must also be designed to withstand full vacuum. The design pressure for the vessels is calculated from

and

where Pr =design pressure to prevent rupture due to internal deflagration, psig P,j = design pressure to prevent deformation due to internal deflagration, psig Pi =maximum initial pressure at which the combustible atmosphere exists, psig R =ratio of the maximum deflagration pressure to the maximum initial pressure;

For gas-air mixtures, R is taken as 9.0; and for dust-air mixtures, R is taken as 10.0

Fu = the ratio of the ultimate stress of the vessel to the allowable stress of the vessel Fy = the ratio of the yield stress of the vessel to the allowable stress of the vessel

If the operating temperature is below 2YC, the value of R is adjusted by

where Ti is the operating temperature in OC. For vessels fabricated of low carbon steel and low alloy stainless steel, Fu = 4.0 and Fy = 2.0.

PROBLEM: A carbon steel 285 vessel is used for avariety of processes using many types of flammable liquids and dusts. The processes are typically run near atmospheric pressure. The relief device set pressure is 20 psig. It is sized to have a maximum pressure accumulation of 20% during upset conditions. What should the vessel design pressures be to prevent rupture if the minimum operating temperatures are either 25°C or O°C? What should they be to prevent deformation?

(This problem was provided by Mr. John Noronha, Eastman Kodak Company.)

Problem No. 35

CHEMICAL ENGINEERING TOPIC: Thermodynamics: Vapor-Liquid Equi- librium

SAFETY AND HEALTH CONCEPT: Toxicology and Industrial Hygiene: Chronic Toxicity

BACKGROUND: When people are exposed to certain chemicals at relatively low but toxic concentrations, the toxic effects are only experienced after prolonged exposures. Mercury is such a chemical. Chronic exposure to low concentrations of mercury can cause permanent mental deterioration, anorexia, instability, insom- nia, pain and numbness in the hands and feet, and several other symptoms. The level of mercury that can cause these symptoms can be present in the atmosphere without a worker being aware of it because such low concentrations of mercury in the air cannot be seen or smelled.

Federal standards based on the toxicity of various chemicals have been set for the "Permissible Exposure Limit," or PEL. These limits are set by the Occupation- al Safety and Health Administration (OSHA). The PEL is the maximum level of exposure permitted in the workplace based on a time-weighted average (TWA) exposure. The TWA exposure is the average concentration permitted for exposure day after day without causing adverse effects. It is based on exposure for 8 hours per day for the worker's lifetime.

The present Federal standard (OSHAIPEL) for exposure to mercury in air is 0.1 mg/m3 as a ceilin value. Workers must be protected from concentrations !I greater than 0.1 mg/m if they are working in areas where mercury is being used.

PROBLEM: Mercury barometers are filled and calibrated in a repair shop. Mercury is stored in a small room that has no ventilation. A mercury spill occurs in the storeroom and is not completely cleaned up. What is the maximum mercury concentration that can be reached in the storeroom if the temperature is 20°C? You may assume that the room has no ventilation and that the equilibrium concentration will be reached. What would the concentration be if the room had a temperature of 1OoC? Is either level acceptable for worker exposure?

Problem No. 36

CHEMICAL ENGINEERING TOPIC: Thermodynamics: Constant Pressure Ex- pansion

SAFETY AND HEALTH CONCEPT: Explosions: Unconfined Burning of Vapor Cloud

BACKGROUND: If a flammable material is burned, there will be an increase in either thevolume of thegas produced (if the pressure is constant) or in the pressure in the container (if the volume is constant). The change in volume during isobaric combustion is caused by two changes in the system: the number of moles of products is usually greater than the number of moles of reactants, and the temperature of the system increases as the exothermic combustion reaction occurs. The change in the volume of the system can be quite large. Most combustion systems are designed so that the combustion reaction goes essentially to completion. Otherwise, the energy produced by the reaction is partially wasted. Pollution by release of fuel vapors also occurs if the reaction is incomplete. Carbon monoxide and other toxic gases that may be difficult to detect are also produced and released.

The calculations for the product gas temperature and volume can be made by assuming that all the gases behave ideally. The assumption does not cause any significant error because the system pressure will be low, about 1.0 atm if burning is at ambient pressure, and the system temperature will be quite high.

PROBLEM: Calculate the volume formed during the adiabatic combustion of 1 g-mol of n-butane in air.

1. Assume that the amount of air is the stoichiometric amount. 2. Assume that the amount of air is 150% of the stoichiometric requirement.

You may assume that the gases are ideal and that the combustion process is carried out at atmospheric pressure. The specific heats will not be constant, Assume that the reaction goes to completion. The reactants enter at 25OC.

Problem No. 37

CHEMICAL ENGINEERING TOPIC: Thermodynamics: Adiabatic Expansion of an Ideal Gas

SAFETY AND HEALTH CONCEPT: Explosions: Energy Release, Bursting Gas Container

BACKGROUND: Gases may be compressed and stored in tanks under pressure. If the pressure of the gas entering the tank is higher than the pressure that the tank can withstand, and if the overpressure protection devices normally provided for a tank are either inoperative or have not been installed, the tank can fail. The resulting explosion can damage the surroundings as the pressure wave from the explosion hits objects near the tank. The amount of work done by the expanding pressure wave can be estimated by assuming that the expanding gas does not mix with the surrounding air and that no heat is transferred to the gas as it expands. For such a case, the relationship between pressure and volume is

pVY = Constant

where P = pressure in expanding gas, psia V = molar vo!ume of expanding gas, ft3/lb-mole y = C p L

Cp = specific heat at constant pressure, BtuPb-moleOR Cv = specific heat at constant volume, BtuPb-moleOR

Once the amount of work done by the expanding gas has been calculated, the effect of that work on the surroundings can be estimated. The estimation of the effects of the pressure wave depends on the pressure in the wave at any point. The best estimates can be made based on the experimental data taken from tests in which a known quantity of TNT is detonated and the effects on the surroundings are measured. The work done by the expanding gas is converted to the equivalent work done by a quantity of TNT and the quantity of TNT is used to estimate the damage to be expected.

PROBLEM NO. 37 75

(a) Show that the work done on the surroundings by the expanding gas when a tank containing a compressed ideal gas explodes is given by

whe& PI, = burst pressure of the tank, psia VT = tank volume, ft3 Pa = atmospheric pressure, psia

You may assume that the expanding gas is ideal, that the specific heats are constant and that the process is isentropic.

(b) One pound of TNT releases about 2000 Btu of energy when it detonates. Estimate the equivalent energy release rate in pounds of TNT if a tank having a volume of 500 ft3 fails at 200 psig. The tank contains air, for which y = 1.4.

Problem No. 38

CHEMICAL ENGINEERING TOPIC: Thermodynamics: Isentropic Expansion of Pressurized Liquid

SAFETY AND HEALTH CONCEPT: Explosions: Rupture of Tank Containing Superheated Liquid

BACKGROUND: Most pressure vessels are designed according to one of the ASME pressure vessel codes. For tanks designed for use under conditions of ordinary severity and moderate pressures, the design criterion is that the tank be designed for pressures of four times the normal working pressure of the vessel. The design basis is the tensile strength of the steel in the vessel, so there is a four-to-one safety factor. Relief valves (or sometimes rupture disks) are provided to keep the vessel from exceeding the working pressure. On occasion, the relief valve will fail to open, or it may be taken off and replaced by a valve with a higher setting. The consequence of having an improper or nonworking relief valve can be substantial. Although most systems designed for heating liquids in tanks have thermostats to limit the temperature in the system, the thermostat can fail and allow heating to continue. If there is liquid in the tank, the pressure in the tank will rise as the temperature rises, and the tank pressure will equal the vapor pressure of the liquid. Thus, if the thermostat fails and the relief valve fails, the tank can rupture. If the tank ruptures, the superheated liquid in the tank will partially flash to vapor. The vapor and remaining liquid will expand rapidly, causing an overpressure wave that can damage the surrounding buildings or equipment. Personnel in the area may be injured or killed. Pressurized liquids can store substantial amounts of energy.

Although the explosion results in a rapidly expanding vapor-liquid system, most of the pressure drop in the expanding cloud is at the system boundary. Thus, for a simplified approach to the problem, the expansion can be considered isentropic. The internal energy change of the system can then be calculated, and since the expansion is also adiabatic, the work done against the surroundings during the expansion can also be calculated. That work can be compared to the work done by an equivalent amount of TNT during an explosion to estimate the damage potential of the energy released by the tank contents when the tank bursts.

PROBLEM: A 50-gallon hot water tank has a working pressure of 75 psia and a burst pressure of 300 psig. During maintenance operations, the tank is emptied, the tank and firing system are cleaned and repaired, and the tank is put back into service. When returned to service, the relief valve that is installed is incorrectly set at 500 psig. The tank is filled and the workers go to lunch after filling the tank. When they leave for lunch, one worker closes both the fill valve and the exit valve.

PROBLEM NO. 38 77

Another worker decides to heat the water during lunch so that the system will be hot and ready to test when they return. Neither tells the other what he has done. Before they return, the maintenance personnel are called to perform some emergency repairs at another location in the plant. The water tank heating system continues to heat the water until the water pressure reaches the failure pressure of the tank. The tank explodes when the pressure reaches 300 psig.

Estimate the blast energy in terms of the TNT equivalent. You may assume that the explosion is adiabatic. The water flashes to steam and water at 100°C; you may assume the expansion is isentropic. Assume that no air mixes with the steam during expansion and that the tank is just filled with liquid water which is saturated at 300 psig when the tank explodes. The work equivalent of exploding TNT is about 2000 Btullb.

Problem No. 39

CHEMICAL ENGINEERING TOPIC: Thermodynamics: Constant Volume Gas Phase Reaction

SAFETY AND HEALTH CONCEPT: Explosions: Pressure Rise for Enclosed Combustion Reaction

BACKGROUND: Refineries and chemical plants use a variety of low pressure vessels as knockout drums and seal drums. Most of these vessels are operated at very low pressures, but they may contain flammable mixtures of vapor and air. It is quite unlikely that ignition will occur in such a vessel because there is usually no source of ignition. However, there is always a chance that ignition might occur, so the American Petroleum Institute's Recommended Practice 521 (API RP 521), 1st edition, states: "Most knockout drums and seal drums will be operated at relatively low pressures. To ensure safe conditions and sound construction, a minimum design pressure of 50 psig is suggested. . . . A vessel with 50 psig design pressure should not rupture if an explosion occurs. Stoichiometric hydrocarbon-air mixtures can produce peak explosion pressures in the order of 7 to 8 times operating pressure, most flare seal drums operate in the range of 0 to 5 psig, and ASME code-allowable stresses are based on a safety factor of 4 to 1." Section 8, Division 1 of the ASME (American Society of Mechanical Engineers) pressure vessel code specifies a safety factor of four to one (applicable at low pressures only). That safety factor implies that a vessel with a stated mechanical design of 50 psig should not rupture at pressures up to 200 psig.

PROBLEM: Show that the 50 psig design pressure suggested by API RP 521 will contain the explosive combustion of a mixture of air and n-hexane with initial conditions of 77°F (25°C) and 5 psig and stoichiometric concentration of n-hexane in air. Compare your result to the estimated pressure rise of "7 to 8 times operating pressure" referred to in the API standard. You may assume that the reaction proceeds to completion and that the products of combustion are carbon dioxide and water.

(This problem was suggested by Mr. J. R. PhiIIips, a graduate student at the University of Arkansas.)

Problem No. 40

CHEMICAL ENGINEERING TOPIC: Thermodynamics: Combustion

HEALTH AND SAFETY CONCEPT: Hazardous Materials Generation and Dis- posal

BACKGROUND: The handling of hazardous waste materials is covered by a number of laws and regulations that are intended to minimize the possibility of hazardous materials being released to the environment where they might have an adverse effect on the environment or people. Several methods have been developed for the disposal of hazardous materials, and, of course, the nature of the waste is important to the suitability of any particular method.

Among the methods used are l a n d f i g , which means burying the waste in the ground; deep well disposal, which is pumping it into a deep underground formation; recovery and recycle, which is reclaiming the material for reuse; biological treatment, which is allowing microorganisms to break down the waste into harmless, or less harmful materials; and incineration, which is burning it.

All the methods have their advantages and disadvantages, depending upon the nature of the waste. Waste minimization is the preferred way, but when there is waste, probably recovery and recycle would generally be preferable when it is a practical solution.

If a material can be made nonhazardous by biological treatment, then this would be a desirable disposal method if it could be carried out without release to the environment, since this is usually a fairly inexpensive method and serves to break down the waste, rather than just to provide perpetual storage as would be the case with the landfilling.

Incineration is often a desirable option if the material can be made nonhazardous by thermal treatment. In general, incineration is a practical solution to the disposal of organic materials, including halogenated materials. The Environmental Protection Agency (EPA) has developed regulations concerning incineration, including the temperatures and residence times required for the destruction of particular types of waste. An approved incineration process will be specified both with respect to the thermal conditions imposed, but to the degree of destruction required as well. The hazardous waste incinerators, indeed, any incinerator, must not emit more than specified amounts or concentrations of some combustion products. Usually this will require the use of scrubbers to remove gaseous materials and/or some particulates. Often filters (baghouses) or electrostatic precipitators will be required to remove very fine particulates.

EPA regulations require that hazardous waste incinerators must have destruction and removal efficiencies such that 99.99% of the principal hazardous con-


stituent is destroyed. Furthermore, the minimum temperature may be specified as well. A typical destruction temperature is 2000°F. In a great many cases, these conditions may be obtained only by the addition of an auxiliary fuel, such as natural gas, since the heating value of the waste may prove to be inadequate to create the necessary temperature.

The following problem deals with such a situation.

PROBLEM: A low concentration of a hazardous component is in solution in a mixture of methyl alcohol and water. The hazardous component distributes between the alcohol and the water in such a manner that, if the alcohol and water were to be separated by distillation, each product would be contaminated to such an extent that it could not be reused. It has therefore been decided that the waste, with the water and the alcohol, will be incinerated. A temperature of 2000°F is required.

The mixture of methanol and water is 30% methanol by weight, and the amount of contaminant is low enough that its heating value can be ignored. You are to determine what natural gas (methane) rate will be required to incinerate this material if the latter is burned with 100% excess air. You may assume that the combustion is adiabatic.

Assume that all the entering streams are at 25°C. The water and alcohol is a liquid solution.

Problem No. 41

CHEMICAL ENGINEERING TOPIC: Thermodynamics: Vapor Liquid Equi- librium

SAFETY AND HEALTH CONCEPT: Properties of Materials: Flash Point

BACKGROUND: Most combustion reactions occur in the gas phase. In order for any flammable material to burn, there must be both fuel and oxidizer present. There must also be a minimum concentration of the flammable gas or vapor in the oxidizer. For most fires to occur, minimum fuel concentration must exist in air at ambient temperature. The minimum concentration at which ignition will occur is called the lower flammable limit (LFL). If the flammable material is normally liquid, the liquid must be warm enough to provide a vapor-air mixture equal in fuel concentration to the LFL concentration. The liquid temperature at which the vapor concentration reaches the LFL can be found experimentally. It is usually measured using a standard method called a "closed cup flash point" test. The "flash point" of a liquid fuel is thus the liquid temperature at which the concentration of fuel vapor in air is large enough for a flame to flash across the surface of the fuel if an ignition source is present.

The flash point and the LFL concentration are closely related through the vapor pressure of the liquid. Thus, if the flash point is known, the LFL concentration can be estimated, and if the LFL concentration is known, the flash point can be estimated. In most cases, the calculation can be made for pure liquids using Raoult's law. However, if the liquid is a mixture, particularly one in which the components are dissimilar, the liquid solution may be nonideal. Then the liquid phase activity coefficients may need to be determined if an accurate estimate of the relationship between flash point and LFL is to be made. The system total pressure is ambient, so it is low enough for the vapor (or gas) phase above the liquid surface to be considered ideal.

PROBLEM: Estimate the flash point of n-octane and compare it with the experimental value given in the literature. (See NFPA Standard 325M, "Properties of Flammable Liquids" or Sax's Dangerous Properties of Industrial Materials.) Start with the basic tenet that the fugacity in the liquid phase must equal that in the vapor phase. Justify each assumption required to arrive at Raoult's law. Then use Raoult's law to estimate the flash point. The lower flammable limit of n-octane is 1.0%.

Problem No. 42

CHEMICAL ENGINEERING TOPIC: Thermodynamics: Vapor Liquid Equi- librium


BACKGROUND: Most combustion reactions occur in the gas phase. In order for any flammable material to burn, there must be both fuel and oxidizer present. There must also be a minimum concentration of the flammable gas or vapor in the oxidizer. For most fves to occur, a minimum fuel concentration must exist in air at ambient temperature. The minimum concentration at which ignition will occur is called the lower flammable limit (LFL). If the flammable material is normally liquid, the liquid must be warm enough to provide a vapor-air mixture equal in fuel concentration to the LFL concentration. The liquid temperature at which the vapor concentration reaches the LFL can be found experimentally. It is usually measured using a standard method called a "closed cup flash point" test. The "flash point" of a liquid fuel is thus the liquid temperature at which the concentration of fuel vapor in air is large enough for a flame to flash across the surface of the fuel if an ignition source is present.

The flash point and the LFL concentration are closely related through the vapor pressure of the liquid. Thus, if the flash point is known, the LFL concentration can be estimated, and if the LFL concentration is known, the flash point can be estimated. In most cases, the calculation can be made for pure liquids using Raoult's law. However, if the liquid is a mixture, particularly one in which the components are dissimilar, the liquid solution may be nonideal. Then the liquid phase activity coefficients may need to be determined if an accurate estimate of the relationship between flash point and LFL is to be made. The system total pressure is ambient, so it is low enough for the vapor (or gas) phase above the liquid surface to be considered ideal.

PROBLEM: The flash point of a liquid mixture can be estimated by finding the temperature at which the equilibrium concentration of the flammable vapors in air reaches a concentration such that

where yi is the vapor phase mole percent of component i and LFLi is the lower flammable limit concentration of component i expressed in mole percent. Estimate 1 the flash point of a liquid mixture containing 60 mole percent n-octane, 15 mole

PROBLEM NO. 42 83

percent n-nonane, and 25 mole percent n-decane. The LFL values are 1.0% for n-octane, 0.8% for n-nonane, and 0.8% for n-decane. Vapor pressure data can be found in Perry's ChemicalEngineersl Handbook, Reid, Prausnitz, and Poliig's The Properties of Gases and Liquids, or other literature sources.

Problem No. 43

CHEMICAL ENGINEERING TOPIC: Thermodynamics: Vapor Liquid Equiliirium


BACKGROUND: Most combustion reactions occur in the gas phase. In order for any flammable material to burn, there must be both fuel and oxidizer present. There must also be a minimum concentration of the flammable gas or vapor in the oxidizer. For most fires to occur, a minimum fuel concentration must exist in air at ambient temperature. The minimum concentration at which ignition will occur is called the lower flammable Limit (LFL). If the flammable material is normally liquid, the liquid must be warm enough to provide a vapor-air mixture equal in fuel concentration to the LFL concentration. The liquid temperature at which the vapor concentration reaches the LFL can be found experimentally. It is usually measured using a standard method called a6'closed cup flash point" test. The "flash point" of a liquid fuel is thus the liquid temperature at which the concentration of fuel vapor in air is large enough for a flame to flash across the surface of the fuel if an ignition source is present.

The flash point and the LFL concentration are closely related through thevapor pressure of the liquid. Thus, if the flash point is known, the LFL concentration can be estimated, and if the LFL concentration is known, the flash point can be estimated. In most cases, the calculation can be made for pure liquids using Raoult's law. However, if the liquid is a mixture, particularly one in which the components are dissimilar, the liquid solution may be nonideal. Then the liquid phase activity coefficients may need to be determined if an accurate estimate of the relationship between flash point and LFL is to be made. The system total pressure is ambient, so it is low enough for thevapor (or gas) phase above the liquid surface to be considered ideal.

PROBLEM: Estimate the flash point of a mixture made by mixing 600 ml of methanol and 400 ml of water. The solution is not ideal, and the activity coefficients must be estimated. For the estimation of activity coefficients, first determine the activity coefficients for methanol-water solutions from vapor-liquid equilibrium data. Assume that the activity coefficients are a function of composition only, and do not depend on the system pressure and temperature. Is such an assumption justified? Vapor liquid equilibrium data can be found in Perry's Chemical Engineers'Handbook. Vapor pressure data can be found in the Handbook as well. The LFL of methanol can be found in NFPA 325M, "Properties of Flammable Liquids" or Sax's Dangerous Properties of Industrial Materials. Once the mixture is ignited, will it continue to burn?

Problem No. 44

CHEMICAL ENGINEERING TOPIC: Thermodynamics Vapor LiquidEquilibrium


BACKGROUND: Most combustion reactions occur in the gas phase. In order for any flammable material to burn, there must be both fuel and oxidizer present. There must also be a minimum concentration of the flammable gas or vapor in the oxidizer. For most fires to occur a minimum fuel concentration must exist in air at ambient temperature. The minimum concentration at which ignition will occur is called the lower flammable limit (LFL). If the flammable material is normally liquid, the liquid must be warm enough to provide a vapor-air mixture equal in fuel concentration to the LFL concentration. The liquid temperature at which the vapor concentration reaches the LFL can be found experimentally. It is usually measured using astandard method called a "closed cup flash point" test. The "flash point" of a liquid fuel is thus the liquid temperature at which the concentration of fuel vapor in air is large enough for a flame to flash across the surface of the fuel if an ignition source is present.

The flash point and the LFL concentration are closely related through the vapor pressure of the liquid. Thus, if the flash point is known, the LFL concentration can be estimated, and if the LFL concentration is known, the flash point can be estimated. In most cases, the calculation can be made for pure liquids using Raoult's law. However, if the liquid is a mixture, particularly one in which the components are dissimilar, the liquid solution may be nonideal. Then the liquid phase activity coefficients may need to be determined if an accurate estimate of the relationship between flash point and LFL is to be made. The system total pressure is ambient, so it is low enough for thevapor (or gas) phase above the liquid surface to be considered ideal.

PROBLEM: Estimate the flash point of n-decane that contains 5.0 mole percent propane. Vapor pressures can be found in Perry's Chemical Engineers'Handbook. The LFL for propane is 2.2% and that for n-decane is 0.8%. The LFL of a mixture of gases can be estimated by finding the concentrations in the gas phase such that

where yi is the vapor phase mole percent of component i and LFLi is the lower flammable limit concentration of component i expressed in mole percent. You may assume the liquid mixture is ideal and use Raoult's law, or you may use the K-factor charts for estimatingylx or for finding the temperature for a knownylx for either component.

Problem No. 45

CHEMICAL ENGINEERING TOPIC: Thermodynamics


BACKGROUND: Many gases are commonly compressed from normal atmospheric pressure to quite high pressures, either to facilitate transfer or storage or to take part in a chemical reaction that is run more favorably at high pressures. As anyone who has ever used a bicycle pump knows, as a gas is compressed its temperature rises if no heat is removed. The reason is explained by the first law of thermodynamics. If the gas is co~pressed, and if no heat is removed, the energy put into the gas in the form of work to increase the pressure becomes part of the internal energy of the gas. The increase in internal energy increases the temperature of the gas. For an ideal gas with a constant specific heat, for example, the energy balance requires that, for an adiabatic process,

where AU = change in internal energy W = work Cv = specific heat at constant volume

AT = change in temperature If the compression process is reversible as well as adiabatic, the entropy does not change during compression of the gas, and, as a consequence,

where TI and Tz are beginning and ending temperatures, Pi and P2 are beginning and endingpressures, and y is the ratio Cp/Cv, with Cp the specific heat at constant pressure. If the specific heats are not constant, but the process is still reversible, the temperature can be calculated by integrating the equation for the entropy change

where AS is the entropy change, S2 - S I , and R is the gas law constant.

PROBLEM NO. 45 87

If the process is not reversible, and if the entropy change can be calculated, the temperature can still be calculated. The procedure becomes even more complicated if the gas is not ideal and the specific heat depends on the pressure as well as the temperature.

If the compression ratio (P2tP1) is small, there is usually no consequence to the temperature rise that accompanies compression. However, in some cases there can be a potential hazard. Consider the compression of air in which there is a small amount of fuel, for example. If the compression ratio is large enough, the mixture of fuel and air will have a temperature high enough after compression that the autoignition temperature of the fuel-air mixture will be reached and the mixture will explode. (The autoignition temperature, sometimes called AIT, is the temperature at which a fuel-air mixture will ignite without external energy being applied. No spark or flame is needed for ignition because the temperature is high enough to initiate combustion.) In fact, the diesel engine operates by injecting diesel fuel into the combustion chamber after the air has been compressed to a high pressure as the piston rises to the top of its stroke. The compression ratio in a diesel engine is about 22 to 1.

In an air compressor where the final pressure is quite high, there can be a possibility of explosion if there is flammable lubricating oil in the compressor or in the discharge from the compressor. Thus, if very high pressures are required, the air may have to be removed from the compressor at an intermediate pressure and cooled before completing compression to the fmal pressure. Intercoolers also make the compression process more efficient because they reduce the amount of work required for compression to a given pressure.

An explosion might also occur if certain gases are compressed to high pressures and attain high temperatures. Such gases as ethylene and acetylene that have positive heats of formation may undergo spontaneous explosions at the temperatures found during compression. Combustible gases can explode if they are drawn into an air compressor and compressed to a pressure high enough to reach the AIT.

PROBLEM: What is the final temperature reached after compressing ethylene and air from 14.7 psia to 1000 psia if the initial temperature is 100°C? You may assume the ratio of CplCv to be 1.22. The AIT of ethylene is 490°C. Will an explosion occur if the mixture is a flammable concentration?


Problem No. 46


SAFETY AND HEALTH CONCEPT: Hazard Reviews; Explosions

BACKGROUND: Most people have heard the tale of the loss of a kingdom because of the loss of a nail. In its simplest form, the tale tells that a horseshoe nail was lost from the shoe of the king's horse. Because the nail was lost, the horseshoe was also lost. The loss of the horseshoe meant the horse was no longer available for the king, and the king in turn could not reach the site of an important battle. Without the king, there was no one at the battle to direct the forces assembled there to protect the kingdom. Without direction, the army lost the decisive battle, and the kingdom was conquered by the opposing army.

While the tale may seem simple, a more recent true story illustrates the potential for damage resulting from a seemingly small occurrence. In 1969, a worker was walking across a high walkway when he stumbled. To save himself from falling, he grabbed a nearby valve stem. The valve stem was not strong enough to support the stress imposed on it by the worker's weight, and it failed. Flammable liquid spewed out of the broken valve and formed a cloud of flammable vapor that was ignited when it reached a nearby truck. The resulting explosion and fire spread to other equipment nearby, and the fire lasted for six days, completely destroying the plant and doing more than $4 million worth of damage.

In order for a fire or explosion to occur, there must be a fuel, an oxidizer, and an ignition source present. These three items are called the "fire triangle," and if any one is missing, ignition is not possible. Most fire and explosion prevention schemes attempt to keep either the fuel or the ignition source from being present. Oxygen is always present in the open atmosphere. In closed systems, inert gases are frequently used to preclude flammable mixtures from being formed, There is a minimum fuel concentration required for ignition. It is called the lower flammable limit concentration (LFL), and it is thevolume or mole fraction of flammable vapor in air that must be reached if ignition is to occur. There is also an upper flammable limit (UFL) concentration. If the fuel concentration is above the UFL, there is too little oxygen in the mixture to support combustion, and themixture will not ignite. The flammability limits of many materials are included in references such as the NIOSH Pocket Guide to Chemical Hazards; the National Fire Protec- tion Association Standard 325M, "Properties of Flamable Liquids"; and Sax's Dangerous Properties of Industrial Materials.

The concentration of a volatile material in air may be calculated easily using basic thermodynamic relationships such as Raoult's law, if it is assumed that the

PROBLEM NO. 46 89

concentration in air reaches equilibrium. Thermodynamic calculations can also be used to determine the energy released if the fuel-air mixture is ignited. By making such calculations, an engineer can estimate the potential for damage should a spill and fire occur. Of course, the damage calculations show what might happen, not what must happen. The calculations become part of a hazard review, and a decision must then be made to determine what action must be taken to reduce the consequence of a flammable material spill or to reduce the probability the spill will occur. Experience has shown a number of methods of reducing spills and their consequences. They include the use of proper storage aod handling methods, the elimination of ignition sources, reduction in the quantity of material stored and handled, and installation of equipment to detect spills and shut down transfer operations when spills occur. Substantial effort and resources are expended to reduce the risk from flammable liquids.

PROBLEM: A worker is transporting a can containing 1.5 L of carbon disulfide from one laboratory to another late on a Friday afternoon. He stops at his office on the way between the labs and sets the can on his desk. Noticing the late hour, he grabs his briefcase and rushes to join his car pool for the trip home. The lid on the can is loose, and when the janitor arrives to clean up a short time later, he notices an odor faintly like garlic in the office. He cleans up as quickly as possible and leaves, closing the door behind him to keep the disagreeable odor from spreading. An energy-saving policy has dictated the shutdown of air conditioning and heating systems over the weekend, and the temperature in the office stays at 8S°Fover the weekend. All the carbon disulfide evaporates into the air in the office. There is no ventilation because the air conditioning system is shut off, so all the carbon disulfide vapor remains in the office. The worker returns early on Monday morning, and turns on the lights.

What surprise greets the worker when he turns on the lights? Name a few simple precautions that might have been taken to prevent the surprise. The energy released by one pound of exploding TNT is about 2000 Btu. Is the surprise likely to be noticed anywhere else in the building?

You will need some additional information to solve this problem, and you may have to make some assumptions as well. The references Sited above will provide flammability limit data. Physical property data can be found in the Chemical Engineers'Handbook If the worker's office is about the usual size, it will be about 10 ft by 12 ft with a 9-ft-high ceiling. Carbon disulfide burns easily in air, and it has a very low ignition temperature. Some references list the ignition temperature as 80°C, and most authorities agree it is 100°C or less. Carbon disulfide burns to form carbon monoxide, carbon dioxide, sulfur dioxide, and sulfur trioxide, depending

90 SAFEVY, HEALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

on the conditions of combustion. In this case, you may assume the combustion products are carbon dioxide and sulfur dioxide. Since you need only an estimate of the energy from the explosion, you may assume the reaction occurs at the standard state, 25°C.

Problem No. 47

CHEMICAL ENGINEERING TOP 'IC: Thermodynam


BACKGROUND: One of the fundamental findings of the application of the concepts of thermodynamics is that as a gas is compressed its temperature increases. That increase in temperature can sometimes lead to problems in the operation of air compressors. For example, the lubricating oils used for air compressors are frequently based on mineral oils, and they may have autoignition temperatures (AIT) as low as 500°F at atmospheric pressure. The autoignition temperature decreases at higher pressures. There is usually little of the lubricating oil present in the air stream leaving the compressor, so the chances of an explosion are quite low. However, the small amount of lubricating oil in the air may collect on the inside of the piping downstream of the compressor and result in an explosion that bursts the piping. Severe damage may occur. If workers are nearby, they may be injured or killed.

Several methods may be used to prevent explosions in air compressors. They include limiting the compression ratio, cooling the air between compressor stages, using oils that have a higher ignition temperature, and keeping the system clean so there is not enough oil on the piping walls to form a film thick enough to allow a mist to form.

The temperature of the air leaving the discharge of a compressor may be calculated through thermodynamic relationships. The energy and entropy balances can be used for the purpose. Quick estimates can be made using ideal gas properties and the assumption of reversible operation.

PROBLEM: Air is compressed in a chemical plant. The air enters the compressor at ambient temperature; for this problem, the ambient temperature may be assumed to be as low as 10°F and as high as 95°F. Air enters the compressor at atmospheric pressure. What is the maximum pressure the air can reach without causing an explosion if the explosion may occur at a compressor exit temperature of 450°F? Make the calculation under two sets of conditions:

1. Assume the air to be an ideal gas with a constant specific heat of Cp = 7.0 caVgmole K and with the compression occurring adiabatically and reversibly. 2. Consider the air to be an ideal gas with Cp given by


where Tis inK and Cp is in caVgmole K. The compressor still operates adiabatically and reversibly.

Determine the outlet temperature for the air having properties of Part 2 if the air is compressed to 150 psig. You may neglect the effect of the small amount of oil that might be present in the air going through the compressor.

Problem No. 48


SAFETY AND HEALTH CONCEPT: Vapor Releases

BACKGROUND: Many flammable and toxic substances are used in chemical processing plants. Some of the substances are used directly in the processes, and others are used indirectly for such purposes as heat exchange or in separations. Frequently, there is no viable alternative to using the flammable or toxic solvent or intermediate material, although a search should always be made to find the least hazardous material for a particular duty.

When it is necessary to use a flammable or toxic material (some chemicals are both flammable and toxic), the properties of the material should be well known, and the plant design and operation should be chosen to minimize the probability that any of the hazardous material will be spilled or will leak from piping and tanks. Then, even though stringent precautions have been taken to avoid releases, studies of the consequences of spills or leaks should be made so that emergency operations can be planned for any potential accident at the plant. The probability of a spill can be estimated as part of the formal process hazard review that should be performed for each plant. The process hazard review is part of the information required for administrative decisions concerning the safety of plant personnel and the cooperative effort needed to protect neighboring areas from potential accidents at the plant.

Releases of materials in the plant can be either as solids, liquids, or gases. Unless the materials are very fine dusts, solid releases will seldom have an immediate effect outside the local area of the spill and are even less likely to leave the plant. Gas releases cannot be prevented from leaving the area of the leak and eventually the gas will leave the plant boundaries. Liquid releases occupy a middle position, because as the liquid vaporizes, the vapor will be blown beyond the location of the spill and then blown outside the plant by the wind. One of the important estimates that needs to be made in determining the potential effects of a large liquid spill is to estimate the rate of vaporization of the liquid.

Since further consideration of the possibility for ignition of a flammable vapor cloud or the toxic effects of a vapor cloud depend strongly on the rate of the spill and the rate of vaporization, it is important to place realistic bounds on the estimates of vaporization rates. The simplest analysis to begin with is that which can be determined with a simple adiabatic energy balance on the liquid stream. For example, if a heated liquid is discharged from a pipe, the process will be essentially adiabatic. Then, the enthalpy of the vapor-liquid mixture leaving the pipe will be the same as the enthalpy of the liquid flowing through the pipe. The

94 SAFEI"Y, HEALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

problem arises in trying to determine the temperature of the liquid after it leaves the pipe. Assuming the liquid is above its normal boiling point in the pipe, does it cool to its boiling point on leaving the pipe? Or does it cool to the temperature of the ambient air? Or may there be circumstances where the liquid will cool to a temperature even lower than the ambient temperature?

Finally, is the effect of cooling of the liquid important in determining the fraction of liquid that flashes to vapor? Some of these questions are explored in the following problem.

PROBLEM: Two chemicals are being considered for use as a solvent in a processing plant, n-heptane and methyl alcohol. Both will be suitable for the purpose, and a decision must be made on which to use in the plant. Both are flammable, of course, and both are pollutants if released to the atmosphere or spilled onto the ground. Precautions have been taken in the plant design to keep any of the solvent from spilling, but the process hazard review group must assume that spillage has occurred and determine the potential consequences. You are to estimate the rate of vapor generation that could accompany a spill of the solvent. Regardless of which solvent is used, it must be used at a relatively high temperature. Solvent flow at some points in the process will be at normal rates of 500 galJmin. You have estimated the flow through a broken pipe will be double the normal rate. The li quid temperature in the flowing stream will be as high as 400°F. In order to help in the choice of which solvent to use, you decide to estimate the fraction of the liquid that will vaporize under three assumed conditions: (a) the liquid and vapor are at the normal boiling point following the release, (b) the liquid and vapor cool to ambient temperature of 80°F following the release, and (c) the liquid and vapor cool to 25OF following the release. The last estimate is based on an analysis of heat transfer between the liquid-vapor mixture and the surrounding air. Your first task is to determine the fraction of the liquid that will vaporize based on the three assumed conditions. In order to compare the two candidate solvents, you then estimate the volume of the vapor-air mixture that would be formed if the reIease continues for 10 min. You do that by determining the vapor cloud size based on having a mixture at its lower flammable limit, and you assume the mixtures of vapor and air will be ideal because the mixture will be at a low pressure. The solvent that generates the smaller vapor-air cloud will be chosen for the process. Using these criteria, which solvent will you choose for the process? Might you also consider the heat of combustion of each candidate solvent? Why?

Problem No. 49


SAFETY AND HEALTH CONCEPT. Storage, Handling, and Transport

BACKGROUND: Most of the materials stored, transported, and processed in modern chemical plants are stable under the storage and handling conditions and are unlikely to undergo any spontaneous transformations that cause unsafe operations. However, some materials have the potential for chemical reaction when mixed with other chemicals, such as occurs if a flammable material is mixed with air. Spontaneous heating of solids is also known to occur when reactions such as oxidation are possible and there is no pathway for the heat to escape.

There are also a few materials such as certain unsaturated hydrocarbons that are stable at low temperatures, but as the temperature increases, they begin to react. As they react, the energy released heats the mixture further until a rapid, even explosive, reaction occurs. Acetylene and ethylene are two gases used frequently in organic chemical reactions. Either can be used safely, but either has the potential for rapid decomposition. If the heat of formation of either acetylene or ethylene is found in a table, it will be seen to be positive. A positive heat of formation indicates that the substance will liberate energy if it is decomposed into its constituent molecules. The system temperature will have to be raised high enough for the reaction to begin, of course, and if the substance is kept cool and away from ignition sources, it may be handled safely. However, if the decomposition reaction begins, it will continue until equilibrium is reached. In many cases, equilibrium will be reached only at avery high temperature or when the decomposition is essentially complete.

The property of decomposition of a pure material is illustrated in the following problem for acetylene. However, you should keep in mind that any material that has a positive heat of formation may be capable of undergoing an exothermic decomposition if the conditions under which it is held allow the reaction to start. Such decomposition reactions are the basis for design of certain explosives, so it is easy to see that the destructive effect may be quite large.

PROBLEM: Acetylene may decompose with a rather large liberation of energy. You are to determine the temperature and pressure that would be reached by acetylene that decomposes in a tank, starting with pure acetylene. Assume the decomposition produces only carbon (in the form of graphite) and hydrogen. You may also assume the starting temperature to be 25°C in the bulk of the acetylene, even though a higher temperature, perhaps in the form of a spark, would usually be required to initiate the decomposition reaction. The initial pressure in the tank

% SAFETY, HEALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

is 5 atm absolute. Once the decomposition begins it will be very rapid, so you may assume the reaction is adiabatic. Keep in mind that there will be no increase in the volume of gas or in the number of moles of gas in the tank and that the solid produced will not influence the pressure to any great extent. However, if the temperature rises, the pressure of the confined gas will also rise. You may assume the gases to behave ideally, but you should justify your assumption. You should also determine the composition of the gas in the tank at the end of the reaction.

Problem No. 50

CHEMICAL ENGINEERING TOPIC: Heat Transfer

SAFETY AND HEALTH CONCEPT. Process Control, Interlocks and Alarms

BACKGROUND: A complex chemical process requires the use of automatic controls to maintain a successful operation. The modern plant is much too complex for human operators to maintain all process variables under adequate control. Many students are aware of the difficulty of keeping even a very simple process under control in the laboratory. For example, if an individual were trying to maintain a constant liquid level in a tank with one stream leaving and one entering, it would require his or her constant attention to maintain the level if the outlet stream were to fluctuate at all with time. Only the very simplest chemical process could be held within acceptable control without instrumentation and process control devices.

Modern chemical processing relies heavily upon instrumentation, not only to maintain product quality and to keep the production schedule, but to prevent accidents as well.

Instrumentation for accident prevention takes many and varied forms, and only examples can be mentioned here. A typical application would be controls to detect and respond to a rapid temperature rise in a reactor in order that the feed may be interrupted, if that would slow the reaction. Another application might be a level sensor in a tank to prevent overfilling, since overfilling might cause overpressure or spillage of a toxic or flammable material.

Instruments that function to control the input of material or energy to a process are particularly critical. Whenever possible, such control loops should be designed to be "fail-safe"; that is, in the event of failure of the instrument or a component of the control loop, the manipulated variable would go to a safe condition. For example, if an exothermic reaction is controlled, at least in part, by control of the input of a reactant, then a fail-safe control loop would be one that would shut off the flow of that reactant if the control equipment failed. Unfortunately, many of the typical process control devices are not inherently fail safe.

When controls are used to prevent accidents, the response time is frequently critical to the success of the prevention. The exothermic reaction is a good example of such a case, because if the temperature begins to rise out of control, the reaction rate increases exponentially with the temperature, so it is critical that the loss of control be detected quickly. The following problem considers the response time of two different temperature sensors.


PROBLEM: A thermocouple assembly is to be installed in a stirred tank reactor. Two different designs are being considered, one is a cylinder that is made of 316 stainless steel; it is 7.0 mm in diameter, and is 100 mm long. The alternate design is also a cylinder made of an alloy similar to bronze; it is 4.5 mrn in diameter and 80 mm long.

At the location where the thermocouples are to be installed, the velocity of the stirred liquid is expected to be 0.3 d s , essentially at right angles to the thermocouples.

If the temperature of the vessel contents were to suddenly rise by 10°C, how long would it take for each of these thermocouples to respond by indicating 90% of the increase, that is, how long will each take to increase by 9"C?

Some properties and model assumptions are required for this problem: Assume that the thermocouples are solid metal, although that is not the case,

because the large diameter is really a shield around the actual sensor.

St. Steel Bronze

Density 8238 8800 kg/m3 Conductivity 13.4 52 W/m K Heat capacity 468 420 J/kg K

The liquid properties are:

Density 985 kglm3 Heat capacity 2.5 W/kg K Viscosity 0.006 kg/m s Thermal conductivity 0.09 Wlm K

It would be reasonable to expect that the stainless steel thermocoupk will last longer and be Iess liabIe to damage than the bronze one. Which one would you specify? Explain.

Problem No. 51

CHEMICAL ENGINEERING TOPIC: Heat Transfer; Mass Transfer


BACKGROUND: One of the potential hazards to workers in an industrial environment is that of exposure to excessively high temperatures in the workplace. Among the various consequences that may result from excessive heat stress are heat skoke, heat exhaustion, and heat cramps. Heat stroke is a life threatening condition, and recovery may not be complete. Other responses to heat stress are less severe.

The normal metabolic processes generate heat within the body. The magnitude of the generation depends largely upon the level of physical activity, and also upon the body mass of the individual. A typical resting value for heat generation is 350 B t l ~ / l ~ (103 W) for a man of about 150 to 155 lb (68 to 70 kg) of body mass. Very vigorous physical labor, such as working with an axe may have a heat generation rate as high as 5500 Btu/hr (1612 W). It is probable that vigorously competing athletes will have an even higher metabolic generation rate. The metabolic rate for a machinist might be about 750 B t u h (220 W), which is approximately the same for a person walking. Typical values for workers in an industrial activity would range from about 700 to 1500 Btu/hr (205 to 440 W). A worker will rarely exceed a rate of 1200 Btufhr (352 W) over an extended time.

In order to niaintain a proper body temperature, the heating and cooling rates must be in balance over an extended time. A temporary rise of 2°F (l.l°C) temperature is about the maximum that would be tolerable. We might note that a 2°F rise in body temperature for a man who weighed 150 lb would follow from an increase in the energy storage within the body of about 300 Btu (316 kJ), since the body's heat capacity is approximately that of liquid water.

Industrial hygienists use the following relation for the thermodynamic processes concerned:

S = M + R + C - E where

S = the rate of storage of energywithin the body (essentially the rate of change of internal energy of the body)

M = the metabolic rate of generation (discussed above). R = net rate of radiative interchange with the surroundings. (Here R will be

positive if the surroundings are warmer than the skin temperature.) C = the rate of convective heating of the body by the surroundiig air E = Rate of cooling due to evaporation of sweat. All terms must be in consistent units such as Btu/hr or watts.


It is noted that the M term must always be positive, but the i? and C terms may be either positive or negative. The evaporative cooling will always result in cooling of the body, so the term must be positive in the equation above. The rate of storage of energy within the body, S, can be either positive or negative.

The evaporative cooling is the primary mechanism by which the body maintains proper temperature control.

A small cooling effect can be derived from the breathing process, that is, inhaling less than saturated air, and exhaling air that is essentially saturated, but this will normally be only a minor effect, less than 10 Btu/hr in most instances.

PROBLEM: A man is working in an environment where the temperature of the surrounding air is 96°F (35.6"C). The man has a body mass of 160 lb (72.6 kg), and is working at a rate of 1050 Btulhr (308 W) (metabolic rate). A spot cooling fan is set up to blow air past his body at a velocity of 10 ftlsec (3.05 d s ) . The air, also at 96°F has a relative humidity of 60%. What rate of evaporation of sweat would be required to maintain this man's body temperature at a constant value?

Consider the mass and heat transfer rates that might be possible in these circumstances and determine whether the man can work indefinitely or whether his body temperature will rise over time. If his body temperature rises with time, how long can he work in such circumstances before it rises by TF.?

Some model simplifications will be required to work this problem. First, assume that radiation interchange is negligible. Next, we wish to make a reasonable estimate of the skin temperature. We know it will be higher than the wet bulb temperature, but it may not be much higher. Assume that it is 5°F (2.S°C) above the wet bulb. Finally, we need an estimate of his surface area and his shape. We could assume that he is a cylinder, 12 in. (0.305 m) in diameter, and 6 ft (1.83 m) tall. (If you don't think this is a reasonable approximation, then you might make an estimate of his arm dimensions, his leg dimensions, and his trunk dimensions.)

HINTS FOR SOLUTION: It will be necessary to find a suitable model to predict the heat transfer coefficient for a cylinder in a cross-flow of air. From the heat transfer model, the mass transfer coefficient can then be deduced from the Chiton-Colburn analogy.

The body will be either heated or cooled by convection according to the heat transfer coefficient, the area and the driving force. Evaporation will proceed according to the mass transfer of water into the air from the body surface.

Most heat transfer texts will have a suitable relation for the heat transfer coefficient. If you cannot find any other one, use Incropera and DeWitt, Fun- damentals of Heat and Mass Transfer, 2nd ed. (John Wiley & Sons, New York, 1985). The Hilpert equation (Eq. 7.52 in that text) will be adequate.

Problem No. 52


HEALTH AND SAFETY CONCEPT: Toxic Exposure Control and Personal Protective Equipment

BACKROUND: One of the potential hazards from which employees must be protected is thermal burns of the skin. Burns are a frequent problem in foundries where molten metals must be poured because these materials often splash if incorrectly handled. For such cases, clothing must be available which will resist the hot metal and simultaneously protect the employee from s k i burns.

Another situation where employees need protection from burns is where hot items must be handled manually. For such cases, gloves are required. For high temperatures where cotton or other natural fabrics would fail due to thermal breakdown, asbestos was used for many years. Asbestos gloves do not have good abrasion resistance, so they often do not last very long. Also, there is a general reluctance to use asbestos because it is a possible contributor to lung disease, including cancer.

One modern alternative material is fabric made of polybenzimidazole (PBI). This fabric resists thermal degradation to quite high temperatures and is otherwise suitable for making relatively comfortable gloves or other clothing items.

The degree of discomfort andlor injury from contact burns depends on a complex relationship between time of contact and temperature of the contacting surface. However, if the skin surface temperature reaches 48"C, then significant discomfort at least, and probably tissue damage, will occur. Most people will be able to tolerate more high-temperature contact on parts of the hand than on other parts of the body, but the 48°C limit is a reasonably good guideline.

PROBLEM: A worker is wearing a glove made of a PBI material that is 4 mm thick. How long can he safely grasp an object that is at 300°C?

Assume that the object is of high conductivity (metal) and the surface remains at approximately 300°C. Assume that the glove is initially at 30°C and that the worker's s k i temperature will be the same as the inside surface of the glove.

The following properties may be assumed for the PBI fabric:

Thermal conductivity: 0.389 W/m K Density: 360 kg/m3 Heat capacity: 1298 J/kg K

Problem No. 53



BACKGROUND: Most organic chemicals are flammable and many chemicals are toxic. In order to prevent spread of hazardous liquids that might be spilled, storage tanks are usually enclosed by a dike large enough to contain the tank contents. Several tanks in a tank farm may be surrounded by a common dike.

While large spills of flammable liquids are rare, they sometimes occur. If a fire occurs following a spill of flammable liquid into the dike, the tank contents can begin to boil, and the vapor generated must be vented or the tank will fail with catastrophic results. Large tanks are designed to operate at a pressure only slightly above atmospheric pressure, so the fire emergency vents are designed to operate at low pressures for venting high rates of vapor flow due to heat input from a fire. (Tanks are also fitted with pressure and vacuum vents used for filling and emptying the tank and for ambient temperature and pressure changes. Those vents are too small to be used for emergency fire vents.) An alternative to installing large fire emergency vents is to build the tank with a weak roof seam so the roof seam will open and vent the tank in case of fire.

The rate of vapor generation, and therefore the venting rate, can be estimated from heat transfer calculations. Heat can be transferred to the tank walls by both radiation and convection from the fire that surrounds the tank. The rate of heat transfer through the tank walls to the liquid inside the tank is much greater than to the vapor because boiling heat transfer coefficients are much higher then vapor phase heat transfer coefficients. In fact, the temperature gradient through a tank wall is usually only a few percent of the difference between the flame temperature and the temperature of the boiling liquid in the tank. Thus, in estimating the boiloff rate, only the portion of the tank that is wetted by liquid needs to be taken into account. The vent designer must assume that the tankis filled to its maximum liquid level, because the fire may occur when the tank is full. The venting rate can be estimated from basic heat transfer considerations. Heat is transferred to the tank by radiation from the hot soot particles and gases in the flame and by convection from the hot gases. The radiation from the fire varies in intensity because it originates from both solid particles (soot) and from gases. Radiation from soot particles is similar in spectral distribution to blackbody radiation; that from hot gases is primarily in more narrow spectral bands. The complex nature of the radiation and the fluctuations of temperature within the fire make exact flame radiation calculations very difficult. Fortunately, it is possible to simplify the calculations. In fact, it has been found that the radiation flux can be averaged, and

PROBLEM NO. 53 103

that for many hydrocarbon fuels the radiant flux emitted by a flame is about 30,000 ~tu/hr-ft2. The average accounts for fluctuations in temperature, emissivity, composition, and other variables in the flame.

Convective fluxes can be estimated on the basis of natural convection heat transfer. Most of the surfaces of interest are large and vertical, so the appropriate correlations are those for vertical plane surfaces. The flame can be assumed to have transport properties equal to those of hot air because of the large amount of excess air in the flame and the fact that most of the gas in the flame is nitrogen. Flame temperatures may vary from fuel to fuel, but a temperature of 2200°F will provide a reasonable average value for most purposes.

PROBLEM: A tank is to be used for storing 10,000 bbl of toluene. The tank will be cylindrical and will be designed to API 650 standards with a cone roof. The tank walls are to be 27 ft high, and there is to be a maximum liquid level of 26 ft in the tank. For what venting rate should the tank be designed? How large a vent area will be required if the maximum tank pressure can be 10 in. of water gauge? Would you recommend a weak seam roof? A standard barrel contains 42 gal.

Problem No. 54

CHEMICAL ENGINEERING TOPIC: Heat Transfer: Radiation and Convection; Design

SAFETY AND HEALTH CONCEPT: Fire protection

BACKGROUND: Fires do not occur very frequently at chemical plants, but they do occur occasionally. Fires maybe caused by relatively minor failure of equipment or storage vessels in which a flammable material is spilled and then ignited. If the fire occurs in a process area where there is a substantial amount of process equipment, piping, valves, instruments, or the like. These pieces of equipment can fail, leading to propagation of fire from one plant area to another. In order to prevent this propagating damage from occurring, several types of fire protection equipment are provided.

The fire protection system and equipment can be divided into two large categories: passive and active. Active equipment includes such things as water sprays, foam, and dry chemicals. It requires that some action be taken, either by the plant operators and fire brigades or as a response by an automatic fire protection system. Passive fire protection equipment does not require any action at the time of the fire. It is designed and installed at the time the plant is built and remains passively in place until needed. Only routine maintenance is required to keep it operable.

One example of passive fire protection is insulating material (called fire-proofing) that is applied to steel structural members and equipment supports in the plant. The time required for unprotected steel supports to fail during a fire is very short. Fire-proofing can extend the failure time significantly and provide enough time for fire fighters to reach the scene, apply cooling water to the supports, and bring the fire under control.

Fire-proofing applied for structural fire protection is often provided only for the structural supports 30 ft above the fire base. Higher elevations can also be exposed to high heat fluxes. Individual judgments must be made in specific cases.

The heat transfer rate in a fire depends on two mechanisms: convection and radiation. Calculation of the heat transfer rate must be made by considering each of the mechanisms separately and then combining the result. If the fire is large, it will radiate at a constant flux; for most hydrocarbons and combustible chemicals, the radiant flux can be taken as 30,000 ~tu/hr-ft2. Most of the radiation inside fires is due to emission from hot carbon particles. In fact, the characteristic red-orange color of a flame is due to the emission from hot particles at visible wavelengths. There is also some band radiation emitted by hot gases at discrete wavelengths. The band radiation is usually invisible to humans, but can be detected by instru-

PROBLEM NO. 54 LO5

ments. The radiation from the hot particles is emitted along a continuum that approximates graybody radiation.

Convective heat transfer in a fire occurs at rates that can be estimated from standard convective correlations. In making the estimation, the properties of nitrogen can be used to approximate the flame properties. Flame temperatures vary considerably from place to place within the flame and as a function of time at a fured position in the flame. A reasonable average for convective heat transfer is 2200°F. The effective blackbody radiation temperature of a flame is much lower.

PROBLEM: An engineer has just arrived on a new job for which the process and piping layout has been completed. She is asked to determine whether fire-proofing will be required for the support structure for piping and equipment. One of the process vessels is a large cylindrical tank with dished ends that is mounted vertically. The tank is supported by a skirt at the bottom. The skirt surrounds the entire vessel, and totally encloses the space beneath the vessel. The figure below shows the cross-section of the bottom of the vessel and the supporting skirt. The

STEEL '1 SlPPORT 1

106 SAFETY, HW'I'H, AND LOSS PREVENTION IN CHEMICAL PROCESSES

supporting skirt will fail when its temperature reaches 1100°F. The engineer is told that at least 2 hr of fire resistance are required to assure that fire fighters have enough time to arrive and begin active fire protection. Ultimately, her decision on the kind and thickness of fire-proofing insulation to be used will depend on initial cost, maintenance cost, and the reliability of the insulation during its lifetime. She begins her study by assuming that the insulation has the thermal properties of cinder concrete.

a. If there is no insulation on the supporting skirt, how long will it be before the skirt fails? You may assume the steel skirt has a uniform temperature across its thickness, even though the temperature increases as the skirt is heated by the fire. The skirt diameter is large enough that it can be considered one-dimensional. The transport properties required for estimating the convective coefficient may be assumed to be constant, but should be taken at the film temperature. The fdm temperature willvary with time, souse an average temperature for the steelsupport half way between the initial temperature of 80°F and the flame temperature. Transport properties for the steel can be assumed constant at their ambient temperature properties. All the data required can be found in the background material or in the sixth edition of Perry's Chemical Engineers' Handbook.

b. What insulation thickness will be required to provide protection to the structural members for 2 hr? The temperature of the structural member must remain below 1100°F. Start your solution by writing a rigorous equation for the temperature gradient in the insulation. Specify the boundary conditions on both sides of the insulation and the initial condition. You will not be able to solve this set of equations unless you are a computer whiz and can do difficult numerical simulations. Even though you cannot solve the equations, the time required for the skirt to fail must still be estimated. Therefore, simplify the problem by making it into a pseudo-steady state problem. To do this, calculate the insulation surface temperature assuming that the steel temperature is its average value between ambient and the failure temperature. The exposed surface temperature will be a weak function of insulation thickness when thus calculated. Use the heat required to heat the steel to its failure temperature and the failure time to estimate the average heating rate. Then use the average heating rate to estimate the thickness of fire-proofing insuiation required.

Problem No. 55

CHEMICAL ENGINEERING TOPIC: Heat Transfer: Radiation

SAFETY AND HEALTH CONCEPT: Fire Protection: Separation

BACKGROUND: Many large storage tanks for flammable liquids are used in the chemical process industries. These tanks are usually enclosed by d i e s that are designed to contain the contents of the tank in case of spill or tank failure. If a large spill occurs and the liquid is ignited, a very large fire will result because the d i e area is very large. (Dikes may be several hundred feet long on each side. For example, a 6-ft high dike around a 100,000 bbl tank would have to be 300 ft2.) Such a large fire will emit radiation that can be transmitted through the atmosphere for long distances, and the radiant flux incident on surrounding buildings and equipment may be large enough to damage or destroy them. If people are exposed to the radiant energy, they may be killed or injured.

In order to protect people and equipment from the effects of thermal radiation, the storage areas are frequently separated from other equipment and from in- habited buildings. Remote impounding is also used. In remote impounding, the ground around the tank is graded so that any spill can be drained to an impounding area where the tanks are not exposed. In order to determine the separation distances required from the dike or remote impounding basin to other areas, an estimate of the radiant heat flux at any point around the fire must be made. The estimate is based on simple models of the fire and the transmission of radiant heat through the atmosphere.

The radiant heat transfer from a fire can be estimated from

q = a t F ~ a T 4

where q = absorbed radiant flux, ~tulhr-ft2 a = absorptivity of the surface where the radiation is incident, unitless z = transmissivity of the atmosphere, unitless E = emissivity of the flame, unitless

F = radiation view factor, unitless a = Stefan-Boltzmann constant, 0.1714(10~) ~ t u / h r - f t ~ - ' ~ ~ T = absolute temperature, OR

The absorptivity of the surface receiving the radiation is usually near 1 for nonmetallic materials. It depends on the wavelength of the incident radiation as well. Flame radiation contains both the continuum radiation characteristic of a blackbody or a graybody and the band radiation from gas emission. However, the


band emission from gases is primarily due to emission from water and carbon dioxide. That radiation can be absorbed by the atmosphere to a large extent. The atmospheric transmissivity depends on the wavelength of the radiation, as well, and the incident radiation maybe reducedsubstantially if the atmosphere is humid.

The flame emissivity is closely coupled to the flame temperature. In fact, some methods of measuring flame temperatures are actually based on the measurement of radiation fluxes from the flame. If a thermocouple or resistance bead ther- mometer is used to measure the flame temperature, the result will be in the range of 1800-2200°F for most fires. If the radiant flux from the fire is measured, it will be in the range of 30,000 to 45,000 ~tu/hr-ft2 for most hydrocarbon-based fuels. Thus, emissivities in the range of 0.3 to 0.7 can be expected.

A simplified method for calculating radiant fluxes is simply to combine the flame emissivity, the atmospheric transmissivity, and the flame temperature into a single term that represents the effective surface emittance of the flame. Then the absorbed radiant flux can be calculated as

The effective surface flux, qs, accounts for the effective surface flux from the flame. The incident flux at any location around the fire is then

where qs is an approximate value of the effective surface flux of the fire modified for the flame emissivity, atmospheric transmissivity, and flame temperature.

The radiation view factor, F, depends only on the geometry of the problem. It is a function of the flame size and shape, the distance to the location where the incident radiant flux is desired (the "target"), and the angles between the flame and the target. (Fis sometimes called thegeometricview factor or the angle factor.) The flame size and angle can be estimated based on the results of modeling the behavior of flames. Flame heights can be estimated from the work of Thomas (Ninth International Symposium on Combustion, Academic Press, New York, 1963), for example. He found that the height of a flame was given approximately by

L/D = 42(d,paQT)0'61 where

L = flame height, ft D = flame diameter, ft m = mass burning rate, lblsec pa = air density, lblft 3

g = gravitational acceleration, ft/sec 2

For circular or square dikes, the flame diameter is the dike diameter or length. The equivalent hydraulic diameter may be used for other dike shapes. The flame may

PROBLEM NO. 55 109

also be tilted by the wind. Welker and Sliepcevich (University of Oklahoma Research Institute Report No. OURI-1578-FR, Norman, OK, 1970) have found a method of predicting the flame angle. The flame may be tilted substantially if the wind velocity is high. Notice that the equation for flame height is written in dimensionless form, so any consistent system of units can be used.

Once the flame size and angle are determined, the radiation view factor can be determined from the information in standard radiant heat transfer texts. The view factor will be given in the form of an equation or a graph. The usual rules for view factor geometry apply to the view factors of flames just as they do to any similar geometric shape.

Once the incident radiant flux is known at some location near the fire, a judgment can be made as to whether it is tolerable for various activities or pieces of equipment. For example, the Code of Federal Regulations (Title 49, Part 193) specifies the acceptable radiant flux permitted at various locations around liquefied natural gas plants. The values specified are 10,000 Btu/hr-ft2 at the plant property line, 6700 Btu/hr-ft2 at streets near the plant, 4000 ~tulhr-ft2 for some buildings near the plant, and 1600 Btu/hr-ft2 (excluding solar radiation) for outdoor public areas such as playgrounds. There are specific rules more detailed than Sited here, of course, but the general idea is to protect the public from the danger of fire following a potential spill of liquefied natural gas.

Substantial damage to most plant equipment and buildings can be expected at incident fluxes of 10,000 Btu/hr-ft2, and at fluxes of 1600 Btuthr-ft2 human skin will receive second degree burns if exposed for about half a minute. Wood structures will be charred and damaged if exposed to radiant fluxes of 4000 Btu/hr-ft2 for long periods.

PROBLEM: A liquefied propane storage tank will have a net storage volume of 280,000 bbl. Storage is at atmospheric pressure and the normal boiling point of about -44°F. The tank is to be constructed in the vicinity of an existing plant and the safety department desires to keep the radiant flux from any potential fire in the dike surrounding the tank below 10,000 ~tu/hr-ft2 at any location within the existing facility. The dike to be built surrounding the tank cannot be more then 8 ft high. It has been found (Welker and Cavin, Report No. DOEEP-0042, U.S. Department of Energy, 1982) that the effective surface flux from propane fires is about 50,000 Btu/hr-ft . Estimate the minimum distance from the dike to the present facilities. You may assume that the burning rate of the fire is equal to a decrease in the fuel depth of 0.44 in./minute and that there is no wind.

It is also desired to keep the radiant flux from the fire below 1600 Btu/hr-ft2 (not including solar radiation) at an outdoor parking lot for the plant employees. How far must the d i e be located from the parking lot? In all your calculations, assume that the exposed surface receiving the radiation is vertical. For the no-wind conditions, the flame will also be vertical.

Problem No. 56

CHEMICAL ENGINEERINGTOPIC: Heat Transfer; Design; Momentum Trans- fer: Fluid Mechanics


BACKGROUND: Many chemical plant operations require that vessels be purged. There are several reasons. For example, when a plant is first constructed, the piping andvessels will be filled with air. The air may have to be purged because it interferes with the process or because it can lead to flammable mixtures with the chemicals within the piping or vessels. If piping or vessels containing a flammable or toxic material must be taken out of service and repaired or inspected, they must first be purged to remove the flammable or toxic material. Otherwise, workers will be exposed to the hazard of working under unsafe conditions.

Piping and vessels may also have to be cleaned following service. The piping or vessel will then have to be purged out of service, cleaned, and purged into service before the process can be continued. In some cases, steam is used both to purge the system and to clean it. For the present, consider only the purging step. It is assumed that most of the liquid or solid residue that might have been in a vessel has been removed. It is also assumed that the remaining vapor is either toxic or flammable. Thus, the vessel must be purged. Steam is available, and it is used to purge the vessel to remove all the hazardous material. The concentration level of hazardous material is monitored to assure it has all been removed and the steam supply is then stopped. At this point, the vessel is hot and full of steam, with no air or other noncondensible gas present. The vessel will gradually cool, and as it does, the steam will condense. When the steam condenses, the pressure in the tank will decrease. A vent is provided to supply air or nitrogen to prevent drawing excessive vacuum in the vessel. Otherwise the vessel will collapse when the maximum allowable vacuum in the tank is reached. Therefore, special care must be used when using steam as a purge gas.

PROBLEM: An uninsulated steel process tank is purged with steam before it is cleaned. The tank is 15 ft in diameter and 30 ft high, and it has walls and roof that are 0.25 in. thick. The tank will begin to collapse when the vacuum reaches 0.3 psig. What vent area is required to prevent vacuum collapse caused by steam conden- sation? The ambient temperature is 25°F and the wind is blowing at 20 mph. Consider two cases:

Case A: The steam used to purge the tank is saturated at 40 psig when it arrives at the valve entering the tank. After being expanded through the valve, the enthalpy

PROBLEM NO. 56 111

of the steam entering the tank will be the same as the enthlapy of saturated steam at 40 psig. The pressure in the tank will be 1.0 atm absolute. Thus, the steam in the tank will be superheated at 1.0 atm. The steam must be cooled to 212OF before it begins to condense. Assume the steam in the tank is at a uniform temperature.

Case B: Assume that the steam in the tank following purging is saturated at 1.0 atm pressure (and therefore at 212OF). In this case, the steam will begin condensing immediately.

In either case, the heat transfer through the bottom of the tank can be neglected, but heat transfer through the top and sides must both be considered. Make your venting calculations assuming that the venting rate required will be the rate calculated as soon as the steam is shut off. In both cases, start by calculating how long it will be before the tank begins to collapse. Collapse will start when the tank pressure is reduced to 0.3 psi below atmospheric pressure.

Problem No. 57


SAFETY AND HEALTH CONCEPT. Process Design

BACKGROUND: Many chemical engineering process designs require heating. The process heaters and furnaces may cover a wide range of sizes and be used for a variety of purposes, including generation of steam and direct heating of reactors. Heaters and furnaces must be designed to generate the required amount of heat and to transfer it to the place where it is needed. In addition, they must be designed to operate efficiently and safely. Most heaters use either gas or liquid fuels, but a signif~cant fraction use solid fuels. Some use a combination fuel, such as finely powdered coal suspended in oil.

Regardless of the type of fuel or the rate of heat generation, most heaters and furnaces have one thing in common: somewhere in the design, there is a fire. The fire may be small or large, depending on the rate of heat generation, but the burner temperature will nearly always be very hot. Temperatures inside the furnace or heater will be above 2000°F, and the walls of the furnace or heater will frequently be above 2000°F as well. Thus, the walls of the heater or furnace must be well insulated to minimize heat loss.

There is also another reason for insulating the walls. Generally, plant operators must be near the heaters or furnaces and they must be protected from the heat. Protection takes at least two considerations. First, the environment must be cool enough for operators to work in the area for whatever time is necessary. Second, even though the environment near the heater or furnace is cool, the actual surfaces may be hot enough to burn unprotected sk i .

PROBLEM: A furnace is designed to reach interior wall temperatures of 2500°F. It is lined with firebrick 4 in. thick having an average thermal conductivity of 0.2 Btu/hr-ft-OF. The firebrickis attached to asteel shell 0.25 in. thick. The temperature of the steel shell currently reaches 220°F, and workers must avoid the furnace because of the danger of being burned. A new engineer is asked to determine the thickness of a layer of insulation that will be applied to the outside of the steel shell toreduce the outside temperature to 120°F. She knows the temperature of the steel shell must not exceed 500°F because the steel will begin to weaken and the bonding agent used to attach the insulation to the steel will start to deteriorate. What thickness of external insulation with a thermal conductivity of 0.35 Btukr-ft-OF will be needed? By what percent will the heat loss from the furnace be reduced? The engineer assumes the temperature drop across the steel to be negligible because the steel layer is thin and the thermal conductivity of steel is about 25 Btu/hr-ft-OF,

PROBLEM NO. 57 113

which is about 100 times the value for the insulating materials. Once she has determined the thickness for the outer insulation layer, she calculates the temperature drop across the steel layer. Was her assumption of negligible temperature drop valid?

Problem No. 58


SAFETY AND HEALTH CONCEPT: Process Design

BACKGROUND: Although most materials that are processed in chemical plants remain in the same phase throughout the process, there are some materials and processes in which a change of phase occurs. Within the processes, a liquid may be boiled to form avapor, or avapor condensed to a liquid. Crystals may be formed from a liquid solution, and solids may be dissolved. If a change of phase occurs outside the control of the operators and in circumstances where no change of phase was anticipated in the plant design, serious problems can occur. For example, if liquid condenses in a l i e leading to a compressor and the liquid enters the compressor, severe damage can be caused. Similarly, cavitation in a pump can damage the pump. Most process equipment is designed to handle only a single phase, and is unable to handle two phases efficiently.

Another problem can be encountered as well. Freezing of liquids in piping or equipment can be a serious problem, particularly in colder climate. Sometimes, freezing of liquid in piping can result in serious safety hazards, such as might be encountered if cooling water flow was stopped for a critical process. Freezing might also lead to overpressure in piping as pumps try to force liquid through a pipe that is plugged by frozen material. One of the most common materials to be used, water, also expands when it freezes. When it does, piping and equipment can be damaged or destroyed.

Process piping is frequently "traced" with heating wires or heating coils to prevent freezing. Steam tracing or electrical tracing is frequently used. Electrical tracing has the advantage that it is not subject to freezing, whereas steam may condense and freeze ifflow is not kept at the proper rate. However, steam tracing may be cheaper if waste heat is available in the plant. In any case, careful attention should be given to the potential for freezing in piping, and steps should be taken to prevent it.

PROBLEM: Cyclohexane is used in a chemical processing plant. It is circulated through 1-in. schedule 40 steel piping. Flow may be interrupted at any time. To prevent freezing of the cyclohexane, the piping is insulated with a V2-in.-thick layer of insulation having a thermal conductivity of 0.23 Btu/hr-ft-OF. An electrical resistance wire between the insulation and the piping supplies heat to the piping to keep the temperature from decreasing below 50°F.

a. Assuming that the ambient temperature can drop to -lO°F, how many watts of electrical energy should be supplied to the piping per foot of length to keep the

PROBLEM NO. 58 115

temperature of the piping and cyclohexane from dropping below 45"F? Assume both the piping and the cyclohexane are at 45°F and no heat flows to either.

b. If the thermostatic switch that controls the electrical heating fails and the heating system fails to go on, how long will it take for the cyclohexane to begin to freeze? You may assume the insulation has negligible heat capacity and that the temperature of piping and cyclohexane are uniform throughout.

c. How long will it take for the cyclohexane to freeze completely? Assume there is no flow in the piping when the power goes off and that the piping is full of cyclohexane.

Problem No. 59


HEALTH AND SAFETY CONCEPT: Fire Protection Systems

BACKGROUND: Almost all commercial buildings, and certainly buildings on site at a chemical production facilitywill be equipped with some type of fire protection system. Components of the system range from alarms to sprinklers to COz-dispens- ing devices.

An automatic system, that is a system that would respond to a fire in the absence of human interaction, will be some type of device that will respond to a temperature rise or perhaps to smoke, as for example the "smoke alarm" that is frequently used in homes, schools, hotel rooms, and stores.

A relatively common type device is a sprinkler system that responds by flooding the room or area with a water spray in the event of a fire. These types of systems have been in use for many years, and are still widely used when the situation is one in which one would use water to extinguish a fire.

One of the oldest, yet more dependable ways the automatic sprinkler system is activated is by the melting of a "fusible link." Such a device is suggested by the sketch of a typical release mechanism shown in Figure 1.

The two elements of the link are held together by a low melting point soldering alloy, which, upon melting from the heat of a small fire, will release the lever arms, which then release the valve, allowing water to be sprayed on the fire.

The fusible links are also frequently used to provide door closing automatically so as to isolate fires.

There are a number of so called "fusible alloys," ranging in melting point from just under 500°F to less than 120°F. One typical alloy melts at 158°F and consists of 50% Bismuth, 26.7% lead, 13.3% tin, and 10% cadmium. In the problem that follows, you are asked to determine how long an alloy that melts at 158°F will take to respond to a sudden rise in temperature caused by a fire in the room where it is being used as the fusible alloy in a sprinkler system.

PROBLEM: A fusible link in a fire protection system responds to a fire by melting when its temperature reaches 158°F. When it melts, it releases a device similar to that shown in Figure 1, thus allowing the opening of the valve of a water sprinkling system. For the purposes of this problem, assume that if a fire breaks out in the room, the ventilation system will remove heat fast enough to keep the temperature from going above 220°F for several minutes.

If the room temperature reaches a temporarily steady220°F, how long will it take for the fusible link to melt if its initial temperature is 78"F?

PROBLEM NO. 59 117

In order to work this problem, you will need to know some properties of the materials and some conditions in the room.

Assume that the metal link which is to separate consists of two strips of brass, held together by a very small amount of the alloy. When the interface between the brass strips reaches 158"F, they will separate and release the valve. The amount of solder is small enough that the heat of fusion may be neglected.

Each of the brass strips is 314 iu. long, 114 in. wide and 1/16 in. thick. The properties of the brass may be taken as given below:

Density: 8500 kglm3 Heat capacity: 380 J/kgK Conductivity: 110 W/mK The heat transfer coefficient between the room and the link is 10 ~ t u l h r - f t ~ " ~ .

Force Exerted by Valve Body

Link

Pivot (held together

by fusible alloy1

Force Exerted by Water Pressure

Figure 1. Fusible link release mechanism.

Problem No. 60

CHEMICAL ENGINEERING TOPIC: Heat Transfer; Design


BACKGROUND: Almost all chemical processes require some storage of raw materials, intermediates, and finished products. The nature of chemical processes requires that the materials be reactive. Many are toxic and many are flammable; frequently they are both toxic and flammable. Modern practice usually follows the tenet that the quantity of material stored should be reduced to a minimum. Storing smaller quantities reduces the hazard in case an accidental release occurs. Smaller inventories also reduce the cost of handling the materials, both because smaller (and cheaper) storage tanks can be used and because money invested in inventory can be minimized.

The probability of tank rupture or leakage is very low. However, the consequences of a release can be very costly, both in terms of human injury or death and in terms of damage to the plant, damage to the surrounding neighborhood, and loss of production. When the consequence of a release is costly, special protection may be provided to prevent an accident from escalating from a small consequence to a catastrophe. Such protection may be relatively simple control systems, operating procedures, and other prevention methods. Control methods may include rupture disks and relief valves to vent tanks and process vessels to prevent failure. Proper design of the vessel is also required, along with proper testing before the vessel is put into service and proper maintenance during the service life of the vessel.

If a spill of flammable material occurs despite the efforts to prevent it, and if a fire occurs, fire protection systems may be provided to control or extinguish the fire and prevent it from causing the damage to propagate from one area to another. Design of fire protection and safety systems may require estimates of the potential heat transfer rates in a fire. If storage vessels are located close together, failure of one vessel can easily lead to failure of another, either through mechanical effects or fire damage. The original design of a storage tank will affect its behavior in a fire, and the design must be consistent with the intended use.

If a fire surrounds a tank, the heat transfer to the tank is due to both radiation and convection. The radiant heat transfer is due primarily to the hot soot particles in the fire, but there is also some radiation from hot gases. Heat transfer from the fire varies in intensity because the fire fluctuates in size, shape, and temperature. However, for a period of a minute or so, the average heat transfer rate can be estimated because the average flame properties can be used. Radiation from the soot in fires is similar in spectral distribution to blackbody radiation; radiation from hot gases occurs primarily in narrower spectral bands. While the exact calculation

PROBLEM NO. 60 119

of radiant emission is very complicated, it has been found that for many hydrocarbon fuels the radiant flux emitted by a large fire is an average of about 30,000 Btulhr-ft2. That average accounts for the average temperature, the average emissivity, composition of the flame, and flame turbulence.

Convective heat transfer rates can be estimated on the basis of natural convection heat transfer correlations. The flame is assumed to have transport properties equal to those of hot air because free-burning flames draw in excess air and most of the gas in the combustion zone is nitrogen. Hame temperatures vary from fuel to fuel, but for most purposes, a flame temperature of 2200°F will be a reasonable average. Note that the temperature suggested for convection heat transfer calculations is different than the effective blackbody temperature that would correspond to the radiant heat flux. A blackbody temperature of 2200°F would provide a radiant flux nearly three times that found in measurements for flames. The difference indicates that a flame must not behave as a blackbody.

Likewise, an object exposed to flame radiation may not absorb all the radiation that impinges on it. However, if the object is in the flame, it will collect soot on its surface, and the absorption of radiation will be quite efficient. Thus, it is usually acceptable to assume that an object surrounded by a fue absorbs all the radiation incident of it.

PROBLEM: Tanks fabricated for the storage of liquefied petroleum gas (LPG) are designed for a working pressure of 250 psig and a safety factor of four based on the tensile strength of the steel used in the tank. The dimensions of the tank depend on its design volume, but tanks are normally cylindrical. Their diameters are usually less than their length because the steel thickness required for the tank depends more on the tank diameter than the length. In this problem, you may assume the tank has a wall thickness of 0.625 in., which is the wall thickness for a tank about 10 ft in diameter. Suppose there is a fue and that the fue surrounds the LPG tank. Eventually, the temperature of the tank walls will be high enough to cause the steel to weaken. The temperature at which the tank walls will weaken enough to fail will be about 1200°F, which is the temperature at which the tensile strength is about one-quarter of its ambient temperature value. If the fire is not fully developed and flickers around the tank, the average radiant flux from the fire will be less than the 30,000 Btu/hr-ft2 discussed above. For this problem, assume the average incident radiant flux is 15,000 Btulhr-ft2. There will be convective heat transfer from the fire to the tank as well. The heat transfer coefficient can be estimated from natural convection correlations for heat transfer from hot air to a cooler surface. In this case, the heat transfer coefficient will be about 2 to 3 Btu/hr-ft2-OF. You may use a value of 2.5 Btu/hr-ft2-"F unless your instructor asks you to find a more precise value from one of the correlations. The absorptivity of the tank surface for flame radiation will be about 0.95, the same as its emissivity. You are to determine how long it will take the fue to heat the tank from ambient temperature, say 80°F, to the failure point.

120 SAFEI*Y, HEALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

You can start by writing an energybalance on the tank shell. You know that the shell will not heat very much if there is liquid in contact with the steel because boiling heat transfer coefficients are large and the heat will be transferred to the liquid. As the liquid boils, the tank pressure will rise, the tank vents will open, and the pressure will be relieved. However, if the steel of the vapor space in the tank is heated, the heat transfer rate to the vapor inside the tank will be quite slow, so the tank will begin to heat. Your energy balance should include terms for the radiant energy incident on the tank from the fire, the convective energy from the fire, the energy reradiated from the tank to the surroundings as the tank heats, and the energy to heat the tank itself. Assume the specific heat and density of the steel remain constant. Heat transfer to the interior of the tank will usually be small and can be neglected for our present purposes. When you've completed your energy balance, you will have a first order differential equation with one term linear in temperature and one term fourth order in temperature. It will have to be solved numerically. A simple Euler technique will be satisfactory, but you may also use a Runge-Kutta method or other method suggested by your instructor.

Problem No. 61



BACKGROUND: Anyone who watches newscasts on television or reads a daily newspaper is aware of frequent transportation accidents, which cause chemicals to be spilled into the environment. Very often, the chemical is one that evaporates at an appreciable rate, which then causes a vapor cloud that spreads either downwind or down gradient (downhill) of the spill site. There are a number of models available to predict the downwind concentration of the chemical if the emission rate is available. For an evaporating pool of chemical, the rate at which it emits vapor is, of course, the evaporation rate.

When the vapor cloud is at or near ambient temperature, the cloud will be neutral density and will behave as a tracer gas with the wind. However, if the cloud is quite dense compared to the air - due either to its being very cold or from the formation of aerosol droplets, which cause the cloud to be dense- the cloud will tend to move down gradient. In some cases, this down gradient movement will be against the wind. Either type of problem may occur, and will depend on the type of material being spilled, the conditions of storage, and the local topography and weather conditions.

The processes involve the release of chemicals, flash evaporation of superheated liquids, formation and growth of a liquid pool, transfer of heat to the pool by solar radiation, either cooling or heating by radiative exchange with the sky, either cooling or heating of the pool by convection with the ground or soil, convective heating or cooling by the air movement across the pool, and cooling by evaporation of the liquid.

The various heat transfer processes are coupled by an overall energy balance and the mass transfer process. The mass balance for the system is then solved simultaneously with the energy balance. Heat transfer and mass transfer models for the various processes are generally available in the literature.

Emission rate models for such processes are considered by Hanna, Guidelines forthe Use of Vapor Cloud Dispersion Models, published by the AIChE/CCPS. An emissions from spills model by Wu and Schroy is available from the Chemical Manufacturers Association.

The student will no doubt note that the process, taken as a whole, may be quite complicated. However, it is instructive to consider certain aspects of such a problem and find solutions based on some s impl ing assumptions and on some realistic zssumed conditions.


Except in rare instances, a pool of liquid that might remain after the initial period of active discharge from a leak or spill has terminated will be at a temperature less than its boiling point. An estimation of the actual temperature will normally require a rigorous solution of the complete coupled energy and mass balance equations. Nevertheless, many times it is observed that a pool of more volatile material will frequently exist below the ambient temperature. Indeed, we often find pools of water to be cooler than the surrounding air, even if there is a considerable influx of solar radiation. A still more volatile material, such as a typical organic solvent, might exist at a temperature well below the ambient.

PROBLEM: Given that a pool of benzene is contained within a diked reservoir following a spill from a storage tank. The diameter of the pool is large compared to its depth, which is 6 in. (0.15 m). At a time when the soil surface temperature is 60°F (15.6"C), and the pool temperature is 46°F (7A°C), estimate the rate at which the pool is being heated (the heat flux) from contact with the ground.

Note that this is a very much simplified model. In reality, the soil temperature will be changing with time, as will the pool temperature. A much more realistic model element for soil temperature variation is that of a semi-infmite solid.

The temperature of the pool of liquid should be considered constant throughout, and the convective heat transfer model appropriate to this case is that of natural convection.

Problem No. 62



BACKGROUND: We are all aware that various instances of equipment failure, or accidents, especially transportation accidents, frequently result in the release of volatile chemicals to the surroundiigs such that the chemicals evaporate, resulting in a release of vapor to the atmosphere. If the chemical is one that evaporates at an appreciable rate, the result will be a vapor cloud that spreads either downwind or down gradient (downhill) of the spill site. There are a number of models available to predict the downwind concentration of the chemical if the emission rate is available. For an evaporating pool of chemical, the rate at which it emits vapor is, of course, the evaporation rate.

When the vapor cloud is at or near ambient temperature, the cloud will be neutral density and will behave as a tracer gas with the wind. However, if the cloud is quite dense compared to the air- due either to its being very cold or from the formation of aerosol droplets, which cause the cloud to be dense - the cloud will tend to move down gradient. In some cases, this down gradient movement will be against the wind. Either type of problem may occur, and will depend on the type of material being spilled, the conditions of storage, and the local topography and weather conditions.



Emission rate models for such processes are considered by Hanna, Guidelines forthe Use of Vapor Cloud Dispersion Models, published by the AIChEICCPS. An emissions from spills model by Wu and Schroy is available from the Chemical Manufacturers Association.

The student will no doubt note that the process, taken as a whole, may be quite complicated. However, it is instructive to consider certain aspects of such a problem and find solutions based on some simplifying assumptions and on some realistic assumed conditions.

124 SAFETY, HEALTH, AND LOSS PREWENI'IONIN CHEMICALPROCESSES

Except in rare instances, a pool of liquid that might remain after the initial period of active discharge from a leak or spill has terminated will be at a temperature less than its boiling point. An estimation of the actual temperature will normally require a rigorous solution of the complete coupled energy and mass balance equations. Nevertheless, many times it is observed that a pool of more volatile material wvill frequently exist below the ambient temperature. Indeed, we often find pools of water to be cooler than the surrounding air, even if there is a considerable influx of solar radiation. A still more volatile material, such as a typical organic solvent, might exist at a temperature well below the ambient. In the following problem however, we consider a time period during a release when the liquid is at, or very near, its normal boiling point, having been cooled by flash evaporation upon exposure to the atmosphere.

PROBLEM: Ethylene oxide is a widely used industrial intermediate, being used for the manufacture of ethylene glycol (antifreeze) and a number of other com- mercially important products. It is also used as a disinfecting fumigant. Its normal boiling point is 10.7"C. If a discharge of ethylene oxide were to occur on a rather warm day when the ambient temperature is about 30°C, it would probably remain at about its boiling point during the course of the discharge, and would change temperature slowly for a short time thereafter.

Ethylene oxide is a relatively toxic material, as one might expect from the fact that it is used as a disinfectant. OSHA has established a permissible exposure limit (PEL) of 50 ppm as a time weighted average. It has a lower flammability limit of 3% in air, and is subject to detonation decomposition so that its upper explosive limit is 100%.

Assume that a spill of ethylene oxide is occurring from a ruptured containmernt vessel. As it is in contact with the soil inside a retaining dike, it boils at 1 atm. The soilis initially at a temperature of 32°C. It maybe assumed that the soil temperature is initially constant, but, of course it will change with time as the surface is cooled by the boiling liquid. Determine the time required for the soil surface temperature under the pool to drop to within OS°C of the boiling point of the liquid. You may assume the temperature of the pool and the heat transfer coefficient remain constant during this time.

The soil properties may be assumed to be as given below:

Thermal Conductivi : 0.52 W/m K 7 Density: 2050 kgfm Heat Capacity: 1840 J/kg K

The heat transfer coefficients for boiling heat transfer are usually fairly high, although it is not possible to predict them with a high degree of confidence. For the purposes of this problem, you may assume that the coefficient is XKIO w/m2 K.

Problem No. 63



BACKGROUND: We are all aware that various instances of equipment failure, or accidents, especially transportation accidents, frequently result in the release of volatile chemicals to the surroundings such that the chemicals evaporate, resulting in a release of vapor to the atmosphere. If the chemical is one that evaporates at an appreciable rate, the result will be a vapor cloud that spreads either downwind or down gradient (downbill) of the spill site. There are a number of models available to predict the downwind concentration of the chemical if the emission rate is available. For an evaporating pool of chemical, the rate at which it emits vapor is, of course, the evaporation rate.

When the vapor cloud is at or near ambient temperature, the cloud will be neutral density and will behave as a tracer gas with the wind. However, if the cloud is quite dense compared to the air - due either to its being very cold or from the formation of aerosol droplets, which cause the cloud to be dense - the cloud will tend to move down gradient. In some cases, this down gradient movement will be against the wind. Either type of problem may occur, and will depend on the type of material being spilled, the conditions of storage, and the local topography and weather conditions.


The various heat transfer processes are coupled by an overall energy balance and the mass transfer process. The mass balance for the system is then solved simultaneously with the energy balance. Heat transfer and mass transfer models . for the various processes are generally available in the literature.

Emission rate models for such processes are considered by Hanna, Guidelines for the Use of Vapor Cloud Dispersion Models, published by the AIChE/CCPS. An emissions from spills model by Wu and Schroy is available from the Chemical Manufacturers Association.

The student will no doubt note that the process, taken as a whole, may be quite complicated. However, it is instructive to consider certain aspects of such a problem and find solutions based on some simplifying assumptions and on some realistic assumed conditions.

126 SAFEI"Y, HEiALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

Except in rare instances, a pool of liquid that might remain after the initial period of active discharge from a leak or spill has terminated will be at a temperature less than its boiling point. An estimation of the actual temperature will normally require a rigorous solution of the complete coupled energy and mass balance equations. Nevertheless, many times it is observed that a pool sf more volatile material will frequently exist below the ambient temperature. Indeed, we often frnd pools of water to be cooler than the surrounding air, even if there is a considerable influx of solar radiation. A still more volatile material, such as a typical organic solvent, might exist at a temperature well below the ambient. In the following problem however, we consider a time period during a release when the liquid is at, or very near, its normal boiling point, having been cooled by flash evaporation upon exposure to the atmosphere. Since the pool is seen here as being in contact with relatively warmer soil, we will assume that it continues to boil during the time period under consideration in the problem.

PROBLEM: Monomethyl mine is an important intermediate in the manufacture of some pharmaceuticals, certain pesticides, surfactants, rubber chemicals, and a photographic developer. It is a strong irritant as well as a significant fire hazard. The lower flammable limit is 4.95% in air. The Federal Standard for industrial exposure (OSHA limit) is 10 ppm, time-weighted average for an 8-hr day. Upon heating, it will evolve even more toxic nitrogen oxides.

If a pool of methylamine forms from a discharge as a result of an accident, it will cause the moisture in the soil below the pool to freeze, because its boiling point is -6°C. We wish to determine what time would be required for the soil to freeze to a depth of 5 mm. If the soil under a pool of spilled liquid freezes, it will prevent the penetration of the liquid into the soil, thus preventing its further contamination of soil or, perhaps, of the groundwater in the region. However, the ultimate release of vapor to the air will be greater by the amount that might have otherwise gone into the soil.

You may neglect the heat of fusion of the moisture in the soil, although doing so does introduce an error if the soil were to have, for example, a moisture content of about 15% water, which would not be untypical.

You may assume that the soil is initially at a uniform temperature of 23°C. The properties of a typical soil are given below. Although you are to use these

properties in the solution of this problem, you should understand that such properties vary over a wide range of values for various soils.

Soil properties: Thermal Conductivity: 0.55 W/m K Heat Capacity 1100 J/kg K Density 2100 kglm3

PROBLEM NO. 63 127

Although you are to understand that boiling heat transfer coefficients cannot be predicted with high confidence because of the complexity of the process, you are to assume that the coefficient in this instance is quite high so that the surface temperature of the soil is approximately constant at the boiling point of the methylamine.

Problem No. 64



BACKGROUND: We have all read of spill accidents due to equipment failure, or accidents, especially transportation accidents, which result in the release of volatile chemicals to the surroundings. When the chemicals evaporate, the result is a release of vapor to the atmosphere. If the chemical is one that evaporates at an appreciable rate, the result maybe a flammable or toxicvapor cloud, which spreads either downwind or down gradient (downhill) of the spill site. Models are available to predict the downwind concentration of the chemical if the emission rate can be estimated. For an evaporating pool of chemical, the rate at which it emitsvapor is, of course, the evaporation rate.

When the vapor cloud is at or near ambient temperature, the cloud will be neutral density and will behave as a tracer gas with the wind. However, if the cloud is quite dense compared to the air- due either to its being very cold or from the formation of aerosol droplets which cause the cloud to be denseOthe cloud will tend to move down gradient. In some cases, this down gradient movement will be against the wind. Either type of problem may occur, and will depend on the type of material being spilled, the conditions of storage, and the local topography and weather conditions.

The processes involve the release of chemicals, flash evaporation of superheated liquids, formation and growth of a liquid pool, transfer of heat to the pool by solar radiation, either cooling or heating by radiative exchange with the sky, either cooling or heating of the pool by convection with the ground or soil, convective heating or cooling by the air movement across the pool, and cooling byevaporation of the liquid.


Emission rate models for such processes are considered by Hanna, Guidelines forthe Use of Vapor Cloud Dispersion Models, published by the AIChEICCPS. An emissions from spills model by Wu and Schroy is available from the Chemical Manufacturers Association.

The student will no doubt note that the process, taken as a whole, may be quite complicated. However, it is instructive to consider certain aspects of such a problem and find solutions based on some simplifying assumptions, and on some realistic assumed conditions.

PROBLEM NO. 64 129

In many instances, a pool of liquid that might remain after the initial period of active discharge from a leak or spill has terminated will be at a temperature less than its boiling point. An estimation of the actual temperature will normally require a rigorous solution of the complete coupled energy and mass balance equations. Common observations, however, suggest to us that a pool of evaporating liquid will often be at a temperature below the ambient. Indeed, we often find pools of water to be cooler than the surrounding air, even if there is a considerable influx of solar radiation. A still more volatile material, such as a typical organic solvent, might exist at a temperature well below the ambient.

If the liquid boiling point is well below the temperature of the surface upon which it is resting, it is possible that active boiling will take place for some time, until the surface is sufficiently cool to reduce the heat input rate from that source. Arigorous solution to this problem is complicated by the fact that the boiling coefficient will vary with the temperature of the soil which we would presume to be changing with time. Although the semi-infiite solid model is useful for a number of common boundary conditions, the variable coefficient condition makes a rigorous solution difficult.

In the present instance, we wish to consider an initial condition where it would be likely that active boilingwill be taking place, and we will attempt to estimate the magnitude of the boiling heat transfer coefficient between the surface on which the liquid rests and the boiling pool.

PROBLEM: Ammonia has a number of important industrial uses as an intermediate in the manufacture of fertilizers, dyes, plastics and fibers, as well as a number of other products. It is also widely used as a fertilizer in the anhydrous form. The OSHA limit for exposure is 50 ppm, time-weighted average over an 8-hr day. The American Conference of Governmental Industrial Hygienists (ACGIH) has set the level that is immediately dangerous to life and health (IDLH) at 500 PPm-

Consider that a distributor for anhydrous ammonia has a small storage facility that is filled from time to time from tank trucks. A storage tank at the facility is surrounded by a concrete pad with a dike that is high enough so that the contents of the tank could be retained witbin the diked enclosure. The diked enclosure is 25 ft by 30 ft and is high enough to retain the contents of the storage tank.

Now a tank truck, somewhat carelessly operated, has just backed into the tank and has not only ruptured the tank, but has also damaged the delivery valve on the truck. As a consequence, liquid ammonia has nearly filled the diked area and is rapidly boiliug, even though it is not an especially hot day.

Estimate the release rate of ammonia vapor due to the heat input from the flat concrete surface, when the surface temperature of the slab is at -lS°C.

You may have some difficulty with this problem due to the need for data that are hard to find, if they exist at all. The main problem here is in estimating the boiling heat transfer coefficient. If you use a model such as the Rohsenow equation,


you may have to guess at some of the surface- and fluid-specific parameters. As an approximation, you could consider that the fluid-specific parameters might be about the same as water. The concrete surface is quite rough, so you might liken it to a roughened surface of some other material. If you use an equation such as the McNelly equation, you do not need some of these data, but the results are probably a little less reliable.

Problem No. 65

CHEMICAL ENGINEERING TOPIC: Mass Transfer; Fixed Bed Adsorption


BACKGROUND: Although respirators, which are devices that are worn over the face to prevent inhaling harmful material, are the "last line of defense" against exposure to airborne contaminants, they are nevertheless very important safety equipment items, and it is vital that they be used appropriately, in recognition of their limitations.

Chemical cartridge respirators provide protection against vapors and gases being inhaled. One type of device uses an adsorbent, such as charcoal, to adsorb organic vapors and thus to purify the air that the wearer inhales. The bed of charcoal will remove essentially all of the contaminant until breakthrough occurs, after which the concentration will rise very rapidly, and substantial exposure can result in a short time. When there is a significant chance that a breakthrough might occur, it would be preferable to use a supplied air respirator, where clean air is delivered to the worker, rather than having a device that purifies the air.

The analysis of the performance of an &purifying respirator can be done by the methods of analysis of any ftved bed adsorber. In this problem we will see what effect the concentration of contaminant has on the service life of an adsorption canister.

The generalized correlation of adsorption potential shows that the logarithm of amount adsorbed is linear with the function (TlV)[log ( fs 1 f )] over a useful range of values. In the above function, T is the temperature (K); Vis the molar volume of liquid at the normal boiling point (cm31g mole); fs is the saturated fugacity (approximate as vapor pressure); and f is the fugacity of the vapor (approximate as partial pressure).

PROBLEM: For a particular charcoal, we have data as follows for dichloropropane (DCP) adsorption capacity:

Amount Adsorbed (TfV)[loglo (fs If )I [cm3 (1iq)llOO g] (units above)

1.0 21 10.0 11


A respirator canister contains 75 g of this type of carbon, and tests have shown that it will allow breakthrough when 82% of the adsorbent is saturated. (Unused length is 18% of the total length of the packed section.)

Regulations permit charcoal canister, full-face-mask respirators at DCP concentrations up to 750 ppm. If a worker were using this respirator in a DCP concentration of 750 ppm when the temperature is 80°F, how long might it be before there is a breakthrough? Assume that the worker breathes at a rate of 45 L/&.

If, due to an accident, a worker is caught in a DCP concentration of 2000 ppm, how long might he have before breakthrough?

Note: A simple adsorbent canister respirator is not adequate for the 2000 ppm level, which has been established as the level that is immediately dangerous to life and health (IDHL). One would choose to use such a device only in an emergency when nothing better were available to assist in escaping the exposure. For such cases, a self-contained breathing apparatus (SCBA) would be preferred. The SCBA is a device much like the SCUBA device worn by divers.

What would be the consequences if the respirator had been used a few days earlier and the cartridge had not been changed?

Suppose the respirator were to be used in an atmosphere that contained hydrogen cyanide (HCN) at 750 ppm? HCN will not be so readily adsorbed by the carbon cartridge. Assume the adsorption capacity to be only about 5% of that for DCP and recompute the time a worker would have in such an atmosphere.

The IDHL for HCN is 50 ppm. It is extremely toxic in acute exposure, and death can result from breathing concentrations that are not detectable by human olfactory senses. That is, by the time the victim smells the gas, it is probably too late to avoid death. For this reason, the air-purifying respirator is not suitable for use in the HCN-containing atmospheres, and either supplied air or SCBA respirators are used for protection against HCN.

Problem No. 66

CHEMICAL ENGINEERING TOPIC: Mass Transfer, Diffusion Through Solids


BACKGROUND: Chemical protective clothing (CPC) is used as a backup to engineering controls to protect workers from exposure to toxic materials that may be encountered. In most cases engineering controls, which are controls that keep harmful agents out of the workplace environment, make the workplace safe without other protection. However, there are inevitably instances in which some protective clothing is needed, including work at hazardous waste disposal sites.

Typically, CPC is made of some kind of polymer membrane, but all such materials are permeable to some extent, and selection is often made on the basis of the permeabiity of a particular polymer membrane to particular chemicals. The permeability of a polymer depends upon the diffusivity of the chemical in the polymer matrix and upon the solubility of the chemical in the polymer. The solubility of the chemical in the polymer is the upper limit of the concentration in the polymer, so that the solubility becomes the driving force for permeation through the membrane.

Other properties of importance include the tensile strength, since it is important that the clothing material resist tearing, and ease with which seams can be made to resist leakage. Of course, it is also important that the material allow for comfort and freedom of movement. The selection of appropriate protective materials and clothing is crucial since there is no backup in the event of failure.

PROBLEM: Methylene chloride is a common ingredient of paint removers, so it sometimes comes in contact with the skin. Besides being an irritant, it also may be absorbed through the skin where it may add with the larger potential exposure frcm inhalation.

Although there is little information available on the dermal dosage that could have immediate adverse effects, let us assume that we might deduce such a lower dose limit from airborne exposure limits. If we do this, we find that a dosage of about 1 or 2 g in 2 hr would be a maximum that should be allowed. Obviously, if one is using this material as a paint remover, and if hand operations are required, protective gloves should be worn. Consider the following situation:

Butyl rubber gloves, 0.04 cm thick, are being worn by an individual working with methylene chloride paint stripper. How long in continuous contact would be


required before one received a dosage of 1 g due to permeation through the glove material. The data needed for this problem are presented below.

Solubility in butyl rubber: 25% by weight (solute-free) 10 2 Diffusivity of solute in butyl rubber: 1.95 x 10- m 1s

Approximate surface area exposed b two hands: 0.08 m2 Density of butyl rubber: 1200 kg/m Y

Problem No. 67

CHEMICAL ENGINEERING TOPIC: Mass Transfer; Diffusion Through Solids


BACKGROUND: Chemical protective clothing (CPC) is used as a backup to engineering controls to protect workers from exposure to toxic materials that may be encountered. In most cases, engineering controls, which are controls that keep harmful agents out of the workplace environment, make the workplace safe without other protection. However, there are inevitably instances in which some protective clothing is needed, including work at hazardous waste disposal sites.

Typically, CPC is made of some kind of polymer membrane, but all such materials are permeable to some extent, and selection is often made on the basis of the permeability of a particular polymer membrane to particular chemicals. The permeability of a polymer depends on the diffusivity of the chemical in the polymer matrix and on the solubility of the chemical in the polymer. The solubility of the chemical in the polymer is the upper limit of the concentration in the polymer, so that the solubility becomes the driving force for permeation through the membrane.

Other properties of importance include the tensile strength, tear resistance, and the ease with which seams can be made to resist leakage. It is also important that the material allow for comfort and freedom of movement. The selection of appropriate protective materials and clothing is crucial since there is no backup in the event of failure.

PROBLEM: In one type of standard permeability test, the material being tested is clamped between two chambers, one holding the liquid chemical, the other being continuously swept with an inert gas. The permeation rate is determined by the change in concentration of the test chemical in the gas side.

One such test* with tetrachloroethylene permeating polyethylene showed a rate of 769.87 mg/cm2 min, whereas the rate through a sample of Teflon was 2.3 mg/cm2 min. If the solubility of tetrachloroethylene in polyethylene is 13% (by weight) and 0.02% in Teflon, estimate the diffusivity of tetrachloroethylene in these two polymers. The membrane thickness in each case was 0.01 cm. Report your results in m2/sec. The density of Teflon may be taken as 2200 kg/m3 and that of polyethylene as 930 kg/m3.

*The data are from Guidelines for Chemical Protective Clothing, 3rd ed., ACGIH, Cincinnati (Feb 1987).

Problem No. 68

CHEMICAL ENGINEERING TOPIC: Mass Transfer


BACKGROUND: The mention of the term PCB, which is an acronym for polychlorinated biphenyl, creates an image of extreme hazard in the minds of many persons. Actually, however, PCB does not present much hazard to humans in incidental contact.

Many chlorinated organic compounds display excellent properties for certain uses, and PCB was just such a case. It was widely used as a dielectric fluid in transformers and capacitors before its manufacture and use were discontinued and ultimately banned. It was also used as a heat transfer fluid in high-temperature service. Most of its uses followed from its exceptional chemical stability and inertness.

Its harmfulness derives from its tendency to "bioconcentrate." It isvery sparingly soluble in water and more dense than water, so that a quantity spilled in a waterway will sink to the bottom and remain to dissolve continuously over the course of many years. However, marine animals (fish, etc.) accumulate the chemical in their tissues at much higher concentration than in the surrounding water because when it is absorbed by the organism it does not readily metabolize. This property is shared by a number of inert, high-molecular-weight, sparingly soluble organic chemicals. Bioaccumulation seems to be a frequent property among chlorinated compounds.

In the case of PCB, the concentration in some fish organs will be several thousands times the concentration in the water in which the fish resides. At these levels, a variety of physiological harmful effects may come to the fish. Predators that feed on the fish will also accumulate the chemical. Such has been the case of some birds of prey, particularly osprey and eagles.

Transportation accidents have sometimes caused contamination of water bodies with bioaccumulating chemicals.

PROBLEM: Consider the following hypothetical accident: 1400 lb of PCB has just been spilled in a river because a rail car hauling a transformer in for replace- ment of the dielectric fluid was derailed while crossing a bridge. It appears that the chemical has collected in large pools on the bottom, over a combined area of about 150 ft2.

PROBLEM NO. 68 137

Estimate the rate at which the chemical will be released into the water, and how long it would require for 1% of the chemical to be dissolved. The following data may be assumed:

Solubility of PCB in water: 0.25 mg/L Mass transfer coefficient, liquid pool to water: 0.5 Ib mole/ft2 hr

PCB is a general term for a number of compounds with similar properties, and these were normally used as mixtures. Assume the average molecular weight in this case is 260.

Problem No. 69

CHEMICAL ENGINEERING TOPIC: Fundamentals; Mass Transfer


BACKGROUND: No matter how carefully workers do their jobs, there remains the possibility of accidents. The more planning and preparation that has gone into accident anticipation and contingency planning, the better the chance of avoiding complications, injury, or property damage if and when an accident occurs. In planning for possible accidents, one of the more likely occurrences might be a release of a large quantity of toxic or flammable vapors or gases. Methods are available for estimating the resulting concentrations from such releases; and from such estimates it is sometimes possible to predict what areas of a plant or of the area surrounding a plant might have to be evacuated.

As a useful guide in the matter of what concentration of a chemical is safe for persons to breathe, many people refer to the "Threshold Limit Values" (TLV) as published by the American Conference of Governmental Industrial Hygienists (ACGIH). The TLV is considered to be a safe upper limit for persons working in an industrial environment. However, for persons off site, the limit usually is considered to be much less for several reasons, among them are the fact that the affected population may be much more susceptible to injury. For example, young children, the very old, and expectant mothers are usually considered to be more susceptible to injury from chemical exposure. Therefore, a frequently used guide is to limit their exposure to no more than 1% of the TLV. These considerations are taken into account when making emergency preparedness plans for the community surrounding a chemical processing plant, and estimates of off-site exposure must be made for accidents that might be anticipated.

Some of the computational methods used for such emission sources as power plant stacks can also be applied to the estimation of the results of accidental releases. Such computations estimate the effect of dilution as a plume leaves the source location. A derivation of any of the methods is beyond the scope of this problem statement.

One of the simpler models to predict dispersion is called the "Gaussian Plume Model" and expresses the average concentration at a location downwind of a continuous source. A Gaussian plume model can be modified to express downwind concentrations of instantaneous releases as well. Such a model is sometimes referred to as a "puff' model. The cloud of gas from a puff release will travel downwind, dispersing somewhat as it travels. Thus, at downwind locations the

PROBLEM NO. 69 139

concentration becomes less as a function of the distance travelled and the degree of turbulence in the mixing process. An equation for a puff model is given below.

C = Q / [ ( ~ ~ ~ ) ~ ~ C J ~ U ~ ]

where C = the concentration at a selected point downwind, rng/m3 (time-weighted

average) s = Diffusion coefficient in thex direction (downwind), m cry = Diffusion coefficient in they direction (cross-wind), m a, = Diffusion coefficient in thez direction (vertical), m Q = Source strength (mass of gas or vapor making up the cloud), mg

This equation is valid for windblown puffs across fairly level ground and is based on an assumptions of how the components of the puffs would be dispersed. It predicts the maximum concentration within the cloud, that is, in the center. The diffusion coefficients depend upon the stability of the atmosphere and the distance downwind from the source. The diffusion coefficients are subject to a number of uncertainties, and disagreements from one author to another. For the present purposes, the values of ax and uymay be taken as equal, but the value of the vertical dispersion coefficient (a,) is different. These values may be estimated from the following abbreviated listing.

Unstable atmosphere (good mixing): u - 0.14?.~~; uZ = 0 . 5 2 2 . ~ ~ - 0 k0.92. Neutral atmosphere (Little mixing): uy = . , a, = o. I~xO.~O

Stable atmosphere (mixing suppressed): ay = 0.02-~~.*~; uZ = 0.05~'.~'

In the above equations, x is the downwind distance from the release in meters, and the a values are the dispersion coefficients, in meters.

PROBLEM: Emergency plans are being formulated so that rapid action can be taken in the event of an accident. It is predicted that if aparticular accident occurs, 1.0 kg of chlorine will be instantaneously released. There is a residential area 500 m away from the prospective release location. For a situation when there is a wind of 2 d s , blowing toward the residential area, estimate the time required for the gas cloud to arrive at the residences and the maximum concentration that would occur in the center of the cloud. How does this concentration compare with 1 percent of the TLV? The TLV for chlorine gas is 1 ppm (a molar ratio).

Determine the worst case situation, assuming the different stabilities presented above. Which case should we plan for?

(This problem is based on a problem in the text, Chemical Process Safety: Fundamentals with Applications, by D.A Crow1 and J.F Louvar, published by Prentice Hall, Englewood Cliffs, NJ.)

Problem No. 70

CHEMICAL ENGINEERING TOPIC: Mass Transfer; Design

HEALTH AND SAFETY CONCEPT: Pressure Relief Systems

BACKGROUND: There are a number of instances when pressure relief systems must be utilized to help ensure safe operation of chemical processing equipment by providing for venting of contents of equipment in the event the pressure exceeds safe limits. Several things come to mind immediately, such as pressure vessels, reactors, heaters, and storage tanks. Generally, if it is possible that a piece of equipment can become overpressured, then some device must be provided to vent the excessive pressure.

Not so obvious, perhaps, but nevertheless important, is the need to prevent underpressure, or unwanted vacuum. Generally, tanks, process vessels, and equipment items that are constructed to withstand only modest internal pressure, will collapse, perhaps with catastrophic consequences, if the pressure is suddenly reduced to below atmospheric. Devices to relieve under-pressure are generally called "vacuum breakers," but the principle is essentially the same, that is, venting is provided to prevent pressure swings of such magnitude that failure will result.

An interesting situation that requires avacuum breaker can arise in a distillation column. Unless the column is constructed to withstand a considerable internal pressure, it will be subject to possible collapse in the event that a vacuum is created within it, since, in most cases, cylindrical vessels will withstand much higher internal pressures than external pressures. If the steam to the reboiler should suddenly be interrupted, while the condenser continues to operate, and especially if a subcooled feed continues to enter, a serious drop in pressure can result that might cause a collapse of the column due to the creation of a vacuum within the column.

There are two general types of devices to relieve higher than desired internal pressures. The first is a device called a "pressure relief valve," which is so designed that it will open when a set pressure is reacbed. If the contents of the equipment are such that additional hazard would follow from a discharge to the atmosphere- and this is the usual case- then a piping system of some kind must be provided to move the vented material to a safe disposal location, a flare, or to a containment vessel. Sometimes, vent lines are to absorbers, where the venting material can be absorbed to prevent its loss to the atmosphere.

A second relief device is called a "rupture disk," which is, as its name suggests, a disk that is designed to burst when a dangerously high pressure is reached, thus preventing the entire equipment bursting in an uncontrolled release and/or explosion. Of course, the same need arises to prevent undesirable discharges to the

PROBLEM NO. 70 141

atmosphere. Rupture disks may be placed within a vent line in order to facilitate the containment of the vented material.

Distillation columns were mentioned above in connection withvacuum breakers. However, there is perhaps a greater possibility for overpressure in a distillation column. In the event of loss of coolant to the condenser, a situation can develop where the reboiler continues to operate and the pressure will rise rapidly. Also, in the case of a cold liquid feed to a distillation column, the feed may account for a considerable portion of the cooling that occurs in the equipment. Thus, loss of feed can also cause overpressure in such a case. A particularly serious problem might arise when there is loss of coolant and the operator shuts off the feed while attempting to bring the process back to control. If the feed is a subcooled liquid, then the pressure rise may be very rapid, indeed.

A situation similar to the loss of coolant mentioned in the paragraph above is considered in the following problem.

PROBLEM: A distillation column is separating a feed mixture of ethanol and water. The feed enters as a saturated liquid at an ethanol mole fraction of 0.035.

The feed rate is 50,000 lblhr, and the overhead composition is 0.83 mole fraction ethanol. Assume that the bottoms is about 0.99 mole fraction water.

The column normally operates at 1 atm as a nominal pressure. However, if the cooling to the condenser is lost, how long will it take for the pressure to rise to 1.5 atm? If the pressure must not rise above 1.5 atm, what venting rate in pounds per second will be required?

Note that there are a number of missing data items in the problem statement. These should be supplied by you if they are not supplied by your instructor. You are to make suitable engineering judgments regarding the height and diameter of the column, and such other items as might be required.

Actually, a rigorous solution to this problem is quite difficult, but some simplifying assumptions may be made. Your instructor may suggest some simplifications, you may wish to make some yourself. You might assume that there are no heat losses from the column, which seems quite reasonable. You will need to make some kind of realistic assumption about the rate of vapor generation that follows the upset. You could assume that the rate of vapor generation remains the same, and you could assume that the column temperature remains approximately the same, even though the pressure increases. Both of these would be what we would call "conservative" because the real rate of pressure rise would be less than estimated by using those assumptions. If you do not wish to assume the vaporization rate is the same, then you will have to estimate what effect the increased pressure (and consequently the increased boiling temperature) would have on the rate of vaporization. You can do this, of course, but you will need to know a typical, or probable value for the heat transfer coefficient in the reboiler. Then as the pressure increases, the boiling rate will decrease due to the decreased temperature driving force in the reboiler.

Problem No. 71



BACKGROUND: Anyone who watches newscasts on television or reads a daily newspaper is aware of frequent transportation accidents that cause chemicals to be spilled into the environment. Very often the chemical is one that wiil vaporize at an appreciable rate, which then causes a vapor cloud to spread downwind of the spill site.

There are a number of models available to predict the downwind concentration of the chemical if the emission rate is available. The better models take into account the vapor cloud density and certain aspects of the terrain. If the cloud is more dense than the air, it may move down gradient under the influence of gravity, and this could be against the wind in some cases. For an evaporating pool of chemical, the rate at which it emits vapor is, of course, the evaporation rate.

To a fairly good approximation, it is possible to predict evaporation rates for some cases from heat transfer and mass-transfer principles. The process involves coupled, simultaneous heat and mass transfer, in which the temperature of the pool is a function of the rate of heating by solar radiation, convective heating (or possibly cooling) from the soil, and convective heating or cooling by the air movement across the pool.

The heat input is then related by an energy balance over the surface of the pool to the heat of vaporization of the chemical that evaporates.

Finally, a simultaneous solution for the surface temperature of the pool, based on the premise that the heat gain from the surroundings is equal to the loss from evaporative cooling, will result in a pool surface temperature from which a mass transfer rate can be evaluated.

Emission rate models for such processes are considered by Hanna, Guidelines for the Use of Vapor Cloud Dispersion Models, published by the AIChEICCPS. An emissions from spills model by Wu and Schroy is available from the Chemical Manufacturers Association.

The student will no doubt note that the process, taken as a whole, may be quite complicated. However, it is instructive to consider certain aspects of such a problem and find solutions based on some simplifying assumptions, and on some realistic assumed conditions.

Our common observations tell us that if a spilled chemical is fairly volatile, as is the case for a number of organic solvents, the surface temperature of an evaporating pool might be well below the ambient temperature. Thus, we would anticipate

PROBLEM NO. 71 143

that a benzene spill, as suggested in the following problem would be at a temperature below the ambient.

PROBLEM: A tank truck hauling benzene has overturned, and a large pool of benzene has formed on the ground. The terrain is fairly flat, and the benzene has spread into a pool that is approximately 20 m in diameter. The wind is blowing across the pool at a velocity of 7 mls.

It is a clear, warm day, with midafternoon temperatures at about 30°C. Because the soil is still wet from the overnight rain, it is probable that not much of the benzene will soak into the ground. This is no doubt fortunate, since it is less likely, therefore, that the benzene will contaminate the groundwater.

Benzene is considered to be a carcinogen, and worker exposure is to be limited to no more than 1 ppm, as a time-weighted average by OSHA regulations. It is therefore of considerable concern in the present instance, because some populated areas lie downwind of this spill. Generally, it is considered that the general population should not be exposed to the OSHA limit, because the members of the general population may not be as resistant as workers in good health.

We wish to estimate the rate of evaporation. We will guess that the temperature of the benzene pool will be about 18°C.

To work this problem, use a flat plate, turbulent boundary layer model for heat transfer and then estimate the mass transfer coefficient from the Chilton-Colburn analogy.

Your instructor may elect to provide some of the property data for your use with this problem.

Problem No. 72

CHEMICAL ENGINEERING TOPIC: Mass Transfer; Design

SAFETY AND HEALTH CONCEPT: Storage, Handling, and Transport

BACKGROUND: Part of any chemical manufacturing operation is to provide an emergency operations plan that details the procedures to be followed if hazardous materials are released from confinement. Two of the major chemical properties that must be accounted for are the toxicity of the chemicals and their flammability. Some materials are both toxic and flammable, but the concentrations that are toxic when the chemical is released to the atmosphere are usually much lower than the flammable concentrations. Storage of hazardous materials should be in the minimum quantities required for the process and the transport and storage facilities should be designed to prevent any leaks. The transport and storage systems should include spill detection systems and methods to shut down transfer piping if any leak is detected.

If the chemical is a liquid, one of the first considerations is to determine the rate at which the chemical will evaporate and enter the atmosphere. Seepage into the soil and drainage into streams are also potential problems; they are usually more long-term considerations. If toxic or flammable material enters the atmosphere, it will be carried outside the plant in a short time. Toxic or flammable materials may be dangerous to either the workers in the plant or to the neighbors if it enters the atmosphere.

Consider the case where a hazardous liquid is spilled. For it to be an immediate threat it must be vaporized into the atmosphere. Its evaporation rate must be known in order to estimate the potential danger to workers or the public. It is unusual to be able to specify exactly how fast a liquid will evaporate if it is spilled because the physical layout of the plant, the size of the pool formed by the leaking liquid, and the atmospheric conditions cannot be forecast exactly. Most data for mass transfer have been taken in laboratories under carefully controlled conditions that do not exactly simulate those found outdoors. However, nontoxic materials might be used in tests on the plant site. Those test results could then be used to predict the evaporation rates of toxic or flammable materials. Straightforward mass transfer correlations might be used for the purpose. For example, evaporation from an open pool could be correlated on the basis of the Sherwood number, the Reynolds number, and the Schmidt number. (The Sherwood number is frequently

PROBLEM NO. 72

called the Nusselt number for mass transfer.) They are defined as follows:

Re = Lvph Sh = k c L / D a and Sc = p l p D a

where L = dimension of the pool in the direction parallel to the wind v = wind velocity

p = air density p = air viscosity

DAB = didfusivity of vapor in the air kc = mass transfer coefficient.

Any system of units may be used, but all the units must be consistent so the Re, Sc, and Sh numbers are dimensionless. A correlation frequently used for determining mass trmsfer coefficients for calculating evaporation rates from spilled liquids is

but it is based on laboratory data where the air flow is smooth and the leading edge of the liquid pool is at the same elevation as the surroundings. In the plant, the wind will be gusty, and the topography may change the evaporation rates. The evaporation rates may also depend on the elevation of the liquid below the top of the impounding space in which it is collected. Data taken at a plant site might be correlated in the same form as shown above. If such a correlation is attempted, it is unlikely that the Schmidt number will vary as much as the Reynolds number or the Sherwood number. In addition, the exponent on the Schmidt number is about 0.33 in a number of correlations, and would be expected to be 0.33 for a first approximation. Thus, the primary coefficients to be evaluated are the constant preceding the Reynolds number and the exponent for the Reynolds number.

PROBLEM: Acrolein is to be used as an intermediate material in a process. It is both flammable and toxic. The acrolein will be stored in tanks, and the possibility of a leak from the tanks is being considered as part of the emergency operations planning for the plant. In the event of a large spill from a storage tank, a drainage systemwill convey the spilled acrolein to a holding trench. The trench will hold the full contents of the tank. It will be 100 ft long and 5 ft wide at the surface. Several tests have been run using water as the evaporating fluid in the trench to determine what evaporation rates might be expected in the event of a spill. The trench was filled with water and the rate at which the liquid surface receded was measured. During the measurements, the pool temperature was recorded, and the tests were run when the wind direction was across the short dimension of the pool and along the long dimension of the pool. The air temperature and humidity were also measured.

The table on the next page includes a summary of the data obtained.


Wind speed Temperatures, O F L Rate (mfir> Air Dew Water (ft) (lb/hr-ft2)

As the new engineer at the plant, you are asked to correlate the data so that the correlation can be used to estimate the evaporation rate of acrolein if a spill occurs at the plant. You have specifically been asked to estimate the acrolein evaporation rate for a wind speed of 5 mph both across the trench and along the trench. The information is needed for determining the concentrations of acrolein that might be found downwind of a spill. A separate atmospheric dispersion model will be used for that purpose. You may assume the air temperature to be 90°F and the liquid acrolein temperature to be 70°F at the time of the spill.

Problem No. 73

CHEMICAL ENGINEER11 gG TOPIC: Mass Trans fer

SAFETYAND HEALTH CONCEPT: Hazardous Waste Generation and Disposal

BACKGROUND: Most chemical manufacturing operations involve the generation, use, and disposal of substances that are hazardous because of their toxicity or flammability. In many cases, the hazardous substance is a chemical that is widely used and not considered dangerous. For example, common salt is used in many processes without causing any particular problems with toxicity. However, if salt is contained in water that is discharged to streams, it may cause irreversible damage to the environment, including both plant and animal life. Thus, while salt is not a particularly hazardous material to handle and to use in processing, it must be kept out of the water that is discharged from a plant. Other substances may be much more toxic, of course, and they must be kept from the environment as well.

The nature of the substances and their effects on the environment are sometimes known quite well, but sometimes it is only known that they have a toxic effect. Sometimes the toxiceffects are inferred for one material through comparison with another similar material. For example, if we know that high concentrations of sodium chloride are toxic to certain plants in the ecosystem, we can infer that high concentrations of potassium chloride will also be toxic to the same plants. (Of course, we must keep in mind that the same substances may be required for life. Sodium and potassium salts are necessary for proper cell growth and reproduction; and potassium deficiencies in particular are encountered in nature. We frequently add potassium salts in the form of fertilizers to enhance plant growth.)

We cannot add large quantities of most substances to the ecosystem without damage. Thus, the kind and amount of materials present in water streams discharged from a plant must be closely monitored to make certain that the discharged water will be within the quality standards specified by the Environmental Protec- tion Agency. The standards for the purity of discharged water depend on the substances in the water and the potential damage to the environment. Specific concentrations can be obtained from the Environmental Protection Agency.

The specific method used to remove hazardous substances from water depends on the nature of the hazardous substance and the concentrations involved. Several different methods can be used. The following problem illustrates one method that might be used.

PROBLEM: The water used in a chemical processing plant contains a mixture of organic acids. The acids are not particularly toxic to the environment in small concentrations, but the concentration in the process water stream is 0.5% by weight


leaving the process, and that concentration is too large for discharge. The acids cannot be used in the plant for any other purpose, and there is insufficient acid for their recovery, purification, and sale. The acids are soluble in hydrocarbons, and it is decided to extract the acids from the water streams by countercurrent liquid-liquid extraction, then use the hydrocarbon as a fuel for one of the plant process heaters. There is no chlorine, nitrogen, or sulfur in the acids, so when they are burned with the fuel, they will not contribute any additional pollution to the air. The hydrocarbon used for the extraction is the fuel oil for the process heaters. Equilibrium data for the acid-water-oil system are given in the table below. Plot the data on a triangular diagram. Determine how much oil will be required to reduce the concentration of acid from the 0.5 mass percent in the feed to 0.05 mass percent, which has been found to be acceptable for discharge. The oil rate used in the process will be 1.5 times the minimum, and the water to be treated will enter the extraction system at a rate of 3500 gallons per day. The oil has a specificgravity of 0.88. Assume the process is to be performed in a countercurrent liquid-liquid extractor having an overall efficiency of 20%. How many stages will be required for the extractor?

Mass percent in water layer Mass percent in oil layer Acid Water Oil Acid Water Oil

Problem No. 74


SAFETY AND HEALTH CONCEPT: Toxic Exposure Control

BACKGROUND: Some chemicals such as benzene, formaldehyde, and asbestos have been implicated as carcinogins. Their use in commerce has been more strictly controlled since their carcinogenic potential was discovered.

As materials are found to be hazardous, precautions must be taken to keep them from entering the workplace and the surrounding neighborhoods. Several processes have been devised to remove hazardous materials from the atmosphere and water. The procedure best suited to a given use might depend on the rate at which the hazardous substance must be removed from a process. In the following problem, you are asked to estimate the rate of discharge for a pollutant that must diffuse to the surface of a porous wall before it can be removed.

PROBLEM: A refrigerated food warehouse is insulated with a layer of urea-formaldehyde foam 18 in. thick. There is some residual formaldehyde in the foam, and it has a concentration of 0.04 lb/ft3 of foam when the material is manufactured. Formaldehyde can diffuse into the air where workers will be exposed to the vapor. The outside of the foam layer is covered with an impermeable layer to prevent moisture migration, so the formaldehyde can only diffuse in one direction, which is into the warehouse. Ventilation for removal of the formaldehyde is not practical because the refrigeration system cannot handle the cooling load imposed by bringing in fresh, warm air. There is a system available that will remove 0.1 lb/hr of formaldehyde from the air during recirculation through the refrigeration system. It will return the air to the warehouse with essentially no formaldehyde, and you can assume the concentration of formaldehyde to be zero in the warehouse as long as the removal system is not overloaded. The warehouse has 24,000 ft2 of insulated surface area and the effective diffusivity of formaldehyde through the foam is 1.5(10-~) cm2/sec at the average temperature in the system. How long will it be before the formaldehyde removal system will be able to keep the air free of formaldehyde? The warehouse must be freely ventilated until that time, so it cannot be used for cold storage. How much formaldehyde must the system be able to remove if the warehouse is to be used within 1 month after the manufacture of the foam?

Problem No. 75


SAFETYAND HEALTH CONCEPT: Hazardous Waste Generation and Disposal

BACKGROUND: Many chemical operations utilize hazardous materials which are processed to form products useful to society. Sometimes the hazardous material is a reactant or an intermediate in the process, and sometimes it is an unwanted byproduct of a reaction. Hazardous materials are also used as solvents or carriers. Regardless of the reason, if a hazardous material appears in an operation, it must be removed from any process stream that is discharged to the environment. Workers and people who live nearby must also be protected from the hazardous effects.

The Occupational Safety and Health Administration (OSHA) has promulgated standards for exposure to many of the substances used in various chemical operations. The Permissible Exposure Limit (PEL) has been defined as the concentration of a hazardous material in the air that can be tolerated on an 8-hr day, 40-hr week basis during the worker's lifetime. The PEL is based on a time-weighted average. OSHA has also established ceiling concentrations that cannot be exceeded at any time. These ceiling values are higher than the PEL, and represent a more dangerous level for exposed people.

Another level, called the IDLH level (Immediately Dangerous to Life or Health), has been established for many substances as well. It represents the level of exposure from which a person can escape within 30 min without experiencing escape-impairing or irreversible health effects. IDLH levels are not specified for potential human carcinogens.

If toxic substances are used in chemical processing plants, care must be taken to assure that both workers and the environment are not exposed to adverse effects. If toxic wastes are generated, they must be removed from waste streams before the waste is discarded. Wastes maybe either gases, liquids, or solids, and the methods of treating or disposing of the waste depends not only on the phase in which the waste exists, but also on the waste itself. Generally, if the waste is a gas or volatile liquid, any air used in processing must be purified before it can be discharged to the atmosphere. The problem that follows indicates one way in which the purification might be accomplished.

PROBLEM: Benzene is present in an air stream in a concentration of 0.05 mole percent. The concentration is to be reduced to 10 ppm (by volume) by absorbing the benzene in a light oil. The air stream will be fed to the absorption column at 1.6 m3/sec and the system operates at 25OC and 1.0 atm. The superficial velocity of

PROBLEM NO. 75 151

the air stream in the column is not to exceed 0.9 mlsec. The tower packing has a gas phase mass transfer coefficient (FG) of 2.4(10-~) kmoles/m2-sec and a liquid phase mass transfer coefficient (FL) of %(lo4) kmoles/m2-sec. The packing area

2 3 is 49.2 m /m . For the dilute solutions and low pressures involved in the process, you may assume that Raoult's law applies. The average molecular weight of the oil is 240, and your column design should be based on an oil rate twice the minimum required rate. Determine the tower diameter and the depth of packing to be used.

Problem No. 76


SAFETY AND HEALTH CONCEPT. Process Control

BACKGROUND: Process control is an important part of chemical processing. With proper control, the quality of product is improved, and production rates will be more constant. Without proper control, reactions may proceed too fast and damage equipment. Workers may be injured or killed by contact with toxic substances or through explosions and fues. Proper process control requires instruments to measure and control the process variables. The instruments must be designed for the property being measured. Usually, thermocouples or resistance temperature detectors will be used for temperature measurements, and pressure transducers and switches will be used for pressure measurements. Their design and location need to be considered carefully to assure they will operate properly throughout the life of the plant.

It is frequently necessary to measure the composition of material in processing systems. Many different instruments and techniques have been devised to measure system composition. Some measure composition changes quickly and accurately, and others require more time for a careful analysis. Composition measurements may be based on periodic samples or on continuous measurements; but whichever method is selected, it must be chosen with the idea in mind that the measurement will be accurate enough for the intended purpose and that the response will be quick enough for adjustments to be made in control parameters. Composition measurements are frequently more difficult to make and are more unreliable than either temperature or pressure measurements.

Composition measurements are based on a number of physical and chemical properties of the material being measured. One method widely used for the measurement of concentrations of flammable gases and vapors in the atmosphere is to detect the temperature change on the surface of a small bead coated with a catalyst. The flammable gas or vapor diffuses through a layer of porous material until it reaches the catalyst surface, where it immediately reacts with oxygen from the air. The oxidation reaction generates heat, raising the temperature of the bead. The bead is part of an electrical bridge circuit, and the bridge becomes unbalanced when the bead is heated. The resulting voltage change is amplified and measured. The following problem is a simplified analysis of some of the design considerations for such a gas detector.

PROBLEM: An instrument is to be designed to measure the concentration of methane in air. Methane has a lower flammable limit concentration of 5 mole

PROBLEM NO. 76 153

percent in air, and the instrument is to be designed for a known response at steady state when the methane concentration is 25% of the lower flammable limit (the LFL for methane is 5.0 mole percent). Methane and oxygen diffuse through a thin porous coating to the catalyst surface, where they react immediately, producing carbon dioxide and water. It has been found that methane must diffuse to the catalyst surface at a rate of 0.1 g,/hr-cm2 to produce the desired heating rate at the catalyst surface. The porous coating for the catalyst surface has an effective diffusivity 36.5% of the molecular diffusivity. The surface temperature of the catalyst reaches 175°C at steady state and the ambient temperature is 25°C. How thick should the porous coating be to provide the correct diffusion rate? The reaction at the surface is

Problem No. 77


SAFETY AND HEALTH CONCEPT. Process Design

BACKGROUND: When hazardous materials are required for chemical processing, they must be very carefully controlled to keep them from entering the environment. Hazardous materials always have one of two general properties: they are flammable or toxic. Frequently, a hazardous material is both flammable and toxic, and the toxic concentrations are usually much lower than the flammable concentrations. Toxicity may further be divided into two other general groups, one including materials that are acutely toxic and whose effects show up within a short time of exposure. Other materials have more insidious effects; their damage to health becomes known only after long exposure. A single material may exhibit both acute and chronic toxicity, depending on concentrations and effects.

It is sometimes very hard to show that a material exhibits long-term health effects. Two good examples are the artificial sweeteners saccharin and cyclamates. Saccharin is generally accepted to be carcinogenic; however, it is still used in a variety of foods and cosmetics because there is no completely acceptable alternative. Some years ago, cyclamates were widely used as artificial sweeteners; however, some evidence seemed to show that cyclamates were carcinogens, and they were banned for use in the United States. Data released in the spring of 1989 indicate that cyclamates are not the carcinogens they were once thought to be, and they may again be allowed to be used for artificial sweeteners.

It is usually much easier to determine a material's flammability characteristics because no living organism needs to be tested to find the properties. For example, the lower flammable limit (LFL) is the lowest concentration at which a vapor or gas will ignite in air. Although the LFL depends on the system temperature and to some lesser extent pressure, the most obvious need is for data at ordinary ambient conditions. Thus, defining LFLs at 1 atm and room temperature provides sufficient data to use for estimating danger of ignition. Such properties as the amount of energy released during combustion are also easily measured. An ignition temperature is more difficult to define and measure. For many materials, an ignition temperature measured as the temperature at which a stoichiometric mixture of flammable vapor and air begins a self-sustaining exothermic reaction has been determined. It is called the autoignition temperature (AIT) and serves as an approximate limit for determining when ignition will occur.

In order for a fire or deflagration to occur, there must be a fuel, an oxidizer, and an ignition source such as a spark or a temperature high enough for ignition. The ignition temperature is low enough for some materials that they will ignite at

PROBLEM NO. 77 155

ambient temperature or below when mixed. Such materials are said to be hyper- golic. Even materials that ignite only at temperatures several hundred degrees above ambient can be ignited by small amounts of energy, frequently less than 1 mJ. Most vapor-air mixtures can be ignited by the spark from a static discharge, for example. Because of the ease of ignition and the rapid combustion reaction that ensues, substantial rules and standards have been written to defme conditions for use of flammable materials. The standards of the National Fire Protection Association are available in most libraries. They form the basis for many laws governing use of flammable materials.

Whether a material is toxic or flammable or both, it must be kept well confined if it is to be used in commerce. The basic process design must account for such materials and assure they are either used during the processes or destroyed so they will not harm workers or the environment. Where it is possible, recovery and recycling of chemicals is the best way to keep them under control. If a material is recycled, it will never be discharged to become a potential pollutant.

PROBLEM: Benzene is used in a process as a solvent for a solid product, and it is dried from the solid at the end of the process. Since benzene is quite flammable (its LFL is 1.3%) and toxic (its permissible exposure limit is 10 ppm), nitrogen is recycled as a carrier gas during drying. Neither the nitrogen nor the benzene is ever to be released from the process. In order to recycle both the benzene (as a liquid solvent) and the nitrogen (as a carrier in the drying process), the benzene in the nitrogen is stripped out in a tray absorber. The benzene entering the absorber is at a concentration of 7.4 mole percent in nitrogen. It must be reduced to a concentration of 0.4mole percent in nitrogen, after which the nitrogen stream will be heated and recycled to dry the product. The benzene will be absorbed in an oil having a molecular weight of 200. The oil enters the absorber at a rate of 0.5 moles of oil per mole of pure nitrogen entering the absorber. Raoult's law can be assumed to apply, and the absorber is designed to operate at 50°C (because the nitrogen- benzene stream entering is hot) and 1.0 atm. The vapor pressure of the oil is negligible, and the nitrogen can be assumed to be insoluble in the oil. Determine the mole fraction of benzene in the liquid leaving the absorption tower and the number of ideal trays required for the process.

Problem No. 78

CHEMICAL ENGINEERING TOPIC: Kinetics; First-Order Reaction

SAFETY AND HEALTH CONCEPT: Toxicology and Industrial Hygiene

BACKGROUND: Toxicology is the study of the effects of toxic materials on organisms, including humans. It is an area of knowledge that is becoming increas- ingly important to chemical engineers. One of the most important tenets of the toxicologist was stated in the sixteenth century by a physician-alchemist named Phillipus Aureolus Theophrastus Bombastus von Hoenheim (who came to be known as Paracelsus) when he was defending his use of mercury to treat a disease with the statement, "What is it that is not poison? AU things are poison and none are without poison. Only the dose determines that a thing is not poison." Various translations later, the last statement is, "The dose alone makes a poison."

So it is with toxicology. All things are toxic, but whether there is harm in exposure depends on the dosage. We are accustomed to thinking of materials as being u t ~ x i ~ 7 7 or "nontoxic," and perhaps the difference is in whether or not one is likely to receive a toxic dose in his or her ordinary conduct in the presence of the material. In any event, we distinguish between "toxic" and "hazardous," because the two are not synonymous.

When a potentially toxic material is taken into one's body, there will be natural processes (((elimination," such as exhaling the material as a vapor, or by passage through the kidneys, or "metabolism," which changes the chemical through reaction) that will tend to rid the body of the material. It is only when the defenses are overwhelmed that toxic effects wiII be seen.

PROBLEM: "Chronic exposure" means continued exposure on a day-to-day basis as might be the case for a worker exposed all the time while he is working. In the case of a particular hypothetical chemical agent, it has been determined that a human can tolerate a level of 10 mg of the agent per kilogram of body weight, that the metabolism is a first-order process, and the half-life for the agent in a human is 3 hr.

If it is assumed that all the agent that is inhaled is adsorbed by the body, what continuous concentration could a person with a body mass of 68 kg tolerate in the workroom air? Assume that the average breathing rate is 45 Llmin.

Note: The student should be aware that this is a hypothetical case and it would be unusual to find a simple model such as this to be adequate. In reality, metabolic rates and processes are not well known, and defining a safe dosage must frequently be done in other ways.

Problem No. 79

CHEMICAL ENGINEERING TOPIC: Kinetics

SAFETY AND HEALTH CONCEPT. Explosions: Runaway ReactionTemperature

BACKGROUND: Many common chemical reactions used in industry are exothermic, that is, they produce energy as the reaction proceeds. In most cases, the reactions are run in reasonably small reactors, and the amount of energy that is producedis small enough that the reaction canbe kept under control without much difficulty. However, in some reactions, particularly those run in large reactors, the rate at which energy is produced can become much larger than the rate at which the energy can be removed from the reaction vessel. Since the rate of reaction increases as the temperature increases, the increasing temperature is reinforced. The reaction rate increases very quickly, sometimes so quickly that the reaction gets out of control and the reactor is damaged or destroyed if it is not adequately vented or protected. In extreme cases, an explosion may occur. The rate at which energy is produced in a reaction can be written as

where Qg = rate of heat generation, caVs

AHr = heat of reaction, cal/g-mole ko = frequency factor, sec-I C = concentration of reactant, g-mole/cm 3

V = reactor volume, cm 3

E = activation energy, cal/g-mole R = gas law constant, cal/g-mole K T = absolute temperature, K If the reaction is exothermic, heat must be removed from the reactor to keep the

temperature from increasing out of control. The heat transfer rate can be written as

where Qr = rate of heat removal, caVsec U = overall heat transfer coefficient, cal/cm2 sec K A = heat transfer area, cm2 To = coolant temperature, K


If heat can be removed as fast as it is generated by the reaction, the reaction can be kept under control. Under steady state operating conditions, the heat transfer rate will equal the energygeneration rate. However, if the heat removal rate is less than the heat generation rate, a condition that might occur because of failure of a cooling water pump, for example, the temperature in the reactor will begin to rise. The net rate of heating of the reactor contents is the difference between Qg and Qr, or

Qn = Qg - Qr

If the net heating rate, Qn, is positive, the reaction will have an increasing temperature. If the rate of increase of Qn is positive, that is, if dQ JdT > 0, the reaction will have the potential to accelerate and become uncontrollable. The rise in temperature will increase both the heat transfer rate and the reaction rate, but the heat transfer rate is a linear function of temperature, and the reaction rate is an exponential function of temperature. If there is sufficient reactant in the reactor, the temperature will increase until the resulting pressure causes the reactor to fail.

PROBLEM

(a) Show that the critical temperature above which heat is generated faster than it can be removed is given by the solution of the equation

where Tc is the temperature at which heat is produced faster than it can be removed.

(b) Assume that a reaction occurs in a continuously stirred tank reactor with the inlet and outlet flows controlled at steady and equal rates. Although the reaction occurs in the liquid phase, the heat of reaction is several times aslarge as the energy required to heat the reacting mixture to its boiling point and vaporize it. The reactant is pure as it enters the reactor, and the reaction goes essentially to completion. The heat exchanger can transfer all the heat from the reactor at a reactant feed rate that is 25% greater than the nominal or design rate. After a few years of operation, it is decided to increase the production rate of the product. The operators know that the product rate can be increased by increasing the reactant flow rate. In fact, on several occasions they have run the reactor at rates up to 20% more than the design rate and have experienced no problems. They also know that the reactor temperature is only a few degrees higher than the cooling water temperature, so they assume that the reactant can be fed at much higher rates, thereby increasing the production rates. The engineering team that designed the reactor is no longer at the plant, and the plant manager is new. He has had no experience with chemical reactions. The pump for the reactor is replaced with a

PROBLEM NO. 79 159

larger pump, and the product pump is also replaced to match. The cooling water pump is also replacedwith a larger model so that the cooling water rate is consistent with the reactant and product rates. The heat exchanger in the reactor is not changed, so the heat removal rate can be increased onlyby increasing the temperature in the reactor. The reaction has an activation energy of 30,OOO cal/g-mole and the coolingwater temperature is 20°C. What is the maximum temperature at which the reaction can operate without having the reaction run away?

(c) What actions might be taken to assure that the reactor will operate safely?

Problem No. 80

CHEMICAL ENGINEERING TOPIC: Kinetics: First-Order Kinetics

SAFETY AND HEALTH CONCEIYT: Toxicology and Industrial Hygiene: Biologi- cal Elimination of Toxic Chemical

BACKGROUND: Many chemicals have a toxic effect on humans. Some of them have acute toxicity, in which case the effect of a given concentration or exposure on the body can be sudden. Some have more long-term effects and take years to cause disease or death. Acommon example of a chemical that has immediate effect on the body is ethyl alcohol. In small amounts, ethyl alcohol does not harm humans. However, if the amount of alcohol ingested becomes large, the effects become noticeable. At first, the effects are primarily a slowing of physical reactions and a gradual loss of mental capacity. If the amount sf alcohol ingested over a short period is large, unconsciousness usually occurs. However, if the alcohol is ingested very rapidly, and in large quantities, it can kill. The effects of alcohol ingestion are usually reversible within a few hours or days, unless the amount ingested is very large. Some death of brain cells and some damage to the liver may result from large quantities ingested over a long period.

Tobacco smoke appears to have minor short-term effects once the body learns to tolerate the smoke. However, the long-term effects can be severe: smoking tobacco has been implicated as the a major factor in causing lung cancer in many countries. It has also been implicated as a factor in heart disease and lung diseases. The symptoms may take many years to appear.

These are only two examples of materials that might cause immediate or long-term health effects. In fact, if either tobacco or alcohol were being introduced into society under today's regulations, it is likely that both would be banned. It is certain that the short-term impairment of mental and physical abilities of alcohol would not be tolerated in a chemical plant (and, in fact, drinking on the job is forbidden). If a chemical is found in the workplace that has the long-term effects caused by smoking tobacco, it would be strictly regulated or forbidden. Many of the present rules limiting hazardous chemical exposures are the result of exposures that cause fewer health effects than tobacco. The chemical engineer must realize that rules will likely be in effect for most of the materials he or she may encounter in the workplace.

The elimination of toxic chemicals from the human body is a complicated procedure that may require the action of several organs. Mathematical modeling of the detoxification process is rarely simple or straightforward. However, in some cases simple models can help our understanding of the elimination process. One such simple model is to consider the rate at which the chemical is eliminated from

PROBLEM NO. 80 161

the body to depend on the concentration in the blood. If that model is assumed to represent the detoxification process, then

dcldt = -kc

where c is the concentration of toxic chemical in moles per liter, mole fraction, or other concentration units; t is the time in hours; and k is the "reaction rate constant," in hr-'

The rate of metabolism of the toxic chemical is thus modeled as a first-order chemical reaction. Of course, the real elimination process is much more complex than a first-order rate equation implies, but the equation does offer some insight into modeling of the elimination process.

PROBLEM: At a party held to celebrate the (successful) completion of a course in kinetics, a student imbibes too much spirits and becomes inebriated. His blood alcohol concentration is 0.21%. The student knows he shouldn't drive home until his blood alcohol concentration drops to 0.02%. If his blood alcohol drops to 0.17% in a half hour and he stops drinking at midnight, at what time can he go home? You may assume that the elimination of alcohol is a first-order process.

Problem No. 81

CHEMICAL ENGINEERING TOPIC: Kinetics; Heat Transfer

SAFETY AND HEALTH CONCEPT: Process Design

BACKGROUND: The problems involved in designing chemical reactors always require attention to both the design of the reactor itself and the reaction that is to be run. All reactions are either exothermic or endothermic, although sometimes the amount of energy used or released by the reaction is relatively small and is easily handled. In addition, the reaction rate is sensitive to the temperature of the system and an exothermic reaction can get out of control if it generates energy faster than the energy can be removed from the system. In practice, that means that the kinetics of the reaction must always be taken into account in determining the heat transfer rates and that the heat transfer rate must be accounted for in studying the reaction kinetics. In the laboratory, it is frequently quite easy to maintain a constant temperature during study of a reaction, but in commercial production, the larger quantities of materials make constant temperature more difficult to attain. The heat transfer from a reacting system will depend on the system geometry primarily as an area function, while the reaction rate (and thus the energy generation rate) will be primarily a function of volume.

If a reaction is to be run, one of the first choices to be made is whether it is to be a batch process or a continuous process. Continuous processes are preferred if the quantity of product is large. There are sometimes advantages in continuous process reactors in ease of control, as well. However, some processes cannot easily be made to operate properly under continuous conditions, and batch or semi-batch reactions are usedinstead. Careful control must be used in either case to avoid potential failure of the reactor. One of the most important things to avoid is a runaway reaction, in which the energy generated by the reaction is greater than the energy that can be removed from the reactor. Exothermic reactions may reach a temperature that limits the equilibrium conversion, of course, but that characteristic cannot be relied on for reaction control if the equilibrium temperature is too high for the system. For example, in a liquid phase reaction, as the temperature increases, the vapor pressure of the liquid increases, which may cause the pressure inside the reactor to become higher than the reactor design pressure. In a gas phase reaction, the pressure also increases as the temperature increases, and if the reaction generates more moles than are in the reactants, the pressure will rise even faster. The following problem illustrates some of the factors that have to be taken into account for a batch reactor design. Relief valves or rupture discs will be required for all reactors to provide added safety in the event of runaway reactions.

PROBLEM NO. 81 163

PROBLEM: A reaction is to be carried out in a stirred batch reactor. The reaction has the stoichiometry

and the reaction is second order and irreversible at any temperature encountered in practice. The reaction rate equation is

where - r ~ = reaction rate, g-moles of AIL-sec CA = concentration of A, g-moles/L CB = concentration of B, g-moles/L

k = reaction rate constant, Llg-mole-sec

The reaction rate constant is a function of the system temperature and is given by

k = ko exp- (fi) where

ko = pre-exponential factor, Llg-mole sec E = activation energy, cal/g-mole R = gas law constant, cal/g-mole K T = absolute temperature, K

The heat of reaction is -90 Wg-mole. The reactor to be used is cylindrical with a diameter of 5 ft and a height of 11 ft. It is well insulated on the top and bottom, and heat transfer through the top and bottom can be neglected. The reactor is jacketed, with heat transfer occurring through the jacket. The maximum working volume of the reactor is 1500 gal, and the jacket extends from the bottom of the reactor to 1.0 ft from the top. The reactants A and B are mixed in the reactor at 75°F to give a concentration of 5 g-moles/L for each at the start of the reaction cycle. The reactor is then heated to initiate the reaction. The heat is supplied by saturated steam at 50 psig. Once the reaction has begun, heat must be removed from the reactor. Heat is removed by cooling with water circulated through the reactor jacket. The cooling water has an average temperature of 75°F. The reactor is well-stirred, and you may assume the liquid in the reactor to be at a uniform temperature and composition at any time (although both will change with time). The reactor has a maximum working pressure of SO psig, so the temperature in the reactor must be kept low enough to keep the vapor pressure of the contents below 50 psig. You may assume the reacting mixture to have the vapor pressure of pure


water. For simplicity, you may also assume the specific heat of the reacting mixture to be constant and e ual to 1.0 caVg-OC. The density of the mixture may be assumed 9 constant at 1.0 g/cm . Heat transfer coefficients may also be assumed constant. For heating, the overall coefficient is 200 ~ t u / h r - f t ~ - ~ F and for cooling, the overall coefficient is 100 Btulhr-ft2-OF. The activation energy for the reaction is 30 kcallg- mole and the pre-exponential factor is l.0(10'~) Llg-mole-sec. You are to do the following:

1. Write the energy balance and kinetic reaction rate equations describing the system. These will involve both heat transfer and the reaction rate. The result will be two nonlinear time-dependent differential equations that can be solved simultaneously to give the temperature and concentration of reactants in the reactor as a function of time. You may assume the reactor contents are uniform throughout the volume of the reactor.

2. Solve the equations to determine if the reaction can be run to 99% conversion of A without exceeding the reactor design pressure. The pressure can be estimated as the vapor pressure of water at the system temperature. You will have to change the heat transfer temperature for the jacket when the steam heat is turned off and the cooling water begins to enter. The heating temperature in the jacket will be constant because you can use condensing steam; you may assume the average cooling water temperature will be constant. We use this simplifying assumption to make the solution easier, although in practice the coolant will heat as it passes through the reactor. In practice, the increase in temperature of the coolant must be taken into account.

3. You will have to solve the equations numerically. Since they will be reasonably simple first-order equations, you may use a technique such as the Euler method or the RungeKutta method. However, be sure your time increments are short enough to keep the temperature rise during each increment reasonable. A reasonable temperature rise for each time increment might be about 1.0 K.

4. Prepare a short table showing how the time for 99% conversion is affected by the temperature of the reaction at which heating is stopped and cooling is started. At what temperature would you begin cooling?

Problem No. 82


SAFETY AND HEALTH CONCEPT: Rupture Discs and Relief Valves

BACKGROUND: Whenever materials are contained in a process vessel, there is a possibility the vessel may rupture because of overpressure. Sometimes the vessel rupture will result in only a small amount of leakage of innocuous material, but sometimes a violent explosion may occur, with substantial destruction. In either case, the vessel will have to be repaired or replaced and any damage to the surroundings will have to be repaired. Workers may be killed or injured by an explosion, or may suffer health effects if the leak contains toxic materials. In order to reduce the probability of tank failure, relief systems must be supplied.

A relief system may incorporate either a rupture disc or a relief valve. Both have similar functions, although there are some important design and operational differences. A rupture disc is a simple device consisting of a thin disc sandwiched between two flanges. If the pressure on one side of the disc becomes greater than the disc can withstand, the disc ruptures, relieving the pressure. A rupture disc can be made to quite close tolerances, and rupture discs have been shown to be reliable for venting vessels and piping. One of the drawbacks to using rupture discs is that once the disc is broken, flow cannot be stopped, and the vessel will have its pressure reduced to the ambient pressure at the end of the discharge line. The rupture disc must also be carefully selected to assure it does not deteriorate because of corrosion and that changing temperature will not cause it to fail at the wrong pressure,

Relief valves are similar in design to ordinary valves except that rather than opening because a handle is turned, they have a spring-loaded stem that remains closed until the pressure in the system gets large enough to overcome the load of the spring. Then the valve will begin to open. A spring-loaded relief valve will usuallybe fully open when the system pressure reaches about 10 percent above the set point pressure. Relief valves have the advantage that they are designed to close once the pressure decreases below the set point. Thus, if the pressure rises again, the relief valve can open again to relieve the pressure. This intermittent operation of the relief valve minimizes the quantity of material discharged from the vessel. Of course, the discharge from the relief valve, like that from the rupture disc, must be directed to a safe location. If toxic or flammable materials are involved, they may have to be routed to a recovery system to eliminate the possibility of further danger.

The choice of what kind of relief device to use and the location for the relief device is based partly on the properties of the materials in the system and partly

166 SAFEI"Y, HEALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

on the level of venting needed. The choice should be integrated with the designers of other parts of the system and should be made in accordance with the standards of the applicable codes and regulations. (One general consideration is that relief valves are frequently used for process protection and rupture discs are frequently used for vessel protection.)

The relief valve design depends on the capacity it must have and on the material being vented. Recent studies sponsored through the Design Institute for Emergen- cy Relief Systems (DIERS) have made substantial progress in providing techniques for specifying relief valves and rupture discs based on the system properties and the material being vented. The paper by H. K Fauske, "Emergency Relief System Design for Reactive and Non-Reactive Systems: Extension of the DIERS Methodology," contained in the July 1988, issue of Plantloperations Progress provides a review of some of the results of the DIERS work and lists a number of more detailed references to the work.

In the following problem, you are to consider how the reaction rate in a gaseous reaction system changes as the pressure increases. The change in reaction rate can have a substantial effect on relief system design.

PROBLEM: The data contained in the following table were taken in the laboratory using a constant volume batch reactor initially charged with a pure reactant. Use the data to determine the volume required for a constant temperature, constant pressure plug flow reactor operated at 500°K and 10 atm absolute pressure. We wish to react an input stream containing 2.0 kmolelmin of pure reactant to give 95% conversion. You may assume that the reactant and the product mixture are ideal gases. Discuss qualitatively how an operating pressure higher than that at which the data were taken might affect the selection of a relief device for the reactor, for example, the type and size of relief device used. You may have to review some of your studies on fluid mechanics to complete the discussion. You do not have to go into detail, but should pick out one or two key points for discussion and mention their effects on the flow through a relief system. The reaction being studied goes essentially to completion at 500°K. The reaction data from the laboratory are as follows:

Time, Reactor Pressure, seconds atm abs

0 1.0 5 1.1 10 1.2 15 1.3 20 1.4 30 1.5 45 1.6 60 1.7

At completion 2.0

Problem No. 83


HEALTH AND SAFETY CONCEPT: Toxic Exposure Control and Personal Protective Equipment

BACKGROUND: It is always required that the air quality within a vessel be known before anyone is allowed to enter it. The fact that the vessel has not been used for some time, or was not used in service with a hazardous material, is not a sufficient reason to enter it without testing the air.

One of the potential hazards of closed vessels is that of an oxygen-deficient atmosphere. Normal air contains approximately 21% oxygen, by volume (or moles). OSHA regulations specify respiratory protection (a supply of breathing air) if the level of oxygen is below 19.5%. Oxygen levels below 16% will cause dizziness, rapid heartbeat, and possible headache. Slightly lower levels will cause an inability to move about readily, and an apathetic view of the impending death. Usually the victim will feel no symptoms, will lose consciousness, and will have no recollection of the incident if rescued.

Oxygen-deficient atmospheres can occur in any closed vessel, such as the holds of ships, vats, tanks, silos, or mines. The following problem will provide an opportunity for you to see how an otherwise harmless appearing situation can lead to serious consequences.

PROBLEM: The interior of a steel tank was taken out of service for subsequent repair. Before closing, it was thoroughly steam cleaned, and the level of residual vapors was found to be acceptably low for entry. It was not immediately entered, however, because of the press of other maintenance work, so it was closed for a time to await its repair priority.

The tank dimensions are 27 ft diameter and 35 ft high. You may assume that both the top and bottom are flat. The material of construction was plain carbon steel (primarily iron).

It is known that the typical corrosion rate for iron (or plain carbon steel) in very moist air is about 0.005 in. per year, but is about first order with respect to the oxygen concentration. Assume that to be the case, and assume that the corrosion reaction is approximated by the reaction:

Estimate how long it will take to reduce the oxygen content in thevessel to 19.5% and to 16%.

Problem No. 84

CHEMICAL ENGINEERING TOPIC: Process Control

HEALTH AND SAFETY CONCEPT: Process Control, Interlocks and Alarms

BACKGROUND: A complex chemical process requires the extensive use, of automatic controls. Process controls are used not only to maintain product quality and production rates, but are widely used to help maintain a safe process. A loss of production may be serious, but a loss of life, or an injury or explosion will be far more serious and generally many times more costly. Thus, when instrumentation is used to maintain process safety, it is important that the devices be adequately maintained.

Many times it is possible to provide a process control scheme with some "fail-safe" attributes, that is, if the controller should fail, the process will be controlled in a safe manner. For example, if a controller is being used to maintain coolant flow to the heat exchanger on an exothermic reactor, the control scheme should be such that in the event of a failure of the controller, the coolant control valve will remain in the open position so that cooling will not be lost. However, the open valve does not necessarily mean that the coolant will flow, because whatever caused the failure might also cause a loss of coolant.

Failure can arise because of a number of reasons, one important one being the loss of utilities, that is, electric power to an electronic controller or compressed air in the case of a pneumatic controller. Also, one must prepare for loss of the process stream that is being controlled, as for example, interruption of coolant supply. There are other modes of control system failure, and these should be anticipated and provision made to maintain control when one or more of them occur.

When a control scheme cannot be made fail-safe, and if failure would result in an accident, then it may be appropriate to use backup, or redundant controls. Studies of failure rates have been made, and of course the rates vary a great deal. Generally, it may be expected that failures will occur at a frequency of from 0.1 to 2.5 times per year, dependingupon the device and the application. The seriousness or severity of any particular failure rate depends upon a number of circumstances. The seriousness of a particular failure depends upon the nature of the process.

A failure of any element of the control system will cause the failure of the system. In the problem that follows we will examine the consequences of a failure, or partial failure, of a control valve.

PROBLEM NO. 84 169

PROBLEM: An exothermic reaction occurs in a 1000-gal reactor. Reactor temperature controlis by cooling water in thejacket. ~ h & transfer function relating reactor temperature to cooling water flow rate is given by:

The linear control valve delivers 480 gaVmin (GPM) of cooling water at 3 psig and 0 GPM at 15 psig.

The controller is a proportional, integral, derivative (PID) controller, tuned by the Zeigler-Nichols technique.

After several weeks of operation the valve performance deteriorates, so that the stem "sticks" in a given position. A dead time of 21 sec is introduced into the control loop by the sticky valve. Valve gain remains at 40 GPMIpsi.

What will happen to the reactor temperature control because of the poor valve performance?

(This problem was submitted by Dr. Jim L. Turpin, Department of Chemical Engineering, University of Arkansas.)

Problem No. 85

CHEMICAL ENGINEERING TOPIC: Process Control

HEALTH AND SAFETY CONCEPTS: Process Control, Interlocks and Alarms

BACKGROUND: A complex chemical process requires the extensive use of automatic controls. Process controls are used not only to maintain product quality and production rates, but are widely used to help maintain a safe process. A loss of production may be serious, but a loss of life, or an injury or explosion will be far more serious and generally many times more costly. Thus, when instrumentation is used to maintain process safety, it is important that the devices be adequately maintained and managed.

The potential for control element failure is a significant problem in the maintaining of safe operations. A regular schedule of testing and maintenance is vital to the continuing satisfactory performance of process control equipment.

The operations personnel must also participate in keeping the instrumentation operating as intended. The instrumentation must be kept tuned to the possible changes in the process, and all personnel should be advised of any changes in the process. The following problem illustrates one of the difficulties that may arise when there is a process change.

PROBLEM: An exothermic reaction occurs in a 1000-gal reactor in which the temperature is maintained by cooling water in the jacket. The transfer function relating reactor temperature to cooling water flow is given by:

Each gallon of cooling water removes 167 Btu. The coolant control valve is a square root valve which delivers 50 GPM wide-open at a controller output of 15 psig and is fully closed at 3 psig controller output.

The steady-state heat removal rate is 450,000 Btu/hr when the reactor is fully loaded. The controller is a PID, tuned by the Cohen-Coon procedure at the fully loaded operating condition.

A few days later the reactor throughput is reduced, with a corresponding reduction in the heat removal rate to 190,000 Btulhr. The controller was not re-tuned.

What will happen to the process control and why? (This problem was submitted by Dr. Jim L. Turpin, Department of Chemical

Engineering, University of Arkansas)

Problem No. 86

CHEMICAL ENGINEERING TOPIC: Laboratory


BACKGROUND: There are a number of areas in and around educational laboratories where there is a potential for hazardous conditions to exist. Many such areas, if they exist, can be identified by what industrial hygienists call a "walk- through survey," which is simply a visual inspection of the facilities with detailed attention being paid to potential sources of hazard, such as places where toxic agents may be emitted into the air or locations where there may be a potential fire hazard.

An industrial hygienist might also conduct some measurements of air flow rates, etc., but without such measurements, one still cangain muchinformation by careful observation.

PROBLEM: You are asked to conduct a walk-through survey of your department's mass transfer laboratory, or that laboratory where students do most of the mass transfer experiments. While you are making the survey, be especially alert:

See if you can locate all of the emission points. (Emission points will be such places as addition points, discharge points, sampling locations, and sealing points. Sealing points include "static seals" such as covers, flanges, etc., and "dynamic seals," which are shaft seals as on pump shafts, valve stems, etc.) Are procedures posted where all can see them? How does the student know what are the safe procedures? Are material safety data sheets available for the chemicals in use? Is there any way for the student to determine what the potential hazards are for the chemicals in use?

Evaluate the ventilation: Is there any use of local exhaust ventilation? Where should local exhaust ventilation be provided? Where or how should hoods be positioned? Where does the supply air come from? Where does the exhaust air go? How many air changes per hour would it provide? What might be a reasonable emission rate for any toxic material in your laboratory? Could there be short circuiting of the ventilation air? What is the probable static pressure with respect to the offices and classrooms?

Who has the responsibility of keeping your laboratory safe? If you feel that your laboratory is unsafe, what will you do about it?

Problem No. 87

CHEMICAL ENGINEERING TOPIC: Laboratory

SAFETY AND HEALTH CONCEPT: Fire Protection

BACKGROUND: Almost all chemical engineering laboratories stock flammable materials for various experiments. The flammable materials should be stored in a cabiinet according to the requirements of National Fire Protection Association (NFPA) Standard 30, "Flammable Liquids Code." A flammable liquids storage cabinet is designed to prevent leakage of flammable materials to the outside, to protect the stored materials from external fires, and to limit access to the materials where necessary. The amount of material stored should be limited in quantity as specified by NFPA 30, and the individual containers of flammable material should be tightly closed when stored in the cabinet to prevent leakage or vaporization. Each container should be labeled according to its contents, and the laboratory data available to each user of the materials should include a Materials Safety Data Sheet (MSDS) for each material stored in the cabiinet. If larger quantities of material are required, they should be stored in separate facilities that meet the requirements of NFPA 30. Personnel who will be using the chemicals should be familiar with their flammability properties as well as their chemical properties. (If the chemicals are also toxic, they should also be handled so that there will be no toxicity danger to the users.)

Experiments should be run in well-ventilated areas, and potential ignition sources should be removed before experiments are started. Ventilation systems should be well designed so that the discharge will not be in a location where ignition might result. In some cases, recovery systems for preventing the vaporized material from reaching the environment may be required.

Each flammable material has a characteristic temperature range over which it can be ignited. The lowest temperature at which a flammable liquid can be ignited is called the flash point. The flash point temperature is the temperature at which the concentration of vapor in air above the liquid surface is high enough to be ignited. Tables of data for flash point temperature can be found in NFPA Manual 325M.

There is a minimum concentration of fuel in air that is required for sustained combustion. It is called the lower flammable limit (LFL), and is measured at atmospheric pressure and ambient temperature. The LFL is given in volume percent of vapor or gas in air. If the fuel concentration increases, it will reach a concentration high enough that there is insufficient oxygen for ignition to occur. The concentration above which ignition will not occur is called the upper flammable limit (UFL). At any concentration between the LFL and the UFL, a

PROBLEM NO. 87 173

vapor-air mixture can be ignited. The flammable range may be relatively narrow, or it may be quite wide, For example, n-heptane has an LFL of 1.05% and a UFL of 6.7%. Carbon disultide has an LFL of 1.3% and a UFL of 50%. Thus carbon disulfide will ignite over a much wider range of concentrations than will n-heptane.

Flammable liquids may be spilled during use in the laboratory, and if they are spilled, a fire may occur. If a fire does occur, some method should be available for controlling or extinguishing it. There are a number of methods available for laboratory fire control, but the method preferred may depend on the fuel being burned. For example, most small hydrocarbon fires can be extinguished by a dry chemical fire extinguisher using sodium bicarbonate or potassium bicarbonate agents. They may also be extinguished or controlled using foam agents. However, some other materials, for example, carbon disdtide, are not easily extinguished with dry chemicals. Foam or water may be used to fight carbon disulfide fires under some circumstances. Foams that are used to control hydrocarbon fires may not be suitable for use on alcohols and other water-soluble flammable liquids. Special alcohol foams are available for such chemicals. The MSDS should list suitable fire control agents. Everyone who is to use flammable materials should read the MSDS and become familiar with the techniques of handling the material as well as methods of preventing, controlling, and extinguishing fires.

If the vapor-air mixture above a flammable liquid is in equilibrium with the spilled liquid, there is a range of temperatures for which the fuel-air mixture will be flammable. The temperature at which the LFL concentration is reached will correspond approximately to the flash point. If the fuel is in a closed system, the UFL concentration will correspond to a temperature above which the fuel-air mixture will not ignite. In most cases, there is negligible solubility of air in the liquid, so the liquid can be assumed to be pure, and since the systems are at low pressure, it can be assumed that the vapor-air mixture is ideal. Thus, Raoult's law can be used to calculate the concentration of fuel in air at any temperature. The calculation can be used to determine the range of temperatures for which the mixture of vapor and air in equilibrium with the liquid is flammable. (A practical consequence of such a calculation is to show that the gasoline-air mixture in the fuel tank of a car is usually too rich to ignite inside the tank. One of the special considerations for burning methanol or methanol-gasoline mixtures is that flammable mixtures may form at ambient temperatures in the fuel tank vapor space. Flame arrestors might be required to reduce the risk of ignition and explosion of the fuel tank if methanol is to be used as a motor vehicle fuel.)

PROBLEM. Flammable liquids are to be stored in a flammable liquids storage cabinet that meets NFPA 30 standards. The laboratory temperature may vary from 60 to 80°F. Determine which of the following materials might form flammable vapor-air mixtures if spilled in the storage cabinet and allowed to reach equi-


librium. Also suggest the type of fire extinguisher(s) that would be appropriate if a small spill is ignited in the laboratory. The materials are

n-pentane ethyl ether carbon disuKde acetone benzene methyl alcohol n-octane isopropyl alcohol p-xylene

Note that the calculations you make are only valid if equilibrium is reached. If \ equilibrium is not attained, the concentrations will be lower than those calculated.

Problem No. 88



BACKGROUND: When toxic materials are used and/or produced by chemical processes, it is necessary to assure that the workers are not exposed to the material(s) to such an extent that they receive a harmful dose. Since the most frequent route of entry for toxic materials is by inhalation, limiting the extent of exposure potential often takes the form of limiting the concentration of toxic material in the air that the workers breathe.

Many of the potentially harmful agents are vapors that may be in the air. Whenever there are volatile materials used in a process, they are likely to escape into the air through various leaks, called emission points, which include sealing points, sample withdrawal points, addition points, and so forth.

Sealing points are those places where the processing equipment components come together. "Static seals" include such seals as flanges and covers, where there is no relative motion between the components. "Dynamic seals" are those places where there is relative motion between components. Equipment with rotating and reciprocating shafts that transmit mechanical energy through seals include pumps, compressors, agitators, and valves. Normally it would be expected that dynamic seals would cause the more serious leakage problems.

In recognition of the exposure potential from inhalation of toxic vapors, the Occupational Safety and Health Administration (OSHA) has established maximum concentration limits for a large number of agents. OSHA regulations, which have the effect of law, specify permissible exposure limits (PELS) for worker exposure as a time-weighted average (TWA) over a work day. In some instances, there are also maximum, or "ceiling" concentrations which must never be exceeded, even for a short time. Other agencies and groups make recommendations, especially the National Institute for Occupational Health and Safety (NOSH) and the American Conference of Governmental Industrial Hygienists (ACGIH). NOSH is a Federal government agency responsible for research and training. ACGIH is an association of technical persons who work in industrial hygiene and are employed by a governmental agency.

PROBLEM: Your company has been a major supplier of a fast drying ink, available in various colors. Your market share has been increasing a bit lately, and one of your competitors has decided to discontinue ink manufacturing. To meet the anticipated increased demand and to forestall foreign competition, your management has decided to install a new process line, parallel to and essentially the same

176 S m , HEALTH, AND LOSS PREVENTION IN CHEMICAL PROCESSES

as the existing line, except with double the capacity. The new process equipment will include 10 pumps along with miscellaneous other equipment in a room that is 48 ft long, 22 ft wide with 14 ft ceiling height. It is planned that the pumps will be equipped with packing glands. The pump type selected has been used for several years in the existing line and has given reliable service. Packing glands, however, always have a finite leak rate because it is necessary to lubricate the packing. As the pumps operate, normal wear will cause the leak rate to increase. With appropriate periodic maintenance the leakage can be held to a low level, however.

To assess the leak rate on the existing pumps, the plant environmental engineer conducted a test as follows: A portable plastic enclosure was placed around the six existing pumps, and 12 Llmin of air was drawn through the enclosure while the pumps were running. This was done until a near steady state was achieved as shown by a constant reading on a hydrocarbon analyzer placed in the exit air stream. The exit air stream was then sampled for 30 min by using a small air pump to draw a sample of the air through a charcoal adsorber tube. The charcoal was assumed to have adsorbed all the organic vapors from the air that passed through it. The sampling rate was 100 rnllmin, the air temperature was 74°F and the barometric pressure was 740 mm Hg. The existing process is in a room with a total volume of 17,000 ft3.

The charcoal tubes were taken to the laboratory, desorbed and the vapors analyzed. The following amounts were found:

Sample Mass TL V- TWA LFL Component (mg) @pm, molar) (mole %)

Toluene 430 100 1.3 1,1,1 Trichloroethane 118 350 6 1,l Dichloroethane 133 200 6 Acetone 675 750 2.6 Methyl ethyl ketone 216 200 2

What is the leak rate for these chemicals in grams per hour for each solvent? If the building has five air changes per hour and there is good mixing, what will be the concentration of solvents in the building? Are these below the TLV in each case? Is the mixture concentration below the TLV?

TLV means "Threshold Limit Value" and is a concentration limit established by the American Conference of Governmental Industrial Hygienists (ACGIH). The TLV-TWA represents the air concentration to which most workers may be exposed for a normal 8-hr day, 40-hr work week, without ill effects.

To determine if a mixture is below the TLV when there are no synergistic effects, but where the health effects are similar and perhaps additive, sum the quotients of each concentration divided by its respective TLV. If the sum is less than 1 then the mixture is below the TLV.

PROBLEM NO. 88 177

LFL is the "Lower Flammable Limit" and represents the lowest concentration in air that would be flammable.

Further evaluation of the existing process showed that with the general ventilation rate of five air changes per hour, the level of acetone in the exhaust was 83 mg/m3. This is the building with the six pumps. What proportion of the emissions do the pumps account for? If the pumps in the new process account for the same proportion of the leakage, what will the expected level of vapors be in the new room if it also has a ventilation rate of five air changes per hour?

Do you feel that the assumption of complete mixing of the ventilation air is a good one? What, if any, importance would there be to a finding that there was only very poor mixing of the ventilation air in the room? The book, Industrial Ventila- tion-A Manual of Recommended Practice, 19th ed. (American Conference of Governmental Industrial Hygienists, Cincinnati, 1986), suggests that the required air flow for this type of ventilation be larger than a calculated requirement by a factor of from 5 to 10 to account for poor mixing.

If there is a problem with possible overexposure in the new facility, what should be done about it? Would you recommerld changing the design? If so, how? Is there any possibility that the total emission might cause an air pollution problem off site?

HINTS FOR PROBLEM SOLUTION: Determine the fraction of the vent air that is sampled, then from the length of the sampling time and the amounts collected, determine the total emission rate in grams per hour.

The dilution air will remove all the contaminants released, but the concentration in the room air and the dilution air depends on the release rate and the dilution air flow rate.

Assume that the pumps in the new building account for the same fraction of the total emissions and that the ratio of acetone to all the other materials is the same as in the test in the old building.

Problem No. 89

CHEMICAL ENGINEERING TOPIC: I-Ieat Transfer; Design


BACKGROUND: Large quantities of hazardous materials must be stored for industrial and commercial use. As an example, liquid hydrocarbons from methane to crude petroleum are stored for various periods between production and end use. Refinery tank farms frequently contain millions of gallons of petroleum products awaiting processing into final products, being used in processing, and awaiting final distribution to the user. Individual storage tanks are often quite large, and storage of several hundred thousand barrels of product in a single tank is not unusual. There are a number of safety and health concerns that must be accounted for in the design and operation of a large storage tank, beginning with the design of the tank itself and ending with the maintenance of the tank and the equipment used to transfer product to and from the tank.

The tank used for storage of product must be designed using materials and techniques that are suitable for the liquid being stored. Temperatures and pressures under which the product will be stored may vary, and economic considerations may dictate the style of tank used. Products that are liquid at ambient temperature will be stored in ambient pressure tanks at ambient temperature, but products that have boiling points below ambient temperature may be stored either as compressed gases, pressurized liquids, or refrigerated or cryogenic liquids. If refrigerated liquids are stored, the tanks will have to be insulated to prevent the liquid from boiling away. Many design factors will have to be considered for each individual case. The following problem considers one aspect of the design of a tank used for the storage of n-butane.

PROBLEM: Large amounts of n-butane must be stored at an industrial site. An economicstudy has shown that a 500,000 bbl tank containingn-butane at a pressure that does not exceed 1.0 psig will be preferred to storage of an equivalent quantity of n-butane at ambient temperature as a pressurized liquid. At this low pressure, n-butane must be refrigerated and stored at subambient temperature. The tank must therefore be insulated. The tank design selected is a double-walled metal tank with a suspended deck. The figure below shows a schematic diagram of the tank. The outer wall is the pressure containment vessel. It must be designed for containment of the gas pressure exerted by the vapor pressure of the n-butane inside the tank. The outer wall and roof of the tank will be in continuous contact with butane vapor. The inner tank is open to the outer tank at the top, so it must be designed only for the hydrostatic pressure of the liquid it contains. The annular space

PROBLEM NO. 89

SUSPENDED

OUTER WALL

INNER TANK

FOAMED =ASS BLOCK INSULATION

between the walls will be filled with insulation. Expanded perlite is the most common insulation used for the purpose. The suspended deck above the liquid insulates the space at the top of the tank. Glass fiber insulation is usually used for insulation on the suspended deck The tank bottom must be insulated as well. Foamed glass blocks are usually used for insulating the tank bottom because the insulation must support the weight of the tank and its contents. The foundation under the tank must be heated to prevent freezing of the soil beneath the tank. If the soil freezes, frost heaving may cause tank failure. A boiloff compressor must be used to withdraw the evaporating vapor from the tank. The vapor may be processed or re-liquefied and returned to the tank.

There are many other safety and operational aspects of tank design. For example, if the ambient temperature decreases, there will be a point at which the butane vapor in contact with the tank roof will become cold enough to begin to condense. The butane that condenses will fall to the suspended deck where it will flow back into the bulk of the liquid below. However if much butane condenses, the tank pressure will drop below ambient and the tank will collapse. You are to calculate the ambient temperature at which butane will begin to condense. That is more difficult than you might imagine because there will be either radiation heating (during the day) or radiation cooling (during the night). You may assume the night sky to be equivalent to a blackbody that radiates at -50°F and that solar radiation during the daytime is 300 ~tu/hr-ft2. If the wind is blowing, the convective coefficient will be different than if the wind is calm. At night, assume the wind is nearly calm and the convective coefficient is2.7 ~ t u / h r - f t ~ - " ~ . During the daytime, assume the wind blows at about 25 mph and the convective coefficient is 4.8 ~ t u / h r - f t ~ - " ~ . (Your instructor may ask you to show these values are reasonable. If so, you may wish to review the article by Kumana and Kothari, "Predict Storage Tank Heat Transfer Precisely," Chemical Engineering, March 22,1982, p. 127.)

Problem No. 90


SAFETY AND HEALTH CONCEPT: Storage and Handling

BACKGROUND: Many industrial materials are produced and utilized in large quantities. They maybe transported from the place of manufacture to the place of consumption in several ways, and they may be used at the point of manufacture as an intermediate. Ammonia is one of the most important commercial chemicals. It is produced and used in large quantities in a number of industries such as plastics, explosives, and fertilizers. The fertilizer industry is the largest user of ammonia, and the use is seasonal, so that large quantities must be stored in order to meet the demand and yet keep the productions levels about constant. Ammonia may be stored in very large insulated tanks at pressures near ambient; in large spheres at moderate pressures, but refrigerated to reduce the pressure; and at ambient temperature but higher pressure, corresponding to the vapor pressure at ambient temperature. Shipment is usually either by railroad tank car, by tank truck, or by pipeline, in which cases it is usually shipped at ambient temperature.

The choice of whether to store ammonia as an ambient temperature liquid, a partially refrigerated liquid, or an ambient pressure liquid depends on economic considerations. One of the factors that determines the storage method is the quantity of ammonia to be stored.

PROBLEM: A company produces ammonia in a 1000-todday plant. It sells the ammonia to customers throughout the year, and the ammonia at the production plant is stored in large tanks that have a maximum working pressure of 2.5 psig. When the ammonia is sold, the buyers frequently ask for guidance on the choice of storage tanks. You are assigned the task of determining an approximate break point for storing the ammonia as a pressurized liquid, a partially refrigerated liquid, or a liquid near ambient pressure. Your analysis should include not only the cost of the tank, but also an analysis of the problems that might arise if the ammonia is transported at ambient temperature and then cooled at the customer's location before storage if storage is not at ambient temperature. You should also consider that the ammonia might have to be warmed before it leaves the production plant if it is transported at ambient temperature. The customers may need to store ammonia in quantities from about 30 tons to 30,000 tons, depending on their usage. The largest pressurized tanks available for ambient temperature storage have a volume of about 50,000 gal. Be sure to provide some ullage space in the tank to allow for liquid expansion as the liquid warms. Be sure to include some considerations of any special safety features you might find necessary.

SACHE Problem Set Volume 1

Documents

Transcript of SACHE Problem Set Volume 1