Handbook of Laboratory Glassware Blowing

8/6/2019 Handbook of Laboratory Glassware Blowing

1/124

A Handbook ofjK05^S8SH

Laboratory Glass-Blowing

Bernard D. Bolas


2/124

MIEMCAL

Gift ofMax S. Marshall, M.D


3/124

**** "


4/124


5/124


6/124

To my FriendsEric Reid

andSidney Wilkinson


7/124

A Handbook ofLaboratory Glass-Blowing

BERNARD D. BOLASWITH NUMEROUS DIAGRAMS IN THE TEXT

BY NAOMI BOLAS

LONDONGEORGE ROUTLEDGE & SONS, LTDNEW YORK : E. P. BUTTON & CO.1921


8/124


9/124

*CONTENTS

CHAP. PAGEI. Introduction and Preliminary Remarks General

Principles to be observed in Glass WorkingChoice of Apparatus Tools and Appliances-Glass i

II. Easy Examples of Laboratory Glass -BlowingCutting and Sealing Tubes, Tubes for HighTemperature Experiments Thermometer-BulbsBulbs of Special Glass, Pipettes, Absorption-Bulbs or Washing Bulbs Joining Tubes,Branches, Exhaustion-Branches, Branches ofDissimilar Glass, Blowing Bulbs, A ThistleFunnel, Cracking and Breaking Glass, Leadingand Direction of Cracks Use of Glass Rod orStrips of Window-Glass, Joining Rod, Feet andSupports Gripping Devices for use in CorrosiveSolutions The Building up of Special Formsfrom Solid Glass 10

III. Internal Seals, Air-Traps, Spray Arresters, Filter-Pumps Sprays, Condensers ; plain, doublesurface, and spherical Soxhlet Tubes and FatExtraction Apparatus Vacuum Tubes, Elec-trode Work, Enclosed Thermometers, AlarmThermometers . . . Recording Thermometers,"Spinning" Glass .' 32

IV. Glass, its Composition and Characteristics Anneal-ing Drilling, Grinding, and Shaping Glass bymethods other than Fusion Stop -cocks


10/124

vi. CONTENTSCHAP. PAGE

Marking Glass Calibration and Graduation ofApparatus Thermometers Exhaustion of Ap-paratus Joining Glass and Metal SilveringGlass 55

V. Extemporised Glass-Blowing Apparatus The useof Oil or other Fuels Making Small Rods andTubes from Glass Scraps The Examination ofManufactured Apparatus with a view to Dis-covering the Methods used in ManufactureSummary of Conditions necessary for SuccessfulGlass-Blowing 80

Index 105


11/124

PREFACETo cover the whole field of glass-blowing in asmall handbook would be impossible. Toattempt even a complete outline of the methodsused in making commercial apparatus wouldinvolve more than could be undertaken with-out omitting the essential details of manipulationthat a novice needs. I have, therefore, confinedmyself as far as possible to such work as willfind practical application in the laboratory andwill, I hope, prove of value to those whoseinterests lie therein.The method of treatment and somewhat dis-

jointed style of writing have been chosensolely with the view to economy of spacewithout the undue sacrifice of clearness.BERNARD D. BOLAS.


12/124


13/124

Handbook of LaboratoryGlass-BlowingCHAPTER I

Introduction and Preliminary Remarks General Principles to beobserved in Glass Working Choice of Apparatus Tools andAppliances Glass.

GLASS-BLOWING is neither very easy nor verydifficult ; there are operations so easy that theyoungest laboratory boy should be able torepeat them successfully after once havingbeen shown the way, there are operations sodifficult that years are needed to train eye andhand and judgment to carry them out ; butthe greater number of scientific needs lie be-tween these two extremes. Yet a surprisinglylarge number of scientific workers fail even tojoin a glass tube or make a T piece that willnot crack spontaneously, and the fault is ratherone of understanding than of lack of ability tocarry out the necessary manipulation.


14/124

2 LABORATORY GLASS-BLOWINGIn following the scheme of instruction

adopted in this handbook, it will be well forthe student to pay particular attention to thereason given for each detail of the desirableprocedure, and, as far as may be, to memoriseit. Once having mastered the underlyingreason, he can evolve schemes of manipulationto suit his own particular needs, although, as arule, those given in the following pages will befound to embody the result of many years'experience.There is a wide choice of apparatus, from asimple mouth-blowpipe and a candle flame to a

power-driven blower and a multiple-jet heatingdevice. All are useful, and all have theirspecial applications, but, for the present, wewill consider the ordinary types of bellows andblowpipes, such as one usually finds in achemical or physical laboratory.The usual, or Herepath, type of gas blow-pipe consists of an outer tube through whichcoal gas can be passed and an inner tubethrough which a stream of air may be blown.Such a blowpipe is shown in section by Fig. i.It is desirable to have the three centringscrews as shown, in order to adjust the position


15/124

LABORATORY GLASS-BLOWING 3of the air jet and obtain a well-shaped flame,but these screws are sometimes omitted. Fig.i, a and b show the effects of defective centringof the air jet, c shows the effect of dirt or

Fig i

roughness in the inside of the air jet, d showsa satisfactory flame.

For many purposes, it is an advantage to


16/124

4 LABORATORY GLASS-BLOWINGhave what is sometimes known as a "quick-change

"blowpipe ; that is one in which jets of

varying size may be brought into positionwithout stopping the work for more than afraction of a second. Such a device is madeby Messrs. Letcher, and is shown by e, and insection by f Fig. i. It is only necessary torotate the desired jet into position in order toconnect it with both gas and air supplies. Asmall bye-pass ignites the gas, and adjustmentof gas and air may be made by a partialrotation of the cylinder which carries the jets.For specially heavy work, where it is

needed to heat a large mass of glass, a multipleblowpipe jet of the pattern invented by myfather, Thomas Bolas, as the result of asuggestion derived from a study of the jetused in Griffin's gas furnace, is of considerablevalue. This jet consists of a block of metal inwhich are drilled seven holes, one being centraland the other six arranged in a close circlearound the central hole. To each of theseholes is a communication way leading to thegas supply, and an air jet is arranged centrallyin each. Each hole has also an extension tubefitted into it, the whole effect being that of


17/124

LABORATORY GLASS-BLOWING 5seven blowpipes. In order to provide a finaladjustment for the flame, a perforated platehaving seven holes which correspond in sizeand position to the outer tubes is arranged toslide on parallel guides in front of these outertubes.The next piece of apparatus for consideration

is the bellows, of which there are three ormore types on the market, although all consistof two essential parts, the blower or bellowsproper and the wind chamber or reservoir.Two patterns are shown in Fig. 2 ; a, is the

form which is commonly used by jewellers andmetal workers to supply the air blast necessaryfor heating small furnaces. Such a bellowsmay be obtained at almost any jewellers'supply dealer in Clerkenwell, but it not in-frequently happens that the spring in the wind


18/124

6 LABORATORY GLASS-BLOWINGchamber is too strong for glass-blowing, andhence the air supply tends to vary in pressure.This can be improved by fitting a weakerspring, but an easier way and one that usuallygives fairly satisfactory results, is to place anordinary screw-clip on the rubber tube leadingfrom the bellows to the blowpipe, and totighten this until an even blast is obtained.

Another form of bellows, made by Messrs.Fletcher and Co., and common in mostlaboratories, is shown by b ; the wind chamberconsists of a disc of india-rubber clamped undera circular frame or tied on to a circular rim.This form is shown by Fig. 2, b.The third form, and one which my ownexperience has caused me to prefer to anyother, is cylindrical, and stands inside thepedestal of the blowpipe-table. A blowpipe-table of this description is made by Enfer ofParis.There is no need, however, to purchase an

expensive table for laboratory use. All thework described in this book can quite well bedone with a simple foot bellows and a quick-change blowpipe. Nearly all of it can be donewith a single jet blowpipe, such as that described


19/124

LABORATORY GLASS-BLOWING 7first, or even with the still simpler apparatusmentioned on page 84, but I do not advise thebeginner to practise with quite so simple a format first, and for that reason have postponed adescription of it until the last chapter.

Glass-blowers' tools and appliances are manyand various, quite a number of them are betterrejected than used, but there are a few essentials.These are, file, glass-knife, small turn-pin,large turn-pin, carbon cones, carbon plate,rubber tube of small diameter, various sizesof corks, and an asbestos heat reflector. Forordinary work, an annealing oven is not neces-sary, but one is described on page 60 in con-nection with the special cases where annealingis desirable.

Fig. 3 illustrates the tools and appliances.a is an end view of the desirable form of file,and shows the best method of grinding theedges in order to obtain a highly satisfactorytool, b is a glass knife, shown both in per-spective and end view, it is made of glass-hardsteel and should be sharpened on a roughstone, such as a scythe-stone, in order to givea slightly irregular edge, c is a small turn-pinwhich may be made by flattening and filing the


20/124

8 LABORATORY GLASS-BLOWINGend of a six-inch nail, d is the large turn-pinand consists of a polished iron spike, about fiveinches long and a quarter of an inch diameterat its largest part. This should be mountedin a wooden handle, e and /"are carbon cones.A thin rubber tube is also useful ; it may be at-

Fig. 3

tached to the work and serve as a blowing tube,thus obviating the necessity of moving the workto the mouth when internal air pressure is to beapplied. In order to avoid undue repetition,the uses of these tools and appliances will bedescribed as they occur.


21/124

LABORATORY GLASS-BLOWING 9Glass, as usually supplied by chemical appar-

atus dealers is of the composition known as"soda-glass." They also supply " hard " or"combustion" glass, but this is only used forspecial purposes, as it is too infusible for con-venient working in the ordinary blowpipe flame.

Soda-glass consists primarily of silicate ofsodium with smaller quantities of silicate ofaluminum and potassium. Its exact composi-tion varies. It is not blackened, as lead glassis, by exposure to the reducing gases whichare present in the blue cone of a blowpipeflame, and hence is easier for a beginner towork without producing discolouration.

Further notes on glasses will be found onpage 55, but for ordinary purposes soda-glasswill probably be used.


22/124

CHAPTER IIEasy Examples of Laboratory Glass-BlowingCutting and SealingTubes for Various Purposes ; Test-Tubes, Pressure-Tubes, Tubes for

High Temperature Experiments Thermometer-Bulbs, Bulbs ofSpecial Glass, Pipettes, Absorption-Bulbs or Washing-Bulbs JoiningTubes ; Branches, Exhaustien-Branches, Branches of Dissimilar GlassBlowing Bulbs ; A Thistle Funnel ; Cracking and Breaking Glass ;Leading and Direction of Cracks Use of Glass Rod or Strips ofWindow-Glass ; Joining Rod, Feet and Supports Gripping Devicesfor use in Corrosive Solutions The Building Up of Special Formsfrom Solid Glass.

PERHAPS the most common need of the glass-blower whose work is connected with that ofthe laboratory is for a sealed tube ; and thesealing of a tube is an excellent preliminaryexercise in glass-blowing.We will assume that the student has adjustedthe blowpipe to give a flame similar to thatshown in dy Fig. i, and that he has learnedto maintain a steady blast of air with thebellows ; further, we will assume that the tubehe wishes to seal is of moderate size, say notmore than half an inch in diameter and with


23/124

LABORATORY GLASS-BLOWING 11walls of from one-tenth to one-fifth of an inchthick.A convenient length of tube for the first trialis about one foot ; this should be cut off fromthe longer piece, in which it is usually supplied,as follows : lay the tube on a flat surface andmake a deep cut with the edge of a file. Do

not "saw" the file to and fro over the glass.If the file edge has been ground as shown in a,Fig. 3, such a procedure will be quite unneces-sary and only involve undue wear ; one move-ment with sufficient pressure to make the file"bite" will give a deep cut. Now rotate thetube through about one-eighth of a turn and


24/124

12 LABORATORY GLASS-BLOWINGmake another cut in continuation of the first.Take the tube in the hands, as shown in a, Fig.4, and apply pressure with the thumbs, at thesame time straining at the ends. The tubeshould break easily. If it does not, do notstrain too hard, as it may shatter and causeserious injuries to the hands, but repeat theoperation with the file and so deepen theoriginal cuts. In holding a tube for breaking,it is important to place the hands as shown insketch, as this method is least likely to causeshattering and also minimises the risk of injuryeven if the tube should shatter. To cut a largetube, or one having very thick walls, it is betterto avoid straining altogether and to break byapplying a small bead of intensely heated glassto the file cut. If the walls are very thin, aglass-blower's knife should be used instead of afile. The tube and glass-blower's knife shouldbe held in the hand, and the tube rotatedagainst the edge of the knife ; this will notproduce a deep cut, but is less likely to breakthe tube. A bead of hot glass should be usedto complete the work.The next operation is to heat the glass tubein the middle ; this must be done gradually


25/124

LABORATORY GLASS-BLOWING 13and evenly ; that is to say the tube must berotated during heating and held some con-siderable distance in front of the flame at first ;otherwise the outer surface of the glass willexpand before the interior is affected and thetube will break. From two to five minutes,heating at a distance of about eight inches infront of the flame will be found sufficient inmost cases, and another minute should betaken in bringing the tube into the flame.Gradual heating is important, but even heatingis still more important and this can only beobtained by uniform and steady rotation. Untilthe student can rotate a tube steadily withoutthinking about it, real progress in glass-blowingis impossible.When the tube is in the flame it must beheld just in front of the blue cone and rotateduntil the glass is soft enough to permit the endsto be drawn apart. Continue to separate theends and, at the same time, move the tubevery slightly along its own axis, so that theflame tends to play a little more on the thickerpart than on the drawn-out portion. If this isdone carefully, the drawn-out portion can beseparated off, leaving only a slight "bleb" on


26/124

14 LABORATORY GLASS-BLOWINGthe portion it is desired to seal. This isillustrated by b, Fig. 4.To convert the seal at b, Fig. 4., into theordinary form of test-tube seal, it is onlynecessary to heat the " bleb " a little morestrongly, blow gently into the tube until thethick portion is slightly expanded, re-heat thewhole of the rounded end until it is beginningto collapse, and give a final shaping by carefulblowing after it has commenced to cool. Ineach case the glass must be removed from theflame before blowing. The finished seal isshown by c, Fig. 4. If desired, the open endmay now be finished by heating and rotatingthe soft glass against the large turn-pin, asillustrated in d, but the turn-pin must not beallowed to become too hot, as if this happensit will stick to the glass. After turningout the end, the lip of glass must beheated to redness and allowed to cool withoutcoming in contact with anything ; otherwise itwill be in a condition of strain and liable tocrack spontaneously. The finished test-tubeis shown by e.When it is necessary to seal a substanceinside a glass tube, the bottom of the tube is


27/124

LABORATORY GLASS-BLOWING 15first closed, as explained above, and allowedto cool ; the substance, if a solid, is now in-troduced, but should not come to within lessthan two inches of the point where the secondseal is to be made. If the substance is aliquid it can more conveniently be introducedat a later stage.Now bring the tube into the blowpipe flamegradually, and rotate it, while heating, at theplace where it is to be closed. Allow the glassto soften and commence to run together untilthe diameter of the tube is reduced to abouthalf its original size. Remove from the flameand draw the ends apart, this should give along, thick extension as shown by f Fig. 4.If any liquid is to be introduced, it may nowbe done by inserting a thin rubber or othertube through the opening and running theliquid in. A glass tube should be used withcaution for introducing the liquid, as any hardsubstance will tend to scratch the inside of theglass and cause cracking. The final closure ismade by melting the drawn-out extension inthe blowpipe flame ; the finished seal beingshown by g, Fig. 4.

If the sealed tube has to stand internal


28/124

16 LABORATORY GLASS-BLOWINGpressure, it is desirable to allow the glass tothicken somewhat more before drawing out,and the bottom seal should also be madethicker. For such a tube, and especially whenit has to stand heating, as in a Carius deter-mination of chlorine, each seal should be cooledvery slowly by rotating it in a gas flame untilthe surface is covered with a thick layer ofsoot, and it should then be placed aside in aposition where the hot glass will not come incontact with anything, and where it will bescreened from all draughts.

Joining Tube. We will now consider thevarious forms of join in glass tubing which aremet with in the laboratory. First, as beingeasiest, we will deal with the end-to-endjoining of two tubes of similar glass, a, b, andr, Fig. 5, illustrate this. One end of one ofthe tubes should be closed, a lip should beturned out on each of the ends to be joined,and both lips heated simultaneously until theglass is thoroughly soft. Now bring the lipstogether gently, until they are in contact at allpoints and there are no places at which aircan escape ; remove from the flame, and blowslowly and very cautiously until the joint is ex-


29/124

LABORATORY GLASS-BLOWING i;panded as shown in b, Fig. 5. Reheat in theflame until the glass has run down to ratherless than the original diameter of the tube, andgive a final shaping by re-blowing. The chief

Fig. 5

factors of success in making such a join are,thorough heating of the glass before bringingthe two tubes together, and avoidance of hardor sudden blowing when expanding the joint.The finished work is shown by c, Fig. 5.To join a small glass tube to the end of a


30/124

i8 LABORATORY GLASS-BLOWINGlarge one, the large tube should first be sealed,a small spot on the extreme end of the sealheated, and air pressure used to expand theheated spot as shown in d. This expandedspot is then re-heated and blown out until itbursts as shown in e, the thin fragments ofglass are removed and the end of the smalltube turned out as shown in f. After this theprocedure is similar to that used in jointingtwo tubes of equal size.When these two forms of joint have beenmastered, a " piece will present but littledifficulty. It is made in three stages as shownin Fig. 5, and the procedure is similar to thatused in joining a large and small tube. Careshould be taken to avoid softening the top ofthe "T" too much, or the glass will bend anddistort the finished work ; although a slightbend can be rectified by re-heating and bendingback. Local re-heating is often useful ingiving the joint its final shape.An exhaustion branch is often made by atotally different method. This method isshown by gt h t and z, Fig. 5 ; g is the tube onwhich the branch is to be made. The end ofa rod of similar glass should be heated until a


31/124

LABORATORY GLASS-BLOWING 19mass of thoroughly liquid glass has collected,as shown, and at the same time a spot shouldbe heated on that part of the tube where it isdesired to make the branch. The mass of hotglass on the rod is now brought in contact withthe heated spot on the tube and expanded byblowing as shown by h. The air pressure inthe tube is still maintained while the rod isdrawn away as shown by i. This will give ahollow branch which may be cut off at anydesired point, and is then ready for connectionto the vacuum pump.If the rod used is of a dissimilar glass, thebranch should be blown much thinner. Sucha branch will often serve as a useful basis forjoining two tubes of different composition, asthe

ordinary typeof branch is more liable to

crack when made with two glasses havingdifferent coefficients of expansion.

Blowing Bulbs. A bulb may be blown on aclosed tube such as that shown by c, Fig. 5, byrotating it in the blowpipe flame until the endis softened, removing it from the flame andblowing cautiously. It is desirable to continuethe rotation during blowing. In the case of avery small tube, it is sufficient to melt the end


32/124

20 LABORATORY GLASS-BLOWINGwithout previous sealing, rotate it in the flameuntil enough glass has collected, remove fromthe flame and blow while keeping the tube inrotation.

Thermometer Bulbs. If the thermometeris to be filled with mercury, it is desirable touse a rubber bulb for blowing, as moisture isliable to condense inside the tube when themouth is used, and this moisture will cause themercury thread to break. In any case, a slightpressure should be maintained inside thethermometer tube while it is in the flame ;otherwise the fine capillary tube will close andit will be very difficult to expand the heatedglass into a bulb.

Large Bulbs. When a large bulb is neededon a small or medium sized tube, it is oftennecessary to provide more glass than would beobtained if the bulb were blown in the ordinaryway. One method is to expand the tube insuccessive stages along its axis, as shown by a,Fig. 6. These expanded portions are thenreheated, so that they run together into onehollow mass from which the bulb is blown ; band c^ illustrate this. Another method, andone which is useful for very large bulbs, is to


33/124

LABORATORY GLASS-BLOWING 21fuse on a length of large, thick-walled, tubing.The heat reflector, g, Fig. 3, should be used, ifnecesssary, when making large bulbs. It

Fig. 6

consists of a sheet of asbestos mounted in afoot, and is used by being placed close to themass of glass on the side away from the blow-pipe flame while the glass is being heated.


34/124

22 LABORATORY GLASS-BLOWINGBulbs of Dissimilar Glass. These may bemade by the second method given under" Large Bulbs," but the joint should be blown

as thin as possible. Further instructions in theuse of unlike glasses are given on page 94.A Bulb in the Middle of a Tube. Unlessthe bulb is to be quite small, it will benecessary to join in a piece of thick glasstubing, or to draw the thin tube out from alarger piece, thus leaving a thick ir.ass in themiddle as shown by d, Fig. 6. This mass ofglass

should now be rotated in the blowpipeflame until it is quite soft and on the point ofrunning together. Considerable practice willbe necessary before the two ends of the tubecan be rotated at the same speed and without" wobbling," but this power must be acquired.When the glass is thoroughly hot, remove fromthe flame, hold in a horizontal position, andexpand by blowing. It is essential to continuethe rotation while this is done. Should onepart of the bulb tend to expand more than theother, turn the expanded part to the bottom,pause for about a second, both in rotating andblowing, in order that the lower portion may


35/124

LABORATORY GLASS-BLOWING 23be cooled by ascending air-currents ; thencontinue blowing and turning as before.

Absorption Bulbs or Washing Bulbs.These are made by an elaboration of theprocesses given in the last paragraph, g, h, andz, Fig. 6, illustrate this.

A Thistle Funnel. This is made byblowing a fairly thick-walled bulb on a glasstube, bursting a hole by heating and blowing,and enlarging the burst-out part by heatingand rotating against a turn-pin.

Bending Glass Tube. Small tubing may bebent in a flat flame gas burner and offers nospecial difficulty. Large or thin-walled tubingshould be heated in the blowpipe flame and aslight bend made ; another zone of the tube,just touching the first bend, should now beheated and another slight bend made. In thisway it is possible to avoid flattening and abend having any required angle can graduallybe produced. A final shaping of the bend maybe made by heating in a large blowpipe flameand expanding slightly by air pressure.

Glass Spirals. If a tube is heated bymeans of a long, flat-flame burner, the softenedtube may be wound on to an iron mandrel


36/124

24 LABORATORY GLASS-BLOWINGwhich has previously been covered withasbestos. The mandrel should be madeslightly conical in order to facilitate withdrawal.It is desirable to heat the surface of theasbestos almost to redness by means of asecond burner, and thus avoid undue chilling

a

of the glass and the consequent production ofinternal strain.A Thermo-Regulatorfor Gas. Fig. 7, ae, shows an easily constructed thermo-regulator.


37/124

LABORATORY GLASS-BLOWING 25The mercury reservoir, a, and the upper part,6, are made by joining two larger pieces oftubing on to the capillary. The gas inletpasses through a rubber stopper, in order toallow of adjustment for depth of insertion, andthe bye-pass branches, d and , are connectedby a piece of rubber tubing which can becompressed by means of a screw clip, thusproviding a means of regulating the bye-pass.

Use of Glass Rod. Apart from its mostcommon laboratory use for stirring ; glass rodmay be used in building up such articles asinsulating feet for electrical apparatus or acid-resisting cages for chemical purposes. Such acage is shown by/, g and h, Fig. 7. Further,by an elaboration of the method of making anexhaustion branch, given on page 18, blownarticles may also be constructed from rod.Note the added parts of 0, Fig. 9.A Simple Foot. The form of foot shown byFig. 7, k, is easy to make and has many uses.First join a glass rod to a length of glasstubing as shown (the joint should be expandedslightly by blowing), cut off the tube and heatthe piece remaining on the rod until it can beturned out as shown by i. This should be


38/124

26 LABORATORY GLASS-BLOWINGdone with the large turn-pin, and care shouldbe taken not to heat the supporting rod toostrongly, otherwise the piece of tube willbecome bent and distorted ; it is better tocommence by heating the edge of the piece oftube and turn out a lip, then extend the heatingby degrees and turn out more and more untilthe foot looks like that shown by z.We now need to make three projections ofglass rod. These are produced as follows :Heat the end of the glass rod until a thoroughlymelted mass of glass has accumulated (the rodmust be rotated while this is being done, other-wise the glass will drop off) ; when sufficientmelted glass has been obtained, the edge of theturned-out foot should be heated to dull rednessover about one-third of its circumference, andthe melted glass on the rod should be drawnalong the heated portion until both are socompletely in contact as to form one mass ofsemi-fluid glass. The rod should now bedrawn away slowly, and, finally, separated bymelting off, thus producing a flat projection.A repetition of the process will give the othertwo projections, and the finished foot may beadjusted to stand upright by heating the


39/124

LABORATORY GLASS-BLOWING 27projections slightly and standing it on thecarbon plate mentioned on page 7. After thefoot is adjusted it should be annealed slightlyby heating to just below the softening point ofthe glass and then rotating in a smoky gasflame until it is covered with a deposit ofcarbon, after which it should be allowed tocool in a place free from draughts and wherethe hot glass will not come in contact withanything. The finished foot is shown by /&,Fig. 7.

Building up from Glass Rod. A glassskeleton-work can be constructed from rodwithout much difficulty, and is sometimes use-ful as a container for a substance which has tobe treated with acid, or for similar purposes.The method is almost sufficiently explained bythe illustration in Fig. 7 ; f shows the initialstage, g the method of construction of thenet-work, and h the finished container. Itis convenient to introduce the substance atthe stage indicated by g. The importantpoints to observe in making this contrivanceare that the glass rod must be kept hot by-working while it is actually in the flame, andthat the skeleton must be made as thin as


40/124

28 LABORATORY GLASS-BLOWINGpossible with the avoidance of heavy massesof glass at any place. If these details areneglected it will be almost certain to crack.

Stirrers. These are usually made fromglass rod, and no special instructions are neces-sary for their construction, except that theglass should be in a thoroughly fused conditionbefore making any joins and the finished joinshould be annealed slightly by covering with adeposit of soot, as explained on page 16. Theflat ends shown in a, Fig. 8, are made bysqueezing the soft glass rod between two piecesof carbon, and should be reheated to dull red-ness after shaping. Fig. 8 also shows variousforms of stirrer.

In order to carry out stirring operations inthe presence of a gas or mixture of gases otherthan air, some form of gland or seal may benecessary where the stirrer passes through thebearing in which it runs. A flask to which isfitted a stirrer and gas seal is shown in sectionby b, Fig. 8. The liquid used in this seal maybe mercury, petroleum, or any other that theexperimental conditions indicate.

If the bearing for a stirrer is made of glasstube, it is desirable to lubricate rather freely ;


41/124

LABORATORY GLASS-BLOWING 29otherwise heat will be produced by the frictionof the stirrer and the tube will probably crack.Such lubrication may be supplied by turningout the top of the bearing tube and filling the

turned-out portion with petroleum jelly mixedwith a small quantity of finely ground or, better,colloidal graphite, and the bearing should also


42/124

30 LABORATORY GLASS-BLOWINGbe lubricated with the same composition. Careshould be taken not to employ so soft a lubri-cant or so large an excess as to cause* it to rundown the stirrer into the liquid which is beingstirred.

Leading a Crack. It sometimes happens thata large bulb or specially thin-walled tube hasto be divided. In such a case it is scarcelypracticable to use the method recommended forsmall tubes on page 12, but it is quite easy tolead a crack in any desired direction. A con-venient starting point is a file cut ; this istouched with hot glass until a crack is initiated.A small flame or a bead of hot glass is nowused to heat the article at a point about aquarter of an inch from the end of the crackand in whatever direction it has to be led. Thecrack will now extend towards the source ofheat, which should be moved farther awayas the crack advances. In this manner a crackmay be caused to take any desired path andcan be led round a large bulb.

CuttingGlass with the Diamond.

Slipsof

window-glass can be used in place of glass rodfor some purposes, and as cutting them involvesthe use of the glaziers' diamond or a wheel-


43/124

LABORATORY GLASS-BLOWING 31cutter, they may well be mentioned under thisheading.

In cutting a sheet of glass with the diamond,one needs a flat surface on which to rest theglass, and a rule against which to guide thediamond. The diamond should be held in analmost vertical position, and drawn over thesurface of the glass with slight pressure. Whilethis is being done the angle of the diamondshould be changed by bringing the top of thehandle forward until the sound changes fromone of scratching to a clear singing note.When this happens the diamond is cutting.A few trials will teach the student the correctangle for the diamond with which he works,and the glass, if properly cut, will break easily.If the cut fails it is better to turn the glass overand make a corresponding cut on the other siderather tHan make any attempt to improve theoriginal cut. The diamond is seldom used forcutting small glass tubes.The use of the wheel-cutter calls for nospecial mention as it will cut at any angle,although the pressure required is somewhatgreater than that needed by most diamonds.


44/124

CHAPTER IIIInternal Seals, Air-Traps, Spray Arresters, Filter-PumpsSprays,

Condensers ; Plain, Double Surface, and Spherical Soxhlet Tubes andFat Extraction Apparatus Vacuum Tubes, Electrode Work, EnclosedThermometers, Alarm Thermometers, Recording Thermometers,"Spinning" Glass.

Internal Seals. It is convenient to classthose cases in which a glass tube passes throughthe wall of another tube or bulb under theheading of " Internal Seals," These are metwith in barometers, spray arresters, and filterpumps, in condensers and some forms ofvacuum tube. The two principal methods ofmaking such seals will be considered first andtheir special application afterwards.An Air Trap on a Barometer Tube. Thisinvolves the use of the first method, and is per-haps the simplest example that can be given.Fig. 9, a, ai and #2, show the stages by whichthis form of internal seal is made. For the firsttrials, it is well to work with fairly thick-walled


45/124

LABORATORY GLASS-BLOWING 33tubing, which should be cut into two pieces,each being about eight inches long.


46/124

34 LABORATORY GLASS-BLOWINGthick walled bulb. Fuse the end of the othertube, attach a piece of glass rod to serve as ahandle, and draw out ; cut off the drawn-outportion : leaving an end like a.Now heat a small spot at the end of thebulb, blow, burst out, and remove the thinfragments of glass. Heat a zone on theother tube at the point where the drawn-out portion commences and expand as shownby ai.The next stage is to join the tubes. Heatthe ragged edges of the burst-out portion untilthey are thoroughly rounded. At the sametime heat the drawn-out tube to just belowsoftening point. Then, while the roundededges of the burst-out portion are still soft,insert the other tube ; rotate the join in theblowpipe flame until it is quite soft, and ex-pand by blowing. If necessary, reheat andexpand again. The finished seal, whichshould be slightly annealed by smoking in asooty flame, is shown by az.A Spray Arrester. This is made by thesecond method, in which the piece of tubewhich projects inside the bulb is fused inposition first and the outer tube is then joined


47/124

LABORATORY GLASS-BLOWING 35on. The various stages of making areillustrated by b, bi and 62, Fig. 9.A bulb is blown between two tubes by themethod given on page 22, the larger tube is thencut off and the small piece of tube introducedinto the bulb after having been shaped asshown in by b, Fig. 9. The opening in thebulb is sealed as shown by bi. The sealedpart is now heated and the bulb inclineddownwards until the inner tube comes incontact with the seal and is fused in position.This operation requires some practice in orderto prevent the inner tube either falling throughthe soft glass or becoming unsymmetrical. Theend of the bulb, where the inner tube comes incontact with it, is now perforated by heatingand blowing, thus giving the form shown by2, and the outer tube is joined on. The

finished spray arrester is shown by ^3.Practice alone will give the power to producea symmetrical and stable piece of work.Two Forms of Filter Pump. That illus-trated by d, Fig. 9, is made by the methodexplained under " An Air Trap on a Baro-meter Tube." That illustrated by c is madeby the method explained under "A Spray


48/124

36 LABORATORY GLASS-BLOWINGArrester." No new manipulation is involved,and the construction should be clear from astudy of the drawings.

Multiple and Branched Internal Seals. Afuller consideration of these will be found onpage 39, but one general principle may well beborne in mind ; that, as far as is possible, atube having both ends fastened inside anothertube or bulb should be curved or have a spiralor bulb at some point in its length, otherwiseany expansion or contraction will put greatstrain on the joints.

Sprays. A spray which is easy to make,easy to adjust, and easy to clean after use isshown by e, Fig. 9. The opening on the topof the bulb is made by melting on a bead ofglass, expanding, bursting, and fusing theragged edges. The two branches which formthe spray producing junction are made by themethod used for an exhaustion branch anddescribed on page 18.A spray which can be introduced throughthe neck of a bottle is shown by h, Fig. 9.The various stages in making this are illustratedby f, and g. If the inner tube is made bydrawing out from a larger piece of glass so


49/124

LABORATORY GLASS-BLOWING 37that two supporting pieces are left on each sideof the place where it is intended to make thefinal bend, that bend can be made in a fiat-flame gas burner without causing the innertube to come in contact with the walls of theouter tube. Care must be taken when joiningon the side piece that the inner tube is notheated enough to fuse it. The small hole inthe side of the outer tube is produced byheating and bursting.A Liebigs Condenser. This consists of astraight glass tube passing through an outercooling jacket. In practice it is better tomake the jacket as a separate piece, and toeffect a water-tight junction by means of twoshort rubber tubes. It may, however, be madewith two internal seals of the class describedunder " A Spray Arrester." There is much lessrisk of these seals cracking if the inner tube ismade in the form of a spiral or has a number ofbulbs blown on it in order to give a certainamount of elasticity.A Double-Surface Condenser. Fig. 10shows a condenser of this nature which issupplied by Messrs. Baird and Tatlock. Itmay be built up in stages as shown by a, b,


50/124

38 LABORATORY GLASS-BLOWINGand c, but the work involved requires consider-able skill, and the majority of laboratory workerswill find it cheaper to buy than to make.

dw Fig. 10A Spherical Condenser. Such a condenseras that shown byf. Fig 10, involves a methodwhich may find application in a number ofcases. The outer bulb is blown from a thickpiece of tubing which has been inserted in a


51/124

LABORATORY GLASS-BLOWING 39smaller piece (see d, Fig. 6) ; then the inner bulbby similar method. It is now necessary to intro-duce the smaller bulb into the larger, and forthis purpose the larger bulb must be cut intohalves. A small but deep cut is made with thefile or glass-blowers' knife in the middle of thelarger bulb, and at right angles to the axis ofthe tube on which it is blown. A minute beadof intensely heated glass is now brought incontact with the cut in order to start a crack.This crack may now be led round the bulb asdescribed on page 30. If the work is carriedout with care, it is possible to obtain the bulbin two halves as shown by d, and these twohalves will correspond so exactly that when thecut edges are placed in contact they will bealmost air-tight. The two tubes from thesmaller bulb should be cut to such a length thatthey will just rest inside the larger, and theends should be expanded. Place the innerbulb in position and fit the two halves of theouter bulb together, taking great care not tochip the edges. If the length of the tubes onthe inner bulb has been adjusted properly, theinner bulb will be supported in position bytheir contact with the tubes on the outer bulb.


52/124

40 LABORATORY GLASS-BLOWINGNow rotate the cracked portion of the outerbulb in front of a blowpipe flame and press thehalves together very gently as the glass softens.Expand slightly by blowing if necessary. If asmall pin-hole develops at the joint it is some-times possible to close this with a bead of hotglass ; but if the bulb has been cut properlythere should be no pin-holes formed. Thecondenser is finished by joining on the sidetubes and sealing the inner tube through by themethods already given. In order to blowbulbs large enough to make a useful condenser,it will be convenient to employ the multiple-jetblowpipe described on page 4.A Soxhlet- Tube or Extraction Apparatus.This involves the construction of a re-entrantjoin where the syphon flows into the lowertube. It is of considerable value as an exerciseand the complete apparatus is easy to make.A large tube is sealed at the bottom and thetop is lipped, as in making a test-tube. Asmaller tube is then joined on by a methodsimilar to that given on page 18, but withoutmaking a perforation in the bottom of the largetube. Heating and expanding by air pressure,first through the large tube, then through the


53/124

LABORATORY GLASS-BLOWING 41smaller tube and then again through the largetube, will give a satisfactory finish to this partof the work.The syphon tube is now joined on to the

large tube as shown by a, Fig. IT, care being

Fig. utaken to seal the other end of the syphon tubebefore joining. The details of the final andre-entrant joint of the syphon tube are shownat the lower part of a. This join is made byexpanding the sealed end of the syphon tubeinto a small, thick-walled bulb, and the bottom


54/124

42 LABORATORY GLASS-BLOWINGof this bulb is burst out by local heating andblowing ; the fragments of glass are removedand the edges made smooth by melting. Asimilar operation is carried out on the side ofthe tube to which the syphon tube is to bejoined. This stage is shown by a. Now heatthe syphon tube at the upper bend until it isflexible, and press the bulb at its end into theopening on the side of the other tube. Holdthe glass thus until the syphon is no longerflexible. The final join is made by heating thetwo contacting surfaces, if necessary pressingthe edges in contact with the end of a turn-pin,fusing together and expanding. The finishedapparatus is shown by c.

Electrodes. A thin platinum wire may besealed into a capillary tube without any specialprecautions being necessary. The capillarytube may be drawn out from the side of alarger tube by heating a spot on the glass,touching with a glass rod and drawing the rodaway ; or the exhaustion branch described onpage 1 8 may be used for the introduction of anelectrode. It is convenient sometimes to carryout the exhaustion through the same tube thatwill afterwards serve for the electrode. The


55/124

LABORATORY GLASS-BLOWING 43electrode wire is laid inside the branch beforeconnecting to the exhaustion pump. Whenexhaustion is completed the tube is heateduntil the soft glass flows round the platinumand makes the seal air-tight. The branch isnow cut off close to the seal on the pump side,a loop is made in the projecting end of theplatinum wire, and the seal is finished bymelting the cut-offend.

Platinum is usually employed for such work,but if care is taken to avoid oxidation it is notimpossible to make fairly satisfactory seals withclean iron or nickel wire. Hard rods of finegraphite, such as are used in some pencils, mayalso be sealed into glass, but it seems probablethat air would diffuse through the graphite inthe course of time.Another method for the introduction of anelectrode is illustrated by d, e,f3.ndg, Fig. n.In this case the bulb or thin-walled tube intowhich the electrode is to be sealed is perforatedby a quick stab with an intensely heated wire-preferably of platinum which is then withdrawnbefore the glass has had time to harden, andthus a minute circular hole is made. Theelectrode is coated with a layer of similar glass,


56/124

44 LABORATORY GLASS-BLOWINGor of the specially made enamel which is soldfor this purpose, inserted into the bulb or tubeby any convenient opening, and adjusted bycareful shaking until the platinum wire projectsthrough the small hole. The bulb or tube isthen fused to the coating of the electrode andthe whole spot expanded slightly by blowing.The appearance of the finished seal is shownby^-. It is well to anneal slightly by smoking.

Thermometers. Apart from the notes onpage 20 with respect to the blowing of asuitable bulb on

capillary tubing there is littleto say in connection with the glass workingneeded in making a plain thermometer. Thesize desirable for the bulb will be determinedby the bore of the capillary tube, the coefficientof expansion of the liquid used for filling, andthe range of temperature for which the thermo-meter is intended.

Filling may be carried out as follows : Fita small funnel to the open end of the capillaryby means of a rubber tube, and pour into thefunnel rather more than enough of the liquid tobe used than is required to fill the bulb.Mercury or alcohol will be used in practice,most probably. Warm the bulb until a few air


57/124

LABORATORY GLASS-BLOWING 45bubbles have escaped through the liquid andthen allow to cool. This will suck a certainamount of liquid into the bulb. Now heat thebulb again, and at the same time heat the capil-lary tube over a second burner. The liquidwill boil and sweep out the residual air, but itis necessary to heat the capillary tube as wellin order to prevent condensation. Allow thebulb and tube to cool, then repeat the heatingonce more. By this time the bulb and tubeshould be free from air, and cooling should givea completely filled thermometer. Remove thefunnel and heat the thermometer to a fewdegrees above the maximum temperature forwhich it is to be used ; the mercury or otherfilling liquid will overflow from the top, and, asthe temperature falls, will recede, thus allowingthe end of the capillary to be drawn out. Re-heat again until the liquid rises to the top ofthe tube, then seal by means of the blowpipeflame. The thermometer is now finishedexcept for graduation ; this is dealt with onpage 75-An Alarm Thermometer. A thermometerwhich will complete an electric circuit when acertain temperature is reached may be made by


58/124

46 LABORATORY GLASS-BLOWINGsealing an electrode in the bulb and introducinga wire into the top, which in this case is notsealed. Naturally, this thermometer will befilled with mercury. There is considerabledifficulty in filling such a bulb without causingit to crack.

Several elaborations of this form are made, inwhich electrodes are sealed through the wallsof the capillary tube, thus making it possible todetect electrically the variation of temperaturewhen it exceeds any given limits.An Enclosed or Floating Thermometer.The construction of this type of thermometer isshown by h and z, Fig n. It is made in thefollowing stages : A bulb is blown on thedrawn-out end of a thin-walled tube as shownby h. A small bulb is blown on the end of acapillary tube, burst, and turned out to form alip which will rest in the drawn-out part of thethin-walled tube but is just too large to enterthe bulb. The capillary tube is introduced andsealed in position, care being taken to expandthe joint a little. The thermometer is filledand the top of the capillary tube closed by theuse of a small blowpipe flame. A paper scalehaving the necessary graduations is inserted,


59/124

LABORATOEY GLASS-BLOWING 47and the top of the outer tube is closed as shownby i.A Maximum and Minimum Thermometer.If a small dumb-bell-shaped rod of glass ormetal is introduced into the capillary tube of ahorizontally placed, mercury-filled thermometerin such a position that the rising mercurycolumn will come in contact with it, the rodwill be pushed forward. When the mercuryfalls again the rod will be left behind and thusindicate the maximum temperature attained.If a similar dumb-bell-shaped rod is introducedinto an alcohol-filled thermometer and pusheddown until it is within the alcohol column, itwill be drawn down by surface tension as thecolumn falls ; but the rising column will flowpassed it without causing any displacement ; thusthe minimum temperature will be recorded.

Six's combined maximum and minimum ther-mometer is shown by b, Fig. n. In this caseboth maximum and minimum records are ob-tained from a mercury column, although thethermometer bulb is filled with alcohol. It isan advantage to make the dumb-bell-shapedrods of iron, as the thermometer can then bereset by the use of a small magnet, another


60/124

48 LABORATORY GLASS-BLOWINGadvantage consequent on the use of metalbeing that the rods can be easily adjusted, byslight bending, so as to remain stationary inthe tubes when the thermometer is hangingvertically, and yet to move with sufficientfreedom to yield to the pressure of therecording column.The thermometer may be filled by thefollowing method : When the straight tubehas been made the first dumb-bell is introducedand shaken down well towards the lower bulb,the tube is now bent to its final shape and thewhole thermometer filled with alcohol as de-scribed on page 44. Now heat the thermometerto a little above the maximum temperature thatit is intended to record, and pour clean mercuryinto the open bulb while holding the ther-mometer vertically. Allow to cool, and themercury will be sucked down. The seconddumb-bell is now introduced, sufficient alcoholbeing allowed to remain in the open bulb toabout half fill it, and the alcohol in this bulb isboiled to expel air. The tube through whichthe bulb was filled in now sealed.

Clinical Thermometers. The clinical ther-mometer is a maximum thermometer of a


61/124

LABORATORY GLASS-BLOWING 49different type. In this case there is a con-striction of the bore at a point just above thebulb. When the mercury in the bulb com-mences to contract, the mercury column breaksat the constriction and remains stationary inthe tube, thus showing the maximum temper-ature to which it has risen.Vacuum Tubes. There are so many formsof these that it is scarcely practicable ordesirable to give detailed instructions formaking them ; but an application of the variousmethods of glass-working which have alreadybeen explained should enable the student toconstruct most of the simpler varieties. Aninteresting vacuum tube is made which has noelectrodes, but contains a quantity of mercury.When the tube is rocked so as to cause frictionbetween the mercury and the glass sufficientcharge is produced to cause the tube to glow.A Sprengel Tump. This, in its simplestform, is illustrated by a, Fig. 12. Such a form,although highly satisfactory in action, needsconstant watching while in action, as shouldthe mercury funnel become empty air willenter the exhausted vessel. Obviously, thefall-tube must be made not less than thirty4


62/124

SO LABORATORY GLASS-BLOWINGinches long ; the measurement being takenfrom the junction of the exhaustion branchwith the fall-tube to the top of the turned-upend.

(V-Q

\\

VJI m

as

Fig. 12

A MacLeod Pump. One form of this isillustrated by b, Fig. 12. It has the advantage


63/124

LABORATORY GLASS-BLOWING 51that the mercury reservoir may be allowed tobecome empty without affecting the vacuum inthe vessel being exhausted.

"Spinning" Glass. By the use of suitableappliances, it is quite possible to draw out acontinuous thread of glass, which is so thin asto have almost the flexibility and apparentsoftness of woollen fibre ; a mass of suchthreads constitutes the " glass wool " ofcommerce.The appliances necessary are : a blowpipe

capableof

givinga well-formed flame of about

six or eight inches in length, a wheel of fromeighteen inches to three feet in diameter andhaving a flat rim of about three inches wide,and a device for rotating the wheel at a speedof about three hundred revolutions per minute.A very satisfactory arrangement may bemade from an old bicycle ; the back wheelhaving the tyre removed and a flat rim of tinfastened on in its place. The chain driveshould be retained, but one of the cranksremoved and a handle substituted for theremaining pedal. The whole device is shownby Fig. 13.The procedure in "spinning" glass is as


64/124

52 LABORATORY GLASS-BLOWINGfollows : First melt the end of a glass rod andobtain a large mass of thoroughly softenedglass, now spin the wheel at such a speed thatits own momemtum will keep it spinning forseveral seconds. Touch the end of the meltedrod with another piece of glass and, without

Fig. 13

withdrawing the original rod from the blow-pipe flame, draw out a thread of molten glassand twist it round the spinning wheel. If thisis done properly, the thread of glass will grip


65/124

LABORATORY GLASS-BLOWING 53on the flat rim, and by continuing to turn thewheel by hand it is possible to draw out acontinuous thread from the melted rod, whichmust be advanced in the blowpipe flame as itis drawn on the wheel. If the rod is notadvanced sufficiently the thread will melt off,if it is advanced too much, so as to heat thethick part and allow the glass to become toocool at the point of drawing out, then thethread will become too thick, but it is easyafter a little practice to obtain the right con-ditions. Practice is necessary also in order tofind the right speed for the wheel.When sufficient glass has been "spun," thewhole "hank " of thin thread may be removedby drawing the thumb-nail across the wheelat any point on its flat rim, thus breaking thethreads, and allowing the "hank" to open.

Brushesfor Use tvttk Strong Acids. Glasswool, if of fine enough texture to be highlyflexible, can be used to make acid-resistingbrushes. A convenient method for mountingthe spun glass is to melt the ends of the threadstogether into a bead, and then to fuse the beadon to a rod ; thus giving a brush. If a pointedbrush is necessary, the point may be ground on


66/124

54 LABORATORY GLASS-BLOWINGan ordinary grindstone or carborundum wheelby pressing the loose end of the spun glassagainst the grinding wheel with a thin piece ofcardboard.When using brushes of this description, it is

well to bear in mind the fact that -there isalways a liability of a few threads of glassbreaking off during use.


67/124

CHAPTER IVGlass, Its Composition and Characteristics. Annealing. Drilling,

Grinding, and Shaping Glass by methods other than Fusion. Stop-cocks. Marking Glass. Calibration and Graduation of Apparatus.Thermometers. Exhaustion of Apparatus. Joining Glass and Metal.Silvering Glass.

THERE are three kinds of glass rod and tubingwhich are easily obtainable ; these are soda-glass, which is that usually supplied by chemicalapparatus dealers when no particular glass isspecified ; combustion-glass, which is suppliedfor work requiring a glass that does not soeasily soften or fuse as soda-glass ; and lead-glass, which is less common. There are alsoresistance-glass, made for use where very slightsolubility in water or other solutions is desirable,and a number of other special glasses ; but ofthese soda-glass, combustion-glass, lead-glass,and resistance-glass are the most importantto the glass-blower whose work is connectedwith laboratory needs.

Soda-Glass. Consists chiefly of sodium


68/124

56 LABORATORY GLASS-BLOWINGsilicate, but contains smaller quantities ofaluminum silicate, and often of calcium silicate ;there may also be traces of several othercompounds.The ordinary soda-glass tubing melts easilyin the blowpipe flame, it has not a longintermediate or viscous stage during fusion, butbecomes highly fluid rather suddenly ; it doesnot blacken in the reducing flame. Bad soda-glass or that which has been kept for manyyears, tends to devitrify when worked. Thatis to say the glass becomes more or lesscrystalline and infusible while it is in the flame ;and in this case it is often impossible to dogood work with that particular sample of glass ;although the devitrification may sometimes beremedied by heating the devitrified glass to ahigher temperature. The presence of aluminumcompounds appears to have some influence onthe tendency of the glass to resist devitrification.Soda-glass, as a rule, is more liable to crack bysudden heating than lead-glass, and articlesmade from soda-glass often tend to crackspontaneouly if badly made or, in the case ofheavier and thicker articles, if insufficientlyannealed.


69/124

LABORATORY GLASS-BLOWING 57Combustion-Glass. Is usually a glass con-taining more calcium silicate and potassium

silicate than the ordinary "soft" soda-glass.It is much less fusible than ordinary soda-glass,and passes through a longer intermediate orviscous stage when heated. Such a glass isnot very suitable for use with the blowpipeowing to the difficulty experienced in obtaininga sufficiently high temperature. If, however,a certain amount of oxygen is mixed with theair used in producing the blowpipe flame thisdifficulty is minimised.

Resistance-Glass. May contain zinc, mag-nesium, and other substances. As a rule it isharder than ordinary soda-glass, and lesssuitable for working in the blowpipe flame. Itshould have very little tendency to dissolve inwater, and hence is used when traces of alkalior silicates would prove injurious in thesolutions for which the glass vessels are to beused.

Lead-Glass. This, or " flint" glass as it isoften called from the fact that silica in the formof crushed and calcined flint was often used inmaking the English lead-glasses, contains aconsiderable proportion of lead silicate. Such


70/124

58 LABORATORY GLASS-BLOWINGa glass has, usually, a particularly brightappearance, a high refractive index, and isspecially suitable for the production of theheavy " cut-glass " ware.

Lead-glass tubing is easy to work in theblowpipe flame, melts easily, but does not be-come fluid quite so suddenly as most soda-glasses ; articles made from it are remarkablystable and free from tendency to spontaneouscracking, although, as is essential for all theheavy or "glass-house" work, the massivearticles need annealing in the oven.The two chief disadvantages of lead-glassfor laboratory work are that it is blackened bythe reducing gases if held too near to the bluecone of the blowpipe flame, and that it is rathereasily attacked by chemical reagents ; thusammonium sulphide will cause blackening.The effect of the reducing flame on lead isnot altogether a disadvantage, however ; be-cause a little care in adjusting the blowpipe anda little care in holding the glass in the rightposition will enable the student to work lead-glass without producing the faintest trace ofblackening. This, in addition to being avaluable exercise in manipulation, will teach


71/124

LABORATORY GLASS-BLOWING 59him to keep his blowpipe in good order, andprove a useful aid in his early efforts to judgeas to the condition of the flame. It preventsdiscouragement if the student does his pre-liminary work with the soda-glass, but heshould certainly make experiments with lead-glass as soon as he has acquired reasonabledexterity with soda-glass.

Annealing. Annealing is a process bywhich any condition of strain which has beenset up in a glass article, either by rapid coolingof one part while another part still remains hot,or by the application of mechanical stress aftercooling is relieved. Annealing is carried outby subjecting the article to a temperature justbelow the softening point of the glass, main-taining that temperature until the whole articlehas become heated through the thicker part,and then reducing the temperature very gradu-ally ; thus avoiding any marked cooling of thethinner and outer parts first.

For thin glass apparatus of the lamp-blownor blowpipe-made variety in which there are nomarked difference of thickness, such as joins ontubes, ordinary seals, bulbs, etc., there is littleneed for annealing ; and even those having


72/124

60 LABORATORY GLASS-BLOWINGrather marked changes of thickness, such asfilter pumps, can be annealed sufficiently bytaking care that the last step in making isheating to just below visible redness in theblowpipe flame and then rotating in a sootygas flame until covered with a deposit ofcarbon. The article should then be allowed tocool in a place free from draughts and wherethe hot glass will not come in contact withanything.A few of the blowpipe-made articles, such,for

example,as glass stopcocks, need more

careful annealing, and for this purpose a smallsheet-iron oven which can be heated to dullredness over a collection of gas burners willserve. Better still, a small clay muffle can beused. In either case, the article to be annealedshould be laid on a clean, smooth, fireclaysurface, the temperature should be maintainedat a very dull red for two or three hours andthen reduced steadily until the oven is cold.This cooling should take anything from three totwelve hours, according to the nature of thearticle to be annealed. A thick article, or onehaving great irregularities in thickness willneed much longer annealing than one thinner


73/124

LABORATORY GLASS-BLOWING 61or more regular. As a rule, soda-glass willneed more annealing than lead-glass.

Drilling Glass. Small holes may be drilledin glass by means of a rod of hard steel whichhas been broken off, thus giving a more or lessirregular and crystalline end.There are several conditions necessary to

enable the drilling of small holes to be carriedout successfully : the first of these is that the" drill " should be driven at a high speed. Thismay be done by means of a geared hand-drillsuch as the American pattern drill, although asomewhat higher speed than this will give iseven more satisfactory. The second conditionis that the pressure on the drill is neither toolight nor too heavy ; this is conveniently re-gulated by hand. The third condition is thatthe drill be prevented from " straying" overthe surface of the glass ; for this purpose asmall metal guide is useful. The fourth con-dition is that a suitable lubricant be used ; astrong solution of camphor in oil of turpentineis perhaps the most suitable. For commercialwork, a diamond drill is often used, but this isscarcely necessary for the occasional work of alaboratory.


74/124

62 LABORATORY GLASS-BLOWINGLarger Holes in Glass. The method of

drilling with a hard steel rod is not highlysatisfactory for anything but small holes.When a larger hole, say one of an eighth of aninch or more, is needed it is better to use acopper or brass tube. This tube may be heldin an American hand-drill, but a mixture ofcarborundum or emery and water is suppliedto the rotating end. Tube or drill must belifted at frequent intervals in order to allow afresh supply of the grinding material to reachthe end. In this case, also, a guide is quiteessential in the early stages of drilling ; other-wise the end of the tube will stray. The speedof cutting may be increased slightly by makinga number of radial slots in the end of the tube ;these serve to hold a supply of the grindingmaterial.

Grinding Lenses. This is scarcely within thescope of a book on glass-blowing for laboratorypurposes, but it may be said that the lens maybe ground by means of a permutating mould ofhard lead or type-metal. The rough shapingis done with coarse carborundum or emery, andsuccessive stages are carried on with finer andfiner material. The last polishing is by the


75/124

LABORATORY GLASS-BLOWING 63use of jewellers' rouge on the mould, now linedwith a fine textile.

Filing Glass. If a new file, thoroughlylubricated with a solution of camphor in oil ofturpentine, is used, there is but little difficultyin filing the softer glasses. A slow movementof the file, without excessive pressure but with-out allowing the file to slip, is desirable. Aftera time the cutting edges of the file teeth willwear down and it will be necessary to replacethe file by another,

Grinding Stoppers. This is, perhaps, themost common form of grinding that the labora-tory worker will need to perform, and for thatreason, rather full details of the proceedure aredesirable.A very crude form of ground-in stopper maybe made by drawing out the neck and the massof glass which is intended to form the stopperto approximately corresponding angles, wettingthe surfaces with a mixture of the abrasivematerial and water, and grinding the stopper inby hand. Frequent lifting of the stopper isnecessary during grinding, in order to allowfresh supplies of abrasive material to reach thecontacts. When an approximate fit is obtained,


76/124

64 LABORATORY GLASS-BLOWINGthe coarse abrasive should be washed off, carebeing taken that the washing is complete, anda finer abrasive substituted. After a while,this is replaced in its turn by a still finergrinding material.

Such a method of grinding may give a satis-factory stoppering if the angles of the plug andsocket correspond very closely before grindingis commenced ; but if there is a wide differencein the original angles, then no amount of grind-ing by this method will produce a good result.The reason for this is that the plug will becomeso worn in the preliminary grinding as toassume the form of a highly truncated cone ;the socket will assume a reverse form, and theend result will be a loose-fitting plug and socket.

Satisfactory grinding may be carried out bythe use of copper or type-metal cones for thepreliminary shaping. Such cones should bemounted on a mandrel which will fit into thechuck of the American hand-drill and turnedon the lathe to the desirable angle for stop,pering. A number of these cones will benecessary. A number of similar moulds, thatis to say blocks of type-metal or hard lead inwhich is a hole corresponding in size and angle


77/124

LABORATORY GLASS-BLOWING 65to the plug desired, should be made also.These must be rotated, either in the lathe orby other means, and are used for the preliminaryshaping of the plug. If but few plugs are to beground it is unnecessary to provide a means ofrotating the moulds, as the plug may be held inthe hand and ground into the mould in amanner similar to that used in the first methodof stoppering.

Fig. 14


78/124

66 LABORATORY GLASS-BLOWINGWhen the socket and plug have been ground,

by the successive use of cones and moulds, tothe desired angle, so that they correspondalmost exactly, the plug is given its final fittinginto the socket by grinding-in with a fineabrasive, in the manner first described.

Stopcocks. Although it would be morestrictly in keeping with the form of this bookto divide the making of stopcocks into twoparts ; shaping by heat and grinding, we willconsider the whole operation here, and take forour example a simple stopcock such as thatillustrated by Fig. 14.The " blank," /, that is the socket beforegrinding, is made by drawing out a piece offairly thick-walled tubing into the form shownby a. Two zones on this tube are then heatedby means of a small, pointed flame, and the tubeis compressed along its axis, thus producingtwo raised rings as shown by b. Two zones,slightly towards the outer sides of these tworaised rings are heated and the tube is drawnwhile air pressure is maintained within. Thisproduces two thin-walled bulbs or extensionssimilar to those shown by c. One of theseextensions is now broken off by means of a


79/124

LABORATORY GLASS-BLOWING 67sharp blow with the edge of a file or other pieceof metal, and the edges of the broken glass arerounded in the flame. The other extension isleft to serve as a handle. We have now apiece of glass like that shown by d. Now heata spot on the side of this, medially between theraised rings, until the glass is on the point ofbecoming deformed, and bring the intenselyheated end of a smaller tube in contact withthe heated spot. Without disturbing the relativepositions of the two tubes, press the smallertube down on a thin steel wire, so that the wirepasses along the tube and enters the soft glass ;thus forming a projection inside the sockets asshown by e. The wire must be withdrawn,again immediately. When the wire has beenwithdrawn, heat the place where it entered todull redness, in order to relieve any strain ;break off the thin extension, which up to thepresent has served as a handle, round off thebroken edges in the flame, and join on andindent a similar piece of small tubing to theopposite side of the socket ; the socket at thisstage being shown by f. The " blank" forthe socket is now completed, but it must beheated to dull redness in order to relieve strain


80/124

68 LABORATORY GLASS-BLOWINGand be placed in an annealing oven, where itshould be annealed for some hours.The "blank" for the plug offers no special

difficulty ; it is made by heating a glass rodand compressing it axially until a mass havingthe form shown by gt Fig. 14, is produced ; theend of this is heated intensely and brought incontact with the rather less heated side of aglass tube which has been drawn to the shapedesired for the handle ; when contact is madea slight air pressure is maintained in the glasstube, thus producing a hollow join. The endsof the tube are sealed and the bottom of theplug is drawn off, thus giving the finished" blank" as shown by h. This blank is nowheld in a pair of asbestos-covered tongs, heatedto dull redness all over, and transferred to theannealing oven.When cold, the socket is ground out by thesecond method given under " Grinding Stop-pers"; that is to say, by means of type-metal orcopper cone, and the plug is ground to fit in acorresponding mould. When the fit is almostperfect, the transverse hole is drilled in the plug,and the final finishing is made with fine abrasivepowder. Great care must be taken in the


81/124

LABORATORY GLASS-BLOWING 69final grinding that there is no accumulation ofabrasive material in the transverse hole of theplug ; if this is allowed to occur there will be aring ground out of the socket where the holesmove, and the tightness of the finished stopcockwill be lost.Marking Glass. As a preliminary to aconsideration of the methods of graduating and

calibrating glass apparatus, it is convenient toconsider the various methods which are avail-able for marking glass. Among these are, thewriting diamond, the carborundum or abrasivepencil, the cutting-wheel, and etching by meansof hydrofluoric acid. Each produces a differentclass of marking and each is worthy of indepen-dent consideration.

The Writing Diamond. This is the namegiven to a small irregular fragment of "bort"which is usually mounted in a thin brass rod.Such a diamond, if properly selected, has noneof the characteristics of a cutting diamond ;although one occasionally finds so-called " writ-ing diamonds " which will produce a definite cut.These should be rejected.The writing diamond is used in much thesame way as a pencil, but is held more perpen-


82/124

;o LABORATORY GLASS-BLOWINGdicularly to the object, and a certain amount ofpressure is necessary. The mark produced isa thin scratch which, although fairly definite,lacks breadth, and this is a disadvantage wherethe marking has to be read at a distance.This disadvantage may to some extent beovercome by making a number of parrallelscratches.

The Abrasive Pencil. A rod of car-borundum composition may be ground or filedto a point, and this forms a very useful pencilfor general work. The marking produced israther less definite than that produced by awriting diamond, but has the advantage ofbeing broader.

The Cutting Wheel. ''Cutting" in this caseis scarcely the ideal expression, it should ratherbe " grinding," but " cutting " is more commonlyused. Exceedingly good graduations may bemade by the edge of a small, thin, abrasivewheel which is mounted on the end of a smallmandrel and driven by a flexible shaft from anelectric motor or any other convenient sourceof power. The depth of the mark can becontrolled, and very light pressure will suffice.

Etching. This is often the quickest and


83/124

LABORATORY GLASS-BLOWING 71easiest way of marking glass apparatus. Theobject to be marked should first be warmedand coated very thoroughly with a thin film ofparaffin wax. When cold, the marking is madethrough the paraffin wax by means of a needlepoint, and the object is then exposed to theaction of hydrofluoric acid. If a shallow butclearly visible marking is desired, it is well touse the vapour of the acid ; this may be doneby bending up a sheet-lead trough on whichthe object can rest with the marked surfacedownwards. A little of the commercial hydro-fluoric acid, or a mixture of a fluoride andsulphuric acid, is distributed over the bottomof the trough, and the whole arrangement isallowed to stand for about an hour. Theobject is washed thoroughly and the paraffinwax removed, either by melting and wiping offor by the use of a solvent, and the marking isfinished.

If a deep marking is desired, in order thatit may afterwards be filled with some pigment,a better result is obtained by the use of liquidcommercial hydrofluoric acid, which is a so-lution of hydrogen fluoride in water. Theacid is mopped on to the object after the


84/124

72 LABORATORY GLASS-BLOWINGmarkings have been made on the paraffin waxfilm, and allowed to remain in contact for a fewminutes. It is advantageous to repeat themopping-on process at intervals during theetching.

In all cases where hydrofluoric acid is used,or stored, it is of great importance to keep itwell away from any optical instruments, as themost minute trace of vapour in the air willproduce a highly destructive corrosion of anyglass surfaces.Methods of Calibration. In the case of

apparatus for volumetric work, this is usuallycarried out by weighing, although some of thesmaller subdivisions are often made by measure-ment. When the subdivisions are made in thisway it is of importance to see that the walls ofthe tube or vessel to be calibrated are parallel.Great errors arise in some of the commercialapparatus from neglect of this precaution. Aconvenient method of testing for parallelism, inthe case of a wide tube, is to close one endand to weigh in successive quantities of mercury.An observation of the length occupied by eachsuccessive quantity will indicate any change inthe bore. In the case of capillary tubes, it is


85/124

LABORATORY GLASS-BLOWING 73convenient to introduce an unweighed quantityof mercury, measure its length accurately, andthen to move it along the tube in stages, eitherby tilting the tube or by the application ofair pressure. A measurement of the lengthat each stage will indicate whether the bore isapproximately parallel or not. Neither of thesemethods is to be relied on without a carefulexamination of the tube, as it may happen thatthere are local irregularities in the bore whichcompensate for each other, and do not, there-fore, affect the volume of a given length.Obviously, the smaller the quantity of mercurywith which the test is carried out and thegreater the number of observations made, theless risk will there be of such an error. Aliquid, such as water or alcohol, which wets theglass is not suitable for such a test, unlessspecial precautions are taken.When, however, a pipette or burette has to

be calibrated to deliver a certain volume ofwater, the final calibration must be made withthis liquid. Thus, the burette would first becalibrated by weighing in definite quantities ofmercury of say 13*54 grammes (i cc at l5 c .),each of the 1 cc divisions should be marked by


86/124

74 LABORATORY GLASS-BLOWINGsome temporary marking. The burette is nowfilled with a solution of potassium bichromateand sulphuric acid and allowed to soak for sometime ; the bichromate is washed out and distilledwater is put in. Successive quantities of waterare run out of the jet, a fixed time beingallowed for draining, and the weights of thequantities delivered are noted. This procedurewill give the necessary data for altering themarking so that it may correspond to i ccdelivered. Each i cc division is now dividedinto tenths by the method described below.A final verification of the markings should bemade when the subdivision is completed.

Subdivision of Graduations. Mark out thespaces to be subdivided on a sheet of paper.Take a reliable ruler on which any convenientlength is divided into the desired number andplace it across the lines at such an angle thatthe limits noted on the rule exactly bridge thegap. Now draw parallel lines through themarkings.

Copying a Scale. When a scale has beenprepared on paper and it is necessary to copythat scale on the waxed-glass surface foretching, a convenient method is to employ a


87/124

LABORATORY GLASS-BLOWING 75long wooden bar having a sharp needle passing-through it at either end. The scale and objectto be marked are fastened in line with oneanother, and the caliper bar is used from stepto step. The mark is made by moving thebar through a minute portion of a circle, whichprovided that the bar is two or three feet inlength, will not introduce any perceptible errorin a scale of say a quarter of an inch in width.The arrangement is shown by Fig. 15.

r~T F^- *s YIGraduating a Thermometer. Assuming that

the thermometer has been made of carefullyselected tubing in which the bore is paralleland free from any small irregularities, we haveonly to fix the freezing point and boiling point.The intervening space may then be dividedinto 100 (if the thermometer is to be Centi-grade) or 1 80 (if Farenheit). This divisionmay be carried out by the method given under


88/124

;6 LABORATORY GLASS-BLOWING" Subdivisions of Graduations." A thermo-meter should not be calibrated until someweeks after making, as the glass bulb tends tocontract.

Joining Glass and Metal. It sometimeshappens that one needs to make a morepermanent and less flexible joint between aglass and metal tube than can be obtained bymeans of a rubber tube. To this end, any oneof three slightly different methods may be em-ployed. In the method of Chatelier one firstcoats the glass with platinum or silver, whichmay be done by moistening the glass withplatinum chloride or silver nitrate and thenheating to redness ; a layer of copper is thendeposited electrolytically on the treated surfaceof the glass, and soldering is carried out in theusual manner.McKelvy and Taylor call attention to two

other methods in the Journal of the ChemicalSociety for September, 1920. In one of thesemethods the glass is coated with platinum bycovering it with a suspension of platinumchloride in oil of lavender and heating until theoil is burnt off. The metal tube is then tinnedon its inner side and soldered to the prepared


89/124

LABORATORY GLASS-BLOWING 77glass, slightly acid zinc chloride being used asa flux.

In the second method, a joint is made bymeans of the Kraus flux, which consists ofequal weights of zinc oxide, borax, and powderedsoda-glass fused together. This is coated onthe inner surface of the metal tube, and the hotglass tube, which has had the end slightlyflanged to give support, is inserted. Fusion ofthe flux is completed by heating the outside ofmetal tube.

Silvering Glass. In all cases where it isintended to deposit a silver mirror on a glasssurface, thorough cleaning is essential. Pro-longed soaking in a hot solution of potassiumbichromate which has been acidified withsulphuric acid will often prove useful. Theglass should then be washed thoroughly, rinsedin distilled water, and the solution should thenbe used.There are many formulae for the silvering

solution, but that used in Martin's method maybe given :


90/124

78 LABORATORY GLASS-BLOWINGA Nitrate of Silver 40 grammes

Distilled Water 1000 c. cm.B Nitrate of Ammonium 60 grammesDistilled Water 1000 c. cm.

C Pure Caustic Potash 100 grammesDistilled Water 1000 c. cm.

D Pure Sugar Candy 100 grammesDistilled Water 1000 c. cm.

Dissolve and add :Tartaric Acid 23 grammes

Boil for ten minutes, and when cool add :Alcohol 200 c. cm.Distilled Water to 2000 c. cm.

For use take equal parts of A and B. Mixtogether also equal parts of C and D in anothervessel. Then mix both liquids together in thesilvering vessel and suspend the glass to besilvered face downwards in the solution. Or ifa vessel has to be silvered on the inside, thesolution is poured in. In this case, the deposi-tion of silver may be hastened by immersingthe vessel to be silvered in warm water.

In working with a silver solution containingammonia or ammonium salts there is some-times the possibility of forming an explosivesilver compound. It is well, therefore, toavoid keeping such solutions longer than isnecessary, and to bear in mind that any deposit


91/124

LABORATOEY GLASS-BLOWING 79formed by solutions containing both silver andammonia may have explosive properties,especially when dry.


92/124

CHAPTER VExtemporised Glass-Blowing Apparatus The Use of Oil or other

Fuels Making Small Rods and Tubes from Glass Scrap The Ex-amination of Manufactured Apparatus with the View to Discovering theMethods Used in Manufacture Summary of Conditions Necessary forSuccessful Glass-Blowing.

IF, in the early stages of his study of glass-blowing, the student should attempt to workwith the very simplest appliances, it is probablethat his progress will be hindered ; the use ofthe apparatus will require an undue amount ofcare and his attention will be distracted fromthe actual manipulation of the glass. The caseis widely different after he has acquired acertain facility in glass-blowing.A Simple Form of Blowpipe. Althoughthere are even more simple forms than thatdescribed here, we are not concerned withthem. The form described is the simplest withwhich any considerable amount of glass-blowingcan be carried out with certainty.


93/124

LABORATORY GLASS-BLOWING 81This form consists of a tube through which

air may be blown with the mouth, a condensa-tion chamber in which any moisture from thebreath can condense, a blowpipe jet, a sup-porting piece and a source of flame.The tube, condensation chamber, and jet arecombined in the ordinary Black's blowpipe,such as is used for blowpipe tests in qualitativeanalysis ; it consists of a conical tin tube havinga mouthpiece at the small end and a side tubewhich carries a brass jet. A support for sucha blowpipe may be cut out of a piece of brassor tin-plate, and should be fastened to a small,flat, wooden board. A source of flame mayconsist of an ordinary brass elbow, such as isused on gas fittings, and into which a piece ofthin brass tube (the body of a fish-tail burnerfrom which the perforated non-metallic plughas been removed will serve quite well) hasbeen fitted. It is an advantage to flatten thebrass tube somewhat and to file the flattenedend to a slope which corresponds with theangle at

Handbook of Laboratory Glassware Blowing

Documents

Transcript of Handbook of Laboratory Glassware Blowing