7/29/2019 h Plca Ceth Amino Phen

1/3

JChemEd.chem.w isc .edu Vo l. 75 N o . 4 Ap r i l 19 98 Jou rna l o f Chemica l Educa tion 467

In the Lab ora tory

Quantitative HPLC Ana lysis of a n Analgesic/ CaffeineFormulation: Determination of Caffeine

Glenda K. Ferguson

Depar tmen t o f Chemist ry and Physics, W esleyan Co l lege , M acon , G A 31 21 0

Mod ern instrumental laboratory experiments that makeuse of consumer products can be most instructive. Throughthese investigations, students can gain hands-on experiencewith vital, state-of-the-art an alytical tools and link scient ificsubject material to their own lives. As this leads to height-ened interest, students are bound to understand and retainthe course material more efficiently. Realizing this, educatorshave developed a number of such experiments. Several thatmake use of high-performance l iquid chromatography(HPLC) (112 ) have appeared in t hisJour nal . Beverages (16 ), food products (2, 6, 7, 8), vegetable oils (9), sunscreenproducts (10), and medications (11, 12 ) have been analyzedquan titatively in t hese endeavors.

In 1983, Kagel and Farwell (11) described an interestingquant itative experiment for th e separation of th e compon entsof Vanqu ish tablets: the analgesics acetaminophen (4-acetamido-phen ol) and aspirin (2-acetoxybenzoic acid), and the cent ralnervous system stimu lant caffeine (1,3,7-trim ethylxanth ine).The structures of these compounds are shown in Figure 1.With salicylic acid as an internal standard, Kagel and Farwell(11) reported a separation time of approximately 8 m in on aZorbax C8 (25cm 4.6 mm i.d.) colum n, with a methanol/

acetic acid/water mobile phase. Their experiment entailedmobile phase optimization and quantitative determinationsof caffeine and aspirin in t hese tablets.

W hile Kagel and Farwells experimen t provided u nd er-

graduates with an interesting and practical hands-on intro-duction t o H PLC, it is not without limitations in todaysinstrum ental analysis laborator y. T he Kagel experiment usedan internal standard to m inimize errors, but replicate injectionsof the t hree standard solution s and t he an algesic samp le werenot performed. Since the mean of a series of replicates is gener-ally more accurate than any single value owing to indetermi-nate error cancellations, replicate analyses should lead to amore accurate caffeine determination. These are more feasiblein the undergraduate lab today, owing to the advent of smaller,more efficient columns capable of much shorter analysistimes. O ther H PLC experimen ts that involve analgesic mix-tures have appeared in this Journal (13, 14), but these do not

incorporate qu antitative analyses. Clearly, an u pdat ed qu an-titative experiment for this drug combination is warranted.T his paper describes a modern experiment for th e quan-

tit ative assay of caffeine in a commercially available medicationthat also includes acetam inoph en and aspirin as active ingre-dients, G oodys Extra Strength H eadache Powders. It shouldbe n oted, however, that ou r liquid chrom atographic separationschem e is not limited to th is particular brand of medication,nor is it limited to caffeine as the analyte. Ot her nam e brandssuch as Excedrin, Vanq uish, or generic versions of th ese three-compon ent m ixtures could be analyzed with the H PLCmeth od described h erein. Since our meth od resolves all threecomponents, only minor procedural modifications would benecessary for determinations of other components or even the

simultaneous assays of all active components in these mix-tures. This separation scheme is also suitable for the analysisof caffeine and /or acetamin ophen in caffeine/acetamin oph encomb inations such as Aspirin-Free Excedrin.

O ur separation scheme is approximately 2.5 times fasterthan that reported by Kagel and Farwell (11 ) and has affordedtime for q uadrup licate injections as well as expansion to fourstandard solution s. We used a shorter analytical colum n an ddeveloped an H PLC p rotocol by which compon ents are re-solved in only 3 min. The experiment detailed herein makes

use of a straightforward H PLC apparatus with a C 18 col-umn, an isocratic mobile phase, UV detection at 254 nm,and a basic integrator; an expensive, soph isticated system isnot required. Figure 2 shows a block diagram of ou r experi-mental setup. The separation is both repeatable and rapid.M oreover, th e experiment can be com pleted in a single thr ee-hour period. The experiment is appropriate for any chemistrystudent who has completed a minimum of one year of gen-eral chemistry and is ideal for an analytical or instrumentalanalysis course.

Prior to the development of this quantitative exercise,the retention properties of these compounds and also salicylicacid, th e major degradation p rodu ct of aspirin, on a C18 col-um n with acetonitrile:water:acetic acid:triethylamine m obilephases were stud ied in ou r laboratory (15 ).

Mobile

Phase

Reservoir

LC Pump

In ject or Colu mn Det ect orIntegrat or /

Printout

to waste

Effluent

Signal

Figure 2 . B lock d iag ram of HPLC exper imenta l setup.

F igure 1 . S tuc tures o f compounds

used in this experiment .

N

N N

NCH3

CH3COOH

OCOCH3

OH

NHCOCH3

Aspirin

Acetaminophen

Caffeine

O

O

CH3

N

NN

NCH3

CH3

Caffeine

O

O

CH3

7/29/2019 h Plca Ceth Amino Phen

2/3

4 68 Jou rna l o f Chemica l Educa tion Vo l . 75 N o . 4 Ap r i l 19 98 JChemEd.chem.w isc .edu

In the Lab ora tory

Overview

Stud ents in ou r quan titative analytical chem istry courseperform this experiment in pairs or groups of 3. We typi-cally have 68 students in the course. All of these studentshave taken a year of general chemistry, and the majority havecompleted a year of organic chemistry.

T he m obile phase for the chrom atographic analyses and

stock solutions of acetaminophen and aspirin are providedfor students. The mobile phase is also used by students as thesolvent in solution preparation. Students are expected to readthe experiment and perform calculations pertaining to caffeinestandard preparation by serial dilution before coming to lab.After the calculations are checked by the instructor, groupsprepare and filter their solutions. This usually requires 3045min . Stud ents then pro ceed with th eir H PLC analyses.

Approximately 1 hour provides sufficient time for agroups actual chromatographic analyses. The instructorequilibrates the system for at least 30 min before the first injec-tion. Groups take turns using the chromatograph, performingquadruplicate injection s of all four caffeine standard solut ionsas well as of the analgesic-mixture solution of unkno wn caffeineconcentration. Each group also runs one injection each ofacetamin oph en and aspirin solution s for retention -time as-sessments. Group members are encouraged to alternate theduties of injecting and operating the integrator.

In the analysis of the d ata, stud ents average the caffeinepeak areas for each solution and calculate the relative peakareas by dividing each by t he smallest. E ach group generatesa graph of relative peak area versus caffeine con centrat ion an dthe relationships best-fit equation. From this, studentsdetermine th e concentration of caffeine in their injectedunknown and ultimately the concentration of caffeine in theoriginal analgesic powder. Each group then calculates the ab-solute standard deviation of the mean result. Students write

stan dard laboratory reports complete with sections on pu rpose,procedure, results, and con clusions.

Exp erimental Procedure

Reagents

Caffeine was purchased from Aldrich C hemical Com pany(M ilwaukee, W I) and used as received. Acetonitrile, triethyl-amin e, and glacial acetic acid, all of H PLC-grade, were pur-chased from Fisher Scient ific (Pitt sburgh, PA). E-Pure-gradewater was produced in ou r laborator y. G oodys Extra StrengthH eadache Powders (Goodys Pharm aceuticals, Mem phis, T N )were pu rchased locally.

Apparatus and Operat ing Condit ions

An Acuflow Series II (Scientific Systems, State College, PA)isocratic pum p is used to deliver the eluent at a constant flowrate of 1.5 mL/min. Solutions are injected to fully fill a 10-Lloop on a Rheodyne (Cotati, CA) 7010 injector. Analysesare carried ou t on a H ewlett-Packard (Avond ale, PA) 10 cm 2.1 mm i.d. (5-m particle size) O D S-Hypersil analyticalcolumn with no guard column. An AcuCon 500 (ScientificSystems) variable-wavelength UVvis detector is set at 254nm to monitor absorbance of the effluent. The detector isset for 10-mV ou tput at 1.0 AU. A H ewlett-Packard Mod el3395 integrator is operated at a chart speed of 1 cm/min to

record chromato grams and calculate peak areas.

Studen t Proced ure

Provided Solut ions: Mobi le Phase and AnalgesicStandards

The mobile phase used in this experiment is a mixtureof 94.1:5.5:0.2:0.2 (v/v/v/v) water:aceton itrile:triethylamin e:acetic acid. It was prepared by placing 110.0 mL of acetonitrile,4.0 mL of triethylamine, and 4.0 mL of acetic acid into a 2-L

volumetric flask and diluting to volume with E-Pure water.The mobile phase was mixed thoroughly, filtered through aM agna 0.45-m nylon filter mem brane (Micron Separations,Westboro, MA), and degassed for 5 min with helium beforeuse. T his mobile phase will be used as the solvent in the p repa-ration o f the solutions described below. In ad dition , filteredsolutions of acetaminophen and aspirin at concentrations of1.00 10 5 g/mLare provided.

Prepa ration o f Caffeine Standa rd Solutions

Weigh 1.0000 g of caffeine and transfer it to a 100-mLvolumetric flask. Add approximately 50 mL of solvent, andswirl and heat gently on a hot-plate. O nce dilution is com-plete, dilute to volume to give rise to a 1.00 10 2-g/mL

stock solution. Perform serial dilutions of this solution toprepare solution s of the following concentrations (in g/m L):1.00 10

3,1.00 104, 5.00 10

5, 1.00 105, 5.00 10

6.T he last four of these will be used as stand ards in the acqu i-sition of the calibration curve data. Filter approximately 3 mLof each solution with the use of Cameo 0.22-m nylon sy-ringe filters (M SI, Westboro, M A) and place in labeled vials.

Prepa ration o f An algesic M ixture Solution

Grind and mix the entire contents of one Goodys anal-

gesic powder with a m ortar and pestle to insure homogeneity.Weigh approximately 0.5 g of the sample to the nearest 0.1mg and transfer to a 100-m L volum etric flask. Add appro xi-mately 50 mL of solvent and swirl to aid dissolution. Dilute

to volume with solvent. Perform a 1:10 dilution to prepare10 mL of the solution to be injected. Filter approximately 3 m Lof th is solution as described for the caffeine stan dards.

Determination of Retention Times of Aspirin,A cetaminophen, and Caffeine

Inject 10 L (the loop volume) of the 1.00 105-g/mL

caffeine solution. Repeat with the provided solutions of aspirinand acetaminoph en at the same concentrations. Record theretention times of these three compounds.

Calib ration Cur ve: Collection and Treatment of Data

Perform q uadru plicate 10-L injections with each of th ecaffeine standard solution s. Du ring each r un , record th e area

of the caffeine peak from the integrator. Calculate the averagearea at each concentrat ion an d t he relative areas by dividingeach average area by the smallest. Plot a calibration curve ofrelative peak area against concent ration (g/m L) with t he useof C A-Cricket Graph III (C omp uter Associates, Malvern, PA).

Q uantitative A nalysis of A nalgesic Pow der

Analyze the analgesic sample solution in quadruplicate10 -L injections. Divide each caffeine peak area value by thesmallest average area of the standards from the previous step.For each injection, determine the concentration of caffeinein the solution. With these data, determine the total mass ofcaffeine in the original 100-mL Goodys powder solution and

calculate the percent by mass of caffeine in the powder sample

7/29/2019 h Plca Ceth Amino Phen

3/3

JChemEd.chem.w isc .edu Vo l. 75 N o . 4 Ap r i l 19 98 Jou rna l o f Chemica l Educa tion 469

In the Lab ora tory

a caffeine standard is shown in Figure 3b. The acetaminophenand aspirin peaks occur at 0.7 m in and 1.4 m in, respectively.Caffeine shows a retention time of 2.2 min . In p revious studies,we have found salicylic acid to elute at 1.3 min. Figure 3ashows no such peak in our analgesic sample. While a salicylicacid peak may be present in analyses of older samples, thisshould not interfere in the quantitation of caffeine.

Figure 4 shows a typical peak area versus concentration

graph indicating the linear relationship over the concentra-tion range of 1.00 10

4 g/mL to 5.00 106 g/mL. Student

data from ou r th ree group s in the spring of 1997 gave rise toexcellent correlation coefficient s of.999 an d greater. T he per-cent b y mass of caffeine in the G oodys Extra Strength pow-der was determined by students to be in the range of 3.053.10%, with absolute standard deviations of the mean in therange of 0.010.05%.

Numerous interesting modifications can be made to theprocedure. Each group could be given a different analgesicpreparation in order to compare various brands. A compari-son between name bran d produ cts and their generic coun ter-parts could be carried out. Alternately, students could perform

quan titative analyses on all of the active ingredients of t hesemedications by incorporating a set of standards composedof all three analytes.

The separation scheme detailed herein involves the useof relatively simple and in expensive chemicals and equipm ent.With proper care, the C18 column has a relatively long life-time. T his concise exercise demo nstrates both the q ualitativeand the quantitative power of liquid chromatography in asingle experiment. Moreover, students can relate to thevaluable skills they acquire in this lab.

Acknowledgments

I thank the Pittsburgh Conference and the College forfinancial support; with a 1994 Memorial National CollegeGrant and matching institutional funds, chromatographicequipment was purchased. I am also grateful to Lesley J.Cown for helpful editorial suggestions and comments.

Literature Cited

1. Goodney, D. E.J. C hem . E du c.1987 , 64 , 187188.2. M unari, M.; Miurin, M.; Goi, G. J. C hem . Edu c. 1991 ,68, 7879.3. DiNunzio, J. E. J. C hem . Ed uc.1985 , 62 , 446447.4. Strohl, A. N. J. C hem . Ed uc.1985 , 62 , 447448.5. D elaney, M. F. ; Pasko, K. M.; Mauro, D . M.; G sell, D . S. ;

Korologos, P. C.; M orawski, J.; Kroliko wski, L. J.; Warren, F. V.,Jr. J. C hem . Ed uc.1985 , 62 , 618620.

6. O ndrus, M.G .; Wenzel, J.; Zimm erman, G. L. J. C hem . Ed uc.1983 , 60 , 776778.

7. Luo, P.; Luo, M . Z.; Baldwin, R. P. J. C hem . E du c. 1993 , 70 ,679681.

8. Silveira, A., Jr.; Koeh ler, J. A.; Beadel, E. F., Jr.; Monr oe, P. A.J. C hem . Ed uc.1984 ,61 , 264265.

9. Farines, M.; Soulier, R.; Soulier, J. J. C hem . E du c. 1 9 8 8 , 6 5 ,464466 .

10. Davis, M. R.; Q uigley, M.N. J. C hem . Ed uc.1995 ,72 , 279281.11. Kagel, R. A.; Farwell, S. O. J. C hem . Ed u c. 1983 , 60 , 163166.12. Mueller, B. L.; Potts, L. W.J. C hem . Ed uc.1988 , 65 , 905906.13. Beaver, R. W.; Bunch, J. E.; Jones, L. A. J. C hem . Ed uc.1983 ,

6 0 , 10001001.14. H addad, P.; Hutchins, S.; Tuffy, M. J. C hem . E du c. 1983 , 60 ,

166168.15. Ferguson, G. K.; Mantsch, R. A. Presented at the 47th Southeast/

51st Southwest Joint Regional Meeting of th e American C hemical

Society, Memphis, TN , Nov. 30 D ec. 1, 1995; Abstract N o. 236 .Figure 4 . Ca f feine ca l ibra t ion curve.

0

5

10

15

20

0 1 2 3 4 5 6 7 8 9 10

RelativePeakArea

Concentration (10 5 g/mL)

Fi gu re 3 . C h rom a t og ram s ob t a i ned f rom (a ) an ana l ges ic pow der

sample solution a nd (b) a ca f feine stand ard solution o f concentrat ion

5 .0 1 05 g/ mL. Uni t s o f re tention t imes are min. C ondi t ions are

given wi thin the text .

(a ) (b )

for each trial. Finally, calculate the average percent caffeineby mass and the absolute standard deviation of the mean.

Results and Discussion

Studentsmust know the retention times of the componentsof th e analgesic/caffeine form ulation in order t o assess theirseparations. We have chosen to provide students with stock

solutions of acetaminophen and aspirin as described above.Alternately, retention times of th ese comp oun ds for th e chro-matographic system at hand may be given. Students could,on the other hand, prepare all required solutions includingthe acetamino ph en and aspirin samp les as well as the mo bilephase/solvent. T his would, however, necessitate approximately1.52 h of additional laboratory time. If this route is chosen,th e instructor may wish to have students prepare their solutionsin one lab period and perform the chromatography in thenext, especially in cases of limited solvent filtration apparatiand/ or chromatographic equipm ent. W hen solutions are keptfor several days, care shou ld be taken to store them in am bervials to prot ect against decom position.

A typical chromatogram of the Goodys powder com-

ponent separation is shown in Figure 3a. That representing