Determination of Carbon and Hydrogen by Combustion VERA L. LESCHER Esso Laboratoriew, Esso Standard Oil Company, Louisiana Dirision, Baton Rouge, La.
A simplified rapid method for the routine semimicrodetermination of carbon and hydrogen in presented. Liquid samples are slowly vaporized, and the vapors are burned in a flame at the tip of the containing ampoule. Combustion is supported by a stream of oxygen at a very high flow rate. 4 technique for handling solid samples is also presented, hj which two sets of apparatus can be operated simultaneously, thereby allowing six to eight determinations per day; explosion hazards are diminished; and a shorter training period for new analysts is required.
I
I ixin by a delivery pipe of sufficient length (80 mm. being the miniS BECEXT years most of t hc piogrebs mtde in the determum length suggested by Clark and Stillson, 6)to allow cooling of mination of carbon and hydrogen has been in the field of the oxygen. A needle valve is provided for each set of apparatus. microchemistry. Desirable features oi the apparatus developed Combustion Tube and Filling. The combustion tube is made by microchemists have often been adapted advantageously by of quartz, has an inside diameter of 20 mm., is 680 mm. long, and is reduced a t one end to a tip 22 mm. long, 4 mm. in inside diameter, analysts to routine samples of semimicro or macro size. and i mm. in outside diameter. The combustion tube filling is Tunnicliff, Peters, Lykken, and Tueiiimler (16) reported that shown in Figure 2. the merits of their routine macromethod employing the n ~ i v Combustion Tube Heaters. Three heaters are used, as shown dual-unitized apparatus “are principally due to the adaptation of i n Figure 1. The sole purpose of the primary heater is to sdpply heat t o burn completely any carbonaceous residue remaining in previously known practices and t o the empharis placed on ad+ the ampoule after the hulk of the sample has burned. All three quate control of the combustion operations, rather than to t h r ~ heaters are hinged and slide on rods attached to the base of the development of new ideas and techiiiqueq.” Speed was obtained multiple unit set. A device, shown in Figure 3, made from sheet mainly by the simultaneous conibuqtion of two samples. Hallett asbestos and laboratory tongs, is placed adjacent to the front surface of the middle heater during the cooling of the tube, in order ( 6 ) , Koyer, Norton, and Sundhrrg (Id), m d Steyermark ( 1 5 ) t n prevent the stream of air from cooling the heater below its have reDorted successful use of niotors to drive niovxhle bui neri crit iral temperature. A similar device remains confitantly in slowly and automatically over sampler. p l a c ~: ~ d , j a c w , tt o thc csit surface of the final heater. The procedure described herein preserits n technique for burning liquid 3:tmplex whereby the bulk of the sample is kept relativt?ly cool CYLINDER OXYGEN and slom-ly vaporized by heat tipplied nmr the exit end of the sample ampoule. By rupporting the exit end a t a higher level t,h:m thr body of the ampoule, it is possible to hurn the vapors in :t small flame a t the tip without the danger of liquids entering the combustion chxnibw. A fast oxygen rate (300 to 350 nil. pcr minute) results in increased pressure whkh, :tcc:ording to Brodie ( 3 ) , aids in producing completeness of comhustion as well as lessening combust ion and Figure 1. Apparatns for Determination of Carbon and sweeping times. No trouble is esperienced n i t h Hydrogen by Combustion condensation of the water in the tip of the coinI . Combustion boat containing arnporile .4 . Preheater J . Middle heater (movable) R. Ascarite ahuorher bustion tube or in the entrance arm of the water K . Final heater Dehydrite absorher C. absorher, as has frequently been reported by L. Dehydrite absorber D. Delivery line Needle v a l v e M. Ascante absorber E. othcrx. The sealed-in capillary feat,ure used by S . Dehydnte-Ascarite absorber F . Oxygen inlet Primary heater (11101 able) 0. Palladium chloride bubbler and trail 0. Baster and Hale ( 1 ) and adopted by others ( 1 3 ) P , Rotameter H . Quartz coinhiiqtinn tiih? w:ts designed to accomplish this qame purpose. A P P A R A T U S A N D PR0CM)URE
Thv apparatus employed i n tiiis 1 ) 1 ~ ~ ) ~ * ~ ( 1 u ~ . c is ilJust,rated in Figure 1. It consists of conventiona1 type equipment arranged to conform to the technique described in this paper. Z D M
Preheater and Purifier. One preheater, which serves both sets of furnaces, consists of a KA-2 pipe 500 mm. long by 25 mm. in inside diameter, filled with copper oxide wire. The purifying train consists of two sections of K.4-2 pipc in series following the preheater and filled with Xscarite and Dehydrite, respectively. The purified oxygen is piped from the preheater arid purifying
M ~
I Figure 2. 1. Roll of copper gauzi B . Quartz chips C . Platinieed asbestos
1246
HEATER
I
HEATER
1
Combustion Tube Filling D. Copper oxide wire E . Load chromate on copper oxide F . G1a.s wool
1247
V O L U M E 21, NO. 10, O C T O B E R 1 9 4 9 Absorption Train. The absorbers used are of thv simple Midvale t,ypc, Stetser-Sorton modification ( 7 ) . The water absorber is tilled 154th Dehydrite, and glass wool is employed in the bottom of the absorber and on top of the tilling to avoid loss of finer particles. The carbon dioxide absorber is filled with hscarite, escept for a short section of Dehydrite which is placed on top of t,he iiscarite to prevent any possible loss of water vapor. Glass wool is plrtwd a h o w and below the filling i l l this atmrber a l x .
S I Z E OF H O L E D E T E R M I N E D BY D I A M E T E R OF
----
0 2 M M . THICK)
Figure 3.
I h i c e to Prevent (holing o F \liddlr
lleater
111addition to the \vater and carbon dioxide a b s o r l ) c ~ rt.lrc. ~ , al)sorption train includes an additional absorber filled with I)I~hydrite arid a bubbler preceded by a trap. The bubbler contains a solution of palladium chloride (0.03 weight, %) through which the exit gas passes. Thus, a continuous test for incomplete combustion is provided. .4 black precipitate indicates the presencr, of carbon monoxide. Because constant exposure to carbon monoxide in the air also discolors this sohti6n, it should not be used for more than 2 to 3 days. A flowmeter is attached to thta final bubbler preceded by an absorber containing Drierite and glass wool. Procedure. LIQCID SA>IPLES..lssemble the combustioil units, fill and install the combustion tubes, fill and attach thc. ahsorhers, and adjust thc heaters to operate at the following t a sample of approsinlately 0.1 gram, weighed tu 1 0 . 1 nig., f o r the analysis. Weigh liquid samples in ampoules madr from 40-inm. sections of 5 - n ~ noutside . diameter Pyrex tubing sealed a t one end to a 30-nun. length of glass rod and drawn out to a capillary at the other end (see Figure 4). +b inside dianieter of the capillary for use in analyses of light saniples (A.P.I. gravity 50+) may be as small as 0.7 mm.; for use in analysis of heavier samples (A.P.1. gravity 50 - ) it should be larger (approxiniately 1.0 mm.). The capillary walls of all ampoules should be of minimuni thickness, 0.3 nun. After proper selection of thc: ampoule for use ill analysis of a given sample, brcak the capillary a t a point 50 mm. from the sample section. \Gpe the ampoult, with a chamois, and migh.
To till thtl ampoule, iva1'111 i t in a small flame and then cool it \vith dry ict. while the capillary is immersed in the sample. Shake the ampoule so that all of tho sample lies in the end opposite the capillary. -1 15-mm. colurnii of heavy material or a 25-11lnl. column of light material in the main part of the ampoule provides thc. dtssired ncight of sample. Wipe the tube with a cleansing tissue to frets it froin condensed \vatel' vapor from the air and from tracw i ) f the sample. Hold the tube approximately 2.5 em. (1 inch) I'rorn the side oi a Bunacn burner flame with the bubble w a t e d between the wanted arid uiirvanted portions of the sample iieaiwt the heat. Hold a clransing tissue a t the end of the capillary t,o absorb the escess sample as it is driven out. Dry the csntire capillary by passing it lightly through the flame, berid it back slightly a t the end to form a hook for hanging it in the balance, and seal the tip. After permitting the ampoule and sample to reach the temperature of the room, weigh the sample. 110 not permit the saniple to enter the capillary tip of the ampoulo at any time after filling. l'lactk t,he weighed absorbers, joined together (glass to glass) by nieans of rubber tubing approximately 1.9 cm. (0.75 inch) in lrngth, in the train. Move the primary heater along the slide rod as far as possible from the middle heater. Place the asbestos de!rice shown in Figure 3 adjacent to the front surface of the middle heater. Clamp a removable air jet to t,he front slide rod and direct a strong stream of air at a point on the tube 10 or 12.5 cm. (4 or 5 inches) distant from the asbestos device. Place cold damp cloths over the sample portion of the tube and the open primary heater until the sample portion is cool to the touch. Break the capillary of the ampoule a t a point 25 mm. from the enlarged portion of glass containing the sample. Place the portion thus removed in i t n ignited boat and place the ampoule in an inclined position in t8heboat (set: Figure 4) so that the capillary remains above the liquid ltvel of the sample, its tip pointing toward the middle heater. Place the boat in the tube so that the tip of the ampoule is within 2.5 cm. (1 inch) of the rolled copper gauze. Remove the asbestos device and air jet. Place the damp cloth on the tube, over thc sample. Xlaintah a very sniall continuous-burning flame a t the t'ip of the ampoule bv gradually moving the middle heater tom-ard the sample as the cloths are moved backward. Raise the lid of the niiddlc htxater slightly from tinit, to time to observe thta flame.
-
TO COMBUSTION CHAMBER
Figure i. Combustion Boat Containing Sample Ampoule
\\'hen the bulk of the sample has burned, move the middle IieaLer to a position adjacent to the stationary final heater; place the primary heater over the ampoule and return the middle heater to a position adjacent to the primary heater. Turn on the primary heater. (Often the capillary tip seals over during an analysis. Application of additional heat, increases the sample vapor pressure and causes a new opening to be blown through the molten glass. The srtrnple vapors continue burning a t this new opening.)
SECTIONS
OF
GLASS ROD
Figure 5. Compartment Boat for Burning Solid Samples tifter 40 minutes, turn off the primary heater, remove the absorbers from the train, apply rubber policemen to the a r m , and allow the absorbers to rest in the enclosed box for 20 to 40 minutes bctfore being reweighed. SOLIDSAMPLES. Burn solids, or extremely heavy liquids, in a specially prepared porcelain boat made into a number of compartments by fusing into it 1-cm. sections of glass rod at Zcm. intervals (see Figure 5). Weigh the same size of sample as in the case of liquids. The rat,e of combustion of the sample can be controlled by watching the flame or by watching the rate of disappearance of the sample, depeiiding upon the nature of t.he material. Other parts of the procedure are the same as described ;tbove. ACCURACY AND PRECISION
The accuracy of the method has been determined by analyses,if one Burrau of St:ind:irds wniple (bcwzoic acid) and several other
ANALYTICAL CHEMISTRY
1248
Table I. Conipoiind Iso-octane
Accuracy and Precision Data for Pure Compounds Carbon, Wt. % 84.22 84.15 84.20 84.16 84.06 84.08 84.33 84.30 84.22
Hydrogen, w t . ”/o 13.96 lZ.83 16.09 16.02 l5,84 15.93 16.01 16,OO 16,06
84.85 85.01 84.77
15.16 15.23
Cetane
Iso-Octane C H Computed value Average value Accuracy6 Precision, S.D.C
84.12 84.19 +0.07 0.09
15.88
15.97 +0.09 0.09
Carbon, Wt. % 91.46 91.40 91 .28
Hydrogen, Wt. % 8.87 8.90
Benzoic acid
68.84 68.76 G8.70
5.01 4.99 5.03
Synthetic mixture a
82.02 81.87 81.04
11.66 11.69 11.iP”
Compound Toluene
8.83
1.5.16
Cetane Toluene C H C H 84.87 1 5 . 1 3 91.25 8.75 8.87 84.88 15.18 91.38 +O.Ol + O 06 + 0 . 1 3 f O . 1 2 0.10 0.03 0.08 0.03
Benzoic Acid
C
H
68.84 4.95 68.77 5.01 -0.07 4-0.06 0.06 0.02
Synthetic Nixture C H 81.92 11.60 81.94 11.69 f0.02 +0.09 0.06 0.03
Acetic acid 4.94 wt. %; ethyl aretate 4.93 wt. % : n-butyraldehyde 3.76 wt. %: methyl isobutyl ketone 3.26 wt. %; isoamyl alcohol 3.87 n t . 70; iso-octane 17.64 wt, % ; n-heptane 17.45 mt. % ; toliiene 44.16 wt. “c Average minus computed value. C
Standard deviation is defined as S.D. =
Table 11.
Sarnple
so. 1 2 3 4 5
6
7 8
9 10 11 12 13 1-1 15 AY.
dZ$-?.
Precision of Procedure for Carbon and Hydrogen by Combustion (Miscellaneous sa.mples) Carbon, Dev. from I-Iydrogen, Wt, % Average Wt. % 13,35 0.05 77.36 13.29 77.26 14.05 0.04 83.23 14.13 83.15 12.74 74.62 0.07 12.93 74.48 13.65 77.84 0 03 13.66 77.79 13.67 77.27 0.06 13.67 77.39 13.70 77.56 0.13 13.53 77.82 14.01 81.11 0.03 14.07 81.05 12.84 83.73 0.07 12.52 83.86 13.40 75.53 0.15 13.28 75.23 13.91 0 11 82.84 13.89 83.07 14.03 0 Oi 83.02 14.06 82.88 82.89 0.14 13.99 13.82 82.62 13.64 78.01 0.11 13.62 77.78 12.80 82.02 0.13 12.78 82.28 13.54 79.73 0.11 13.57 79.50 0.09
Dev. from Average 0.03
Filling Constituent Lead chromate Quartz chips Copper gaiize roll Platinized asbestos Copper oxide nire
0.20/, for carbon and 0.1% for hydrogen. Powers (lb), in his statistical study of accuracy and precision of analytical microdeterminations of carbon and hydrogen, quotes Niederl as saying that *0.2% is acceptable with no values to exceed +0.3%, and found in his own study an existing precision obtained from 349 determinations by 23 experienced microanalysts of 2.9 parts per 1000 of carbon and 22 parts per 1000 of hydrogen. DISCUSSION
C o m b u s t i o n Tube Filling. The purpose of each constib uent of the filling is:
Purpose Sulfur removal Provide large amount of hot contact surface Hot radiating surface enables sample vapors t o reach flash point sooner and allows easy control of their r a t e of burning Catalyst for combustion reaction Catalyst for combustion reaction
0.05
0.14 0.00 0.00 0.09 0.03 0.04 0.06 0.01 0.01 0.05 0.01
0.0’1 0.01
0.04
pure compounds. The results of these tests, shown in Table I, indicate that the accuracy is of the order of 0.06% for carbon aiid 0.08% for hydrogen. The precision of the procedure x i s deierniiiieti from the data in Talile I ; these show an ave1’:rge $ l : i ~ i ( l : ~ deviation d of 0.08% carbon and 0.04% hydrogen. Duplicate analyses of fifteen miscellaneous plant samples are shown in Table I1 t o indicate the precision to be expected on unknown complex Inistures. The precision and accuracy of this method conipare favorably with those of the routine procedures of microchenlists. Steyermark (16) shows an acceptable accuracy of +0.3y0 deviation from theory with many checks within a FC\Y hundredths of 1%. Belcher and Spooner ( 2 )claim rewlts compiirable in accuracy wit8h t,hose obtained with general staridnrd methods of microanalysis-
Combustion Tube Heaters. Hinged-type heat,ers are found advantageous, particularly in the case of the middle heater, because the operator is able to raise the upper half of the heater slightly from time to time in order to view the small flame. Absorbents. Dehydrite and Ascarite were chosen for use in this procedure. They offer the advantages of being solids; as Niederl and Roth (11) reported, Dehydrite absorbs up to 3070 of it,sweight and -4scarite absorbs 10times as much carbon dioxide as does soda lime. Oxygen rates, using previously reported techniques, were necessarily slow in order to allow sufficient residence time for complete coinbu.qt,ion in t’he catalyst-packed combustion tube, inasmuch RS the burning did not all take place a t the capillary tip of the ampoule as is the case in the present method. Rates ranging froni 50 to 250 ml. per minute (8, 8) were considered fast. Millin (8) found complete absorption of water by Dehydrite and of carbon dioxide by Ascarite a t 250 ml. per minute. A test at these laboratories of t.he carbon dioxide absorption capacity of Ascarite showed complete absorption a t rat’es up to 1000 ml. per minute, employing an absorber of tmheMidvale type, StetserSort,onmodification ( 7 ) . Extreme care i n properly packing the Ascarite (8- to 20-mesh) is essential t o prevent channeling. It is advisable to add the re-
Table 111. Time Required for Two Simultaneous .4nalyses by Routine Procedure Operation Weigh absorbers (4) IYeigh samples (2) Connect absorbers, insert s a n i r k Burn all sample vapors Bring sample heater u p t o temp. Burn carbon residue from ampoule R u s h combustion tube Disconnect absorbers and all rubber policemen Equilibrate absorbers T+-eigh absorbers Calculate results Total elapsed time for two analyses Total norking time for two anaiyses
Time, Minutes 8 8 10 26 * 10
*:
15 5
20 8
3
lei 67
V O L U M E 21, NO. 10, O C T O B E R 1949 agent in small portions, tapping the absorber gently after each addition, as suggested by Clark and Stillson (4).Care in selecting, handling, and storing Ascarite is also essential, as some lots of this drying agent were not effective. Time Required for Analysis. Table I11 gives the average time required for the various operations necessary for this procedure. Six to eight determinations can be made in an 8-hour day. The working time can be lessened somewhat by using weighed absorbers from one determination for a subsequent determination and using a smaller sample [50 to 70 ing. have been suggested by Natelson and co-workers (9, l0)l. ACKNOW LEDGBlElVT
The author gratefully Acknowledges the assistance of E. J. Xewchurch, who made inany helpful suggestions in the writing of this paper. LITERATURE CITED
(1) Baxter, G. P., and Hale, A. H., J . Am. Chem. SOC.,58, 510 (1936).
1249 (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)
Belcher, Ronald, and Spooner, C. E., ,J. Chem. SOC.,1943, 313 Brodie, S.S.,IND. ENG.CHEY.,ANAL.ED., 11, 517 (1939). Clark, R. O., and Stillson, G. H., I b i d . , 12, 494 (1940). I b i d . , 17, 520 (1945). Hallett, L. T., Ibid., 10, 101 (1933). I r o n Age, 102, 8 (1918). Millin, D., J . SOC.Chem. I n d . , 58, 215 (1939). Natelson, S.,Brodie, S. S., and Connor, E. B., IND. ENG.CHEM., ANAL.ED.,10, 276 (1938). Ibid., 10, 609 (1938). Niederl, J. B., and Roth, R. T., Ibid., 6, 272 (1934). Powers, F. W.,Ibid., 11, 660 (1939). Renoll, hl. IT., Midgley, T., Jr., and Henne, A. L., Ibid., 9, 566
(1937). (14) Royer, G. L., Norton, A. R., and Sundberg, 0. E., I b i d . , 12, 688 (1940). (15) Steyermark, d l , Ibid., 17, 523 (1945). (16) Tunnicliff, D. D., Peters, E. D., Lykken, Louis, and Tuemiiiler, F. D., I b i d . , 18, 710 (1946). R E C E I V E D>larch 15, 1948. Presented before t h e Southwest Regional SOCIETY,Houston, Tex., December 12 and hleeting, A V K R I C A XCHEXICAL 13, 1947.
Chemical Determination of Tryptophan in Proteins JOSEPH R. SPIES AND DORRIS C. CH.4JIBERS Allergen Research Division, Bureau of Agricultural a n d Industrial Chemistry, U . S. Department of Agriciclture, JVashington 25, D. C. Fundariiental knowledge of the behavior of free and peptide-linked tryptophan is needed for the development of a n accnrate method of analysis of proteins for tryptophan. Described are: a method of alkaline hydrolysis which protects trqptophan from external destruction a t temperatures up to 185’ C. without addition of antioxidants to the solution; the effects of tinie and temperature on the racemization of free tq ptophan heated in 5 N sodium hydroxide; and the effects of free and peptide-linked amino acids on tryptophan heated in 5 N sodium hydroxide. A variable proportion of tryptophan in proteins is destrojed bq cystine, cysteine, lanthionine, serine, and threonine during alkaline hydrolysis but these amino acids do not destroy tryptophan under conditions used for analysis of unhydrolyzed proteins. SeF era1 modifications of a
D
IFFICULTIES in the accurate quantitative determination of tryptophan in proteins and the biological importance of this essential amino acid account for the contentious history of this subject. Whether proteins can be hydrolyzed with alkaline agents without significant destruction of tryptophan is still controversial. I n this laboratory significantly highrr values were obtained on the unhydrolyzed protein than on the alkaline tivdrolyzate, even when hydrolysis was cai r i d out by a procedurc that eliminated external destruction of trvptophan. This observation led to a study of free and peptide-linked tryptophan under alkaline hydrolyzing conditions. The conclusion was that alkaline hydrolysis of proteins preliminary to tryptophan analysis should not be used because of variable amounts of destruction of tryptophan, depending on the amino acid composition of the protein. Results of this study are recorded to explain the fundamental causes of destruction of tryptophan during alkaline hydrolysis of
method for colorimetric analysis of unhydrolyzed proteins are described. The basic method, like t h a t previously described for the determination of free tryptophan, involves two steps-reaction of tryptophan and p-dimethylaminobenzaldehyde in 19 N sulfuric acid to form a colorless condensation product and subsequent development of a blue color by oxidation with sodium nitrite. For greatest accuracy each protein must be analyzed using predetermined optimum conditions for the basic reactions. -4 general procedure, based on studies with eleven proteins, may be used where resulting economy of time outweighs the possible sacrifice of a small degree of accuracy. The precision of the method is +0.880/0 and the accuracy is believed to he *l to 3qo. Tryptophan may be determined in the presence of carbohydrates.
proteins and to provide evidence of the accuracy of the incthod described for the determination of tryptophan in proteins in which these destructive factors are eliminated. A critical evaluation of the accuracy of the values obtained by this method is also presented. Because of its biological importance, the scope and complexity of analytical problems concerning tryptophan are much broader than those involved in protein analysis. These rcsults provide methods and fundamental information based upon \\ hich the analyst and the research worker may dcvise procedures to mcct spccial conditions. APPARATUS AND MATERIALS
A Coleman spectrophotometer, Model 11, was used for colorimetric analyses. A Beckman quartz spectrophotometer was used to obtain the absorption curves shown in Figure 6. Parr nickel microbombs are commercially available. Reagents for colorimetric analysis have been described (SO). Sodium hydroxide was prepared from a saturated “carbonate-
