V O L U M E 2 6 , NO. 2, F E B R U A R Y 1 9 5 4
325 Furman, N. H., and Wallace, J. H., J . Am. Chem. SOC.,51, 1449
tween the two methods was excellent, the averages of four titrations by each solution being found to differ by only 0.06%.
(1929).
Gleu, K., Z.anal. Chem., 95,305 (1933). Greenspan, F. P.,and MacKellar, D. G., -4s.4~.CHEM.,20, 1061
ACKNOWLEDGMENT
(1948).
Huckaba, C.E.,and Keyes, F. G., J . Am. Chem. SOC.,70, 1640
The authors wish to thank V. L. Burger of this laboratory for originally suggesting the investigation of the cerawhydrogen peroxide titration and for much helpful advice regarding the analytical techniques used in the work.
(19481. ~--,_ - .
Shanley, E S., and Greenspan, F. E'., Ind. Eng. Chem., 39, 1536 (1947).
Smith, G . F.. and Gets, C. A., IND.ENG.CAEM.,ANAL.ED., 10,304 (1938).
Thorns, H., Arch. Pharm., 225, 335 (1887). Willard, H. H.,and Young, P., J . A m . Chem. SOC.,55, 3260
LITERATURE CITED
(1) Atanasiu, J. A.,and Stefanescu, V., Ber. deut. ehem. Ges., 61, 1343 (1928). (2) Berry, A. J., Analyst, 58, 464 (1933). (3) Fowler, R.M.,and Bright, H. A., J . Research Natl. Bur. Standurds, 15, 492 (1935).
(1933). RECEIVED for reriew October 1, 1953. Accepted October 23, 1953. Work supported b y the U. S. Navy Bureau of Ships under Contract KObs 531667 Presented before the Division of Analytical Chemistry a t the 124th Meeting of the .kMERICAN CHEMICAL SOCIETY, Chicago, Ill.
Use of Sulfuric Acid in the Detection And Estimation of Steroidal Sapogenins HENRY A. WALENS, ARTHUR TURNER, JR., and MONROE E. WALL Eastern Regional Research Laboratory, Philadelphia 18, Pa.
A rapid method for the identification and estimation of steroidal sapogenins was required. It was found that steroidal sapogenins dissolved in sulfuric acid have characteristic ultraviolet absorption spectra in the region 220 to 400 mp, which can be used for the detection and estimation of these substances. Optimum reaction conditions are the use of 0.1 to 5.0 mg. of sapogenin dissolved in 10 ml. of 94% sulfuric acid and warmed at 40" C. for 16 hours. Under these conditions, the determinations are reproducible and follow Beer's law. Constituents of binary mixtures can be spectrophotometrically determined, using appropriate equations based on previously determined absorptivities. More complex mixtures can be determined after preliminary separation by means of chromatography. A qualitative scheme is suggested for the identification of the 13 most commonly occurring steroidal sapogenins, using chromatographic behavior, spectra of the sulfuric acid chromogens, and melting points. The method can be applied only to purified sapogenins, not to crude mixtures.
D
URIK'G the course of a study on the partition coefficients
of steroidal sapogenins, the need for a method for determining relatively small quantities of these compounds became apparent The method finally adopted was based on the studies of Diaz, Zaffaroni, Rosenkranz, and Djerassi (1). They showed that steroidal sapogenins, on treatment with concentrated sulfuric acid, give characteristic chromogens. It was found necessary to make a thorough study of this reaction, and it has been observed that under suitable conditions the reaction can be used both for identification and for estimation of the various Sapogenins. ,4 qualitative scheme has thus been developed for the identification of the 13 most commonly occurring steroidal sapogenins, using chromatographic behavior, spectra of the sulfuric acid chromogens, and melting points. PROCEDURE
General. The sample, preferably 5.0 mg. (although quantities as low as 0.1 mg. can be used) is weighed into a 10-ml. volumetric
flask. Alternatively, a sample in this weight range is dissolved in chloroform, and transferred to a 10-ml. volumetric flask, and the solvent carefully evaporated to dryness. Sulfuric acid, 94% by volume, is added to the 10-ml. mark. The flask is then immersed for 16 hours in a water bath maintained a t 40' C. The flask and contents are then cooled to room temperature and the contents diluted, if necessary, to volume with 94% sulfuric acid. For the qualitative determination of the sapogenins, the ultraviolet spectrum is obtained over the range 220 to 600 m p using a Beckman DU spectrophotometer or, preferably, a Gary recordin spectrophotometer, with 1.0-cm. matched quartz cells, using 949f sulfuric acid as the blank. Four spectral curves, as determined by means of a Gary recording spectrophotometer, are shown in Figure 1. By comparison of the spectrum obtained with Figure 2 and Table I, most sapogenins can be easily identified. Confirmatory methods have been previously described ( 2 ) . Single-Component Samples. For estimation of the quantity of sapogenin present, the absorbance of the sample is measured a t 250 mg. The concentration of the solution can then be calculated. using the absorptivities listed in Table 11, and the amount of sapogenin present can be determined.
Table I. Wave Length Positions and Intensities of Absorption RZaxima of Sulfuric Acid Chromogens of Steroidal Sapogenins Sapogenin Chlororenin Diosgenin Desoxydiosgenin Gitogenin Hecogenin Desoxyhecogenin Hecogenone Kammogenin Xryptogenin Manogenin Markogenin Rockogenin Samogenin Sarsasapogenin Desoxysarsasapogenin Sarsasapogenone Smilagenin Desoxysmilagenin Smilazenone Tigoggnin Desoxytigogenin Yuccagenin Cholesterol
Absorption Maxima, hlp 270.330.416 271; 415; 514 271,312 272, 308 276,350,396 268,348,394 269,347 233,272,349 280.383 276,348,400,468 270,308 273,379 270,308 271,310 272,308 267,310 272,312 273.308 268.310 270; 312 274.296 240,268,405 308,416
Logarithm of molecular absorptivity.
Log ea a t Corresponding hfaxima 3 . 9 6 . 4 . 0 0 3 71 3 . 9 9 ; 4.06: 3 . 6 4 4.00,3.93 4.08, 4.11 4 . 0 6 . 4 . 1 0 ,4 . 2 4 3.97,3.98,3.61 3.95,3.89 4.11,4 02,3.89 3.77,4.06 4 .OO, 4 . 0 1 , 3 . 6 9 , 3 . 4 4 1.00,3.90 3,85,4.07 4.01,3.91 3.98,3.86 3.97,3.92 4.01.3.76 3.98.3.90 4 .OO, 3 . 9 4 4.06.3.86 3 . 9 4 ;3 . 8 8 3.99,4.03 4.11,4.09,3.83 3.95,3.58
326
ANALYTICAL CHEMISTRY
Table 11. Absorptivities of Steroidal Sapogenins a t 250 and 350 Mba Absorption Coefficients Sapogenin 250 m r 350 mp Chlorogenin 16.0 18.4 Diosgenin 18.4 14.6 Gitogenin 18.1 8.8 Hecogenin 16.3 29.6 Kammogenin 18.3 17.5 9.9 Kryptogenin 12.3 hfanogenin l5,3 23.3 Markopenin 16.6 10.9 Rockonenin 17.9 22.4 Sammogenin 17.2 12.0 Sarsasapogenin 15.3 13.9 Smilagenin 16.4 13.9 15.7 Tigogenin 13.9 Yuccaaenin 12.4 28.0 Absorptivity is defined as a = A / b c where A is the absorbance of a solution of thickness b (centimeters) and c grams per liter compared with a n equal thickness of solvent.
SATURATED
'L5
DIHYDROXY
YUCCAGENIN 1.5
W 0
z 4
1.0 0
Lo
m
4
0.5
ECOG E N I N
rDlOSGENlN
PTOGENIN
I
Y UCCAGEN I N
3 MANOGENI
1
Binary Mixtures. The procedure for the qualitative identi-
fication of binary mixtures is essentially the same as for singlecomponent samples. The estimation of the sapogenin content requires the measurement of the absorbance of the solution a t 250 and 350 mr. The concentration of each component can then be calculated by the use of simultaneous equations, using the absorptivities presented in Table 11. Multicomponent Mixtures. A multicomponent mixture must first be separated into fractions, each fraction being either a single component or a binary mixture. This separation can best be accomplished by means of adsorption chromatography using -80-mesh activated alumina or Florisil (Floridin Co., Warren, Pa.) as the adsorbing agent. Both are heated t o 150" C. before use.
-H
HLOROGE N I N
30
450
350 \VELENGTH,
Figure 2.
Graph of Log
mu
c us. Wave Length of Absorption Maxima
Solid columns indicate single sapogenins; hollow columns indicate group6 of sapogenins. T h e range of log 6 is indicated by the cross hatching. the range of w a w length is roughly indicated by the width of the eoluAn
material is removed by filtration, and the filtrate is placed on the column. The residue is again treated with hot benzene cooled, and filtered, and the filtrate is placed on the column. Tde treatment of the residue is repeated until all of the sample is dissolved and has been placed on the column. The elution with benzene is continued until no more sapogenin is removed from the column . _ with this eluent. The next eluent, 5% chloroform in benzene, is then used and the elution continued.- The concentration of 'the chloroform is increased to 2075, and finally 20% ethyl alcohol in benzene is used. The benzene and the 5% chloroform in benzene eluents remove the monohydroxy sapogenins from the column, leaving the dihydroxysapogenins adsorbed. The 20% chloroform in benzene removes most of the dihydroxysapogenins, and the 20% ethanol in benzene removes any remaining dihydroxysapogenins. The procedure with activated alumina differs only in the choice of eluents. The sample is dissolved in 5% chloroform in benzene, eluting with this solvent and then with 20% chloroform in benzene in order to remove the monohydroxysapogenins from the column. Chloroform is then used to remove the dihydroxysapogenins, followed by 20% ethyl alcohol in benzene. The volume of each eluent that is to be used is determined by the size of the column and by the sample. The point at which the next higher eluent is to be used can be determined by collecting 10 to 20 drops of the eluent on a watch glass and evaporating the solvent to dryness. If a residue remains, the same eluent is continued; if there is no residue, the next higher eluent can be used. Each eluent is collected in a separate container, or containers, and the solvent is evaporated to dryness. Each fraction can then be treated as a single component or as a binary mixture as discussed previously.
0 250
300
350 400 WAVELENGTH, m y
450
500
Figure 1. Absorption Spectra of Hecogenin, Tigogenin, Diosgenin, and Yuccagenin Cary recording spectrophotometer
The amount of adsorbing agent used in the chromatography can be greatly 'varied, depending upon the size of the sample and the size of the columns available. With a relatively small sample (1 gram or less), 50 to 100 times as much adsorbent as sapogenin can be used. This ratio can be lowered to 10 times as much adsorbent as sapogenin with larger samples, but should not be lowered to less than 10 to 1. The column should have a relatively small ratio of cross section to length, and the rate of flow should be low. The elution pattern is essentially the same for both of the adsorbing agents. Alumina is the more active adsorbent, but Florisil is preferable as a resin remover. In the procedure using Florisil, the sample is dissolved in as small a volume of hot benzene as possible, and the solution is cooled to room temperature. Any undissolved or reprecipitated
DISCUSSION
Chromatography. One of the primary requirements of the method is that the samples to be used must be purified. Any plant residues or resins remaining in the sample would reart with the sulfuric acid and render any measurements of the absorbance invalid. Most of the interfering substances are removed in the preliminary processing of the sample (2). The remaining interfering substances which are tenaciously held by Florisil can be removed by an adsorption chromatography using this adsorbing agent. At the same time, the sample can be separated into two or more fractions, each of which is a single component or a binary mixture. Several synthetic mixtures, simulating naturally occurring mixtures, were prepared and chromatographed on Florisil with quantitative recovery of the sapogenins. Factors Mecting the Sulfuric Acid Chromogen. In an effort to standardize the reaction conditions, the effect of the concentration of the sulfuric acid and of the length of heating was
32 7
V O L U M E 26, NO. 2, F E B R U A R Y 1 9 5 4 studied. As commercial sulfuric acid varies in concentration from 95 to 98% acid, it was decided to work with an acid concentration of 94% or lower in order to obtain reproducibility. Using manogenin as a typical example, samples in 85 and 94% sulfuric acid (by volume) m r e immersed for 1 hour in a water bath a t 40" C. The absorptivities were obtained and, as shown in Table 111, the values for the 85% sulfuric acid were much lower than the comparable 94y0 series. Moreover, all these absorptivities were not constant after the 1-hour heating period but increased with time.
Table 111. Variation of Absorptivity with Acid Concentration and Length of Heating Period Wave Length, hlb 250 275 295 310 330 350 396
Absorptivity, 85% 16 hours 1.9 12.0 17.3 2.2 17.2 1.3 13.6 1.7 1.0 10.4 1. o 14.4 13.0 0.40
Absorptivity, 94% 16 hours 4.9 16.0 5.6 24.1 5.2 21.9 4.5 17.6 5,4 17.1 7.0 22.5 1.5 11.6
1 hour
1 hour
sapogenins were determined by two different methods, the results of which appear in Table IV. In the first method, samples of the individual sapogenins were treated with sulfuric acid and the absorptivities were determined. For the second method, binary mixtures of varying percentage composition, composed of hecogenin and tigogenin and of manogenin and gitogenin were treated with sulfuric acid. The absorbance of each solution was measured at the various wave lengths, and the absorptivities of the sapogenins were calculated by means of simultaneous equations. On the whole, the values found for the single components and those calculated for the binary mixtures are in good agreement. These values also agree with the absorptivities determined from the complete spectral curves obtained from the Cary recording spectrophotometer. One exception, tigogenin, s h o w nominal discrepancies of 8% a t 275 and 295 mp.
Table I\'. Comparison of A4bsorptivitiesDetermined from Pure Sapogenin and Calculated from Binary Mixtures Wave Length. JIp
230 275 295 310 330 3.j0
Hecogenin Pure Calcd 16.0 16.4 25.6 23.5 20.9 21.1 20.1 19.2 23.3 23.8 30.8 29.6
Tigogenin Pure Calcd. 16.3 15.8 24.0 22.1 17.7 19.3 20.5 21.0 19.1 19.9 14.1 14.3
Llanopenin Pure Calcd. 16.0 15.9 23.9 23.6 21.6 21.1 17.8 17.7 17.9 18.2 23.3 24.1
Gitogenin ___ Pure 17.6 27.4 27.5 29.6
14.0 9.0
Calcd 18.3 27.7 27.9 30.1 14.2 9.1
Heating overnight for a total period of 16 hours resulted in stable, reproducible values which, as shown in Table 111, were much higher than comparable results obtained after 1hour. This was particularly desirable because the sensitivity of the procedure was increased a t least fourfold. Similar results were obtained with the other common sapogenins. Determination of Absorptivities. The absorptivities of several
i'or-7?J .9
m 4
co NC E N T R A T I o N ,
G M;/L
Figure 4. Absorbance of Sulfuric Acid Chromogen us. Concentration of Hecogenin Measurements a t three wave lengths
Determination of Adherence to Beer's Law. In order to utilize the method for the estimation of steroidal sapogenins, it was necessary to determine whether the sulfuric acid chromogen follows Beer's law. From complete spectral curves, obtained FTith a Cary recording spectrophotometer, several wave lengths corresponding to the maxima and minima of some of the sapogenins were selected. Concentrations of 0.0125 to 0.050 gram per liter were prepared, the sulfuric acid chromogen was formed, and the absorbance was measured a t the selected wave lengths, using a Beckman DU spectrophotometer with 1.0-em. quartz cells. Figures 3 and 4 show that the chromogen follows Beer's
.3
Table V.
L
00
.01
.02 .03 .04
.05
.06
C O N C E N T R A T I O N , G M./L
Figure 3. Absorbance of Sulfuric .kcid Chromogen as a Function of Concentration of Sapogenin Measured a t 250
mp
Estimation of Individual and Total Steroidal Sapogenins in Binary Mixtures
Sapogenin Present, hfg. Hecogenin Tigogenin Total 2.0 3.0 1.o 1 .o 3.0 2.0 1.5 3.0 1.5 0 5 2.5 2.0 2.0 2.5 0.5 hfanogenin Gitogenin Total 0.625 1.875 2.5 1.250 1,250 2.5 1.875 0.625 2.5
Sapogenin Found, M g . Hecogenin Tigogenin Total 1.1 1.8 2.9 2.1 0.8 ' 2.9 1.6 1.4 3 .O 2.1 0.2 2.3 0.6 1.8 2.4 Jfanogenin Gitogenin Total 0.65 1.91 2.56 1.26 1.29 2.55 2.00 0.51 2.51
ANALYTICAL CHEMISTRY
328
in the region of 220 to 600 mp with a Cary recording spec t r o p h o t o me t e r. The maxima of these curves, as I well as log e, are shown in Chromamgrsph y Monohydroxy Dihydroxy Table I. All of the sapo1 genins show a maximum in I Sulfuric acid ultraviolet spectrum the region 270 to 275 mp. Sulfuric acid u1t;aviolet spectrum I This is probably due to the 1 Peak at 350 mp Peak at 350 mp spiroketal structure of the E and F rings or to its open chain derivative, since the 3-desoxy and the 3-keto sapogenins also have peaks in this region, as does kryptogenin, which has the E and F rings open. Cholesterol, a nonsapogenin, which also gives a chromogen upon treatment with sulfuric acid, does not have a peak in this region. Present Absent A peak a t 350 mp occurs only with the 12-keto sapogenins, as shown by the fact Peak at 415 mp that nonketonic and 3-keto I s a p o g e n i n s do not have Diosgenin maxima a t this wave length, nor does kryptogenin, a 16,19diketo compound. The 2,3Present i M. P. M. P. ace tat^ I dihydroxysapogenins R ith a Tigogenin 200OC. 2000-208" C: 5-6 double bond have a charSarsasapogenin 200" C. 140'-150' C. acteristic peak near 235 mp, Smilagenin 184" C. 14Oo-16O0C. which is lacking in the Present Absent spectrum of a 3-hydroxy, 5-6 double bond compound such as diopgenin. However, the Rockogenin 220' C. latter compound has characteristic absorption maxima in the region 400 to 600 mp. law in the concentration range used. Similar data have been The absorption maxima of the latter type compound are probably obtained with other eapogenins. due to dehydration of the 3-hydroxyl, followed by the production of a complex conjugated system. For example, 3-desoxydiosACCURACY AND PRECISION genin, which does not have a 3-hydroxyl group, has a spectrum The quantity of steroidal sapogenin in a one-component sample that lacks absorption in the 400 to 600 mp region and resembles can be determined from the absorbance a t 250 mp. Although that of the saturated compounds. this is not a maximum, the most reproducible values were found The spectra of the saturated, noncarbonyl Eapogenins, whether a t this wave length. The reproducibility was within =k2% and monohydroxy, dihydroxy, or desoxy, are essentially similar, exthe average recovery with a number of samples was within 14%. hibiting a typical absorption maximum near 310 mp. An excep For binary mixtures, determinations were conducted a t 250 tion to this is chlorogenin, which has a characteristic peak a t and 350 mp, where best recoveries were obtained. Various mix330 nip. This peak may be due to the effect of 6-CY-hydroxyl, tures of hecogenin and tigogenin and of manogenin and gitogenin which occurs only in chlorogenin. Many sapogenins have charwere prepared. These two mixtures were selected because of acteristic peaks in the region 375 to 600 mp which can be useful their frequent occurrence together in actual practice. The in identifying them. The data are summarized in Figure 2. amount of each sapogenin present in the mixture was calculated From the foregoing discussion, it is apparent that many of the by means of simultaneous equations, using the appropriate absteroidal sapogenins can be characterized by the ultraviolet sorptivities (Table 11). The results are shown in Table V. spectra of the sulfuric acid chromogens. However, a number of From the data, it is apparent that the total sapogenin content sapogenins, particularly the saturated noncarbonyl compounds, of binary mixtures can be determined. The accuracy in the dehave spectra which are too similar to permit conclusive identificatermination of individual sapogenins is not high, although it is If one uses chromatographic adsorption to distinguish tion. evident that relatively good estimates can be obtained. The merely between the easily eluted monohydroxy sapogenins and amount of the major component can be found with reasonably the more tenaciously adsorbed dihydroxysapogenins, and, in good accuracy, whereas the error in the minor constituent is certain cases, utilizes Koflcr micro melting points, a complete large. The precision of duplicate measurements on binary mixidentification of a t least 13 of the commonly encountered sapotures is within &2%. genins can be made with as little as 5 to 10 mg. A qualitative IDENTIFICATION OF STEROIDAL SiPOGEXINS scheme is given in Table VI using these three criteria: chromatographic behavior, spectrum of the sulfuric acid chromogen, Complete spectral curves of the steroidal sapogenins available and melting point. and of some of the derivatives of these compounds were obtained Table VI.
Qualitative Analysis Scheme for Identification of Commonly Occurring Sapogenins Unknown Sapogenin
1
1 -
I
I
1
V O L U M E 2 6 , NO. 2, F E B R U A R Y 1 9 5 4 ACKNOWLEDGMENT
The authors appreciate the assistance of Mary-Anne Morris in obtaining the complete spectral curves of the sulfuric acid chromogens. LITERATURE CITED
(1) Diaz, G., Zaffaroni, A., Rosenkranz, G., and Djerassi, C., J . Org. Chem., 17, 747 (1952).
329 (2) Wall, M. E., Krider, 11. AI., Rothman, E. S Iand Eddy, C. R., J . Biol. Chem., 198, 533 (1952). RECEIVED for review August 20, 1953. Accepted October 19, 1953. S i n t h in a series on steroidal sapogenins. Work done a s p a r t of a cooperative arrangement between the Bureau of Plant Industry, Soils, and .%vicultural Engineering and Bureau of hgricultural a n d Induatrial Chemistry, U. S. Department of Agriculture and the National Institutes of Health, Department of Health, Education, and Welfare.
Rapid Determination of Aromatics in Petroleum Absorption with Picric Acid-Nitrobenzene EDUARDO A. PASQUlNELLl Destileria de Yacimientos fetroliferos Fiscales, f u e r t o Eva feron, Argentina
A rapid and simple method has been developed for determining the aromatics in a wide variety of petroleum fractions boiling below 600" F. With this method a pair of determinations can be performed in only 20 minutes, using not more than 10 ml. of sample and a minimum of equipment. Its precision is io.;% and its accuracy is of the order of =kl?6. The method is specific for aromatic hydrocarbons. Olefins do not interfere, and thus i t can be applied directly to cracking products without an additional determination, such as bromine number, which is often necessary in many of the existing methods. Furthermore, the test furnishes qualitative information.
M
ANY different methods have been reported for the quanti-
tative determination of aromatics in petroleum fractions. They are most varied and include such techniques as sulfonation (18), nitration (IO), hydrogenation (IQ), oxidation (Q),solubility (II), absorption (6), aniline points ( 9 ) , silica gel adsorption ( 1 7 ) , chromatography ( 7 ) , refractive index and specific dispersion (21), ultraviolet spectrometry ( d ) , and Raman spectrometry (IS). Several of these methods involve the use of more than one technique. Although many of these methods have good features and represent great progress in hydrocarbon analysis, none of these possesses all the advantages that one would wish to see in a single analytical method. Some of the methods need very expensive instruments, and most of them are not specific for aromatic hydrocarbons and need complementary determinations such as bromine number, to correct for olefins. Sone of them is general, they cannot be applied indiscriminately to any petroleum fraction, and sometimes they are complex, laborious, and timeconsuming. Furthermore the petroleum industry lacks an easy test to disclose the presence of aromatics. Many of these methods are useful in industry, and some have become standard procedures, but the ideal of accuracy, precision, rapidity, simplicity, specificity, generalness, and economy has not yet been attained in this field. It was decided to investigate the possibility of developing easier and simpler qualitative and quantitative analytical methods for determining petroleum aromatics. The action of picric acid upon petroleum distillates was tested, taking advantage of its ability t o form molecular combinations with aromatic hydrocarbons. These addition complexes are colored and crystallizable, with characteristic melting points that can be used to identify the respective aromatics. Huntress and Mulliken (14) cite many examples of these complexes. The predominating colors are yellow, orange, and red, and in general the picrates are molecular combinations of 1 mole of picric acid for each mole of hydrocarbon. A4nexception is anthracene which, apart from its normal combination, forms another complex of 2 moles of picric acid for each mole of hydrocarbon. Certain aromatic
hydrocarbons such as diphenyl and diphenylmethane forni no true picrates. A survey of the literature revealed that picric acid has seldom been used in the petroleum industry for analytical purposes. Few investigations of picric acid aa an analytical reagent for hydrocarbons have been undertaken. Jones and Wootton ( 16 ) in 1907 identified 1- and 2-methylnaphthalene in a 180' to 210' C. fraction of a Borneo crude through its picrates. Cosciug (8) in 1935 separated picrates of both methylnaphthalenes from a 30"
5w 10 I
1 P 3 4 5
I
10
I 15
REAGENT ADDED, VOL. %
Figure 1.
Influence of Reagent Concentration on Color of Challaco Kerosine6 1. 10% picria acid in nitrobenzene 2. 3. 4.
15% picric acid in nitrobenzene 20 % picric acid in nitrobenzene 55% picrio acid in nitrobenzene
to 200' C. fraction of a Romania crude. I n 1946, Gambrill and Martin ( 1 1 ) developed a method for' the determination of aromatics in gasoline based on the variation in the solubility of picric acid with the change of aromatic content. Sachanen (80) mentions the use of picric acid in separating
