22 1
V O L U M E 2 8 , NO. 2, F E B R U A R Y 1 9 5 6 Even smaller amounts may be determined by scaling down the volume of dilution. *$ series of determinations carried out on a number of aliquot? from a hydrolysis reaction indicated satisfactory reproducibility with a deviation expectation of 0.5% (Table I). Certain diazotizable primary aromatic amines, such as the sulfonamides, p-aminophenol, and acetanilide interfere in the method, but many of these may be separated through preliminary extraction procedures. Medicaments commonly used in conjunction with p-acetophenetidide such as aspirin, caffeine, codeine, and antihistaminic compounds do not interfere. When applied t o the assay of p-acetophenetidide in tablets or capsules of aspirin, p-acetophenetidide, and caffeine, the method niai. he used directll- on the miuture without separation of any
of the components. Results of a series of a-says on such a combination are given in Table 11. LITER4TURE CITED
(1) Afanas’ev, B. N., Aptechnoe D e b 2, 21 (1953). (2) .Assoc. Offic. Agr. Chemists, “Alethods of Analysis,” 7th ed., 1950. (3) Bratton, A. C., and Marshall, E. K., Jr., J . Biol. Chem. 128, 537 (1939). (4) Casinj, -I., Ann. chim 41, 611 (1951). (5) Degner, E. F., and Johnson, L. T., AKAL.CHEM.19, 330 (1947). (6) Horn, Detlef, Pharm. Zentralhalle 90,296 (1952). (7) AIiller, E. li.,Am. J . Pharm. 89, 156 (1917). R E C E I V Efor D review
M a y 2 3 . 1955.
Accepted Sovember 7, 1955
Standard Potential of Iodine-Iodine Monocyanide Half Cell DAVID F. BOWERSOX, ELIOT A. BUTLER, and ERNEST
H. SWIFT
Gates and Crellin Laboratories, California lnstitute of Technology, Pasadena, Calif.
The standard potential of the half cell I*(*)f 21ICS = 2ICK 2H7 2e- has been measured in cells wit11 negligible liquid junction potentials. The value calculated at 25“ C. in solutions 0.2 to 4.0 volume formal in perchloric acid is -0.711 volt; the value for a hypothetical half cell 1 molal in iodine is -0.625 volt. The constant for the reaction 1 2 H C S = ICN f I I1 was calculated from soluhility measurements to be 0.85.
+
+
+
+
+
T
ITR.1TIONS are in use ivhich involve the oxidation of iodine to iodine monocyanide in acid cyanide solutions ( I O ) . These titrations do not require such high acid concentrations as do the (torresponding iodine monochloride titrations (19). The study of the iodine-iodine monocyanide half-cell potential was Iii:i.de because of the use that this potential n-ould have in the c.alculztion of equilibria and end point conditions during iodirie innnocyanide titrations. Ilarlier measurements have been made of the iodine-iodinr monocyanide half-cell potential, but the lack of consistency of the results leaves the ditta in question. In 1912 Kovach ( 5 ) studied the iodine-iodine monocyanide system and from hrl, data the value -0.60 volt is calculated for t’he half-cell reaction
I?(lMj
+ 2HCS
=
2ICS
+ 2H+ + 2e-
+
+ H C S = IC?; + I - + H +
-
I
2 Figure 1.
3
Cell construction
1. Iodine-iodine monocyanide half-crll compartment 2, 3. Hydrogen electrode compartments
(2)
was made by a study of the solubility of iodine in hydrocyanic acid solutions, and the value 0.870 3~ 0.009 mole per liter v a s obtained. EXPERKMENTAL
Reagents. Reagent grade chemicals were used in all preprtrations. Volumetric apparatus was calibrated prior t o its use.
TO POTEM IOMETER
(1)
However, from Kovach’s data the iodine-iodide half-cell potential is calculated, with corrections for complex ion formation ( I s ) , to be -0.520 volt, 0.016 volt different from the value given b y Latinier (6). More recently Gaugin ( 2 ) reported an averagr value of -0.640 volt for the half cell of Equation 1. In his study, deviations from the mean were large, becoming 0.180 volt at pH 6. In the present investigation potential measurements were made in solutions which ranged from 0.2 to 4VF in perchloric acid and from 0.06 to 0.3VF in total cyanide, and the value -0.625 0.003 volt was obtained for this half cell. I n the course of the potential measurements a re-evaluation of the equilibrium constant for the disproportionation reaction
1,
A 6 V F perchloric acid d u t iou \viis prr1xiwd from t h t , ti()% acid and was standardized against sodium hydroxide solution which had been standardized against potassium hydrogen phthalate. Standard solut’ions of potassium iod:ite, approximately 0.1 and 0.002T’F, and of silver nitrate, O . l V F , were prepared liv . . weight from the d t s , Sodium cyanide clolutions, 0.6T’F, n eic stand:irtli,ml h y :L modified Liebig titration iust before each exaeriment. h 1O.UOml. portion of The cyanide solution as mixed xvith 2 ml. of OVF sodium hydroxide solution, 2 ml. of 6T’F ammonium hydroxide, and 1 ml. of 1.0VF potassium iodide wlution. This solution w:is diluted to 50 ml. and titrated Ivith silvrr nitrate Polution to the appearance of the silver iodidc precipitate. The required quantity of resuhlinied iodine ivas ground in : i i i immediately hefore each esprrimrnt. 1 tank hydrogen was p a s s r d through :i uxshing chain similar to that described by Kolthoff and Laitinrn ( 4 ) . In early experiments the potassium permanganate solution specified by these IvorkerP was troublesome because of the forimtion of large quantities of mmganese dioxide, and was replaced by a solution 0.51% in chromic. triosidr, 4 V F in phosphoric arid, and 9T’F in sulfuric acid.
Apparatus. The cell assembly, shown in Figure 1, consisted of three 200-ml. lipless beakers connected by glass tubing bridges to a single intermediate vessel. The bridges were prepared from 6-mm. outside diameter glass tubing and ivere fitted with groundglass stopcocks. One of the 200-ml. beakers contained the iodine-iodine monocyanide half-cell solution, and the other two,
222
ANALYTICAL CHEMISTRY
contained hydrogen electrode assemblies. The temperature of the half-cell solutions was maintained a t 25.0' i 0.5" C. by means of water baths. The beaker which contained the iodine-iodine monocyanide half-cell solution had a tightly fitting stopper through u hich passed the bridge, a bright platinum wire electrode, mounted in 6-mm. glass tubing, and an inlet tube which was flared and ground to fit the tip of the pipets used for adding the half-cell solution components, thus permitting their addition without loss of hydrocyanic acid. A magnetic stirrer provided constant stirring throughout the measurements. T w o different types of hydrogen electrodes were used ( 3 , 9); these are shown in Figure 1. KOsignificant advantage was found in one type over the other for the purposes of this study The ele'ctrodes were platinized by the procedure of Lorch (8). The potential measurements were made with a Gray Instrument Co. Model E potentiometer with a Leeds 8: Xorthrup enclosed lamp and scale galvanometer. Procedure. I n experiments made to determine the equilibrium constant for the disproportionation of iodine in the presence of hvdrocyanic acid, two sets of solutions were prepared in glassstoppered flasks. One set was prepared by the addition of weighed portions of sodium perchlorate to standard perchloric acid the other was prepared by mixing standard solutions of sodium cyanide and perchloric acid. The find perchloric acid concentration, sodium perchlorate concentration, and ionic strength were the same in both sets of solutions. Finely ground iodine was added in excess t o all solutions and the mixtures were maintained at 23" 0.2" C. for periods of time up to 7 days. Samples were transferred by a 25-ml. pipet, using rubber bulbs for drawing the solution into the pipet, to 300-ml. iodine flasks which contained 5 ml. of carbon tetrachloride and sufficient hydrochloric acid to make its final concentration 4VF. The partial pressure of the hydrogen cyanide above the pipetted solutions was less than that of water, therefore such solutions could be pipetted without significxnt change in concentration. The solutions weie then titrated with 0.002VF potasaium iodate by the procedure described by Swift (12). Titrations were repeated on successive days until the reproducibility of the results indicated that equilibrium had been achieved. I n making the potential measurements, calculated volumes of water, potassium iodate solution, and perchloric acid were added t o an excess of freshly ground iodine in the iodine-iodiae monocyanide half-cell beaker. The beaker was stoppered and standard sodium cyanide solution was added from a pipet through the inlet tube. The bridges, intermediate vessel, and hydrogen electrode chambers were filled with perchloric acid of the same concentration as the final molal concentration of perchloric acid in the iodine-iodine monocyanide half cell. Potential measurements were made a t 5- to 10-minute intervals until the readings remained constant within 0.1 mv. for a t least 30 minutes. In preliminary evperiments it was found that the cell potential remained constant to 0.1 mv. for 24 hours. DI SCUSSIOW
Disproportionation of Iodine in Hydrocyanic Acid Solutions. The equilibrium constant for the disproportionation reaction (Equation 2) must be known if the concentrations of the species present in the iodine monocyanide half cell are to be calculated. Earlier calculations of the equilibrium constant for this equation by Kovach ( 5 ) and by Lewis and Keyes ( 7 ) depended upon conductivity and vapor pressure measurements, respectively. In each case the acid concentrations at which the measurements were made were appreciably lower than those which were used in the present potential measurements. Therefore, it was concluded that a determination of the equilibrium constant should he made by an independent method under conditions comparable to those used in the potential measurements. The solubility of iodine a t 25' C. in solutions l.00VF in perchloric acid and 0.089VF in sodium perchlorate 13 as found from a series of eight experiments to be 11.96 5 0 12 X 10-4VF. The sodium perchlorate was added in order to keep the ionic strengths in these and subsequent experiments eonstant. I n Table I are data and calculations from experiments made to determine the value of the disproportionation constant. The concentrations of iodine and triiodide ions were calculated from the solubility of iodine determined above, the volume of potassium iodide required for the titration of the sample, and the dissociation constant for the triiodide ion, K = 1.3 X 10-3 a t 25'
Table I.
Determination of Disproportionation Constant = UCN) (I-) ( H + ) at (12)
W3"
c.
(Solutions were l.00VF in perchloric acid, 0.089VF in hydrocyanic acid, a n d 0.089 V F in sodium perchlorate)
KIOs
Expt. 1 2 3 4 5
Required, hImole
V.M
I-,
Ia -,
ICN,
HCN, VM
0,02105 0,02100 0.02136 0.02112 0.02141
0.00700 0.00698 0,00711 0.00703 0,00713
0.00643 0.00641 0.00653 0 00645 0.00654
0.0134 0.0134 0,0136 0.0135 0.0137
0.0756 0.0756 0.0754 0.0755 0.0753 0.870
VM
VIM
Av.
K 0.854 0.853
0.884 0.866
+
0.894
0.009
C. (I ). The iodine monocyanide concentration is equal to t h e sum of the iodide ion and triiodide ion concentrations, while the hydrocyanic acid concentration is the difference between the initial formal sodium cyanide concentration and the iodine monocyanide concentration. The value obtained for the disproportionation constant is 0.870 =t0.009. The values obtained for the constant by other methods are near this value. Calculations from the data of Lewis and Keyes (?) give values ranging from 0.91 to 1.67, with a "weighted mean" of 1.4. Kovach's values for the constant range from 1.17 to 1.50, with 1.38 as the average. Yost and Stone (13) have applied their formation constants for the iodine dicyanide and diiodocyanide complexes to Kovach's data and obtained a correrted average value of 1.50. In the present calculations of the standard iodine-iodine monor~ anide half-cell potential the value 0.87 was used for the disproportionation constant. I t was found that the use of 1.50 for the constant would change the calculated standard potential by only approximately 1 mv. Complexes of Iodine Monocyanide. Yost and Stone (13) obtained the values 1.17 and 2.50 for the association constants of the iodine dicyanide and diiodocyanide complexes, K , = [I(CK)2-]/(ICN) (CS-)and K2 = (T,CN-)/(ICN) (I-), respectively. Hence, as the solutions used in the present investigation were acid and the iodide concentrations were low, these complexes were present only in small concentration. For example, in a solution approximately 0.5VF in perchloric acid, 0.05VF in hydrocyanic acid, 0.05VF in iodine monocyanide, and which is saturated with iodine, the calculated molal concentrations of diiodocyanide and iodine dicyanide are 2 X 10-4 and 1 X lo-", respectively. Less than 0.5% of the iodine monocyanide is complexed even to diiodocyanide. Reference Electrodes. Hydrogen electrodes had the major advantage over other common reference electrodes of permitting virtual elimination of liquid junctions from the cell. The electrodes were found to cause no difficulty either in their construction or operation. The special precaution was made of providing an intermediate vessel between half cells and ungreased stopcocks in the bridges. These stopcocks were opened briefly at the time of potential measurements. These steps were taken to minimize diffusion of components of the iodine-iodine monoc>-anidehalf cell into the hydrogen electrodes. In general, about 20 potential measurements could be made with the two hydrogen electrodes before the potential readings obtained began to differ by more than 0.1 mv. When this occurred, the electrodes were replatinized. Perchloric acid was chosen for use in this study because data for its activity are available (II ) and it evidently has only very slight tendency to form complexes with the half-cell constituents. Iodine-Iodine Monocyanide Half Cell. Analyses made over a 24-hour period of solutions a t 25' C. and a t the acid concentrations used shes-ed no significant change in cyanide concentration. Therefore the half-cell reaction shown in Equation 1 was assumed to be the potential controlling reaction.
223
V O L U M E 2 8 , NO. 2, F E B R U A R Y 1 9 5 6 Table 11. Standard Potential, E " , of Half Cell 12(1M) 2HCN = 2ICN 2H+ 2e- Eeell, -EO, HCN, HC104, ICE,
+
+
+
VF 4
1 "
VM
0 050
0 150
2
0 053
0 247
0 0515
0 148
0 050
0 050
0 0501
0 0099
0.0539
0 252
0,0540
0 I56
0.0513
0 0487
0.0481 0,0494
0 0488 0 0126
0 062 0 062 0 061
0 254 0 259 0 239
0.0%
0 152
0 0520
0 0471
0.050li
0 0094
Volt
Volt
0.6871 0.6840
0.6288 0 6257
(3) or, since
E" = Ecell The addition of
1
0 5
0.2
0.072.5
0 2275
0,063
0 135
0.0558
0 0442
0.0513
0 0087
0.6730 0,6710 0.6820 0.6842 0.7119 0.7111 0.7103 0.7411 0.7502
0.6263 0.6243 0.6227 0.6249 0,6240 0.6232 0.6232 0 . 6196a 0,6287'
0.6625 0.6710 0.6769 0.6762 0.6893 0 6870 0 7125 0.7124 0.7475 0,7556
0 6124' 0,6217 0.6255 0.6248 0.6278 0 6253 0.6235 0.6234 0,6178' 0.6259"
Ar. a
The standard potential was calculated from the
rated solutions. expression
0 6248
+
0.0026 volt
S o t included in calculation of average.
RT
CALCULATIONS AKD RESULTS
I n Table I1 are shown the cell concentrations, measured cell potentials (Ecen),and calculated standard half-cell potentials ( E " )for Equation 1. The calculations of the standard half-cell potential were made with the assumption that the volume molal concentrations of iodine monocyanide and hydrocyanic acid are equal to their activities a t the concentrations involved. The reference state for iodine is taken as 1.0 molal in these calculations of the E" value in order to simplify calculations involving unsatu-
'I?
hi (s12j"z, where siz is the molal solubility of
iodine, to the value of E" gives the half-cell potential for Equation 5
12(s)
+ 2HCS
= 2ICK
+ 2 H + + 2e-
(5)
in which solid iodine is taken as the reference state. The calculated values of E" are reproducible with a standard deriation of 2~0.0026volt through the range of concentrations studied, except a t approximately 0.01VM hydrocyanic acid, the lowest concentration of this species considered. A probable cause for the relatively large random errors a t 0.01VM hydrocyanic acid is that Ea is dependent upon the ratio ( I C S ) / ( H C S ) . The iodine monocyanide concentration is fixed by the quantity of iodate added, is approximately constant, and is subject to errors of the same relative magnitude in every experiment. The concentration of hydrocyanic acid, however, depends upon the difference between the sodium cyanide added and the iodine monocyanide produced. Therefore, the absolute errors in the concentration of hydrocyanic acid remain about constant, but become much larger relatively as the concentration of hydrocyanic acid decreases. The effect of this is seen by the consideration of two cases:
('1) H C S (B) HCN In preliminary experiments attempts were made to determine analytically the concentrations of iodide ion, iodine, and iodine monocyanide present in the cell solutions a t the time of the potential measurements. However, the sampling technique did not entirely eliminate all the small particles of iodine that mere present in the system; consequently this method did not yield satisfactory results. Therefore, the concentrations of species present a t equilibrium were calculated in the following manner. The concentration of iodine was determined from the solubilitj measurements discussed above. The concentration of iodine monocyanide was calculated from the volume of standard sodium iodate solution added, with the assumption that the reduction of iodate by iodine was quantitative. The difference between the total c j anide added and that present as iodine monocyanide gave the concentration of hydrocyanic acid. The hydrogen ion Concentration 11as calculated from the quantities of perchloric acid and sodium iodate added, with correction for the hvdrocyanic acid formed. Further refinement of the calculated concentration values was made by application of the diqproportionation constant.
(ICN) + RT - In F (11) (HCS)
= 0.15VM; = 0 01VM;
I C s = 0 05VJI: I C S = 0 05VM
If the concentration of hydrocyanic acid is changed by 0.003 mole per liter, the change in the calculated E a for case A is 0,0005 volt while that for case B is 0.0068 volt. Thus, the same absolute error in the concentration of hydrocyanic acid causes negligible error in the potential of case ,4and causes serious error in case 13. Because of the large random deviations of the potential values in solutions 0.01VJf in hydrocyanic acid, these potential values nere not used in calculation of the average value. .4 check upon the consistency of the data obtained can be made by the calculation of the iodine-iodide potential from the iodineiodine monocyanide potential and the disproportionation constant for iodine in hydrocyanic acid. The following tabulation of equations indicates the calculation:
I1(lJT) 21r(lM)
+ PIICS + 2HCS L(s) 21-
or
+ 211' + 2 e = 2 I C S + 213' + 21= 12(1.11j = 12(s)+ 2 e =
21CS
SI.' 28816 165 3'1 18 24783
E" = -0 5366 volt
This value for the iodine-iodide potential differs by just 0.0011 volt from the value given by Latinier (6). Thus, the results of this study are found to be consistent with independent potential determinations of related systems. AVALYTICAL APPLICATION
By making use of this half-cell potential of -0.63 volt, one can calculate the potential required for the oxidation of any given quantity of iodine (or iodide) to iodine monocyanide under specified conditions. For example, assume that the solution is
224
ANALYTICAL CHEMISTRY
1.0 molal in hydrogen ion, 0.1 molal in hydrocyanic acid, that the iodine monocyanide formed is 0.01 molal, and that the iodine a t the end point is 10-0 molal; the iodine-iodine monocyanide potential in such a solution would be -0.74 volt. Considering only equilibrium conditions, this potential value indicates that the oxidation could be effected by electrolytically generated bromine, chlorine, or tetrapositive cerium; in addition, under similar conditions, quantitative oxidation of tripositive arsenic and antimony should be obtained. Qualitative experiments have indicated that these predictions ran be realized and quantitative studies are in progress. LITERATURE CITED
(1) Davies, .\I.. and Gwynne, E., J . Am. Chem. SOC.74, 2748
(1952). (2) Gaugin, K., BuZl. soc. chini. P’rance 1948, 1052.
(3) Hildebrand, J. H., J . A m . Chem. SOC.35, 847 (1913). (4) Kolthoff, I. l f .and , Laitinen, H. -1., “pH and Electro Titi-+ tions,” p. 90, Wiley, Sew York. 1944. (5) Kovach, L., 2. p h y s i k . Chern. 80, 107 (1912). (6) Latimer, W. AI., ”Oxidation Potentials,” p. 63, Kew T o r k , Prentice-Hall, 1952. (7) Lewis, G. X., and Keyes, D . B., J . A m . Chem. SOC. 40, 472 (1918). (8) Lorrh, A . E., ISD. Esc. CHEW.,ASAL. ED. 6, 164 (1934). (9) Noyes, .i..I.,and Garner, C. S., J . Am. Chem. SOC.58, 1266 (1936). (10) Oesper, R. E., and Bottger, W., ”Sewer Methods in Volumetric Analysis,” Tan Nostrand, Sew York, 1938. (11) Robinson, R. A . . and Stokes, li. H., Trans. Faraday SOC.45, 619 (1949). (12) Swift, E. H., J . Ani. C h e m . SOC.52, 897 (1930). (13) Yost, D. XI., and Stone. W . E., Ibid., 5 5 , 1889 (1933). RECEIVED for review August 8, 1955. Accepted November 18, 1955. Contribution S o . 2025 from the Gates and Crellin Laboratories of Chemistry, California Institute of Technology.
Determination of Sulfuric and Sulfonic Acids in Sour Oil L. 1. CALI and 1. WEST LOVELAND Sun
Oil Co., Marcus Hook, Pa.
In the sulfonation of lubricating oils a sour oil lajer is obtained which contains sulfonic acids and a small amount of sulfuric acid. Methods of analysis were needed for both acids in order to follow the processing of the sour oil to a finished sulfonate. The aniline precipitation method, used b> others on acid sludges where ~ proied to sulfuric acid contents w e r e 20 to 9 0 7 ~has work satisfactorilj on sour oils where the concentration is 0.01 to 0.25qc. This method has been extended so that sulfonic acids can be determined on the same sample. The presence of sulfur dioxide in the sour oil does not appear to interfere in either analjsis. A t least 10 moles of water for everj mole of sulfuric acid plus sulfonic acid should be present to obtain quantitatiie results. For sour oils the repeatabilitj standard deFiation for sulfuric acid is 0.007% in the range 0.01 to 0 , 2 5 7 ~and for sulfonic acid it is 0.0977~in the range of 3 to 8 7 ~ .
T
HE sulfonation of lubricating oils produces commercially valuable oil-soluble sulfonates, usually called mahogany sulfonates. The sour oil layer contains not only the mahogany sulfonic acids but also small amounts of sulfuric acid and sulfur dioxide which are salt formers. Methods were needed t o folloiv the removal of the salt. formers, to predict yields of the finished product, and to determine losses of sulfonic acid during processing. Alethods have been published for the determination of sulfuric acid in acid sludges or spent acids ( 2 , 6, 10); however, the sulfuric acid concentrations were very high, usually 20 to 90%. The sour oils the authors are concerned Tvith contain 0.01 to 0,3070 of sulfuric acid. Bacon ( 8 ) investigated a number of methods for determining sulfuric acid and acid sludges. One of these methods used aniline in chloroform to precipitate the sulfuric acid. The aniline sulfate was dissolved and titrated with potassium bromate. Parke and Davis (8) studied this same method on sulfuric acid in the presence of organic acids and mineral acids. Quantities as small as 1 mg. of sulfuric acid were analyzed with an accuracy of 0.2 mg. These workers improved the method of Bacon ( 2 )by dissolving the precipitate in hot water and titrating the sulfuric acid released with base to a phenolphthalein end point.
More recently, Keiss and others (10) used this technique to det,ermine sulfuric acid in spent acids and acid sludges. Holzman and Suknarowski (6) have determined sulfuric acid in acid sludges by making use of the partition of sulfuric acid and sulfonic acid between water and amyl alcohol. The sulfuric acid in the water layer is precipitated as barium sulfate and weighed. There is little or no information in the literature on the direct determination of sulfonic acids in sour oils. However, methods are available for the indirect determination of sulfonic acids as the neutralized product. Sodium sulfonates have been analyzed by Marron and Schifferli ( 7 ) using the p-toluidine hydrochloride method. Epton ( 5 ) and Barr, Oliver, and Stubbings ( 3 ) have titrated t’he sodium sulfonates with cetyl pyridinium bromide. Brooks, Peters, and Lykken (4)have proposed a scheme for analyzing petroleum sulfonates, but t’his is a rather lengthy method. The ASTM procedure (1) can also be used on oilsoluble sodium sulfonates. Weiss and others (11)have suggested an improved ‘4STll method, but both of these methods are very time-consuming. Sulfonic acids have been determined in spent acids and acid sludges where concentrations of from a few per cent to more t,han 50% are involved. Holzman and Suknarowski (6) have determined the sulfonic acids in the amyl alcohol extract h>removing the alcohol under reduced pressure and weighing the dried acids. This method has the disadvantage of being t,imeronsuming. Von Pilat and Starkel (9) determined total acidity and corrected for the sulfuric acid present to give a measure of the sulfonic acids plus other acidic materials. Any sulfur dioxide present would cause an appreciable error in the results. JTeiss and others (10) determined sulfonic acids in acid sludges by correcting total acidity for sulfuric acid (determined by the aniline precipitation method) and sulfur dioxide (determined by iodine-thiosulfate titration). Their results, however, were not too satisfactory because of uncertainty in equivalent weights. They recommend the use of the ASTM analysis for determination of the sulfonic acids. This work extends the aniline precipitation procedure to t,he analysis of sour oils for sulfuric acid concentrations of 0.01 to 1.0%. I n addition, the same sample is used to determine sulfonic acids in the range of 3 to 8%. The method has the advantage of simplicity and it requires no special equipment. Sulfur dioxide in acid sludges and spent acids has been de-
