90 Analyst, February, 1973, Vol. 98, pp. 90-97 Ionic Polymerisation as a Means of End-point Indication in Non-aqueous Thermometric Titrimetry Part II.* The Determination of Organic Acids BY E. J. GREENHOW AND L. E. SPENCER (Defiartment of Chemistry, Chelsea College, University of London, Manresa Road, London, S. W.3) A method for the thermometric titration of organic acids in non-aqueous solution, involving the ionic polymerisation of acrylonitrile to indicate the end-point, has been evaluated for a range of mono- and polyacidic compounds, including phenols, triazine derivatives and tannic acid. In addition to acrylonitrile, methyl acrylate and dimethyl itaconate can be used as monomers for the end-point indication. A number of titrant - catalysts have been examined, including tetra-n-butylammonium hydroxide, n-butyllithium and potassium hydroxide. The precision of the method ranges from about 0.5 per cent.with 0.1 M titrant, to 2.7 per cent., with 0.001 M titrant, by using the manual and semi-automatic methods described in Part I. Sample sizes down to about 0.0001 mequiv, e.g., about 10 pgof benzoic acid, which corresponds to 100 p.p.m. of benzoic acid in the volumes of sample solution titrated, can be determined with 0.001 M titrant. Calibration graphs show the volume of titrant and mass of sample to be linearly or almost linearly related in the range 0 to 2 ml of titrant. Comparison of the titration values obtained by using the acetone method and ionic polymerisation with potassium hydroxide and tetra-n-butylam- monium hydroxide solutions as titrants enables one to differentiate between acidic groups of different ionicstrengths in the 10 to 12 pK, region.THE difficulty in obtaining a sharply defined end-point in the non-aqueous thermometric titration of weak acids, such as phenols, has been pointed out by Vaughan and Swithenbank.1 These authors showed that the end-point could be clearly defined by using acetone as a thermometric end-point indicator. Acetone is used as the solvent for the sample of acid, and, after the latter has been neutralised by the alkaline titrant, further addition of titrant catalyses the exothermic dimerisation of the acetone and a sharp rise in temperature occurs. The general principles of this approach to thermometric titrimetry, which has been given the name “catalytic thermometric titration,” have been discussed by Vajgand, GaAl, Zarnic, Brusin and Velimirovic.2 Catalytic thermometric procedures in which highly exothermic ionic polymerisation processes are used to indicate the end-point in titrations of organic acids and bases in non- aqueous solution have been discussed in Part I3 and in earlier paper^.^^^ When acrylonitrile is used as the solvent for an organic acid it undergoes polymerisation after the acid has been neutralised by the alkaline titrant, if certain strongly basic titrants such as potassium hydroxide in propan-2-01 and tetra-n-butylammonium hydroxide in toluene - methanol are used, provided that a sufficient excess of the titrant is added. With 0.1 M titrants rises in temperature exceeding 10 “C can be obtained with an excess of titrant of about 0.05 ml.The present paper gives a more detailed evaluation of the procedure described in an earlier paper.4 The precision and sensitivity of the method have been studied, as well as its application to a range of mono- and polyfunctional organic acids. In addition to acrylo- nitrile, methyl acrylate and dimethyl itaconate have been investigated as monomer - solvents. A number of compounds known to be effective catalysts for anionic polymerisation have been examined as possible titrants ; those found to be suitable include n-butyllithium and potassium t-butoxide, in addition to the titrants mentioned above. Results obtained in the titration of polyacidic compounds by using acrylonitrile as the end-point indicator and potassium hydroxide and tetra-n-butylammonium hydroxide as titrants have been compared with those obtained by using the acetone procedure of Vaughan and Swithenbank.1 * For Parts I and 111 of this series, see pp.81 and 98, respectively. @ SAC and the authors.GREENHOW AND SPENCER EXPERIMENTAL REAGENTS- 91 Laboratory-reagent grade acrylonitrile, methyl acrylate, toluene, methanol, propan-2-01, 2-methylpropan-2-01, acetone, dimethylformamide and dimethyl sulphoxide were dried over molecular sieve 4A before use. Dimethyl itaconate-Commercial material supplied by Pfizer Ltd. was used as received. Benzoic acid-AnalaR grade (BDH Chemicals Ltd.) was used as received. Tannic acid-Laboratory-reagent grade (BDH Chemicals Ltd.) material (relative mole- cular mass 1701.23 and empirical formula C76H5@46) was used.Other organic acids were laboratory-reagent grade materials and were used without further purification. Tetra-n-butylammonium hydroxide, 0- 1 M in toluene - methanol-Laboratory-reagent grade (BDH Chemicals Ltd.). This reagent was used as received. Prepare more dilute solutions by adding toluene - propan-2-01 (3 + 1) mixture to the 0.1 M reagent. Tetra-n-butylammonium hydroxide solutions, 0.1 to 0.001 M-Standardise the solutions against benzoic acid in dimethylformamide by the thermometric method. Potassium hydroxide, 0.1 M solution in propan-2-ol-Standardise this solution against benzoic acid, with phenolphthalein as indicator, and also by the thermometric method. Potassium t-butoxide, 0.1 M solution in 2-methyl~ropan-2-ol-Prepare this solution by dissolving potassium metal in 2-methylpropan-2-01 and standardise the solution against benzoic acid in dimethylformamide by the thermometric method.n-Butyllithium, 0.1 M solution in toluene-Prepare this solution by diluting the 16 per cent. solution in hexane (Koch-Light) with toluene and standardise it against benzoic acid in dimethylformamide by the thermometric method. PROCEDURE- The manual and semi-automatic methods (Methods A and B) described in Part 1 3 were used with the following modifications. Titrate organic acids as 0-1 N solutions in toluene or dimethylformamide when 0.1 and 0-025 M titrants are used, and as 0.01 N solutions when 0.01 and 0.001 M titrants are used. In precision determinations add the sample as an aliquot of a standard solution with grade A pipettes or, for amounts of less than 1 ml, with an Agla micrometer syringe.With 0.1 and 0-025 M titrants carry out the titrations in a 50-ml beaker insulated with polystyrene foam or in the closed apparatus (capacity 15 ml) (Fig. 1) by using 10 and 5 ml of monomer, respectively; with 0.01 and 0-001 M titrants use the closed apparatus (capacity Scales apparatus 15-ml 1 crn apparatus 1 1 cm Magnetic stirrer Fig. 1. Titration apparatus92 GREENHOW AND SPENCER : IONIC POLYMERISATION FOR END-POINT [Analyst, VOl. 98 5 ml) and 2 and 1 ml of monomer, respectively. Use the methyl acrylate and dimethyl itaconate monomers as 1 + 1 mixtures with dimethylformamide or dimethyl sulphoxide, and not as the pure monomers. End-points are measured as in Part I at the “upturn” temperature [method (i)], at the intersection point of the tangents to the two component parts of the titration curve [method ( 4 1 or by the method of Vaughan and Swithenbank,6 in which the end-point is taken to be the point where the tangent to the main heat rise leaves the curve at its lower temperature end (see Fig. 1, reference 6).All three methods give values showing similar precision, but that of Vaughan and Swithenbank gives the best stoicheiometry and is preferred when a tangent can be drawn unambiguously. RESULTS AND DISCUSSION The precision of the method has been established by titrating aliquots of standard solutions of 3,5-xylenol and benzoic acid in toluene with 0.1, 0.025, 0.01 and 0.001 M tetra- n-butylammonium hydroxide solutions, and the results obtained are summarised in Table I.Precisions are expressed as coefficients of variation of single points about the mean, and range from about 0.5 per cent. for 0.1 M titrant to 2-7 per cent. for 0.001 M titrant. Co- efficients of variation of the means themselves, which depend on the number of replicate titrations, are correspondingly lower. Higher precisions could, no doubt, be attained by adding the sample by mass instead of by volume. Most of the determinations were carried out by using the manual procedure (A) in which titrant is added from a burette and the temperature is measured with a thermometer. Some improvement in precision was obtained by using the semi-automatic procedure (B) in which the temperature is measured with a thermistor and recorded, but this improvement was not marked.TABLE I RESULTS FOR PRECISION FROM THE THERMOMETRIC TITRATION OF BENZOIC ACID AND 3,5-XYLENOL WITH 0.1 TO 0.001 M SOLUTIONS OF TETRA-n-BUTYLAMMONIUM HYDROXIDE WITH ACRYLONITRILE AS END-POINT INDICATOR Titrant Titra- molarity tion 0-1 A 0.025 A 0.026 A 0.01 A 0.00 1 A 0.1 B 0.01 B * See Procedure. and the remainder in Coefficient of variation, Method* per cent. & Mean F- End- Mono- Sample,’ titre/ Standard Single Mean point mer/ml mgt n# ml deviation points value (ii) 6 B,12*6 4 1.075 0.0068 0.537 0.268 (ii) 10 x , 4 4 1.433 0.013 0.878 0.439 (ii) 10 x , 1 0 6 3.236 0.051 1-64 0.690 2 B, 1.2 5 1.222 0.022 1.78 0.795 1 B,O*036 4 0.810 0.022 2.72 1.36 5 B,12-5 4 1-06 0.0044 0.416 0.208 2 B, 1.2 5 1.225 0.018 1.47 0.660 (i) (4 (4 (2) Experiments with 10 ml of monomer were carried out in the 60-ml beaker the closed apparatus (Fig.1). t B = benzoic acid; X = 3,5-xylenol. n = number of determinations. The effect of carbon dioxide and moisture in the atmosphere contributes to the total error in determinations when the 50-ml beaker is used. This effect is great with 0.001 M titrant and it was not possible to obtain reproducible results when an open container was used for the titration. Sharp, reproducible end-points were, however, obtained when a closed system that was purged with dry nitrogen was used (Fig. 1). Typical titration curves obtained with titrants of different molarities are shown in Fig. 2. Although satisfactory results could be obtained with 15 ml of acrylonitrile and a total volume of monomer plus sample of up to about 30 ml when 0.1 and 0.025 M titrants were used, with the 0.01 and 0.001 M titrants it was necessary to take smaller amounts of monomer and sample solution to achieve the required concentration of the catalyst for initiation of polymerisation at the end-point.Suitable volumes were 2 ml of acrylonitrile and not more than 1 ml of sample solution, and 1 ml of acrylonitrile and not more than 0.1 ml of sampleFebruary, 19731 INDICATION IN NON-AQUEOUS THERMOMETRIC TITRIMETRY. PART 11 93 solution, for the 0.01 and 0.001 M titrants, respectively. The blank titration value, which gives a measure of the amount of catalyst needed to initiate polymerisation under the con- ditions used, ranges from less than 0.05 ml with 0.1 M titrant to about 0.3 ml with 0901 M titrant. Tetra-n-butylammonium hydroxide reagent/ml ( 1 division = 1 mi) Fig.2. Thermometric titration of benzoic acid with 0.1 to 0.001 M tetra-n-butylammonium hydroxide reagent. a b C d Molarity . . . . 0.1 0.01 0.001 0.001 Acrylonitrile/ml . . 10 2 1 1 Apparatus . . .. A B B C Benzoic acid/mg . . 12.2 1.22 0.036 0.036 A = 60-ml beaker; B = 12-ml beaker; and C = 6-ml closed apparatus (Fig. 1) With the more dilute titrants one might expect the volume of “free” titrant required to initiate polymerisation to increase significantly with increase in titrant volume , because any volume increase effectively reduces the over-all concentration of the catalyst. If this was so the calibration graph would be exponential in form.However, calibration graphs for all the titrants studied were found to be linear, or almost linear, in the range 0 to 2 ml of titrant, even though a titrant volume of 2 ml represents a three-fold increase in total volume when 1 ml of monomer is used. Fig. 3 shows the calibration graph for benzoic acid and nominally 0.001 M (actually 0.000 65 M) tetra-n-butylammonium hydroxide solution. I 1 ! 0.001 M Tetra-n-butyiammonium hydroxide reagent/ml Fig. 3. Calibration graph for the titra- tion of benzoic acid with 0.001 M tetra- n-butylammonium hydroxide reagent94 GREENHOW AND SPENCER : IONIC POLYMERISATION FOR END-POINT [Analyst, vol. 98 It was noted earlier4 that water and methanol inhibited the polymerisation process but that, with 0.1 M titrant, about 1 per cent.of water and 5 per cent. of methanol could be tolerated in the solutions to be titrated. With 10-mg samples in JOml of solution this is 25 and 125 times the sample size, respectively. Tertiary amines and polar solvents, such as dimethylformamide and dimethyl sulphoxide, do not inhibit polymerisation but primary and secondary amines undergo an exothermic addition (cyanoethylation) with acrylonitrile, which reduces the end-point sharpness (Fig. 4). With 0.1 M titrant and 10 ml of acrylonitrile this effect does not become serious, from the aspect of end-point determination, until the amine to sample ratio approaches 100 : 1 with morpholine. In addition to acrylonitrile, other monomers that are capable of undergoing anionic polymerisation have been examined as end-point indicators.Satisfactory initiation of poly- merisation for the titration could not be achieved with styrene, isoprene or methyl meth- acrylate, but methyl acrylate and dimethyl itaconate polymerised readily when dissolved in dimethylformamide or dimethyl sulphoxide. Successful titrations were carried out with potassium hydroxide in propan-2-01 and tetra-n-butylammonium hydroxide as titrant - catalysts. The titration curves were similar to those obtained with acrylonitrile (Fig. 5). Dimethyl itaconate is an attractive monomer for routine analysis in that it is non-toxic and, unlike methyl acrylate, has a tolerable odour. .t i 1 1 I I I 0.1 M Tetra-n-butylammonium hydroxide reagent/ml (1 division = 1 ml) Fig. 4. Effect of morpholine on the titration of benzoic acid with 0.1 M tetra-n-butylammoniuni hydroxide reagent.Conditions: 12 mg of benzoic acid and 5 nil of acrylonitrile. Morpholine/mg: a, 75; b, 160; c, 200; d, 320; e , 430; and f, 530. Arrow indicates theoretical end-point Tetra-n-butylammonium hydroxide was found to be a satisfactory titrant - catalyst for most of the monofunctional acids determined. It has the advantage that with many acids the salts formed are soluble in the titration solution, and the question of occlusion of sample in precipitated salts does not arise. Potassium hydroxide in propan-2-01 and potassium t-butoxide in 2-methylpropan-2-01 are also satisfactory titrant - catalysts but the potassium salts formed in the titrations are generally insoluble and sample occlusion can occur.n-Butyl- lithium, a well established catalyst for anionic polymerisation, does not polymerise acrylo- nitrile, presumably because of complex formation, but it is an effective catalyst for the polymerisation of dimethyl. itaconate in dimethylforniamide and dimethyl sulphoxide. Titration curves obtained with 0.1 M potassium hydroxide in propan-2-01, 0.1 M potassium t-butoxide in 2-methylpropan-2-01 and 0.1 M n-butyllithium in toluene as titrants are shown in Fig. 5. In Table 11, a range of monobasic organic acids that have been titrated with 0.1 M tetra-n-butylammonium hydroxide solution is listed. Most of the acids examined combineFebruary, 19731 INDICATION IN NON-AQUEOUS THERMOMETRIC TITRIMETRY. PART 11 96: with the titrant in a 1 : 1 molar ratio, but 2,6-di-t-butyl-4-methylphenol, succinimide and phthalimide react sub-stoicheiometrically while 2-hydroxyquinoline is unreactive.The low values obtained with the first three of these four compounds would appear to result partly from steric hindrance as all of them react to a greater extent, but still not stoicheiometrically, with potassium hydroxide. An explanation for the non-reactivity of 2-hydroxyquinoline would lie in the strong hydrogen bonding of the hydroxyl group with the nitrogen atom. As was noted in Part l , 3 the titration process involves competition between monomer and organic acid for the titrant. Apparently acrylonitrile, succinimide, phthalimide and, possibly, 2,6-di-t-butyl-4-methylphenol have similar reactivities with respect to the titrant.T i t r a n t h l (1 division = 1 ml) Fig. 5. Catalytic thermometric titration with different monomer and titrant - catalyst combinations. a b c d e f g Titrant . . .. K KB B K B L L Monomer/ml .. A,10 A,15 M,2 M,2 T,2 T,l T , l Solvent/ml . . - - F,2 S,2 F,2 F,1 S,1 Compound/mg . . 2,12 X,4 Y,15 R,11 Y,15 2,12 H,9 Titrants (0.1 M)-K, KOH in propan-2-01; KB, potassium t-butoxide in 2-methylpropan-2-01; B, tetra-n-butylammonium hydroxide in toluene - methanol; and L, n-butyllithium in toluene. Monomers-A, acrylonitrile; M, methyl acrylate; and T, di- methyl itaconate. Solvents-F, dimethylformamide ; and S, dimethyl sulphoxide. Compounds-2, benzoic acid ; X, 3,S-xylenol; Y, salicylic acid ; R, resorcinol ; and H, 3,4-dihydroxybenzoic acid A number of polyfunctional acids have been titrated with 0.1 M tetra-n-butylammonium hydroxide and 0.1 M potassium hydroxide solutions in conjunction with acrylonitrile as end-point indicator.In Table 111, the titration values are compared with those obtained by the acetone method of Vaughan and Swithenbank.l It can be seen that with their method the second acidic group in hydroquinone and salicylic acid is determined while with the polymerisation technique it is not. Further, with the potassium hydroxide titrant the second acidic group in resorcinol and pyrogallol is determined while with the tetra-n-butylam- monium hydroxide titrant it is not. The first and second acidic groups in cyanuric acid and the first two and third acidic groups in trithiocyanuric acid can be distinguished in a similar way.With tannic acid, twenty, twelve and ten acidic groups are determined by using the acetone procedure, the polymerisation method with potassium hydroxide titrant and the polymerisation method with tetra-n-butylammonium hydroxide titrant, respectively. The above results with polyfunctional acids suggest that catalytic thermometric titration96 GREENHOW AND SPENCER : IONIC POLYMERISATION FOR END-POINT [Analyst, Vol. 98 TABLE I1 MONOFUNCTIONAL ORGANIC ACIDS TITRATED THERMOMETRICALLY WITH 0.1 M TETRA-n-BUTYLAMMONIUM HYDROXIDE SOLUTION WITH ACRYLONITRILE AS THE END-POINT INDICATOR Conditions: Add 0.1 mequiv of the acid in 1 ml of toluene or dimethylformamide to 10 ml of acrylonitrile in a 50-ml beaker and titrate by using Method A Phenylacetic acid, benzilic acid, hippuric acid and cinnamic acid Benzoic acid, 9-toluic acid, o-nitrobenzoic acid, 3,5-dinitrobenzoic acid, P-aminobenzoic acid, m-aminobenzoic acid, p-methoxybenzoic acid and acetylsalicylic acid 3,5-Xylenol, 2,6-xylenol, 2,6-di-t-butyl-4-methylphenol (0.6) , * o-nitrophenol, o-aminophenol, salicylaldehyde, methyl salicylate and 3-hydroxypyridine 1-Naphthol, l-amino-7-naphthol, 8-hydroxyquinoline and 2-hydroxyquinoline (0) Dirnedone Succinimide (0.17) * and phthalimide (0.35) * titration, the figure being the fraction of the acid function titrated a t the end-point.and 0.89, respectively, with 0.1 M potassium hydroxide in propanol-2-01 as the titrant. Figures in parentheses following the names of compounds indicate non-stoicheiometric * The values for 2,6-di-t-butyl-4-methylphenol, succinimide and phthalimide are 0.73, 0.69 can be made selective by using different titrant - catalysts and different monomers (including acetone).Although acids, or acidic groups within a compound, that are distinguishable by the three methods cannot be classified into groups in a simple manner according to their pK, values, i.e., their acid-dissociation constants in aqueous solution, it is apparent from the pK, values given in Table I11 that the methods are selective within the pK, range 10 to 12 for most of the compounds examined. The anomalous result obtained with succinic acid, which is titrated as a monobasic acid in the acetone method and as a dibasic acid by the polymerisation technique, may be caused by 1 : 1 condensation of the acid with acetone, which would eliminate one carboxyl group. With dichloroisocyanuric acid, a source of “active” chlorine, two equivalents of titrant are consumed in the displacement of the chlorine atoms.TABLE I11 THERMOMETRIC TITRATION OF POLYFUNCTIONAL ORGANIC ACIDS WITH ACRYLONITRILE AND ACETONE AS END-POINT INDICATORS Compound Succinic acid . . .. Resorcinol . . . , . . Pyrocatechol . . .. Hydroquinone . . .. Phloroglucinol . . .. Pyrogallol . . .. . . Salicylic acid . . .. p-Hydroxybenzoic acid . . 3,4-Dihydroxybenzoic acid 3,4,6-Trihydroxybenzoic acid Tannic acid .. .. 2,6-Pyridinedicarboxylic acid Trithiocyanuric acid . . Dichloroisocyanuric acid . . Cyanuric acid . . .. .. . . .. .. .. .. .. .. .. .. .. . . . . .. . . Acidic groups 2 2 2 2 3 3 2 2 3 4 2 3 3 - 1 (3) Groups titrated by method* - P1 P2 K 2 2 (1) 1 1 (1) 1 2 (2) 1 2 (2) 1 1 (22) 1 1 (22) 2 2 2 2 2 2 2 2 2 2 10 12 20 2 2 1.74 1 1 2 2 3 3 2 3 2 pK, valuest -7 PKI PK2 4.21 5-64 9.85 12.08 9.81 11.32 10.85 11.39 8-40 8.88 9.01 11.64 2.98 12-38 4.68 9.3 4-34 8.85 - - - - - - 6-6 10.6 4-90 8-O$ - c * P1: acrylonitrile end-point indicator with tetra-n-butylammonium hydroxide titrant (0.1 M).P2 : acrylonitrile end-point indicator with potassium hydroxide titrant (0.1 M) . K: acetone method with potassium hydroxide titrant (1.0 M) ; values in parentheses are from reference 1. Data are from reference 7, except for t Acid-dissociation constants in aqueous solution. thiocyanuric acid (Mr. J. Chudy, Chelsea College). $ pK, 10.8.February, 19731 INDICATION IN NON-AQUEOUS THERMOMETRIC TITRIMETRY. PART 11 97 It can be concluded that the polymerisation technique offers a higher degree of sensitivity than has been achieved previously in the non-aqueous titration of organic acids and, in conjunction with the acetone procedure, enables one to obtain selectivity in the study of weaker acids. The sharpness and magnitude of the temperature changes at the end-point when 0.001 M titrant is used suggests the possibility of carrying out determinations in the sub-microgram region with 0.O001 M and weaker titrants. A limitation to the use of weaker titrants is the requirement for a catalyst concentration sufficient to initiate anionic poly- merisation, which automatically limits the final total volume of sample, monomer and titrant at the end-point. REFERENCES 1 . 2. Vaughan, G. A., and Swithenbank, J. J., Analyst, 1966, 90, 694. Vajgand, V. J., Gail, F. F., Zarnic, Lj., Brusin, S., and Velimirovic, D., in Ruzas, I., Editor, “Proceedings of the IIIrd Analytical Chemical Conference,” Volume 2, AkadCmiai Kiad6, Budapest, 1970, p. 443. Greenhow, E. J., and Spencer, L. E., Analyst, 1973, 98, 81. Greenhow, E. J., Chem. & Ind., 1972, 422. - , Ibid., 1972, 466. Vaughan, G. A., and Swithenbank, J. J., Analyst, 1970, 95, 890. Kortum, G., Vogel, W., Andrussow, K. , Editors, “Dissociation Constants of Organic Acids in Aqueous Solution,” International Union of Pure and Applied Chemistry, Butterworths, London, 1961. NOTE-Reference 3 is to Part I of this series. 3. 4. 5 . 6 . 7 . Received June 28th, 1972 Accepted September 28th, 1972