DETERMINATION OF FREE AND COMBINED PLANT ACIDS Maija-Liisa Salo and Kaija Kotilainen Department of Animal Husbandry, University of Helsinki Received July 11, 1969 Organic acids are of great significance in plant and animal metabolism. The non- volatile organic acids or plant acids (Ranson 1965), as they are called in the book Plant Biochemistry, are of such abundance in many plant materials that they constitute an important taste factor. Examples among human foodstuffs are fruits and vegetables. The acids probably also affect the palatableness of animal forage; for instance, such palatable feeds as clover, roots and green tops are rich in acids. Moreover, plant acids are a noteworthy factor in the preparation of silage; the acids, which in the plant occur partly in a free state and partly as salts, form an efficient buffer system (Playne & McDonald 1966, Whittenbury et al. 1967). Malic and citric acids are present in the highest concentrations in most vegetable materials (Hulme & Richardson 1954, Playne & McDonald 1966, Freeman 1967). Other acids found in larger quantities are quinic acid (Hulme & Richardson 1954, Hirst & Ramstad 1957, Jones & Barnes 1967), glyceric acid (Lessard & McDonald 1966, Playne & McDonald 1966) and oxalic acid (Sanderson & Selvendran 1965, Freeman 1967). Several acids occur in small quantities (Wall et al. 1961, Houston et. al 1963, Sanderson & Selvendran 1965, Jones & Barnes 1967). The most common cations associated with the plant acids are K + , Na + , Ca + + and Mg + . Individual plant acids are determined by chromatographic techniques. Separation and determination may be effected by gradient elution ion-exchange (Wall et al. 1961, Davies et al. 1965, Jones & Barnes 1967) or silica gel chromatography (Wall et al. 1961, Sanderson & Selvendran 1965, Lessard & McDonald 1966). In addition to a knowledge of the sequence of the elution of the acids from the column, paper and thin- layer chromatography are taken as a basis for the identification (Wall et al. 1961, Davies & Windsor 1967, Jones & Barnes 1967). Gas-liquid chromatography has also been attempted in the determination of plant acids (Hautala 1966, Li & Woodroof 1968). Detailed investigations of this kind offer a good picture of the plant acid compo- https://www.c-info.fi/en/info/?token=sK70q4xHjQOZjRkB.DxGT4_5mZn3mPyygifnmqQ.2pJ6H5SjFvlD-L0KzyBZHZuYizEGOuPTXhQkU677LgwAUe4c7jybC1LNdyHwZZcoMWwZ_QGberGvu7CPYWjFDPnpO4g8V6uiMWiH1ZZc6AJR4YxTX-LQaT-Qb6x71829tzP3KZPLAgRtK_CYtnSUNFFgcQYQGjiGrqTixwdCoEpn87Yu-La4f4b-Qsg06mSr42rE 278 sition; nevertheless in routine work the procedures are complicated and can be applied in well-equipped laboratories only. In this investigation attempts were made to develop a useful method for the deter- mination of the total amount of plant acids occurring in a free state and as salts. PROPOSED SCHEME FOR THE DETERMINATION OF FREE AND COMBINED PLANT ACIDS Principle The acids and salts are extracted with water att room temperature. A portion of the extract is used for the determination of the acidity by titration to pH 9(= titratable acidity). From a second portion, the cations are retained by a strong cation exchanger using batch technique, whereupon the acidity is determined by titration to pH 9.0. Since chlorides, phosphates and nitrates are also dissolved in the water and are converted by the cation exchanger into corresponding acids, the second titration-value is corrected accordingly. The remainder presents the total of plant acids. The cations are eluted from the ion exhanger with hydrochloric acid and determined by flame photometry. On the basis of the acid and cation equivalents, the total amount of plant acids, occurring in a free state and as salts, is calculated per g dry substance. Dissolving of acids 2 g of the powdered sample is put into an Erlenmeyer flask, and 100 ml of cation-free water is added (distilled water treated in a cation-exchange column). The mixture is shaken mechanically for two hours, poured into a 200 ml centrifuge tube, and centri- fuged for 15—20 minutes at 3000 r.p.m. The liquid is sucked through a sintered-glass filter (porosity G2) into a suction flask. If the solution is kept overnight, a few drops of 0.01 % Na-merthiolate should be added. The obtained solution (filtrate A) is used for the following determinations: Titratable acidity 25 ml of the filtrate A is pipetted into a 250 ml beaker and titrated electrometrically with 0.1 N sodium hydroxide to pH 9.0. The result is calculated as mequiv./g dry- substance. Total acidity, cations, and correctionsfor inorganic anions 50 ml of the filtrate A is pipetted into an Erlenmeyer flask and 1 g of Amberlite IR- -120H+-resin added. The mixture is shaken mechanically for 30 minutes. The solution is then decanted into a G 1 sintered glass funnel and filtered into a 200 ml measuring flask, the major part of the ion-exchange granules remaining in the Erlenmeyer flask. The ion-exchange resin is then washed several times with small amounts of water and kept for the determination of the cations. The solution (filtrate B) is used in the following determinations: Acidity. 100 ml of the filtrate B is pipetted into a beaker and titrated to pH 9.0 279 as described above. The total amount of plant acids is calculated from the consumption of 0.1 N NaOH, after correction for the inorganic anions present (see below). Chloride correction. I —2 ml of a 10 per cent K2Cr04 solution is added to the previously mentioned neutralized solution, and the chloride content determined by titration with 0.05 N AgNOa . A blank value, determined after the corresponding treat- ments, is substracted from the titration value obtained. The resulting figure corresponds to the amount of chlorides which have been converted to hydrochloric acid during the treatment in the cation exchanger. The result is calculated as mequiv. HCI/g dry substance. Phosphate correction. The P-contentof filtrate B is determined for a sml sample as ammonium phosphomolybdate (Kuttner & Lichtenstein 1932), which is measured colorimetrically using a Beckman Model B spectrophotometer (wavelength 690 mji, blue-sensitive phototube, red filter) or an EEL portable colorimeter. The result is calculated as mequiv. H3P04/g dry substance. The correction used is two thirds of the value calculated, since at pH 9.0 only two of the protons in phosphoric acid are dis- sociated. Nitrate correction. The nitrate content of filtrate B is determined from a 20 ml sample according to the method of Follet and Ratcliff (1963). The reduction with cadmium is effected in an ion-exchange column, and the colorimetric determination carried out at 474 mjx using Beckman Model B spectrophotometer. The result is calculated as mequiv. HNOa/g dry substance. The total amount of plant acids is arrived at by substraction of the Cl”, P04 3~ and N03~-corrections from the total acidity milliequivalents. Extract for chromatographic analysis. 25 ml of filtrate B is evaporated to dryness in vacuo at a temperature of 40°C. The residue is dissolved in 1 ml of water containing a small amount of thymol. The resulting extract may be used for thin-layer or paper chromatographic analysis of the plant acid composition, and tested for the possible precence of galacturonic acid. Cations in the water extract The ion-exchange resin mass obtained during the course of the determination of the total acidity, now containing the retained cations, is transferred from the sintered glass funnel back to the Erlenmeyer flask by rinsing with 50 ml of 3 N HCI. The mixture is shaken mechanically for 30 minutes, and filtered through the G 1 glass-sinter into an eva- poration bowl, avoiding as far as possible the passing of resin granules on the filter. 25 ml of 3 N HCI is added to the resin, the shaking is repeated for another 30 minutes, and the mixture is filtered through the same filter into the evaporation bowl. The filtrate is eva- porated to dryness on a water bath, and the residue is dissolved and rinsed with four 5 ml aliquots of 0.2 N HCI into a small Erlenmeyer flask. The cations are determined from this solution, using a Beckman Model DU spectrophotometer equipped with a flame photometer. To the investigated solution is added such an amount of an appropriate buffer solution that the concentration of extraneous cations in the buffer exceeds that of the corresponding substances in the investigated solution. This is also done in conjunction with the construction of calibration curves by the aid of standard solutions. The buffer solutions contain potassium, sodium, calcium, magnesium and iron, respectively, as 280 chlorides. The analysed solutions are diluted with 0.2 N HCI to yield transmittance reading within the range 20—80. If the reading is below 20 in the determination of Ca, the accuracy can be improved by the addition of a fixed amount of a calcium chloride solution. RESULTS AND DISCUSSION OF PROPOSED METHOD The behaviour of different acids and salts In connection with the ion-exchange treatment and titration the behavior of different acids and salts was studied. It was noted that in titration to pH 9.0 the protons of all carboxyl groups in the most common plant acids dissociated almost completely. This is also the case with galacturonic acid. In the case of phosphoric acid, only two of the pro- tons became dissociated; glutamic acid and aspargic acid dissociated to a minor extent only. Thus, the acids behaved in titration as could be expected on the basis of their pK- values (Kolthoff & Furman 1949). It was noted that the accuracy of the titration could be much improved by using a pH-meter instead of colour indicators. The equipment used was a Beckman pH-meter and an automatic burette. Tests with pure salts disclosed that although the calcium salts of malic and citric acids dissolve in water only slowly, they may dissolve rather quickly in the water extract of the plant sample, on account of the acidity of the extract. Also CaHP04 dissolves to some extent in the extract, but Ca3(P04 ) 2 and calcium oxalate remain undissolved. The behaviour of the calcium salt of pectic acid was also going to be investigated; however, the only preparation available (Fluka, Phytic acid, Calcium salt) was too impure to yield any results. In the tests with pure salts it was noted that the solubility of slightly soluble salts was enhanced by the cation exchanger. Nevertheless, the acid contents ob- tained, when the plant samples were extracted with water and the cations then separately removed from the extract, were higher than in a treatment, where the cation exchanger was present during the extraction stage (Table 1). The former alternative was preferred for the proposed analytical procedure because of the higher values obtained and the shorter time required by the determination. Thus, according to the first alternative, titratable acidity, total acidity and cations can be determined from the same sample. Several cation-exchange resins were tested for the retaining of the cations. All the strong sulphonic acid type cation-exchange resins tested yielded practically the same re- sults as the resin chosen for the procedure. The weak carboxylic acid type resin, Amber- lite IRC-50, was unsatisfactory, The resin chosen, Amberlite IR-120 (p.a., regenerated with 6 N HCI) has an exchange capacity of 4.5 mequiv./g. The amount used, 1 g/g dry plant sample, therefore represents a considerable excess. The water extract was in the beginning separated by filtration, but centrifugation was later preferred because of the low filtration rate of vegetables and fruits. The results obtained by centrifugation are generally slightly higher (Table 1), since in the filtration- alternative the washing is inevitably insufficient. In the proposed procedure the washing is omitted altogether; instead, the amount of water is measured accurately and the result calculated on the basis of the volume measured. This working stage is in fact simple as well as quick, and the repeatability of the results is good. 281 Table 1. Examples of the total acid contents obtained after different extractions and filtrations. Values expressed as mequiv. per 1 g dry substance. Extraction with water with cation without cation exchange exchange filtration centrifugation Cocksfoot 1.284 1.488 1.491 Cabbage 1.133 1.245 1.273 Leek 0.681 0.773 0.781 Bean 1.021 1.053 1.063 Parsley 1.623 1.685 1.594 Potato 0.664 0.775 0.796 Testing of the procedure for the determination of cations A series oftests were made withpure solution-mixtures containing KH 2P04—Na2HP04 CaCl2 MgCl2, to check the procedure for the determination of the cations. The solutions were treated in the same manner as were the water extracts of the plant samples. Examples of the results are presented in Table 2. It appears that one rinsing, after the elution-shaking treatment, is sufficient in the batch technique. However, the fact that the plant samples contain a large excess of potassium, in comparison with the other cations present, and that the amount of sodium is very minute, introduces complications in the determination. To determine the sodium and calcium, a 2-g sample must be taken; this means that a high dilution must be effected in conjunction with the determination of potassium. The tests disclosed that excessively high values were obtained at very low concentrations of Na + , Ca + + and Mg ++ . The transmittance readings should range bet- ween 20 and 80; 30—70 was taken as the objective. It seems that in the determination of potassium relatively satisfactory results are obtained even in the upper scale-region, as shown in example II of Table 2. Na + is the most difficult cation, the results obtained being systematically too high. Fixed amounts of calcium and sodium were also added to very dilute solutions to move the meter reading to a more accurate range. The example presented in Table 2 shows that for calcium this is a feasible approach, whereas in the case of sodium the result cannot be improved by it. Two examples of the influence of phosphates upon the determination of the cations are furthermore given in Table 2. The measurements have been made direct on a phos- phate solution. The results are in conformity with general experience (Wenner 1958); the Ca-value dropping down to an entirely wrong level in the presence of phosphates. The same trend, although weaker, can be noted for magnesium. On the other hand, potassium and sodium are determinable without error also in solutions containing phos- phate. The good sodium-values obtained in these tests imply that the apparent surplus of sodium, noted previously, has been introduced as a contamination in connection with the cation-exchange treatment and eluate evaporation. At the end ofTable 2 some examples are given of parallel analyses ofplant materials. The results, which were obtained on separate samples, display a sufficient repeatability. The ion-exchange treatment was also compared with the ferric acetate treatment 282 Table 2. Check of the procedure for the determination of cations. For explanations, see text. Cations, mg/20 ml 0.02 N HCI’ 1ransmittance K+ Na+ Ca ++ Mg f ! readings Pure phosphate solution No. I. Calculated 8.00 0.20 1.00 0.30 0.25 1.00 0.32 0.29 0.93 0.33 [ 0.34 1.09 0.32 Found 1) 7.98 2) 8.02 3) 7.98 30—70 1) 0.35 1.39 0.30 1.00 0.37 0.94