S T A T U S OF S T A N D A R D I Z I N G pH M E A S U R E M E N T S D. K. K o l l e r o v , L. V. V r u b l e v s k a y a , V. M. M o k h o v , a n d L. P. P i s k u n o v a

M. Yu. G o r i n a ,

UDC

389.6 : 541.132.3

Questions of the unity, regularity, and accuracy of pH measurements have great significance for controlling technological processes and production in a number of branches of industry and agricultural economy. These processes can be satisfactorily classified as water purification, extraction of milk-acid products, supervision over thermal energy equipment systems, and determination of soil acidity, pH-Meters were found to be extremely popular analytical instruments. A special survey showed, for example, that, in the Ukrainian SSR, pH-meters comprise close to 83% of all the analyzers of the properties and content of liquids. Thus, standardizing pH measurements should be considered as one of the most important problems of physieochemieal measurements. The concept of standardizing pH measurements had its inception in our country with the development of an operating version of a checking circuit for pH-meters, which included an upper standard unit intended for reproducing original measures, the transmission of the meanings and magnitudes of pH units from the original measures to sample means and apparatus for checking operational measurement means. This checking circuit design was proposed as the basis of our metrological projects and standardization of pH measurements. As a result, the checking circuit was adopted with some refinements and simplifications which did not alter the basic design [1]. All the elements comprising the circuit for reproducing the original measures and the transmission of measurement units to the operating instruments remained unchanged. A description of these elements is detailed below. Reproducing Original Measures. The carriers of the pH unit-buffer solutions with a known pH value-form the basis of standardizing pH measurements [2]. In order to establish the pH values of these buffer solutions, the most accurate measurement methods are used. The electrometrical method of measuring pH in circuits without transfer, by using a hydrogen-chlorsilver element, was recognized as such a method. The method has a well-established theoretical basis and, therefore, was adopted for certifying standard buffer solutions (original measures) and also for developing new original measures pertaining to the interesting limits of pH values at various temperatures. This method and other questions associated with deriving pH standards were developed and described in Soviet and foreign works [3, 4]. On the basis of this method an evaluation was made of the meanings of the pH value of original measures with an error of • 0.01 pH unit [4], whereas at one time basic practical measurements, using pH-meters, were made with an error of 0.05-0.1 pH units. By using five buffer solutions [4], the pH scale from 1.68 to 9.18 pH at 25~ in the 0 to 95~ range was reproduced, and this was reflected in the All-Union State Standard 10170-62 for the pH scale and in the All-Union State Standard 10171-62 for buffer solutions which reproduce this pH scale. Later, the need for pH measurements in medicine and biology, and also the development of pH-meters with increased accuracy by domestic industry, posed the problem of producing pH measurement means with an error of thousands of parts of a pH unit. This appeared possible to achieve in the case where, as the basis of the method of calculating the activity factors of the chlor-ion, the Beits-Guggenheim condition was stipulated, by virtue of which the value of the Debye-Htickel equation is assumed equal to 1.5 kg t / z mole -I/z with an ion strengthI ~ 0.!. Although the acceptance of this condition permitted increasing pH measurement accuracy, nevertheless it made an accurate pH scale conditional. This fact did not lessen the practical significance of introducing a more accurate pH scale.

Transited ~omIzmeriteI'naya Tekhnika, No. 9, pp. 13-16, September, 1970. Original a r ~ c ~ submit~d May 14, ]970.

I

9 1971 Consultants Bureau~ a division o/ Plenum Publishing Corporation, 227 West 17th Streetl New York, N. u 10011. All rights reserved. This article cannot be reproduced for any purpose whatsoever [ without permission of the publisher. A copy of this article is available from the publisher for $15.00.

1306

Fig. 1

Fig. 2

TABLE 1

PH Values Temperature, ~

15 20 25 30 35 38 40

molar

phosphate ratio

1:1

1:3,5

6,901 6,883 6,866 6,855 6,845 6,841 6,840

7,447 7,429 7,415 7,402 7,393 7,388 7,387

In order to carry out a l l the projects enumerated above, and to maintain henceforth the required l e v e l of a c curacy of p r a c t i c a l pH measurements, a standard unit was produced, whose general appearance is shown in Fig. ! and Fig. 2. The arrangement includes a measurement unit, a unit for purifying hydrogen, and a thermostating unit for p r e l i m i n a r i l y impregnating the solution with hydrogen. A P-37 t~gh-resistance ' potentiometer, a standard 0.005 class e l e m e n t , and an M - 1 9 5 / 2 galvanometer are used in the arrangement's compensating circuit. The unit permits m a k ing e m f measurements with an error up to 0.02 mV, which corresponds approximately to 0.0003 pH units. The independent part of the unit is a specially developed TB-3 thermostat, which is a t h r e e - c e l l system with a then-nostating liquid c a p a c i t y close to 140 liters. The thermostat has a cooling chamber (from + 4 to 20~ with an FAK-07 cooling plant evaporator mounted in it, a heating chamber (from 20 to 95~ where two heaters are l o c a t e d - t h e basic one and a general one with a power of "~ 3 kW for forced heating, and an operating chamber with three screens which provide thermostating r e l i a b i l i t y . All the chambers have heat insulation [6]. Three e l e c t r o l y t i c elements are i m m e r s e d simultaneously in the thermostat's operating chamber. The glass e l e c t r o l y t i c e l e m e n t consists of two c o m m u n i c a t i n g vessels, connected in the lower part by a glass tube. A hydrogen electrode is immersed in one of the vessels, and two chlorsilver electrodes are immersed in the other. The e l e m e n t ' s structure permits i t to be placed in the thermostat at a depth up to 35 cm, which provides the necessary constant temperature of the solution being examined. The solution's temperature is measured directly in the e l e m e n t by equiscaled mercury thermometers having a scale value of 0.01 ~ in the 0-60~ range and 0.02 ~ in the 60-95~ range. The construction of the thermostat and of the e l e c t r o l y t i c elements ensures the solution's constant t e m p e r a tures in the elements up to 0.02 ~ in the 10-60~ range and 0.03-0.04 ~ in the 0-10~ and 60-95~ ranges. A t e l n perature change of 0.02 ~ corresponds to a change in the pH value of ~ 0.0002 pH units. In the precision arrangement described above, pH values of buffer solutions were obtained with an error up to 0.002 pH units in the 6-8 pH region, which is important for m e d i c a l - b i o l o g i c a l research. In this region, pH values

1307

are reproduced by using two buffer solutions which are solutions of a monosubstituted potassium phosphate mixture and a duosubstituted sodim-n phosphate solution taken in different molar ratios: 1 : 1 and 1 : 3.8. Table 1 shows the results of the measurements for the 14-40~ temperature region. The data prepared by the D. I. Mendeleev All-Union Scientific Research Metrological Institute agrees favorably with the results of NB [8] within the limits of 0.001-0.002 pH u n i t s - f o r a solution with a phosphate ratio of 1 : 1 in all the examined temperature ranges and also for a ratio of 1 : 3 . 5 in the 18-30~ temperature range. For temperatures of 35-40~ the deviation of our data from the NB data a mounted to 0.005-0.006 pH units; the reason for the latter is not as yet explained.

Fig. 3

The Unit for Transmitting pH Units of Standard Buffer Solutions to Sample Solutions. The derivation of standard buffer solutions whose pH values are established by measurements in circuits without transfer is quite a difficult process. Therefore, they cannot be a sample means (which by its very designation should be the output of mass production) of checking and adjusting pH-meters. This gives rise to the necessity of using a more e c o n o m i c a l method of certifying buffer substances and solutions. Such measurements, having achieved wide dissemination in practice, are m a d e in circuits with transfer. However, as a result, measurement accuracy should be reduced because of the presence in these circuits of an indeterminant diffusion potential. The problem of using circuits with transfer for c e r t i fying sample buffer solutions was the subject of a special investigation [7], while the possibility was established of such a use in investigating the hydrogen-chlorsilver e l e m e n t and standard buffer solutions as comparison elements. This possibility was realized by developing a sample p H - m e t e r with a hydrogen measurement electrode, shown in Fig. 3 and described in detail in [8]. The sample p H - m e t e r consists of a unit for preparing and purifying hydrogen, thermostating means, a galvanic element, and an e m f measurement means. The galvanic e l e m e n t with transfer can have a different structure depending on the purposes of the measurement. But the basic stipulation is the presence of a liquid compound of the free diffusion type in a wide glass tube (0 10 mm), which is better than the others for providing standardization of measurement conditions and the possibility of a theoretical analysis of diffusion processes in the future. The investigations conducted on this arrangement showed that the transmission of a pH unit from circuits without transfer to circuits with transfer is achieved with a reproducibility error of :k 0.01 pH unit [7-9]. Additional investigations were m a d e for the purpose of increasing pH measurement accuracy up to thousands of parts. As a result, it was discovered that pH measurement reproducibility within the limits of + 0.002 pH units can be obtained on the described equipment in circuits with transfer and having a hydrogen electrode by using chlorsilver and chlorthallium comparison electrodes, but it was not attained with a mercurous: chloride comparison electrode.

Fig. 4

1308

Fig. 5 The basic conditions of such measurements are: the maintenance of an established thermostatic system, the necessary saturation time of the solution by hydrogen, the use of good quality electrodes and the maintenance of a hydrogen purification system. The sample pH-metrical arrangement described has, in addition' a second purpose. RegularityofpH measurementisensurednot only by usingbuffer solutions certified as to pH value, but also by the constancy of the electrode potentials which comprise the galvanic measurement element, If the possible discrepancies in the electrode potential values can be compensated for by adjusting the pH-meter according to the buffer solutions, the possible instability of the electrode potentials are the source of the measurement errors and therefore should be kept under control. This.is basically important for accurate pH measurements. The hydrogen electrode is the most stable electrode accepted as a standard in electrochemistry and, therefore, is also used in the described arrangement for certifying sample electrodes which are used as a means of checking operating electrodes. The certification is made on the basis of the emf measurements of the galvanic elements with the use of sample and standard buffer solutions. The Unit for Checking pH-Meters. The units described above are related to means which ensure the reproduction of original measures, and to the means of transmitting pH values from these original measures (standards) to sample measurement means. Operating pH-meters consist of two independent elements: a detector of the pH value and a converter of e m f to an output signal. The converter is an electronic millivoltmeter with a high-resistance input (Rin not less than 10 lz ohms), whose sensitivity matches the temperature dependence of the electrode system characteristic, and a scale calibrated in pH units. The converter's electrical characteristics can be varied with time due to the natural drift of the parameters of the electrical circuit's individual elements. The converter's pH-meter can be checked by using an electrode system simulator-an e m f source with the controlled parameters of an electrical circuit. For this purpose a special measurement unit was built (Fig. 4). It makes it possible to check pH-meters for all the basic parameters. The accuracy of the assigned e m f values is provided by the class of the potentiometer. The principle of the arrangement's operation is contained in simulating the e m f of an electrode system by using a P-37 potentiometer, in simulating the resistances in the circuits of the glass and auxiliary electrodes by using corresponding resistance, and also in simulating the parasitic voltages in the detector's circuits and connecting lines. The possibility is provided for checking the variations of the pH-meter readings with the variation of the supply voltage from the nominal, for recording the output signal with a PS1-01 type self-recording potentiometer for checking the stability of the readings, and also for the operational accuracy of the controlling equipment of the pHmeters. The unit permits checking practically all types of laboratory and industrial pH-meters. The Unit for Checking pH-Meter Electrode Detectors. The accuracy of the pH-meter readings depends on the degree of matching the converter with the characteristic of the electrode s y s t e m - t h e dependence of emf on the pH value and temperature of the solution. Operation of the assembly should be considered as most efficient when the electrodes correspond (within normalized limits) to the nominal characteristics, and the converter is adjusted a c -

1309

cording to tabular emf values of this same electronic system. In this case matching the converter with the detector is reduced merely to adjusting the beginning of the scale (the so-called "adjustment according to the buffer"). With prolonged storage, operation under experimental conditions, and for a number of other reasons, the electrode characteristics are irreversibly altered. The converter's adjustment to the actual electrode system characteristic is in principle possible, but inexpedient, since ~it eliminates the operative replacement of the electrodes upon their sudden failure. Thus, checking the electrodes for the correspondence of their characteristics by certified data is one of the problems of standardizing pH measurements. Figure 5 Shows the unit for checking the electrodes. The operating principle of the unit involves measuring pH-meter glass electrode potentials with respeet to the sampte comparison electrode in buffer solutions at various temperatures, measuring the electrical resistance of the glass electrode's sensitive membrane, the insulation resistance of the shielded output, and the resistance of the auxiliary electrodes. The e m f measurement of the electrochemical system, "the checked electrode-sample comparison electrode," is made by the compensation method using a P-37 potentiometer and a high-resistance null-indicator designed on the basis of a UEV-1 amplifier. This same amplifier is used in the resistance measuring circuit. The limits of resistance mea.*urement are 1" 103-1.10 ~ a. The arrangement is intended for simultaneously checking ten electrodes located in a special chamber with autonomous thermostating. All the basic parameters which characterize glass and auxiliary electrodes in pH-meters can be determined on the unit. Thus, all the basic apparatus which provides standardization of pit measurements was built in our country. The improvement of smnple measurement electrodes, the dissemination of pit measurements with an accuracy up to thousands of parts in a wider pH range, and also research on the possibility of transmitting pH units with an even higher accuracy are problems of future metrological projects in this measurement area. LITERATURE CITED I, 2.

Checking Circuit for pH Measurement Means. Recommendations for Standardization SEV RC1932-69. V. V. Alexandrov, D. K. Kollerov, and I. L. Skorik, Tr. inst-Komitera, Standartgiz, No. 68 (128), Moscow

(1963). 3. 4.

5. 6. 7.

D. K. Kollerov, N. V. Kuznetsova, and I. L. Skorik, ibid., pp. 42-58. V. V. Aleksandrov, L. V.Vruvlevskaya,D. K. Kollerov, N. V. Kuznetsova, and I. L. Skorik, ibid., pp. 50-79. R. Belts, Determining pH, Theory and Practice, Khimiya, Moscow (1968). A. Kh. Fayans and V. P. Chekulaev, Izmeritel'. Tekh., No. 8 (1968). N. P. Barabanova and D. K. Kollerov, Transactions of Metrological Inst. SSSIL, Izd. Standartov, No. 96 (156)

(1968). 8.

9.

1310

D. K. Kollerov, ibid., pp. 25-36. E. M. Malkova, L. K. Khalturina and E. V. Shestopalova, ibid., pp. 36-45.