1322
ANALYTICAL CHEMISTRY
ever, in both case8 it !-as necessary to heat the reaction mixtures for varying periods of time or even evaporate and then redissolve in water.
tion oi a class of compounds, under the modified conditions used, differentiation within a class of compounds is possible. I n one case, color differentiation betn-een the Etereoisomere quinine and quinidine has been shon-n (Table 11).
Table 111. nliscellaneous Xitrogen Compounds
LITERATURE CITED
-4cidAcid-Toluene Plus Toluene Ethyl Alcohol Yellow Tellom Ethanolamine Octadecylamine Dark amber Purple Urea Yellow Orange-yellow Biuret Yellow Pale orange Methylurea Yellow Orange-yellow Phenylurea Orange Orange-yellow Hippuric acid Yellow Pale purple Diphenylamine Broffnish Yellow-green Benzylamine Yellowish Reddish purple, dissipated b y agitation Twenty-one primary aromatic amines, three containing naphthalene ring systems, were also tested and all gave positive color reactions.
( I ) B l u m e n t h a l , F., Biochem. 2. 19, 521 (1909). (2) B u r m i s t r o v , 9 . I., J . Gen. Chena. 19, 1511-14 (1949'1. (3) Burr, G. O., G o r t n e r , R. A . , J . -4m. C'ilem. SOC.46, 1224-46 (1924). (4) Chenz. Revs. 26, 325-7 (1540). ( 5 ) Fischer, E., Ber. 19, 2988 (1886). (6) Fleig, C., Bull. SOC. chim. France (4) 3, 1038-45 (iPOS). (7) Frieber, W., Centr. Bakteriol, Paraaitenk. 87, 254-77 (1922) (8) R a y m o n d - H a m e t , Bull. sci. phawnacol. 33, 447-56 '1928). (9) Ibid.,pp. 518-25. (10) R e m , C., Loew, Ii, Ber. 36, 4326 (1903 . (11) Urk, H . W.van, Phaiin. Weekblad 6 6 , 101-8 (1929 (12) Wasicky, R., Z . anal. C'hem. 54, 3 9 3 4 (1915). (13) TTerner, =1. E. .L,Sci. Proc. Roy. Dilbliit .Sot. 23: 214-21 (19441.
Khen the modification indicated iE employed, Ehrlich's reagent will react a t room temperature with every class of nitrogen compounds tried. Although this precludes its use in identifica-
RECEIVED for reriew Sovember 8, 195.5. . i c c e p t r c .%pril I S .
:$.ji3
Alkaline Solutions for pH Control ROGER G. BATES and VINCENT E. BOWER National Bureau o f Standards, Washington 25,
D. C.
A series of alkaline solutions useful for control of pH at any point between pH 7 and 13 can be prepared by adding standard solutions of hydrochloric acid or sodium hydroxide to portions of the following stock solutions: 0.1M tris(hydroxymethyl)aminomethane, 0.02832 borax, 0.05M sodium bicarbonate, 0.05M disodium hydrogen phosphate, and 0.2M potassium chloride. The compositions and buffer values of the solutions are given for intervals of 0.1 pH unit, and the dilution values and approximate temperature coefficient of pH are also indicated. The estimated accuracy is within 10.02 pH unit. The measured pH values agree well with those computed by the mass law.
C
gH R a n g e
Systein
7 . 0 - 9 , 0 Tris(hydrosymethy1)aminonietliane-h~drc ',L:loric acid 8.0- 9 . 1 9.2-10.8 5.6-1 1, 0 10.g-12.0
Borax-hydrochloric acid Borax-sodium hydroxide S o d i u m bicarbonate-sodium hydroxide D i s o d i u m hydrogen phosphate-sodium ?,>-droxide
12.0-13.0
Potassium chloride-sodium h7-droxide
The pH is based on the conventional activiry sm.e defined by the SBS standards'
ONTROL of p H in the range from 7 to 13 is most commonly
accomplished by means of buffer solutions containing phosphates, borates, barbiturates, ammonia, carbonates, or glycine ( 5 , 6, 7 , 9, 11, IS). Because of side reactions with proteins or carbohydrates, however, the phosphate and borate solutions intended for use in the p H range from 7 t o 9 are often not well suited to studies in physiological media, while phosphates and carbonates are incompatible n-ith calcium salts. Ammonia buffers are not highly stable, but triethanolamine buffers may be used from p H 7 to 8.5 ( 4 ) . The low solubility of barbituric acid in cold water and the anomalous behavior of the silver-silver chloride electrode in barbiturate buffers (12) are sometimes of concern. Likewise, glycine is considered unsuitable in certain applications, in vien of the uncertain effect of ampholytes upon the ionic strength. It does not appear possible at the present time to find a series of solutions for the alkaline range that will be entirely free from these objections. The present paper presents the results of a determination of the pH a t 25" C. of some solutions that will prove useful in many instances for p H control in the range from 7 . 0 to 13.0. These are divided into six series, as follows:
Figure 1. Cell vessel
V O L U M E 28, NO. 8, A U G U S T 1 9 5 6
1323
with Equation 1. As a consequence of the &ell-iecognized liquid junction error near the ends of the pH scale, however, the p H of the 0.1~1.1solution of sodium hydroxide, measured with respect t o the borax or phosphate standards, appeared to be about 0.05 unit too low-that is, the measured E was too small, and a p H value of 12.83 a t 25" C., instead of 12.88, was indicated for the hydroxide solution. A correction was applied for this defect in the experimental pH measurement, on the assumption that the voltage departure is a linear function of pH betneen the standardizing points of 9.18 and 12.88. Hence, in effect, the p H in this highly alkaline region mas computed b)
The pH value.! of the six series of solutions were deteimined a t 25' C. hy measurement of the electromotive force of the cell (5') satd. Pt; H2(g:is, 1 a t m j , soln. X(pHs) I ~ c l statidard(pHs), Hs(gas,l atin. j : Pt in which the unkno\\-n solution is designated by X and the pH standard bj- S . The cell vessel is shon-n in Figure 1. T o enhance the convenience and versatility of the asjembly, the cell is provided with a saturated calomel electrode that is brought into contact \\-ith the intermediate salt bridge. The value of pHx is related t o pIls and the electromotive force, E , of the cell (all a t 25' C.! hy
pHx = pHs
Table I.
+ 16.004 E
z nil. O.l.M HC'l diluted t o 100 nil.) =z
-0.026
unit deg.-l; I
x
8
7.00
4ti
i.20
45 7 44 7
7.10 i.30 i.40
7,50
!.(io i
70
7.80
7.90
8.00
8.10 8.20
I,
43.4 42.0
40.3 38.5 36.6
34.6 32.0
29.2 26.2
22.Y
8.30 8.40 8.50
19.9 17.2 14.7 12.4 10.3
8 80
8 5 7 0
8.60 8.70
Y
90
Y 00
5.7
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
=
0.001~
AgHvzb -0.02
012 -0.02
018
020
023 027 029
-0.0''
031
031 029 026
022 020
10 30 10 40 10 50
10 60 10
io
10 80 i n 90 11 00
7fi
i i
10.7
12 2
13.5 15 2 16.5 17.6 19.1 20.2 21.2
014
22.0
22
,
0 013 0 014 0 015 0 016
0 016 0 015
0 014 0 0 0 0 0 0
Boiax, p H 8 00 to 9.10 0.0?5.1f borax, z ml. 0.1M HC1, diluted t o 100 ml.) H/dt = -0.008 unit deg. I 0.025
1111.
-1;
T
B
A.pHi/za
0:OOS 0.010 0.011 0.012 0.015
+ O , 07
20.5
19.7 18 8
17.7 16.6 15.2 13.5 11.6
9 4 7 1 4
2 (1
0.018 0 020
0,029 0.024 0.oL'lj
..
10.05 +0.04
Borax, p H 9.20 t o 10.80 (50 ml. 0.025.M borax, a ml. 0 . l M XaOH, diluted to 100 ml.) d p H / d t z= -0.008 unit deg.-l; f = 0.001(23 x) 9.20 0.9 ... 9.30 3 . 6 0.027 AO,Ol= 9.40 6.2 0.026 +0.01b 8.8 0,023 9.50 9.60 11.1 0.022 10.01,a+ O . O l a 9.70 13.1 0,020 9.80 15 0 0.018 +0.01b 9.90 16.7 0.01l~ 10.00 1 8 . 3 0.014 10.10 1 9 . 5 0.011 -0.01~ 10 20 20.5 0,009 0,006 10.30 2 1 . 3 0,008
+
10.40 10.50
+0.02
10.60 10.70 10.80
22.: 22.c 23.3
23.80 ",25
0,007 0 OOli 0.00.5 0.004
...
-0.01
016
-0.01
+
22)
5.0 6.2
(2)
024
1111.
9 60 9 70 9 80 9 90 10 00 10 10 10 20
+ 17.111 E
There !vas a slight unavoidable disturbance of the two junctions when a free connection was made betn-een the two hydrogen
-0.01
Carbonate 0.05.M SaHCOa, z ml. 0 . l M S a O H , diluted t o 100 ml.) d[iWdL = -0.009 unit d e g . 9 ; I = 0.001 ( 2 5 (50
(50
PH 8.00 8.10 8.20 8.30 8.40 8.50 8.60 8.70 8.80 8.90 9.00 9.10
010
013 015 017
pHs
Compositions, Buffer Values @), and Dilution Values (ApHl/*)of Buffer Systems
(30 1111. 0.1 .lf trisl h ~ d r o x ~ r n e t l i y 1 ) a n i i n o n i e i l a n e .
pH
=
EXPERIM EYT 4 L PROCEDURE
(1)
T r i ~hydroxi iilethy1)ammometliane
dpH, dt
pHx
+0.03b
+0.04. fO.04.
+0.026
0 1
010
+0.030 0.006
009 008
+
2x1
10.90
013
013
Phosphate (50 nil. 0.05.U l-a2HPOj, x ml. 0.l.U S a O H , diluted t o 100 nil.) dpH/dt = -0.015 unit deg.-1; I = O . O O l ( 7 7 11.00 11.10 11.20
11.30 11.40 11.50 11.60 11.70
3.3 4 1
0 : 009
6.3 7.5 9.1 11.1
0,014
5.1
13.3 16.2
11.80 19.4 11.90 23.0 12.00 2 6 . 9 a Measured 6 Calculatei
+
0.011 0.012
-0.06,a
0.017
- 0 . 0 9 , ~-0.10b
0,026 0.030 0.034 0,037
-0.15 b -0 1 3 . -0.17b ~
0.022
..
\Vhen pHx exceeds pHs the sign of E is positive (positive electrode on the right), whereas E is negative when pH.Y is less th:in pHs. STAYDARD s
For measurements between pH 7 and p H 13, the KB3 pliosphate (pHs = 6.86 at 25' C.) and borax (pHs = 9.18 a t 25' C.) dtandards Tvere supplemented with a 0.ldJ solution of sodium hj-droside, to whicli a pH of 12.88 waS assigned ( 2 ) . The phosphate standard was 0.025JI with respect to both potassium dihydrogen phosphate and disodium hydrogen phosphate. It \\-a3 used for the tris(hydroxymethy1)aniinoniethane series, and the 0.013~ borax standard was chosen for the carbonate series and the two series of borax buffer solutions. For the highly alkaline jolutions of the phosphate-h?-droside and chloride-hydroxide seyies: both the borax etandard and the supplementary standard of sodium hydroxide were employed. Khen the two cell solutions xere the phosphate and borax stmdards, the observed value of E Kas completely consistent
Hydroxide-C hloride ( 2 5 ml. 0.2.M KCI, x ml. 0.2.1f XaOH, diluted t o 100 ml.) dpH,'dt = -0.033 unit de..-]; I = O.OOl(50 2a)
-0.07b
PH 12.00 12.10 32.20
12.30 12.40 12.50 12.60
12.70 12.80 12.90 13.00
B 6.0 8.0
10.2 12.8 16 2 20.4 25.6 32.2 41.2
53.0
66.0
APHLIIZO
0.028
-0.28
0.048
-0
28
-0
28
0.042 0.060 0.076
0.094 0.1.'
O.Ifi 0.21 0.25
0.30
-0.28
-0.27
electrode compartments (Figure 1). These junctions were located in the centers of the bulbs just below each compartment. Hence the potential difference betneen the two hydrogen electrodes was usually determined by the measurement of each electrode individually against a saturated calomel reference electrode located in the intermediate bridge solution. One hydrogen electrode was isolated by a closed stopcock dut ing the measurement of the potential of the other, so that no disturbance of the liquid junctions resulted. T h e electromotive force of the cell usually remained constant within 1 0 . 2 mv. (0.0035 pH) for one half hour or longer. The potential of the reference electrode cancels out a hen the difference between the tTvo measurements is taken. T o obtain stable hydrogen electrode potentials in the carbonate buffer solutions, it was found necessary t o pass the hydrogen gas through a presaturator containing a portion of the same carbonate solution; othernise, removal of carbon dioxide by the bubbling hydrogen caused a sloiv drift in the direction of increasing alkalinity. Tris(hpdroxymethy1)aminomethane of high purity was furnished by Commercial Solvents Corp , and the sodium bicarbonate
1324
ANALYTICAL CHEMISTRY
was reagent grade. Both were found to assay 100.O~oby titration with standard acid. The stock solution of disodium hydrogen phosphate was prepared from KBS Standard Sample 186IIb. Hydrated borax usually loses its water of hydration slowly during storage. Although this change in composition is almost without effect on the p H of the 0.01M standard (I?),it might have a larger influence upon the p H of buffer solutions formed from the borax and strong acid or alkali. Accordingly, the stock solution of borax was prepared from material that had been dehydrated by drying overnight a t 110’ C. and then raising the temperature gradually to 400’ C. The water from which the solutions were prepared was purged with air that had been freed of carbon dioxide. RESULTS
The p H values calculated by Equations 1 and 2 m-ere plotted on a large scale as a function of the volume of strong acid or alkali added to 50 ml. (or 25 ml.) of stock solution. Smooth curves were drawn through the points, and the volumes of reagent nere read at intervals of 0.1 p H unit. The results are summarized in Table I. The estimated accuracy of the p H vnlues is f0.02 unit. Gomori (8) has determined the pH of six solutions containing tris(hydroxymethy1)aminomethane and hydrochloric acid at 23“ and 37” C. His pH values, interpolated a t 25’ C., agree with those obtained here within =!=0.02unit. The Van Slyke buffer value, p , defined as db/dpH (14), where b represents a number of moles of strong alkali added to 1 liter of buffer solution, is given in the table. Also listed are a few values of the dilution value, ApH1i2 ( I ) , a quantity that expresses the change of pH resulting from dilution of a portion of the buffer solution with an equal volume of pure Yater. The value of ApHl/z is positive when the p H increases on dilution and negative when i t decreases. I n view of the good agreement between the measured and calculated p H values for the tris( hydroxymethy1)aminomethane buffers and for the potassium chloride-sodium hydroxide solutions (see below) the dilution values of the solutions in these two series were computed from the composition, equilibrium constants, and the Debye-Huckel equation. Dilution values for several compositions in the other buffer systems were determined by measuring the p H before and after dilution. Some calculated values are also included in the table for comparison with those obtained by direct measurements. The effect of temperature changes (in p H units per degree C. near 25” C.) is indicated by the approximate calculated value of dpH/dt ( 8 ) given for each buffer along with a formula for the ionic strength, I . Like that of other alkaline solutions, the pH of these buffer mixtures is rather sensitive to temperature changes.
A high buffer value, a low dilution value, and a small value of dpH/dt are desirable properties of buffer solutions for p H control. The p H values given in the table were compared with the p H calculated from the dissociation constant, the compositions of the solutions, and conventional activity coefficients defined by the Debye-Huckel equation in its “second-approximation” form, n-hich contains an ion-size parameter, a,. For the mixtures composed of univalent ions and uncharged species alone, the p H calculated with u, betn-een 4 and 6 A . differed by no more than 1 0 . 0 1 unit from the measured values. I n view of the complexity of the equilibria in solutions containing appreciable concentrations of boric acid (IO),no calculation for the borax-hydrochloric acid series was attempted. When the solutions contain bivalent or multivalent ions, however, the activity-coefficient term is considerably larger than when only univalent ions are present, and the calculated pH is more sensitive to the value chosen for a,. However, the calculated pH was again consistent w t h the measured pH, but only when somewhat larger values of a,-namely 10 for the carbonate series and 8 for the phosphate series-wre used. While such large values for bivalent and trivalent ions are certainly not surprising, they enhance the difficulty of interpreting measured p H values simply and accurately in terms of chemical equilibria. LITERATURE CITED
Bates, R. G., d s . . i L . CHEX. 26, 871 (1954). Bates, R. G., “Electrometric pH Determinations,” chap. 4 and 5, Wiley, Yew York, 1954. (3) Bates, R. G., Pinching, G. D., Smith, E. R., J . Research .\-aatl. (1) (2)
BUT.Standards 45, 418 (1950). (4)
Bates, R. G., Schwarzenbach, G., Helu. Chim. Acfa 37, 1437
(5)
Britton,’H. T. S., “Hydrogen Ions,” 3rd ed., vol. I, chap. 16, Van Yostrand, Ken. York, 1943. Clark, W. hZ.. “Determination of Hydrogen Ions,” 3rd ed., chap. 9, Williams and Wilkins Co., Baltimore, 1928. Clark, W. hf.. Lubs, H. A,, J . Biol. Chem. 25, 479 (1916). Gomori, G., Proc. SOC.Exptl. Bid. Med. 62, 3 3 (1946). Kiehl, S. J., Loucks, R. D., Trans. Electrochem. Soc. 6 7 , 8 1 (1935). Kolthoff,I. M., Bosch, W., Rec. trav. chim. 46, 180 (1927). Kolthoff, I. M., Vleeschhouwer, J. J., Biochem. 2 . 189, 1 9 1
(1954).
(6)
(7) (8) (9) (10) (11)
(1927).
(12) Manov, G. G., Schuette, K. E., Kirk, F. S., J . Research Natl. BUT.Standards 48, 84 (1952). (13) SZrensen, S. P. L.. Biochem. 2. 21, 131 (1909); 22, 352 (1909); Ergeb. Physiol. 12, 393 (1912). (14) Van Slyke, D. D., J . Bid. Chem. 52, 525 (1922). RECEIVED for review November 25, 1955.
.4ccepted April 17, 1933.
Determination of Methylene Chloride in Aqueous Solutions B. M. TEMPLEMAN and JACQUES JUNEAU M a i n Laboratory, Canadian Celanese, Ltd., Drummondville,
A quantitative method for determining small quantities of methylene chloride in aqueous solutions has been developed, in which the methylene chloride is saponified with sodium hydroxide in a peroxide Parr bomb ignition-type apparatus. This method is simple and rapid and has a relative mean deviation of 2.6 parts per thousand and a relative mean error of 0.010%.
A
METHOD for the quantitative determination of small quantities (0.5 to 1.5%) of methylene chloride in aqueous solutions was desired for routine control analysis. Two things were required-speed and a reasonable degree of accuracy.
P. Q., Canada
The standard method of combustion ( 3 ) and subsequent determination of the chlorides by volumetric or gravimetric procedures did not lend itself readily to this type of sample, and a considerable amount of time was required for each determination. -4review of the literature showed several detection methods (5-7), but these were qualitative in nature. An infrared and mass spectrometric method was likewise noted ( I ) , but this required apparatus not currently in use in this laboratory. Vogel ( 4 ) gives a saponification method with alcoholic potassium hydroxide solution as a characterization test for alkyl halides. As this seemed a simple procedure, that would be applicable to methylene chloride, it was tried quantitatively. Very low results nere obtained, with a poor degree of accuracy, possibly because
