Polarographic Determination of Nitrite DAVID TUNG-WHEI CHOW AND REX J. ROBINSON Chemical and Oceanographic Laboratories, University of Washington, Seattle, Wash.
the saturated cslomel electrode (SCE), and all currents have been corrected for residual current of molybdate. Catalytic Reduction of Nitrite in Sulfate Medium. It hae long been known that nitrite and nitrate give well-defined reduction waves only in the presence of certain polyvalent oations (13, f4), and also that molybdate can be reduced polarographioally in acidic solution, Johnson and Robinson ( 4 ) observed that molybdate gave an enhanced diffusion current in the presence of nitrate and that this might be used a6 the basis for its determination, They used a solution of 0.1 M sulfuric acid, 0.2 M sodium sulfate, and 8.75 X M sodium molybdate. In the present investigation, polwagrams have been obtained for similar solutions containing varying concentrations of sodium nitrite ranging from 1 x 10-4 M to 5 X 10-3 M . Typical polarograins aze shown in Figure 1, in which curve a was obtained with molybdate alone and curve b with nitrite preqent.
This work was undertaken to investigate the polarographic reduction of molybdate in the presence of nitrite. It has been found that nitrite gives a current enhancement starting at zem applied potential which differs from that of nitrate recently reported. This current enhancement is a function of nitrite concentration and may be used as the basis for its determination. The molybdate-catalyzed reduetion of nitrite in both sulfate and chloride solutions and the reduction of nitrate in chloride medium have been studied. It has been found that the current enhancement of both nitrite and nitrate is greater in chloridesolutions. The method is suitable for the estimation of low concentrations of nitrite and also for the determination of both nitrite and nitrate ions in the presence of eaoh other, for which a procedure has been proposed.
~ ~ ~ ~ I YJonn8on I L I ana ILUUIIISULI ( y ,
K
I~C~ULYW
ua n r +
lybdate gave an enhanced polarographic diffusion current when nitrate was present in the solution and that this enhmcement in current was a function of the nitrate concentration. They were able t o determine nitrate by measuring the current at the applied potential of -0.75 volt us. SCE. Kolthoff, Harris, and Matsuyama (7) had used 8 potential of -1.20 volts vs. SCE for the determination of nitrate in uranyl acetate solutions. Likewise, Keilin and Otvos ( 6 ) had used the same potential for the determination of nitrite in the presence of uranyl ions. Further investigation of the polarographic reduction of molybdate has shown that the diffusion current is enhanced also by the presence of nitrite, and that this may be used as the basis for its determination. Figure 1. Polarograms of Sulfuric Acid Solutions Containing Varying Concentrations of Sodium Nitrite
APPARATUS
The pal~rogramspresented in this paper have been obtained with a Sargent polarograph, Model XXI. An H-shaped electrolytic cell (9)with a saturturated calomel reference electrode ax one arm connected b y . a potassium chloride bridge was used. The capillary ehmacteristios, m*/at"', were 1.53 rng.PI*se~.-''* The electrolytic cell was immersed in a thermostat adjusted to 25.0" 2z 0.1" C. by a Mero-to-Mero thermoregulator and a Mero-bMerc relay control box (Precision Scientific Co.).
(I. 0.1 M
sulfuric acid, 0.2 M sodium sulfate, end 8.15 X 10-1 M sodium molybdate b. 0.1 M nulfuric acid, 0.2 M sodium sulfate, 8.75 X 10.1 M s o d i m molybdate, and 8.00 X 10-1 M sodium nitrite E. 0.1 M sulfuric acid, 0.2 M sodium sulfate, 8.75 x 1 0 - E M s o d i u m "."luh.l.rp ."A 7 Y . ,"-I M "odiumnitrite ,.
""
REAGENTS
All chemicals used in this investigation were of reagent grade. The various solutions were prepared with distilled water. EXPWMENTAL
The oxygen dissolved in the various solutions wa6 removed by washing with nitrogen gas. However, when the nitrogen gas wm bubbled through the acidic nitrite solutions, there was a p preciable loss of oxides of nitrogen. Keilin and Otvos ( 8 ) estimated that this error ranged up to 3% in their work. In the present investigation, loss of oxides of nitrogen was prevented by separately washing with nitrogen gas the sodium nitrite solution and the medium containing the indifferent electrolyte and the molybdate. The desired amount of nitrite was then added t o the indifferent electrolyte medium, under a nitrogen atmosphere, and carefully mived with a minimum of surface. agitation. Polar+ graphic analysis was made immediately after the prepazation of this solution. Potentials reported in this work are referred t o
ACIa
Nitrite Molarity
3oluUon"
Reduction Current for Nitrite, Pa.
R Pa..
330
380
325 343
33s 325 306 287
1493
1494
ANALYTICAL CHEMISTRY
could he conveniently measured a t the applied potentials of -0.15 volt and -0.50 volt os. SCE. These applied potentials u-ere much less negative than reported by previous investigators for their work. The results obtained a t -0.50 volt us. SCE are given in Table I. It is seen that there is a definite relationship between the reduction current and nitrite Concentration though i t is not linear. Thus i t is necessary to establish and use a Calibration curve when determining nitrite.
nitrik andmolybdenum,(III) likewisegives an increase in current a t these potentials. When the nitrite concentration was greater than 2 X M, a minimum w m observed in the polarogram a t a potential of about -0.18 volt os. SCE with a preceding maximum a t zero xpplied potential. A typical polarogrm is shown in Figure 1, e. The ciause of this maximum and minimum is unknown. It is different from the usual msximum with the dropping mercury electrode, caused by stirring effeot of the growing drop or by adsorp tion of eleotroactive material on the electrode surface, since it is quite reproducible and is a function of the nitrite concentration, Kolthoff and Parry (8) also ohsewed minima, probably of a similar nzture, in the polarographic reduction of molybdate and h3rdrogen peroxide. They too offered no explanation as t o tlle C Suse. Catalytic Reduction of Nitrite and Nitrate in Chloride Mediurn. It mas found that in hydrochloric acid-sodium chloride mediu m th e current enhancement caused by nitrite in the presence ,of ... molyQdate was greater tnan LnaL in suime meamm. roiarograms were obtained for a series of different nitrite solutions in a medium containing 0.1 M hydrochloric acid, 0.2 M sodium chloride, and 8.75 X M sodium molybdate, as shown in Figure 2, m which curve a was obtained with molybdate only and ruwe b with nitrite mesent. The current enhancement also s t a t e d a t the applied potential where molybdenum(V1) was reduced to molybdenum(V), a~ in the case of the sulfate medium. The minimum phenomenon was also observed when the nitrite roncentration was greater than 4 X M. The currents could be measured a t the potentials of -0.15 and -0.40 volt us. SCE. The results obtained at -0.40 volt us. SCE are given in Table 11. Because of the greater current enhancement in chloride medium, it W ~ Bpossible t o determine smaller nitrite concentrations in chloride than in sulfate solutions. Other concentrations of sodium chloride, hydrochloric acid, and sodium molybdate were investigated regarding their effects on the current. While variations in the sodium chloride concentration were found t o have no effect, an increase in hydrogen ion concentration increased the current enhancement considerably. Although the sensitivity of thc method may he increased by increasing the hydrogen ion concentration, it was felt undesirnhle to operate a t a higher hydrogen ion concentration €or nitrite solutions, because of the resulting decrease in stability of the nitrous acid solutions and because ample sensitivity was obtained nt a hydrogen ion concentration of 0.1 M . It was also found that varying molybdate concentration did not affect the current enhancement when the nitrite concentration was smll, hut, for high nitrite concentrations, higher current enhancement values were obtained with higher molybdate concentrations. It was determined that a molybdate concentration of 8.75 X 10-6 M was the most suitable far this work since the
. . .
Figure 2. Polamgrams of Hydrochloric Acid Solutions Containing Varying Concentrations of Sodium Nitrite a.
b.
0.1 M hydrochloric add, 0.2 M sodium chloride, and 8.75 X 10-5 M sodivm molybdate 0.1 M hydmchloria add, 0.2 M sodium chloride, 8.75 X 10-6 M sodivm molybdate, and 1 10 - I M sodium nitrife
x
By comparing the reduction currents of nitrate as obtained b j Johnson and Robinson (4) with those of nitrite as given in Tahlt I, it was readily apparent that nitrite, for a given concentration, gave the greater current enhancement in the presence of malyb. date. This may have been due t o the fact that nitrite is mort easily reducible than is nitrate. Greater ease of reduction of nitrite also accounts far the catalytic determination of nitrite a t a smaller negative applied potential. From the results in Table I, it is obvious that, as in the ease of nitrate, the current-molarity ratio for nitrite wan not constant and the reduction current was apparently not diffusion controlled. Consequently, the'IlkoviE (S)equation could not he applied here to calculate the number of eleotrons involved in the reduction process and thus t o determine the reduotion products. It has been proposed by Johnson and Robinson (4)that the current enhancement with nitrate is due t o the chemical reactioa between nitrate and the molybdenum(II1) formed a t the dropping mercury electrode. Kitrate oxidizes malybdenum(II1) to molyhdenum(V) which is then reduced again a t the cathode, resulting in an increase in current. The molybdate-catalyzed reduction of nitrite can also be explained by this type of mechanism, mzith the difference that the current enhancement starts a t the applied potential where molyhdenum(V1) is re:luced to molyhdenum(V). This suggests that molybdenum(V), formed a t the cathode, is oxidized by nitrite to molyhdenum(VI) which is reduced again a t the dropping mercury electrode. The oxidation of molybdenum (V) t o molybdenum(V1) by nitrite but not by nitrate ala0 indicates that nitrite is more easily reducible than is nitrate. Since the increase in current is dependent upon chemical reaction, a higher current value is expected with a faster reaction rate, which is in agreement with the observed current enhancement with nitrite. When the applied potential is more negative than -0.3 volt us. SCE, malyhdenum(V1) is reduced at the electrode t o molybdenum(III), and the chemical reaction between
-.
.I
Table 11. Catalytic Determination of Nitrite in Hvdroohloric Acid Solution"
Nitrite Molarity
Reduction Current for Nitrite. Pa.
K = i/C. pa./Mole.nty
a 0.1 M hydroohlorio acid. 0.2 24 sodium Chloride 8.75 x 1"-1 M ao(1~urn molybdate. m ~ T U l , 8= 1.53 rnz.2'~ aee.-l/* Rebiduhl current at -0.4 volt
-
V O L U M E 25, NO. 10, O C T O B E R 1 9 5 3 Tahle 111. Catalytic
Molarity
x
of Nitrate'in Chloride Reduotion Current for Nitrate.
Nit+
1495
Pa.
K
VC. pa./MolantY 1000 =
Reduotion Current
for Nitrateb. "a.
10-5 5.00 0.050 0.083 0:biS 1.00 x 10-4 830 0.153 0.042 765 2.00 x 10-4 4.00 X 10-4 0.080 600 0.240 ... 0.303 6.00 x 10-4 Y)s 0,380 450 ... 8 . 0 0 x 10-4 0.165 1.00 x 10-3 0.408 408 321 ... 0.641 2.00 x IC-* 3.50 X 14.5 255 0.893 1.08: 0 598 5.00 x i o - a 217 0.1 M hydrochloric acid 0.2 M sodiiim chloride, 8.75 x 10-6 M sodiu. molybdate. m*WJV = 1.$3 mg.z/( seo.-L'* Residual ourrent at -0.7 volt " 8 . SCE = 0.583 pa. Dsts of Johnsou and Robinson (4) fpr nitrate in 0.1 M sulfurio acid. 0 . 2 M sodium sulfate. 8.75 X 10-1 M sodium molybdate. s t -0.76 volt. v e. 8CE m ~ l W , *= 1.30 mg.*f*seo:"~
:
m
maximum galvanometer sensitivity could be utilized far low cot centrations of nitrite. Since the current enhancement from nitrite was greater in chloride than in sulfate solutions, it wa8 thought desirable t o ot>tain data for nitrate in chloride medium. A typical polarogai n is shown in Figure 3. From the results, i t was found that ans applied potential more negative than -0.40 volt us. SCE could he used for measuring the current enhancement from nitrat#3. The results at -0.75 volt us. SCE are shown in Table 111. Th,e data of Johnson and Robinson ( 4 ) for nitrate in sulfate mediuin are also given in Table I11 for comparison. It is seen that in chloride medium, nitrate likr nitrite gave a greater current el1haneement than in sulfate solutions. This suggests that t hLe reaction rate between molybdate and nitrate or nitrite was fasts:r in chloride than in sulfate medium. In a chloride solution, a minimum nitrate concentration of 2 X 10-6 M may be detectei1, as compared to 1 x 10-4 M in sulfate medium. Catalytic Determination of Nitrate and Nitrite in a Mixtur e. By utilizing the fact that nitrite gives a current enhancement 1tt the applied potential where molyhdenum(V1) in reduced t,O molybdenum(V) whereas nitrate does not, i t was possible to di . 1
Figure 3. rO.l
Polarogram of Solution Containing Nitrat,e in Chloride Solution
M hydrochloric acid, 0.2 M sodium chloride, and sodium nitrate
1.50 X 10-3
M
termine nitrite and nitrate in a mixture. The concentration of nitrite can be estimated by determining the current a t -0.15 volt vs. SCE in either a sulfate or a chloride solution. However, chloride medium is recommended because of the higher sensitivity. Nitrate in the mixture could not be determined directlyeven at apotential great enoughtoform molybdenum(II1) because of the lack of additivity of the current enhancements caused by nitrate and nitrite. However, by oxidizing nitrite to nitrate with hydrogen peroxide, as described by KeilL, and Otvos (e), the total concentration of nitrate could be estimated. Then by subtracting the nitrite concentration previously determined, nitrate in the original mixture could be obtained. Molybdate in the Presence of Nitrite. The enhancement in current wa8 also applied t o the determination of molybdate in the presence of nitrite. In a 2 X 10-8 M solution in nitrite, the lower limit of detectable molyhdate concentration was found t o be about 1 X IO-' M and 5 X IO-' M in sulfate and chloride solutions, respectively. This is considerably less sensitive than the 1
I
11.4
0.1 il4 aulfuric acid, 0.2 M aodiurn sulfate. rn'lltII8 = 1.53 mg.*/l seo:l~~. Residual onrrent a t -1.25 voltsua. SCE = 0.366 (LB.
Polarography of Nitrous Acid. In the polarographic determination of nitrite in the presence of molybdate, a wave other thsn that of molybdate was observed t o start at m applied potential of about -0.76 volt us. SCE. This differed from the polarographic reduction of nitrite in the presence of uranyl in which no such wave had been observed by Keilin and Otvos (6). Heyrovskj. and Nejedly ( d ) had observed the same wave with an acidic solution of nitrite. This they attributed t o the reduction of nitric oxide because they obtained the same wave when an alkaline solution, through which hydrogen gas had been bubbled after passing 6rst through an acidic solution of nitrite, wa8 acidi6ed. They concluded that nitric oxide was reduced t o ammonia by comparing the wave height with that of B thallous solution of the same concentration, They did not, however, report the halfwave potential nor the potential at, which the current was measured. Further studies regarding this wave were made in this investigation. It was determined that this wave occurred even in the absence of molybdate. Consequently, the use of molybdate was eliminated in later work and palarograms were obtained with acidic solutions of nitrite. It was found that this wave waa independent of the nature of the mineral acid used and also of the supporting electrolytes. Studies were made of 0.1 and 0.2 M hydrochloric acid, 0.1 and 0.2 M sulfuric acid, mixtures of hydrochloric acid and sodium chloride, and of sulfuric acid and sodium sulfate solutions. Virtually the same results were obtained with only a slight difference in the half-wave potentials in sulfate and chloride media. The appliedpotential of - 1.25 volts YS. SCE was used to measure the reduction currents. The results are recorded in Table IV. It was observed that a well-defined wave was obtained only
:
ANALYTICAL CHEMISTRY
1496
when the concentration of nitrite was higher than 1 X lo-' M , and that the wave was rather drawn out, extending from about -0.76 to - 1.1 volts us. SCE. A typical polarogam is shown in Figure 4. Despite the fact that the waves were not well-defined vith low concentrations of nitrite, the current a t -1.25 volts us. SCE could be used as well for estimating the nitrite concentration as shown by the results in Table IV.
0.5
.-
3
'-Fo-o:/
-z?
I
-0.5
-1.0 -0.8
-0.9
-1.0
-1.1
P O T E N T I A L , v. vs. S.C.E.
Figure 5. Analysis of Nitrous Aoid Reduction W a v e 0.1 M sulfuric acid and 0.2 M d i u m sulfate
Figure 4. Polarogram of Solution Containing Nitrite in Sulfate Solution 0.1 M sulfuric acid. 0.2 M sodium auIfafe, and 4.00 nitrite
x
10-3 M sodium
The half-wave potential was determined to be -0.98 volt us. i against the applied
SCE in a sulfate medium by plotting log
potential as illustrated in Figure 5. A value of n = 0.4 was determined from the slope of this straight line, indicating that the reduction was irreversible. A value of -0.96 volt us. SCE was found a8 the half-wave potential in a chloride medium by a similar plot. It was concluded that this wave was due t o the reduction of nitrous acid rather than of nitric oxide, as proposed by Heyrovskji reasons. itroua acid deoomposes according to
parently, Heyrovskg and Nejedly had not appreciated the fact that nitric oxide is only slightly soluble in alkaline solutions. According to Moser (lO),even water may slowlyreact with nitric oxide forming nitrous acid. In view of the above considerations, i t was concluded that the wave actually was due to nitrous acid. By assuming the diffusion coefficient far nitrous acid t o be the same as that for nitrite ions, 1.92 X lo-' em.* see.-' (5),and hy usingthe IlkoviE equation (s), a value of 2.8 was obtained for n. This agrees with the value obtained for the reduction of nitrite in the presence of uranyl ion by Keilin and Otvos (6) who interpreted this as indicating that the nitrite was reduced t o nitrogen. Since this investigation was completed, Kaufman et al. ( 5 ) have reported a value of 3.4 forn or nitrite in the presence of uranyl ion. ACKNOWLEDGMENT
This investigation was partly supported by the Office of Navd Research under contract No. N8onr-520/III with the University of Washington. The authors are indebted to M a n i n G. Johnson for valuable criticisms. LITERATURE CITED
_-
" I "
". ".._
" l
...".""" _.",
(1) Bray. y.uJ
stated that Reaotion 2 was the 6rst step to occur. Thus, in the experiment of Heyrovskj. and Nejedly, both NO and NIOa gases would be carried out of the acidic nitrite solution by the hydrogen gas. The latter oxide of nitrogen, when passed through an alkaline solution, would react with the alkali to form nitrite again. Upon acidification, the wave of nitrous acid would be obtained. Nitric oxide is not very soluble in water, acidic or alkaline solutions. However, it reacts, though slowly, with alkali to form nitrite and nitrogen g a ~(fa). If nitric oxide remained only as dissolved gas in the alkaline solution, i t would rertain1.v be washed out upon prolonged passing of an inert gas throurh the solution. In a repetition of the Heyrovskj. and Nejedly experiment, nitrogen gas was passed through a similar solution for I.? hours. It was assumed then that no nitric oxide remained diraolved. However, the wave was still obtained after the aridifiration of the alkaline solution. Furthermore, by colorimetric determination ( I f ) , nitrite was found to exist in this alkaline solution. Ap-
W.C.,Chem. Reos., 10, 161 (1932).
(2) Heyrovskl, J.. and Nejedly,
Y.,Collection Csechoslou. Chem. Communs., 3, 126 (1931). (31 IlkoviO. D., Ibid.. 6, 498 (1934). (4) Johnson, M. G.,and Robinson. R. J., 0 ilnC"I
illY11.
( 5 ) Kaufman, F., Cook, H. J.. and Davis, Y. M.. J. Am. Chem. Soe., 74, 4997 (1952). (6) Keilin, Bertram, and Otvos, J. W., &id,, 68, 2665 (1946). (7) Kolthoff. I. M.. Harris, W. E., and Matsuy;m a . G.. Ibid., 66 1782 (1944). (8) Kolthoff, I. M., and Parry, E. P.,Ibid., 73, 5,315(1951). -""" ..,.A (9) Lingme, J. J., and Laitinen. H. A,, IND.E*.- P _lll_, En.,11, 504 (1939). (10) Moser, L.,2. anal. Chem.. 50,401 (1911). (11) Robinson. R. J., and Thompson, T. G., J . M a h e Resea& (Sears Fozmdation). 7,42 (1948). (12) Russell, W.J.. and Lepraik, W., J . Chem.Soc.. 31.37 (1877). (13) Tokuoka. M., Collectin C ~ h o s l o u .Chm. Cmmuns., 4, 4W )._
.-.^".
(1932).
(14) Tokuoka, M., and Ruaicks, J., Ibid., 6, 339 (1934). R a c ~ i v s ofor revierv March 22, 1952. Accepted June 27, 1953. Presented! in Part before the Northwest Regional Meeting. A u s a m ~C~EMIOAL. ~ SOO~ETT, Seattle, Wash.. 1951.
