ANALYTICACHIMICA ACTA 3x7

SPECTROPHOTOMETRlC DETERMINATION OF NITRATE

USING CHROMOTROYIC ACID

PHILIP W. \VI?ST AND 1’. I’. I<A~IACI-IhNDRAN

Codes Chewicnl Lnborn!orics. Lostisicrrra Stole U2livcrsify, Uato,~ i~orcgc, La. 70803 (U.S..4.)
(Rcceivcrl January tst, 1gG6)

Nitrate and the oxides of nitrogen are recognized as important pollutants in

water and air respectively. There are many sources of pollution, including the
manufacture and use of fertilizers, automobile exhausts and various industrial
operations. Nitrates in water often indicate the final stages of biological oxidation.
Infant methemoglobincmia has been traced to high nitrate content in well waters.
In Minnesota alone, from 1947 to 1950, there wcrc x3g cases with 14 deaths traced
to the use of contaminated farm water supplies 1. Oxides of nitrogen in air are espc-
cially significant in smog formation, Several spectrophotometric methods for the
determination of nitrate have been reported in the literature which involve the use
of common reactions such as reduction followed by diazotization”J, or Ncssler’s
reaction&i, nitration reactions”-10 or othcrsri. Determinations of nitrates using redox
reactions are not selective and are generally inferior to nitration methods’“. The
most widely used method at present is based on the nitration of phenoldisulfonic
acidla-15. This method is time-consuming because a relatively large volume of
sample must be cvaporatcd carefully and the reaction is not sensitive enough to be
applicable for low levels of nitrate, Even more critical, the method suffers from a
large number of interferences, the most serious being chloride, and is generally quite
unreliable. Other methods are reported utilizing z,4-xylenot4~5, z,G-xyleno17-0, or
phenarsazenic acid’0 as reagents. The e,6-xylenol method has been improved17 to
tolerate up to 700 mg/l of chloride using mercury(I1) sulfate as a masking agent. It
involves a solvent extraction separation procedure.
WEST AND SARMA~~ introduced a specific and sensitive spot test for nitrate
based on its reaction with a solution of chromotropic acid (x,8-dihydroxy-3,G-
naphthalene disulfonic acid) in concentrated sulfuric acid. WEST AND LYLE+~ have
shown that this reaction is free of interferences from most common ions. Of special
interest was the observation that chlorides do not form a color with the reagent
<aIone, but in the presence of nitrate they do increase the intensity of the yellow color
produced. Using a large excess of chloride in the form of hydrochloric acid (I ml of
concentrated hydrochloric acid added to xoo ml of a 0.01 o/0solution of the reagent
in concentrated sulfuric acid), they found that the reaction complied with Beer’s
law in the range of 0.x to I mg/l of nitrate and applied this to the determination of
nitrate in samples of water or air. Above I mg/l the curve showed a discontinuity
and a positive deviation from Beer’s law. BATTEN 20 has used this method for the
determination of nitrate in samples containing traces of formaldehyde but no chloride,
the interference due to the former being removed by oxidation to formic acid.

Anal. China. Ada, 35 (IgCG) 3x7-324

 318 I’. W. WEST, T. P. RAMACHANDRAN

A further investigation of the reaction between nitrate and chromotropic acid

and the influence of chloride, nitrite and oxidants has been carried out. A procedure
is now proposccl in which the discontinuity previously noted has been eliminated and
reliable results can be obtained directly on z&ml samples of water over the range
0.2-20 mg/l of nitrate. The procedure is free of significant interferences.

EXPERIMENTAL

Reagents and apparalats

All aqueous solutions were made using double-distilled water.
Stagtdavd n&ate solution. A stock solution was prepared by dissolving 1,371 g
of reagent-grade sodium nitrate and diluting to I 1 to give a x.0 mg nitrate/ml
solution. By suitable dilution, a series of standards was prepared for the concentration
range 0.2-20 mg/l.
Szclficrz’c acid. Analytical reagent-grade (Mallinckrodt) concentrated acid was
found to be free from nitrate and was used throughout the procedure.
P&fz’ed chtromolvopic acid. A saturated solution of disodium salt of x,8-
dihydroxy-g,G-naphthalene disulfonic acid (technical grade, Eastman Organic
Chemicals) was processed twice using dccolorizing charcoal. The purified reagent was
crystallized from the filtered solution by the addition of sulfuric acid. The product
was filtered, washed several times with ethanol and dried below 80”. A 0.x0/b solution
of the reagent in concentrated sulfuric acid was prepared and then stored in low
actinic bottles. The reagent solution was colorless (a check on freedom from nitrate
contamination of the sulfuric acid) and was found to bc adequately stable for two
weeks.
Sulfite-urea solution. 5 Q of urea and 4 g of reagent-grade anhydrous sodium
sulfite were dissolved and diluted to IOO ml with distilled water. The solution was
used for the elimination of interfering substances such as nitrite or oxidizing agents.
Antimony so&ion. The solution was prepared by heating 0.3 g of antimony
metal in So ml of concentrated sulfuric acid until all of the metal was dissolved. The
solution was cooled and added to 20 ml of ice water. Any salt which crystallized when
kept overnight was redissolved by heating. Two milliliters of this solution added to
2.5 ml of sample can mask up to 2000 mg/i of chloride.
A&‘xzrutz~s. A Beckman Model DK recording spectrophotometer was used in the
study of absorption spectra. Absorbance measurements were made in il Beckman
Model DU spectrophotomcter equipped with matched r-cm quartz cells.

Pveliminavy ex:perimen.ts
The absorption maximum of the yellow reaction product of nitrate with
chromotropic acid was found to be at 410 m,u, Preliminary experiments were made
by pipetting 2.5 ml of nitrate solution into ro-ml volumetric flasks and diluting to
3 ml by adding 0.5 ml of double-distilled water. Next, I ml of concentrated sulfuric
acid Was added, the solution was swirled and the flasks were allowed to stand for
4 min in a tray of cold water at IO-2o”, in which the water level was slightly above
that in the flask. One ml of chromotropic reagent was added to each flask which
was swirled and then allowed to stand for an additional 3 min. The volume of
solution in each flask was next made up to IO ml by the addition of concentrated

A+lal. C/bit?&
Actn, 35 (Ig66) 317-324
 4400’

0350, P 0fl.r 24 hr.

____________

Ju l otter 24 Iv*.
-_________
0.150 ,’

0.100
I 2 3 4 5 6 7 8
Ho UII

Fig. I. Vnriatiorl of absorbnncc with respect to time and the influcncc of chloritlc on color clcvcl-
opmcnt.
(1) 15 mg of nitrate/l and no chloride;
(II) 15 mg of nitrate/l with zoo mg of chloritlc/l;
(III) 15 mg of nitrate/l with 400 mg of chloritlc/l;
(IV) 15 mg of nitrate/l with IOOO mg of chloride/l;
(V) 20 mg of nitrate/l with 2000 rng of chloritlc/l;
(VI) 9.15 mg of nitrate/l with 4000 mg of chloritlc/l ant1 zo mg of antirnony(II1).

sulfuric acid, the flasks were stoppered and the contents mixed thoroughly. Sufficient
heat was developed to promote the reaction without the solutions boiling or spattering.
After the flasks were cooled to room temperature the volume was again adjusted to
IO ml. Blank experiments were run substituting double-distilled water instead of
nitrate solution.
Studies of the reaction revealed that the maximum color was devclopcd
within IO min and was found to be stable for 24 h as shown in Fig. I. The temperature
of the cooling bath was found not to be critical and may vary anywhere from IO to
2o”. Heat developed during mixing has no harmful effect as w<as shown by re-
producibility of the final absorbances. The cooling bath served simply to dissipate
sufficient heat to avoid boiling of the solutions with resultant formation of bubbles.
Maximum and constant absorbance was obtained when the total aqueous phase was
between 2.7 and 3.1 ml and the final volume was made up to IO ml using concentrated
sulfuric acid. The mole ratio method showed that two moles of nitrate react with one
mole of chromotropic acid.
Stfcdy off~otential interferences. Substances which ordinarily interfere seriously
in nitrate determinations such as nitrite, oxidizing agents, chloride and iron(II1)
were studied specially.
Chlorine and certain other oxidizing agents can react with chromotropic acid
producing yellow products. The absorption spectrum appears similar to the nitrate-
chromotropic acid system with the maximum shifted to 460 rnp. However, addition
of excess sodium sulfite before the addition of the reagent completely eliminated

Anal. Chim. Ada, 35 (x966) 3X7-324

 &LIhIINATXON OIr INTILRIWRILNCBS
.._-- ________~___________________ _... __. ____ .-..-..--
17zlarfcrcnce Nitrute Masltitrg ugenl A bsorl~~ncc Corrected
(wil) (owls) ubsorbawe
”.___....--- ..--.-- __-.._.-..
-.. ___.__- ____.__ ____.--____________
____.
--__--
Nil rite
0.0 0.0 2.5 1116 urea o.or5 13lnnk
30 0.0 I.2 1.2
30 0.0 2.5 111gurc’a 0.015 0.000
30 3.0 2.5 ~ng urc’a o.oGg 0.054
30 5.5 2.5 rng urc’a 0.113 0.098

Oxidnut (M.trO.r-)
0.0 0.0 I.2 mg sona- Cl.015 Mank
10 0.0 0.620 O.<JO5
40 0.0 1.2 ni& SO33- 0.015 0.000
40 3.0 1.2 mg S&2- 0.069 0*0.54
40 5.5 I .2 rng sag- 0.113 0.098

Irotr(lZI) (+ cldoridcn)
0.0 0.0 IO rng SW+ 0.015 13lank
-10 0.0 0.115 0. IO0
-10 0.0 IO mg SW+ 0.015 0.000
40 0.8 10 lllfi sw+ 0.030 o.org

u Acltlctl with the iron in cnch cnsc in the amollnt of 1000 m&/l.

the intcrfercncc due to oxidizing agents, as shown in Table I, and was without
influence on the nitrate-chromotropic acid reaction.
Nitrite produced an intense color similar to that produced by nitrate. The
absorption maximum in this case occurred at 440 mp. Addition of urea before the
addition of the reagent eliminated all nitrite as nitrogen gas, yet did not affect the
nitrate-chromotropic acid system. This provided an effective and simple means for
avoiding interferences clue to nitrite.
Chloriclc may react with nitrate in highly aciclic solutions giving the following
products
HN03 c 3 HCl -+ NOCl + Cla + z HnO.
This is a potential disaclvantage inherent in all methods employing high acid con-
centrations. In the present case the products react with chromotropic acid and might
account for the increase in sensitivity as observed by WEST AND LYLES, in which
chloride ancl chromotropic acid were introduced simultaneously into a nitrate solution.
The color cleveloped was similar to that produced with nitrate and reagent alone
but was not stable. In Fig. I, plots II, III, and IV show the chloride interference at
the 15 mg/l nitrate level when the study was carried out according to the preliminary
procedure. The rate of color development was suppressed hy chloride initially, since
nitrate and chloride reacted with each other in acidic solution before the reagent
was added. It can be seen from Fig. I, plot IV, that with IOOO mg/l of chloride, more
than son/0 of the nitrate reacted with chloride when such mixtures were kept for
4 min in a I :2 (v/v) sulfuric acid-water medium,
Recause of the unreliability of the reactions involving both nitrate and
chloride with chromotropic acid, a procedure which would eliminate chloride or the
effect of chloride was indicated. Precipitation of chloride would be effective but very

A,UZi.Chi7n. ACta, 35 (tg66) 3X7-324’

SPECTROPHOT03fETRIC DETERMINATION OF NITRATE 321

tedious. The elegance of masking techniques was tempting but challenging. However,
gratifying results were obtained using the sequestering action of antimony and it has
been established as an effective and convenient means for conditioning the test
reaction.
Among possible masking agents for chloride that were studied, mercury(II)
and antimony(W) each showed promise. However, mercury could not be used in
the present system because it was easily reduced by the reagent. Fortunately, anti-
mony was found to mask chloride so completely that true test colors developed and
remained stable even after 24 h (Fig. I, plots V and VI). Antimony did not interfere
with the nitrate-chromotropic acid system, yet it effectively masked up to 2000 mg
of chloride/l when used in the recommended procedure (Table II).
Iron(III) causes serious interference in many calorimetric procedures due to
the yellow color of its chloro-complex. In the prcscnt method this interference was
overcome by the addition of antimony which completely discharged the color by
demasking the chloroferrate(III) complex. Up to 40 mg of iron(III) per 1 showed no
in terfercncc (Table I).

Recommended procedwe
Pipet 2.5 ml of nitrate solution (standard or sample) having a nitrate content
in the range 0.2-20 mg/l into dry xo-ml volumetric flasks. To each flask add I drop of
sulfite-urea solution, place the flasks in a tray of cold water (10-20’) and add 2 ml
of the antimony sulfate solution. The flasks should be swirled during the addition
of each solution. After the mixtures have stood in the bath for about 4 mm, add I
ml of chromotropic acid reagent, swirl the flasks again and then allow to stand in
the cooling bath for an additional 3 min. Next add concentrated sulfuric acid to

TABLE 11
ARSORUANCIE OF NITHATE-CHROMOTHOPIC ACID SYSTEhl

(At 410 mp using x-cm cell)

-___~--~--._--~I~_--_.-___-__-_ -_...__-_ __
Nilrule Corrected CorrccLed Corrected Correcled
(wlr) horbance absorbance ulsorbmrce absorbance
series # x series # a series # 3 series # 4
-~--
0.0 0.015 0.015 0.015 o.org
0.2 0.0035 0.0035 0.004 0.0035
0.3 0.005 0.005 0.005 0.005
0.4 0.007 0.007 0.008 0.007
0.8 o.or5 0.014 0.0x5 O.OIG
I.6 0.028 0.028 0.029 0.028

2:: 0.07 I 0.073 0.072 0.07 I

0.107 0.107 0.107 0.107
8.0 0.142 0.141 0.144 0.142
12.0 0.21 I 0.2x2 0.21 I 0.21 I
X6.0 0.277 0.280 0.278 0.278
20.0 0.347 0.347 0.350 0.351
--
Scrics # 1. With nitrate only.
Scrics # 2. Nitrate in prcscncc of urea and sulfite.
Series # 3. Nitrate in prcscncc of urea and sulfite and IO mg of antimony(lI1).
Scrics # 4. Nitrate in prcsuncc of urea and sulfite and IO mg of antimony(II1) togcthcr with
2000 mg of chloride/l.

Amal. Chiwa. Acta, 35 (1966) 3I7-324

 322 I’. W. WEST, T. P. RAMACHANDRAN

adjust the volume to the ro-ml mark, stopper the flasks and mix the contents by
inverting them 4 times. Finally allow the solutions to stand for 45 min at room tem-
perature and again adjust the volume to the IO-ml mark with concentrated sulfuric
acid. Final mixing should be done very gently so as not to introduce gas bubbles.
Double-distilled water was substituted for nitrate solution for running blank exper-
iments.
The absorbance reading should be taken 15 min or more after the last adjust-
mcnt of volume. The cell should be rinsed with the solution and then filled carefully
so as to avoid trapping bubbles. For this operation it was found expedient to keep
the cell in a slanting position (30” to horizontal) with a ground side facing up and
pouring the solution very slowly down the side of the cell. The absorbance may be
measured in a Beckman DU or other suitable spectrophotometer at 410 rnp using
water in the reference cell.

RESULTS AND CONCLUSIONS

Table II gives the corrected absorbances of the nitrate-chromotropic acid

system at various concentrations of nitrate, both in the absence and in the presence
of chloride. Beer’s law was obeyed up to 20 mg of nitrate/l. Table III presents results
of a statistical evaluation of nitrate determinations at various levels containing
different amounts of chloride (0-2000 mg/l). The standard deviation and relative
standard deviation for a 95% confidence limit is given at various levels. It is also
possible to get better results at low nitrate levels using a 5-cm cell and Beckman
model B spectrophotometer for absorbance measurement. Color may bc developed
from 5 ml of standards or samples by adding a proportional amount of reagents and
adjusting the final volume to 25 ml.
The method can be directly applied to the analysis of water without recourse
to evaporation or precipitation steps. However, appreciable amounts of suspended
impurities should be removed by centrifugation or filtration. Table IV presents the
data from analyses of turbid samples. Tolerance to chloride can be extended from
2000 to 4000 mg’/l by doubling the strength of the antimony solution. Interferences
due to iron(III), nitrite, and oxiclizing agents have also been eliminated. In addition,
studies have been made of cliverse ions in concentrations of 200 mg/l. both in the

TABLE III
STATISTICAL ANALYSIS OF AIISORBANCE AT VARIOUS LEVELS OF NITRATR

Number of N itrda A verugo & Standard f Coefficient of

experiments take>8 absor*bnnce deviutiolr variutio7t for 95%
(wit) differewce (w/t) confidence limit (%)

20 0.3 0.0045) 0.030 20

10 0.6 0.0115 0.030 IO
IO I .2 0.0227 0.024 4
IO 3.1 0.05Go 0.0042 2.7

IO 6.5 0.115 o-093 3

IO 10.0 0.178 0.100 2
IO 1G.G 0.290 o.rGo 2
10 X9.5 0.340 0.180 2

Anal. Ckinr. Acla. 35 (1966) 317-$24

SPECTROPHOTOMETRIC DETERMINATION OF NITRATE 323

TABLE IV
ANALYSIS OF WATER SAMPLES

Natllre of nmourrr of Chloride Nitrate Number of

sample NOS- &ded present fOil?bd experiments
(mgll) (WC) (mgll)
Baton Hougc
tap water 0.00 50 Nil 5
Baton Xougc
tap water 2.50 50 2.5f0.02 10
Synthetic
sample l 2.00 4000 2.0f0.02 5
.+*oo 4000 4.0 ho.05 5
8.00 4000 8.0 rto.05
No. 37b 0.00 I I20 4.30 :
No. 37 3.00 1120 7.30 3
No. 38” 0.00 I 176 4.35 3
No. 38 3.00 117G 7.35
No. 3gb 0.00 1006 4.70 3
No. 39 3.00 I 000 7.70 3
No. 4ob 0.00 IOIG 8.10 3
No. 40 3.00 1016 II.10 3
No. 4x” 0.00 950 4.Go 3
No. 41 3.00 950 7.Go 3
No. 42” 0.00 1016 +G0 3
No. 42 3.00 1016 7.Go 3
- --.-______ --__ ---_--_ -__- ~---- .__._
p The nntimony(ll3) solution used was double strcnah because of the high chloride concentration.
1) Water samples rcccivcd from I&n-Tech Laboratoricu, Inc.. Baton Rouge. Wcrc nnalyecd
according to rccommcnded procedure after removal of suspcndcd matter by ccntrifugation.

absence and in the presence of 4 mg of nitrate/l. There were no significant interferences

disclosed when the recommended test procedure was applied to the following:

Li+, Na+, I<+, Cu2+

Be2+, Mg2+, Ca”*, Zn”+, Cd”+, H@+

BO2-, BqO,+, Ala+

COa”-, Ti4+, 2r4+, Sn”+. Sn4+

NHa+, HP042-, P4074-. VO2+. Asa+, Bi3+

St-, SOS”-, Cr3+, CrO43-, MOO&-, WO4+

F-, Cl-, C103-, Clod-, Mn2+, Mn04-, Br-, BrOa-, OCI-

Fez+, Co2+, Ni”+.

It should be noted, however, that Ba2+, Sr2+, Pb3+, I-, IOr, SeOa2- ad SeOG- are
incompatible with the skstem and form precipitates. They would not be expected
to be present in significant amounts in ordinary waters and so do not offer any
complications. If present in concentrations greater than 20 mg/l, Cr3+ interferes t0
the absorbance of its sulfato-complex.

This investigation was supported by U.S. Public Health Service Grant

Number AP 00117.

Anal. Chim. Acta, 35 (1966) 317-324

 324 I’. W. WEST, T. P. RAMACHANDRAN

SUMMARY

A spectrophotometric method is described for the determination of nitrate in

the 0.5 to 50,ug range using chromotropic acid as the reagent and masking agents
for the elimination of possible critical interferences due to chloride, chlorine, iron(II1)
and oxidants. The method can be applied directly for the determination of 0.2-20 mg
of nitrate/l in 2.5-ml samples of water. Over 4000 mg of chloride/l can be tolerated.
The method is simple, rapid and reliable. At the I mg/l level for nitrate the coefficient
of variation at the g5o/0 confidence limit is 4%, There are no interferences.

Les auteurs decrivent unc methode pour lc dosage spectrophotometrique des

nitrates (0.5 h 50& au moycn cl’acide chromotropique, comme rdactif, et d’agents
masquants pour empScher les interferences dues aux chlorures, au chlorc, au fer(II1)
et aux oxydants. Ce procede peut ftrc appliqub directement au dosage dcs nitrates
(0.2 h 20 ms/l) dans des Bchantillons dc 2.5 ml d’eau. On peut toldrer jusqu’a 4000
mg/l de chlorures. La methocle est simple, rapide et stirc. I1 n’y a pas d’interferences.

ZUSAMhlENPASSUNG

Es wird eine cinfachc und schnelle sl~cktra1pl~otometrische Methode zur

13estimmung von 0.5-50 ,ug Nitrat beschrieben, bci der Cln-omotropsaure als Reagcnz
und mnskierende Iicagcnzien zur Elimination mijglicher Stiirungen durch Chlorid,
Chlor, Eisen(II1) und Oxiclationsmittel vcrwendet werden. Bit Methode kann direkt
zur Bestimmung von 0.2-20 mg Nitrat/l in 2.5 ml einer Wasserprobe vcrwendet
werdcn. fher 4000 mg Cl-/l sincl zulassig. Bci einem Gehalt von I mg Nitrat/l
betrtigt der Variationskoeffizicnt 4% bei einem Vertrauensbereich von 95%. Stijrun-
gen tretcn nicht auf.

I I-l. M. I~OSCII. A. 13. ROSENPIEL~, Ii. HUSTON, I-1. 12. SHIPhlAN AND I<. %I. WOOD~ARD, J. Am.
CVufw Works Assoc., 42 (1950) 161.
2 J. W. C~row AND M. S. JOIINSTOND, .4,1ral. Cllirw A&r, 27 (rg62) ‘141.
3 17. I-lARI<OW, J. Ckc?~. .SoC., 55) (1891) 320.
4 A. M. G. MACIIONALD, ltrd. Ckuwz. Rlf~., 31 (1955) 566.
5 I-I. BARNIIS, /I~rnl.vst, 75 (1950) 388.
0 1;. I_. ZQSHER, I<. rt. IIII~RT AND H. I3xcKhf.m, .rlmzl. Chem., 30 (~958) 1972.
7 A. ivl. HARTELY AND 1L 1. As~r, J. Aw. CV/rrlcv Il'orks Assoc., 52 (~gbo) 255.
8 A. M. HARTELY AND R. 1. &XI, rl w-zl. Clrew., 35 (1963) 1207.
g 1-I. A. C. MONTGOMERY ANI) J. F. DYMOCK. klrrcrlysl. 87 (19Gz) 374.
IO J. C. SWAIN, Clrcm. I&. (Lodon). 16 (1957) 479.
I I R. BASTAIN, IC. WELXRLING AND Ii. P,\LILLA, iltial. Cltcna.. 29 (1957) 1795.
12 R. MULLER AND 0. WEDEMANN, I’owr Wasscv, 22 (1955) 247.
13 A. GIRANDVAL AND H. LAJOUS. J. Ckens. Sot., 48 (1885) 1903,
14 C. M. JOI~NSON AND A. Unwc~r, Awul. Cl~o~t., 22 (1950) 1526.
15 M. J, TAI~,\s. A?&. Chem.. 32 (1950) 1020.
16 R. lDra~sc~i, Mikvockinz. Ada, IO (1956) 1490.
17 I>. W. W. ANDRDWS, /I?ia1ysI, 8g ( I#.~) 730.
18 P. W. WIIST AND 1’. SARMA, Mikrockitrr. Acts, 4 (1957) 50G.
ICJ 1’. w. WEST AND G. L. LYLE, nncil. Chim. Acta. 23 (1gGo) 227.
20 J. J. I~ATTEN, Aural. Ckem.. 36 (rgG.+) 939.

A?ld. Chin,. /ida, 35 (196(i) 3X7-324