4-108 INORGANIC NONMETALS (4000)

4500-NH3 NITROGEN (AMMONIA)*

4500-NH3 A. Introduction

1. Selection of Method known addition (E), a phenate method (F), and two automated
versions of the phenate method (G and H) are presented. Meth-
The two major factors that influence selection of the method to ods D, E, F, G, and H may be used either with or without sample
determine ammonia are concentration and presence of interfer- distillation. The data presented in Tables 4500-NH3:I and III
ences. In general, direct manual determination of low concen- should be helpful in selecting the appropriate method of analysis.
trations of ammonia is confined to drinking waters, clean surface Nesslerization has been dropped as a standard method, al-
or groundwater, and good-quality nitrified wastewater effluent. though it has been considered a classic water quality measure-
In other instances, and where interferences are present or greater ment for more than a century. The use of mercury in this test
precision is necessary, a preliminary distillation step (B) is warrants its deletion because of the disposal problems.
required. The distillation and titration procedure is used especially for
A titrimetric method (C), an ammonia-selective electrode NH3-N concentrations greater than 5 mg/L. Use boric acid as the
method (D), an ammonia-selective electrode method using absorbent following distillation if the distillate is to be titrated.
The ammonia-selective electrode method is applicable over
the range from 0.03 to 1400 mg NH3-N/L.
* Approved by Standard Methods Committee, 1997.
Joint Task Group: 20th Edition (4500-NH3.H)—Scott Stieg (chair), Bradford R. The manual phenate method is applicable to both fresh water
Fisher, Owen B. Mathre, Theresa M. Wright. and seawater and is linear to 0.6 mg NH3-N/L. Distill into
NITROGEN (AMMONIA) (4500-NH3)/Preliminary Distillation Step 4-109

sulfuric acid (H2SO4) absorbent for the phenate method when fied at 4°C. For preservation for up to 28 d, freeze at ⫺ 20°C
interferences are present. unacidified, or preserve samples by acidifying to pH ⬍2 and
The automated phenate method is applicable over the range of storing at 4°C. If acid preservation is used, neutralize samples
0.02 to 2.0 mg NH3-N/L. with NaOH or KOH immediately before making the determina-
2. Interferences tion. CAUTION: Although acidification is suitable for certain types
of samples, it produces interferences when exchangeable ammo-
Glycine, urea, glutamic acid, cyanates, and acetamide hydro- nium is present in unfiltered solids.
lyze very slowly in solution on standing but, of these, only urea
and cyanates will hydrolyze on distillation at pH of 9.5. Hydro-
lysis amounts to about 7% at this pH for urea and about 5% for 4. Bibliography
cyanates. Volatile alkaline compounds such as hydrazine and
amines will influence titrimetric results. Residual chlorine reacts THAYER, G.W. 1970. Comparison of two storage methods for the anal-
with ammonia; remove by sample pretreatment. If a sample is ysis of nitrogen and phosphorus fractions in estuarine water. Ches-
likely to contain residual chlorine, immediately upon collection, apeake Sci. 11:155.
treat with dechlorinating agent as in 4500-NH3.B.3d. SALLEY, B.A., J.G. BRADSHAW & B.J. NEILSON. 1986. Results of Compara-
tive Studies of Preservation Techniques for Nutrient Analysis on Water
3. Storage of Samples Samples. Virginia Institute of Marine Science, Gloucester Point.

Most reliable results are obtained on fresh samples. If samples

are to be analyzed within 24 h of collection, refrigerate unacidi-

4500-NH3 B. Preliminary Distillation Step

1. General Discussion Some anion-exchange resins tend to release ammonia. If this

occurs, prepare ammonia-free water with a strongly acidic cat-
The sample is buffered at pH 9.5 with a borate buffer to ion-exchange resin. Regenerate the column according to the
decrease hydrolysis of cyanates and organic nitrogen com- manufacturer’s instructions. Check ammonia-free water for the
pounds. It is distilled into a solution of boric acid when titration possibility of a high blank value.
is to be used or into H2SO4 when the phenate method is used. 2) Distillation—Eliminate traces of ammonia in distilled water
The ammonia in the distillate can be determined either colori- by adding 0.1 mL conc H2SO4 to 1 L distilled water and
metrically by the phenate method or titrimetrically with standard redistilling. Alternatively, treat distilled water with sufficient
H2SO4 and a mixed indicator or a pH meter. The choice between bromine or chlorine water to produce a free halogen residual of
the colorimetric and the acidimetric methods depends on the 2 to 5 mg/L and redistill after standing at least 1 h. Discard the
concentration of ammonia. Ammonia in the distillate also can be first 100 mL distillate. Check redistilled water for the possibility
determined by the ammonia-selective electrode method, using of a high blank.
0.04N H2SO4 to trap the ammonia. It is very difficult to store ammonia-free water in the labora-
tory without contamination from gaseous ammonia. However, if
2. Apparatus storage is necessary, store in a tightly stoppered glass container
to which is added about 10 g ion-exchange resin (preferably a
a. Distillation apparatus: Arrange a borosilicate glass flask of strongly acidic cation-exchange resin)/L ammonia-free water.
800- to 2000-mL capacity attached to a vertical condenser so that For use, let resin settle and decant ammonia-free water. If a high
the outlet tip may be submerged below the surface of the receiving blank value is produced, replace the resin or prepare fresh
acid solution. Use an all-borosilicate-glass apparatus or one with ammonia-free water.
condensing units constructed of block tin or aluminum tubes. Use ammonia-free distilled water for preparing all reagents,
b. pH meter. rinsing, and sample dilution.
b. Borate buffer solution: Add 88 mL 0.1N NaOH solution to
3. Reagents 500 mL approximately 0.025M sodium tetraborate (Na2B4O7)
solution (9.5 g Na2B4O7 䡠 10 H2O/L) and dilute to 1 L.
a. Ammonia-free water: Prepare by ion-exchange or distilla- c. Sodium hydroxide, 6N.
tion methods: d. Dechlorinating reagent: Dissolve 3.5 g sodium thiosulfate
1) Ion exchange—Prepare ammonia-free water by passing (Na2S2O3 䡠 5H2O) in water and dilute to 1 L. Prepare fresh
distilled water through an ion-exchange column containing a weekly. Use 1 mL reagent to remove 1 mg/L residual chlorine in
strongly acidic cation-exchange resin mixed with a strongly 500-mL sample.
basic anion-exchange resin. Select resins that will remove or- e. Neutralization agent.
ganic compounds that interfere with the ammonia determination. 1) Sodium hydroxide, NaOH, 1N.
4-110 INORGANIC NONMETALS (4000)

2) Sulfuric acid, H2SO4, 1N. immediately transfer sample flask to distillation apparatus. Dis-
f. Absorbent solution, plain boric acid: Dissolve 20 g H3BO3 till at a rate of 6 to 10 mL/min with the tip of the delivery tube
in water and dilute to 1 L. below the surface of acid receiving solution. Collect distillate in
g. Indicating boric acid solution: See Section 4500-NH3.C.3a a 500-mL erlenmeyer flask containing 50 mL indicating boric
and b. acid solution for titrimetric method. Distill ammonia into 50 mL
h. Sulfuric acid, 0.04N: Dilute 1.0 mL conc H2SO4 to 1 L. 0.04N H2SO4 for the ammonia-selective electrode method and
for the phenate method. Collect at least 200 mL distillate. Lower
4. Procedure distillation receiver so that the end of the delivery tube is free of
contact with the liquid and continue distillation during the last
a. Preparation of equipment: Add 500 mL water and 20 mL minute or two to cleanse condenser and delivery tube. Dilute to
borate buffer, adjust pH to 9.5 with 6N NaOH solution, and add 500 mL with water.
to a distillation flask. Add a few glass beads or boiling chips and When the phenate method is used for determining NH3-N,
use this mixture to steam out the distillation apparatus until neutralize distillate with 1N NaOH solution.
distillate shows no traces of ammonia. d. Ammonia determination: Determine ammonia by the titri-
b. Sample preparation: Use 500 mL dechlorinated sample or metric method (C), the ammonia-selective electrode methods (D
a known portion diluted to 500 mL with water. When NH3-N and E), or the phenate methods (F and G).
concentration is less than 100 ␮g/L, use a sample volume of
1000 mL. Remove residual chlorine by adding, at the time of 5. Bibliography
collection, dechlorinating agent equivalent to the chlorine resid-
ual. If necessary, neutralize to approximately pH 7 with dilute NICHOLS, M.S. & M.E. FOOTE. 1931. Distillation of free ammonia from
buffered solutions. Ind. Eng. Chem., Anal. Ed. 3:311.
acid or base, using a pH meter.
GRIFFIN, A.E. & N.S. CHAMBERLIN. 1941. Relation of ammonia nitrogen
Add 25 mL borate buffer solution and adjust to pH 9.5 with 6N to breakpoint chlorination. Amer. J. Pub. Health 31:803.
NaOH using a pH meter. PALIN, A.T. 1950. Symposium on the sterilization of water. Chemical
c. Distillation: To minimize contamination, leave distillation aspects of chlorination. J. Inst. Water Eng. 4:565.
apparatus assembled after steaming out and until just before TARAS, M.J. 1953. Effect of free residual chlorination of nitrogen com-
starting sample distillation. Disconnect steaming-out flask and pounds in water. J. Amer. Water Works Assoc. 45:47.

4500-NH3 C. Titrimetric Method

1. General Discussion b. Indicating boric acid solution: Dissolve 20 g H3BO3 in

water, add 10 mL mixed indicator solution, and dilute to 1 L.
The titrimetric method is used only on samples that have been Prepare monthly.
carried through preliminary distillation (see Section 4500- c. Standard sulfuric acid titrant, 0.02N: Prepare and standard-
NH3.B). The following table is useful in selecting sample vol- ize as directed in Alkalinity, Section 2320B.3c. For greatest
ume for the distillation and titration method. accuracy, standardize titrant against an amount of Na2CO3 that
Ammonia Nitrogen Sample has been incorporated in the indicating boric acid solution to
in Sample Volume reproduce the actual conditions of sample titration; 1.00 mL ⫽
mg/L mL 14 ⫻ normality ⫻ 1000 ␮g N. (For 0.02N, 1.00 mL ⫽ 280 ␮g
N.)
5–10 250
10–20 100
20–50 50.0
50–100 25.0 4. Procedure

a. Proceed as described in Section 4500-NH3.B using indicat-

2. Apparatus ing boric acid solution as absorbent for the distillate.
b. Sludge or sediment samples: Rapidly weigh to within ⫾1%
Distillation apparatus: See Section 4500-NH3.B.2a and b. an amount of wet sample, equivalent to approximately 1 g dry
weight, in a weighing bottle or crucible. Wash sample into a
3. Reagents 500-mL kjeldahl flask with water and dilute to 250 mL. Proceed
as in ¶ 4a but add a piece of paraffin wax to distillation flask and
Use ammonia-free water in making all reagents and dilutions. collect only 100 mL distillate.
a. Mixed indicator solution: Dissolve 200 mg methyl red c. Titrate ammonia in distillate with standard 0.02N H2SO4
indicator in 100 mL 95% ethyl or isopropyl alcohol. Dissolve titrant until indicator turns a pale lavender.
100 mg methylene blue in 50 mL 95% ethyl or isopropyl alcohol. d. Blank: Carry a blank through all steps of the procedure and
Combine solutions. Prepare monthly. apply the necessary correction to the results.
NITROGEN (AMMONIA) (4500-NH3)/Ammonia-Selective Electrode Method 4-111

5. Calculation Sample 1 contained 200 ␮g NH3-N/L, 10 mg Cl⫺/L, 1.0 mg

NO3⫺-N/L, 1.5 mg organic N/L, 10.0 mg PO43⫺/L, and 5.0
a. Liquid samples: mg silica/L. The relative standard deviation and relative error
for the 21 participating laboratories were 69.8% and 20%,
(A ⫺ B) ⫻ 280 respectively.
mg NH3-N/L ⫽ Sample 2 contained 800 ␮g NH3-N/L, 200 mg Cl⫺/L, 1.0 mg
mL sample
NO3⫺-N/L, 0.8 mg organic N/L, 5.0 mg PO43⫺/L, and 15.0 mg
b. Sludge or sediment samples: silica/L. The relative standard deviation and relative error for the
20 participating laboratories were 28.6% and 5%, respectively.
(A ⫺ B) ⫻ 280 Sample 3 contained 1500 ␮g NH3-N/L, 400 mg Cl⫺/L, 1.0 mg
mg NH3-N/kg ⫽
g dry wt sample NO3⫺-N/L, 0.2 mg organic N/L, 0.5 mg PO43⫺/L, and 30.0 mg
silica/L. The relative standard deviation and relative error for the
where: 21 participating laboratories were 21.6%, and 2.6%, respec-
A ⫽ volume of H2SO4 titrated for sample, mL, and tively.
B ⫽ volume of H2SO4 titrated for blank, mL.

7. Bibliography
6. Precision and Bias
MEEKER, E.W. & E.C. WAGNER. 1933. Titration of ammonia in the
Three synthetic samples containing ammonia and other con- presence of boric acid. Ind. Eng. Chem., Anal. Ed. 5:396.
stituents dissolved in distilled water were distilled and analyzed WAGNER, E.C. 1940. Titration of ammonia in the presence of boric acid.
by titration. Ind. Eng. Chem., Anal. Ed. 12:711.

4500-NH3 D. Ammonia-Selective Electrode Method

1. General Discussion 2. Apparatus

a. Principle: The ammonia-selective electrode uses a hydro- a. Electrometer: A pH meter with expanded millivolt scale
phobic gas-permeable membrane to separate the sample solution capable of 0.1 mV resolution between ⫺700 mV and ⫹700 mV
from an electrode internal solution of ammonium chloride. Dis- or a specific ion meter.
solved ammonia (NH3(aq) and NH4⫹) is converted to NH3(aq) by b. Ammonia-selective electrode.*
raising pH to above 11 with a strong base. NH3(aq) diffuses c. Magnetic stirrer, thermally insulated, with TFE-coated stir-
through the membrane and changes the internal solution pH that ring bar.
is sensed by a pH electrode. The fixed level of chloride in the
internal solution is sensed by a chloride ion-selective electrode 3. Reagents
that serves as the reference electrode. Potentiometric measure-
ments are made with a pH meter having an expanded millivolt
a. Ammonia-free water: See Section 4500-NH3.B.3a. Use for
scale or with a specific ion meter.
making all reagents.
b. Scope and application: This method is applicable to the
b. Sodium hydroxide, 10N.
measurement of 0.03 to 1400 mg NH3-N/L in potable and
c. NaOH/EDTA solution, 10N: Dissolve 400 g NaOH in 800
surface waters and domestic and industrial wastes. High concen- mL water. Add 45.2 g ethylenediaminetetraacetic acid, tetraso-
trations of dissolved ions affect the measurement, but color and dium salt, tetrahydrate (Na4EDTA 䡠 4 H2O) and stir to dissolve.
turbidity do not. Sample distillation is unnecessary. Use standard Cool and dilute to 1000 mL.
solutions and samples that have the same temperature and con- d. Stock ammonium chloride solution: Dissolve 3.819 g anhy-
tain about the same total level of dissolved species. The ammo- drous NH4Cl (dried at 100°C) in water, and dilute to 1000 mL;
nia-selective electrode responds slowly below 1 mg NH3-N/L; 1.00 mL ⫽ 1.00 mg N ⫽ 1.22 mg NH3.
hence, use longer times of electrode immersion (2 to 3 min) to e. Standard ammonium chloride solutions: See ¶ 4a below.
obtain stable readings.
c. Interference: Amines are a positive interference. This may
be enhanced by acidification. Mercury and silver interfere by 4. Procedure
complexing with ammonia, unless the NaOH/EDTA solution
(3c) is used. a. Preparation of standards: Prepare a series of standard
d. Sample preservation: Refrigerate at 4°C for samples to be solutions covering the concentrations of 1000, 100, 10, 1, and 0.1
analyzed within 24 h. Preserve samples high in organic and
nitrogenous matter, and any other samples for longer storage, by
lowering pH to 2 or less with conc H2SO4. * Orion Model 95-10, EIL Model 8002-2, Beckman Model 39565, or equivalent.
4-112 INORGANIC NONMETALS (4000)

mg NH3-N/L by making decimal dilutions of stock NH4Cl 4500-NH3:I. PRECISION AND BIAS OF AMMONIA-SELECTIVE ELECTRODE
solution with water.
Precision
b. Electrometer calibration: Place 100 mL of each standard
solution in a 150-mL beaker. Immerse electrode in standard of Mean Single
lowest concentration and mix with a magnetic stirrer. Limit Level Recovery Overall Operator
stirring speed to minimize possible loss of ammonia from the mg/L Matrix % % RSD % RSD
solution. Maintain the same stirring rate and a temperature of 0.04 Distilled water 200 125 25
about 25°C throughout calibration and testing procedures. Add a Effluent water 100 75 0
sufficient volume of 10N NaOH solution (1 mL usually is 0.10 Distilled water 180 50 10
sufficient) to raise pH above 11. If the presence of silver or Effluent water 470 610 10
mercury is possible, use NaOH/EDTA solution in place of 0.80 Distilled water 105 14 5
NaOH solution. If it is necessary to add more than 1 mL of either Effluent water 105 38 7.5
NaOH or NaOH/EDTA solution, note volume used, because it is 20 Distilled water 95 10 5
Effluent water 95 15 10
required for subsequent calculations. Keep electrode in solution
100 Distilled water 98 5 2
until a stable millivolt reading is obtained. Do not add NaOH Effluent water 97 — —
solution before immersing electrode, because ammonia may be 750 Distilled water 97 10.4 1.6
lost from a basic solution. Repeat procedure with remaining Effluent water 99 14.1 1.3
standards, proceeding from lowest to highest concentration. Wait
until the reading has stablized (at least 2 to 3 min) before
Source: AMERICAN SOCIETY FOR TESTING AND MATERIALS. Method 1426-79. Amer-
recording millivolts for standards and samples containing ⱕ 1 ican Soc. Testing & Materials, Philadelphia, Pa.
mg NH3-N/L.
c. Preparation of standard curve: Using semilogarithmic
graph paper, plot ammonia concentration in milligrams NH3-N 6. Precision and Bias
per liter on the log axis vs. potential in millivolts on the linear
axis starting with the lowest concentration at the bottom of the For the ammonia-selective electrode in a single laboratory
scale. If the electrode is functioning properly a tenfold change of using surface water samples at concentrations of 1.00, 0.77, 0.19,
NH3-N concentration produces a potential change of about 59 and 0.13 mg NH3-N/L, standard deviations were ⫾0.038,
mV. ⫾0.017, ⫾0.007, and ⫾0.003, respectively. In a single labora-
d. Calibration of specific ion meter: Refer to manufacturer’s tory using surface water samples at concentrations of 0.10 and
instructions and proceed as in ¶s 4a and b. 0.13 mg NH3-N/L, recoveries were 96% and 91%, respectively.
e. Measurement of samples: Dilute if necessary to bring The results of an interlaboratory study involving 12 laboratories
NH3-N concentration to within calibration curve range. Place using the ammonia-selective electrode on distilled water and
100 mL sample in 150-mL beaker and follow procedure in ¶ 4b effluents are summarized in Table 4500-NH3:I.
above. Record volume of 10N NaOH added. Read NH3-N con-
centration from standard curve. 7. Bibliography

5. Calculation BANWART, W.L., J.M. BREMNER & M.A. TABATABAI. 1972. Determination
of ammonium in soil extracts and water samples by an ammonia

冋 册
100 ⫹ D electrode. Comm. Soil Sci. Plant Anal. 3:449.
mg NH3-N/L ⫽ A ⫻ B ⫻ MIDGLEY, C. & K. TERRANCE. 1972. The determination of ammonia in
100 ⫹ C condensed steam and boiler feed-water with a potentiometric am-
monia probe. Analyst 97:626.
where: BOOTH, R.L. & R.F. THOMAS. 1973. Selective electrode determination of
A ⫽ dilution factor, ammonia in water and wastes. Environ. Sci. Technol. 7:523.
B ⫽ concentration of NH3-N/L, mg/L, from calibration curve, U.S. ENVIRONMENTAL PROTECTION AGENCY. 1979. Methods for Chemical
C ⫽ volume of 10N NaOH added to calibration standards, mL, Analysis of Water and Wastes. EPA-600/4-79-020, National Envi-
and ronmental Research Center, Cincinnati, Ohio.
D ⫽ volume of 10N NaOH added to sample, mL. AMERICAN SOCIETY FOR TESTING AND MATERIALS. 1979. Method 1426 –79.
American Soc. Testing & Materials, Philadelphia, Pa.

4500-NH3 E. Ammonia-Selective Electrode Method Using Known Addition

1. General Discussion suring occasional samples because no calibration is needed.

Because an accurate measurement requires that the concentration
a. Principle: When a linear relationship exists between con- at least double as a result of the addition, sample concentration
centration and response, known addition is convenient for mea- must be known within a factor of three. Total concentration of
NITROGEN (AMMONIA) (4500-NH3)/Ammonia-Selective Electrode Method Using Known Addition 4-113

ammonia can be measured in the absence of complexing agents TABLE 4500-NH3:II.* VALUES OF Q VS. ⌬E (59 MV SLOPE) FOR
down to 0.8 mg NH3-N/L or in the presence of a large excess (50 10% VOLUME CHANGE
to 100 times) of complexing agent. Known addition is a conve- ⌬E Q ⌬E Q ⌬E Q ⌬E Q ⌬E Q
nient check on the results of direct measurement.
b. See Section 4500-NH3.D.1 for further discussion. 5.0 0.297 9.0 0.178 16.0 0.0952 24.0 0.0556 32.0 0.0354
5.1 0.293 9.1 0.176 16.2 0.0938 24.2 0.0549 32.2 0.0351
2. Apparatus 5.2 0.288 9.2 0.174 16.4 0.0924 24.4 0.0543 32.4 0.0347
5.3 0.284 9.3 0.173 16.6 0.0910 24.6 0.0536 32.6 0.0343
5.4 0.280 9.4 0.171 16.8 0.0897 24.8 0.0530 32.8 0.0340
Use apparatus specified in Section 4500-NH3.D.2.
5.5 0.276 9.5 0.169 17.0 0.0884 25.0 0.0523 33.0 0.0335
3. Reagents 5.6 0.272 9.6 0.167 17.2 0.0871 25.2 0.0517 33.2 0.0333
5.7 0.268 9.7 0.165 17.4 0.0858 25.4 0.0511 33.4 0.0329
Use reagents specified in Section 4500-NH3.D.3. 5.8 0.264 9.8 0.164 17.6 0.0846 25.6 0.0505 33.6 0.0326
Add standard ammonium chloride solution approximately 10 5.9 0.260 9.9 0.162 17.8 0.0834 25.8 0.0499 33.8 0.0323
times as concentrated as samples being measured.
6.0 0.257 10.0 0.160 18.0 0.0822 26.0 0.0494 34.0 0.0319
4. Procedure 6.1 0.253 10.2 0.157 18.2 0.0811 26.2 0.0488 34.2 0.0316
6.2 0.250 10.4 0.154 18.4 0.0799 26.4 0.0482 34.4 0.0313
6.3 0.247 10.6 0.151 18.6 0.0788 26.6 0.0477 34.6 0.0310
a. Dilute 1000 mg/L stock solution to make a standard solution
6.4 0.243 10.8 0.148 18.8 0.0777 26.8 0.0471 34.8 0.0307
about 10 times as concentrated as the sample concentrate.
b. Add 1 mL 10N NaOH to each 100 mL sample and imme- 6.5 0.240 11.0 0.145 19.0 0.0767 27.0 0.0466 35.0 0.0304
diately immerse electrode. When checking a direct measure- 6.6 0.237 11.2 0.143 19.2 0.0756 27.2 0.0461 36.0 0.0289
ment, leave electrode in 100 mL of sample solution. Use mag- 6.7 0.234 11.4 0.140 19.4 0.0746 27.4 0.0456 37.0 0.0275
netic stirring throughout. Measure mV reading and record as E1. 6.8 0.231 11.6 0.137 19.6 0.0736 27.6 0.0450 38.0 0.0261
c. Pipet 10 mL of standard solution into sample. Thoroughly 6.9 0.228 11.8 0.135 19.8 0.0726 27.8 0.0445 39.0 0.0249
stir and immediately record new mV reading as E2.
7.0 0.225 12.0 0.133 20.0 0.0716 28.0 0.0440 40.0 0.0237
5. Calculation 7.1 0.222 12.2 0.130 20.2 0.0707 28.2 0.0435 41.0 0.0226
7.2 0.219 12.4 0.128 20.4 0.0698 28.4 0.0431 42.0 0.0216
7.3 0.217 12.6 0.126 20.6 0.0689 28.6 0.0426 43.0 0.0206
a. ⌬ E ⫽ E1 ⫺ E2. 7.4 0.214 12.8 0.123 20.8 0.0680 28.8 0.0421 44.0 0.0196
b. From Table 4500-NH3:II find the concentration ratio, Q,
corresponding to change in potential, ⌬ E. To determine original 7.5 0.212 13.0 0.121 21.0 0.0671 29.0 0.0417 45.0 0.0187
total sample concentration, multiply Q by the concentration of 7.6 0.209 13.2 0.119 21.2 0.0662 29.2 0.0412 46.0 0.0179
the added standard: 7.7 0.207 13.4 0.117 21.4 0.0654 29.4 0.0408 47.0 0.0171
7.8 0.204 13.6 0.115 21.6 0.0645 29.6 0.0403 48.0 0.0163
Co ⫽ Q Cs 7.9 0.202 13.8 0.113 21.8 0.0637 29.8 0.0399 49.0 0.0156

8.0 0.199 14.0 0.112 22.0 0.0629 30.0 0.0394 50.0 0.0149
where:
8.1 0.197 14.2 0.110 22.2 0.0621 30.2 0.0390 51.0 0.0143
Co ⫽ total sample concentration, mg/L,
8.2 0.195 14.4 0.108 22.4 0.0613 30.4 0.0386 52.0 0.0137
Q ⫽ reading from known-addition table, and
8.3 0.193 14.6 0.106 22.6 0.0606 30.6 0.0382 53.0 0.0131
Cs ⫽ concentration of added standard, mg/L. 8.4 0.190 14.8 0.105 22.8 0.0598 30.8 0.0378 54.0 0.0125
c. To check a direct measurement, compare results of the two
8.5 0.188 15.0 0.103 23.0 0.0591 31.0 0.0374 55.0 0.0120
methods. If they agree within ⫾4%, the measurements probably
8.6 0.186 15.2 0.1013 23.2 0.0584 31.2 0.0370 56.0 0.0115
are good. If the known-addition result is much larger than the 8.7 0.184 15.4 0.0997 23.4 0.0576 31.4 0.0366 57.0 0.0110
direct measurement, the sample may contain complexing agents. 8.8 0.182 15.6 0.0982 23.6 0.0569 31.6 0.0362 58.0 0.0105
8.9 0.180 15.8 0.0967 23.8 0.0563 31.8 0.0358 59.0 0.0101
6. Precision and Bias
* Orion Research Inc. Instruction Manual, Ammonia Electrode, Model 95-12,
Boston, MA. 02129.
In 38 water samples analyzed by both the phenate and the
known-addition ammonia-selective electrode method, the elec-
trode method yielded a mean recovery of 102% of the values
obtained by the phenate method when the NH3-N concentrations of these samples were analyzed, the mean standard deviation was
varied between 0.30 and 0.78 mg/L. In 57 wastewater samples 1.32 mg N/L. In three measurements at a sewer outfall, distilla-
similarly compared, the electrode method yielded a mean recov- tion did not change statistically the value obtained by the elec-
ery of 108% of the values obtained by the phenate method using trode method. In 12 studies using standards in the 2.5- to 30-mg
distillation when the NH3-N concentrations varied between 10.2 N/L range, average recovery by the phenate method was 97%
and 34.7 mg N/L. In 20 instances in which two to four replicates and by the electrode method 101%.
4-114 INORGANIC NONMETALS (4000)

4500-NH3 F. Phenate Method

1. General Discussion TABLE 4500-NH3:III. PRECISION DATA FOR MANUAL PHENATE METHOD
BASED ON TRIPLICATE ANALYSES OF AMMONIUM SULFATE
a. Principle: An intensely blue compound, indophenol, is
Relative
formed by the reaction of ammonia, hypochlorite, and phenol
NH3-N Standard
catalyzed by sodium nitroprusside. Lab/ Concentration Optical Deviation
b. Interferences: Complexing magnesium and calcium with Analyst mg/L Density %
citrate eliminates interference produced by precipitation of these
ions at high pH. There is no interference from other trivalent 1/1 0.1 0.129 1.55
forms of nitrogen. Remove interfering turbidity by distillation or 1/2 0.1 0.114 9.66
filtration. If hydrogen sulfide is present, remove by acidifying 2/1 0.1 0.100 10.2
2/2 0.1 0.122 2.36
samples to pH 3 with dilute HCl and aerating vigorously until
3/1 0.1 0.112 3.61
sulfide odor no longer can be detected. 3/2 0.1 0.107 1.94
1/1 0.3 0.393 0.39
2. Apparatus 1/2 0.3 0.364 0.32
2/1 0.3 0.372 2.64
Spectrophotometer for use at 640 nm with a light path of 1 cm or 2/2 0.3 0.339 0.90
greater. 3/1 0.3 0.370 0.31
3/2 0.3 0.373 0.46
3. Reagents 1/1 0.5 0.637 0.77
1/2 0.5 0.630 0.56
2/1 0.5 0.624 1.65
a. Phenol solution: Mix 11.1 mL liquified phenol (ⱖ89%)
2/2 0.5 0.618 0.86
with 95% v/v ethyl alcohol to a final volume of 100 mL. Prepare 3/1 0.5 0.561 0.27
weekly. CAUTION: Wear gloves and eye protection when han- 3/2 0.5 0.569 0.91
dling phenol; use good ventilation to minimize all personnel
exposure to this toxic volatile substance.
640 nm. Prepare a blank and at least two other standards by
b. Sodium nitroprusside, 0.5% w/v: Dissolve 0.5 g sodium
diluting stock ammonia solution into the sample concentration
nitroprusside in 100 mL deionized water. Store in amber bottle
range. Treat standards the same as samples.
for up to 1 month.
c. Alkaline citrate: Dissolve 200 g trisodium citrate and 10 g 5. Calculations
sodium hydroxide in deionized water. Dilute to 1000 mL.
d. Sodium hypochlorite, commercial solution, about 5%. This Prepare a standard curve by plotting absorbance readings of
solution slowly decomposes once the seal on the bottle cap is standards against ammonia concentrations of standards. Com-
broken. Replace about every 2 months. pute sample concentration by comparing sample absorbance
e. Oxidizing solution: Mix 100 mL alkaline citrate solution with the standard curve.
with 25 mL sodium hypochlorite. Prepare fresh daily.
f. Stock ammonium solution: See Section 4500-NH3.D.3d. 6. Precision and Bias
g. Standard ammonium solution: Use stock ammonium solu-
tion and water to prepare a calibration curve in a range appro- For the manual phenate method, reagent water solutions of
priate for the concentrations of the samples. ammonium sulfate were prepared and analyzed by two analysts
in each of three laboratories. Results are summarized in Table
4. Procedure 4500-NH3:III.
7. Bibliography
To a 25-mL sample in a 50-mL erlenmeyer flask, add, with
thorough mixing after each addition, 1 mL phenol solution, 1 mL SOLORZANO, L. 1969. Determination of ammonia in natural waters by the
sodium nitroprusside solution, and 2.5 mL oxidizing solution. phenolhypochlorite method. Limnol. Oceanogr. 14:799.
Cover samples with plastic wrap or paraffin wrapper film. Let PARSONS, T.R., Y. MAITA & C.M. LALLI. 1984. A Manual of Chemical
color develop at room temperature (22 to 27°C) in subdued light and Biological Methods for Seawater Analysis. Pergamon Press,
for at least 1 h. Color is stable for 24 h. Measure absorbance at Elmsford, N.Y.
NITROGEN (AMMONIA) (4500-NH3)/Automated Phenate Method 4-115

4500-NH3 G. Automated Phenate Method

1. General Discussion b. Sulfuric acid, H2SO4, 5N, air scrubber solution: Carefully
add 139 mL conc H2SO4 to approximately 500 mL water, cool
a. Principle: Alkaline phenol and hypochlorite react with to room temperature, and dilute to 1 L.
ammonia to form indophenol blue that is proportional to the c. Sodium phenate solution: In a 1-L erlenmeyer flask, dis-
ammonia concentration. The blue color formed is intensified solve 93 mL liquid (ⱖ89%) phenol in 500 mL water. In small
with sodium nitroprusside. increments and with agitation, cautiously add 32 g NaOH. Cool
b. Interferences: Seawater contains calcium and magnesium ions flask under running water and dilute to 1 L. CAUTION: Minimize
in sufficient concentrations to cause precipitation during analysis. exposure of personnel to this compound by wearing gloves and
Adding EDTA and sodium potassium tartrate reduces the problem. eye protection, and using proper ventilation.
Eliminate any marked variation in acidity or alkalinity among d. Sodium hypochlorite solution: Dilute 250 mL bleach solu-
samples because intensity of measured color is pH-dependent. Like- tion containing 5.25% NaOCl to 500 mL with water.
wise, insure that pH of wash water and standard ammonia solutions e. EDTA reagent: Dissolve 50 g disodium ethylenediamine
approximates that of sample. For example, if sample has been tetraacetate and approximately six pellets NaOH in 1 L water.
preserved with 0.8 mL conc H2SO4/L, include 0.8 mL conc For salt-water samples where EDTA reagent does not prevent
H2SO4/L in wash water and standards. Remove interfering turbidity precipitation of cations, use sodium potassium tartrate solution
by filtration. Color in the samples that absorbs in the photometric prepared as follows:
range used for analysis interferes. Sodium potassium tartrate solution: To 900 mL water add
c. Application: Ammonia nitrogen can be determined in po- 100 g NaKC4H4O6 䡠 4H2O, two pellets NaOH, and a few boiling
table, surface, and saline waters as well as domestic and indus- chips, and boil gently for 45 min. Cover, cool, and dilute to 1 L.
trial wastewaters over a range of 0.02 to 2.0 mg/L when photo- Adjust pH to 5.2 ⫾ 0.05 with H2SO4. Let settle overnight in a
metric measurement is made at 630 to 660 nm in a 10- to 50-mm cool place and filter to remove precipitate. Add 0.5 mL polyoxy-
tubular flow cell at rates of up to 60 samples/h. Determine higher ethylene 23 lauryl ether* solution and store in stoppered bottle.
concentrations by diluting the sample. f. Sodium nitroprusside solution: Dissolve 0.5 g
Na2(NO)Fe(CN)5 䡠 2H2O in 1 L water.
2. Apparatus g. Ammonia standard solutions: See Sections 4500-NH3.D.3d
and e. Use standard ammonia solution and water to prepare the
Automated analytical equipment: An example of the required calibration curve in the appropriate ammonia concentration
continuous-flow analytical instrument consists of the inter- range. To analyze saline waters use substitute ocean water of the
changeable components shown in Figure 4500-NH3:1. following composition to prepare calibration standards:

3. Reagents Concentration
Constituent g/L
a. Ammonia-free distilled water: See Section 4500-NH3.B.3a. NaCl 24.53
Use for preparing all reagents and dilutions. MgCl2 5.20
CaCl2 1.16
KCl 0.70
SrCl2 0.03
Na2SO4 4.09
NaHCO3 0.20
KBr 0.10
H3BO3 0.03
NaF 0.003

Subtract blank background response of substitute seawater from

standards before preparing standard curve.

4. Procedure

a. Eliminate marked variation in acidity or alkalinity among

samples. Adjust pH of wash water and standard ammonia solu-
tions to approximately that of sample.
b. Set up manifold and complete system as shown in Figure
4500-NH3:1.
c. Obtain a stable base line with all reagents, feeding wash
water through sample line.

Figure 4500-NH3:1. Ammonia manifold.

* Brij-35, available from ICI Americas, Wilmington, DE.
4-116 INORGANIC NONMETALS (4000)

d. Typically, use a 60/h, 6:1 cam with a common wash. at concentrations of 0.16 and 1.44 mg NH3-N/L, recoveries were
107 and 99%, respectively.
5. Calculation
7. Bibliography
Prepare standard curves by plotting response of standards
processed through the manifold against NH3-N concentrations in HILLER, A. & D. VAN SLYKE. 1933. Determination of ammonia in blood.
standards. Compute sample NH3-N concentration by comparing J. Biol. Chem. 102:499.
sample response with standard curve. FIORE, J. & J.E. O’BRIEN. 1962. Ammonia determination by automatic
analysis. Wastes Eng. 33:352.
AMERICAN SOCIETY FOR TESTING AND MATERIALS. 1966. Manual on Indus-
6. Precision and Bias
trial Water and Industrial Waste Water, 2nd ed. American Soc.
Testing & Materials, Philadelphia, Pa.
For an automated phenate system in a single laboratory using O’CONNOR, B., R. DOBBS, B. VILLIERS & R. DEAN. 1967. Laboratory
surface water samples at concentrations of 1.41, 0.77, 0.59, and distillation of municipal waste effluents. J. Water Pollut. Control
0.43 mg NH3-N/L, the standard deviation was ⫾0.005 mg/L, and Fed. 39:25.

4500-NH3 H. Flow Injection Analysis

1. General Discussion

a. Principle: A water sample containing ammonia or ammo-

nium cation is injected into an FIA carrier stream to which a
complexing buffer, alkaline phenol, and hypochlorite are added.
This reaction, the Berthelot reaction, produces the blue indophe-
nol dye. The blue color is intensified by the addition of nitro-
ferricyanide. The resulting peak’s absorbance is measured at 630
nm. The peak area is proportional to the concentration of am-
monia in the original sample.
Also see Section 4500-NH3.F and Section 4130, Flow Injec-
tion Analysis (FIA). Figure 4500-NH3:2. FIA ammonia manifold.
b. Interferences: Remove large or fibrous particulates by fil-
tering sample through glass wool. Guard against contamination
from reagents, water, glassware, and the sample preservation 3. Reagents
process.
Also see Section 4500-NH3.A. Some interferents are removed Use reagent water (⬎10 megohm) to prepare carrier and all
by distillation; see Section 4500-NH3.B. solutions. To prevent bubble formation, degas carrier and buffer
with helium. Use He at 140 kPa (20 psi) through a helium
degassing tube. Bubble He through 1 L solution for 1 min. As an
2. Apparatus alternative to preparing reagents by weight/weight, use weight/
volume.
Flow injection analysis equipment consisting of: a. Buffer: To a 1-L tared container add 50.0 g disodium
a. FIA injection valve with sample loop or equivalent. ethylenediamine tetraacetate and 5.5 g sodium hydroxide,
b. Multichannel proportioning pump. NaOH. Add 968 mL water. Mix with a magnetic stirrer until
c. FIA manifold (Figure 4500-NH3:2) with tubing heater and dissolved.
flow cell. In Figure 4500-NH3:2, relative flow rates only are b. Phenolate: CAUTION: Wear gloves. Phenol causes severe
shown. Tubing volumes are given as an example only; they may burns and is rapidly absorbed into the body through the skin. To
be scaled down proportionally. Use manifold tubing of an inert a tared 1-L container, add 888 g water. Add 94.2 g 88% liquefied
material such as TFE.* phenol or 83 g crystalline phenol, C6H5OH. While stirring,
d. Absorbance detector, 630 nm, 10-nm bandpass. slowly add 32 g NaOH. Cool and invert to mix thoroughly. Do
e. Injection valve control and data acquisition system. not degas.
c. Hypochlorite: To a tared 500-mL container add 250 g
5.25% sodium hypochlorite, NaOCl bleach solution† and 250 g
* Teflon or equivalent. water. Stir or shake to mix.
NITROGEN (AMMONIA) (4500-NH3)/Flow Injection Analysis 4-117

d. Nitroprusside: To a tared 1-L container add 3.50 g sodium TABLE 4500-NH3:IV. RESULTS OF SINGLE-LABORATORY STUDIES WITH
nitroprusside (sodium nitroferricyanide), Na2Fe(CN)5NO 䡠 2H2O, SELECTED MATRICES
and 1000 g water. Invert to mix. Known Relative
e. Stock ammonia standard, 1000 mg N/L: In a 1-L volumet- Addition Standard
ric flask dissolve 3.819 g ammonium chloride, NH4Cl, that has Sample/Blank mg Recovery Deviation
been dried for 2 h at 110°C, in about 800 mL water. Dilute to Matrix Designation NH3-N/L % %
mark and invert to mix. Wastewater Reference — 104 —
f. Standard ammonia solutions: Prepare ammonia standards treatment sample*
in desired concentration range, using the stock standard (¶ 3e), plant Blank† 0.4 100 —
and diluting with water. influent 0.8 102 —
Site A‡§ 0 — 0.5
0.4 108 —
4. Procedure 0.8 105 —
Site B‡§ 0 — 0.5
0.4 105 —
Set up a manifold equivalent to that in Figure 4500-NH3:2 and 0.8 105 —
follow method supplied by manufacturer or laboratory standard Site C‡§ 0 — 1.1
operating procedure for this method. Follow quality control 0.4 109 —
procedures described in Section 4020. 0.8 107 —
Wastewater Reference — 106 —
treatment sample*
5. Calculations plant Blank† 0.4 105 —
effluent 0.8 105 —
Prepare standard curves by plotting the absorbance of stan- Site A‡㛳 0 — ND
0.4 90 —
dards processed through the manifold versus ammonia concen-
0.8 88 —
tration. The calibration curve is linear. Site B‡㛳 0 — ⬍0.1
0.4 93 —
0.8 94 —
6. Precision and Bias Site C‡㛳 0 — ⬍0.1
0.4 89 —
a. Recovery and relative standard deviation: The results of 0.8 91 —
single-laboratory studies with various matrices are given in Landfill Reference — 106 —
Table 4500-NH3:IV. leachate㛳 sample*
b. MDL: A 650-␮L sample loop was used in the method Blank† 0.4 105 —
described above. Using a published MDL method,1 analysts ran 0.8 106 —
Site A‡# 0 — 1.9
21 replicates of a 0.020-mg N/L standard. These gave a mean of 0.4 125 —
0.0204 mg N/L, a standard deviation of 0.0007 mg N/L, and an 0.8 114 —
MDL of 0.002 mg N/L. Site B‡# 0 — 0.4
0.4 96 —
0.8 106 —
7. Reference Site C‡# 0 — 0.2
0.4 102 —
1. U.S. ENVIRONMENTAL PROTECTION AGENCY. 1989. Definition and Pro- 0.8 107 —
cedure for the Determination of Method Detection Limits. Appendix ND ⫽ not detectable.
B to 40 CFR 136 rev. 1.11 amended June 30, 1986. 49 CFR 43430. * U.S. EPA QC sample, 1.98 mg N/L.
† Determined in duplicate.
‡ Samples without known additions determined four times; samples with known
additions determined in duplicate.
§ Site A and C samples diluted 20-fold, Site B sample diluted 100-fold. Typical
relative difference between duplicates 0.2%.
㛳 Samples not diluted. Typical relative difference between duplicates ⬍1%.
# Site A and C samples diluted 50-fold; Site B sample diluted 150-fold. Typical
relative difference between duplicates 0.3%.