MIDWEST RESEARCH INSTITUTE
IDENTIFICATION OF ANALYTICAL PROCEDURES FOR NITRILOTRIACETIC ACID (NTA)
IN WATER, SEDIMENT, AND DRINKING WATER
TASK 40
FINAL REPORT
November 2, 1981
EPA Prime Contract No. 68-01-5915
MRI Project No. 4901-A(40)
Prepared for:
U.S. Environmental Protection Agency
Office of Pesticides and Toxic Substances
Field Studies Branch TS-798
401 M Street, S.W.
Washington, D.C. 20460
•
Attn: Dr. Frederick Kutz, Project Officer
Mr. Daniel Heggem
MIDWEST RESEARCH INSTITUTE 425 VOLKER BOULEVARD. KANSAS CITY. MISSOURI 64110 • 816753-7600
-------
MRI-NORTH STAR LABORATORIES 10701 Red Circle Drive, Minnetonka, Minnesota 55343 • 612933-7880
MRI WASHINGTON, D.C. 20006-Suite 250,1750 K Street, N.W. • 202 293-3800
-------
IDENTIFICATION OF ANALYTICAL PROCEDURES FOR NITRILOTRIACETIC ACID (NTA)
IN WATER, SEDIMENT, AND DRINKING WATER
by
Viorica Lopez-Avila
John E. Going
TASK 40
FINAL REPORT
November 2, 1981
EPA Prime Contract No. 68-01-5915
MRI Project No. 4901-A(40)
Prepared for:
U.S. Environmental Protection Agency
Office of Pesticides and Toxic Substances
Field Studies Branch TS-798
401 M Street, S.W.
Washington, D.C. 20460
Attn: Dr. Frederick Kutz, Project Officer
Mr. Daniel Heggem
MIDWEST RESEARCH INSTITUTE 425 VOLKER BOULEVARD, KANSAS CITY, MISSOURI 64110 • 816753-7600
-------
DI S CLAI MER
This docunierit has been reviewed and approved for publication by the
Office of Toxic Substances, Office of Pesticides and Toxic Substances,
U.S. Environmental Protection Agency. Approval does not signify that
the contents necessarily reflect the views and policies of the Environ-
mental Protection Agency, nor does the mention of trade names or com-
mercial products constitute endorsement or recommendation for use.
-------
PREFACE
This report presents the results of the literature review accomplished on
MRI Project No. 4901-A, Task 40, “Identification of Analytical Procedures for
Nitrilotriacetic Acid (NTA) in Water, Sediment, and Drinking Water,’ for the
Environmental Protect.ion Agency (EPA Prime Contract No. 68-OF-5915). The re-
view was performed by Dr. Viorica Lopez-Avila. The report was prepared by
Dr. Lopez-Avila with assistance from Dr. John Going. Dr. Mitcheil Erickson
contributed to the final editing.
Midwest Research Institute would like to thank Drs. Larry Games and Gil
Cloyd from Procter and Gamble for providing valuable assistance during the
course of this task. The cooperation of personnel from Monsanto Company
(Hr. Ed Halec) and W. R. Grace Company (Mr. J. Amairsakris) is sincerely
appreciated.
Approved:
James L. Spigarelli, Director
Analytical Chemistry Department
E. Going, Head
Environmental Analysis Section
11
-------
1. Summary.
2. Recommendations .
3. Introduction
4. Analytical Techniques for NTA in Water, Sediments
Drinking Water
Sample Preservation
NTA Isolation from Sample Hatrix
Chemical Derivatization
Gas Chromatography
High Pressure Liquid Chromatography . .
Other Analytical Techniques
5. NTA Tionitoring Data
Canadian Drinking Water Study
Canadian Wastewater Study
Other Studies: Removal of NTA by Activated
Preface
Figures
Tables.
CONTENTS
ii
iv
V
1
3
4
6
6
12
16
• . . . 19
• . . . 21
• . . • 24
27
• . . . 27
• . • 35
• . . 35
and
Sludge.
References.
39
iii
-------
FIGURES
Number Title Page
1 GC/NPD Chromatograms Showing the Effect of Removing Positive
Interferences to Obtain a Clean Separation of NTA - The
Concentrations of HCOOH used were: A. 0.1 H and 16 H;
B. 1 H and 10 H; C. 2 N and 8 H; D. 4 H and 8 N 14
2 Electron Impact Mass Spectrum of Tri-n-butyl ester of NTA.
The molecular ion, m/e 350, can be detected if the
spectrum is magnified Ca. lOx 22
3 Possible Pathways for Biodegradation of Nitrilotriacetic
Acid in Sewage 38
iv
-------
TABLES
Number Title Page
1 Summary of Published Methods for NTA Analysis 7
2 Effect of Vial Pretreatment and Mode of Sample Introduction:
pg of NTA Recovered from 10 pg Samples 11
3 Ion Exchange Resins 12
4 % Recovery of NTA by Concentration of DOWEX l-X8 Resins . . . . 13
5 Results of the Validation Study for the Modified Method Using
Samples Collected from Various Locations in the USA . 15
6 Derivatization Methods Used for NTA . 17
7 Esterification of NTA: assignment of Factors . 19
8 Packing Materials Reported for Analysis of NTA Esters by
Gas Chromatography 20
9 Comparison of GC and GC/MS/SIH Results from Several Sewage
Samples Using Games’ Method 23
10 National Monitoring Program Samples 28
11 Summary of Ontario Samples 28
12 Summary of Quebec Samples 29
13 Groundwater Samples 31
14 NTA in Municipal Groundwaters 31
15 Levels of NTA Found in the 70 Municipalities Across Canada. . . 32
16 Mean Concentrations of NTA (in pg/liter) Estimated to be
Discharged into the Major Drainage Basins 34
17 Comparison of the Levels of NTA (in pg/liter) Found in Ten
Municipal Waters Surveyed in or Prior to 1975 and in 1976 . . 34
18 NTA Data from Four Treatment Plants in the Netherlands 36
19 Effect of Temperature on Removal of NTA Under Steady Conditions 37
V
-------
SECTION 1
SUMIIARY
The purpose of this task was to conduct a comprehensive literature review
of analytical techniques for nitrilotriacetic acid (NTA) in water, sediment,
and drinking water and to summarize the readily available monitoring data.
This literature review identified promising analytical techniques for NTA and
at the same time pointed out deficiencies and voids in the current methods.
Fifty articles dealing with analysis of NTA in various matrices (e.g.,
water, drinking water, sewage sludge, etc.) were retrieved from the open lit-
erature following a computerized literature search and contacts with private
companies. Areas which were specifically addressed in obtaining the pertinent
information were: compound stability, isolation from the matrix (extraction
efficiency, removal of matrix interferences), derivatization procedures for
gas chromatographic analysis, selectivity and sensitivity of the detection
system, and examples of monitoring studies.
The most promising analytical procedure, still under evalution by Procter
and Gamble Company, involves concentration of NTA from water samples using
anionic ion exchange resins, elution with formic acid, derivatization to tn—
n-butyl NTA, and analysis by gas chromatography with nitrogen selective detec-
tor (GC/NPD). Confirmation is done by gas chromatography/mass spectrometry
(GC/HS). Use of 14 C-labeled NTA spiked into water samples directly in the
field or prior to the isolation step allows correction of NTA recovery from
the sample matrix by scintillation counting measurement. Quantitation of NIA
by GC/NPD and GC/HS in the selected ion monitoring mode was comparable, al-
though daily fluctuations of the NPD response were reported. Use of an inter-
nal standard such as nitrilotripropionic acid (NPT) resulted in improved con-
sistency of the response factors. This method gave improved precision and
accuracy in influent sewage, effluent sewage, river water, and tap water at
concentrations < 100 pg/liter. Other analytical procedures discussed in the
report involve polarography, colonimetry, and potentiometry. Difficulties in
the determination of NTA by these techniques which arise from interference by
metals and naturally occurring ligands, such as ethylenediamine tetraacetic
acid (EDTA), are also discussed.
No analytical procedures to analyze for NTA in river sediments were re-
ported in the literature. Extraction of NTA from sediment with water, dilute
aqueous base, or organic solvent may be applicable. Once extracted in an
aqueous solution or an organic solvent, analytical procedures described for
water can be applied.
1
-------
1onitoring data for NTA in the Canadian environment and data regarding
removal of NTA by activated sludge treatment are presented. The results of
the Canadian monitoring program indicate relatively few locations where NTA
levels occurred above 10 pg/liter. The authors of the study concluded that
use of NTA as detergent builder is not likely to increase the levels of NTA
in the Canadian environment.
2
-------
SECTION 2
RE COMMENDAT IONS
• The analytical procedure involving concentration of NTA from water by
ion exchange chromtography, elution with formic acid, derivatization to a
volatile ester, and analysis by gas chromatography is recommended to analyze
for NTA at microgram-per-liter concentrations in water samples.
• Additional method development should address sample preservation pro-
cedures and NTA isolation from waters and sediments. Preservation of samples
by addition of formaldehyde solution was frequently reported; however, no jus-
tification for selection of this preservative was provided. Other preserva-
tion reagents, such as formic acid, should be investigated. Investigation of
other ion exchange resins that would allow concentration of larger volumes of
water samples and thus lower detection limits for NTA should also be consid-
ered. The necessity and methods for treatment of glassware to prevent NTA
losses should be further studied.
• Use of ‘ 3 C-labeled NTA to be spiked in the field is strongly recom-
mended; this is expected to improve the method precision and accuracy since
the recovery of NTA can be corrected by the recovery of 13 C-labeled NTA added
to the sample, as measured by GC/HS.
• The GC/ 1S technique is preferred to GC/NPD technique especially if 13 C-
labeled NTA is to be added to the sample. This will allow both positive con-
firmation and quantitation of NTA.
Use of capillary columns to chromatograph the water or sediment ex-
tracts containing the NTA ester should be investigated. The higher resolving
power of a capillary column over the packed column may help to reduce the need
for sample cleanup and increase the method sensitivity.
3
-------
SECTION 3
INTRODUCTION
Interest in the determination of NTA first occurred as a consequence of
the observation that small amounts of NTA had adverse effects on the quality
of indicator color change in some complexometric titrations. 1 After its in-
troduction as a partial replacement for sodium tripolyphosphate in detergent
formulations, the interest in the NTA determination shifted to drinking water,
inland waters, sewage, etc., because of the increased concern over effects of
NTA in the environment.
In the United States NTA was marketed as a detergent builder in the 1960’s.
However, in December 1970, the use of NTA in detergent formulations was discon-
tinued in the United States as the result of reported adverse health effects.
Despite this fact, Canada and Sweden continued to use NTA after a thorough
evaluation of all available toxicological data.
Currently NTA is manufactured by Monsanto and W. R. Grace who export it
to Canada and supply it for use in certain industrial processes. 2 When re-
strictions were placed on the use of phosphates in Canada, the use of NTA in-
creased. For example, early in 1972 the NTA content in detergents was 6%;
whereas in 1973 the average content was 15%. Consequently, the discharge of
NTA to the environment increased. It was estimated that a city with a popula-
tion of one million would discharge 5,000 lb of NTA per day with an average
concentration in the wastewater of 5 ppm. 3
In order to determine whether NTA discharged with the wastewaters would
build up in the aquatic system, a National Surveillance Program was initiated
in Canada to monitor levels of NTA in raw and drinking waters. 3 The results
of this study demonstrated that NTA did not appear to accumulate in the Canadian
water sources. During a 5—year period, the accumulated data indicate relatively
few locations where NTA levels occur above 10 pg/liter.
In the United States, at industry’s request, the Office of Testing and
Evaluation (aTE) reviewed risks associated with NTA. Based on findings of
this risk assessment, the Environmental Protection Agency (EPA) has decided
not to take any regulatory action against the resumed use of this compound in
laundry detergents. They recommend that NTA manufacturers and processors con-
duct monitoring and environmental studies in order to confirm predictions that
NTA does not accumulate in the environment.
The purpose of this task was to conduct a comprehensive literature re-
view of analytical techniques for NTA in water, sediment, and drinking water.
4
-------
This literature review was intended to identify techniques available for the
analysis of NTA in water and sediment and, at the same time, to point out the
deficiencies and voids in the current methods. The information regarding NTA
analysis was retrieved from the open scientific literature, government reports,
and unpublished literature from private companies contacted directly. The
next section summarizes the analytical methods, and the final section presents
NTA monitoring data.
5
-------
SECTION 4
ANALYTICAL TECHNIQUES FOR NTA IN WATER, SEDIMENTS
AND DRINKING WATER
A summary of published methods for NTA analysis is given in Table 1.
Examination of this table shows a wide range of detection methods (e.g. , gas
chromatography, high pressure liquid chromatography, polarography, colorimetry,
potentiometri.c titration, etc.) that vary in sensitivity, selectivity, complex-
ity, degree of evaluation, ease of operation, etc. A detailed discussion of
each of these techniques follows. Sample preservation and NTA isolation from
water and sediment are discussed first.
SAMPLE PRESERVATION
The importance of proper sample preservation cannot be overemphasized.
NTA was reported to be easily biodegradable and loss during sample shipment
and storage is expected if samples are not preserved or maintained at low tem-
peratures (< 5°C). The reported values of the half-life of NTA in receiving
waters vary from minutes to months, 3 being affected by the presence of heavy
metals, and the biological activity of the matrix.
This section describes procedures reported in the literature for preser-
ving water samples containing pg/liter levels of NTA. These include: addi-
tion of formaldehyde, sterilization by filtration and freezing, and ultraviolet
light treatment.
The most comi on preservation technique uses formaldehyde (1% solution of
37% solution) and storage at 5°C prior to analysis. 8 ’ 9 Storage at room temper-
ature in polyethylene bottles following preservation with 1% formaldehyde solu-
tion was also reported.’ 2 In spite of the extensive study surveying NTA in
the Canadian drinking waters, 3 ’ no details about the stability with time of
low levels of NTA in distilled water and natural water, preserved with formal-
dehyde solution, were given. An unpublished memorandum from L. M. Games at
Procter and Gamble, Cincinnati, Ohio, reports that formaldehyde is stabilized
for “long periods of time” by addition of 1% formalin to samples stored in
polyethylene.
6
-------
TABLE 1. SUMMARY OF PUBLISRED METHODS FOR NTA ANALYSIS
Description Reference
Gas chrornatographic techniques
Specific method for NTA, ethylenediaminetetraacetic acid 4
(EDTA) and diethylenetriaminepentaacetic acid (DTPA).
Aqueous sample is evaporated to dryness. NTA is deriva-
tized with CH 3 OH/BF 3 . Analysis by GC/FID. Concentrations
lower than 200 pg/liter NTA could not be detected.
Gas chromatographic determination of NTA as trimethyl 5
ester. Inland water samples. Lower limit of detection
25 pg/liter. Identity of ester confirmed by GC/HS.
Gas chromatographic determination of NTA as tripropyl 6
ester. Aqueous sample is evaporated to dryness and treated
with an n-propanol/acetyl chloride (10:1) mixture. The re-
sulting NTA-propyl ester is analyzed using a nitrogen phos-
phorus detector. Concentrations down to 1 pg/liter NTA
were detected in tap and river water.
Gas chromatography of tri-propyl ester with prior concentra- 7
tion on ion exchange resin. Microgram level. Determina-
tion in lake waters. No interference from fatty acids.
NTA was determined in aqueous systems ranging from tap 8
water to sewage effluents by use of anion-exchange resins,
derivatization with n-butanol/HC1 and gas chromatography.
Concentrations at 1 to 10,000 pg/liter were reported.
Determination of NTA using gas chromatography with nitro— 9
gen specific detector was evaluated and applied to esti-
mate the levels of NTA in Canadian raw and potable waters.
Concentration of NTA from water using ion exchange resins
and derivatization to its tri-n-butyl ester were used
(Aue’s Method, Ref. 8). Limit of detection 0.2 pg/liter.
NTA was analyzed at sub-pg/liter levels in raw water and 10
drinking water using gas chromatography with nitrogen-
selective detector. Au&s Method (Ref. 8) was used to con-
centrate NTA from the aqueous sample.
Gas chromatographic determination as tri-n-butyl ester 11
after anion-exchange chromatography. Quantitative analysis
at 10 pg/liter demonstrated NTA was separated as its tn-n-
butyl ester from 24 other fatty acid, phenolic, and poly-
carboxylic acids using OV-210 columns.
(continued)
7
-------
TABLE 1. (continued)
Description Reference
Gas chromatographic determination with nitrogen selective 12
detection of NTA as its tri-n-butyl ester after anion ex-
change chromatography. Method gives improved precision
and accuracy in influent sewage, effluent sewage, river
water, tap water at concentrations < 100 pg/liter. Varia-
tions in recovery due to matrix effects are overcome by
the use of ‘ 4 C-NTA as a tracer. Identification of NTA con-
firmed by GC/MS in the selected ion monitoring (SIN) mode.
Gas chromatographic determination of NTA and related amino- 13
polycarboxylic acids as n-butyl and n-trifluoroacetyl,
n-butylester derivatives. Limit of detection 10 pg/liter
for all compounds. Identification by GC/HS. Applied to
sewage influent, inland waters.
Method based on the gas chromatographic determination of 14, 15
NTA as its trimethylsilyl derivative. The sample is evapo-
rated to dryness directly in the reaction vial with an ex-
cess of the ammonium salt of EDTA to mask interfering cat-
ions and derivatized with bis(trirnethylsilyl)trifluoro-
acetarnide in dimethylformamide. Method is suitable to
samples containing 1 to 100 mg/liter NTA without a prior
concentration step.
Gas chromatographic determination of NTA esterified with
N,0-bis(trimethylsilyl)acetamide. Confirmation by GC/HS.
Gas chromatographic determination in natural waters as 17
tris(2-chloroethyl ester). Specificity claimed in sewage
samples. Preconcentration on anion-exchange resin; elu—
tion with HC1. Identity of ester confirmed by GC/NS.
High pressure liquid chromatography
* Anion exchange chromatography was investigated to separate 18
NTA from other amino-acid chelates. Possible interferences
from metallic ions were overcome by converting all metal-
NTA chelates to ferric chelate. Method applied to analysis
of sewage samples and solutions of detergent formulations
with a sensitivity of 1 mg/liter.
Polarographic techniques
* Polarography with cadmium. NTA is converted to its 1:1 19
cadmium complex by adding a 100% excess of cadmium. De-
termination of NTA in lake waters (1 to 10 mg/liter).
(continued)
8
-------
TABLE 1. (continued)
Description Reference
Differential cathode-ray polarography was used to deter- 20
mine “free” NTA in natural water as the lead-NTA complex.
Determination of “total” NTA is based on releasing NTA
from metal complexes under acidic conditions. Subsequent
addition of EDTA and adjustment of pH to 8 preferentially
bind the metals with EDTA. Detection limits of 10 pg/liter
were reported without any sample preconcentration.
Automated method for the determination of NTA in natural 21
waters, detergents, and sewage samples, based on the for-
mation of bismuth NTA complex at pH 2 using twin cell
oscillographic DC polarography; 15 samples/hour can be
processed. Nethod can be used to detect 10 pg/liter with-
out any sample preconcentration.
Polarographic analysis of NTA as In(III)-NTA complex. 22
The method was evaluated with river water and sewage sam-
ples at 25.7 ppb to 2.57 ppm using anion exchange concen-
tration and isotope dilution to correct for incomplete
recovery. No interference observed with a 10-fold molar
excess of EDTA and EHDP (ethane-l-hydroxy-l,l-diphospho-
nate), citrate, maleate, phosphate, and carbonate. No
apparent interference from surfactants.
Polarographic determination of NTA as Pb-NTA complex, Cd-
NTA complex, and Bi-NTA complex was evaluated. Pb-NTA
complex was investigated in three buffer systems: tris-
hydro-chloride-ammonium hydroxide, glycine-ammonium hydrox-
ide, and ammonium chloride-ammonium hydroxide, at pH 7 to 8.
Cd-NTA complex was measured in ammornum chloride-ammoniuin
hydroxide buffer at pH 9. Bi-NTA complex was investigated
in NaC1 media at pH 2. Polarographic determination of NTA
as the Bi-NTA complex was found to be the most sensitive
(50 pg/liter detection limit).
NTA was determined in natural and wastewaters at 3 pg/liter 23
as the bismuth complex using dual cell pulse differential
polarography. Effects of possible interference by metal
ions and complexants were evaluated.
Differential pulse polarography of NTA and EDTA as Cd- 24
complex in synthetic seawater and phytoplankton media.
The presence of competing metal cations, including copper
is not detrimental if the method of standard additions is
used.
(continued)
9
-------
TABLE 1. (concluded)
Description Reference
Colorimetric techniques
Determination in soils. Soil sample extracted with 25
0.025 Il NaOH and CaC1 2 solution. Extract analyzed for
NTA by photometric measurement of blue color (borate
buffer, pH 9.2) developed by modifying the zinc-
Zincon method. Calibration was linear to less than
10 pg/g, but no detection limit was reported.
Calorimetric determination of blue-green color of 26
Ni(II)-NTA chelate. Automatic determination with
Technicon Auto-Analyzer in detergent formulations.
* Colorimetric determination of Fe-NTA chelate in acti- 27
vated sludge effluents at concentrations between 5
and 50 mg/liter using the phenanthroline method.
Potentiometric titration
Potentiometric titration with Fe(III). Determination 28
of EDTA, NTA, and DTPA (diethylenetriaminepentaacetic
acid). Allows determination of 1 to 2% NTA in EDTA.
Potentiometric titration with Tl(III) and Cu(II). De— 29
termination of NTA in water and sewage. The lower de-
tection limit was 100 pg/liter of NTA. The relative
standard deviation in sewage samples was 5%, and the
recovery varied from 90 to 110%.
Atomic absorption spectroscopic technique
* Compleximetric atomic absorption spectroscopic method 30
based on the formation of a soluble Pb-NTA chelate.
10
-------
Freezing of water samples following sterilization by filtration (Whatman
GF/C, 2 to 3 |J pore size) was reported by Murray.5 Water samples containing
20 to 2,000 |jg/liter NTA were filtered, acidified with HC1 and kept frozen in
screw-capped bottles until analyzed. No change in NTA concentration was re-
ported for samples kept frozen up to 6 weeks. NTA in unfiltered lake water
was found to degrade in about 8 days at an unspecified temperature.7 If,
however, the sample was filtered through a 0.45 |J membrane filter, NTA was
found to be stable for at least 20 days in Pyrex and polyethylene containers
without deterioration.7 Therefore, samples should be filtered immediately
after collection if storage at room temperature is anticipated.
Stolzberg14 reported that river water samples spiked with NTA at 1 to 50
mg/liter and incubated at room temperature under normal fluorescent lighting
showed constant concentration of NTA for 1 to 3 days, after which undetectable
levels of NTA were found.
A vial pretreatment procedure was investigated by Stolzberg.14 Results
in Table 2 indicate that vial pretreatment is unnecessary when handling milli-
gram quantities of NTA. Simple washing with distilled water and drying is
sufficient. However, when handling microgram quantities of NTA, it is espe-
cially important to preleach the reaction vials and to include ammoniacal EDTA
masking reagent in the procedure, since NTA may complex with alkali and alka-
line earth metal ions from glass.
TABLE 2. EFFECT OF VIAL PRETREATMENT AND MODE OF INTRODUCTION
Mg OF NTA RECOVERED FROM 10 (Jg SAMPLES3
Vial Introduction technique
pretreatment
None
Water washed
0.1 M HC1 soak
0.1 M EDTA soak
A
0.0
0.3
1.7
2.7
B
0.0
6.0
6.9
0.0
C
1.9
2.7
6.0
2.9
D
7.1
8.8
8.9
9.3
E
10.0
9.9
7.7
10.0
a Data taken from Reference 14.
b Conditions :
A - Sample added as 1.0 ml of 10 |Jg/ml NTA.
B - 1 ml water evaporated in vial, then
10 |Jl of 1 (Jg/ml NTA added.
C - 1 ml 10 |Jg/ral NTA + 1 drop cone. HC1.
D - 10 (Jl 1 (jg/ul NTA.
E - 1 ml 10 [Jg/ml NTA + 1 ml 150
EDTA.
Procedures for preservation of sediments for organic analysis include
freezing or sterilization by autoclaving. Addition of acid, such as formic
acid, that may inhibit bacterial activity for short period of time until
11
-------
analysis may also be applicable. No information dealing with analysis of NTA
in sediments is available; therefore, preservation procedures for sediments
need to be established.
NTA ISOLATION FROM SAMPLE MATRIX
Reported techniques for concentration of NTA from water, include ion ex-
change columns, freeze drying, and evaporation. Following is a brief review
of each of these techniques.
Ion Exchange Columns
Table 3 summarizes the type of ion exchange resins and their character-
istics considered important in the concentration process. In order for ex-
change to take place, the solution containing the compound of interest, in
this case NTA, is passed through the resin and the resin counterions are re-
placed by the sample ions of like charge. Neutral molecules and those having
the same charge as the resin functional groups pass through the resin. Fol-
lowing cleanup with small volumes of 0.1 H formic acid to wash—off other inter-
ferences, 16 H formic acid is then used to elute the sorbed ions. Formic acid
is then removed completely from the sample prior to derivatization.
TABLE 3. ION EXCHANGE RESINS
Resin
Base
Mesh
Exchange
group
Exchange
capacity
(ECV)
(meq/mL)
Form
Dowex 1-X8
Styrene-
divinylbenzene
50/100
CH 2 N(CH 3 ) 3
1.4
HCQO
Cl
Bio-Rad AG1-X2
Styrene-
d ivinyib enzene
50/100
CH 2 N(CH 3 ) 3
0.7
HCOO
Several studies which used ion exchange resins to concentrate NTA from
water will be briefly described below.
Davies 31 concentrated a number of carboxylic acids from water using an
ion exchange column and found that NTA can be concentrated on Dowex-1 resin
in the formate form and eluted with 5H HCOOH. Recovery of NTA by elution
with HCOOH of different inolarities and the effect of pH on recovery of NTA
using Dowex l-X8 resin was reported by Chau. 7 Results indicate that NTA can
be quantitatively retained by resin at pH 2.5 to 10 (see Table 4). Recovery
data of NTA by elution from resin with HCOOH of different molarities (1 to
8 H) indicate quantitative recovery with acid molarities > 2 H.
12
-------
TABLE 4. % RECOVERY OF NTA BY CONCENTRATION
ON DOWEX 1-X8 RESINS
Elution with
HCOOH
Effect
of pH
Acid
Volume
Recovery
Recovery
molarity
(ml)
(¼)
pH
(%)
1
100
56
1
0.3
2
75
96
2.2
90
2.5
75
99
2.6
101
3
75
99
3.5
99
4
75
101
5
102
5
50
102
6
100
6
50
101
8
101
7
50
99
8.6
102
8
50
101
9.5
10.3
103
102
Aue 8 used a similar resin, Bio-Rad AG1-X2 (different percentages of
crosslinking agent), to concentrate NTA. A resin with a lower percentage of
crosslinking has an open structure which is more permeable to higher molecular
weight substances than a high crosslinking resin. It also swells more when
in use.
Despite the fact that ion exchange columns are recommended to concentrate
NTA from water, no systematic study of the concentration process was reported.
Nost studies recommend use of 50 mm length x 5 mm ID resin bed for a 50 to
100 ml volume of water sample. Before runs, 16 N or 25 N formic acid is passed
through the column, followed by deionized water. The sample is poured into
the reservoir and allowed to flow at 3 mi/mm. Following the sample, 0.1 N
formic acid is used to wash off the impurities and NTA is eluted from the col-
umn with 16 N formic acid. 8 ’ 9 Malaiyandi 9 used this procedure to analyze for
NTA in drinking water and reported recoveries > 90¼ at concentrations ranging
from 1 to 25 fig/liter of water. Games 12 reported that use of 0.1 N HCOOH to
wash off interferences prior to elution of NTA led to positive interferences
which had a significant effect upon the recovery of NTA. Use of 2 N HCOOH
(10 mL) for initial cleanup and 8 N HCOOH (10 ml) to elute NTA gave good and
reproducible recoveries. Figure 1 is a series of chromatograms that show the
effect of removing positive interferences in order to improve the chromatog-
raphy of NTA. Each chromatogram was obtained on the same sample, which was
subjected to different cleanup and eluted with a different concentration of
HCOOH. NTA recovery was > 95% in A through C and 67% in D. The concentra-
tions of formic acid used were A, 0.1 N and 16 N; B, 1 N and 10 N; C, 2 N and
13
-------
A B
C
D
Figure 1 — GC/NPD chromatograms showing the effect of removing positive interferences to obtain
a clean separation of NTA — the concentrations of HCOOH used were: A. 0.1 M and 16 M;
B. 1 M and 10 M; C. 2 M and 8 H; D. 4 M and 8 M. Taken from Reference 12.
Acid
-------
8 N; D, 4 N and 8 N. Results of the validation study using samples from vari-
ous locations in the USA by the Games procedure,’ 2 slightly modified to allow
correction of recovery using ‘ 4 C-NTA, are given in Table 5. It can be noted
that relative standard deviations are about 10% at concentrations ranging from
1 to 50 pg/liter. Corrected NTA recoveries vary from 77 to 114%.
TABLE 5. RESULTS OF THE VALIDATION STUDY FOR THE
MODIFIED METHOD USING SAMPLES COLLECTED FROM
VARIOUS LOCATIONS IN TIlE USAa
NTA
added
(pg/liter)
No. of
samples
%
Recovery
Influent sewage
50
7
114 ± 10
Effluent sewage
10
8
106 ± 17
River water
5
5
77 ± 11
Tap water
1, 2, 3
9
88 ± 10
a Data taken from Reference 12.
Freeze Drying
Murray used freeze drying or lyophilization to concentrate small volumes
of samples (10 ml). Prior to lyophilization, 0.1 ml concentrated HC1 was
added to the sample. The ampules were shell-frozen in a dry-ice acetone mix-
ture and lyophilized on a freeze dryer. Lyophilization is applicable to all
water types and allows the separation of precipitated inorganic salts and by
redissolving the residue in an organic solvent. However, the technique is
very time consuming if larger volumes of water have to be handled. No recov-
ery data was available.
Evaporation
Evaporation of water samples in a porcelain dish at 115°C in a drying
oven or at 150°C in a sand bath under a moderate stream of nitrogen was re-
ported. 6 The residue was rinsed with 16 N formic acid, and the formic acid
was completely evaporated by heating the flask in a sand bath at 110°C under
nitrogen. No recovery data were reported.
Solvent Extraction
When NTA is present in a solid matrix, such as river sediment, use of
any of the techniques described above is not possible. Extraction from the
matrix with water or a dilute aqueous base might be possible since NTA is
15
-------
very soluble in water. Once extracted in an aqueous solution or an organic
solvent, the compound can be treated as a water sample as described above.
No data were found on the partition ratio of NTA between water and sediment,
so the probability that NTA is present in sediment cannot be predicted.
CHEMICAL DERIVATIZATION
Techniques available to derivatize NTA prior to gas chromatographic
analysis are reviewed. Derivatization is done to reduce compound polarity,
and to increase is volatility and its chemical stability. Derivatization,
however, is quite often avoided because it can seriously affect the precision
and accuracy of the method if not performed properly. Furthermore, derivati-
zation may be a potential source of contamination if reagents used are not
pure or may lead to undesirable by-products or artifacts. Since NTA is not
volatile, gas chromatographic analysis requires prior esterification. Table 6
summarizes the reagent procedures and the detection systems reported for NTA.
Each derivatization technique will be discussed briefly below.
Rudling 4 developed a method based on the derivatization of NTA in the
presence of EDTA using BF 3 in methanol (10% w/v). Concentrations lower than
200 pg/liter of NTA were difficult to analyze owing to the poor chromatographic
properties of the trimethyl ester of NTA.
Murray 5 used methanol/HC1 and dimethyl sulfite to derivatize NTA to its
trimethyl ester. Esterification was found to be at its maximum after 3 hr at
68°C. Shorter esterification times gave peaks that were thought to be mono-
and dimethyl esters of NTA.
Esterification of NTA with n-propanol/HC1 was reported by Chau. 7 Since
n-propanol refluxes efficiently at 100°C, this temperature was recommended.
Esterification times between 30 mm and 1 hr were found to give complete reac-
tion over the range of 1 to 20 pg.
Reichert 6 used n-propanol/acetyl chloride to esterify NTA to its propyl
ester. The derivatizing reagent in this case is the HC1-saturated n—propanol
which forms by mixing n—propanol with acetyl chloride in a ratio 10:1. The
propyl acetate which forms as by-product does not interfere with the esterifica-
tion. By use of n-propanol/acetyl chloride, quantitiative esterificatiori of
NTA occurs within 30 mm, and NTA is easily soluble in this mixture at the boil-
ing point.
Several papers 813 report derivatizatmon of NTA to its tributyl ester using
n-butanol/HC1. The esterification reaction was efficiently optimized by use of
a factorial experimental design.’ 3 Two HC1 concentrations, two reaction temper-
atures, and two reaction times were investigated. The results are given in
Table 7. The optimum conditions were found to be at: acid concentration 3 N,
temperature 65°C, and reaction time 45 mm. An increase in temperature to 150°C
caused the cleavage of NTA to iminodiacetic acid (IDA). Longer reaction times
also cause conversion of NTA to IDA.
16
-------
_________________________ ‘ IABLE 6. DERIVATIZA1 ION IIE1IIODS WEI ) FOR NIA ________
l)et.ect i oil
I)er i vati ye Reagent Proc e ,Itui-e system Re Fe ri’ui C
tii—MoLliyI ester BE 3 in methanol Evatiorale sample to dryness Add psL riftca— CC/Eli)
10% (w/v) lieu reagelit (1 (1 ml), seal auiipiile and place
in an ultrasonic hatch at 100°C for 40 mm
After cool tog, add chloroform (I ml) amid trans—
ter to i tube eoimlammuiuig 3 ml miller solution
(NaOilJI( 2 1 1P0 4 , p 11 7) SIm.i lie I mmii mud remove
chiorolorm layer.
HeLiiauiol/IIC I amid Evaporate sanipit, to dryness Add I nil HeOhl/ CC/FiB 5
dimethylsiml lie lICE amid 0 2 nil dumethmylsiulfite. Seal v,a s, bC/MS
plae iii water bath at. 68°C for 3 Imu Remove
excess reagents on a motam y evaporator
tri—Propyl estet n-Proiianol/am.etyl Evapouaie sample to dryness Add 2 ml GC/NPI) 6
chloride (10 1) n—Iiropamio [ /acetyl Llmloride Heat at 110°C for
30 miii in a sand bath heat mixture at 110°C
fot 10 miii while passing N 2 thiough to evapo—
rate solvent to I ml Cool Add 0 5 ml methyl-
cyclohexauie and 2 ml I N NaOhl Separate phases
n-Propanol/llC l I ml Al mqiiot of l IlA (10 g) evaporated to dry— GC/Fhl) 7
ness at 90 to 100°C Add 2 nil ii—propanol
salmirateil with IlCI Ref lux br 30 miii
tr i-n—Bulyl ester mi-Butanol/IICI 2 nil n-Bimlaiiol/3N lICl is added to the dry sample CC/11) 8
Cap tightly aiud ma miutaiui 25 miii iii au ultra— CC/NI’D 9
sonic bath at 75°C CC/MS 10
I I
12
13
(c-on l i iuiic ii)
-------
I-
TAI )I.E 6. (Lout. I iJ(lcd)
I)oL ct iou
Dertvative Reagecul Pro eduare system Ilpii’iu e
tri—rrirsetiiylstlyl ester Bis(truunetl iyl- Mesk interfering itiouis with EDIA (2 ing HYIA CC/ElI) 16
silyl)trtlluoro— per ml sdnhIIle) Evaporate to (Ilyiless (pieseuiue
acetainide (BSIFA) of moisture would c iuse hydrolysis of illS ulrrivj— 16
t i ye). Add 0 2 in I BS’I FA and 0 2 m I)MF and agi —
tale heal, scaled vial at 70°C for 30 miii, cool
I’yridi uie or acetouuu tn Ic can be used instead of
D 1IF
tri(2—Cluloroethyl) ester Boron trufluoride Evaporate sample to dryness at 100°C Add 1 ml CC/IC 17
In 2—chioroetluanol esterification reagent, seal ampule, place iii au CC/MS
10% w/v oil bath for 2 lii Dissolve iii benzeuue, wash
with waler
-------
TABLE 7. ESTERIFICATION OF NTA: ASSIGNNENT OF FACTORSa
Factors
Low level
High level
Acid concentration
1.25 N
3 N
Reaction temperature
65°C
100°C
Reaction time
10 mm
45 mm
a Data taken from Reference 13.
Games 12 reported that residual HC1 left in the vial after evaporation
contributes to rapid loss in the performance of the nitrogen selective detec-
tor. Therefore, 1 H HC1 was used, and the temperature was increased to 85°C.
Final traces of HC1/butanol were removed by adding methanol to the dried sample
and immediately evaporating it at 85°C.
Stolzberg’ 4 reported derivatization of NTA to its trimethylsilyl deriva-
tive. The sample is evaporated directly in the reaction vial with an excess
of the ainmonium salt of EDTA to mask interfering cations and to connect NTA
to a more volatile compound. To efficiently remove water from the vial, addi-
tion and evaporation at 70°C of two portions of dry methylene chloride is rec-
ommended. The importance of complete removal of water is demonstrated by the
fact that 1 mg of water can consume approximately 20 mg of BSTFA reagent.
Effects of vial pretreatment with microgram amounts of NTA was investigated,
since leaching of alkali and alkaline earth metal ions from the glass was
thought to cause incomplete derivatization. The results, summarized in
Table 2, show that preleaching of the reaction vials and masking the inter-
fering cations with ammonium salt of EDTA are required in order to avoid
erratic results.
To increase the sensitivity of the detection technique derivatization of
NTA to its tri(2-chloroethyl) ester and electron capture detection was re-
ported. 17
GAS CHROMATOGRAPHY
All gas chromatographic techniques reported in the literature require
derivatizatmon of NTA to its volatile ester. Derivatization techniques were
discussed in the preceeding section. A list of packing materials that have
been used to chromatograph the ester derivatives is given in Table 8.
19
-------
TABLE 8. PACKING MATERIALS REPORTED FOR ANALYSIS OF
NTA ESTERS BY GAS CHROMATOGRAPHY
GC packing material GC conditions Reference
5% OV-17 on Aeropak (100/120 mesh) 150-285°C 4
at 10°C/mm
2°h Ethylene glycol adipate on Chromosorb W isothermal 5
at 195°C
2% FFAP on Chrornosorb W isothermal 5
at 195°C
3% OV-1 on Chromosorb WHP (80/100 mesh) isothermal 6
at 195°C
3% OV-1 on Chromosorb WHP (80/100 mesh) 180-225°C 7
at 3°C/mm
Carbowax 2011 on Celite 545 (100/120 mesh) isothermal 8
(similar to 0.3% Carbowax 2011 on acid at 183°C
washed Chromosorb W)
5% OV-101 on Chromosorb WHP (80/100 mesh) isothermal 10
at 235°C
3% OV-210 on Chromosorb IIP (80/100 mesh) isothermal
at 200°C
3% OV-210 on acid washed Chromosorb W (60/80) 145°C (8 ruin) 11
to 240°C at
6°C/ruin
Ultrabond 2011 (100/200 mesh) 215°C (8 ruin) 12
to 250°C at
20°C/ruin
20
-------
The possibility that other compounds may interfere in the analysis of
NTA was investigated. 5 The esters of saturated, unsaturated, and branched
chain fatty acids were chromatographed with the NTA trimethyl ester. Inter-
ference was found with the C 18 iso ester. Possible breakdown products of NTA
such as methyliminodiacetic acid (tIIDA) and iminodiacetic acid (IDA) were re-
ported to have shorter retention times then the trimethyl NTA and would not
interfere in the analysis.
Several detectors, including flame ionization detector (FID), 4 ’ 5 ’ 7 ’’’’ 13 ’ 6
nitrogen selective detector (NPD or TSD), 6 ’ 9 ’ 10 ’’ 2 electron capture detector
(ECD)’ 7 and mass spectrometer (MS), 5 ’’°’’ 7 have been used to detect the NTA
esters. Both internal standard and external standard quantitation were re-
ported. Nitrilotripropionic acid was used as internal standard by Games. A
set of three standards at concentrations similar to those expected in the sam-
ples were run each day to establish the standard curve. The response to NTA
using NPD was linear from 1 pg/liter to 5,000 pg/liter.
Detection of NTA using an element selective detector such as NPD has a
higher degree of reliability over FID, however confirmation of NTA by GC/MS
is preferred. MS can be used at the same time to quantitate the samples. A
comparison of quantitation using NPD and GC/MS in the selected ion monitoring
mode (Sill) was reported operating the mass spectrometer in both the electron
impact (El) and the chemical ionization (CI) mode. Ammonia was used as the
CI gas. Ions selected for El were m/e 158 and m/e 258 (see mass spectrum in
Figure 2). For CI, ions monitored were at m/e 360 (N + 1) and m/e 246.
Table 9 shows a comparison of GC and GC/MS/SIM data for four sewage samples
(both influent and effluent). Analyses were done on the same extract after
derivatization so that variations in recovery would not affect the comparison.
Although daily fluctuations of the NPD response were reported, no solution
was found. It was recommended that a standard curve be obtained with each
set of four to five samples. Use of an internal standard of similar struc-
ture, such as nitrilotripropionic acid (NTP) resulted in improved consistency
of the response factors over the concentration range of interest. Methanol
was preferred over acetone as the solvent since the tributyl ester of NTP was
not sufficiently soluble in acetone. It is also recommended 12 that sample
extracts that had been derivatized and were ready to be analyzed for NTA,
should be stored dry at room temperature and addition of methanol should be
done immediately prior to analysis.
HIGH PRESSURE LIQUID CHROMATOGRAPHY
Analysis of NTA in sewage samples by high pressure liquid chromatography
using a strong anion exchange resin coated onto Zipax and 0.02 N Na 2 B 4 0 7 10 H 2 0
as the mobile phase was reported.’ 8 Possible interferences from metallic ions
were overcome by converting all metal—NTA chelates to ferric chelate. Following
addition of the ion salt, 100 p1 of the water sample was injected directly
onto the column. The column effluent was monitored at 254 nm. Although the
utility of this method has been demonstrated using both sewage samples and
detergent formulations, additional confirmation of NTA is required. For posi-
tive identification the hydrolyzed chelate had to be further purified and con-
firmed by both infrared and gas chromatographic techniques. Furthermore,
21
-------
1
U
C
0
-D
C
a)
>
a
a,
Figure 2 - Electron impact mass spectrum of tri-n-butyl ester of NTA.
The molecular ion, m/e 350, can be detected if the
spectrum is magnified Ca. lOx.
co
‘3- )
0 50 100 150 200 250 350
m/e
22
-------
TABLE 9. COMPARISON OF GC AND GC/HS/SIM RESULTS FROM
SEVERAL SEWAGE SANPLES USING GAHES’ METHODa
Sample type
GC quantitation (pg/ . )
(NPD detector)
SIN/GC/tlS quantitation
(pg/i)
Sewage influent
A
B
C
D
16
6
52
13
14
10
56
10
Sewage effluent
A
B
C
D
3
<1
6.1
1.2
5
0
11
0
a Data taken from Reference 12.
23
-------
method sensitivity is only 1 mg/liter and the response is linear only through
0.1 to 1.5 pg range. Beyond 1.5 pg the exchange capacity of the column was
exceeded and the NTA eluted early. At lower levels, peaks were quite broad
and difficult to measure. Because of this, the precision of the method was
quite poor. At 1 pg level the relative standard deviation was 2%, whereas at
0.2 pg the relative standard deviation was 25%.
OTHER ANALYTICAL TECENIQIJES
As briefly described in Table 1, there are various other analytical tech-
niques, such as polarography, colorimetry, and potentiometric titration, that
can be used to analyze for NTA. They are all based on the complexing ability
of NTA. Some of these methods (e.g., colorimetric, potentiometric titration)
lack the sensitivity and the selectivity required to detect trace levels of
NTA in environmental samples. Nonetheless, they were quite popular in the
late 1960’s due to their simplicity and the widely accepted reliability of
these colorimetric techniques. Following is a review of these techniques.
Polarographic techniques will be discussed first, followed by colorimetric
and potentiometric titration methods.
Polarographic methods are based on the reduction of a suitable metal ion
complexed with NTA. Since the half-wave potentials for the reduction of com-
plexes of the same ion vary appreciably with the ligand, the procedure is al-
most as specific for NTA as the GC methods. Daniel and LeBlank 32 described a
polarographic method for the analysis of NTA in presence of EDTA. A+cadmium
complex was formed by addition of excess cadmium salt. The free Cd 2 and the
Cd-NTA complex give two well resolved polarographic wav s. The half-wave
potential is -0.97 V for Cd-NTA complex, -0.6 V for Cd 2 , and -1.4 V for
Cd-EDTA complex. This method was used by Asplund’ 9 to analyze samples of
lake water containing NTA at concentrations from 1 to 10 mg/liter.
Afghan 2 ° used differential cathode ray polarography to detect NTA as
“free” NTA and as “total” NTA at levels as low as 10 pg/liter without any sam-
ple preconcentration. “Free” NTA in natural water is determined using lead
as reagent and pH 8 and varying the potential from -0.5 V to -1.0 V versus
mercury pool reference electrode. The Pb-NTA complex has a half-wave poten-
tial at -0.9 V. Determination of “total” NTA is based on releasing NTA from
metal complexes under acidic conditions. Addition of EDTA and adjustment of
pH to 8 will preferentially bind the metals with EDTA. NTA is then measured
as indicated above by forming the Pb-NTA complex. Interferences were found
from ethylene glycol-bis(p-aminoethylether)-N,N-tetraacetic acid (EGTA) and
tripolyphosphate which have half-wave potentials very close to Pb-NTA complex.
Afghan 2 ’ also reported an automated method using dc polarography for the
determination of NTA in waters, sewage, and detergents. This method is based
on the formations of Bi-NTA complex at pH 2 and has a sample throughput of
15 samples/hr. A detection limit of 10 pg/liter and a coefficient of varia-
tion at 100 pg/liter of 1.3% was reported. Interference from other compounds
including amino acids, K 2 HPO 4 , glucose, nutrient broth, peptone, urea, etc.
was investigated.
24
-------
Another differential polarographic method was reported by Haberman 22 using
an In(III)-NTA complex. The technique was evaluated with river water and sewage
samples containing NTA at levels between 26 pg/liter and 2.6 mg/liter. 14 C-NTA
was added to samples to correct for incomplete recovery. No interference was
found in distilled water with a tenfold molar ratio of EIJTA, citrate, maleate,
phosphate, carbonate or sulfate to 2.6 mg/liter Na 3 NTA. One disadvantage of
this technique was the high concentration of indium required for complete NTA
recovery. At a mole ratio of 5:1 of In-NTA the polarographic peak of In-NTA
complex is not well resolved from the indium reduction peak.
Taylor et al. 16 evaluated palarography using Pb-NTA, Cd-NTA, and Bi-NTA
complexes to determine which gave the best defined polarographic peaks. Pb-NTA
complex was investigated in three buffer systems: tris-hydrochloride-ammonium
hydroxide, glycine-ammonium hydroxide, and ammonium chioride-ammonium hydroxide,
at pH 7-8. Cd-NTA complex was measured in ammonium chloride-arnmoniuni hydroxide
buffer at pH 9. Bi-NTA complex was investigated in NaC1 media at pH 2. Polaro-
graphic determination of NTA as the Bi-NTA complex was found to be the most
sensitive (50 pg/liter detection limit).
1ore modern polarographic techniques such as differential pulse polaro-
graphy have been evaluated by Hoover 2 and Stolzberg. 24 Differential pulse
polarography is more sensitive than dc polarography and can be used to measure
small currents at a potential cathodic of an electrochemically active species
present at much greater concentration. Both Bi-NTA complex 23 and Cd-NTA com-
plex 24 were investigated. Relative standard deviations of < 5% were reported
by Hoover 23 at 1 mg/liter levels of NTA and calibrations were linear from
10 pg/liter to 4 mg/liter. An acetate electrolyte was much better than chlo-
ride electrolyte for Bi-NTA complex in sewage samples.
Colorimetric methods for NTA analysis are based on the measurement of
the absorbances of the colored metal chelates that are formed with excess
metal ions. For example, a method by Tabatai 25 measures the blue color ob-
tained by treating the sample with borate buffer (pH 9.2), zinc sulfate and
2-carboxy-2’ -hydroxy-5’-sulfoformazylbenzene (Zincon). Vanwelssenaer 26 mea-
sured the absorbance at 600 nm of the blue-green color of the Ni-NTA chelate,
and Swisher 27 developed a method that utilizes the formation of a stable 1:1
complex of NTA with ferric ions and is measured at 502 run. Although rapid,
calorimetric techniques are not very sensitive (mg/liter sensitivity) and are
subject to interference by other chelating agents. Colored material in samples
can also interfere.
Another instrumental method used is potentiometric titration with iron
(III) chloride, 28 thallium (III) nitrate 2 u and copper (II) nitrate. 16 ’ 29 The
titration curves (mV versus ml titrant) consisted of two equivalence points,
the first belonging to thallium (or copper) consumed by ammonium pyrrolidine-
dithiocarbamate (APDC) and the second end point belonging to NTA or other
ligands in the sample. The lowest concentration for titrants was 10
Since metal cations were present in sewage usually as complexes with NTA, the
sample was first titrated at pH 5.5 to eliminate Ca, 1g, Zn, Ba, followed by
precipitation and filtration of hydroxides at strongly alkaline pH, and addi-
tion of APDC which precipitat d cation in the presence of NTA. The excess
of APDC was titrated with Ti 3 and Cu 2 . The detection limit was 0.1 ppm. 29
25
-------
Although potentiometric titration techniques offer somewhat greater sensitiv-
ities than colorimetric techniques, they are elaborate and time—consuming.
A compleximetric atomic absorption spectrometric (AAS) method based on
the formation of a soluble Pb-NTA chelate that is determined by AAS was re-
ported. 30 Pretreatment of sample was minimal requiring only pH adjustment
and dilution. Matrix effects were counteracted by using the method of stan-
dard additions.
26
-------
SECTION 5
NTA MONITORING DATA
CANADIAN DRINKING WATER STUDY
The program for monitoring NTA in Canadian waters began in 1971 and was
carried out by the Inland Waters Directorate, Fisheries and EnvironmentA in
Canada in cooperation with Procter & Gamble Company of Canada, Limited. The
purpose of this program was fourfold:
* Survey of NTA in the finished drinking water in municipalities across
Canada (four annual samplings).
* Survey of NTA in raw water above and below 13 selected Ontario munici-
palities and the municipal tap water (monthly sampling).
Study Hamilton Harbour, the western end of Lake Ontario, rivers in
the metropolitan Montreal area, and several selected groundwaters.
Survey NTA in the international section of St. Lawrence River.
The survey of NTA in finished drinking water program was undertaken by
Canada Centre for Inland Waters (CCIW); the program to survey NTA in 13
Ontario municipalities was undertaken by Procter & Gamble Company in coopera-
tion with the Ontario Ministry of the Environment and CCIW. The other two
programs involved sampling over shorter periods of time than those covered by
the monitoring program. No detailed sampling rationale or design was given.
However, some of the special study sites were purposely selected on the basis
of prior knowledge or prediction of the NTA discharge into the water system.
Two analytical methods were employed. CCIW used the polarographic
method by Afghan 2 ° (detection limit 10 p8/liter) and Procter and Gamble used
the gas chromatographic method by Warreni 3 (detection limit 1 pg/liter). In
both laboratories, the samples were preserved with formaldehyde. No other
analytical details were given.
Table 10 summarizes the data for NTA showing the number of samples exa-
mined and the number of samples in which NTA was present at or above 10 pg/
liter (method detection limit). It can be noticed that the largest number
of samples were from Ontario and Quebec. The summaries of the Ontario and
Quebec samples are given in Tables 11 and 12, respectively.
27
-------
TABLE 10. NATIONAL MONITORING PROGRAM SMPLESa,b
Number of samples examined Number of NTA positive
samples
Province 1972 1973 1974 1975 1972 1973 1974
1975
British Columbia 3 78 9 36 0 0 0
1
Alberta 6 7 6 7 0 1 1
0
Saskatchewan 6 8 5 7 0 2 0
2
Manitoba 6 8 7 7 0 1 0
Ontario 76 76 73 60 3 1 3
0
4
Quebec 60 58 54 57 0 2 11
16
New Brunswick 5 10 0 5 0 0 0
0
Nova Scotia 6 15 0 9 0 0 0
0
Newfoundland 0 4 5 0 0 0 0
0
168 264 159 188 3 7 15
23
a Polarographic method by Afghan 2 ° and gas chromatographic method
by Warren 13 were used to analyze samples.
b Data taken from Reference 3.
TABLE 11. SUMMARY OF ONTARIO SAMpLESa,b
Great Lakes Number of samples examined Number of NTA positive
samples
system 1972 1973 1974 1975 1972 1973 1974
1975
Lake Superior to
Detroit River 7 7 7 6 1 0 0
0
Lake Erie 4 4 6 4 0 0 0
0
Welland Canal 6 6 6 5 0 0 1
0
Niagara River 2 2 2 2 0 0 0
0
Lake Ontario 21 23 22 13 0 0 1
2
St. Lawrence River 2 2 2 2 0 0 0
0
Ottawa River 6 5 3 3 0 0 0
0
Small lakes 7 5 5 5 0 0 0
0
Small rivers 8 8 8 8 2 1 1
1
Groundwaters 13 14 12 12 0 0 0
1
7 I
a Gas chromatographic method by Warren - was used to analyze samples.
b Data taken from Reference 3.
28
-------
TABLE 12. SUMMARY OF QUEBEC SAMPLES3'b
Source
Number of samples examined
1972 1973 1974 1975
Number of NTA positive samples
1972 1973 1974 1975
St. Lawrence 4444
River*
Ottawa River- 2222
Montreal area 15 14 13 13
Richelieu River 5445
Yamaska River 3333
Riviere 2101
1'Assomption
0
0
0
0
0
0
1
0
1
0
0
8
0
2
0
0
6
1
2
0
Small rivers
Small lakes
Groundwater
13
14
2
60
11
17
2
58
11
15
2
54
13
14
2
57
0
0
0
0
0
0
0
2
0
1
0
11
4
3
0
16
"' Excluding Montreal area.
a Polarographic method by Afghan20 and gas chromatographic method by Warren13
were used to analyze samples.
b Data taken from Reference 3.
29
-------
Seventy-eight samples of municipal drinking water samples collected from
groundwater from four Canadian provinces during the 4 years of the national
monitoring program were also monitored for NTA. Results given in Table 13
indicate that deeper aquifers are not likely to contain significant amounts
of NTA. However, when municipal groundwaters were analyzed for NTA (see data
in Table 14) 90% of the samples were found to contain 5 pg/liter or less and
the median value was 0.5 pg/liter.
The other studies (e.g. , Hamilton Harbour, Lake Ontario, St. Lawrence
River, etc.) showed much lower levels of NTA due to environmental degradation.
The Hamilton Harbour, for example, can be considered as a completely mixed
reactor with an input of 100 to 125 pg/liter, a retention time of 15 months,
and an effluent with an average concentration of 15 pg/liter. Likewise, the
St. Lawrence River, under the extreme conditions of the total NTA usage by
10 million people and allowing for no degradation in the wastewater treatment
or in the river, the NTA concentration would be about 30 pg/liter with a more
realistic figure of 10 pg/liter or less. 3
Despite the considerable amount of data collected under this study, there
was difficulty in interpreting the data because of the difference in values
reported on comparison samples analyzed by the two laboratories. Furthermore,
the polarographic method used by CCI1 was sensitive to only 10 pg/liter and,
therefore, most locations were reported as showing no detectable levels of
NTA.
Malaiyandy et al. 9 has continued the study, sampling 70 municipalities
across Canada. The sampling locations are given in Table 15. Grab samples
were taken at each public water supply system from the raw water influent at
the immediate vicinity of the water treatment plant (sample A), treated water
samples (sample B), treated water samples from stations at 0.5 to 1.0 mile
from the treatment plant (samples C and D). Further consideration was given
to groundwater sources, disinfection methods, and major and minor drainage
basins.
The sampling locations were selected to cover a statistically significant
representation of population density in Canada. Nineteen private homes and
two commercial establishments in the Ottawa-Carleton Region, Boulbourn and
Lanark were also sampled. All samples were collected during winter months
from November 1976 to February 1977. A gas chromatographic method with
detection limit of 0.2 pg/liter was used to analyze these samples.
Data shown in Table i is summarized in Table 16 where the mean values
of NTA from all sources are given. The highest levels of NTA were found in
the St. Lawrence and Great Lakes basins which receive higher levels of NTA
compared to other basins.
A comparison of the levels of NTA found in 10 municipal water surveyed
prior to 1975 and in l976-1977 is shown in Table 17. Examination of the
data indicates that NTA levels have decreased in five municipalities and re-
mained the same in other five. Investigation of the various treatment pro-
cesses at these locations (prechiorination, pre- and postchlorination, and
ozonation) concluded that none of the treatment processes appears to be com-
pletely effective for NTA removal.
30
-------
TABLE 13. GROUNDWATER
Number of sam
Number of sam
pies examined NTA (10 ppb
pies containing
or > 10 ppb)
1974 1975 1972 1973
1974 1975
Province 1972 1973
British Columbia 0 11
1 4 0 0
0 0
Saskatchewan 1 1
0 1 0 0
0 0
Ontario 13 14
12 12 0 0
0 0
Quebec 2 2
16
2 2 0 0
15
0 0
1
a Polarographic method by Afghan
and gas chromatographic method
by Warren 13
were used to analyze samples.
b Data taken from Reference 3.
TABLE 14. NTA IN
MUNICIPAL GROUDWATERSa,b
Number of
Area Population samples
Number of samples in range
Maximum
(ppb Na 3 NTA)
0-1 2-5 6-10 10
Frankford 1,600 29
20 7 2 0
8
Guelph 51,000 29
20 5 2 2
12
Kemptville 2,000 29
13 11 5 0
9
St.ouffville 3,200 29
22 7 0 0
5
Stratford 23,000 29
20 7 2 0
8
Uxbridge 2,300 29
22 3 1 3
36
Number of 174
117 40 12 5
samples
a Poiarographic method by Afghan 20 and gas chromatographic method by Warren 13
were used to analyze samples.
b Data taken from Reference 3.
31
-------
TABLE 15. LEVELS OF NTA FOUND IN TIlE 70 MUNICIPALITIES ACROSS CANADAa
b Population
City Province served A
Levels of NTA, g/
B
C D
Barrie ON 30,756 d + + +
Bathurst NB 16,674 0.7 0.6 0.5 0.5
Belleville ON 35,507 9.2 13.9 7.0 5.6
Brandon MN 36,000 0.8 0.3 0.5 0.3
Brautford ON 70,000 14.7 11.5 16.7 24.5
Calgary AL 457,112 1.0 1.6 • 1.2
Cap-de-Madelaine QU 33,000 + + + +
Cayuga ON 1,070 33.5 30.4 15.4 7.4
Charlottetown PE 19,500 0.5 + 0.3 +
Clarenville NF 1,900 0.4 ÷ + ÷
Dartmouth NS 72,000 0.4 + + +
Dawson Creek BC 18,000 0.3 + + 0.3
Drui mondville QU 44,200 21.3 + 16.9 22.0
Edmonton AL 530,000 1.6 1.7 1.8 2.0
Fredericton NB 44,000 1.2 0.5 0.5 0.5
Granby QU 34,000 18.9 0.3 + +
Grand Falls NF 9,153 + + 0.3 +
Grand Prairie AL 16,618 + ÷ + +
Halifax NS 111,500 0.3 + + +
High Prairie AL 5,000 + 0.7 0.9 0.7
Hudson QU 4,500 ... + + +
Hull-Gatineau QU 100,000 6.1 4.7 3.8 4.1
Kamloops BC 28,000 + ... ÷ +
Kelowna BC 21,000 + + ÷ +
Lachine QU 73,900 27.3 21.8 25.6 23.0
Laval QU 250,000 14.3 12.6 13.0 11.8
Lethbridge AL 45,000 3.4 2.3 4.3 3.3
Levis QU 23,300 14.4 9.6 5.0 7.7
Longueil QU 220,000 3.7 3.3 3.1 3.7
Magog QU 15,155 0.8 + + +
Medicine Hat AL 301,000 2.8 2.8 2.8 3.4
tioncton NB 58,000 0.5 0.4 .. . 0.3
Montreal QU 1,888,000 1.1 0.9 0.7 0.9
Nanaimo BC 38,000 + + ÷ +
Niagara-on-the-lake ON 3,670 1.2 0.9 0.7 0.9
North Vancouver BC 41,000 + 0.4 0.3 +
Ottawa-Carleton ON 442,000 2.5 2.1 2.3 2.5
Penticton BC 50,000 + + + ÷
Pierrefonds QU 70,000 2O•O l2 9 l2•6 l 2 2
Portage la Prairie MN 14,000 162.0 101.3 83.3 124.4
Port Hawkesbury NS 4,000 0.4 + + 0.5
(continued)
32
-------
TABLE 15. (concluded)
a Data taken from Reference 9; gas chromatographic method was used to analyze
samples (method of detection 0.2 pg/liter).
b NP, Newfoundland; NS, Nova Scotia; PE, Prince Edward Island; NB, New Brunswick;
QU, Quebec; ON, Ontario; MN, Manitoba; SK, Saskatchewan; AL, Alberta; BC,
British Columbia.
c Symbols used are: A,
at treatment plant;
treatment plant; D,
ment plant.
d The symbol + represents levels of NTA at 0.2 pg/liter or below.
e . . . , sample unavailable.
f 1.0 pg/liter in repeat samples obtained in November 1977.
g Well water supplies; only one residence showed this level.
.
b Population
City Province served A
C
Levels of NTA, pg/ Z
B
C D
Prince Albert
SK
30,000
5.6
...
7.6
5.6
Prince George
BC
45,000
+
0.3
+
+
Quebec
QU
225,000
3.9
3.5
2.7
0.3
Regina
SK
148,200
+
+
+
+
Rimouski
QU
32,716
0.8
0.4
0.5
0.4
Riviere du Loup
QU
15,000
6.2
...
1.9
...
St. Eustache
QU
5,000
11.7
. . .
11.4
13.3
Saint John
NB
103,000
+
+
+
+
St. Johns
NP
97,550
0.6
0.7
0.4
0.3
Saskatoon
SK
134,000
0.4
+
0.3
0.3
Selkirk
MN
10,000
1-
+
1-
+
Shawinigan
QU
44,000
+
+
+
+
Sherbrooke
QU
92,000
0.8
+
+
+
Smith Falls
ON
11,812
0.8
+
+
+
Sudbury
ON
91,500
0.5
0.4
0.4
0.3
Swan River
MN
4,000
7.9
12.8
8.0
8.1
Swift Current
SK
16,000
0.5
0.6
+
+
Thetford Mines
QU
26,000
0.6
0.6
0.5
0.4
Thunderbay
ON
102,529
1.5
0.5
0.4
0.6
Toronto
ON
2,000,000
1.6
4.8
4.4
4.7
Trois Rivieres
QU
60,080
3.4
2.8
3.1
2.0
Truro
NS
13,268
0.4
0.4
0.3
0.3
Vancouver
BC
1,000,000
+
0.3
+
+
Vermillion
AL
2,969
+
+
+
+
Victoria
BC
75,000
0.3
+
+
+
Windsor
ON
188,900
2.7
1.2
1.5
1.3
Windsor
QU
6,100
1.2
0.8
0.9
0.7
Winnipeg
MN
577,923
+
+
+
+
Yorktown
SK
ON
14,500
21 sites
0.5
+
169 g
+
1-
raw water sample; B, treated water sample collected
C, treated water sample collected 0.5 mile from
treated water sample collected 1.0 mile from treat-
33
-------
TABLE 16. IEAN CONCENTRATIONS OF NTA (in pg/liter) ES IHATED
TO BE DISCHARGED INTO THE 1AJOR DRAINAGE BASINS
?lajor drainage
Raw
1120
Finished
1120
no. of cities
no. of cities
basins
sampled
mean
sampled
mean
Atlantic Sea Board
12
0.5
12
0.3
Exploits
1
0.2
1
0.2
St. Lawrence
21
7.7
22
4.9
Great Lakes
9
7.2
9
5.8
Lake Winnipeg
14
1.6
14
21 a
tlackenzie
3
0.2
3
0.4
Frazer
1
0.2
2
0.2
Columbia
2
0.2
2
0.2
Pacific Sea Board
4
67
0.2
8
4
0.2
82
a Data taken from Reference 9.
TABLE 17. COHPARISON OF THE LEVELS OF NTA (in pg/liter) FOUND
IN TEN MUNICIPAL WATERS SURVEYED IN OR
PRIOR TO 1975 AND IN 1976 _ 77 a
Levels
of
NTA
in
1975
Location
Province
and
before
in
1976-77
Granby
Quebec
80
0.2
Druinmondvil le
Quebec
20
19.5
Shawinigan
Quebec
20
0.2
Sherbrooke
Quebec
10
0.2
Quebec
Quebec
20
1.9
Lachine
Quebec
20
23.5
Pierrefonds
Quebec
10
12.5
Brantford
Ontario
30
17.6
Brandon
tianitoba
82
0.4
Prince Albert
Saskatchewan
10
6.6
a Data taken from Reference 9.
34
-------
The results of these two studies indicate relatively few locations where
NTA levels occur above 10 pg/liter, and it was concluded that the use of NTA
at that rate will not increase the levels of NTA in the major basins.
CANADIAN WASTEWATER STUDY
A long-term study of 13 Ontario cities measured NTA, nine metals and
phosphorus in grab samples from sewage treatment plant influent and effluent
and the receiving streams above and below the sewage outfall. 33 The monthly
sampling from November 1971 to March 1975 spanned the changeover date in early
1973 when the NTA content of Canadian household detergents increased from 6%
to 15%. This increased usage of NTA was reflected in the NTA content of sewage
influent which increased from a median level of 1.3 mg/liter to 3.2 mg/liter.
The sewage effluent also reflected the change in NTA usage, but less markedly.
Following the change, the median level of NTA in the receiving stream below
the sewage outfall was 0.05 mg/liter, with 97% of all samples containing less
than 0.5 mg/liter. The levels were significantly higher (following changeover)
in only 3 of the 13 locations.
OTHER STUDIES: REMOVAL OF NTA BY ACTIVATED SLUDGE
Several field studies to determine the removal of NTA during wastewater
treatment have been reported. 34 36 Controlled dosages of NTA were added
directly to the raw influent wastewater of an activated sludge plant unit,
and NTA concentrations were monitored at various points within the plant and
in the receiving system. Several conclusions were drawn from these studies.
Average removal varied as a function of NTA concentration and type of
treatment, for typical wastewaters being approximately 90% at dosages up to
8 mg/liter and about 75% at 16 pg/liter. Games 12 reported 90% removal of NTA
at an average influent NTA concentration of 110 mg/liter when using activated
sludge treatment and only 60-70% removal when using trickling filters (see
Table 18).
Removal of NTA by primary sedimentation is negligible (8%).36 At pH
values and NTA concentrations typical for wastewater (mg/liter range) adsorp-
tion of NTA on wastewater solids is negligible (< 2%)36 due to its high water
solubility and chelating properties. The 8% apparent removal reported in refer-
ence 36 may have been due to sampling procedure and not to adsorption on solids.
The rate of NTA removal improved over a several month period. 37 This
indicates that the efficiency of the sludge increases with acclimatization.
Temperature appeared to be one of the most important factors in NTA
degradation. 38 Table 19 gives the percent of NTA removal (concentrations
5-20 mg/ liter) at various temperatures using activated sludge.
35
-------
IAIlI.k 18 N’IA l)ATA IRON l.OIJR IREATHLRT PlANTS IN •fIlE NI .TIlERLANflS’ ’
Avea ge NTA Perceiil Pert& iiL
lull jui I Aver. ge Aver4ge % aiiIliieiit AvCC4 8 C °h 1fIueiit (li L ) ______ NTA
p I ault type Soapi e Lump i eiovez y (rduuge) recove ry (lange) iii fluevt effluent removal reniova I
Aiiiersloort ‘Ir ick l ing A II I 95(85—108) 92(86-99) 59 14 75 64
I’iIit i I I 9 5 90(18-98) 105(97-114) 68 17 72 75
Suest Iiaaiui Trickling A 12 6 9i(74— 108) 98(83-108) 61 21 63 76
Filte, II 9 8 75(63-90) 88(83-98) 64 20 61 75
l iiius hutCh Au ivaleil A II 4 84(60- 105) 94(89- 108) 53 7 86 82
Stiilge I i 8 6 94(84-100) 84(76-91) 37 8 78 77
Il,uu il , ’rwu k ArLIVaLeIJ A 10 2 100(89- 107) 95(90—104) Iii 12 89 90
Sludge ,iiul B 8 2 68(46-85) 94(87-100) 110 6 91 86
iii t u.iiy Ti aLiiieiit
the f t . ,ImIIIe5 were live Loui euuLive 24-hr composule sdiflI)!Cs collected in November 1980, jnd the B samples Were uollected
in l)ct ember 1980
Ii Iki Li t akeut I rain Rule reitue 12
-------
TABLE 19. EFFECT OF TEMPERATURE ON REMOVAL
OF NTA U1 DER STEADY CONDITIONSa
% Removal of initial
concentration
Temperature (°C) 20 mg/i 5 mg/2
5
3
3
7.5
66
82
10
94
96
20
>98
>98
a Data taken from Reference 38.
It appears from data in Table 19 that while NTA is almost completely degraded
during favorable conditions (temp. ) 20°C), removal is incomplete during winter
months. Similar temperature dependency of acclimated sewage has been reported
both in a trickling filter and an oxidation pond 38 and also in an activated
sludge treatment plant 4 ° where degradation was reduced significantly in winter.
A more recent study showed that first order NTA degradation occurred over the
temperature range studied, 2 to 24°C. 4 ’ Kinetic data were given.
* Requirements for optimal biodegradation include moderate temperatures
(> 20°C), prior and continued acclimation of sludge to NTA, aerobic conditions,
and relatively low NTA loadings. NTA does not degrade under anaerobic condi-
tions such as in septic tanks. 42 Acclimation of sludge to NTA generally re-
quired several weeks in field syst.ems. 42 Temporary fluctuations or interrup-
tions in NTA loadings produced prolonged reductions in degradation efficiency. 42
Possible pathways for biodegradation of NTA in sewage have been sug-
gested (see Figure 3).4 As can be seen, the first step in the degradation
of NTA may involve a decarboxylation to N-methylnitrilodiacetic acid or de-
acetylation to irninodiacetic acid (IDA). N-Methylnitrilodiacetic acid can
decarboxylate to dimethylglycine or demethylate to IDA. Dimethyiglycine can
also further demethylate to form N-methylglycine. Additionally, IDA can de-
acetylate to glycine. The possibility of accumulation of certain NTA inter-
mediates and their nitrosation to form nitrosamines would have to be considered
especially since nitroso-IDA forms readily at acidic pH. There are no available
data on carcinogenicity testing of nitroso-IDA; nitroso-sarcosine is carcinogenic
to rat. 42
Degradation of NTA was reported to continue in the receiving stream
reaching 60% at 0.5 miles below outfall, and was essentially complete within
2 miles downstream. 36
37
-------
H - N CH2 - COOH
CH 2 - COOH
Imino—Diacef Ic Acid
C
0
0
>‘
a,
U
0
a,
-D
- CH2 - COOH
Glycine
CH2 - COOH
N—CI-12 - COOH
CH2 - COOH
Nitrilotriacetic Acid
C H 2
N—CH2
C H3
- COOH
- COOH
N—Methyl Nitrilodiacetic Acid
C
0
a
>.-
x
0
a
U
a,
-o
- CH2 - COOH
Dimethyl Glycine
C
0
a
>‘
-c
a,
E
a,
H3CN
N-CH2-COOH
H 7
N—Methyl Glycine (sarcosine)
Figure 3 - Possible pathways
acid in sewage.
for biodegradation of nitrilotriacetic
Data taken from Reference 40.
H3CN
H 3 C
HN
H”
38
-------
References
1. Farrow, R. N. P., and A. G. Hill, Analyst , 90, 210 (1965).
2. ileiners, A. F., T. W. Lapp, C. E. Mumma, E. G. Podrebarac, R. P. Reidsdorf,
H. R. Owens, and H. Gadberry, “Potential Worker and Consumer Exposures to
Nitrilotriacetic Acid (NTA) in Detergents, EPA 560/11-79-008, June 1979,
45 pp..
3. Natheson, D. H., “Nitrilotriacetic Acid (NTA) in the Canadian Environment,”
Sci. Ser. - Can., Inland Waters Dir. , 74 (1977), 46 pp.
4. Rudling, Lars, “Simultaneous Determination of Nitrilotriacetic Acid,
Ethylenediaminetetraacetic Acid, and Diethylenetriaminepentaacetic
Acid as Their Methyl Ester Derivatives by GLC (gas-liquid chrornatog-
raphy),” Water Res. , 6(7), 871—6 (1972).
5. Murray, D., and D. Povoledo, “Determination of Nitrilotriacetic Acid in
Inland Waters by Gas Chromatography of the Trimethyl Ester,” J. Fish.
Res. Bd. Can. , 28(7), 1043-7 (1971).
6. Reichert, J. K. and A. H. H. Linckens, “A Simple and Sensitive Detection
Method for NTA in Drinking Water and River Water,” Environ. Technol. Lett. ,
1(1), 42—9 (1980).
7. Chau, Y. K., and N. E. Fox, “A GC Method for the Determination of Nitrilo-
triacetic Acid in Lake Water,” J. Chroinatogr. Sci. , 9(5), 271-5 (1971).
8. Aue, W. A., C. R. Hastings, K. 0. Gerhardt, J. 0. Pierce, II, H. H. Hill,
and R. F. Moseman, “The Determination of Part-per-billion Levels of
Citric and Nitrilotriacetic Acids in Tap Water and Sewage Effluents,”
J. Chromatogr. , 72(2), 259-67 (1972).
9. Malaiyandi, H., D. T. Williams and R. O’Grady, “A National Survey of
Nitrilotriacetic Acid in Canadian Drinking Water,” Environ. Sci. Technol. ,
13(1), 59-62 (1979).
10. Williams, D. T., F. Benoit, K. Huzika, and R. O’Grady, “Gas Chromatographic
Determination of Nitrilotriacetic Acid Using a Nitrogen Selective Detector,”
J. Chromatogr. , 136, 423-7 (1977).
11. Larson, R. A., J. C. Weston, and S. H. Howell, “Quantitative Gas Chromato-
graphic Determination of Nitrilotriacetic Acid in the Presence of Other
Carboxylic Acids,” J. Chromatogr. , 111(1), 43—9 (1975).
12. Games, L., and J. A. Staubach, Procter and Gamble Co., “Analysis of
Nitrilotriacetic Acid in Environmental Waters,” submitted for publica-
tion to Environmental Science and Technology.
13. Warren, C. B., and E. J. Malec, “Quantitative Determination of Nitrilo—
triacetic Acid and Related Aminopolycarboxylic Acids in Inland Waters.
Analysis by Gas Chromatography,” J. Chromatogr. , 64(2), 219-37 (1972).
39
-------
14. Stolzberg, R. J., and D. N. Huxne, “Determination of Nitrilotriacetate in
Environmental Water by Gas Chromatography of the Trimethylsilyl Ester,”
Anal. Chem. , 49(3), 374-8 (1977).
15. Stolzberg, R. J., and D. N. Hume, “Determination of NTA in Environmental
Water by Gas Chromatography,” Anal. Lett. , 6(9), 829-37 (1973).
16. Taylor, J. K., W. L. Zielinski, Jr., E. J. Naienthal, R. A. Durst, and
R. W. Burke, “Development of Method for NTA (nitrilotriacetic acid)
Analysis in Raw Water,” Nati. Bur. Stand. , Washington, D.C., U.S. Nat.
Tech. Inform. Serv., PB Rep., 72 (No. 219035/3), 30 pp.
17. Rudling, L., “Determination of Nitrilotriacetic Acid,” Water Res. , 5(10),
831—8 (1971).
18. Longbottom,J. E., “Determination of Nitrilotriacetic Acid by High-Speed
Ion-Exchange Chromatography,” Anal. Chem. , 44(2), 418-20 (1972).
19. Asplund, J., and E. Wanninen, “Polarographic Determination of Low Concen-
trations of Nitrilotriacetate in Lake Water,” Anal. Lett. , 4(5), 267-
75 (1971).
20. Afghan, B. K., and P. D. Goulden, “Determination of Trace Quantities of
NTA by Differential Cathode-Kay Polarography,” Environ. Sci. and Technol. ,
5(7), 601-6 (1971).
21. Afghan, B. K., P. D. Goulden, and J. F. Ryan, “Automated Method for Deter-
mination of Nitrilotriacetic Acid in Natural Water, Detergents, and
Sewage Samples,” Anal. Chem. , 44(2), 354-9 (1972).
22. Haberman, J. P., “Polarographic Determination of Traces of Nitrilotri-
acetate in Water Samples,” Anal. Chem. , 43(1), 63-7 (1971).
23. Hoover, T. B., “Polarographic Determination of NTA,” Southeast Environ.
Res. Lab., Athens, Ga., U.S. Environ. Prot. Agency, Of f. Res. Dev.,
(Rep.) EPA, 73, (EPA-R2-73-254), 44 pp.
24. Stolzberg, R. J. , “Determination of Ethylenediaminetetraacetate and
Nitrilotriacetat.e in Phytoplankton Media by Differential Pulse Polaro-
graphy,” Anal. Chim. Acta , 92, 139—48 (1972).
25. Tabatabai, N. A., J. N. Bremner, and R. R. Storrier, “Extraction and
Colorimetric Determination of Nitrilotriacetate (NTA) in Soils,”
Comm. in Soil Sci. and Plant Analysis , 3(1), 57-64 (1972).
26. Vanwelssenaers, N., and G. G. Clinckenmaillie, “The Determination of
Nitrilotriacetic Acid in Granular Detergent Formulations with a Technicon
AutoAnalyzer,” Anal. Chim. Acta , 58, 243—5 (1972).
27. Swisher, R. D., N. N. Crutchfield, and D. W. Caldwell, “Biodegradation
of Nitrilotriacetate in Activated Sludge,” Environ. Sci. and Technol .
1(10), 820—7 (1967).
40
-------
28. Horacek, J., and R. Pribil, “Potentiometric Determination of NTA, EDTA,
and DTPA in Mixtures,” Talanta , 16, 1495-9 (1969).
29. Sekerka, I., J. Lechner, and B. K. Afghan, “Potentiometric Determination
of Nitrilotriacetate in Water and Sewage,” Anal. Lett. , 6(11),
977-83 (1973).
30. Lautenbacher, H. W., and H. W. Baker, “Determination of NTA in Aquatic
Toxicity Studies: A Compleximetric Method Using Atomic Absorption
Spectroscopy,” Bull. Environ. Contam. Toxicol. , 11(1), 57 (1974).
31. Davies, C., R. D. Hartley, and G. J. Lawson, “Chromatographic Behavior
of Organic Acids on Dowex 1-X10,” J. Chromatogr. , 18, 47—63 (1965).
32. Daniel, R. L., and R. B. LeBlanc, “Polarographic Determination of
Nitrilotriacetic Acid in (Ethylenedinitrilo)tetraacetic Acid,”
Anal. Chem. , 31, 1221-3 (1959).
33. Woodiwiss, C. R., R. D. Walker, and F. A. Brownridge, “Concentration of
Nitrilotriacetate and Certain Metals in Canadian Wastewaters and
Streams: 1971—1975,” Water Res . , 13, 599—612 (1979).
34. Rudd, J. W. M., B. E. Townsend, and R. D. Hamilton, “Discharge of NTA
From Two Sewage Treatment Facilities Located in a tlidcontinental
Climate,” J. Fish. Res. Board Can. , 30(7), 1026-30 (1973).
35. Rudd, J. W. H., and R. D. Hamilton, “Biodegradation of Trisodiurn Nitrilo-
triacetate in a Model Aerated Sewage Lagoon,” J. Fish. Res. Board Can. ,
29(6), 1203-8 (1972).
36. Shwuate, K. S., J. E. Thompson, J. D. Brookhart, and C. L. Dean, “NTA
Removal by Activated Sludge-Field Study,” J. Water Pollut. Contr. Fed. ,
42(4), 631—40 (1970).
37. Renn, C. E., “Biodegradation of NTA Detergents n a Wastewater Treatment
System,” J. Water Pollut. Control Fed. , 46(10), 2363-71 (1974).
38. Eden, G. E., G. E. Culley, and R. C. Rootham, “Effect of Temperature on
the Removal of NTA (nitrilotriacetic acid) During Sewage Treatment,”
Water Res. , 6, 877-83 (1972).
39. Bouveng, H. 0., G. Davisson, and E. H. Steinberg, “NTA in Sewage Treatment,”
Vatten , 24, 348-359 (1968).
40. Colgate—Palmolive, unpublished data (1970).
41. Larson, R. J., G. G. Clinckermaillie, and L. Van Belle, “Effect of
Temperature and Dissolved Oxygen on Biodegradation of Nitrilotriacetate,”
Water Res. , 15(5), 615—20 (1981).
42. Epstein, S. S., “Toxicological and Environmental Implications of the Use
of Nitrilotriacetic Acid as a Detergent Builder. 1,” mt. J. Environ.
Stud. , 2(4), 291-300 (1972).
41
------- |