See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/221951693

Modification of iodometric determination of total and reactive sulfide

in environmental samples

Article in Talanta · February 1999

DOI: 10.1016/S0039-9140(98)00253-7

CITATIONS READS

106 7,009

2 authors, including:

Zenon - Pawlak
Wyzsza Szkola Gospodarki
143 PUBLICATIONS 1,869 CITATIONS

SEE PROFILE

All content following this page was uploaded by Zenon - Pawlak on 09 December 2019.

The user has requested enhancement of the downloaded file.

 Talanta 48 (1999) 347 – 353

ISSN 0039-9140
Chem. Abstr. 130:176841
109

Modification of iodometric determination of total and reactive

sulfide in environmental samples
Zenon Pawlak a,*, Agnieszka S. Pawlak b
a
State Health Laboratories, Utah Department of Health, 46 North Medical Dri6e, Salt Lake City, UT 84113, USA
b
Department of Surgery, Uni6ersity of Texas Health Science Center, 6431 Fannin, Suite 4.268, Houston, TX 77030, USA

Received 9 December 1997; received in revised form 15 July 1998; accepted 21 July 1998

Abstract

In iodometric determination of sulfide two reactions are taking place when alkaline solution is added to HCl
acid–iodine. The main oxidation reaction (1), H2S +I2 = 2HI+S; and side reaction of sulfide (2), S − 2 + 4I2 +
8OH − =SO24 − +8I − +4H2O. Preference of reaction (2) over (1) is dependent on pH increasing to \ 7. When
sulfide solution of pH 9 was mixed with HCl acid – iodine, the recovery exceeded 120%, but the recovery of a solution
with a pH of 13 exceeded 200%. To eliminate the side reaction in iodometric titration, the sulfide solution must be
acidic when it is mixed with HCl–iodine. To avoid the side reaction (2), the pH of sulfide solutions were adjusted with
acetic acid to pH 5.5, mixed with HCl–iodine solution and then titrated with standard thiosulfate with precision and
accuracy B 93%. © 1999 Elsevier Science B.V. All rights reserved.

Keywords: Reactive sulfide; Sulfide side reaction error; pH of mixing sulfide – iodine

1. Introduction N). For reactive sulfide extraction processes

Sulfide concentrations in water and waste water
solutions are analytically determined as total
sulfide or as reactive sulfide concentrations [1–6].
The total sulfide measurements may be performed
on samples prepared by precipitating sulfide in
forms of ZnS or CdS [7]. The reactive sulfide
analysis is based on the measurement of the
amount of hydrogen sulfide released upon contact
with aqueous acid (pH 2) in a closed system and
then captured in weak alkaline NaOH ( : 0.25
* Corresponding author. Tel.: +1 801 584 8400; fax: +1
801 584 8486; e-mail: ZPawlak@doh.state.UT.US
NaOH solutions will absorb hydrogen sulfide, but
in the presence of OH − ions will cause a diver-
gence in iodometric sulfide determinations [8]. The
differences between total sulfide and reactive
sulfide is in method of extraction, not in method
of sulfide analysis.
Alternative absorption reagents for hydrogen
sulfide have been studied, such as: a solution of
diluted 0.015 N sodium hydroxide +30% ethanol
[9]; and 0.1 N sodium hydroxide + 1% tri-
ethanolamine + 0.1 M EDTA [10].
Useful quantitative tests for measuring sulfide
include titrometric, colorimetric, potentiometric,
coulometric and ion-selective electrode methods

0039-9140/99/$ - see front matter © 1999 Elsevier Science B.V. All rights reserved.
PII S0039-9140(98)00253-7
348 Z. Pawlak, A.S. Pawlak / Talanta 48 (1999) 347–353
[3,11,16]. The most common titrometric method is
the iodine oxidation technique, which is desig-
nated for hazardous samples containing more
than 1 mg l − 1 of sulfide. The methylene blue
method of colorimetric determination is based on
the reaction of sulfide, ferric chloride, and
dimethyl-p-phenylenediamine; which results in the
formation of methylene blue. The methylene blue
method is used for samples of low sulfide concen-
trations between 0.1 and 20 mg l − 1. The potentio-
metric method, utilizing silver/silver sulfide
electrode in titration of dissolved sulfide, is appli-
cable for contamination \ 0.03 mg l − 1.
Many contradictions appear in the literature
regarding the accuracy of the iodometric determi-
nation of hydrogen sulfide. The iodometric
method of determination of hydrogen sulfide and
sulfides utilizes the following reaction (reaction 1):
H2S +J2 =2HJ+ S (1)
which takes place in an acidic medium.
According to the procedure [13], for reasonably
satisfactory results, the sulfide solution must be
diluted with DI water to a low concentration of
about 0.01 mol l − 1. The sulfide solution should
be added to the excess of acidified iodine solution,
not conversely.
Hence, if alkaline solutions (pH 9 – 13) contain-
ing sulfide are added to HCl acid – iodine, there is
a very high risk that part of the sulfide will be
oxidized to sulfate (reaction 2) [6,8,12]:
S − 2 +4I2 + 8OH − =SO24 − +8I − +4H2O (2)
The purpose of our study was to evaluate the
pH resulting from the mixing of sulfide solutions
in pH range 9 – 13 with HCl acid – iodine solu-
tions, and the impact of various conditions on the
accuracy of the results of iodometric determina-
tion of sulfide. We are going to prove in our
experiment that by adjusting the pH of alkaline
sulfide solutions with acetic acid, an accuracy of
9 3% is achieved. To obtain quality data, the
solution of sulfide must be brought to an acidic
range of pH 5 – 6, mixed with HCl acid–iodine
solution, then the excess of iodine must be back
titrated with thiosulfate until the blue, starch-io-
dide color disappears.
2. Materials and methods
2.1. Stock solution
Sulfide reference solutions were prepared from
Na2S · 9H2O in 1l DI (deionized) water, which
were standardized using 0.025 N iodine solution
and 0.025 N sodium thiosulfate solutions. Sulfide
reference solutions contained 6.9 and 16.1 mmol
l − 1. These standards are unstable and should be
prepared weekly, and refrigerated when not used.
2.2. Working solution
The amounts of 10, 20, 40 and 80 ml of refer-
ence sodium sulfide solution were diluted to 1000
ml volume in DI water or 0.25 N sodium hydrox-
ide solution. A total of 100 ml of the working
solution was titrated with thiosulfate using au-
thor’s procedure, standard procedure, non-stan-
dard procedure or EPA 9030 method.
2.3. Methods of titration
2.3.1. Author’s procedure
A total of 100 ml of sulfide (working solution)
in 0.25 N NaOH solution (pH 13) was added to a
flask containing 1.5 ml of glacial acetic acid and
15 ml of DI water. Case study A: The buffered
sulfide solution of pH 5.5 was added to the mix-
ture of 8 ml 6 N HCl +excess of 0.025 N iodine
+ 15 ml of DI water and then titrated with thio-
sulfate with starch acting as the indicator.
2.3.2. Standard procedure
Case study C: 100 ml of sodium sulfide (work-
ing solution) in 0.25 N NaOH solution (pH 13)
was added to the mixture of 8 ml 6 N HCl, 15 ml
of DI water, and excess of 0.025 N iodine solu-
tion, and then titrated with thiosulfate solution,
using starch as the indicator. Case study B: Sam-
ples of sodium sulfide dissolved in DI water,
resulting in pH 9.2–10.5, were analyzed.
2.3.3. Non-standard procedure
Case study D: 100 ml of sodium sulfide (work-
ing solution) in 0.25 N NaOH solution (pH 13)
was added to the mixture of excess 0.025 N iodine
 Z. Pawlak, A.S. Pawlak / Talanta 48 (1999) 347–353
solution and 15 ml of DI water, acidified with 8
ml 6 N HCl and then titrated with thiosulfate
solution, using starch as indicator.
2.3.4. EPA Method 9030 (sulfide determination)
and 7.3.4 interim guidance for reacti6e sulfide
(extraction procedure)
‘Test method to determine hydrogen sulfide re-
leased from wastes’, SW-846, chapter 7, section
7.3.4.2, provides a procedure to release hydrogen
sulfide upon contact with an aqueous acid. The
trapping solution of 0.25 N NaOH solution is
quantified for content of sulfide by method 9030.
For reactive sulfide concentration, part 7.2.3 in
method 9030 is substituted as follows: ‘The trap-
ping solution must be brought to pH of 2 before
proceeding with iodometric titration’ (see section
7.3.4.2 part 7.7 in test methods SW-846). A 100
ml aliquot of sulfide in 0.25 N NaOH solution
was used for the determination of sulfide by
iodometry.
Case study E (reactive sulfide, SW-846, chapter
7, section 7.3.4.2 extraction, and EPA 9030 deter-
mination procedure): The amount of 1, 2, 4, and
8 ml of sodium sulfide (stock solution) were ex-
tracted and titrated according to the SW-846
chapter 7 and EPA 9030 procedures. A 100 ml of
trapping solution in a 0.25 N NaOH solution was
adjusted to pH 2 with HCl, the sulfide solution
was added to the mixture of excess 0.025 N iodine
solution, 2 ml 6 N HCl and 20 ml DI water, then
titrated with thiosulfate solution. If the sulfide
solution was brought to pH 2 using HCl, part of
the sulfide was relased to the air as hydrogen
sulfide.
Case study F (total sulfide, EPA method 9030):
Sulfide was precipitated as ZnS (using five drops 2
N zinc acetate per 100 ml sample and a few drops
of NaOH solution to produce a pH \10), filtered
out, with precipitate returned to the original bot-
tle, then 100 ml of DI water was added, resulting
in a solution with pH of 10.5. The sulfide solution
was subsequently mixed with HCl acid–iodine
solution, resulting in a solution with a pH 10.5
and titrated with thiosulfate solution (EPA
method 9030).
Sulfide samples were investigated at different
pH conditions for six cases; A, B, C, D, E and F
349
(Table 1). Calculations of sulfide concentration
were made: 1 ml of 0.0250 N iodine reacts with
0.4 mg sulfide. The results for the six cases studied
are shown in Table 1 and Fig. 1.
3. Results and discussion
One of the critical points of iodometric titration
of sulfide is ignoring the pH value of sulfide
solutions in the mixing process with the HCl
acid–iodine mixture. Most methods found in the
literature [1,6,8,13] recommend that the sulfide
solution, which has pH over 7, be added to an
acidified iodine solution, and not the reverse.
However, some current standard iodometric
methods [2,3,5] for titration of precipitated ZnS
ignore this principle and recommended adding
iodine first, and acidifying next. An error may
occur due to the side oxidation of sulfide to
sulfate, because the sulfide solutions usually are
weakly alkaline. Mixing an alkaline sulfide solu-
tion with an HCl acid–iodine solution causes the
side reaction (2) to take place to a considerable
extent. Analyzing sulfide solutions with pH of
about 9 and above during the mixing procedure,
serves as the basis for these experiments. Also, the
EPA 9030 method describing the use of HCl acid
to bring the alkaline sulfide solution (pH of 13) to
a pH of 2 must be changed, because of the
hydrogen sulfide that is released.
Table 1 shows the results of titration of sulfide
solutions using author’s method, standard proce-
dure (method approved), non-standard procedure,
and the EPA 9030 method of iodometric titration.
The content of sulfide was measured iodometri-
cally for samples of sulfide in solutions of pH’s
9.2–10.5, pH 13, and also precipitated as total
sulfide in form of ZnS.
In Case study A it was found, that decreasing
the pH of the sulfide solution from 13 to 5.5
before mixing with HCl acid–iodine, can improve
sulfide analysis. Neutralization of hydroxide ions
in sulfide solution by acetic acid showed no re-
lease of hydrogen sulfide, but treatment with hy-
drochloric acid of pH 2 (EPA method 9030),
causes a release of hydrogen sulfide. An important
step in eliminating the error resulting from side
Table 1
350

Iodometric determination of sulfide in alkaline samples

Sample status (titration procedure) Set c 1 Set c 2

Case study Standard solu- Recoverya Standard solu- Recoverya

tion (mmol l−1) tion (mmol l−1)

(mmol l−1) (%R) (mmol l−1) (%R)

(A) Standard solution in 0.25 N sodium hydroxide, treated with acetic acid, pH 6.9 7.0 101 16.1 16.3 101
5.5 (this procedure)
13.8 13.7 99 32.3 32.3 100
27.7 27.4 99 64.5 63.9 99
55.4 55.4 100 129.0 129.0 100
Average (%D)b 92.5 92.3
(B) In DI water, pH 9.2–10.5 (standard procedure) 6.9 8.3 120 16.1 18.4 114
13.8 15.7 114 32.3 36.5 113
27.7 32.4 117 64.5 68.5 106
55.4 58.2 105 129.0 131.6 102
Average (%D) 96.8 93.2
(C) In 0.25 N sodium hydroxide solution, pH 13 (standard procedure) 6.9 16.1 234 16.1 32.4 201
13.8 29.0 210 32.3 56.8 176
27.7 49.9 180 64.5 93.5 145
55.4 87.0 157 129.0 160.0 124
Average (%D) 98.4 94.8
(D) In 0.25 N sodium hydroxide solution, pH 13 (non-standard procedure) 6.9 18.3 265 16.1 39.0 242
13.8 34.5 250 32.3 71.1 220
27.7 63.7 230 64.5 118.7 184
55.4 108.0 195 129.0 188.3 146
Average (%D) 95.3 94.3
Z. Pawlak, A.S. Pawlak / Talanta 48 (1999) 347–353

(E) Reactive sulfide in 0.25 N NaOH, treated with HCl, pH 2. (EPA 9030 6.9 6.0 87 16.1 13.8 86
procedure)
13.8 11.9 86 32.3 27.1 84
27.7 23.6 85 64.5 54.2 84
55.4 47.6 86 129.0 109.7 85
Average (%D) 93.5 92.2
(F) Total sulfide precipitated ZnS, solution pH 10.5 (EPA 9030 procedure) 6.9 7.6 110 16.1 17.4 108
13.8 14.8 107 32.3 34.2 106
27.7 28.8 104 64.5 66.4 103
55.4 55.4 100 129.0 127.7 99
Average (%D) 92.6 92.2

a
Recovery or accuracy (%R) is expressed as a ratio: B/A; C/A; D/A; E/A and F/A×100%.
b
Precision (%D) is the ‘relative percent difference’ between duplicate test results, (%D) =(difference/average value)×100%, the mean of 2–4 duplicates.
 Z. Pawlak, A.S. Pawlak / Talanta 48 (1999) 347–353 351

Fig. 1. Influence of pH on sulfide solutions in mixing process with HCl acid – iodine in iodometric titration with thiosulfate: Reactive
sulfide solution (pH 13) treated with HCl to pH 2, followed by HCl acid – iodine (curve E); sulfide solution (pH 13) treated with
acetic acid to pH 5.5 and than HCl acid–iodine (curve A); total sulfide precipitated as ZnS, treated with HCl acid – iodine (curve
F); sulfide solution (pH 9.2–10.5) treated with HCl acid – iodine solution (curve B); sulfide solution (pH 13) treated with HCl
acid – iodine (curve C); sulfide solution (pH 13) treated with iodine and then HCl solution (curve D).

reaction (2) is to change the pH from pH 13 to 5.5
by using acetic acid (as shown in this work). The
obtained recoveries of sulfide were in range 99–
101% and precision B 9 3%. The above experi-
ments have confirmed that slight modifications in
the methods may produce great differences in the
results.
The standard solution of sodium sulfide in DI
water resulted in pH 9.2 – 10.5 in the measured
range of concentrations. In Case study B (Table 1),
mixing a weakly alkaline sulfide with HCl acid–
iodine solution using standard titration procedure
(Std-Titr-Proc) resulted in side reaction (2). How-
ever, the 120% recovery of this technique is not
quite sufficient for ordinary analytical work. Ex-
periments have shown that the precision were from
9 3.2 to 9 6.8% and recoveries from 120 to 102%.
In Case study C, sulfide in a high alkaline
solution (pH 13) was added to an excess of HCl
acid–iodine solution and back titrated with thio-
sulfate (Std-Titr-Proc). In the mixing process,
sulfide was oxidized by reaction (1) and partially
by reaction (2). It was found that the recoveries
were in range of 234–124% for low and high
standard sulfide solutions, respectively. The results
have poor precision, 9 8.4% in the lower range of
the sulfide concentration. During the mixing pro-
cess, the pH value of the solution changed slowly
and unevenly, resulting in partial oxidation of
sulfide to sulfate.
352 Z. Pawlak, A.S. Pawlak / Talanta 48 (1999) 347–353
In Case study D, a high alkaline sulfide solution
(pH 13) was added to an excess of non-acidified
iodine solution, then 8 ml of 6 N HCl was added
(Non-Std-Titr-Proc). The observed recoveries of
sulfide were in the range of 265 – 146% for low
and high standard solutions, respectively These
results can be explained by the occurrence of
reaction (2). Comparison of reaction (2) with
reaction (1) shows that 1 mol of sulfide in alkaline
solution uses 4 mol of iodine, compared to 1 mol
in the acid solution. The reaction between sulfide
and iodine in an alkaline solution is a kinetic
process and is dependant on the concentrations of
hydroxide ions and the concentration of oxidizer
[14]. Considering Case study D, where sulfide in
0.25 N NaOH solution was mixed with excess of
iodine solution first and then acidified, we found
reaction (2) occurred to a higher extent for lower
concentrations of sulfide in the sample. Reaction
(2) is quite extensive in alkaline solutions with
much stronger oxidants than iodine [8]. However,
reaction (2) should be avoided and eliminated in
sulfide determination. The way to eliminate the
side reaction is to neutralize the hydroxide ions by
bringing the sulfide solution into an acidic range
with acetic acid and then mixing with HCl acid–
iodine solution.
Results for the reactive sulfide test are shown
for Case study E in Table 1. 100 ml samples were
acidified with HCl to pH 2, mixed with HCl
acid–iodine and titrated with thiosulfate. Our re-
coveries of sulfide were 84 – 87% and passed 50%
of the EPA method requirements. It may be con-
cluded that strong hydrochloric acid causes the
release of hydrogen sulfide, therefore resulting in
lower values.
Results of determination of total sulfide are
shown for Case study F in Table 1. The recoveries
were in the range of 110 – 99%, higher recoveries
in low ranges of concentrations are related to the
partial oxidation of sulfide to sulfate.
In Fig. 1 the percentage recovery of sulfide is
plotted against standard concentrations of sulfide
for six cases. The graph shows iodometric results
of: Reactive sulfide solution (pH 13) treated with
HCl to pH 2 and than HCl acid – iodine (curve E);
sulfide solution (pH 13) treated with acetic acid to
pH 5.5 and than HCl acid – iodine (curve A); total
sulfide precipitated as ZnS, treated with HCl acid
–iodine (curve F); sulfide solution (pH 9.2–10.5)
treated with HCl acid–iodine solution (curve B);
sulfide solution (pH 13) treated with HCl acid–
iodine (curve C); sulfide solution (pH 13) treated
with iodine and then HCl solution (curve D).
Fig. 1 illustrates the dependence of recovery of
the sulfide reaction over the concentration range
6.9–129 mmol l − 1. In the studied concentration
range (Case study C and D) sulfides showed high
recoveries in the beginning, then decreased in the
range of interest. As ratio [sulfide]/[iodine] in-
creases with sulfide concentration, the sulfate for-
mation (side reaction 2), decreases. Only under
similar conditions, with pH constant, and [io-
dide] [sulfide] is the recoverable total sulfur in
the form of sulfate [14]. In our experiment, excess
of [iodide] was only 3× higher than [sulfide], but
it should be a 10- to 20-fold excess.
The kinetics of the oxidation of hydrogen
sulfide and hydrosulfide ion to sulfur and sulfate
in aqueous solution by hydrogen peroxide was
investigated more than the same reaction with
iodine [14,15]. Hydrogen peroxide is recom-
mended for the treatment of odors due to the
generation of hydrogen sulfide in municipal
sewage treatment systems, in concrete sewer lines,
and in other anaerobic environments where or-
ganic matter and sulfides are present.
With pH 8 and above, the predominant
product of oxidation is sulfate, and 10-fold excess
of oxidizing agent is required for the reaction to
go to high yield or completion. The results are
summarized in Table 2.
In alkaline solution, in presence of hydrogen
peroxide, under the pseudo first-order rate con-
stant kobsd (rate= kobsd [S2 − ]T, kobsd = kc [H2
Table 2
Sulfate yield for some oxidants in alkaline media
Oxidizing agent E° (V) pH % SO2−
4
Hydrogen peroxide [14] 1.77 8.5 99
Potassium permanganate 0.59 13.5 97
[[8](a)]
Sodium hypochlorite [[8](a)] 0.89 13.5 69
Iodine (this work) 0.54 13 B50
 Z. Pawlak, A.S. Pawlak / Talanta 48 (1999) 347–353
O2][H + ]), sulfate was the only product observed
[14]. The reaction of hydrogen sulfide and hydro-
gen peroxide has been studied kinetically, at pH
8.1, 25°C, m =0.4, the overall rate constant (kc)
for the formation of sulfate, kc =2.61 min − 1. and
only 0.021 at pH 5.05 [14].
4. Conclusions
On the basis of experimental results, the follow-
ing conclusions can be drawn:
1. When alkaline sulfide solution is mixed with
HCl acid – iodine solution, there is a high risk
that besides the main reaction (1), the side
reaction (2) will also influence our results,
giving poor accuracy especially for low sulfide
content.
2. It was found that adjusting the pH of alkaline
sulfide samples to pH 5.5 with acetic acid
before mixing with HCl acid – iodine solution,
gave results with precision and accuracy B 9
3 in sulfide determination (Case study A).
When alkaline sulfide solution was brought to
a pH of 2 with HCl (EPA 9030 method) part
of the sulfide was released as hydrogen sulfide
(Case study F).
3. The very high variation of side reaction (2)
observed in Case study C (where sulfide solu-
tions a pH of 13 were mixed with HCl acid–
iodine to give recoveries of 234–124%),
indicates of importance of pH in iodometric
determination of sulfide.
Acknowledgements
The authors thank Don Gentry for helpful
discussions.
.
View publication stats
353
References
[1] United States Environmental Protection Agency, Test
Methods for Evaluating Solid Waste: Physical and Chem-
ical Methods, 3rd ed., SW – 846, United States Environ-
mental Protection Agency, Washington DC, 1986,
Methods 9030, and chapter 7 section 7.3.4.2.
[2] National Institute for Occupational Safety and Health,
NIOSH Manual and Analytical Methods, National Insti-
tute for Occupational Safety and Health, Cincinnati, OH,
1977, Method No. S.4.
[3] American Public Health Association, Standard Methods
for the Examination of Water and Wastewater, 19th ed.,
American Public Health Association, Washington DC,
1995, pp. 4 – 122, pp. 4 – 131.
[4] M.W. Skougstad, M.J. Fishman, L.C. Friedman, D.E.
Erdmann, S.S. Dunkan, (Eds.), Methods for Determina-
tion of Inorganic Substances in Water and Fluvial Sedi-
ments: Techniques of Water-Resources Investigations of
the United States Geological Survey, Book 5, 1979, pp.
619 – 620.
[5] F.J. Wechler (Ed.), Standard Methods of Chemical Anal-
ysis, 6th ed., vol. 2, part B, Van Norstrand, 1962, pp.
2484 – 2488.
[6] I.M. Kolthoff, B. Belcher,V.A. Stenger, G. Matsuyama,
in: Volumetric Analysis, vol. 3, Interscience Publishers,
1957, pp. 291 – 292.
[7] W.L. Bamesberger, D.F. Adams, Environ. Sci. Technol.
47 (1969) 258 – 261.
[8] O. Bethage, Anal. Chimica Acta (a) 9 (1953) 129–139;
ibid., (b) 10 (1954) 113 – 116.
[9] L.P. Van DenBerge, A. Devreese, M. Vanhoorne, Am.
Ind. Hyg. Assoc. J. 46 (1985) 693 – 695.
[10] J.T. Purdham, L. Yongyi, Am. Ind. Hyg. Assoc. J. 51
(1990) 269 – 272.
[11] I.M. Kolthoff, P.J. Elving, E.B. Sandell, in: Treatise on
Analytical Chemistry, vol. 7, part 2, Interscience Publish-
ers, New York-London, 1961, pp. 75 – 79.
[12] I.M.Z. Kolthoff, Anal. Chem. 60 (1921) 448 – 457.
[13] G.H. Jeffry, J. Baset, J. Mandham, R.C. Denney, in:
Vogel’s Textbook of Quantitative Chemical Analysis, 5th
ed., Longman Scientific and Technical, New York, 1989
pp. 398 – 399.
[14] M.R. Hoffmann, Environ. Sci. Technol. 11 (1977) 61–66.
[15] Y.G. Adewuyl, G.R. Carmichael, Environ. Sci. Technol.
21 (1987) 170 – 177.
[16] D.T. Pierce, M.S. Applebee, C. Lacher, J. Bessie, J.
Environ. Sci. Technol. 32 (1998) 1734 – 1737.
.