The pei was found to decrease the mutual influences of the

The PEI was found to decrease the mutual influences of the

components of the binary Ag(I)-Cu(II) system (Fig. 3, Table 3)

and ternary Pb(II)-Cu(II)-Cd(II) system (Fig. 4, Table 3) at the

electrode surface. Therefore, the anodic peak heights of the

components increased in the presence of PEI. (Fig. 4). The dif-

ferent complex stabilities of Ag(I), Pb(II), Cd(II) and Cu(II)

with PEI allowed Ag(I), Pb(II) or Cd(II) to be determined in

the presence of an excess of Cu(II) (Table 3).

Conclusion

The water-soluble polymers PEI and TU-PEI diminish the in-

teractions of the components of the Cd(II)-Pb(II)-Cu(II) and

Ag(I)-Cu(II) systems at the carbon-paste electrode surface and

reduce the influence of Cu(II) on the anodic peak heights of

Cd(II) and Pb(II) in the ternary system and on the anodic peak

height of Ag(I) in the binary system. It is therefore possible to

determine Ag(I), Pb(II) and Cd(II) in aqueous solutions of PEI

in the presence of larger excesses of Cu(II) than when PEI is

absent from the solution.

References

1. Geckeler K, Lange G, Eberhardt H, Bayer E (1980) Pure

Appl Chem 52 : 1883

2. Shkinev VM, Spivakov BYa, Geckeler K (1989) Talanta 36 :

861

3. Bard AJ, Faulkner LR (1980) Electrochemical methods: Fun-

damentals and applications. Wiley, New York, pp 718

4. Osipova EA, Kamenev AI, Sladkov VE, Shkinev VM (1997)

J Anal Chem 52 : 242

313

Fresenius J Anal Chem (1998) 361 : 313–318 – © Springer-Verlag 1998

Rainer Haas · Torsten C. Schmidt · Klaus Steinbach

·

Eberhard von Löw

Chromatographic determination 

of phenylarsenic compounds

Received: 17 July 1997 / Accepted: 9 October 1997

Abstract Gas chromatographic (GC) and liquid chromato-

graphic methods for the investigation of phenylarsenic com-

pounds are presented. With gas chromatography using an elec-

tron capture detector (ECD), the chemical warfare agents PFIF-

FIKUS, CLARK I and CLARK II can be detected. After de-

rivatization with mercaptans and dimercaptans the sum of

diphenylarsenic compounds resp. phenylarsenic and phenylar-

sonic compounds can be detected as the mercapto resp. dimer-

capto derivatives. High performance liquid chromatography

(HPLC) analysis may be used for the detection of triphenyl-

arsenic compounds and ADAMSITE.

1 Introduction

Phenylarsenic compounds as chemical warfare agents were

produced in large amounts during the world wars I and II. Af-

ter World War II the production sites and filling plants were de-

stroyed and the chemical warfare agents were sunk in the North

Sea and the Baltic Sea or deposited in the production sites and

filling plants [1]. Residues of these chemical warfare agents are

still present and contaminate soil and water.

The most important phenylarsenic compounds, which were

used as chemical warfare agents, were:

– Diphenylarsine chloride (Ph

2

AsCl), called CLARK I

– Diphenylarsine cyanide (Ph

2

AsCN), called CLARK II

– Phenarsazine chloride (Ph

2

As(NH

2

)Cl), called ADAMSITE

– Phenylarsine dichloride (Ph

2

AsCl

2

), called PFIFFIKUS and

– arsine oil, a technical mixture of arsenic(III) chloride, phenyl-

arsine dichloride, diphenylarsine chloride and triphenylar-

sine [2].

CLARK I, CLARK II and ADAMSITE are strong irritants,

which are called “sternutators”. Toxic effects of these com-

pounds occur from concentrations in air of approx. 0.1 mg/m

3

on. PFIFFIKUS is toxic by inhalation (irritant) and by skin

contact [2].

In water, soils and sediments the phenylarsenic compounds

can be metabolized via hydrolysis and oxidation [2, 3]. In Fig.

1 the main reactions of the compounds and their metabolites

are given.

A chromatographic determination of some phenylarsenic

compounds was achieved mainly with hyphenated techniques.

These utilized GC or HPLC and element specific detectors like

photooxidation-hydride-atomic absorption spectrometry (AAS)

[4], inductively coupled plasma (ICP) with atomic emission

spectrometry (AES) [5–8] or, more recently, mass spectrome-

try (MS) [9–11]. For phenylarsonic acid, ion chromatography

[12, 13] and capillary electrophoresis (CE) [14] were described

as additional separation methods. The matrices studied include

river-water [13], sediments [11], oil-shale [7, 15], incineration

effluents [6] and marine organisms [9]. However, very few

chromatographic methods were described for the separation of

the sternutators themselves rather than their possible metabo-

R. Haas (

౧)

Büro für Altlastenerkundung und Umweltforschung, 

Stadtwaldstrasse 45a, D-35037 Marburg, Germany

T. C. Schmidt · E. v. Löw

Environmental Hygiene and Immunology, 

Medical Centre of Hygiene, Philipps-University, 

Pilgrimstein 2, D-35037 Marburg, Germany

K. Steinbach

Department of Chemistry, Philipps-University, 

Hans-Meerwein-Strasse, D-35032 Marburg, Germany

lites. In a thorough review of the chromatographic analysis of

chemical warfare agents [16] of the analytes dealt with here,

only ADAMSITE was mentioned at all. An environmental ap-

plication can be found in [17]. Therein, ADAMSITE was stud-

ied in snow samples together with nerve agents. Other pub-

lished methods were based on TLC [18] or gaschromatography

without [19, 20] or after derivatization [21–23]. Thioglycolic

acid methyl ester (TGM) was used as derivatization agent for

some diphenylarsines and phenylarsines [21, 22]. ADAMSITE

can be detected with GC/ECD after derivatization by bromina-

tion of the aromatic ring [23].

The aim of our study was to develop chromatographic meth-

ods for the separation of the phenylarsenic compounds presented

in Fig. 1. We were especially interested in the possibility of de-

rivatizing phenylarsenic compounds with mercaptans to yield

derivatives, which can be sensitively detected with GC-ECD.

The same reaction was already used vice versa for the separa-

tion of mercaptans after derivatization with phenylarsine oxide

[24].

2 Experimental

2.1 Chemicals

The following chemicals were used: methanol p.a. (Merck,

Darmstadt, Germany); acetone p.a. (Merck); t-butyl methyl

ether p.a., 99.8% (Aldrich, Deisenhofen, Germany); diphenyl-

arsine chloride [CAS No.: 712-48-1] (97%, with 3% impurity

of diphenylarsine cyanide and 0.1% of triphenylarsine);

diphenylarsine cyanide [CAS No.: 23525-22-6] without de-

tectable impurities (GC, HPLC); phenarsazine chloride [CAS

No.: 578-94-9] without detectable impurities (HPLC); phenyl-

arsine dichloride [CAS No.: 696-28-6]; phenylarsine oxide

[CAS No.: 637-03-6] (Aldrich); phenylarsonic acid [CAS No.:

98-05-5] 97% (Aldrich); triphenylarsine [CAS No.: 603-32-7]

>98% (Aldrich); triphenylarsine oxide [CAS No.: 1153-05-5]

97% (Aldrich), thioglycolic acid methyl ester 95% (Aldrich);

thioglycolic acid ethyl ester 97% (Aldrich); 1-ethanethiol 97%

(Aldrich); 1-propanethiol 99% (Aldrich); 1,2-ethanedithiol

90% (Aldrich); 1,3-propanedithiol 99% (Aldrich).

Bis(diphenylarsine)oxide [UPAC: tetraphenyl diarsoxane,

CAS No.: 2215-16-9] and diphenyl arsonic acid [UPAC: di-

phenyl arsinic acid, CAS No.: 4656-80-8] were not available as

reference substances.

Stock solutions of the phenylarsenic compounds in t-butyl

methyl ether were prepared in the concentration range of 0.5

mg/mL to 4.4 mg/mL. Stock solutions of the mercaptans in

acetone were prepared in a concentration of 10.0 mg/mL 

2.2 Derivatization and gas chromatographic equipment 

and conditions

All derivatizations were done at a temperature of 20° C in ace-

tonic solution in 1.2 mL vials mixing 0.5 mL acetone, 20 

µ

L

phenylarsenic compound (stock solution) and 20 

µ

L mercaptan

(stock solution). The reactions are completed in 15 min.

For the separation of the derivatives an HP 5890 series II+

gas chromatograph (Hewlett Packard, Waldbronn, Germany)

with HP 7673 autosampler and ECD were used. The tempera-

tures of the injection block and the detector were 250° C and

300° C, respectively. The injection volume was 1 

µ

L (split in-

jection, split approx. 1 : 10). A (5%-phenyl)-methylpolysilox-

ane column, 30 m, 0.25 mm i.d., 0.25 mm d

f

(DB-5 from J&W,

Köln, Germany) was used. The carrier gas was nitrogen with a

column headpressure of 100 kPa. The temperature of the col-

umn was 230° C. Data acquisition was accomplished with

Gynkosoft, v. 5.32 (Gynkotek, Germering, Germany).

For identification of the derivates a GC/MS-system VG

Trio 2 in the EI mode (70 eV) was used. The injection volume

was 1 

µ

L (splitless injection). A (5%-phenyl)-methylpolysilox-

ane column, 30 m, 0.25 mm i.d., 0.25 mm d

f

(DB-5 from J & W,

Köln, Germany) was used. The carrier gas was helium. The

314

Fig. 1 a–d Scheme of hydrol-

ysis and oxidation reactions

of CLARK I and CLARK II

(a), ADAMSITE (b), 

PFIFFIKUS (c) and triphenyl-

arsine (d)

temperature of the column was started at 40° C and was raised

to 250° C at 10° C/min.

2.3 HPLC equipment and conditions

The HPLC equipment consisted of a gradient pump M-480, an

on-line degasser GT-103, an autosampler GINA 50 and diode

array detector UVD 340-S. Data acquisition was accomplished

with Gynkosoft, v. 5.32 (all Gynkotek, Germering, Germany).

The injection volume was set to 20 

µ

L. For separation of the

organoarsenic compounds two different systems were used.

For the first one an RP-18 column (250 

×

3 mm, NUCLEOSIL

120-5) with the following linear gradient was used: acetoni-

trile/water 50/50 (v/v) to acetonitrile 95/5 (v/v) in 22.5 min,

then reequilibration for 15 min. The second one was carried

out with a cyanopropyl column (250 

×

3 mm, NUCLEOSIL

CN 100-5) with an eluent gradient from acetonitrile/water

10/90 (v/v) to acetonitrile/water 70/30 (v/v) in 30 min and sub-

sequently applying the starting eluent for 15 min.

3 Results and discussion

3.1 GC analysis

From the investigated phenylarsenic compounds only phenyl-

arsine dichloride, diphenylarsine chloride and diphenylarsine

cyanide and, in high concentrations (more than 200 ng/

µ

L),

triphenylarsine can be detected with gas chromatography using

an ECD without derivatization.

Diphenylarsines

All mercaptans and dimercaptans react with diphenylarsines in

forming stable diphenylarsine thioether (Fig. 2). The deriva-

tives were identified using GC/MS and are given in Table 1.

The derivatization reactions are nearly quantitative. The equi-

librium concentrations of the mercapto derivatives are 90–98%,

if the diphenylarsine concentrations are 100 ng/

µ

L or lower.

When derivatizing diphenylarsines with more than one mer-

captan or dimercaptan, the yields of the resulting diphenylar-

sine thioether are proportional to the molar ratio of the used

mercaptans. Figure 3 shows the gas chromatographic separa-

tion of the six diphenylarsine derivatives at 230° C.

Phenylarsines

Phenylarsine oxide and dichloride form phenylarsine thioether

with mercaptans. These products are not stable in acetonic so-

lution. Therefore mercaptans cannot be used for the analytical

detection of phenylarsines.

1,2-Ethanedithiol reacts with phenylarsine oxide and di-

chloride in forming the stable 2-phenyl-<1,3,2>dithiarsolane

315

Table 1 Retention times (t

R

)

and limits of detection (LOD)

of phenylarsine and diphenyl-

arsine derivatives; abbr. see

text

Phenylarsenic

Mercaptan

Derivative

t

R

/min

LOD/ng

compound

for deriv.

PhAs-Cl

2

–

–

2.70

4.0

Ph

2

As-Cl

–

–

4.29

0.6

Ph

2

As-CN

–

–

4.60

0.3

Ph

2

As-X

EtSH

Ph

2

As-SEt

5.85

0.6

Ph

2

As-X

PrSH

Ph

2

As-SPr

6.82

0.6

Ph

2

As-X

TGM

Ph

2

As-SGM

10.89

0.4

Ph

2

As-X

TGE

Ph

2

As-SGE

12.54

0.4

Ph

2

As-X

Et(SH)

2

Ph

2

AsSEtSH

12.29

0.6

Ph

2

As-X

Pr(SH)

2

Ph

2

AsSPrSH

15.65

0.9

PhAsO

Et(SH)

2

PhAsS

2

Et

4.50

0.1

PhAsO

Pr(SH)

2

PhAsS

2

Pr

5.53

0.2

Fig. 2 a–c Scheme of the de-

rivatization reactions; a: reac-

tion of diphenylarsenic com-

pounds with thiols and dithi-

ols; b: reduction of phenylar-

sonic acid by dithiols; c: re-

action of diphenylarsine ox-

ide with dithiols

[CAS No.: 4669-53-8, PhAsS

2

Et], 1,3-propanedithiol reacts

with phenylarsine oxide and dichloride in generating the stable

2-phenyl-<1,3,2>dithiarsinane [CAS No.: 55883-62-0, PhAs-

S

2

Pr]. Both reactions are quantitative. Phenylarsonic acid is re-

duced by 1,2-ethanedithiol and 1,3-propanedithiol to phenylar-

sine oxide. Therefore, the sum of the phenylarsenic compounds

is determined after derivatization. The derivatives were identi-

fied by GC/MS. Since the reaction of phenylarsine oxide with

a 1 : 1 molar mixture of 1,2-ethanedithiol and 1,3-propanedi-

thiol yields 95% PhAsS

2

Et and only 5% PhAsS

2

Pr, a chro-

matogram with both derivatives similar to the one in Fig. 3 can-

not be given. Instead, Fig. 4 shows two overlayed gas chro-

matograms of these derivatives.

In Table 1 the retention times and limits of detection (LOD)

of the investigated stable phenylarsenic and diphenylarsenic

compounds are presented. The LOD were determined with the

3

σ

-method using a consecutive dilution series.

3.2 HPLC analysis

Table 2 gives the retention times and limits of detection for the

HPLC separation for the six investigated phenylarsenic com-

pounds with the two systems used. In Fig. 5 the separation

achieved with the RP-18-system is shown, in Fig. 6 the corre-

sponding result with the CN-column. In both figures separation

of only five compounds is shown. The general elution order on

both columns is, as one might expect, phenylarsenic < diphenyl-

arsenic < triphenylarsenic compounds. Because the polarity of

the investigated compounds differs remarkably, a rather steep

gradient had to be used with both systems. In general, retention

was stronger on the less polar RP-18-column.

CLARK I and CLARK II react with the water in the eluent

by forming diphenylarsine hydroxide (Ph

2

AsOH) and give only

one peak. With both systems, phenylarsonic acid (PhAsO(OH)

2

)

is eluted near to or at the column dead time. For a stronger re-

tention of this compound ion-pairing chromatography has to be

used, but since the influence of an ion-pair reagent on the reac-

tions of the phenylarsenic compounds is not known, this was

not applied. In addition, separation of all six compounds was

not possible with both systems. Phenylarsine oxide (PhAsO)

and triphenylarsine (Ph

3

As) are separated with both systems.

With the RP-18 column is not possible to analyze Ph

3

AsO. The

retention time of the eluted peak differed remarkably from one

chromatographic run to the next and sometimes no peak at all

was detected, even when using measurement standards in high

concentrations. Although there is no rational explanation for

the behavior of Ph

3

AsO at the moment, we tentatively suggest

that irreversible binding to the column might have occured.

With the CN column, Ph

2

AsOH and Ph

2

As(NH

2

)Cl were coe-

luted. All attempts to separate the overlapping peak pairs (us-

ing methanol as organic modifier or decreasing the slope of the

gradient) did not improve the results in both systems, because

peak broadening increased considerably as can already be seen

in Fig. 6 (peak 1). Interactions of the diphenylarsenic compounds

with the cyano group of the stationary phase might be respon-

sible for this behavior. Because of the problems occuring with

both columns, the choice of the column depends on the investi-

gated substances. Interestingly the limits of detection achieved

with the two systems differ remarkably and are always lower

with the first separation system. They were determined with the

3

σ

-method using a consecutive dilution series.

In Fig. 7 the UV spectra of the six compounds are com-

pared. The UV spectra of all compounds except ADAMSITE

316

Fig. 3 Gas chromatogram of six diphenylarsine derivatives by

using 400 ng Ph

2

As-Cl; 1: Ph

2

As-SEt, 2: Ph

2

As-SPr, 3: Ph

2

As-

SGM, 4: Ph

2

AsS

2

Et, 5: Ph

2

As-SGE, 6: Ph

2

AsS

2

Pr

Fig. 4 Overlayed gas chromatograms of PhAsS

2

Et [1] and

PhAsS

2

Pr [2], injected amounts: 40 ng

Table 2 Retention times (t

R

)

and limits of detection (LOD)

for both HPLC systems 

(1: RP-18, 2: CN-column);

abbr. see text

Substance

Peak No

t

R

1/min

LOD1/ng

t

R

2/min

LOD2/ng

Ph

2

AsOH

1

6.0

7.5

13.3

52

Ph

2

As(NH

2

)Cl

2

4.5

8.5

15.2

140

Ph

3

As

3

19.8

6.6

25.9

11.2

Ph

3

AsO

4

not possible

16.5

4.6

PhAsO(OH)

2

5

1.8

34

1.9

n.d.

PhAsO

6

2.9

7.3

4.5

25.4

are similar, but nevertheless they can still be used for peak

identification. The very characteristic spectrum of ADAMSITE

can be explained with the many mesomeric structures of the tri-

cyclic compound. This causes a decrease in the energy gap be-

tween the HOMO and the LUMO, which subsequently leads to

a bathochromic shift of the absorption bands.

3.3 Comparison of GC and HPLC analysis

Advantages of the GC/ECD analysis are 10 to 100 times lower

detection limits, the shorter time needed for an analysis (less

than 20 min compared to at least 35 min including equilibration

time), the greater selectivity due to the derivatization and the

possibility of easy confirming of results with the use of differ-

ent derivatization reagents. With GC analysis the distinction

between CLARK I and CLARK II is possible. After derivatiza-

tion the phenylarsine compounds on the one hand and the

diphenylarsine compounds on the other hand give only two de-

rivatives, hence the sum of these compounds is determined.

Use of HPLC is necessary for the analysis of triphenylarsine

compounds and ADAMSITE, which cannot be detected by GC/

ECD. The distinction between As(III) and As(V) compounds,

triphenylarsine and triphenylarsine oxide resp. phenylarsine

oxide and phenylarsonic acid, is only possible with HPLC.

Therefore the choice of one particular separation method

depends on the analytical problem that has to be solved.

4 Conclusions

For arsenic containing chemical warfare agents and their hy-

drolysis and oxidation products rather simple chromatographic

methods are available, which might be applied to the analysis

of soil and water samples. All stable mercapto and dimercapto

derivatives can be used as reference compounds in the investi-

gation of such samples by GC analysis. Most of these com-

pounds cannot be detected with GC/ECD without derivatiza-

tion. Our future efforts will concentrate on improving the sepa-

ration with HPLC (e.g. temperature gradients, choice of

columns), including a general investigation of the behaviour of

phenylarsenic compounds on reversed-phase chromatographic

columns. Besides, the use of an atomic emission detector in gas

chromatography of the mercapto derivatives will be studied for

comparison and the methods will be applied to the study of en-

vironmental matrices.

References

1. Office of the chief of chemical corps, headquaters European

command (1947) The history of captured enemy toxic mu-

nitions in the American zone. European theater may 1945 to

june 1947

2. Jackson KE (1935) Chem Rev 17 : 251–292

3. Haas R (1996) Umweltmed Forsch Prax 1 : 183–189

4. Howard AG, Hunt LE (1993) Anal Chem 65 : 2995–2998

5. Gast CH, Kraak JC, Poppe H, Maessen FJMJ (1979) J

Chromatogr 185 : 549–561

6. Spall WD, Lynn JG, Andersen JL, Valdez JG, Gurley LR

(1986) Anal Chem 58 : 1340–1344

7. LaFreniere KE, Fassel VA, Eckels DE (1987) Anal Chem

59 : 879–887

8. Roychowdhury SB, Koropchak JA (1990) Anal Chem 62 :

484–489

9. Caroli S, La Torre F, Petrucci F, Violante N (1994) Environ

Sci Pollut Res Int 1 : 205–208

10. Kumar UT, Vela NP, Caruso JA (1995) J Chromatogr Sci

33 : 606–610

11. Pritzl G, Stuer-Lauridsen F, Carlsen L, Jensen AK, Thorsen

TK (1996) Int J Environ Anal Chem 62 : 147–159

12. Hirayama N, Kuwamoto T (1988) J Chromatogr 447 : 323–

328

317

Fig. 6 HPLC chromatogram

of five phenylarsenic com-

pounds on a CN column, in-

jected amounts: 1400 ng

(CLARK I), 700 ng (all oth-

ers). For peak identification

see Table 2

Fig. 7 Comparison of UV spectra of the six phenylarsenic

compounds studied in HPLC. Numbers of spectra refer to the

compound numbers in Table 2

Fig. 5 HPLC chromatogram

of five phenylarsenic com-

pounds on an RP 18 column,

injected amounts: 700 ng

(CLARK I), 350 ng (all oth-

ers). For peak identification

see Table 2

13. Hirayama N, Kuwamoto T (1988) J Chromatogr 457 : 415–

420

14. Lopez-Sanchez JF, Amram MB, Lakkis MD, Lagarde F,

Rauret G, Leory MJF (1994) Fresenius J Anal Chem 348 :

810–814

15. Fish RH, Brinckman FE, Jewett KL (1982) Environ Sci

Technol 16 : 174–179

16. Witkiewicz W, Mazurek M, Szulc J (1990) J Chromatogr

503 : 293–357

17. Johnsen BA, Blanch JH (1984) Arch Belg Med Soc, Hyg,

Med Trav Med Leg, Suppl : 22–30

18. Skolowski M, Rozylo JK (1993) J Planar Chromatogr –

Mod TLC 6 : 467–471

19. Paulig G, Gielsdorf W (1981) Arch Kriminol 167 : 65–69

20. Zerba EN, Ruveda MA (1972) J Chromatogr 68 : 245–247

21. Schoene K, Steinhanses J, Bruckert HJ, König A (1992) J

Chromatogr 605 : 257–262

22. Schoene K, Bruckert HJ, Steinhanses J (1995) Analytik

Kampfstoff-kontaminierter Rüstungsaltlasten. Erich-

Schmidt-Verlag, Berlin

23. Schoene K, Bruckert HJ, Juerling H, Steinhanses J (1996) 

J Chromatogr A 719 : 401–409

24. Hannestad U, Soerbo B (1980) J Chromatogr 200 : 171–177

318

Fresenius J Anal Chem (1998) 361 : 318–323 – © Springer-Verlag 1998

Peter L. Neitzel · Wolfgang Walther 

·

Wolfgang Nestler

In-situ methylation 

of strongly polar organic acids 

in natural waters supported 

by ion-pairing agents 

for headspace GC-MSD analysis

Received: 17 July 1997 / Revised: 29 September 1997 / 

Accepted: 2 October 1997

Abstract Strongly polar organic substances like halogenated

acetic acids have been analyzed in surface water and ground-

water in the catchment area of the upper Elbe river in Saxony

since 1992. Coming directly from anthropogenic sources like

industry, agriculture and indirectly by rainfall, their concentra-

tions can increase up to 100 

µ

g/L in the aquatic environment of

this catchment area. A new static headspace GC-MSD method

without a manual pre-concentration step is presented to analyze

the chlorinated acetic acids relevant to the Elbe river as their

volatile methyl esters. Using an ion-pairing agent as modifier

for the in-situ methylation of the analytes by dimethylsulfate, a

minimal detection limit of 1 

µ

g/L can be achieved. Problems

like the thermal degradation of chlorinated acetic acids to halo-

genated hydrocarbons and changing reaction yields during the

headspace methylation, could be effectively reduced. The

method has been successfully applied to monitoring bank infil-

trate, surface water, groundwater and water works pumped raw

water according to health provision principles.

Introduction

There is a widespread pollution of the aquatic environment by

many anthropogenic organic substances in Central Europe due

to their wide range of application in industry, agriculture and

households and their biological persistence. Especially strongly

polar organic compounds like synthetic amino acids (chelating

agents EDTA, DTPA, NTA), aromatic sulfonic acids and halo-

genated carbon acids have become a subject of investigation in

the past and for the next decade [1, 2].

Many of these substances appear in surface water of Central

European rivers like Elbe, Rhine and Danube, in groundwater,

river bank infiltrate and in drinking water [3]. Most of the

haloacetic acids like mono-, di- and trichloroacetic acid (MCA,

DCA, TCA) come from sources like chemical industries (semi-

products of chemical synthesis, solvents for polymers and de-

tergents for metallic surfaces) or as metabolites from plant pro-

tection agents applied in agriculture. In 1992 chlorinated acetic

acids were detected in Elbe river water up to a concentration of

70 

µ

g/L [4], in tap and drinking water in the USA up to 160

µ

g/L in 1983/86 [5, 6]. By rainfall (degradation of halogenated

hydrocarbons in the troposphere) or river bank infiltration these

substances can reach the groundwater and can become a danger

potential for drinking water treatment [7]. Because of the very

high toxic and carcinogenic risks of some of these substances

such as di- and trichloroacetic acid, a fast and exact analytical

method for chloroacetic acids is needed to control their con-

centration, behavior and fate in surface, drinking and ground-

water according to health provision principles [8].

Analytical methods of choice to detect halogenated acetic

acids for levels less than 50 

µ

g/L are gas chromatography with

electron capture or mass selective detectors and sample pre-

concentration on solid phases or liquid-liquid extraction fol-

lowing diazomethane derivation [9–11]. Methyl esters of

haloacetic acids are so much less soluble in water than free

acids, giving headspace methods analytical possibilities. Using

the method of static headspace and gas chromatography with

dimethylsulfate (DMST) in-situ derivation [12], we have cal-

culated detection limits for the haloacetic acids with an MSD 

of about 50 

µ

g/L. That means, they are higher than the concen-

trations found in drinking and groundwater today. Main rea-

sons of this situation are to be found in the equilibrium level of

the distribution between liquid and gaseous phase for the micro

pollutants (as their methyl esters), in the yields of the deriva-

tion reaction and in thermal instability of the analytes during

the headspace procedure.

P. L. Neitzel · W. Walther

Dresden University of Technology, 

Institute for Groundwater Managemant, Mommsenstraße 13, 

D-01062 Dresden, Germany

W. Nestler (

౧)

Institute for Technology and Economics, 

Department of Civil Engineering and Architecture, 

Friedrich-List-Platz 1, D-01069 Dresden, Germany