Investigation of the behaviour of haloketones in water samples

Article abstract of DOI:10.1016/S0045-6535(00)00536-1

The behaviour of the haloketones (HKs) 1,1-Dichloropropanone (1,1-DCP), 1,1,1-Trichloropropanone (1,1,1-TCP) and 1,3-Dichloropropanone (1,3-DCP) in ultrapure water solutions and in fortified drinking water samples was invest/gated. Their concentrations were determined at regular time intervals by the use of a gas chromatography-electron capture detector (GC-ECD) method. Two different temperatures were studied. The results have shown that HKs decompose both in ultrapure water solutions and in drinking water samples. The decomposition rates are higher in the drinking water samples, especially at higher temperature. 1,1,1-TCP is the compound which decomposes fastest followed by 1,3-DCP and 1,1-DCP. Chloroform was formed both in the ultrapure water solutions and in the drinking water samples, probably due to the decomposition of 1,1,1-TCP. In the drinking water samples, formation of chloral hydrate was also observed.

Full text of DOI:10.1016/S0045-6535(00)00536-1

Chemosphere 44 +2001) 907±912
www.elsevier.com/locate/chemosphere
Investigation of the behaviour of haloketones in water samples
*
Anastasia D. Nikolaou , Themistokles D. Lekkas, Maria N. Kostopoulou,
Spyros K. Gol®nopoulos
Department of Environmental Studies, University of the Aegean, Karadoni 17, 81100 Mytilene, Greece
Received 1 September 2000; accepted 11 October 2000
Abstract
The behaviour of the haloketones +HKs) 1,1-Dichloropropanone +1,1-DCP), 1,1,1-Trichloropropanone +1,1,1-TCP)
and 1,3-Dichloropropanone +1,3-DCP) in ultrapure water solutions and in forti®ed drinking water samples was in-
vestigated. Their concentrations were determined at regular time intervals bythe use of a gas chromatography±electron
capture detector +GC±ECD) method. Two di€erent temperatures were studied. The results have shown that HKs
decompose both in ultrapure water solutions and in drinking water samples. The decomposition rates are higher in the
drinking water samples, especiallyat higher temperature. 1,1,1-TCP is the compound which decomposes fastest fol-
lowed by1,3-DCP and 1,1-DCP. Chloroform was formed both in the ultrapure water solutions and in the drinking
water samples, probablydue to the decomposition of 1,1,1-TCP. In the drinking water samples, formation of chloral
hydrate was also observed. Ó 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Haloketones; 1,1-dichloropropanone; 1,1,1-trichloropropanone; 1,3-dichloropropanone
1. Introduction
propanone, 1,1,1-trichloropropanone and 1,3-dichloro-
propanone +Lekkas, 1996). Theyhave been detected at
concentrations an order of magnitude lower than tri-
halomethanes and haloacetic acids +Nieminski et al.,
1993; LeBel et al., 1997; Williams et al., 1998; Simpson
and Hayes, 1998) and they have been reported to di-
minish at high pH, decrease with contact time and fur-
ther degrade throughout the distribution system +Arora
et al., 1997). 1,1,1-trichloropropanone decomposes to
form chloroform +Stevens et al., 1989; Koch et al., 1991;
Singer, 1994; Chen and Weisel, 1998).
Trihalomethanes were the ®rst class of chlorination
by-products to be identi®ed in drinking water +Bellar
et al., 1974; Rook, 1974; Gol®nopoulos et al., 1998),
followed byhaloacetic acids +Quimbyet al., 1980;
Christman et al., 1983; Miller and Uden, 1983; Reckow
and Singer, 1984; Krasner et al., 1989) and haloaceto-
nitriles, haloketones +HKs), and chloropicrin at lower
concentrations +Trehyand Bieber, 1980; Oliver, 1983;
Krasner et al., 1989; Williams et al., 1997; Nikolaou et
al., 1999, 2000). HKs belong to the volatile chlorination
by-products and have been found to be subject to base-
catalyzed hydrolysis and reactions with residual chlorine
+Krasner et al., 1989; Singer, 1994). The main HKs
identi®ed as chlorination by-products are 1,1-dichloro-
1,1-dichloropropanone and 1,1,1-trichloropropanone
have been found to have carcinogenic and mutagenic
e€ects on mice +Bull and Robinson, 1986). The con-
centration of these compounds in drinking water has not
yet been regulated.
In the present paper, the behaviour of HKs in ul-
trapure water solutions and in forti®ed drinking water
samples is investigated. The kinetics of HKs decompo-
sition and simultaneous chloroform formation are also
determined.
* Corresponding author. Tel.: +30-25136-000-227; fax: +30-
0251-36226.
E-mail address: nnikol@aegean.gr +A.D. Nikolaou).
0045-6535/01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 4 5 - 6 5 3 5 + 0 0 ) 0 0 5 3 6 - 1

908
A.D. Nikolaou et al. / Chemosphere 44 -2001) 907±912
Table 1
GC±ECD conditions
2. Experimental
2.1. Glassware
Carrier gas ¯ow
Oven temperature
1.6 ml/min
39°C +12 min)
175°C
Injector temperature
Split ratio
Detector temperature
All glassware used during analysis were prepared as
follows: washed with detergent, rinsed with tap water,
ultrapure water +Millipore: Milli-Ro 5 plus and Milli Q
plus 185), acetone +Mallinckrodt Chemical Works, St.
Louis) and placed in an oven at 150°C for 2 h.
1:25
300°C
Table 2
GC±MS conditions
2.2. Standard solutions
Carrier gas ¯ow
Oven temperature
1.3 ml/min
35°C +9 min)
to 40°C +3 min) at 1°C/min
to 148°C at 6°C/min
175°C
Stock solutions were prepared in Methyl-tert-butyl-
ether +MTBE) +Merck, for organic trace analysis) by the
addition of 1,1-dichloropropanone, 1,1,1-trichloro-
propanone and 1,3-dichloropropanone +Chemservice,
purity> 99%). The concentrations of the stock solutions
were calculated bythe volumetric ¯ask weight change.
These stock solutions were used to prepare a solution
with the three HKs +100 mg/l) in MTBE, known vol-
umes of which were injected into ultrapure water +Mil-
lipore: Milli-Ro 5 plus and Milli Q plus 185) in order to
obtain standard solutions with ®nal concentrations 0.25,
1.00, 4.00, 10.00, 25.00, 50.00 and 100.00 lg/l for system
calibration. The stock solution of 100 mg/l HKs was also
used for preparation of standard solutions 50 lg/l in
ultrapure water and in drinking water These solutions
were divided and stored in 40 ml glass vials capped with
PTFE-faced silica septum +Pierce 13075). The vials were
carefully®lled just to over¯owing without trapping air
bubbles in the sealed vial. Two di€erent temperatures
were tested: 21°C +ambient) and 30°C +in a water bath).
Injector temperature
Split ratio
1:25
MS conditions
Solvent delay12 min
MS temperature
EMV
280°C
2200
1.9
Scan/s
MS scan programme
35±265
0:32 mm i:d: Â 0:25 lm ®lm thickness. A fused silica
capillaryHP-VOC column +60 m  0:32 mm  1:8 lm)
was also used for gas chromatography±mass spectrom-
eter +GC±MS) veri®cation. The injection technique was
split/splitless and the carrier gas ¯ow 1.6 ml/min.
2.5. Analytical conditions
The analytical conditions of the gas chromatograph
and the mass spectrometer are presented in Tables 1
and 2.
2.3. Sample preparation
A modi®cation of EPA Method 551.1, which in-
cludes liquid±liquid extraction +LLE) with MTBE +EPA,
1998) was performed. 6 g of anhydrous sodium sulfate
+Merck) and 2 ml of MTBE were added to 35 ml of HKs
solution in a 40-ml vial. The vial was sealed and shaken
byhand for 1 min and left undisturbed for 2 min. 1 ll of
the ether phase was then injected into the gas chro-
matograph. Each solution was analyzed every half an
hour in order to observe the e€ect of time on HKs
concentrations.
3. Results and discussion
The recoveries of the GC±ECD method for HKs
determination have been estimated at six concentration
levels and are given in Table 3.
The recoveries of the method ranged from 66.2% to
121.6%. The lowest value was observed for 1,1,1-TCP at
a concentration of 1.0 lg/l. 1,3-DCP was not recovered
at concentrations below 1.0 lg/l and the relative SD for
this compound was high.
2.4. Apparatus
The instrumentation used included a Hewlett Pack-
ard Gas Chromatograph 5890 Series II with a 63Ni
Electron Capture Detector and a Hewlett Packard Mass
Selective Detector 5971. The system was supported by
HP G1034C system software. The carrier gas used was
Helium and the make-up gas, nitrogen. The column used
for gas chromatography±electron capture detector +GC±
ECD) analysis was fused silica DB-1, 30 m Â
The detection limits were calculated as follows:
DL ˆ t  S;
ꢀ1†
where DL is the detection limit +lg/l), t the Student's t
value for a 99% con®dence level and an SD estimate
with n À 1 degrees of freedom +t ˆ 3747 when n ˆ 5) and
S is the SD of the replicate analyses.

A.D. Nikolaou et al. / Chemosphere 44 -2001) 907±912
909
Table 3
Mean recovery+%), relative SD +%, in parenthesis) and detection limits +DLs) of HKs at six concentration levels ꢀn ˆ 5†
Concentration levels +lg/l)
DL +lg/l)
HKs
0.5
1.0
2.0
5.0
10.0
20.0
1,1-DCP
120
+2.6)
79
+2.4)
83.7
+5.6)
112.48
+1.7)
107.0
+4.5)
102.54
+6.1)
0.06
1,1,1-TCP
108
+7.29)
66.2
+11.2)
75.8
+5.6)
111.32
+5.3)
111.7
+1.3)
105.49
+4.9)
0.15
a
1,3-DCP
121.6
+35.1)
100.7
+13.0)
98.4
+15.8)
88.0
+20.9)
97.35
+3.2)
0.98
a Not recovered.
The concentrations of HKs versus time in a forti®ed
water sample at 21°C, in a forti®ed water sample at 30°C
and in an ultrapure water solution at 30°C are plotted in
Figs. 1+a)±+c), respectively.
k of the reaction. Representative lnC±t plots are pre-
sented in Figs. 2+a)±+c) for HKs concentrations in a
forti®ed drinking water sample at 21°C. The decompo-
sition rates are presented in Table 4, while the linear
regression equations that best describe the reactions in
each case and the corresponding con®dence intervals are
shown in Table 5.
The rate of the decomposition was determined for
each compound bythe use of linear regression after the
concentration values were ln-transformed. The loga-
rithms of the concentration values plotted versus time
give a linear curve, the slope of which represents the rate
The fact that the zero value is not included in the
estimated con®dence intervals for k at 95% con®dence
Fig. 1. Decomposition of HKs in: +a) forti®ed water sample at 21°C; +b) forti®ed water sample at 30°C; +c) ultrapure water solution at
30°C.

910
A.D. Nikolaou et al. / Chemosphere 44 -2001) 907±912
Fig. 2. Decomposition of HKs in forti®ed drinking water at 21°C.
Table 4
Decomposition rates of HKs
1,3-DCP shows a higher decomposition rate in forti®ed
drinking water sample at 21°C.
Chloroform formation was observed during the de-
composition of HKs, as is presented in Fig. 3 for a
forti®ed drinking water sample at 21°C. The kinetic
equations describing chloroform formation have been
estimated using the procedure described above for HKs
and are shown in Table 6.
Water sample
Ultrapure
water, 30°C
21°C
30°C
1,1-DCP
1,1,1-TCP
1,3-DCP
0.022
0.182
0.220
0.071
1.393
0.099
0.010
0.062
0.098
Chloroform is formed faster in ultrapure water so-
lutions than in drinking water samples. This fact indi-
cates that HKs decomposition in drinking water may
lead to other products as well in addition to chloroform.
A faster chloroform formation rate in drinking water
sample was observed at 21°C than at 30°C.
Chloral hydrate formation was also observed in
forti®ed water samples only. Chloral hydrate concen-
tration versus time in a forti®ed drinking water sample
at 21°C is plotted in Fig. 4.
level statisticallycon®rms that the concentrations of
HKs are dependent on time.
In ultrapure water solutions at 30°C and in forti®ed
drinking water samples at 21°C, the compound which
decomposes fastest is 1,3-DCP, followed by1,1,1-TCP
and 1,1-DCP. However, in forti®ed water samples at
30°C the highest decomposition rate is that of 1,1,1-TCP
followed bythe rates of 1,3-DCP and 1,1-DCP decom-
position. The three HKs studied have higher decompo-
sition rates in forti®ed drinking water samples than in
ultrapure water solutions. 1,1-DCP and 1,1,1-TCP de-
compose faster at 30°C than at 21°C. On the contrary,
The formation of chloral hydrate only in forti®ed
drinking water samples and not in ultrapure water so-
lutions suggests that chloral hydrate is not produced
directlyfrom the decomposition of HKs but can be

A.D. Nikolaou et al. / Chemosphere 44 -2001) 907±912
911
Table 5
Equations describing the decomposition of HKs ꢀln C ˆ ln C0 À kt†
Equation
n
tn À 2; a=2
Con®dence
interval for ln C0
Con®dence
interval for k
Con®dence
level +%)
Forti®ed water sample, 21°C
1,1-DCP
1,1,1-TCP
1,3-DCP
ln C ˆ 3:264 À 0:022t
36
1.645
1.645
1.645
+3.235±3.293)
+3.580±3.662)
+3.156±3.340)
+0.021±0.023)
+0.181±0.183)
+0.218±0.222)
95
95
95
ln C ˆ 3:621 À 0:182t
ln C ˆ 3:248 À 0:220t
36
35
Forti®ed water sample, 30°C
1,1-DCP
1,1,1-TCP
1,3-DCP
ln C ˆ 3:299 À 0:071t
36
6
1.645
1.943
1.645
+3.252±3.346)
+3.406±3.592)
+1.333±1.647)
+0.070±0.072)
+1.366±1.420)
+0.096±0.102)
95
95
95
ln C ˆ 3:499 À 1:393t
ln C ˆ 1:490 À 0:099t
35
Ultrapure water solution, 30°C
1,1-DCP
1,1,1-TCP
1,3-DCP
ln C ˆ 3:286 À 0:010t
36
36
36
1.645
1.645
1.645
+3.253±3.319)
+3.365±3.573)
+3.910±4.188)
+0.009±0.011)
+0.060±0.064)
+0.096±0.100)
95
95
95
ln C ˆ 3:469 À 0:062t
ln C ˆ 4:049 À 0:098t
Fig. 3. Formation of chloroform in a water sample forti®ed with HKs at 21°C.
Table 6
Equations describing the formation of chloroform +ln C ˆ ln C0 ‡ kt)
Equation
n
tn À 2; a=2
Con®dence
interval for ln C0
Con®dence
interval for k
Con®dence
level +%)
Forti®ed water sample, 21°C
CHCl3
ln C ˆ 1:846 ‡ 0:092t
34
35
29
1.645
1.645
1.703
+1.740±1.952)
+2.875±3.009)
+0.009±0.237)
+0.090±0.094)
+0.008±0.010)
+0.132±0.136)
95
95
95
Forti®ed water sample, 30°C
CHCl3
ln C ˆ 2:942 ‡ 0:009t
Ultrapure water solution, 30°C
CHCl3
ln C ˆ 0:123 ‡ 0:134t
Chloral Hydrate
Time (h)
Fig. 4. Chloral hydrate formation in a forti®ed water sample at 21°C.

912
A.D. Nikolaou et al. / Chemosphere 44 -2001) 907±912
Gol®nopoulos, S.K., Xylourgidis, N.K., Kostopoulou, M.N.,
Lekkas, T.D., 1998. Use of a multiple regression model for
predicting trihalomethane formation. Water Res. 32, 2821±
2829.
formed through another pathwaywhen natural organic
matter is present.
Koch, B.K., Krasner, S.W., Sclimenti, M.J., Schimp€, W.K.,
1991. Predicting the formation of DBPs bythe simulated
distribution system. J. AWWA, October 1991, 62±70.
Krasner, S.W., McGuire, M.J., Jacangelo, J.G., Patania, N.L.,
Reagan, K.M., Aieta, E.M., 1989. The occurrence of
disinfection by-products in US drinking water. J. AWWA
81, 41±53.
4. Conclusions
The behaviour of the HKs 1,1-DCP, 1,1,1-TCP and
1,3-DCP in ultrapure water solutions and forti®ed
drinking water samples was investigated at two di€erent
temperatures. The reaction kinetics of HKs decompo-
sition and simultaneous chloroform formation were
determined in each case. The results have shown that the
decomposition rates of HKs are higher in forti®ed water
samples than in ultrapure water solutions. In ultrapure
water solutions at 30°C and in forti®ed drinking water
samples at 21°C, the highest decomposition rate is that
of 1,3-DCP, followed by1,1,1-TCP and 1,1-DCP.
However, in forti®ed water samples at 30°C 1,1,1-TCP
decomposes faster than 1,3-DCP and 1,1-DCP. While
the rates of 1,1-DCP and 1,1,1-TCP decomposition are
higher at 30°C than at 21°C, the rate of 1,3-DCP de-
composition is higher at 21°C. Chloroform formation is
faster at 21°C than at 30°C and faster in ultrapure water
solutions than in drinking water samples. The latter in-
dicates that HKs decomposition in drinking water may
lead to other products as well in addition to chloroform.
Chloral hydrate formation was also observed only in
forti®ed water samples and not in ultrapure water so-
lutions, which suggests that chloral hydrate is not pro-
duced directlyfrom the decomposition of HKs but can
be formed through another pathwayin the presence of
natural organic matter.
LeBel, G.L., Benoit, F.M., Williams, D.T., 1997. A one-year
surveyof halogenated DBPs in the distribution system of
treatment plants using three di€erent disinfection processes.
Chemosphere 34, 2301±2317.
Lekkas, T.D., 1996. Environmental Engineering I: Manage-
ment of Water Resources. Department of Environmental
Studies, Universityof the Aegean, Mytilene, Greece.
Miller, J.W., Uden, P.C., 1983. Characterization of non-volatile
aqueous chlorination products of humic substances. Envi-
ron. Sci. Technol. 17, 150±157.
Nieminski, E.C., Chaudhuri, S., Lamoreaux, S.-T., 1993. The
occurrence of DBPs in Utah drinking waters. J. AWWA 85,
98±105.
Nikolaou, A.D., Gol®nopoulos, S.K., Kostopoulou, M.N.,
Lekkas, T.D., 2000. Decomposition of dihaloacetonitriles in
water solutions and forti®ed drinking water samples.
Chemosphere 41, 1149±1154.
Nikolaou, A.D., Kostopoulou, M.N., Lekkas, T.D., 1999.
Organic by-products of drinking water chlorination: a
review. Global Nest: The Int. J. 1, 143±156.
Oliver, B.G., 1983. Dihaloacetonitriles in drinking water: algae
and fulvic acid as precursors. Environ. Sci. Technol. 17,
80±83.
Quimby, B.D., Delaney, M.F., Uden, P.C., Barnes, R.M., 1980.
Determination of the aqueous chlorination products of
humic substances bygas chromatographywith microwave
emission detection. Anal. Chem. 52, 259±267.
Reckow, D.A., Singer, P.C., 1984. The removal of organic
halide precursors bypreozonation and alum coagulation.
J. AWWA 76, 151±158.
References
Arora, H., LeChevallier, M.W., Dixon, K.L., 1997. DBP
occurrence survey. J. AWWA 89 +6), 60±68.
Bellar, T.A., Lichtenberg, J.J., Kroner, R.C., 1974. The
occurrence of organohalides in chlorinated drinking water.
J. AWWA 66, 703±711.
Rook, J.J., 1974. Formation of haloforms during chlorination
of natural waters. Water Treat. Exam. 23, 234±242.
Simpson, K.J., Hayes, K.P., 1998. Drinking water disinfection
by-products: an Australian perspective. Water Res. 32,
1522±1528.
Bull, R.J., Robinson, M., 1986. Carcinogenic activityof
haloacetonitrile and haloacetone derivatives in the mouse
skin and lung. In: Water chlorination: Chemistry, Environ-
mental Impact and Health E€ects, vol. 5. Lewis Publishers,
USA, pp. 221±227.
Singer, P.C., 1994. Control of disinfection by-products in
drinking water. J. Environ. Eng. 120, 727±744.
Stevens, A.A., Moore, L.A., Miltner, R.J., 1989. Formation
and control of non-trihalomethane disinfection by-prod-
ucts. J. AWWA 81, 54±60.
Chen, W.J., Weisel, C.P., 1998. Halogenated DBP concentra-
tions in a distribution system. J. AWWA 90, 151±163.
Christman, R.F., Norwood, D.L., Millington, D.S., 1983.
Identityand yields of major halogenated products of aquatic
fulvic acid chlorination. Environ. Sci. Technol. 17, 625±628.
EPA, 1998. EPA Method 551.1. Determination of chlorinated
disinfection by-products, chlorinated solvents, and halogen-
ated pesticides/herbicides in drinking water byliquid±liquid
extraction and gas chromatographywith electron capture
detection. USEPA, Oce of Water, Technical Support
Center, 26 W. Martin Luther King Dr., Cincinnati, OH
45268.
Trehy, M.L., Bieber, T.I., 1980. In: Proceedings of the ACDS
Division of Environmental Chemistry, San Fransisco, CA,
24±29 August. American Chemical Society, Washington,
DC.
Williams, D.T., Benoit, F.M., LeBel, G.L., 1998. Trends in
levels of disinfection by-products. Environmetrics 9, 555±
563.
Williams, D.T., LeBel, G.L., Benoit, F.M., 1997. Disinfection
by-products in Canadian drinking water. Chemosphere 34,
299±316.