OCTOBER 2001

VOLUME 49, NUMBER 10

© Copyright 2001 by the American Chemical Society

Determination of Dithiocarbamate Fungicide Residues in Food by a

Spectrophotometric Method Using a Vertical Disulfide Reaction
System
Eloisa D. Caldas,*,†,‡ Maria Hosana Conceição,§ Maria Clara C. Miranda,‡
Luiz César K. R. de Souza,‡ and Joaquim F. Lima§
College of Health Sciences and Chemistry Institute, University of Brası́lia, 70919-970 Brası́lia, DF, Brazil,
and Public Health Central Laboratory of Federal District-LACEN-DF, 70830-010 Brası́lia, DF, Brazil

Dithiocarbamates are a class of fungicides extensively used in many crops worldwide. The current
residue definition of dithiocarbamates in food for compliance with maximum residue limits, at
national and international levels, is total residues arising from the use of any or each dithiocarbamate
fungicide, determined as CS2. The analytical method most frequently used to analyze dithiocarbamate
residues in food for monitoring purposes was proposed more than 30 years ago. In this method,
total dithiocarbamates are decomposed to CS2, which is purified and reacted with a cupric reagent.
The yellow complex formed is quantified by spectrophotometry. In this paper, a new reaction system
for the purification and complexation of CS2 is proposed. The new system is less fragile than the
traditional design, is easier to assemble, and allows for a higher sample throughput, in addition to
being of low cost. Recovery of added mancozeb, thiram, or ziram (0.15-8.0 mg/kg) in rice, beans,
apple, banana, orange, papaya, tomato, cucumber, and potato ranged from 82 to 120%, with relative
standard deviations from 0 to 10% (n ) 3 or 5). Analysis of apple, tomato, and papaya samples with
field-incurred dithiocarbamate residues showed comparable results using both the traditional and
the new reaction systems.

Keywords: Dithiocarbamate fungicides; CS2 reaction system; spectrophotometry; food analysis

INTRODUCTION To guarantee the use of pesticides in the field accord-

ing to good agricultural practices and to protect con-
Dithiocarbamate fungicides are used worldwide on a
sumer health, regulatory agencies establish tolerances,
range of crops, mainly due to their high efficiency in
controlling plant fungal diseases and relatively low or maximum residue limits (MRL), in treated crops. The
mammalian acute toxicity. The toxicological significance methodology needed to analyze residues of dithiocar-
of the ethylenebis(dithiocarbamate) residues in food bamates in food is not always trivial, mainly due to the
stems from the metabolite or degradation product polymeric character and insolubility of some of these
ethylenethiourea, known to be carcinogenic and terato- compounds (Figure 1). Reversed-phase HPLC methods
genic in laboratory animals (1). have been proposed for the analysis of thiram and other
dithiocarbamates in food using UV (2, 3) or electro-
chemical detection (3, 4). However, these methods are
* Author to whom correspondence should be addressed
[e-mail eloisa@unb.br; fax +55(61)321-9995]. not selective enough to distinguish among members
† College of Health Sciences. within the ethylenebis(dithiocarbamate) subclass (na-
‡ Public Health Central Laboratory of Federal District. bam, maneb, zineb, and mancozeb) or within the N,N-
§ Chemistry Insitute.
dimethyldithiocarbamate subclass (ziram and ferban).
10.1021/jf010124a CCC: $20.00 © 2001 American Chemical Society
Published on Web 09/05/2001
4522 J. Agric. Food Chem., Vol. 49, No. 10, 2001 Caldas et al.

Figure 2. Traditional in-series two-trap CS2 reaction system.

tial of decreasing analysis times and increasing sample

throughput while maintaining good analytical perfor-
mance and low cost.

MATERIALS AND METHODS

Reagents. All chemicals were of analytical grade, obtained
from Vetec or E. Merck. Solutions of copper(II) acetate mono-
hydrate (color reagent) were prepared (0.48 g in 100 g of
diethanolamine/1 L of water) and kept at 5 °C protected from
light. An acid stannous chloride solution was prepared by
dissolving 31.25 g of SnCl2‚2H2O in 500 mL of concentrated
Figure 1. Dithiocarbamate fungicides representative of three
subclasses. HCl and bringing the volume to 2.5 L. A stock solution of
carbon disulfide (0.002 mg/mL) was prepared in ethanol.
Verma et al. (5) describe the use of tetraacetonitrile- Nitrogen gas was of commercial grade (White Martins). A
copper (I) perchlorate for the spectrophotometric deter- 0.543 mg/g stock mixture of mancozeb (74.5% purity and 41.8%
CS2; Dr. Ehrenstorfer), a 0.541 mg/g stock mixture of maneb
mination of thiram in grain, and Malik (6) proposed the (95% purity; Ultra Scientific, North Kingstown, RI), and a
determination in grain of ferbam as its iron(II)-batho- 0.612 mg/g stock mixture of ziram (98% purity; Ultra Scien-
phenanthroline-tetraphenylborax complex. tific) were prepared in milk sugar lactose (E. Merck). A
A multiresidue methodology that could discriminate 0.0150% stock solution of thiram (99.8% purity; Uniroyal
among all of the dithiocarbamate compounds in a single Chemical, Middlebury, CT) was prepared in acetone. The
analysis would be advantageous for monitoring and traditional and proposed reaction systems were made by
dietary risk assessment purposes. As this methodology Vidrolima Comércio, Importação e Serviços em Vidros Ltda,
is not currently available, the MRLs of dithiocarbam- Brası́lia, DF, Brazil.
ates, at both the national (7-10) and international Standard Curve. Known volumes of the CS2 stock solution
were transferred to 25 mL volumetric flasks, 15 mL of the color
levels (11), are defined as total residues arising from
reagent was added, the volume was completed with ethanol,
the use of any or each dithiocarbamate fungicide, and the solution was allowed to stand for 15 min. The
determined as CS2. absorbance was measured in a continuous flow Gilford Stasar
Gas chromatographic methods have been described III spectrophotometer (Gilford Instruments, Oberlin, OH) at
for the determination of the CS2 formed after acid 435 nm. A typical standard curve ranged from 0.38 to 23 µg
hydrolysis of dithiocarbamates in food using a head- of CS2/mL, corresponding to 0.10-6 mg of CS2/kg in the food
space technique (12-14) or after the partitioning of the sample.
CS2 into an organic solvent (15, 16). However, the Apparatus. The acid decomposition of the dithiocarbamates
method most frequently used worldwide, mainly be- occurs in a two-neck round-bottom boiling flask placed in a
cause of its low cost and minimal analytical require- heating mantle. The flask is connected to a nitrogen inlet tube
and to a condenser, which is then connected to the traditional
ments, was proposed by Cullen (17) and modified by or proposed CS2 reaction system.
Keppel (18). This method is based on the spectrophoto- Traditional CS2 Reaction Systems. A system based on the
metric determination of the cupric complex formed with design of Cullen (17) consists of two traps in series (Figure 2).
the CS2 evolved from the acid decomposition of dithio- Trap 1 contains 20 mL of 10% NaOH solution, and trap 2
carbamates in the presence of stannous chloride as a contains 15 mL of color reagent. A modification of this design
reducing agent. Recently, a flow injection system for the was proposed by Thier and Zeumer (21) and includes a third
spectrophotometric quantification of the CS2 partitioned trap containing 10 mL of a 30% Pb(CH3COO)2‚3H2O solution,
into hexane was reported (19). which is connected to the condenser and to the NaOH-
The apparatus for the purification of the CS2 formed containing trap. The purpose of both NaOH and lead acetate
solutions is to purify the CS2 formed, removing volatile
after hydrolysis and formation of the CS2 cupric complex
interferences, mainly H2S and SO2.
used in the Cullen and Keppel method has been utilized Proposed Vertical CS2 Reaction System. The new system is
by laboratories over the years with no significant composed of two segments assembled in a vertical format
changes from its original design (7, 8, 20). The objective (Figure 3). The first segment (32 cm total length) contains an
of this work was to propose a new CS2 reaction system, antireflux retention valve, a diffuser tube, and a trap contain-
less fragile than the traditional design, with the poten- ing the 10% NaOH solution (trap 1). The second segment (27
Dithiocarbamate Residues in Food J. Agric. Food Chem., Vol. 49, No. 10, 2001 4523

AOAC International to analyze dithiocarbamates in

fungicide formulations utilizes a similar system, con-
taining three traps in series and a vacuum line to pull
gas through the system (22). In this methodology, CS2
is reacted with methanolic KOH in the last trap and
xanthate formed is titrated with I2 solution.
The two-trap and three-trap in series systems were
tested for the analysis of dithiocarbamates in tomato.
The traps contain a spiral (Figure 2) to give a homoge-
neous contact between the gases and the reagents and
to guarantee the efficiency of the reactions, namely, the
removal of volatile interferences and the complexation
of CS2 with the cupric reagent. The performance of the
two-trap system was satisfactory. Recovery of added
mancozeb, ziram, or thiram (0.15-4.0 mg of CS2/kg)
ranged from 81 to 105%, and relative standard devia-
tions of triplicate analyses were from 0 to 6%. When
the three-trap system was used, recovery of added
mancozeb, maneb, or thiram (0.8-5.0 mg of CS2/kg)
ranged from 45 to 75%, and relative standard deviations
of triplicate analyses were from 0 to 21%. The low
performance of the three-trap system is probably due
to the higher number of connections, which can increase
the chance of gas leaking. The good performance of the
two-trap system shows that the NaOH solution is
sufficient to purify the CS2 prior to complexation and
an additional trap with lead acetate solution is not
necessary.
The two-trap system was used for two years in the
laboratory for monitoring dithiocarbamates in tomato
Figure 3. Proposed vertical CS2 reaction system.
collected in the field and at the market of the Federal
cm total length), which is inserted inside the first segment, District area. During this routine work, however, this
contains the reaction tube (trap 2), with the color reagent. system showed some drawbacks. The assembling and
Sample Preparation and Analysis. Dithiocarbamates are deassembling procedure is time-consuming and has to
nonsystemic fungicides, and their residues are expected to be done very carefully to prevent damage. In addition,
remain on the surface of the crop. Polished rice and dry beans the seal of the spherical joints that connect the traps
were analyzed without any prior processing. The other com- (Figure 2) had to be checked before each analysis to
modities (apple, orange, papaya, banana, cucumber, tomato, guarantee a leak-free system.
and potato) were frozen and subsequently cut into pieces.
Spiked samples were used to test the traditional and new CS2 Proposed Vertical CS2 Reaction System. To solve
reaction systems. Of each sample, 100 g was transferred to some of the problems present in the CS2 reaction system
the two-neck round-bottom flask, and a known volume (thiram) based on the traditional design, we developed a new
or mass (mancozeb, maneb, or ziram) of the dithiocarbamate reaction system (Figure 3). The system is very compact,
stock solution or mixture was added. The flask was agitated, relatively robust, easy to assemble and clean, and
175 mL of stannous chloride acid solution was added, and the practically free of losses at the connections. The same
condenser and the N2 inlet were connected. The traditional sample preparation, reaction times, and volumes and
two-trap (Figure 2), three-trap in series, or new CS2 reaction concentrations of the reagents were used with both the
system (Figure 3) was connected to the condenser and the N2
gas and heating mantle turned on. Gas flow (100-180 mL/
traditional two-trap and the new reaction systems.
min) and heat were controlled to produce a gentle and constant The system is composed of two segments connected
gas flow and reflux. After 45 min, the system was turned off, vertically to the condenser (Figure 3), which is connected
the solution in trap 2 was transferred to a 25 or 50 mL to the two-neck round-bottom hydrolysis flask. From the
volumetric flask, the volume was completed with ethanol, and condenser, the N2 gas carrying the CS2 formed by the
the absorbance was measured at 435 nm. For all crops, a acid hydrolysis of the dithiocarbamates passes through
control sample with no dithiocarbamate added was analyzed. the antireflux retention valve and the diffuser tube,
No dithiocarbamate was found in any control sample tested.
reaching the NaOH solution (trap 1). After leaving trap
1, the gases enter the tube covered by a spiral in the
RESULTS AND DISCUSSION second segment through an upper entrance and reach
Traditional CS2 Reaction Systems. The system the color reagent in trap 2. The spiral is intended to
proposed initially for the spectrophotometric analysis guarantee a homogeneous contact between the gases
of dithiocarbamates consists of a two-neck round-bottom and the color reagent and an efficient complexation
flask connected to a condenser, which is connected to reaction.
two traps in series (Figure 2). Air or nitrogen is The most critical component of the new system is the
introduced through the system, and a vacuum line is diffusion tube. The gas flow leaving the diffusion tube
connected to the last trap (17, 18, 20). Thier and Zeumer reaches trap 1 through four small holes in the base of
(21) modified this design by the addition of a third trap the tube that contains the diffuser tube. This creates a
containing lead acetate solution to aid in the removal smoother gas flow through the NaOH solution and
of interferences. They used air to carry gases through increases the extraction efficiency of the interferences
the system. The methodology recommended by the formed in the hydrolysis flask. When the diffusion tube
4524 J. Agric. Food Chem., Vol. 49, No. 10, 2001 Caldas et al.

Table 1. Recovery of Added Dithiocarbamate Fungicides Table 2. Analysis of Samples with Field-Incurred
in Food Using the Proposed Vertical CS2 Reaction Dithiocarbamate Residues Using the Traditional
System In-Series and the Proposed Vertical Reaction Systemsa
amount added av % RSDb traditional proposed
food compound (mg of CS2/kg) recoverya (%)
mg of RSD mg of RSD RSDb
apple mancozeb 0.15 102 9.9 crop CS2/kg (%) CS2/kg (%) (%)
1.0 82 7.8
apple 1.2 4.7 0.98 12 14
3.0 98 5.1
0.96 13 0.87 10 6.9
thiram 1.0 110 0
0.29 11 0.34 6.7 11
ziram 1.0 90 5.1
tomato 0.45 6.9 0.47 4.3 3.1
papaya mancozeb 0.15 120 2.8 0.49 10 0.43 7.3 9.2
1.0 106 4.9 papaya skin 0.46 13 0.48 7.1 2.7
thiram 0.15 103 2.8 a Each residue value is the average of three replicate samples.
1.0 91 1.5 b Relative standard deviation of residues in both systems.
3.0 106 3.0
orange mancozeb 0.15 86 4.5 foods chosen represent the different crop classes that
1.0 95 7.3 are normally treated with dithiocarbamates: cereal
banana mancozeb 0.15 105 3.8 grains (rice), legume vegetables (beans), fruits with
1.0 93 5.1 edible and inedible peels (apple, banana, orange, and
2.0 106 8.2 papaya), fruiting vegetables (tomato and cucumber), and
thiram 1.0 110 0 root vegetables (potato). The results are shown in Table
ziram 1.0 87 4.0
1. For all of the compounds and foods tested, the
dry beans mancozeb 0.15 102 9.9 recoveries fell within the range of 82-120%, with
1.0 96 1.0 relative standard deviations (n ) 3 or 5) ranging from
thiram 1.0 110 0 0 to 10%.
ziram 1.0 90 3.9
The vertical CS2 reaction system described in this
polished rice mancozeb 0.15 120 3.3 paper has been applied successfully in our laboratory
1.0 101 7.5 for monitoring dithiocarbamates in food for compliance
8.0 95 6.6 with Brazilian MRLs and for dietary exposure assess-
thiram 1.0 91 2.8
ziram 0.15 108 1.5 ment. The system represents a considerable improve-
1.0 110 0 ment from the traditional in-series two-trap system in
3.0 106 7.4 three main aspects: (a) reduced space requirements in
a laboratory; (b) reduced time needed for assembling/
potato mancozeb 0.15 98 10
1.0 86 4.0 deassembling, resulting in a higher sample throughput;
thiram 1.0 107 2.0 and (c) lowered repair costs. In a laboratory bench space
ziram 1.0 90 2.3 of 1.5 × 0.5 m, where there were two in-series two-trap
systems, we now have five new systems. The two-trap
tomato mancozeb 0.15 107 0
1.0 105 2.0 traditional system has at least four segments, which
4.0 103 1.4 have to be handled carefully to avoid damage, demand-
ing ∼10 min for assembling, including checking for gas
cucumber mancozeb 0.15 104 7.2 leaks. The vertical system has only two segments, which
1.5 102 3.9
thiram 0.15 100 0 are much less fragile, requiring <2 min to assemble,
1.5 93 0 and does not need to be checked for gas leaks. The
ziram 0.15 100 0 robustness of the new system also reduced considerably
1.5 91 3.8 its maintenance due to damaged parts, decreasing the
a n ) 3 in all cases except for thiram in papaya and ziram in costs per analysis.
rice at 0.15 and 3.0 mg/kg, where n ) 5. b Relative standard From over the 100 food samples of rice, beans, tomato,
deviation. potato, banana, papaya, and orange collected at local
market and analyzed in the laboratory using the vertical
was removed from the system, large gas bubbles were system, dithiocarbamates were detected in only tomato,
formed, carrying basic vapor from trap 1 to the color apple, and papaya skin. Analysis of samples with field-
reagent solution in trap 2, changing its pH. The ef- incurred dithiocarbamate residues was performed using
ficiency of the complexation reaction in trap 2 decreased both the traditional in-series and the proposed vertical
as the pH increased, yielding low recoveries (<60%). systems (Table 2). The two systems yield comparable
The N2 flow rate was also found to be critical and results, with the relative standard deviations (RSD) of
should not be >180 mL/min; otherwise, it could also the values falling within the same range of each system
carry NaOH from trap 1 to trap 2. Higher flow rates RSD.
can decrease the volume of the alcoholic color reagent Conclusion. Although chromatographic methods
solution in trap 2, also affecting the performance of the have been applied successfully in the analysis of dithio-
system. The antireflux retention valve helps to regulate carbamates as CS2 in food, the spectrophotometric
the gas flow, in addition to preventing any reflux in case method proposed by Cullen (17) and Keppel (18) is still
of a sudden decrease in pressure. widely used. The method can achieve the sensitivity
The efficiency and reproducibility of the new system required for compliance with national and international
(Figure 3) were tested with different food crops spiked MRLs, and it is especially suitable for monitoring
with dithiocarbamates (Figure 1) at levels from 0.15 to laboratories in developing countries, due to its low costs
8.0 mg of CS2/kg. These concentrations are within the and minimal laboratory and analytical expertise re-
range of MRLs for dithiocarbamates in various crops quirements. We have proposed a new CS2 reaction
in Brazil (10) and in the Codex Alimentarius (11). The system for use with the spectrophotometric method. The
Dithiocarbamate Residues in Food J. Agric. Food Chem., Vol. 49, No. 10, 2001 4525

new system is less fragile than the old system, is easier (11) Codex Alimentarius. Pesticide Residues in Foods
to assemble, and allows for higher sample throughput. Maximum Residue Limits; Food and Agriculture Orga-
This new reaction system can potentially be applied for nization of the United Nations, World Health Organi-
the analysis of dithiocarbamates in fungicide formula- zation: Rome, Italy, 2000; Vol. 2B.
(12) Hill, A. R. C. Headspace Methods for Dithiocarbamates.
tions to replace the apparatus recommended by the
In Modern Methods for Pesticide Analysis. Emerging
methodology of the AOAC International (22). Strategies for Pesticide Analysis; Cairns, T., Sherma, J.,
Eds.; CRC Press: London, U.K., 1992; pp 213-231.
ACKNOWLEDGMENT (13) Jianren, Y.; Yongquan, Z.; Shuzshen, J.; Zhengguo, W.;
Fuzhen, Z. Gas Chromatographic Headspace Determi-
We acknowledge the assistance of Monica Moraes nation of Dithiocarbamate Fungicide Residues on Veg-
Nunes with sample preparation and Carol R. Johnson etables. Sci. Agric. Sin. 1989, 22, 76-80.
for reviewing the English language. (14) Friedrichs, K.; Winkeler, H. D.; Gerhards, P. Determi-
nation of Dithiocarbamate Residues in Foodstuff by
LITERATURE CITED Headspace Gas Chromatography and Flame Photomet-
ric Detection. Z. Lebensm. Unters. Forsch. 1995, 201,
(1) WHO. Pesticide Residues in Foods1993. Evaluations. 69-73.
Part IIsToxicology; International Programme on Chemi- (15) Pelegrı́, R.; Gamón, M.; Coscollá, R.; Beltrán, V.; Cuñat,
cal Safety; World Health Organization: Geneva, Swit- P. Determinación de residuos de mancozeb en frutas y
zerland, 1994. hortalizas por cromatografı́a de gases. Modificación del
(2) Gustafsson, K. H.; Fahlgren, C. H. Determination of método de recogida de disulfuro de carbono producido
Dithiocarbamate Fungicides in Vegetable Foodstuffs by em la descomposición ácida. Rev. Agroquim. Tecnol.
High-Performance Liquid Chromatography. J. Agric. Aliment. 1991, 31, 573-579.
Food Chem. 1983, 31, 461-463. (16) Harrington, P.; Horner, J.; Hird, S.; Griffiths, T.; Rey-
(3) van Lishaut H.; Schwack, W. Selective Trace Determi- nolds, S. L. Modification of the Method for Measurement
nation of Dithiocarbamate Fungicides in Fruit and of Dithiocarbamate Residues as Carbon Disulfide in
Vegetables by Reverse-Phase Ion-Pair Liquid Chroma- Fruit and Vegetables; Central Science Laboratory: York,
tography with Ultraviolet and Electrochemical Detec- U.K., 1998; pp 1-34.
tion. J. AOAC Int. 2000, 83, 720-727. (17) Cullen, T. E. Spectrophotometric Determination of
(4) Silva, Ma del P.; Procopio, J. R.; Hernández, L. Electro- Dithiocarbamates Residues on Food Crops. Anal. Chem.
chemical Detection in the Determination of Several 1964, 36, 221-224.
Dithiocarbamates by Reverse-Phase Liquid Chromatog- (18) Keppel, G. E. Collaborative Study of the Determination
raphy. J. Liq. Chromatogr. Relat. Technol. 1999, 22, of Dithiocarbamate. Residues by a Modified Carbon
463-475. Disulfide Evolution Method. J. Assoc. Off. Anal. Chem.
(5) Verma, B. C.; Sood, R. K.; Sharma, D. K.; Sidhu, H. S.; 1971, 54, 528-532.
Chauhan, S. Improved Spectrophotometric Method for (19) Bohrer, D.; Nascimento, P. C.; Gomes, H. M. Improve-
the Determination of Thiram Residues in Grains. Ana- ment in the Determination of Mancozeb Residues by the
lyst 1984, 109, 649-650. Carbon Disulfide Evolution Method Using Flow Injec-
(6) Malik, A. K. Determination of the Fungicide Ferbam in tion Analysis. J. Agric. Food Chem. 1999, 47, 212-216.
Wheat Grains after Adsorption onto Microcrystalline (20) Vuik, J.; van Dinter, R.; de Vos, R. H. Improved Sample
Naphthalene. J. AOAC Int. 2000, 83, 971-975. Pretreatment of the Carbon Disulfide Evolution Method
(7) Dogheim, S. M.; Alla, S. A. G.; El-Marsafy, A. M.; for the Determination of Dithiocarbamate Residues in
Fahmy, S. Monitoring Pesticide Residues in Egyptian Lettuce. J. Agric. Food Chem. 1992, 40, 604-606.
Fruits and Vegetables in 1995. J. AOAC Int. 1999, 82, (21) Thier, H. P., Zeumer, H., Eds. Dithiocarbamate and
948-955. Thiuram Disulfide Fungicides. In Manual of Pesticide
(8) Ripley, B. D.; Lissemore, L. I.; Leisheman, P. D.; Residues Analysis; Pesticide Commission: Germany,
Denommé, M. A. Pesticide Residues on Fruits and 1987; Vol. 1, 432 p.
Vegetables from Ontario, Canada, 1991-1995. J. AOAC (22) Cunniff, P., Ed. Official Methods of Analysis; AOAC
Int. 2000, 83, 196-213. International: Arlington, VA, 1995; Chapter 7.
(9) Sadlo, S. Quantitative Relationship of Application Rate
and Pesticide Residues in Greenhouse Tomatoes. J.
AOAC Int. 2000, 83, 214-219. Received for review January 29, 2001. Revised manuscript
(10) ANVISA. Agência Nacional de Vigilância Sanitária. received June 15, 2001. Accepted June 17, 2001.
Toxicologia (cited in 2000), available at URL http://
www.anvisa.gov.br. JF010124A