Decontamination of Pesticide Residues on Fruits and Vegetables

Md. Wasim Aktar asked:


Preferably, RAC samples used in processing studies should contain field treated quantifiable residues as close as possible to the MRL, so that measurable residues are obtained, and transfer factors for the various processed commodities can be determined. A transfer factor gives the ratio of the residue concentration in the processed commodity to that in the RAC. For example if the residue concentration is 0.5 mg/kg in olives and 0.2 mg/kg in olive oil, the transfer factor is 0.2/0.5=0.4. A factor 1 (= concentration factor) indicates a concentration effect of the processing procedures. Enhancing the residues either by increasing the application rates, shortening the pre-harvest interval (PHI) or spiking the RAC with the active ingredient and its metabolites in vitro is not, as and rule, desirable. Spiking is only acceptable if the RAC residues can be shown to consist only of surface residues. However, in some cases, especially where residues in the RAC are close to the analytical limit of determination, field treatment at exaggerated rates or shortened PHIs is advisable to obtain sufficient residue levels for the processing studies.

The first step in household or commercial food processing is the preparation of food using various mechanical processes, such as removing damaged or soiled items or parts of crops, washing, peeling, trimming or hulling. This often leads to significant declines in the amount of pesticide residues in the remaining edible portions (Petersen et al., 1996; Celik et al., 1995; Schattenberg et al., 1996).

WASHING

Household washing procedures are normally carried out with running or standing water at moderate temperatures. Detergents, chlorine or ozone can be added to the wash water to improve the effectiveness of the washing procedure (Ong et al., 1996). If necessary, several washing steps can be conducted consequently.

The effects depend on the physiochemical properties of the pesticides, such as water solubility, hydrolytic rate constant, volatility and octanol-water partition coefficient (Pow), in conjunction with the actual physical location of the residues; washing processes lead to reduction of hydrophilic residues which are located on the surface of the crops. In addition, the temperature of the washing water and the type of washing has an influence on the residue level. As pointed out by Holland et al. (1994), hot washing and the addition of detergents are more effective than cold water washing. Washing coupled with gentle rubbing by hand under tap water for 1 min dislodges pesticide residues significantly (Barooah and Yein, 1996). Systemic and lipophilic pesticide residues are not removed significantly by washing.

Table (1) shows examples of the effects of washing on the residue levels of different pesticides applied to fruits and vegetables.

PEELING

The outer leaves of vegetables often contain residues of pesticides applied during the growing season. Therefore, peeling or trimming procedures reduce the residues levels in leafy vegetables. Peeling of root, tuber and bulb vegetables with a knife is common household practice. Many examples show that most of the residues concentration is located in or on the peel. Peeling of the RACs may remove more than 50% of the pesticide residues present in the commodity. Thus, removal of the peel achieves almost complete removal of residues, so leaving little in the edible portions. This is especially important for fruits which are not eaten with their peels, such as bananas or citrus fruits. Reynolds (1996) showed that peeling or trimming of carrot reduced the residues of chlorfenvinphos, primiphos-methyl, quinalphos, triazophos resulting a transfer factor of 0.2. However, the peel from commercial peeling processes can be used as animal feed or for the production of essential oils (citrus) or pectin (citrus, apple etc.). For such industrial processes, it is important to realize that especially non-systemic surface residues are often concentrated in the peel. For systemic pesticides, peeling may not be as effective as shown by Sheikhorgan et al (1994). After application of thiometon on cucumbers, no reduction of residue levels could be detected in the peeled cucumbers.

Under the Codex Alimentarius, as in other international standards, MRLs refer to the whole fruits, which is appropriate for assessing compliance with GAP. These MRLs are of limited significance, however, in assessing dietary exposure to pesticides from fresh fruits, which are peeled (Holland et al.,1994).

COOKING

Cooking procedures at different temperatures, the duration of the process, the amount of water or food additives, and the type of system (open or closed) may have an impact on the residue level. Normally, residues are reduced during the cooking process by volatilization in open systems or by hydrolysis in closed systems. In any case, adding cooking liquid dilutes the residues. Several studies were reported on the dissipation of pesticides in crops during cooking. In addition to the studies summarized in table 1 the behavior of the organophosphorus pesticides chlorfenvinphos, fenitrpothion, isoxathion, methidathion and prothiophos during cooking was examined by Nagayama (1996) with green tea leaves, spinach and fruits. These pesticides decreased during the cooking process corresponding to the boiling time. According to their water solubility, some pesticides were translocated from the raw materials into the cooking water. On the other hand, the pesticide remained in the processed food according to their octanol-water partition coefficient, which is an indicator of hydrophilic or lipophilic properties of the compound. In exceptional cases, cooking processes may cause pesticide degradation, yielding a reaction product of toxicological significance. For e.g., daminozide is degraded to UDMH (1, 1-dimethylhydrazine), which is much more potent than the parent compound (Leparulo-Lofus et al.,1992). Another example is the formation of ETU (Ethylenethiourea) from EBDCs (Ethylene bisdithiocarbamate) fungicides like mancozeb, during heating processes (Petersen et al., 1996).

Dipping in chemical solution

Sodium chloride solution is largely used to decontaminate the pesticide residues from different fruits and vegetables .there are several studies to prove the efficacy of salt water washing to dislodge the pesticides from crops. In this process, sample of chopped fruits and vegetables is put in a beaker containing 5% sodium chloride solution. After 15 minutes the plant samples are gently rubbed by hand in salt solution and alt water is decanted. The examples of the effect of salt solution treatment on the residue levels of different pesticides applied to vegetables have been shown in table 1.

Kumar et al (2000) reported that dipping of green chillies in 2% salt solution for 10 minute followed by water wash prove to be effective, facilitating the removal of 32.56 and 84.21% residues correspondingly at 0 and 5 days after spray of triazophos (700g a.i./ha) while the acephate residues were removed to an extent of 78.95% at zero day. Following same technique Kumar et al (2000) observed the 90.56 and66.93% reduction correspondingly on 0 and 5 days after spraying of cypermethrin in chillies.

Dip treatment of fruits in NaCl solution, HCl, acetic acid, NaOH solution, potassium permanganate removed 50-60% of surface residues of synthetic pyrethroids compared to 40-50% removal by hydrolytic degradation with NaOH (Awasthi, 1986b).

Water solution of NaOH, acetic acid potassium dichromate and soap solution used as decontaminating agents for tom ………….

The treatment of fruits with 2% tamarind solution dip for 5 minute followed by tap water wash and steam cooking for 10 min. was found to remove the residues of monocrotophos, carbaryl and fenvalerate to an extent of 41.81, 100 and 100% respectively. Treatment with 2% salt solution was equally effective.

Dip treatments of the brinjal fruit wioth water, sodium chloride, HCl solution, acetic acid solution or potassium permanganate solution were all found to remove 30-33% of the residues of fenvalerate, permethrin, cypermethrin and deltamethrin; NaOH solution 40-45% and Teepol (a detergent) solution 50-60%. The effect of washing in reducing the residues decreased progressively at the second and third harvests.

Many experiments were carried out with the three common household preparations viz. washing with water, salt water washing and cooking to ***** their relative efficiencies in reducing the pesticide residues in different vegetables. The results have been summarized in the following table.

Table: Effect of washing, salt water washing and cooking on pesticide residue levels.

Crop Pesticide % of Residue dislodged * Result Reference

Washing with

water Salt water washing Cooking

Cauliflower Methamidophos 41-48 46-47 46.94

-53.54 Largest reduction was brought about by cooking. Jacob and Verma (1990)

Okra

Methamidophos

64-72

19-58

58-64 Washing with water could remove maximum residues indicating its maximum solubility in water though all the processes lower down the TMRL values. Jacob and Verma (1990)

Cauliflower Alpha-cypermethrin

7-38 _

12-17 Washing was found to be more efficient than cooking probably due to the thermal stability of cypermethrin. Malik et al (1997)

Cabbage

Chlorpyriphos

Quinalphos

38

41

52.13

56.50

54.3

55 With the three processes residues were reduced to some extent. They can not reduce the residue below the MRL. Thus a waiting period of a minimum of one and two weeks, respectively, was suggested irrespective of washing cooking for quinalphos and chlorpyriphos on cabbage. Nagesh and Verma (1997)

Cow pea

Metasystox

Carbalyl

84.3

87.5

86.4

88.7

83.4

80.8 Only boiling of the pod samples could decontaminate the residues present of surface or inside the tissue to the extent of safe limits by 10th day of treatment. Dikshit et al (1984)

Cauliflower

Malathion

60

70

80 Cooking was found to be most effective and lowered the TMRL value from one week to zero days. Jacob and Verma (1989)

Bhindi

Quinalphos

61.84-64.35

43-53

78-82

Both washing with water and salt water washing brought down the residues below the MRL at zero days, cooking also did this resulting maximum reduction of residues.

Jacob and

Verma (1985)

Cabbage

Malathion

Carbaryl
Pyrethroids

64.60

75.40

22.06 (av.)

-

-

-



83.97

89.62

56.72 (av.)

The extent of decontamination was higher due to cooking compared to washing for all insecticides.

Bhatia and

Verma (1994)

Leaves and curds of cauliflower heads of cabbage and pods of Indian colza

Green beans

Methamidophos

DDT

Malathion

Carbaryl



65.71-77.67

71

96

52



-

-

-

-

80-88.88

52(cooked)

66 (pressure cooked)

99(cooked)

99(p.cooked)

77cooked

69(p.cooked)



Cooking dislodges maximum residues.

Water wash removed maximum DDT residues whereas cooking is effective to remove malathion and carbaryl residues.

Dikshit et al (1986)

Elkins et al (1968)

From the above table it can be said that cooking is most effective to reduce the residues of different pesticides from various vegetables though in some cases washing with water was found to be effective to reduce the initial residues of pesticides and it has been found that with the ageing of residues or with the increase in the sampling days over treatments the effect of washing decreases to remove the toxicant to the same extent as that of samples collected immediately after spray where boiling or cooking is found to be effective. One of the possible reason for high percentage of removal of toxicant from immediately collected samples as most of the residues are present of the surface of the samples and hence it is very easy to remove by simple washing as observed by Dikshit et al (1984,86) Elkins et al (1968), Bhatia and Verma (1994) and Malik et al (1998). With the time elapsed the residues are migrated inside the deeper tissues or strongly adhere on the rough surface of some vegetables. Moreover, the washing cannot reduce the residues to the safe level as compared to boiling.

There are some studies where all the three culinary processes proved to be inefficient to reduce the residues below the MRL value. According to Jacob and Verma (1991) residues of quinalphos in the treated cauliflower crop would be reduced only to some extent by various home processing methods like washing and cooking. Nagesh and Verma (1997) opined that the inefficiency of the home processes for decontaminating the treated cabbage might be due to the strong adsorption properties of quinalphos and chlorpyriphos.

Effect of household preparation for decontamination of pesticide multiresidues in fruits and vegetables

Low levels of pesticide residues were detected in 97(40%) of mt 243 samples analyzed after following normal household washing, peeling and cooking procedures. The number of samples containing detectable residues dropped to 47(19%) after household preparation. These results indicate that residue level in most commodities are substantially reduced after household preparation (Schattenberg et al., 1996)

Ramesh and Balasubramanian (1999) performed a study with fruits and vegetables collected from Chennai local markets and fortified with known concentrations of various pesticides followed by decontamination study with different household preparations like washing, cooking , peeling resulting 65-95% decontamination of pesticide residues at different stages of 512 raw market samples analyzed, the organochlorine and organophosphorus pesticides present in the 12 samples were removed resulting in residues well below the toxicologically acceptable limits.

A short rinse in tap water reduces pesticide residues on many types of produce (Krol et al., 2000). Rinsing removed residues for nine of the twelve pesticides studied. Among captan, chlorothalonil, iprodione, vinclozolin, endosulfan, permethrin, methoxichlor, malathion, diazinon, chlorpyriphos, bifenthrin and DDE; residues of vinclozolin, bifenthrin and chlorpyriphos were not removed. This study confirms that the water solubility of pesticides does not play a significant role in the observed decrease. The majority of pesticide residues appear to reside on the surface of produce where it is removed by the mechanical action of rinsing.

Earlier studies of the effects of commercial and home preparation on pesticide residue in fruits and vegetables were summarized by Zabik (1987). The early studies showed residue reduction to be substantial, with percentage reduction of chlorinated hydrocarbons ranking from 50 to 99+ % for commercial preparation and from 14 to 99+ % for home preparation with the exception of parathion in spinach and broccoli, commercvial and home prewparation substantially reduced organophosphate residues, with the reduction generally being in the high 80 or 90% range. Carbamate residues were reduced by 58 to 99+ % when the vegetables were commercially processed but only by 11 to 92% in home preparation.

A recent study in Korea supports these earlier studies (Lee and Lee, 1997). These authors found that 45% of the organophosphate residues were eliminated when the foods were washed in water, 56% with detergent washing, 91% with peeling, and 51% with blanching or boiling.

Methods of multiresidue analysis of pesticides in fruits and vegetables

Analysis by gas chromatography

Nakamura et al (1994) developed a method for multiresidue analysis of 48 pesticides (20 organophosphorus, 7 organochlorine, 14 organonitrogen and 7 pyrethroid pesticides ) permitted in Japan on the basis of capillary GC after extracting the pesticides with nacetone from vegetable and fruit samples or with acetonitrile from lipid containing crops followed by reextraction into ethyl acetate (test solution). Organophosphorus pesticides were directly determined by GC-FPD. Organonitrogen pesticides were determined by GC-FTD (GC-NPD) following clean up by silica gel chromatography. Organochlorine and pyrethroid pesticides were measured by GC-ECD after clean up by florisil column chromatography. Recoveries for ten crops at fortification levels of 0.05-0.25 ppm were 42.5-128.5%. the detection limits were 0.001 ppm for organophosphorus and organochlorine pesticides and 0.01 ppm for organonitrogen and pyrethroid pesticides.

A multiresidue method was used by Dejonckheere et al (1996) for determination of organochlorine, organophosphorus and organonitrogen pesticides in vegetables and fruits which were extracted with acetone followed by liquid-liquid partitioning with water:apolar pesticides in petroleum ether phase, polar pesticides extracted from aqueous layer with dichloromethane and analyzed by gas chromatography with electron capture (GC-ECD), flame photometric (GC-FPD) and thermoionic specific (GC-TSD) detection.

The method used for multiresidue determination of 52 pesticides including organophosphorus, organochlorine, organonitrogen, certain pyrethroids and dithiocarbamate pesticides in vegetables and fruits was described by Dogheim et al (1999) utilizing gas chromatography. Samples were extracted with acetone followed by partitioning with hexane and dichloromethane and estimated by GC-ECD and GC-NPD. Dithiocarbamates were digested in mixture of concentrated HCl, SnCl2 and water for evolution of CS2 which is collected in an ethanolic solution of copper acetate and diethanolamine to form a yellow complex. The absorbance of yellow product was determined spectrophotometrically at 435 nm. The average recoveries and CVs of the 52 pesticides were 72-118 and 1-20%, respectively at the spiking levels of 0.01-1 ppm. A similar kind of method was also described by Kole et al (1998).

Krol et al (2000) used a multiresidue procedure for determination of 12 pesticides in vegetables where samples were extracted with 2 propanol and petroleum ether followed by washing with distilled water 3 times. Final analysis of the samples was performed by GC-ECD, FPD, XSD and/or ELCD.

Ramesah and Balasubramanian (1999) described a method to determine organochlorine, organonitrogen and organophosphorus pesticides in vegetables and fruits following extraction with 2-propanol and petroleum ether by mechanical shaker followed by partitioning with distilled water and column cleanup over florisil for OC and OP pesticides. For organonitrogen pesticides the extraction was done with acetone followed by partitioning with 10%NaCl and ethyl acetate and column clean up over silica gel. organochlorine, organophosphorus and organonitrogen compounds were analyzed by GC-ECD,GC-FPD and GC-NPD, respectively.

Using GC-ECD, the efficiencies of acetonitrile and acetone to extract the 8 pyrethroids from 6 fruits and vegetable samples were compared by Pang et al (1997). The extraction efficiency of acetone was competitive with that of acetonitrile for the 6 fruit and vegetable samples. The ruggedness tests demonstrated further that the proposed method is simple, accurate with good precision and suitable for multiresidue analysis of pyrethroid in various agricultural products.

Organophosphorus and organochlorine pesticide residues from fruit and vegetables by capillary GC with electron capture detector (ECD), nitrogen phosphorus detector (NPD), flame photometric detector (FPD) in the sulfur and phosphorus modes, and mass spectrometry detector (MSD) in selected ion monitoring (SIM) mode were determined by Torres et al (1995) following extraction by Matrix Solid Phase Dispersion (MSPD) resulting recoveries of 41-108% with relative SD of 2-14% in the conc. range 0.5-10 µg/liter in oranges, lemons, grapefruit, pears, plums, lettuces and tomatoes.

A multiresidue method as described by Sannino et al (1995) for quantitative determination of 39 organophosphorus compounds (parent pesticides and their major metabolites) in 7 fatty processed foods based on automated gel permeation chromatography with a Biobeads SX3 column and a methylene chloride-cyclohexane (15 + 85) eluant after extraction with methylene chloride. Organophosphorus compounds are quantitated by GC-FPD using OV-1701 and DB-5 columns. Average recoveries from samples fortified at 0.025-1 mg/kg ranged from 50.6% for dichlorvos to 185% for malaoxon. Determination limits were between 0.005 and 0.040 mug/mL. Results were confirmed by gas chromatography/mass spectrometry with selected-ion monitoring.

Gas chromatographic conditions for separation and identification of the compounds were selected using two capillary columns of different polarities and two detectors, ECD and NPD for multiresidue quantitative determination of 37 pesticides in fruit and vegetables and to study the efficiency of gel-permeation chromatography clean-up after ethyl acetate extraction (Balinova,1999).

Trova et al (1999) performed liquid chromatographic determination of pesticide residues (including azinphos-ethyl, azinphos-methyl, carbaryl, diflubenzuron, dinocap and teflubenzuron) in vegetables after extraction using an ethyl acetate/n-hexane solvent system instead of the widely employed methylene chloride. Recoveries as required by ‘Guidelines for residues monitoring in the European Union’ were observed; the new solvent system may be considered as an alternative to halogenated compounds, dangerous for their toxicity and harmful for their environmental behaviors, in extraction of HPLC-determinable active compounds.

A wide range screening method was proposed by Gelsomino et al (1997) for multiresidue analysis of 77 pesticides (12 organohalogens, 45 organonitrogens, 11 organophosphorus and 9 pyrethroids) in agricultural products using gas chromatography equipped with long, narrow-bore fused-silica open-tubular columns and electron-capture detector (ECD). Residues were extracted with acetone followed by dichloromethane partitioning and gel permeation chromatographic clean up. Recoveries of the majority of pesticides from spiked samples of carrot, melon and tomato at fortification levels of 0.04-0.10 mg/kg were 70-108%. Limits of detection were less than 0.01 mg/kg for ECD.

Beena et al (2002, 2003) carried out monitoring of vegetable samples adopting a multiresidue analytical technique employing GC-ECD and GC-NPD systems with capillary columns.

Ueno et al (2003) studied an efficient and reliable multiresidue method for determining 52 nitrogen- and/or phosphorus- containing pesticide residues in a large number of vegetable samples in which samples were extracted with acetonitrile, and the separated acetonitrile layer was purified by gel permeation chromatography that divided the pesticide eluate into 2 fractoions, the pesticide fractions were respectively purified by a 2-step minicolumn cleanup, the second fraction through silica gel minicolumn; first fraction through the tandem minicolumn (florisil minicolumn, inserted on silica gel minicolumn) which was eluted with acetone-petroleum ether (3+7). The combined eluate was subjected to dual column gas chromatography with nitrogen-phosphorus and flame photometric detection. Recoveries of 52 pesticides from fortified samples ranged from 72 to 108% with relative standard deviations of 2-17%, except for the recoveries of methamidophos and chorothalonil. The detection limits of the pesticides were satisfactory (0.001-0.009 mg/kg) for monitoring of pesticide residues in vegetables.

Menkissoglu et al (2004) performed a study of the matrix induced effect for 16 common pesticides, most frequently found in monitoring studies in tomato pepper and cucumber, using a simple multiresidue method with GC-ECD or NPD, without a previous cleanup step. Anomalously high GC responses and subsequently very high recoveries for several pesticides in the extracts were obtained by a conventional calibration with pesticide solution in ethyl acetate.

A faster, less effective, environmentally safer supercritical fluid extraction (SFE) method was evaluated by Garcia et al (1996) over conventional sonvent extraction methods for the extraction of imidacloprid, methiocarb, chlorpyrifos, chlorothalonil, endosulfan-1, endosulfan-2 and endosulfan sulfate, from pepper and tomato using vegetable sample: anhydrous magnesium sulfate (5:7) mixtures to carryout the extraction with supercritical CO2 and HPLC/DAD,GC/ECD and GC/FPD for analysis. The chosen SFE conditions were 300 atm, 500C, 200?l of methanol static modifier, 1 minute static time, and dynamic extraction with 15 ml of CO2 and collection in 3 ml of ethyl acetate. Except for imidacloprid, which was not recovered under any of the assessed conditions, pesticide recoveries were greater than 80%.

A simplified method is described by Chaput (1987) where reverse phase liquid chromatography was utilized with post column derivatisation and fluorescence detector to determine 7 N-methyl carbamates (aldicarb, carbaryl, carbofuran, methiocarb, methomyl, oxamyl and propoxur) and 3 related metabolites in fruits and vegetables after extraction of the sample with methanol followed by gel permeation chromatography (GPC) or GPC with on-line Nuclear-celite clean up for crops with high chlorophyll and/or carotene content (e.g. cabbage and broccoli). Recovery data were obtained by fortifying 5 different crops (apples, broccoli, cabbage, cauliflower and potatoes) at 0.05 and 0.5 ppm. Recoveries averaged 93% at both fortification levels. The coefficient of variation of the method at both levels is