ethyl formate fumigation as a replacement for methyl

Ethyl formate fumigation as a replacement for methyl bromide fumigation in dried vine fruit. Dried Fruit Research and Development Council, Project CSH 64D

C.R. Tarr1, R. Reuss2, P.R.Clingelefer1, P. Annis2, 1CSIRO Plant Industry 2Stored Grain Research Laboratory Horticulture Unit CSIRO Entomology Private Mail Bag GPO Box 1700 Merbein Vic. 3505 Canberra ACT 2601 AUSTRALIA AUSTRALIA Ph (03) 50513100 Fax (03) 50513111 Email [email protected] Purpose To develop effective and safe methods to fumigate cartons and bulk bins of dried vine fruit with ethyl formate. This includes :- • investigation of the penetration efficiency of ethyl formate into large masses of dried vine

fruit. • exposure to ethyl formate of cultured ‘wild’ insect populations of major pests to identify

effective dosage and identify ‘tolerance’ compared to unexposed populations. • consult with packers re the development of suitable fumigation protocols, • develop and demonstrate this technology to industry.

Acknowledgement DFRDC $65,776 CSIRO $36,377 Report finalised April 2004

Disclaimer All use of the information in the report shall be entirely at the risk of the recipient. CSIRO and DFRDC, their officers, employees and agents accept no responsibility for any person acting on or relying upon any opinion, advice, representation, statement, or information contained in the report and disclaim all liability for loss, damage, cost or expense incurred or arising by any reason of any person using or relying on the information contained in the report or by reason of any error, omission, defect or mis-statement. 1

mailto:[email protected]

Table of Contents Media Summary .....................................................................................................................................3 Technical Summary ...............................................................................................................................4 Introduction ............................................................................................................................................5 Methodology ...........................................................................................................................................6 Section 1: Investigations into penetration of ethyl formate into unprocessed and processed sultanas .................................................................................................................................................................7

Relative Humidity of sultanas .............................................................................................................9 Bulk density of sultanas ....................................................................................................................10 Natural levels of ethyl formate, other volatiles and accumulation of carbon dioxide ......................12 Fate of ethyl formate applied to sultanas ..........................................................................................16 Movement of ethyl formate through columns of sultanas.................................................................18

Section 2: Mortality studies of pest populations sourced from Sunraysia to fumigation with ethyl formate and comparisons with an unexposed population to identify any development of ‘tolerance’ to the fumigant......................................................................................................................................25

Materials and Methods ......................................................................................................................25 Results and Discussion ......................................................................................................................28

Section 3: Ethyl formate fumigation as a replacement for methyl bromide fumigation in dried vine fruit........................................................................................................................................................39

Trial 1. Ethyl formate fumigation of a shipping container load of unprocessed dried sultanas in June 2000...........................................................................................................................................39 Trial 2: A combined ethyl formate and CO2 fumigation of a shipping container of unprocessed dried sultanas in April 2001. .............................................................................................................41 Discussion .........................................................................................................................................44

Section 4 : Draft method of application of Ethyl formate (eranol) to shipping containers to fumigate dried vine fruit.......................................................................................................................45 Technology Transfer ............................................................................................................................50

Bibliography ......................................................................................................................................50

Recommendations ................................................................................................................................52

2

Media Summary Methyl bromide, a scheduled ozone depleter, has been used for fumigation of many food commodities. However, its use is being phased out as part of the international Montreal protocol. This leaves many industries including the dried vine fruit (dvf ) industry with a need to develop alternative treatment strategies.

Previous dried vine fruit research has highlighted several alternatives, including the use of ethyl formate as a space fumigant. However, ethyl formate may be explosive under certain conditions and research on dosage levels and application methods for space fumigation needed to be undertaken. This project examined :- • the penetration efficiency of ethyl formate into dried vine fruit • dosages needed for 100% mortality of dried vine fruit pests in space fumigation • whether tolerance to ethyl formate has developed in the Sunraysia insect populations • the mechanics of fumigation in large spaces • development of a method for fumigation of dvf in a shipping container sized space. There was very little adsorption of ethyl formate by the fruit, a good characteristic for fumigation, but it was found to move slowly through compact dried vine fruit. This slow dispersion was alleviated by including it in a carrier gas of CO2, a concept developed during the project. Laboratory studies showed dried vine fruit pests were susceptible to dosages of between 20-40gm-3 of ethyl formate. An effective dosage, after allowing for sorption, of 40gm-3 should provide acceptable mortality of dried vine fruit pests. After more than 50 years treatment in the Sunraysia district there was no evidence of tolerance to ethyl formate having developed in pest populations. Two fumigations with ethyl formate were carried out in shipping containers. The first used liquid ethyl formate and demonstrated that this fumigant worked in cold conditions. The second demonstration used CO2 as a carrier gas, thereby reducing flammability and improving distribution of ethyl formate. These fumigations produced 99.9% and 100% mortality on complete lifecycles of all dried vine fruit pests included. It is proposed that industry consider the potential to adopt ethyl formate as a space fumigant as an alternative to Methyl Bromide. A draft application protocol has been developed to assist industry in giving consideration to this possibility.

3

Technical Summary Methyl bromide, a scheduled ozone depleter, is used for fumigation of dried vine fruit (dvf). Its availability for use on products other than for quarantine reasons is being phased out by 2005 as part of the international Montreal protocol. This leaves the dvf industry with a need to replace methyl bromide with alternative treatment strategies.

Previous research has highlighted several alternatives including the use of ethyl formate as a space fumigant. Ethyl formate is registered for use in dried fruit in small packages (approx 15kg). It has proven effective on the major insects present in the Australian industry. Research has shown it leaves no discernable residues in the fruit 7 days after treatment. Ethyl formate has a high daily exposure level due to it’s low human toxicity (TLV 100 ppm) and may offer a suitable replacement strategy in dried vine fruit for methyl bromide. However, ethyl formate is explosive under certain conditions and research on dosage levels and application methods for space fumigation needed to be undertaken. This project examined :- • the penetration efficiency of ethyl formate into dried vine fruit • dosages needed for 100% mortality of dvf pests in space fumigation • whether tolerance to ethyl formate has developed in the Sunraysia insect populations • the mechanics of fumigation in large spaces • development of a method for fumigation of dvf in a shipping container sized space. There was very little adsorption of ethyl formate by the fruit when compared to grains, a good characteristic for fumigation (Desmarcheliar, Johnston and Le Trang Vu 1999, Hilton and Banks 1997). However it was found to move slowly through compact dried vine fruit. This slow dispersion was found to be alleviated by including it in a carrier gas of CO2 a concept application method developed during the project (Allen, S.E., Desmarchelier J.M., 2002). Laboratory toxicity studies showed dvf pests were susceptible to dosages of between 20-40gm-3 of ethyl formate. The same studies showed no evidence of tolerance to ethyl formate having developed in the Sunraysia pest populations. Further research on insect toxicity using the most resistant lifecycle stages of dvf pests should be carried out. An effective dosage after allowing for sorption onto fruit of 40gm-3 should provide acceptable mortality of dvf pests. A draft application protocol is included in this report. Two fumigations with ethyl formate were carried out in shipping containers. The first used liquid ethyl formate and demonstrated that this fumigant will work in cold conditions (average 11oC) at the lower temperature limit for methyl bromide application. The second demonstration used CO2 as a carrier gas, thereby reducing flammability and improving distribution of ethyl formate. The first fumigation killed all stages of O. surinamensis and all stages of T. confusum with the exception of 2 pupae from a total of 3000 test insects. The second fumigation produced 100% mortality of the complete lifecycle of T. confusum and pupae, larvae and adults of P. interpunctella as well as O. surinamensis present in the fruit. It is proposed that industry investigate adoption of ethyl formate as a space fumigant. The use of a proprietary CO2/ ethyl formate gas mix for fumigation should be investigated as this new development will significantly increase the safety of application and make commercial use a reality.

4

Introduction

The phase out of methyl bromide for use as a fumigant of products other than those requiring fumigation for quarantine reasons is rapidly approaching (2005). This will leave the dried fruit industry with a need for alternative treatment strategies to replace methyl bromide. The final reports from the DFRDC funded project CSH 34 and 44 presented several alternative treatment strategies to replace and reduce the need for methyl bromide fumigation in the control of insect infestations in dried vine fruit.

As part of these strategies, the use of ethyl formate as an alternative space fumigant was highlighted

as having potential. Ethyl formate is at present registered for use in packaged dried fruit where it is used as an insecticide for the final disinfestation of the product during packaging. It has proven to be effective on the major insects present in the Australian industry (Hilton and Banks 1997, Tarr and Clingeleffer 1993), and leaves no discernable residues in the fruit 7 days after treatment (unpublished internal CSIRO report). Ethyl formate has a high daily exposure level due to it’s low human toxicity (TLV 100ppm) and may be suitable as a replacement in dried fruit for methyl bromide when used in combination with other alternative strategies.

A report on ethyl formate highlighted its explosive nature at concentrations between 2.8-16% in air, if a source of combustion is provided (Pearson, R.D., Apte, V.B., 1998). However safe application to larger enclosures has been proposed and demonstrations carried out by grain researcher Dr. J. Desmarchelier at CSIRO Entomology, with the application of ethyl formate in water as both a fumigant and surface spray treatment for grain. Desmarchelier demonstrated in initial experimentation that ethyl formate like ethanol is not flammable when mixed with water (Desmarchelier, et al. 1998).

The potential application of ethyl formate as a space fumigant leads to questions about its ability to penetrate through large bulks of fruit. Hence, a method of application to bulk sultanas as a fumigant needs to be developed and its effective penetration tested prior to industry recommendations being made. Research is being carried out at CSIRO Entomology into the application of ethyl formate to grains. As a part of this project CSIRO Entomology and Plant Industry investigated its application for dvf.

Objectives • To develop efficacious and safe application methods for the fumigation of dried vine fruit in large

carton or bulk bin stacks using ethyl formate. • Investigate the penetration efficiency of ethyl formate into large masses of dried vine fruit, ie.

bulk bins, stacks and palletised carton stacks. • To consult with packers regarding the development of suitable fumigation protocols, then develop

and demonstrate this technology to industry.

5

Methodology Laboratory studies were conducted to test the efficacy of ethyl formate penetration into large masses of sultanas, e.g. bulk bins and palletised cartons. The studies were carried out on both processed and unprocessed fruit. The penetration of ethyl formate was measured using a column of fruit to simulate bulk dried fruit. This was carried out at CSIRO Entomology’s, Stored Grain Laboratory in Canberra. See section 1 for more detail. An insect population, obtained from the Sunraysia district, was cultured until large enough populations were raised for use in laboratory mortality trials and experimental space fumigation treatments with ethyl formate. Insects were trapped during the autumn of 1999 and cultured to obtain sufficient numbers by spring of 1999. The major pests Plodia interpunctella, Oryzaephilus surinamensis and O. mercator were cultured, plus other pests where enough adults were captured to establish a viable population. These cultures were used to prove the efficacy of the fumigant on local insect populations and to pinpoint resistance (tolerance) which may have developed in local populations. See section 2 for more detail. Application methodologies for fumigation were developed and demonstrations of fumigation methods using ethyl formate in shipping containers were scheduled. The methods demonstrated the use of ethyl formate in a larger space equivalent to a fumigation tent and was timed to occur during the peak infestation periods of the season where possible. These methods took advantage of natural infestations for the demonstration of efficacy as well as cultured bioassays. See section 3. A draft application protocol was developed for the dried vine fruit industry. See section 4.

6

Section 1: Investigations into penetration of ethyl formate into unprocessed and processed sultanas Measurement of ethyl formate by gas chromatography

Background

Two different instruments and three different columns were used to separate headspace containing ethyl formate, ethanol and a variety of other compounds.

Results and conclusions

The Shimadzu GC 6AM fitted with a SP 2401/SE 30 glass column rapidly detected high levels of ethyl formate. Other compounds, such as ethanol, methanol and methyl formate could not be separated with this instrument. While the Shimadzu may be suited for measuring fumigation levels of ethyl formate in the absence of ethanol it is unsuited for low levels of the gas and in applications that contain other esters or alcohols. The Varian 3600 fitted with a 30 m DB-FFAP column can separate ethyl formate from ethanol and to some extent from the other components. However, identification of peaks by retention times is not easily achievable, as all peaks are eluted in less than two minutes. The column has some application where there is ethyl formate and Ethanol contained in samples but no (or very little) other esters and alcohols are present. Separation of ethyl formate and other esters, alcohols, ketones and aldehydes was best achieved with a 60 m AT-WAX column. Series of formates and alcohols were separated to the baseline (Figures 1.1 and 1.2). At 80°C compounds relevant to the ethyl formate work could be separated in less than 8 minutes. Separation was further improved by lowering oven temperature. At 125ºC analysis can be carried out in less than 5 minutes, with adequate separation of the most relevant compounds. The AT-WAX column is considered to be the column of choice for ethyl formate work. It is especially suitable for natural levels of ethyl formate and other compounds, studies on sorption and decay of the fumigant and any other application where methyl formate, methanol and ethanol as well as other compounds need to be separated from ethyl formate.

7

Figure 1.1. Separation of ester series (formates)

Figure 1.2. Separation of alcohol series

8

Relative Humidity of sultanas

Background

Moisture content of sultanas is determined by methods different to those commonly used for the moisture contents determination in grain at the SGRL. Present and future work in this area requires the use of these techniques to determine one of the vital parameters in sultana storage and for developing fumigant treatments for dried fruit.

Methods

Processed and unprocessed sultanas were sealed in 250 mL glass jars with a lid modified to allow relative humidity (rh) measurement with an electronic rh meter (accuracy ±1%, RS Relative Humidity Meter, calibrated with lithium chloride solution standard). The sultanas were then stored at four temperatures and rh was measured after a minimum of 24 hours of equilibration. The moisture content of the sultanas was estimated from a published isotherm (Figure 1.3).

y = 100x2 - 70x + 22

5

10

15

20

25

0.3 0.4 0.5 0.6 0.7 0.8 0.9

water activity

moi

stur

e co

nten

t %

Figure 1.3. Sultana isotherm. Source: Pixton and Warburton 1972.

Results

Processed sultanas had a higher water activity than unprocessed fruit (Figure 1.4). Increases in rh with storage temperature followed a linear relationship. The estimated moisture content of sultanas is shown in Table 1.1.

9

48

49

50

51

52

53

54

55

18 22 26 30 34 38 42 46

Storage temperature °C

rh %

Unprocessed sultanas Processed sultanas

Figure 1.4. Relative humidity of storage environment for sultanas stored at different temperatures Table 1.1. Measured rh and estimated moisture contents of sultanas Commodity Temperature ºC RH % Estimated moisture contents % Processed 18.5 51.2 12.4 26.0 52.1 12.7 33.6 52.7 12.9 44.0 53.6 13.2 Unprocessed 18.7 48.8 11.7 28.0 49.6 11.9 33.8 49.7 11.9 43.1 50.2 12.1

Bulk density of sultanas

Background

The compressibility of sultanas could play an important role in the way ethyl formate moves through a column of dried fruit.

Methods

The bulk density of unprocessed and processed sultanas was measured with a bulk density balance (manufactured by CBH). Processed and unprocessed sultanas were compressed by impact 20 times. After each compression, the bulk density was determined.

Results

Unprocessed sultanas had considerably less bulk density than processed sultanas (Table 1.2). Even after 20 impact compressions the bulk density of the commodity was still increasing slightly (Figure 1.5). Expressed as a percentage, bulk density increase by compression was more substantial in unprocessed compared to processed fruit (Figure 1.6). By avoiding compression of unprocessed

10

sultanas it should be possible to increase the ease of fumigant penetration through interstitial spaces into the bulk and to insure a more even distribution of fumigant. Table 1.2. Bulk density of processed and unprocessed sultanas. Type uncompressed (kg/hL) compressed (*5) (kg/hL) unprocessed 48.50 53.75 46.75 51.75 48.75 54.00 46.50 50.00

Average 47.75 52.50 processed 62.50 68.25 62.75 70.50 66.75 71.25 62.00 69.75

Average 63.50 70.00

455055606570758085

0 5 10 15 20

Number of compressions

Bul

k de

nsity

(kg/

hL)

unprocessed processed

Figure 1.5. Increases in bulk density (kg/hL) of processed and unprocessed sultanas as a result of progressive manual compression

11

0

5

10

15

20

25

30

35

0 5 10 15 20

Number of compressions

Incr

ease

in b

ulk

dens

ity %

unprocessed processed

Figure 1.6. Increases in bulk density (%) of processed and unprocessed sultanas as a result of progressive manual compression Natural levels of ethyl formate, other volatiles and accumulation of carbon dioxide

Background

Measurement of respiratory gases gives an estimate of biochemical activity in stored products. In turn, this influences the interaction between product and fumigant. Natural levels of fumigants and their products may occur in storage atmospheres. Information on biochemical reactions occurring in the untreated product forms part of developing an understanding of the fumigation process, quality changes and the likelihood of significant fumigant residues. As carbon dioxide is a likely end product of ethyl formate breakdown it is important to develop knowledge of the amount of gas produced by untreated sultanas.

Materials and Methods

Processed and unprocessed sultanas were sealed in flasks and stored at 25, 35, and 45°C. Ethyl formate and related compounds as well as CO2, O2 and CO were monitored over several weeks. Analysis of alcohols and esters Analysis of gas samples was carried out using a Varian 3600 CX GC with an AT-WAX column (60 m × 0.53 mm ID × 1.0 µm) and a FID. Chromatographic conditions used: carrier gas helium, pressure 6.5 psi, oven temperature 80ºC, detector temperature 275ºC, injector temperature 150ºC. Injection volume was 10 µL delivered with a gastight syringe. Peaks were identified by comparison to standard injections. Concentrations were calculated on the basis of peak areas from regression equations calculated from standards of various dilutions. Analysis of CO2, O2 and CO Analysis of gas samples was carried out using a Fisher model 1200 Gas Partitioner with 80-100 mesh Columpak™ PQ (6.5 ft x 1/8 in) and 60-80 mesh Molecular Sieve 13X (11 ft by 3/16 in) columns in series and a Thermal Conductivity Detector. Chromatographic conditions used: carrier gas, helium with a flow rate of 30 mL/min and oven temperature 50°C. Injection volume was 1 mL delivered with

12

a gastight syringe. CO2 and O2 concentrations were calculated based on adjusted peak areas. Peak areas were calibrated periodically using a standard gas mixture with known CO2, O2, CO and N2 composition. The limit of detection for CO2 was below atmospheric concentrations.

Results and discussion

CO2, O2 and CO Accumulation of CO2 and loss of O2 was temperature dependent and followed trends commonly seen in stored grain (Figures 1.7. and 1.8.). CO also accumulated in a manner similar to canola (Figure 1.9.). For all gases changes were more pronounced in the unprocessed sultanas. Unprocessed sultanas had a higher rh than processed fruit (about 2% higher) but this does not completely account for the differences. RQ increased with increasing storage temperature and was considerably below unity at lower storage temperatures (Table 1.3.). This suggested that respiration of carbohydrates was not the primary source of changes in carbon dioxide and oxygen. The source of CO accumulation should be investigated, but it can be assumed that the considerable levels of this gas measured are an indicator of ongoing oxidation reactions. The comparatively high impact of stored sultanas on storage atmospheres suggests a high level of oxidative activity in the product. The impact of this phenomenon on ethyl formate breakdown needs to be investigated in more detail.

0123456789

0 1 2 3 4 5storage time (weeks)

CO

2 %

v/v

Unprocessed 25°C Processed 25°C Unprocessed 35°CProcessed 35°C Unprocessed 45°C Processed 45°C

Figure 1.7. Changes in carbon dioxide concentrations in hermetically sealed sultanas

13

6

8

10

12

14

16

18

0 1 2 3 4 5

storage time (weeks)

O2

% v

/vUnprocessed 25°C Processed 25°C Unprocessed 35°CProcessed 35°C Unprocessed 45°C Processed 45°C

Figure 1.8. Oxygen consumption of hermetically sealed sultanas

0

1000

2000

3000

4000

5000

6000

0 1 2 3 4 5

Storage time (weeks)

CO

ppm

v/v

Unprocessed 25°C Processed 25°C Unprocessed 35°CProcessed 35°C Unprocessed 45°C Processed 45°C

Figure 1. 9. Carbon monoxide production of hermetically sealed sultanas

Table 1.3. Respiratory quotients (RQ) of sultanas after 5 weeks of storage

Temperature °C Material RQ after 5 weeks 25 Processed 0.3 Unprocessed 0.4

35 Processed 0.6 Unprocessed 0.5

45 Processed 0.7 Unprocessed 1.1

14

Volati le compounds The alcohols methanol and ethanol were present in higher concentrations than methyl and ethyl formate (Table 1.4.). Ethanol predominated at 25 and 35ºC but high levels of methanol were found at 45ºC. No clear trend for higher levels of these compounds in processed compared to unprocessed produce was apparent. On average, the levels of ethyl formate and the alcohols increased with increasing storage temperature. In processed sultanas methyl formate levels peaked after 3 weeks of storage at 25 and 35ºC. High levels were found after 1 week of storage at 45ºC, in processed as well as unprocessed sultanas. On average, more methyl formate was found in processed sultanas compared to unprocessed fruit. Ethyl formate levels seemed to be temperature dependent in processed sultanas, a trend that was less clear in unprocessed fruit. Methanol levels were almost always highest at 45ºC in processed and unprocessed sultanas. Concentrations peaked after 3 weeks storage. In contrast, ethanol concentrations in processed sultanas were lower with increasing storage temperature. However, there were high concentrations of ethanol in the headspace of unprocessed sultanas stored at 45ºC for 1 week. Table 1.4. Alcohol and esters levels in the headspace of sealed flasks containing sultanas. Figures are averages of measurements taken over 5 weeks. (MeF = methyl formate, Etf = ethyl formate, MeOH = methanol, EtOH = ethanol ) Temperature °C Commodity MeF mg/L EtF mg/L MeOH mg/L EtOH mg/L

25 processed 0.08 0.06 0.31 0.42 unprocessed 0.04 0.13 0.13 0.30 average 0.06 0.10 0.22 0.36



The emerging profile of volatile alcohols and esters in sultanas provides an interesting challenge for the analysis of low levels of ethyl formate on sultanas. Chromatographic separation of alcohols from ethyl formate is crucial to achieving realistic estimates of residual fumigant post fumigation. At 25 and 35ºC, natural levels of ethyl formate were far below levels considered safe for consumption (less than 0.3 mg/L). At 45ºC levels were about half of the safe level. This is unlikely to be due to previous treatments as the unprocessed sultanas had similar levels of organic ethyl formate.

15

Fate of ethyl formate applied to sultanas

Background

Understanding the breakdown or sorption of ethyl formate on dried fruit is crucial in developing fumigation techniques and understanding the risk of fumigant residues. Ethyl formate applied as liquid or vapour Materials and methods

120 g of processed fruit was sealed in flasks and ethyl formate was applied directly onto fruit (liquid) or onto a filter paper in the flask’s neck (vapour).

Results

Ethyl formate breakdown occurred quite rapidly. However, the fumigant was quite persistent in comparison with the rates of loss experienced when the fumigant is applied to cereal commodities such as rice and wheat. It may be that this fact is what makes ethyl formate particularly suited as a treatment for dried fruit. There were no substantial differences due to the method of application (Figure 1.10.), that is applying the fumigant directly to the commodity did not increase loss of vapour. From a sorption perspective there would be little advantage in vaporising the fumigant before application.

5 7 9

11 13 15 17 19 21 23

1 2 5 24 Exposure (hours)

EtF

ppm

v/v

applied as liquid applied as gas

Figure 1.10. Loss of ethyl formate(EtF) applied as gas or liquid to processed sultanas

16

Processed and unprocessed sultanas Materials and methods

250 g of processed and unprocessed sultanas were sealed in 500 mL flasks and dosed with 100 µL of ethyl formate. Headspace was monitored with Varian GC and AT-WAX column as described.

Results

Little ethyl formate was lost during the early stages of fumigation (Table 1.5.). Over 24 hours approximately half the fumigant was lost. There was no difference between unprocessed and processed fruit. Loss of fumigant can be estimated by a regression (Figure 1.11). The vapour of ethyl formate applied to fruit is persistent compared to application to cereal commodities. Consequently, some benefit may be derived by applying ethyl formate as a space type of fumigant to processed or unprocessed dried fruit stored in sealed containers. Further investigations are necessary to explore this aspect of dried fruit/ethyl formate interactions, paying particular attention to filling ratios, temperature and water activity and their effect on the persistence of ethyl formate vapour. Table 1.5. Percentage loss of ethyl formate in processed and unprocessed fruit over a 24 h period.

Ethyl formate loss % Exposure Time Minimum Maximum Average

3 1 13 7 4 8 14 10 24 45 55 48

y = -1.3335x + 63.83R2 = 0.9229

0

10

20

30

40

50

60

70

80

0 4 8 12 16 20 24

EtF

mg/

empt

y fla

sk

Processed Unprocessed

Exposure (hours)

Figure 1.11. Loss of ethyl formate (EtF) from flasks containing processed and unprocessed sultanas

17

Movement of ethyl formate through columns of sultanas Ethyl formate (eranol) applied to surface of processed and unprocessed fruit in small glass columns Background

To investigate the way ethyl formate penetrates through sultanas glass columns were tightly packed with processed and unprocessed fruit. To differentiate between vapour and combined vapour-liquid movement of fumigant ethyl formate was applied to a layer of glass beads (vapour) or directly to the fruit (vapour-liquid).

Materials and methods

Glass columns (ID 30mm, length 450 mm) with four sample ports (headspace, 50 mm, 160 mm, 270 mm) were filled with 60 g of glass beads followed by 120 g of processed or unprocessed sultanas. Eranol was then applied either to the surface of the fruit or on top of a 20 g layer of glass beads. The dose was 50 µL of ethyl formate (equivalent to 0.42 mL/kg, within lower range of label). Results

The majority of ethyl formate vapour remained in the headspace of the column for both methods of application and types of commodities (Figures 1.12. and 1.13.). Some penetration occurred at all levels of the column with the highest concentration at the distance closest to the point of application. Direct application led to higher headspace concentrations than application to glass beads. Ethyl formate vapour appeared to penetrate processed sultanas better than unprocessed sultanas. However the liquid-gas ethyl formate penetrated columns of both processed and unprocessed fruit more successfully, possibly by liquid seepage through interstitial spaces and a wick effect of sultanas exposed to high levels of liquid fumigant. From this data it appears that fumigant does not easily penetrate tightly packed columns of sultanas and that ethyl formate vapour does not penetrate as successfully as liquid-vapour mixtures occurring during direct application of the fumigant.

Figure 1.12. Movement of ethyl formate through columns of sultanas. Ethyl formate was applied to a layer of glass beads at 0.42 mL/kg (equivalent to 10 mL/25kg). Column contained 120 g of sultanas. Two replicates are shown. Numbers are averages of measurements taken over a 24 hour period.

18

Figure 1.13. Movement of ethyl formate through columns of sultanas. Ethyl formate was applied directly to fruit at 0.42 mL/kg (equivalent to 10 mL/25kg). Column contained 120 g of sultanas. Two replicates are shown. Numbers are averages of measurements taken over a 24 hour period. Ethyl formate (eranol) applied at different depths of a column containing processed sultanas

Background

To further investigate the way ethyl formate penetrates through sultanas, glass columns were tightly packed with processed and unprocessed fruit and fumigant was applied at different depth.


Glass columns (ID 30mm, length 450 mm) with four sample ports (headspace, 50 mm, 160 mm, 270 mm) were filled with 120 g of processed or unprocessed sultanas. Eranol was then applied at four different depths. The dose was 50 µL of ethyl formate (equivalent to 0.42 mL/kg, within lower range of label).

Results

When fumigant was applied to the surface most of the material remained in the headspace with some penetration 50 mm into the column. (Figure 1.14). When ethyl formate was applied 50 mm into the column a more even distribution of fumigant was found (Figure 1.15). Further into the column distribution was skewed towards the bottom of the column (Figures 1.16 and 1.17). There appears to be a clear advantage to applying the fumigant below the surface of the fruit to achieve a more even distribution of the chemical. However, a gradient skewed towards the top of the column will result. Application towards the centre to bottom of the column leads to a bottom heavy distribution. Clearly simultaneous application at several depths would be the ideal way to achieve even fumigant distribution.

19

050

100150200250300350400

1 4 24

Exposure time (hours)

EtF

mg/

L0 50 160 270

Figure 1.14. Penetration of ethyl formate through a column of sultanas. Fumigant applied directly to surface. Legend shows sample depth in mm.

050

100150200250300350400

1 4 24


EtF

mg/

L

0 50 160 270

Figure 1.15. Penetration of ethyl formate through a column of sultanas. Fumigant applied 50 mm from surface. Legend shows sample depth in mm .

20

050

100150200250300350400

1 4 24


EtF

mg/

L0 50 160 270

Figure 1.16. Penetration of ethyl formate through a column of sultanas. Fumigant applied 160 mm from surface. Legend shows sample depth in mm.

050

100150200250300350400

1 4 24


EtF

ppm

v/v

0 50 160 270

Figure 1.17. Penetration of ethyl formate through a column of sultanas. Fumigant applied 270 mm from surface. Legend shows sample depth in mm.

21

Ethyl formate movement through 85 mm columns of unprocessed and processed sultanas Background

Large PVC columns were filled with processed and unprocessed fruit. To investigate upward and downward movement of vapour, fumigant was applied to the centre of the column.


Four PVC columns (ID 85 mm, length 1 m, half-hole septa sample point every 50 mm) were loosely filled with 2.5 kg of unprocessed sultanas or 3.5 kg of processed sultanas to leave a headspace of 180 mm. Ethyl formate in the form of Eranol was applied according to label to the centre of the compressed column (injection point). Measurements were then taken at various distances above and below the injection point. Analysis was carried out using Varian GC with AT-WAX column as described previously, except that oven temperature was increased to 100ºC to shorten retention times.

Results

Distribution of ethyl formate vapour through a column of unprocessed fruit was much more even than through the processed commodity (Figures 1.18 and 1.19). Higher levels were found below the application point, probably due to the effect of gravity. Application of ethyl formate to the centre of a loosely packed column of unprocessed fruit seems to achieve a good distribution of the fumigant. It is likely that the greater ease of penetration of liquid and vapour ethyl formate through the less dense unprocessed fruit accounts for this effect.

0

20

40

60

80

100

120

140

160

180

350 250 150 -150 -250 -350 -400

Distance from application (mm)

Ethy

l For

mat

e m

g/L

Figure 1.18. Penetration of ethyl formate through a column of processed sultanas 6 h after application. Negative numbers show distance below application point, positive numbers show distances above application point.

22

0

20

40

60

80

100

120

140

160

180

350 250 150 -150 -250 -350 -450

Distance from application (mm)

Ethy

l For

mat

e m

g/L

Figure 1.19. Penetration of ethyl formate through a column of unprocessed sultanas 6 h after application. Negative number show distance below application point, positive numbers show distances above application point.

Effect of compression on movement of ethyl formate through 85 mm columns of unprocessed

sultanas

Background

Compression seems to play an important part in the penetration and distribution of ethyl formate liquid and vapour. To investigate this effect in unprocessed sultanas, columns were manually compressed to simulate compaction.


Four PVC columns (ID 85 mm, length 1 m, half-hole septa sample point every 50 mm) loosely filled with approximately 2.5 kg of unprocessed sultanas to leave a headspace of 180 mm. Column 1 remained non-compressed. Columns 2, 3 and 4 were compressed by 50, 100 and 150 mm respectively. Ethyl formate in the form of Eranol was applied according to label to the centre of the compressed column (injection point). Measurements were then taken at 100, 200 and 300 mm distance above and below the injection point. Analysis was carried out using Varian GC with AT-WAX column as described previously, except that oven temperature was increased to 100ºC.

Results

After 3 hours of fumigation the majority of fumigant was found at or below the point of application (Figure 1.20). There was no clear relationship between compression and fumigant distribution. However, after 24 hours high levels of ethyl formate were associated with low compression of sultanas (Figure 1.21). In particular this was clear at the point of application and at the points furthest away from application. It seems that compression plays no or only a minor role during the early stages of fumigation but has an impact on fumigant distribution in the longer term. Further

23

investigations with detailed sampling at various stages during the fumigation are necessary. Toxicological data on necessary times of exposure to ethyl formate in sealed systems is needed to guide these experiments.

0 50 100 150

-300 mm

-200 mm

-100 mm

0

+100

+200

+300

Compression (mm)

Distance from

injection point

Figure 1.20. Ethyl formate measured in compressed columns of unprocessed sultanas 3 h after application. Data shown as 3-D cone graph. Height of cone shows ethyl formate concentration.

0 50 100 150

-300 mm

-200 mm

-100 mm

0

+100

+200

+300

Compression (mm)

Distance from

injection point

Figure 1.21. Ethyl formate measured in compressed columns of unprocessed sultanas 24 h after application. Data shown as 3-D cone graph. Height of cone shows ethyl formate concentration.

24

Section 2: Mortality studies of pest populations sourced from Sunraysia to fumigation with ethyl formate and comparisons with an unexposed population to identify any development of ‘tolerance’ to the fumigant. Background Insect populations obtained from the Sunraysia district, were raised and subjected to space fumigation treatments with ethyl formate. Insects were trapped during the autumn of 1999, several cultures were started at this time to obtain sufficient numbers by the spring of 1999. The major pests Plodia interpunctella, Oryzaephilus surinamensis and O. mercator were targeted, other species which were captured in abundance were also cultured. These included Tribolium confusum, Tribolium castaneum (Herbst) and Ephestia figulilella. Insect cultures after multiplication were transported to CSIRO in Canberra for mortality studies. The aim of this research was to prove the efficacy of the fumigant on local insect populations and pinpoint any resistance (tolerance) to ethyl formate which may have developed in local populations. It included: • establishing whether the local Sunraysia pest populations had developed tolerance to ethyl

formate fumigation. This was carried out by establishing a range of dosages which caused a range of mortalities for insect populations from Sunraysia (eg. 0, 50, 90, 99, and 100% mortality) and then comparing these results with results from cultures raised in Canberra which had no history of ethyl formate exposure.

• establishing dosage parameters to obtain 95 and 100% kill on local Sunraysia and Canberra populations.

CSIRO Entomology (SGRL) background research has shown that ethyl formate is sorbed rapidly onto many grains and onto sultanas although at a much lower rate (section 1). This means that a dosage observed to kill insects exposed without sultanas present will be the dosage that needs to reach the insects in the fruit after sorption has been taken into account. At present sorption has not been studied in sufficient detail to know how much fumigant will be sorbed in every scenario. It was decided that at this stage mortality studies without fruit would be of more use than mortality studies with fruit present. Materials and Methods

Bioassay methods

Cultures of local insect populations were established by collecting insects from Sunraysia dried vine fruit packing establishments during the 1999 season and breeding up populations of these insects. The major pest species present were identified under an Olympus stereo zoom microscope and placed in incubation containers with their favoured food substrate. Insects were cultured at 25oC in an incubation chamber under a 14 hour daylight - 10 hours dark regime. The chamber had no humidity controller and this was modified slightly by the addition of steam from a steam humidifier ( Eucybear brand ). The insects captured and cultured included, the beetles, Oryzaephilus surinamensis (common name sawtoothed grain beetle), Tribolium confusum (common name confused flour beetle), Tribolium castaneum (common name red flour beetle), Oryzaephilus mercator (common name merchant grain

25

beetle), the moths Plodia interpunctella (common name Indianmeal moth) and Ephestia figulilella (common name raisin moth). The Tribolium species were raised on a mixture of wholemeal stone ground flour, unprocessed sultanas and yeast. The rest of the insects were raised on unprocessed sultanas. All food substrates were frozen for 3 months prior to use to eliminate mite infestation and cross infestation of the cultures. Prior to use in the dosage trials insects were separated from the food substrate, either by sieving, running adults off the food or hand selecting the desired life cycle stage. After exposure to the fumigant a recovery time of 24 hours was allowed prior to estimating mortalities. During this period insects were kept at 25oC on suitable food substrate in controlled temperature rooms.

Fumigation methodology

Insect lifecycle stages were separated from food substrate and other life cycle stages and placed in 150 ml glass jars with Bakelite lids. Ventilation was through mesh incorporated into the lid. The jars were placed in glass vacuum desiccators and exposed to a predetermined concentration of ethyl formate for 24 hours at 25oC. Gas dosage was calculated using gas laws and applied to a liquid. In this case Ethyl formate dose (ml) = required Concentration (mg/l) x Volume of container (l) liquid density(g/l) the volume of the desiccators had been measured gravimetrically prior to the experiment by measuring the volume of water held by the sealed desiccator. In this case the Volume of the container (l) = weight of water at 25oC x weight of water (g) 1.000027 (cc to ml) The weight of water at 25oC is found in standard tables, supplied by SGRL (fumigant handling and measurement booklet pg 122). The liquid density of ethyl formate (eranol) was taken from the manufacturers specifications; for eranol this was 901 g/l. The addition of a gas (or liquid, which becomes a gas) to a sealed desiccator increases the pressure inside the desiccator. To make allowances for this the volume the gas will occupy in the desiccator is removed prior to addition. The volume to be removed is calculated using gas laws. Based on the general gas law PV = nRT The dosage of the fumigant in mls (for a gas) and hence the amount removed from the desiccator can be calculated either using the gas density or the volume of 1 mole: Dosage (ml) = C x V x 1/ gas density x T / TGD x PGD / P Or Dosage (ml) = C x V x Volume of 1 mole / molecular weight x T / Tmol x Pmol / P,

26

Where gas density (g/l) and volume of 1 mole (l) are found in tables, C = required concentration, mg/l V = volume, (L) T = Lab temperature, oK P = Lab pressure, mb TGD = temperature for gas density, oK PGD = pressure for gas density, mb Tmol = temperature for 1 mole, oK P mol = pressure for 1 mole, mb In this experiment we used the volume of 1 mole Pressure and temperature were measured every day prior to fumigation. eg. Gas removed (ml) = C x V x 24.11 (v of 1 mole) / 74.08 (mol wt) x 296.7/ 293 (t/tmol) x 1013.25/ 956.24 (Pmol/P) (only the blue, bold, italic, text changes day to day) Verification of the gas dosage was carried out using a Shimadzu gas chromatograph GC-6AM, with FID (flame ionisation detector at 280oC) and a SP 2401/SE 30 glass column. Helium was used as the carrier gas at 80oC with an injection port temperature of 80oC. Dosage was with headspace gas of 40ul per injection. Retention time for ethyl formate was 0.62 minutes. Gas concentration in the desiccators was checked 1 hour after dosage and 24 hours after dosage prior to unsealing the desiccators. Gas standards for the GC were made up each day. The range of the standards depending on the dosage. The standards were made up in ‘Quickfit’ Erlenmeyer flasks with ground glass joints, of known volume (1.165, 1.197, 1.056 l) and ‘Quickfit’ cone adaptors with plastic cap and rubber septa. The dosage calculations for standards were calculated as above for gas dosage and gas volume (mls) removed prior to gas addition. Standard curves were graphed using Microsoft Excel with area/1000 on the x-axis and concentration on the y-axis. A trendline and R2 equation was plotted from the standards data, with the slope equation used to estimate y intercept for each point. Figure 2.1. is an example of the generated curve and associated slope.

27

Shimadzu -EtF- 40 ul injection 8/2/00 - 10/2/00y1 = 0.145x - 2.6857

R2 = 0.9819

y3 = 0.1306x - 1.0789R2 = 0.9949

y2 = 0.0935x + 0.4667R2 = 0.9938

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

0 100 200 300 400 500 600 700

Area/1000

Conc

entra

tion

EtF

(mg/

L)

y1 y2 y3 Linear (y1) Linear (y3) Linear (y2)

Figure 2.1 Examples of standard curves generated during the experiments. EtF = Ethyl formate. Insect mortality was calculated after the insects had been removed from the fumigation desiccators for 24 hours. Insects were kept at 25oC on a small amount of food during this time. At this stage all dead insects were removed and counted. After a further 24 hours any insects which died were also removed and added to the mortality. (In these experiments mortality at 24 hours did not change with further incubation.) Results and Discussion Canberra insect cultures

Initial observations were made with Canberra (SGRL) cultures of T. castaneum (Tc4), T. confusum (Tco37) and Ephestia kuehniella (Ek ; chosen due to it’s similarity to raisin moth). Both adult and larval stages of the two beetles were observed along with the larval stage of the moth. Dosages of 0, 5, 10 and 20 gm-3 were selected as starting point concentrations for mortality studies, based on recent observations with Sitophilus species: a grain pest. From discussion it was felt that these dosages should give a range of mortalities including 100% (figure 2.2. and table 2.1). During the experiments the dosage range was adjusted as needed. Initial results showed quite low mortalities for the adult and larval stages of the flour beetles T. castaneum (Tc4) and T. confusum (Tco37). The mortality range for larval E. kuehniella (Ek) in the initial dosages was within expectations (Figure 2.2. and table 2.1). As a result a higher range of concentrations was selected for a second exposure (on Tuesday 8/2/00). This consisted of 0, 20, 40, 60 gm-3 of ethyl formate.

28

Table 2.1. Mortality of larvae, adults and pupae of three insect species when exposed to varying concentrations of ethyl formate in desiccators for 24 hours. Twenty insects from SGRL were in each sample, 5 samples per desiccator giving a total of 100 insects per desiccator. Carried out on 7th and 8th of February 2000. % mortality Sample Desired

Conc. gm-3

Final Conc. gm-3

Ephestia kuehniella (larvae)

Tribolium castaneum (adults)

Tribolium castaneum (larvae)

Tribolium confusum (adults)

Tribolium confusum (larvae) (pupae)

4 0 0 0.0 0.0 0.0 0.0 0.0 1 5 4.5 6.7 0.0 5.0 0.0 5.0 2 10 7.9 72.2 0.0 9.5 0.0 4.5 3 20 16.9 100.0 65.0 73.7 52.4 76.2

Figure 2.2. Mortality of three insect species exposed to varying concentrations of ethyl formate in desiccators for 24 hours, 20 insects per sample 7/2/00. EK = E. kueheniella, TC4 = T. castaneum, TCo37 = T. confusum.

29

Table 2.2 and Figure 2.3 show gas concentrations used and mortalities achieved during the second 24 hour fumigation on the 8th and 9th of February. Twenty T. confusum (Tco37) adults and larvae were placed in each desiccator. Then treated with ethyl formate at concentrations of 0, 20, 40, 60 gm-3. Table 2.2 Mortality of larvae and adults of T. confusum when exposed to varying concentrations of ethyl formate in desiccators for 24 hours on the 8th and 9th of February 2000. Twenty insects from SGRL were in each sample, N= 2 samples per desiccator giving a total of 40 insects per desiccator.

8/2/00 % mortality Sample

Desiccator Desired Conc. gm-3

Start Conc. gm-3

Final Conc. gm-3

T. confusum Tco37 (adults)

T. confusum Tco37 (larvae)

5 0 0 0 0.0 5.3 13 20 20.7 11.1 57.9 95.0 9 40 62.8 36.6 100.0 100.0 7 60 81.7 47.3 100.0 100.0

Figure 2.3. Mortality of larvae and adults of T. confusum exposed to ethyl formate on 8/2/00 for 24hours Similar mortalities at 20gm-3 were achieved when compared to the first trial (Figure 2.1.) but 100% mortality at a dosage of 40 gm-3 indicated that at this stage a dosage range which included increments between 20 and 40gm-3, should probably give a range of mortalities from 0 to 100%, including 50 and 95% mortality for either Tco37 adults or larvae. To obtain a more replicable result larger numbers of insects were exposed to a dose range of 0, 20, 25, 30, 40 gm-3 in the 3rd fumigation on the 9th of Feb. The insects selected were Tco37 adults, Tc4 larvae and Oryzaephilus surinamensis (Os) adults. The adult Tco37 beetles were chosen because T. confusum are more numerous in Sunraysia than T. castaneum, while selection of Tc4 larvae was due

30

to a large number of these being available in a clean state already removed from food substrate and saving at least ½ a days work. O. surinamensis was included because it is the most numerous beetle pest in Sunraysia. Insects were weighed into the sample jars, using pre-estimated insect weights with the aim to add approximately 200 insects per jar. This was replicated twice and gave approximately 600 insects per desiccator. Adding insects above this level per desiccator was discussed but work at SGRL showed that both live and dead insects selectively sorb ethyl formate and it was feared that a higher quantity of insects may have caused reduced mortalities due to reduced fumigant concentrations and effectiveness. Thus we did not wish to investigate this at present, therefore the lower but still statistically significant levels of approximately 200-300 insects per jar was chosen. Results are shown in table 2.3 and Figure 2.4. Table 2.3. Mortality of T. confusum adults, T. castaneum larvae and O. surinamensis adults when exposed to varying concentrations of ethyl formate in desiccators for 24 hours on 10/2/00. 260 Tc4 larvae weighed ~ 0.6g, 190 TCo37 adults weighed ~ 0.46g, 320 Os adults weighed ~ 0.16g each sample was replicated in separate desiccators giving approximately 600-700 insects per desiccator.

% mortality Sample Desired

Conc. gm-3

Start Conc. gm-3

Final Conc. gm-3

Tribolium castaneum Tc4 (larvae)

Tribolium confusum Tco37 (adults)

Oryzaephilus surinamensis Os (adults)

1 0 0 0 0.0 0.5 1.3 2 20 19.9 15.8 16.6 78.6 100.0 3 20 20.4 15.4 42.5 95.7 100.0 4 25 24.9 18.9 66.2 98.6 100.0 5 25 25.9 19.8 85.6 99.5 100.0 7 30 30.5 24.0 93.8 100.0 100.0 9 30 29.9 23.7 95.7 100.0 100.0

11 40 38.8 33.5 100.0 100.0 100.0 13 40 38.6 32.6 99.6 100.0 100.0

31

Figure 2.4. Mortality of insects exposed to varying concentrations of ethyl formate in desiccators for 24 hours on 10/2/00 ( 200-300 insects per sample in 2 reps using Canberra lab cultures ) It can be seen that a good range of mortalities was achieved for Tc4 larvae, but the adult Tco37’s in this case died at slightly lower gas concentrations than expected and although the mortality curve is acceptable it is not optimal without a 50 percentile position. The Tribolium species were considerably more tolerant of ethyl formate than the adult O. surinamensis which reached 100% mortality at 20gm-3. Having established initial mortality curves for T. confusum adults, T. castaneum larvae and observed O. surinamensis adults and E. kuehniella larvae, sourced from SGRL in Canberra, initial mortality curves for insects sourced from the Sunraysia district were investigated. Sunraysia insects

The insects used included T. confusum adults, O. surinamensis adults, Plodia interpunctella (P.i.) larvae and Ephestia figulilella (E.f.) late stage larvae/pupae. Twenty insects per sample were placed in jars in desiccators, which were dosed with 0, 5, 10, 20 or 30 mg/l (gm-3 ) of eranol. These insects were placed in jars with 1 or 2 sultanas (their food substrate). It was presumed that this small amount of food substrate would not adsorb a significant amount of fumigant. This seemed to be the case with concentrations 1 hour after addition hardly changing in the next 24 hours (start conc. vs final conc. from table 2.4).

32

Table 2.4. Mortality of T. confusum adults, O. surinamensis adults, P. interpunctella larvae and E. figulilella late stage larvae/pupae, when exposed to varying concentrations of ethyl formate in desiccators for 24 hours on 16-17 Feb 2000. Initial tests with insects sourced from Sunraysia, 20 insects per sample were added unless available numbers were lower giving approx 80 insects per desiccator.



Start Conc. gm-3

Final Conc. gm-3

T. confusum

(adults)

O. surinamensis

(adults)

P.interpunctella

P.i. (larvae)

E. figulilella E.f. (late stage larvae/pupa.)

1 0 0.0 0.0 0.0 0.0 0.0 8.3 2 5 4.7 4.5 0.0 5.0 23.1 50.0 3 10 8.9 8.5 0.0 78.3 66.7 75.0 4 20 17.9 17.5 100.0 100.0 91.7 100.0 5 30 26.7 27.2 100.0 100.0 100.0 100.0

Figure 2.5. Mortality of insect cultures exposed to varying concentrations of ethyl formate in desiccators for 24 hours, 16-17 Feb 2000. Insects sourced from Sunraysia district From Figure 2.5. it can be seen that all insects tested had 100% mortality at a concentration of 30 gm-3 or lower. Both larvae of the moth species (P.i. and E.f.) had gradually increasing mortality with increasing gas concentration. O. surinamensis showed a rapid increase in mortality from 5 to 78% mortality when fumigant concentration was increased from 5 to 10 gm-3. Whereas T. confusum adults showed no mortality at all up to 20 gm-3 at which point 100% of insects died.

33

These results led us to try to produce a mortality curve for larger numbers of O. surinamensis adults, T. confusum adults and larvae from Sunraysia and to compare these mortality curves with previous curves for SGRL raised insects. A dosage range of 0, 10, 15, 20, 25, 30 gm-3 was used. Due to restraints in insect numbers O. surinamensis was observed at 0, 10, 15, 20, 25 gm-3 and T. confusum at 0, 15, 20, 25, 30 gm-3. Insects were weighed and a pre-estimated weight added to each jar giving approximately 100-150 insects per replicate since the cultures available would not yield 200 insects per sample. Results are displayed in table 2.5 and Figure 2.6. Table 2.5. Mortality of T. confusum adults, larvae and O. surinamensis adults when exposed to varying concentrations of ethyl formate in desiccators for 24 hours on the 17/2/00. All insects originating from Sunraysia district Victoria, Australia. Tco adults weighed ~ 0.38g, Tco larvae weighed ~ 0.22g, Os adults weighed ~ 0.1g giving 150-200 insects per sample and less than 600 insects per desiccator.



Start Conc. gm-3

Final Conc. gm-3

T. confusum (adults)

T. confusum (larvae)

O. surinamensis

(adults) 1 0 0.0 0.0 0.0 0.0 6 10 10.2 10.7 - - 95.0 2 15 14.9 13.8 89.7 79.5 100.0 3 20 20.3 18.4 99.3 90.4 100.0 4 25 25.3 23.8 100.0 89.4 100.0 5 30 30.3 28.2 100.0 98.1

Figure 2.6. Mortality of insects exposed to varying ethyl formate concentrations in desiccators over 24 hours, 17/02/00. Insects originating from Sunraysia district.

34

The results shown in Table 2.5 and Figure 2.6 mirror the previous results (Figure 2.5) giving the initial results from a small sample of insects more credence due to the larger insect numbers exposed in this experiment (100-200 insects). More work on other life cycle stages need to be carried out to find the most tolerant life cycle stage, which is expected to be the pupae or late stage larvae. Fumigant dosages can then be adjusted for this life cycle stage. Previous work by Hilton and Banks had identified Oryzaephilus species as relatively tolerant of ethyl formate, however these results indicate Tribolium species are more tolerant when no food residues are present, this was the case with both the Sunraysia and Canberra cultures. The results from the mortality trials for insects raised at SGRL in Canberra and insects sourced from Sunraysia and raised in Merbein are plotted and compared in Figure 2.7. These results indicate that there was no increase in tolerance to ethyl formate in the Merbein cultures for any of three insect species tested including T. confusum adults, T. confusum larvae and O. surinamensis adults when compared to the unexposed cultures sourced form Canberra.

Figure 2.7. Mortality of insect cultures after exposure to varying concentrations of ethyl formate in desiccators for 24 hours. Insects sourced from Merbein and SGRL Canberra cultures, 150 - 250 insects per sample. 10 -17 Feb 2000. In each case the insects from Merbein died at the same or lower dosage than the insects cultured in Canberra. For O. surinamensis the Canberra insects were not exposed to a dosage below 20gm-3 so the dosage was higher initially than the dosage the Merbein insects were exposed to. It is likely that they have similar mortalities at similar concentration rates. It can be seen from these results that there is no evidence of increased tolerance to ethyl formate in the insect population in Mildura, when compared to cultured populations in Canberra. This may have been helped by the present work

35

practices which means the fumigant is used solely on goods which are leaving the factory, therefore taking any tolerant insects out of the area. However, if ethyl formate is to be used on-site as a fumigant, with survivors possibly returning to the general pest population, a more accurate estimate of 99.9% mortality would be needed. This will allow a good estimate of dosage rates needed to reduce the chance of tolerance occurring. Hence further mortality curves should be explored for the Merbein cultures to help quantify the actual dosage needed to obtain 99-100% mortality of pests. At present in the commercial situation insects are treated at what is probably a reasonable dosage for 100% kill. It can be concluded that there is a high likelihood that no tolerance to ethyl formate has developed in the Sunraysia district dried fruit pests. Furthermore an effective dose of 40gm-3 of ethyl formate should cause 100% mortality of common pests.

36

Section 3: Ethyl formate fumigation as a replacement for methyl bromide fumigation in dried vine fruit. Aim: To develop efficacious and safe application methods for the fumigation of dried vine fruit in large carton or bulk bin stacks using ethyl formate. Two trials were carried out to demonstrate and refine fumigation technologies. Ethyl formate fumigation of a shipping container load of unprocessed dried sultanas in June 2000 and a combined ethyl formate and CO2 fumigation of a shipping container of unprocessed dried sultanas in April 2001. Background Ethyl formate is commonly used for insect control in the dried fruit industry, where it is added to processed fruit during packaging. During storage of the unprocessed fruit in 0.5 t bins, stack fumigation with methyl bromide has been the preferred pest control option. However, with the pending phase-out of methyl bromide, alternatives such as ethyl formate need investigation. Ethyl formate is already registered for application to dried fruit and is therefore a highly attractive option for bulk disinfestation of unprocessed fruit during warehousing. Work on the effect of ethyl formate on pests of dried fruit (section 2) and the physio-chemical behaviour of ethyl formate when applied to dried fruit (section 1) has progressed to the point where several field trials were carried out. Trial 1. Ethyl formate fumigation of a shipping container load of unprocessed dried sultanas in June 2000 Methodology The trial commenced on 29th June 2000 on site at the Mildura Cooperative Fruit Company in Irymple. A shipping container was loaded with approximately 4.5 t of unprocessed dried sultanas in nine 0.5 t bulk bins and treated with 60 g m-3 of ethyl formate vapour for 44 hours. The 20 ft container was in good condition, had plywood flooring, and passed a pressure test applied after additional sealing of the floor, prior to in-loading. The fumigant was applied in the afternoon in liquid form by gravity feed under a blanket of carbon dioxide and left to vaporise from a channel within the container. All fittings were carefully earthed. On 1st July 44 hrs after application, the remaining ethyl formate vapour was removed from the container by natural ventilation after the doors were opened. The average ambient temperature during the fumigation was 11.6°C with a maximum of 16.4°C and a minimum of 6.6°C. Within the commodity temperatures averaged 9.2°C with minimum night time temperatures of 3.7°C and maximum daytime temperatures of 14.7°C. Relative humidity averaged 61% (minimum: 51.9%, maximum: 72.5%). Throughout the fumigation process ethyl formate concentrations were monitored with a portable GC-PID (Photovac gas chromatograph with photo-ionisation detector). Samples were taken from 7 points in the container by the instrument’s sample pump via copper sample lines; gas was drawn through the lines after the lines had been flushed using a 1 litre syringe. Concentrations were calculated based on peak areas compared to standards. Caged bioassays of Oryzaephilus surinamensis and Tribolium confusum (both field strains ex CSIRO Merbein) had been placed in the headspace and buried amongst the fruit. Laboratory and field

39

controls were maintained for comparison. Both bioassays included all lifecycle stages including eggs, larvae, pupae and adults.

a b c Figure 3.1. a: Shipping container with bins and sample lines prior to addition of ethyl formate, b: the portable GC used in gas concentration monitoring, c: Trial dosing system for ethyl formate. Results Within minutes after application concentrations in the headspace reached 5-10 g m-3. Concentrations rose slowly approaching 30 g m-3 within 4 hrs after application. Over the next 3 hrs concentrations varied from 30-50 g m-3 and reached 65 g m-3 eight hours after application, indicating the end of the vaporisation phase. After 24 hrs fumigant levels in the container were between 30-65 g m-3 throughout the vessel. Higher concentrations were found in the headspace while lower concentrations were measured inside the bins of fruit, although generally fumigant distribution was fairly even. After 44 hours headspace concentrations were somewhat lower but still above 50 g m-3. Airing off was easily achieved by natural ventilation with concentrations quickly reduced to below the TLV of 100 ppm (Dräger detector tube). In summary, target concentrations were easily reached after a prolonged vaporisation phase. This prolonged vaporisation was possibly due to the low fumigation temperatures experienced. Bioassay results of O .surinamensis and T. confusum placed in the headspace and buried inside the bins of fruit showed complete control of adults, larva and eggs for both T. confusum and O. surinamensis and pupae of O. surinamensis, however some low survival of the T. confusum pupae occurred, these being the more resistant life stage. After full assessment there were 2 surviving pupae from 3,135 treated insects. This equates to 100% mortality for eggs, larvae and adults and 99.68 ± 0.2 % mortality for pupae. The survivors were from samples buried within the fruit, indicating poorer penetration of ethyl formate through the bulk fruit when added at low temperatures or as a liquid. Overall, ethyl formate promises to be a viable option for the disinfestation of dried fruit in shipping containers. However, as with methyl bromide, efficacious treatment at temperatures below 10°C may be difficult.

40

Figure 3.2. Ethyl formate concentrations in headspace and within sultana fruit bins during a 44 hour fumigation in June 2000. Average ambient temperature 11.6oC, commodity average temperature 9.2oC. EtF = ethyl formate. Trial 2: A combined ethyl formate and CO2 fumigation of a shipping container of unprocessed dried sultanas in April 2001. Methodology The trial commenced on 2nd April 2001 on site at the Mildura Cooperative Packing Company in Irymple, Victoria. A shipping container was loaded with approximately 4 t of unprocessed dried sultanas in 0.5 t open crates. The fruit was naturally infested with Oryzaephilus sp., Tribolium sp. and Plodia interpunctella. The container (a refrigerated type, with volume of 27.3 m3) was in good condition and passed a pressure test applied prior to in-gassing after sealing the rear refrigeration ports. During fumigation, the average ambient temperature was 17.5°C with a maximum of 27.2°C and a minimum of 10.2°C. Within the commodity, temperatures averaged 19.7°C with minimum temperatures of 17.6°C and maximum temperatures of 27.0°C. Relative humidity averaged 51.1% (minimum: 49.8%, maximum: 51.5%). On 4th April, 48 hours after commencement of the fumigation, the remaining ethyl formate vapour was removed from the container by natural ventilation. The ethyl formate/CO2 fumigant mix was produced by directing gaseous CO2 from a gas cylinder through an inlet into a sealed vessel that contained liquid ethyl formate. After bubbling through the liquid fumigant, a mixture of ethyl formate and CO2 gas left the vessel through an outlet attached directly to the shipping container (Figure 3.3 a,b,c). The fumigant gas was applied to the commodity through the rear wall of the sealed container at a constant flow rate, obtained from the measured flow rate of CO2. All fittings were carefully earthed. To facilitate evaporation of ethyl formate, the sealed vessel was kept at 35-40°C by immersion in a water bath. CO2 concentrations were measured with a portable GC-TCD (Gas chromatograph with thermal conductivity detector) fitted with a sample pump. Ethyl formate concentrations were monitored with a portable GC-PID (Gas chromatograph with photo-ionisation detector). Samples were drawn at several points in the container by the GC-TCD pump followed by the GC-PID via copper sample

41

lines. CO2 concentrations were measured after TCD calibration with standard gases. Concentrations for ethyl formate were calculated based on peak areas compared to standards (Figure 3.4.a,b). The fumigant was applied in two stages. The first stage began immediately after sealing. CO2 was passed through the chamber containing 1600 mL of ethyl formate at a flow rate of 60 mL min-1 for a period of 3 hours. Measurements of ethyl formate concentrations during this time showed that the flow rate may have been too high and that ethyl formate had been vented from the system, meaning that a peak concentration above 50 gm-3 was unlikely to occur (the aim was 60 gm-3). The second stage began 21 h after sealing. CO2 was passed through the chamber containing 1000 mL of ethyl formate at a flow rate of 30 mL min-1 for a period of 5 hours. This additional ethyl formate allowed test insects to be exposed to dosage levels expected to cause 100% mortality. Caged bioassays of P. interpunctella and T. confusum (both field strains ex CSIRO Merbein) and Sitophilus oryzae had been placed in the headspace and buried inside the fruit bins. Laboratory and field controls were maintained for comparison. P. interpunctella and Tribolium species are important pests of dried fruit. The T. confusum bioassay included all lifecycle stages including egg, larvae, pupae and adults. The P. interpunctella bioassay included adults, pupa and larva in smaller numbers. Sitophilus oryzae, while not a pest of dried fruit, was included because the species, in particular the immature stages, are known to be hard to control using fumigants, including ethyl formate.

a b. c. Figure 3.3. a: Experimental ethyl formate dosage vessel, b & c : ethyl formate vessel attached to CO2 cylinder and kept warm in water bath during gas addition.

a. b. Figure 3.4. a: Container and bulk bins set up prior to fumigation, b: gas concentration monitoring with Gas Chromatographs for the experiment.

42

Results The average ethyl formate concentration over the 48 h treatment period was 34.2 gm-3. The average CO2 concentration was 41% (Figure 3.5). In the first hour of the first stage of application, ethyl formate concentration reached 27 gm-3 while CO2 rose above 30%. When addition of gas was stopped after three hours, ethyl formate concentrations were above 45 gm-3 and CO2 at 42%. The average pressure in the container during application was 152 Pa. For the next 18 hours the CO2 concentration remained constant while the ethyl formate concentration declined to below 30 gm-3. When application was resumed for 5 h at a lower flow rate, ethyl formate concentrations were restored to 40 gm-3 while CO2 concentrations reached 50%. The container was pressurised to 37 Pa. Over the next 20 hours CO2 levels varied little while ethyl formate concentrations declined gradually with readings fluctuating around 30-40 gm-3 (Figure 3.5 & 3.6). Airing off was easily achieved by natural ventilation with concentrations reduced below the TLV of 100 ppm within less than an hour after opening the container doors (Dräger detector tube).

0

10

20

30

40

50

60

70

0 12 24 36 48Exposure time (h)

EtF

gm

-3/C

O2 %

Ethyl formate Carbon dioxide

Figure 3.5. Ethyl formate(EtF) and carbon dioxide concentration in a container filled with dried fruit in bulk bins. (White markers and full lines show container sealed and under gas, dotted line and grey markers show periods when gas was added to the container).

43

Figure 3.6. Ethyl formate concentrations recorded at 7 sites around and within fruit bins during a 48 hour fumigation with a mix of CO2 and ethyl formate. Bioassays showed 100% mortality of adults and immature life stages of dried fruit pests placed within the fruit and complete control of the natural infestation was observed. A total of 3,900 T. confusum adults, pupa and larvae were exposed and killed along with an estimated 480 eggs, this being calculated from the field controls. P. interpunctella test insects (the more resistant pupa and last stage larva), all died but only limited numbers were available with a total of 13 insects exposed to ethyl formate. Adult S. oryzae were completely controlled. The treatment also substantially reduced the number of adults developing from immature S. oryzae pupae, but failed to completely control this stage of the grain pest at this fumigant concentration. It should be noted that pupal S. oryzae are only found within whole grains and this result is not relevant for insect control in dried vine fruit. Discussion The use of CO2 as an adjunct to ethyl formate has much to recommend it. The reduced occupational risk due to CO2’s fire retardant properties and the potential (unquantified) increase in insect toxicity make this technique an attractive option for the disinfestation of dried fruit in gas-tight containers. From recent research (Allen & Desmarchelier 2002, Damcevski & Annis 2002, Hilton & Banks 1997) an ethyl formate dosage rate of 35 gm-3 would not be expected to completely control insect infestation, however with the added CO2, this was the case. Laboratory data on the toxicity of a range of CO2 and ethyl formate mixtures to insects is required to determine the ideal mix that should be used in commercial practice. Future work should include trials on optimal flow rates, pressures and vaporisation method for application to containers and gas-tight silos. Subsequent to this research BOC gases have produced an ethyl formate/CO2 mixture in a gas cylinder. This will be registered for use by June 2004. Providing the mixture produces a lethal concentration for dried fruit pests it has the potential to make safe fumigation in shipping containers or other sealed enclosures a reality.

44

Section 4 : Draft method of application of Ethyl formate (eranol) to shipping containers to fumigate dried vine fruit. To consult with packers regarding the development of suitable fumigation protocols, then develop and demonstrate this technology to industry. Applications In-container disinfestation with ethyl formate is suitable for processed and unprocessed dried fruit where insect control to meet phyto-sanitary standards is achievable. Dose

• By weight of fruit: 14 mL of Eranol per 25 kg pack fruit (0.6 mL/kg) • Calculation of liquid ethyl formate dose (mL) when applied as a space fumigant:

)901(25@)()/60(

CdensityliquidLcontaineremptyvolumeLmgionconcentratrequired

°×

Speed of treatment In-container disinfestation with ethyl formate must maintain toxic concentrations throughout the container and the commodity for a period of time sufficient to achieve efficacious insect control. In practice this means between 24-48 hours (dose dependant). Limitations Treatment should not be attempted where:

• Commodity is stored in impervious packaging • Containers fail to meet standards for gas-tightness • Container is lined with materials that substantially break down or adsorb ethyl formate • Treatment is likely to result in residues above the suggested MRL (Maximum residue limit)

Container selection Containers should be:

• Clean and structurally sound • Welded steel walls and roof • Dry plywood or steel floor • Free from contaminant residues • Free from residual infestation

45

Pressure testing

• The only way to insure the gas-tightness of a container is to pressure test it immediately before use

• Pressure decay time should exceed 10 seconds (0.2-0.1 kPa) • Pressure must exceed 0.25 kPa • If pressure test fails (3-10 seconds) check for leaks, seal them and retest

Fumigant addition With an ethyl formate in CO2 gas mixture Currently ethyl formate in CO2 gas mixture is not produced in bulk commercial quantities but limited quantities may be sourced from BOC gases. This can be added to a shipping container with reduced flammability risks. Application should be through copper lines or a purpose made sealed application port (see Figure 4.1).

Figure 4.1. Application of CO2/ ethyl formate gas to a shipping container with liquid ethyl formate Ethyl formate can be applied as illustrated in Figure 4.2. Liquid fumigant contained in an earthed drum is run through a copper line with a carbon dioxide blanket supplied from a CO2 cylinder. Gravity feed is used to dose the container with the appropriate dose of ethyl formate while reducing flammability by including CO2 as a adjunct. The liquid will then enter a channel running the length of the container containing cotton wadding where it will evaporate. Figure 4.2. Application of liquid ethyl formate to a shipping container

CO2 Cylinder

Drum with Ethyl formate

Copper tubing

Shipping container

Channel

Earth

Shipping container

CO2 / ethyl formate cylinder

Copper tubing

46

Minimising risk of ignition Above flashpoint (-20°C) vapour/air mixtures are explosive within flammable limits. Vapours within flammability limits are sensitive to static discharge. LEL (lower explosive limit) 2.8 % v/v = 85 mg/L, UEL (upper explosive limit) 16 % v/v = 485 mg/L

• Avoid exposure to heat or flames. • Avoid contact with strong oxidisers. • Apply under a blanket of carbon dioxide. • Apply at a slow rate. • Constantly monitor fumigation to assure mixture is not within flammable limits. • Uses only copper lines. • Earth all equipment. • Use non-sparking tools and equipment.

Clearance and ventilation The Threshold Limit Value TLV (permissible exposure limit PEL) for airborne exposure to ethyl formate is 100 ppm

• Before entering container that has been treated with ethyl formate the doors should be unlocked, opened wide and left for several hours.

• Container must be checked to ensure ethyl formate level has fallen below 100 ppm before risk area is declared safe.

Monitoring

• Ethyl formate concentrations should be monitored with a suitable Gas Chromatograph (e.g. Photovac PID) or other suitable apparatus at appropriate points throughout the fumigation period and during airing-off, until levels have fallen below the TLV (100 ppm).

• Environmental levels can be monitored with Dräger tubes (ethyl acetate type). Workforce requirements

• Container inspections must never be performed alone. • During fumigation in container yards someone must be present to watch out for hazards. • At least one of the members of the workforce should be an experienced fumigator holding a

fumigation licence issued by the appropriate authority (where applicable). • A minimum of three workers is recommended for treatments.

The fumigators must :

• Understand the requirements of the application technique. • Be aware of the hazards associated with fumigation using ethyl formate. • Be able to carry out a pressure test. • Know how to monitor in-container and environmental concentrations of ethyl formate.

47

Responsibilities of the fumigator-in-charge (where required)

• Notify civil authorities. • Identify risk area, post warning notices, ensure area clear of unauthorised personnel. • Assure appropriate personal protective equipment (PPE) available. • Certify risk area is safe to enter after fumigation. • Put in place health checks as required by appropriate health authorities.

PPE (Personal protective equipment) See chemical MSDS.

• Overalls • Gloves (barrier cream with polyethylene gloves, or either PVC, Neoprene or Butyl rubber

gloves) • Hard hat (where required) • Safety shoes • Gas detecting apparatus • SCBA or appropriate respirators

First aid measures

• Always get medical attention • Inhalation: remove to fresh air. CPR if unconscious. Give Oxygen if breathing is troubled. • Ingestion: Do not induce vomiting. Give water to drink. Seek immediate medical assistance. • Skin contact: Flush with plenty of soap and water. • Eye contact: Flush with plenty of water for at least 15 minutes.

Emergency and rescue operations Three persons are required if emergency access is to be gained to areas under gas.

• Two of these must be equipped with (and preferably trained in the use of) SCBA. • The third person must remain outside the risk area in case further help has to be summoned.

Fire fighting measures Use dry chemical, alcohol foam or carbon dioxide. Water may be ineffective.

• Use Class B fire extinguishers • Flood container with carbon dioxide • Cool container with water spray

Accidental release and spills

• Ventilate area • Remove all ignition sources • Wear appropriate PPE • Isolate hazard

48

• Contain and recover • Absorb spill onto vermiculite, dry sand, earth • Use water spray to disperse vapours • Do not flush to sewer

General procedure for application of ethyl formate

1. Undertake all necessary steps to ensure that fumigation conforms with regulations of the appropriate local, state and Commonwealth authorities.

2. Selection of suitable container. 3. Initial pressure testing of container. 4. Floor sealing, if required. 5. Installation of fumigant dosing port. 6. Loading of fruit. 7. Placement of monitoring lines. 8. Final pressure testing of container. 9. Additional sealing, if required. 10. Sealing container. 11. Apply ethyl formate under carbon dioxide. 12. Monitor fumigant concentration in container. 13. Monitor environmental levels of ethyl formate. 14. After 24-48 hours release gas. 15. Ventilate. 16. Check levels are below 100 ppm before declaring risk area free of gas.

Note: It is recommended that industry consider using this draft to develop its own protocol with prior advice from fumigation experts (eg. BOC, statutory authorities). Should the procedure be adopted by industry it is imperative that Safe Operating Procedures be developed and key personnel be pre-trained by fumigation experts and accredited before undertaking fumigation.

49

Technology Transfer During the project 2 demonstrations of ethyl formate fumigation were carried out. Initial results were communicated by a PowerPoint presentation to DFRDC Council on 9 Nov 2000. A PowerPoint presentation to a dried fruit quality seminar on January 29th 2001 presented initial results to industry representatives. Results of the whole project and previous research was extended to industry members at a ½ day workshop ‘Alternatives to Methyl Bromide for the Dried Vine Fruit Industry’, on December 17th 2001 at Mildura Cooperative Ltd. premises in Irymple. This presentation outlined several alternatives to methyl bromide fumigation and gave industry leaders information on the ethyl formate project, including fumigation methods used in the trials. A paper ‘Reuss, R., Tarr, C., Annis, P., 2001. ‘ Fumigation of a shipping container load of unprocessed dried sultanas with a combination of ethyl formate and carbon dioxide’, was presented by R Reuss to The National Working Party on Grain Protection, which has an on-going research program evaluating potential chemical protectants and application techniques that may maintain efficacy while reducing residue levels. This was presented after the Australian Postharvest Technical Conference, Adelaide Convention Centre, Adelaide, South Australia, 1-4 August 2000. During the project life it was found that the USA federal register was proposing to revoke the FFDCA tolerances for residues to this fumigant due to it no longer being used in USA. Background data and information was collected for the industry to present an objection to this revocation. Useful research papers published during and after the project life or used as references during this project are included in the bibliography.

Bibliography Allen, S.E., Desmarchelier J.M., 2002. Ethyl formate as a fast fumigant for disinfestation of

sampling equipment at grain export. In: Proceedings of the Australian Postharvest Technical Conference, Adelaide Convention Centre, Adelaide, South Australia, 1-4 August 2000, pp 82-88.

Annis, P.C., 2002. Ethyl formate-What are we up to? In: Proceedings of the Australian Postharvest Technical Conference, Adelaide Convention Centre, Adelaide, South Australia, 1-4 August 2000, pp 74-77.

Annis, P., 1999. The relative effects of concentration, time, temperature, and other factors in fumigant treatments. In: Jin Zuxun, Liang Quan, Liang Yongsheng, Tan Xianchang and Guan Lianghua, eds, Proceedings of the 7th International Conference on Stored Product Protection. 14-19 October 1998, Beijing, PR China. Sichuan Publishing House of Science and Technology, Chengdu, Sichuan Province, PR China.

Damcevski K.A., Annis P.C., 1998. Two old fumigants for the new millennium. In: M.P. Zalucki, R.A.I. Drew and G.G. White, (Eds), Pest Management- Future Challenges. Proceedings of the 6th Australasian Applied Entomological Research Conference, Brisbane, Australia, p 307.

Damcevski K.A., Annis P.C., 2002. The response of three stored product insect species to ethyl formate vapour at different temperatures. In: Proceedings of the Australian Postharvest Technical Conference, Adelaide Convention Centre, Adelaide, South Australia, 1st –4th August 2000, pp 78-81.

Desmarchelier, J.M., 1999. Ethyl formate and formic acid: occurrence and environmental fate. Postharvest News and Information 10, 7N-12N.

Desmarchelier, J.M., Allen, S.E., Ren, Y.L., Moss, R. , Le Trang Vu, 1998. Commercial scale trials on the application of ethyl formate, carbonyl sulphide and carbon disulphide to wheat. CSIRO Entomology Technical Report No 75. CSIRO, Canberra, Australia, pp 8-25.

50

Desmarchelier, J. M., Johnston, F.M., Le Trang Vu, 1999. Ethyl formate, formic acid and ethanol in air, wheat, barley and sultanas: analysis of natural levels and fumigant residues. Pesticide Science, 55. pp815-824.

Hilton, S.J., Banks, H.J., 1997. Ethyl formate as a fumigant of sultanas: sorption and efficacy against six pest species. In: E.J. Donahaye, S. Navarro and A. Varna (Eds), Proceedings of the International Conference on Controlled Atmosphere and Fumigation in Grain Storage, Nicosia, Cyprus, 1996, pp 409-422, Printco Ltd.

Hilton, S., Banks, H.J., Desmarchelier, J., Johnston, F., L. T. Vu, 1997. The effect of storage and transport on ethyl formate and methyl bromide residues in dried vine fruit. In: Technical Talk No 5. Research Report., Sunraysia Horticultural Centre, 68-74.

Pearson, R.D., Apte, V.B., 1998. A report on the explosibility of ethyl formate vapour in air. WorkCover NSW, Londonderry Occupational Safety Centre. Report No TR 17022. January 1998.

Ren, Y.L., Mahon, D., 2003, Field trials on ethyl formate for fumigation of on-farm storage. in E. J. Wright, M. C. Webb, E. Highley, eds, Stored Grain in Australia 2003. Proceedings of the Australian Post Harvest Technical Conference, Canberra, 25-27 June 2003. CSIRO Stored Grain Research Laboratory, Canberra, p210-216.

Reuss, R., Annis, P. 2000. Fumigation of rice products with the potential methyl bromide replacements, ethyl formate and carbonyl sulphide. CSIRO Entomology Contracted Report No. 55, May 2000.

Reuss, R., Annis P. 2002 (in press). Interaction of ethyl formate (EtF) with stored products. In: Proceedings of the 8th International Working Conference on Stored Product Protection, University of York, York, UK, 22 - 26 July, 2002.

Simmons, P., Fisher, C.K., 1945. Ethyl formate and isopropyl formate as fumigants for packages of dry fruits. Journal of Economic Entomology 38, 715-716.

Tarr, C. R., Clingeleffer P.R., 1993. Disinfestation of dried vine fruit to reduce reliance on chemicals, in particular methyl bromide. Final report DFRDC Project No. CSH34DF, CSIRO Division of Horticulture, Merbein Vic., 21pp, Appendix 3.

Tarr, C. R., Clingeleffer, P. R., 1997. Implementation strategies for dried vine fruit disinfestation technologies. Final report DFRDC Project No. CSH34, CSIRO Plant Industry, Horticulture unit, PMB Merbein Vic.3505., 34 pp, Appendix 4.

Trang, L. V., Ren, Y. L., 2004, Natural levels of ethyl formate in stored grains determined using an improved method of analysis. Journal of Stored Products Research, 40, 77-85.

Vincent, L.E., Lindgren, D.L., 1972. Hydrogen phosphide and ethyl formate: Fumigation of insects infesting dates and other dried fruits. Journal of Economic Entomology 65, pp 1667-1669.

51

Recommendations From this research it can be seen that there is no increased tolerance to ethyl formate in the insect population in Mildura, when compared with a cultured population from Canberra. This result supports the supposition that since the fumigant has been solely used on goods which are leaving the factory the chances of increased tolerance of resident populations due to insect survival of ethyl formate treatment are quite low. At present insects are dosed at what is probably an effective dosage for 100% kill. This is supported by the data presented here, where dosages of 30-40 gm-3 produced 100% mortality of tested insect species. However if ethyl formate is to be used on-site as a fumigant, with survivors possibly returning to the pest population, a more accurate estimate of the 99.9% mortality would be needed. Further research to produce more detailed mortality curves should be explored for the Sunraysia cultures to help quantify the actual dosage needed to obtain 99-100% mortality of pests. This will allow a good estimate of dosage rates needed to reduce the chance of tolerance occurring in the future. Furthermore an effective dose of 40gm-3 for 24 hours of ethyl formate should cause 100% mortality of common pests. The use of CO2 as an adjunct to ethyl formate has much to recommend it. The reduced occupational risk due to CO2’s fire retardant properties and the potential increase in insect toxicity make this technique an attractive option for the disinfestation of dried fruit in gas-tight containers. From recent research (Allen & Desmarchelier 2002, Damcevski & Annis 2002, Hilton & Banks 1997)an ethyl formate dosage rate of 35 gm-3 would not be expected to completely control insect infestation, however with the added CO2 this was the case in the second fumigation. Laboratory data on the insect toxicity of a range of CO2 and ethyl formate mixtures should be investigated to determine the ideal mix that should be used in commercial practice. Future work should include trials on optimal flow rates, pressures and method for application to containers and gas-tight fumigation tents. Subsequent to this research BOC gases have produced an ethyl formate/CO2 mixture in a gas cylinder (Vapormate). Providing the mixture produces a lethal concentration for dried fruit pests it has the potential to make safe fumigation in shipping containers and sealed fumigation chambers a reality. Vapormate is expected to be registered for use in 2004, studies of its efficacy are continuing on grains. Contact with BOC about research trials and supply of the product should be through Mr Robert Ryan of BOC gases. Contact details : Bob Ryan, Manager Market Support, BOC Limited, 428 Victoria Street, Wetherill Park NSW 2164 Australia. Fax: +61 2 9616 3428; Tel:+61 296163365 w or 0401718590 Email [email protected] Initial research studies since this project on Vapormates’ use in grain can be seen in the following references: References Damcevski, K., Dojchinov, G., Haritos, V., 2003, VAPORMATE™, a formulation of ethyl

formate with CO2, for disinfestation of grain., in E. J. Wright, M. C. Webb, E. Highley, eds, Stored Grain in Australia 2003. Proceedings of the Australian Post Harvest Technical Conference, Canberra, 25-27 June 2003. CSIRO Stored Grain Research Laboratory, Canberra, p199-204.

52

Haritos, V. S., Damcevski, K. and Dojchinov G. 2003, Toxicological and regulatory information supporting the registration of VAPORMATE™, as a grain fumigant for farm-storages., in E. J. Wright, M. C. Webb, E. Highley, eds, Stored Grain in Australia 2003. Proceedings of the Australian Post Harvest Technical Conference, Canberra, 25-27 June 2003. CSIRO Stored Grain Research Laboratory, Canberra, p193-198.

Mahon, D., Burrill, P. R., Ren, Y.L., 2003, Seed store disinfestation trials with

VAPORMATE™ (ethyl formate + CO2)., in E. J. Wright, M. C. Webb, E. Highley, eds, Stored Grain in Australia 2003. Proceedings of the Australian Post Harvest Technical Conference, Canberra, 25-27 June 2003. CSIRO Stored Grain Research Laboratory, Canberra, p205-209.

Ryan R.F., Bishop S.R. and Pearson R.D. 2003 VAPORMATE™ :Non-flammable Ethyl

Formate / Liquid Carbon Dioxide Fumigant Mixture., in E. J. Wright, M. C. Webb, E. Highley, eds, Stored Grain in Australia 2003. Proceedings of the Australian Post Harvest Technical Conference, Canberra, 25-27 June 2003. CSIRO Stored Grain Research Laboratory, Canberra, p 190-192.

53

ethyl formate fumigation as a replacement for methyl

Documents