Joint Experiment Technical Study (JETS) Final Report 16/1 … · 2017-04-06 · AA-075-CTR JETS Final Report 16/1 Dithiocarbamates in Tobacco – April 2017 5/32 4. Participating

Agrochemical Analysis Sub-Group

Joint Experiment Technical Study

(JETS) Final Report 16/1

Dithiocarbamates in Tobacco

April 2017

Author:

Masahiro Miyoshi, Japan Tobacco Inc., Japan

Table of Contents

1. Introduction ......................................................................................................................... 3

2. Test Material ....................................................................................................................... 3

2.1 Preparation ................................................................................................................. 3

2.2 Homogeneity ............................................................................................................. 4

2.3 Distribution ................................................................................................................ 4

3. Test Procedure .................................................................................................................... 4

4. Participating Laboratories ................................................................................................... 5

5. Limit of Quantitation and Recoveries ................................................................................. 6

6. Statistical Evaluation of Results ......................................................................................... 8

6.1 Determining the assigned value................................................................................. 8

6.2 Setting the target standard deviation ......................................................................... 8

6.3 Calculating individual z-scores.................................................................................. 9

2.4 Interpretation of z-scores [6] ..................................................................................... 9

7. Analytical Methods ........................................................................................................... 12

8. Results ............................................................................................................................... 12

9. Conclusion ........................................................................................................................ 12

10. References ......................................................................................................................... 13

APPENDIX 1: Homogeneity Test .......................................................................................... 14

APPENDIX 2: Sample Preparation ......................................................................................... 15

APPENDIX 3: Instrumentation ............................................................................................... 24

APPENDIX 4: Standard Material & Calibration .................................................................... 26

APPENDIX 5: Questions & Comments .................................................................................. 29

AA-075-CTR JETS Final Report 16/1 Dithiocarbamates in Tobacco – April 2017 3/32

1. Introduction

Annual proficiency tests on tobacco containing CPA residues have been conducted by

CORESTA-FAPAS since 2005. To further evaluate the various results on several

agrochemicals in the proficiency tests during 2006, 2007, 2008 and 2011 Joint Experiments

Technical Studies (JETS) have been performed. Two JETS targeting Chlorothalonil and

Dimethomorph in 2009, one JETS focusing on Cyfluthrin in 2010 and one JETS focusing on

Endosulfans (sum) in 2012 have been successfully performed. Useful information was

obtained on the stability of Chlorothalonil and the storage conditions for Endosulfan standard

solutions. Improved results for Dimethomorph and Cyfluthrin were also obtained.

During the 2015 Sub-Group meeting in Victoria Falls, Zimbabwe, a member of the Sub-

Group proposed performing a JETS on Dithiocarbamates due to the fact that only seven out of

the thirty laboratories that participated in CORESTA-FAPAS test in 2015 were able to obtain

a satisfactory z-score within ±2. Also at this time, a spiked Dithiocarbamates sample was

successfully prepared.

This JETS describes a mini-proficiency test to further evaluate Dithiocarbamates on tobacco

using both an incurred sample and an artificially spiked sample.

2. Test Material

2.1 Preparation

One Dithiocarbamates agronomically incurred tobacco (Burley; Mexico, 2014) and two

Dithiocarbamates-free tobaccos (Burley; Japan, 2009 and Flue-cured; Japan, 2012) were

supplied. The supplied tobaccos were ground and sieved using a 0.8 mm mesh screen.

One artificially spiked material was prepared blending powder of Maneb reference material

(Wako Pure Chemical Industries Ltd., 90.2 %) with Dithiocarbamates-free Flue-cured tobacco

by hand and thoroughly mixed by a tumbler for around 4 hours. Assuming that there was no

degradation of spiked Maneb during the preparation, the material should contain approx. 2.00

mg/kg concentration as CS2.

The below-mentioned tobacco samples were prepared in this JETS:

Blank samples: 16/1 Control 1 (Burley; Japan, 2009)

16/1 Control 2 (Flue-cured; Japan, 2012)

Test samples: 16/1 Sample 1 (Burley; Mexico, 2014), agronomically incurred

16/1 Sample 2 (Flue-cured; Japan, 2012), artificially spiked


2.2 Homogeneity

Two test samples, Sample 1 and Sample 2 were tested for homogeneity by a laboratory in

accordance with FAPAS protocol [1] in 2016 and 2014, respectively.

The results, together with their statistical evaluation [2], [3], are described in Table A1.1 and

Table A1.2 for Sample 1 and Sample 2, respectively. These data indicated sufficient

homogeneity.

2.3 Distribution

Four tobacco samples (approximately 50 g each) were dispatched to 18 participants on

February 5, 2016. Each participant was requested to store these samples at room temperature

and in the dark until testing.

One parcel was sent on March 2, 2016 due to the preparation of the necessary documents for

importation of tobacco samples.

3. Test Procedure

3.1 Each of the two test samples were analyzed in triplicate (n=3) for Dithiocarbamates (as

CS2) using the laboratory’s method of choice.

3.2 Two Burley recoveries and two Flue-cured recoveries using the corresponding blank

samples supplied were extracted and analyzed in parallel with the test samples.

3.3 Recoveries were spiked such that their final fortification levels were equivalent to 5.00

mg/kg (the current GRL for Dithiocarbamates (as CS2)) [4].

3.4 When matrix-matched calibration or procedural standard calibration was used, it was

recommended that the standard should be prepared from the supplied blank sample.

Laboratories which use methods that are “blind” to different tobacco types were free to

choose either the supplied blank or in-house blank when preparing their standard.

3.5 All analysis results were quoted in mg/kg on an ‘as is’ basis (i.e. not corrected for the

recovery or moisture).

3.6 Each laboratory was also asked to provide detailed information on the method used in

the provided template spreadsheet.


4. Participating Laboratories

19 laboratories from fourteen different countries participated in this JETS. Of these, 17

laboratories, i.e. 89.5 %, submitted the template spreadsheets.

Table 1: Participants list

Each participating laboratory was coded with a unique lab number once their results were

submitted. The results from one of the laboratories, which was assigned Lab 17, were found

to be incomplete and it was therefore excluded from this JETS.

Laboratory Country

AGROLAB-RDS, Athens Laboratories Greece

AGROLAB-RDS, Thessaloniki Laboratory Greece

Analytica Alimentaria GmbH Germany

ARGEFAR, Ege University Turkey

China National Tobacco Quality Supervision and Test Center China

Eurofins Dr. Specht Laboratorien GmbH Germany

Eurofins Food & Agro Sweden AB Sweden

Global Laboratory Services, Inc. USA

JT Leaf Tobacco Research Center Japan

JTI Ökolab Austria

LAnaRT Argentina

Microbac Laboratories Inc. Southern Testing Division USA

NVWA – Food and Consumer Product Safety Authority Netherland

PT HM Sampoera Tbk. Indonesia

Souza Cruz SA Brazil

Tobacco Research Board Zimbabwe

UFAG Laboratorien AG Switzerland

USDA AMS S&T FLS National Science Laboratory USA

Zhengzhou Tobacco Research Institute of CNTC China


5. Limit of Quantitation and Recoveries

Participating laboratories measured and reported two recoveries using the supplied blank

samples. The recovery values, together with limits of quantitation (LOQs) are shown as

follows:

Table 2: LOQs and recoveries

Fortification

level

(mg/kg)

Replicate 1

(%)

Replicate 2

(%)

Mean

(%)

1 0.9 5.2 94 92 93

2 1 5 91 94 93

3 1.00 5 90 87 89

4 0.25 5.0 92 101 97

5 0.06 5 97.9 100.6 99

6 0.01 5 61.2 60.8 61

7 0.50 6.0965 100 90.2 95

8 0.22 5 105.62 107.21 106

9 0.2 5.3 105 70 88

10 1 5.08 97 101 99

11 0.18 5.0 102.7 103.5 103

12 0.01 5.125 75 75 75

13 1 5.01 91.643 99.82948527 96

14 1 5 82 75 79

15 1 5 83 85 84

16 0.06

18 0.05 5 69 64 67

LOQ

(mg/kg)

Lab

number

Dithiocarbamates (expressed as CS2) in Burley

Not tested


Table 2: LOQs and recoveries (cont’d)

All submitted LOQs were significantly less than the current GRL for Dithiocarbamates as CS2

(5.00 mg/kg) [4].

Recoveries were measured from 1.2 -6.0965 mg/kg. The target concentration was 5.00 mg/kg

(the GRL for Dithiocarbamates) in the protocol.

Lab 16 reported the recoveries of the Flue-cured type sample only, while the other 17

laboratories reported the recoveries in both Burley and Flue-cured types. The recoveries

between both types were comparable [Table 2]. Lab 16 and Lab 18 reported recoveries in

Flue-cured outside the range of 60 % - 140 %, while the other labs reported recoveries in both

types within the range.

The results submitted from both Lab 16 and Lab 18 were excluded in determining the

assigned values in Sample 2 (Flue-cured), while the result from Lab 16 only was excluded in

determining the assigned value in Sample 1 (Burley).

Fortification

level

(mg/kg)

Replicate 1

(%)

Replicate 2

(%)

Mean

(%)

1 5.2 104 97 101

2 5 97 88 93

3 5 95 92 94

4 5.0 94 92 93

5 5 99.08 97.76 98

6 5 62.40 59.6 61

7 6.0965 99.6 100.4 100

8 5 102.39 101.58 102

9 5.3 94.2 96.2 95

10 5.08 102 103 103

11 5.0 98.4 103.1 101

12 0.469 75 75 75

13 5.01 91.55 93.56541739 93

14 5 79 84 82

15 5 79.00 75 77

16 1.2 30 27 29

18 5 49 51 50

Dithiocarbamates (expressed as CS2) in Flue-cured

Lab

number


6. Statistical Evaluation of Results

6.1 Determining the assigned value

The influence of outliers was minimized by using a robust statistical procedure to drive the

robust mean [1]. The median of valid results was also obtained. The uncertainty (u) of the

robust mean and median was assessed:

is the the standard deviation of the robust mean or the median absolute

deviation (sMAD)

n is the number of data points used to calculate the median or the robust mean

These measures of central tendency were compared.

For both test samples in this JETS, the robust mean was considered the most appropriate

measure of central tendency.

6.2 Setting the target standard deviation

The target standard deviation, P, was derived using the appropriate form of the Horwitz

equation [1], [5]:

C is the concentration, i.e. the assigned value, expressed as a

dimensionless mass ratio.

mr is a dimensionless mass ratio (e.g. 1ppm 1 10-6

).

The assigned value, the target standard deviation and their related parameters are given in

Table 3.

Table 3: Assigned value and Target standard deviation

Sample

Assigned value

(mg/kg)

Target standard deviation

(mg/kg)

Data points

n

Robust mean

Robust standard deviation

Uncertainty

u

Derived from

p

1 16 6.36 1.179 0.334 Horwitz 0.770

2 15 1.46 0.340 0.100 Horwitz 0.221

mr

C02.0σ

8495.0

p =

X̂

n

σu

ˆ=

σ̂


The uncertainty of the assigned value, u, was given by:

is the standard deviation of scaled median absolute deviation.

n is the number of data points used to calculate the median.

For Sample 2, the uncertainty of the assigned value (the robust mean) was large. Therefore z-

scores are given for information only and NOT for evaluative purpose.

6.3 Calculating individual z-scores

Participating laboratories’ z-scores were calculated as:

x is the laboratory’s reported result.

is the assigned value.

is the target standard deviation.

All results were quoted in mg/kg, uncorrected for the moistures or recoveries. The results for

Sample 1 and Sample 2, together with z-scores are given in Tables 4 and 5, respectively.

Distribution charts of the results in both samples are shown in Figures 1 and 2, respectively.

2.4 Interpretation of z-scores [6]

In normal circumstances, about 95 % of z-scores will lie in the range |z| ≤ 2. Scores in this

range are designated “satisfactory”.

Occasional z-scores in the range 2 < |z| ≤ 3 would be expected at a rate of 1 in 20. Scores in

this range are designated “questionable”.

Scores where |z| > 3 are to be expected at a rate of about 1 in 300. Given this rarity, such z-

scores very strongly indicate that the result is not fit-for-purpose and almost certainly requires

investigation. Scores in this class are designated “unsatisfactory”.

n

σu

ˆ=

σ̂

σ p

Xxz

)ˆ(=

X̂

σ p


Table 4: Sample 1 (Burley, agronomically incurred) Dithiocarbamates results & z-scores

Figure 1: Sample 1 (Burley, agronomically incurred) Dithiocarbamates results & z-scores

Replicate 1

(mg/kg)

Replicate 2

(mg/kg)

Replicate 3

(mg/kg)

Mean

(mg/kg)

SD

(mg/kg)z -score

1 5.8 6.2 5.8 5.9 0.23 -0.6

2 5.62 5.79 5.81 5.7 0.10 -0.8

3 4.88 5.14 5.17 5.1 0.16 -1.7

4 6.84 6.74 7.13 6.9 0.20 0.7

5 3.60 3.47 3.32 3.5 0.14 -3.8

6 7.22 7.23 7.29 7.2 0.04 1.2

7 7.2 7.2 7.1 7.2 0.06 1.0

8 5.43 5.67 5.35 5.5 0.17 -1.1

9 7.5 7.8 7.7 7.7 0.15 1.7

10 7.01 6.86 7.15 7.0 0.14 0.8

11 5.33 5.45 5.38 5.4 0.06 -1.3

12 5.015 5.017 5.343 5.1 0.19 -1.6

13 8.48 8.36 8.37 8.4 0.07 2.7

14 7.36 7.81 6.81 7.3 0.50 1.3

15 7.94 7.51 7.74 7.7 0.22 1.8

16 4.6 4.7 4.7 4.7 0.06 -2.2

18 5.905 5.152 5.024 5.4 0.48 -1.3

Lab

number

Dithiocarbamates (CS2)

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

10.00

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Co

nc

en

tra

tio

n (

mg

/kg

)

Laboratory


Table 5: Sample 2 (Flue-cured, artificially spiked) Dithiocarbamates results & z-scores

Figure 2: Sample 2 (Flue-cured, artificially spiked) Dithiocarbamates results & z-scores

Replicate 1

(mg/kg)

Replicate 2

(mg/kg)

Replicate 3

(mg/kg)

Mean

(mg/kg)

SD

(mg/kg)z -score

1 1.2 1.2 1.2 1.2 0.00 -1.2

2 1.25 1.25 1.21 1.2 0.02 -1.0

3 1.74 1.64 1.67 1.7 0.05 1.0

4 1.43 1.46 1.39 1.4 0.03 -0.2

5 1.72 1.67 1.71 1.7 0.03 1.1

6 1.64 1.64 1.58 1.6 0.03 0.7

7 1.3 1.3 1.3 1.3 0.00 -0.7

8 1.08 1.09 1.12 1.1 0.02 -1.7

9 1.8 1.7 1.8 1.8 0.06 1.4

10 2.24 2.10 2.19 2.2 0.07 3.2

11 1.22 1.21 1.18 1.2 0.02 -1.2

12 0.437 0.504 0.466 0.5 0.03 -4.5

13 1.83 1.92 1.90 1.9 0.05 1.9

14 1.18 1.23 1.28 1.2 0.05 -1.1

15 1.73 1.75 1.64 1.7 0.06 1.1

16 1.17 0.94 1.06 1.1 0.12 -1.8

18 0.607 0.614 0.655 0.6 0.03 -3.8

Lab

number

Dithiocarbamates (CS2)

0.00

0.50

1.00

1.50

2.00

2.50

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Co

nc

en

tra

tio

n (

mg

/kg

)

Laboratory


7. Analytical Methods

Submitted information on the methods used for Dithiocarbamates analysis are compiled and

documented in the following appendices:

Appendix 2: SAMPLE PREPARATION

Appendix 3: INSTRUMENTATION

Appendix 4: STANDARD MATERIAL & CALIBRATION

Appendix 5: QUESTIONS & COMMENTS

5 laboratories (Lab 1, Lab 3, Lab 6, Lab 10 and Lab 13) used the ISO [7] or the similar

method using spectrophotometer [Table A3.1]. The remaining 12 laboratories used a different

method.

8. Results

The outcome performance for this JETS was higher than that of the 2015 CORESTA FAPAS

proficiency test (FT0111). The percentage of satisfactory z-scores was increased from 47 % to

82 % in both Sample 1 (agronomically incurred) and Sample 2 (artificially spiked).

Out of the 17 participating laboratories, 14 laboratories obtained satisfactory evaluations for

Sample 1 and for Sample 2 (for information only) [Table 4 and 5]. 11 laboratories (Lab 1, Lab

2, Lab 3, Lab 4, Lab 6, Lab 7, Lab 8, Lab 9, Lab 11, Lab 14 and Lab 15) obtained satisfactory

z-scores for both samples.

Lab 5 and Lab 16 had z-scores less than -2 for Sample 1. However, both laboratories obtained

satisfactory z-scores for Sample 2. Lab 13 had a z-score more than +2 for Sample 1. However,

this laboratory obtained the satisfactory z-score for Sample 2.

Lab 10 had a z-score more than +3 for Sample 2, but obtained a satisfactory z-score for

Sample 1. Lab 12 and Lab 18 had with z-scores less than -3 for Sample 2. Both laboratories

obtained the satisfactory z-scores for Sample 1.

9. Conclusion

Although some laboratories performed poorly, the overall outcome for this JETS looked good.

It is recommended that these laboratories obtaining questionable or unsatisfactory evaluation

should check and/or improve their own analytical methods.

No further study is planned for the analysis of Dithiocarbamates.


10. References

1. FAPAS Proficiency Testing Protocol, Organization and Analysis of Data, Sixth Edition,

September 2002, FAPAS.

2. Analytical Methods Committee, Test for ‘sufficient homogeneity’ in a reference

material, Editor: Thompson, M., No.1 Jul 2004.

3. Analytical Methods Committee, Robust statistics: a method of coping with outliers,

Technical Brief No.6, Apr 2001.

4. CORESTA Guide Nº1, The Concept and Implementation of Agrochemical Guidance

Residue Levels, July 2013, CORESTA.

5. Thompson, M., Recent trends in inter-laboratory precision at ppb and sub-ppb

concentrations in relation to fitness for purpose criteria in proficiency testing, Analyst,

2000, 125, 385-386.

6. Thompson, M., Ellison, S.L.R., and Wood, R., The International Harmonized Protocol

for the Proficiency Testing of Analytical Chemistry Laboratories, Pure & Applied

Chemistry, 2006, 78, 145-196.

7. Tobacco and tobacco products –Determination of dithiocarbamate pesticides residues –

Molecular absorption spectrometric method, First edition -1983-06-15, Ref. No. ISO

6466-1983 (E).


APPENDIX 1: Homogeneity Test

Table A1: Homogeneity data for Sample 1 and Sample 2

Sample identity

Replicate 1 Replicate 2 Replicate 1 Replicate 2

1 8.2 9.0 4.2 5.0

2 8.5 8.0 4.3 4.5

3 8.7 8.1 4.4 4.7

4 8.9 8.7 4.2 4.5

5 7.0 8.2 4.2 4.7

6 8.7 8.3 4.7 4.2

7 8.3 8.7 4.0 4.1

8 8.1 8.8 4.1 5.0

9 8.0 7.9 4.7 4.2

10 8.4 8.5 4.5 4.9

mean, n 8.353 20 4.455 20

original of target sd (σp)

abs. target sd (σp) & as RSD% 0.971 11.62 0.569 12.78

S an

S sam2

σ all2

critical

S sam2<critical?

0.1856

ACCEPT

16/1 Sample 1

(Agronomically incurred)

(mg/kg)

Horwitz

0.4113

0.0337

0.0848

0.3303

ACCEPT

16/1 Sample 2

(Artificially spiked)

(mg/kg)

Horwitz

0.3599

0.0000

0.0292


APPENDIX 2: Sample Preparation

Lab 1

Outline of the method

Heating dithiocarbamates with a solution of stannous chloride and hydrochloric acid yields

carbon disulphide which is distilled, purified and collected in a methanolic potassium

hydroxide solution. Under these conditions, carbon disulphide forms potassium xanthogenate.

The absorption of this reaction product is measured spectrometrically at a wavelength of 302

nm with baseline correction at wavelengths of 272 nm and 332 nm (xanthogenate method).

Modifications to DIN EN 12396-3, 2000

1) The stannous chloride solution (40 g/100 mL) is not diluted with hydrochloric acid and

water. Tobacco sample is heated to boiling with water and concentrated hydrochloric acid and

then the concentrated stannous chloride solution is added.

2) An additional absorption tube filled with concentrated sulfuric acid is used between the

lead acetate solution and the sodium hydroxide solution.

Preparations

The distillation device is shown in the following Figure.

The first absorption tube is filled with 20 mL of a solution of lead acetate in water. The

second tube contains 20 mL of concentrated sulphuric acid and the third is filled with 20 mL

of a 10 % sodium hydroxide solution. The forth absorption tube contains 8 mL of the

xanthogenate reagent and has to be cooled with ice water in order to avoid losses of methanol.

All ground joints are covered with a thin fat film. To control the gas flow a vacuum pump is

fixed to the last absorption tube. The vacuum has to be adjusted to obtain a slight air stream

through the glass frit. Care has to be taken that no leaks are between the connections of the

experimental device.

Liebig condenser

dropping f unnel absorption tubes

v acuum pump

tube with f rit


Decomposition and Distillation

20 g sample material is transferred into the 1000 mL round bottom flask and the flask is

sealed. 200 mL of diluted hydrochloric acid 3.4 % is heated in a 400 mL glass beaker nearly

to the boiling point on a hot plate. The hot solution is added to the sample material using a

dropping funnel. For fortification experiments the respective amount of tituration is given to

the sample before adding the diluted hydrochloric acid. Subsequently, 20 mL of a tin (II)-

chloride solution is added and the round bottom flask is closed again.

The mixture is heated up to the boiling point and after 30 min the connection between

absorption tubes 3 and 4 is opened in order to ventilate the distillation device. The pump is

switched off and the solution in tube 4 is made up to 10 mL with methanol and used

immediately for the spectral-photometric measurements.

Evaluation

Absorbance is measured at 302 nm with baseline correction at 272 nm (or the wavelength of

the minimum of the absorption nearby) and 332 nm in a 5 cm-light path cell and the

concentration of CS2 is calculated using calibration curves of CS2.

Spectrophotometric Detection

Photometer: Carry 100

Light path cell: 5 cm (Concentration range: 0.51 µg/10mL - 7.62 µg/10 mL)

Wavelength range: 230 - 400 nm

Specific wavelengths: 302 nm (maximum of the CS2 xanthogenate)

272 nm, 332 nm (for baseline correction)

Slit width: 2 nm

Data interval: 1 nm

Scan speed: 451 nm/min

Lab 2

1． 0.5 g +/-1mg tobacco is extracted in a 20 mL Headspace vial with 5 mL Isooctane and

10 mL Tin (II) chloride-Solution (15 g/L in 5M HCl) for 3.5 h at 80 °C in an ultrasonic

bath.

2． During cooling down the Headspace vial is shaken by hand every 15min. The Isooctane

phase is measured by GC-MS.

Lab 3

1． Weigh 2.00 g of Tin (II) chloride, tare the balance.

2． Weigh 5.00 g of sample to be analyzed.


3． Transfer the sample and Tin (II) chloride to a round bottom flask in the distillation

apparatus.

4． Add approximately 50 mL of distilled water and swirl the round bottom flask for

approximately one minute or until all of the tobacco is impregnated.

5． Connect the flask to the condenser.

6． The condenser is connected to the first gas wash bottle (bubbler) containing

approximately 20 mL of sulfuric acid.

7． This wash bottle is then connected to a second gas wash bottle (bubbler) containing

approximately 25 mL of potassium hydroxide solution.

8． Heat the flask for at least 10 minutes to allow complete impregnation of the tin (II)

chloride into the tobacco and allow the oxygen to be expelled from the system.

9． Make sure that the condenser is well cooled to prevent water from crossing into the first

gas wash bottle.

10． Place 100 mL of HCl solution into the reservoir.

11． Slowly add the HCl solution to the round bottom flask by opening the three-way

stopcock so that the flow of nitrogen passes through both the inlet tube and the reservoir.

12． As you are opening for the HCl, open red three-way stopcock bottom knob by turning

the knob to your left.

13． You should see bubbles in HCl flask.

14． When all of the HCl solution has been added to the flask turn the three-way stopcock so

that the nitrogen flow is coming from the inlet tube only.

15． Make sure that the tube gas traps are bubbling.

16． Gently boil for at least 40 minutes.

17． Remove the heat source.

18． Transfer the contents of the second gas wash bottle to a 50 mL volumetric flask.

19． Wash the glass tube and the fritted thistle end with distilled water and transfer the

washings to the flask.

20． Dilute the volumetric flask to the mark with distilled water, mix and allow to stand for

at least 15 minutes.

21． Place sample in cuvette and analyze using UV instrument.

Lab 4

1． 5 g of sample in a plastic coated bottle.

2． Add 150 mL of tin chloride/HCl solution.


3． Add 25 mL iso-octane.

4． Seal with screw cap fitted with rubber septa.

5． Shake each bottle and put into water bath at 80 ± 5 °C for 60 minutes.

6． Invert bottles ten times at 20, 40 and 60 minutes.

7． Place bottles in cold water bath until the sample reached room temperature.

8． Transfer a portion of the iso-octane into a vial.

Lab 5

1． 3 grams of aluminium chloride in 100 mL flask is weighed flask. purified water up to

100 mL.

2． 0.3 grams % +-1, 20 millilitres of sample is weighed into headspace vials.

3． 3 millilitres of aluminium chloride is added to the solution prepared in the above

example vials.

4． Vial cap is tightly closed.

5． Examples are ready to make injection in 17 minutes

Lab 6

1． 5 g of tobacco (b), weighed to the nearest 10 mg, was placed in flask A. 2 g stannous

chloride was added followed by 50 mL distilled water.

2． The flask was shaken until all the tobacco has been impregnated. Immediately after this

has been done, flask A was connected to condenser B, which was connected with wash-

bottle E containing 20 mL concentrated sulphuric acid, and wash-bottle F containing 25

mL potassium hydroxide reagent. Reservoir C and inlet tube D were put in position, and

a current of nitrogen, 50 mL per minute (c), was allowed to pass through the whole

apparatus via D. Flask A was heated to 30-40 °C. For all of the tobacco to be

impregnated by the stannous chloride solution, flask A was allowed to remain for at

least 10 minutes in the conditions just described.

3． 100 mL hydrochloric acid solution was placed in reservoir C and slowly added to flask

A. Whilst the acid is being added to the reaction flask the 3-way tap was turned so that

the nitrogen supply is connected to reservoir C as well as passing into flask A via inlet

tube D. The contents of flask A were then heated to boiling point whilst maintaining a

nitrogen flow of 50 mL per minute through inlet tube D. Boiling was sustained for 30

minutes. Condenser B was well cooled to prevent steam water passing into the

concentrated sulphuric acid in trap E.

4． At the end of the boiling period, wash-bottles E and F are disconnected and the nitrogen

flow is turned off. The content of wash-bottle F is transferred to a 50 mL volumetric

flask. Flask F is thoroughly rinsed with distilled water which is also added to the


volumetric flask. The volume of the combined solutions is adjusted to 50 mL with

distilled water.

Lab 7

1． Weigh 1g of tobacco sample into an extraction vessel.

2． Add 10 mL of cyclohexane and 30 mL of hydrolysis solution (1.5 % SnCl2 + 1.5 % L-

cysteine in 5N HCl).

3． Close and place in the microwave oven.

4． Irradiate with microwave for 40 min.

5． Cooling

6． Take 1 mL from the upper phase of cyclohexane.

Lab 8

1． 2.0 g tobacco sample was weighed into a 100 mL conical flask.

2． Add 25 mL of isooctane, and then add 30 mL of 1.5 % SnCl2 made in 5 mol/L HCl.

3． Seal the conical flask with screw cap, and shake it slightly by hand.

4． The sample was extracted by ultrasound under 60 °C and 500 W for 1 hour. Shake it

every 20 minutes during sonication.

5． Take 1.5 mL of extracting solution to filter into vial for analysis.

Lab 9

1． 0.6 g Sample in Headspace Vial (20 mL)

2． Add 6 mL SnCl2 (4 %) and crimp.


Lab 10

Tobacco samples (5 grams) are heated in the presence of a stannous chloride reducing agent

(2 grams), concentrated hydrochloric acid (100 mL), and deionized water (50 mL) to destroy

and liberate any dithiocarbamate residue as carbon disulfide. Nitrogen gas is purged

continuously through the system and sweeps the carbon disulfide through a scrubber solution

of sulfuric acid (20 mL) where interferences are removed. It then travels into a trap containing

methanolic potassium hydroxide (25 mL) where potassium o-methyl dithiocarbamate is

formed. Following the reaction period, the trap is disconnected and the trap solution is

quantitatively transferred and diluted to volume in a 50 mL volumetric flask with deionized

water.

Lab 11

1． 2.0 g tobacco sample was weighed into a 100 mL conical flask

2． Add 25mL of isooctane, and then add 30 mL of 1.5 % SnCl2 dissolved in 5 mol/L HCl

3． Seal the conical flask with screw cap, and shake it slightly by hand.

4． The sample was extracted by ultrasound under 60 °C and 500 W for 1 hour. Shake it

every 15 minutes during sonication.

5． Take 2 mL of extracting solution to filter into vial for analysis.

Lab 12

1． 5 g of the homogenized material are weight in a headspace vial (20 mL).

2． The SnCl2-HCl solution is heated at 50 °C.

3． 10 mL of that solution is added to the headspace vial containing the sample.

4． Air is evacuated by adding Argon and the vial is closed.

5． The vial is kept in the laboratory oven at 70 °C for 2 hours, where the conversion of

Dithiocarbamates to CS2 takes place by reaction with the SnCl2-HCl solution.

Lab 13

1． Weigh 5 g of tobacco (b) to the nearest 10 mg, place in flask A. Add 2 g stannous

chloride and followed by 100 mL distilled water.

2． Shake the bottle until all tobacco submerged. Connect flask A to condenser B.

Condenser B is connected to E wash-bottle that containing 20 mL concentrate sulphuric

acid and F wash- bottle, containing 25 mL potassium hydroxide reagent. Then, put

reservoir C and inlet tube in position, and allow a current of nitrogen 50 mL per

minute(c), pass through the whole apparatus via D.

3． Heat flask A in 30-40 °C for at least 10 minutes.


4． Add 50 mL hydrochloric acid solution 25 % in reservoir C slowly to flask A. While the

acid is being added, flask the 3-way tap should be turned off so the nitrogen supply is

connected to reservoir C as well as passing into flask A via inlet tube D.

5． Heat flask A to boiling point while maintaining a nitrogen flow 50mL/minute for 30

minutes.

6． Turned off the nitrogen flow transfer and rinse the flask F thoroughly with distilled

water, adjust the volume into 50 mL with distilled water. And, stand for 15 minutes

before determine using spectrophotometric at 272, 302, and 332 nm.

Lab 14

1. Sample weight: 10gr

2. Add: 40 mL H2O, 25 mL iso-octane and 150 mL solution SnCl2 in HCl 12N

3. Shake in water bath: 2 hours in 80 °C

Lab 15

1. In a 250 mL Shotte bottle weight 10.0 g sample, add 40 mL Η2Ο and wait 10-15'.

2. Add 25 mL iso-octane and 150 mL hydrolysis reagent (tin (II)-chloride in hydrochloric

acid) and close immediately.

3. Put the Shotte bottle in a shaking-water bath for 2h at 80 °C.

4. After the 2h the reaction mixture is cooled down to 30 °C in a cooling water bath.

5. Take 1 mL of the iso-octane phase, filtrate with PTFE filter 0.45 µm

6. Ready for GC-FPD analysis.

Lab 16

Samples are weighed out 3 grams each into separate properly labeled Teflon tubes with screw

on lids. Isooctane and leaching solution are added to each tube. The samples are placed on an

80 °C hot block for one hour. Samples are mixed well every 15 minutes. After the samples are

removed from the hot block, they are allowed to cool to room temperature and the centrifuged

for 10 minutes at 3000 rpm. One mL of each centrifuged sample is transferred to properly

labeled auto sampler vials and given to the analyst for testing. The leaching solution in

combination with the heat liberates CS2 gas, which is detected via GC/PFPD) with a sulfur

filter installed instrumental analysis.

Lab 18

1. Weigh 10 g (± 1 %) tobacco sample in a sample flask of 250 mL, add 40 mL water

(wetting) and let stand for 15 min.

2. Then add 25 mL iso-octane, with a dispenser.

3. Add 150 mL Tin (II) chloride (15g/L) in HCl 4M, with a dispenser and tightly close the

flask immediately with a screw cap with septum inlay.


4. Place the flask in a water-shaking bath, set at 80 ˚C. Shake for 2 hours.

5. Let the flask contents cool down to room temperature (eventually with a cooling bath)

6. Pipet 0.5 mL of the upper iso-octane layer into an auto sampler vial, add 0.5 mL iso-

octane, mix and close the vial with a snap-cap.

7. Inject 4 µL into the GC-MS TQ system (used in the MS-SIM mode)

Sample weight, extraction solvent, the solvent volume, extraction technique and the time in

each laboratory are described in Table A2.1.


Table A2.1: Sample weights and extraction/clean-up procedures in each laboratory

Participating laboratories used various extraction techniques (distillation, ultrasonic, heating, shaking and headspace) during from

17 minutes to 3.5 hours.

Lab

number

Sample weight

(g)

Extraction

solvent & reagent

Extraction

technique

Temperature

(ºC)

Extraction

timeClean-up

1 20 HCl (200mL) & SnCl2 solution (20mL) Distillation Boiling 30 minsWashing with lead acetate,

sulfuric acid & NaOH

2 0.5 (±1mg) Isooctane (5mL) & SnCl2/HCl solution (10mL) Ultrasonic 80 3.5 hours No-clean-up

3 5.0 Distilled water (50mL), SnCl2 (2g) & HCl (100mL) Distillation Boiling 40 minsWashing with sulfuric acid &

KOH

4 5 Isooctane (25mL) & SnCl2/HCl solution (150mL) Heating 80±5

60 mins

(inverting every

20 mins)

No-clean-up

5 0.3 (±1%) Water (3mL) & 3% AlCl3 (3mL) Headspace 17 mins No-clean-up

6 5 (to the nearest 10mg) Distilled water (50mL), SnCl2 (2g) & HCl (100mL) Distillation Boiling 30 minsWashing with sulfuric acid &

KOH

7 1 Cyclohexane (10mL) & SnCl2/HCl solution incl. L-cystein (30mL) Microwave 40 mins No-clean-up

8 2 Isooctane (25mL) & 1.5 % SnCl2/HCl (30mL) Ultrasonic Under 601 hour (shaking

every 20 mins)No-clean-up

9 0.6 4% SnCl2 in water (6mL) Headspace No-clean-up

10 5 Deionized water (50mL), SnCl2 (2g) & conc. HCl (100mL) Distillation Heating Washing with sulfuric acid

11 2 Isooctane (25mL) & 1.5 % SnCl2/HCl (30mL) Ultrasonic Under 601 hour (shaking


12 5 SnCl2/HCl (10mL) Headspace 70 2 hours No-clean-up

13 5 (to the nearest 10mg) Distilled water (100mL), SnCl2 (2g) & HCl (50mL) Distillation Boiling 30 minsWashing with sulfuric acid &

KOH

14 10 Water (40mL), isooctane (25mL) & 1.5 % SnCl2/HCl (150mL) Shaking 80 2 hours No-clean-up

15 10.0 Water (40mL), isooctane (25mL) & SnCl2/12N HCl (150mL) Shaking 80 2 hours PTFE filter 0.45um

16 3 Isooctane Heating 801 hour (mixing


18 10 (±1%) Water (40mL), isooctane (25mL) & SnCl2/4M HCl (150mL) Shaking 80 2 hours No-clean-up


APPENDIX 3: Instrumentation

12 participating laboratories used gas chromatography mass spectrometry (GC-MS), Flame

Photometric Detector (GC-FPD) or Electron Capture Detector (GC-ECD) for the analysis of

Dithiocarbamates in this JETS, while 5 laboratories employed spectrophotometer.

Instrument, detector and parameters applied in each laboratory are described in Table A3.1.


Table A3.1: Instrument, detector and parameters in each laboratory

Volume

(µL)Mode Type

Dimensions

(length, internal diameter,

depth)

Stationary phase

1302 nm corrected by

272 nm & 332 nm

2 GC MS (EI+) 76/78 1 Split (20:1)Split, with cup,

glass wool

Agilent, DB-

5MS30 m x 0.25 mm, 0.25 µm

5% Phenyl 95% dimethyl

arylene siloxane

50°C (2 min) , 20°C/min -> 100°C (0 min),

100°C/min -> 300°C (5 min)


272 nm & 332 nm

4 GC FPD - 5 SplitlessStraight, non-

deactivated

RTX-502.2,

längd 30 m x 0.53 mm, 3 µm

diphenyl/dimethyl

polysiloxane phase?

5 GC MS (EI+) 76 1000 Split (20:1)4 mm ID tap

GW Pk 1

HP-

INNOWAX60 m x 0.320 mm, 0.25 µm Silica 4 min.50 ºC,40 ºC increment 130 ºC 1 min.


272 nm & 332 nm

7 GC FPD - 1 Split (1:1) ? GS-GasPro 15 m x 0.32 mm, 0.1µmunique bonded PLOT

column 100℃ (30℃/min) 200℃

8 GC MS (EI+) 76/78 2 Split (10:1)Split, cone, with

glass wool

Agilent HP-

VOC60 m x 0.32 mm, 1.8 µm

6% cyanopropylphenyl -

94% dimethyl polysiloxane

from 45℃ (hold for 2 min) to 100 ℃ at 10 °C/min,

then hold for 15 min at 230 °C in post run mode

9 GC MS (EI+) 76 100 Split (90:1)Split with glass

wool

Agilent, DB-

5MS30 m x 0.25 mm, 0.25 µm

5% Phenyl 95% dimethyl

arylene siloxane?


272 nm & 332 nm

11 GC MS (EI+) 76/78 2 Split (10:1)cone, glass

wool

Agilent DB-

62460 m x 0.32 mm, 1.8 µm

6% cyanopropylphenyl -

94% dimethyl polysiloxane

from 45℃ (hold for 2 min) to 100 ℃ at 10 °C/min,

then hold for 15 min at 230 °C in post run mode.

12 GC ECD - 2000 Split (5:1)

Ultra inert,

universal, low

pressure drop,

glass wool

Agilent

19095J-

123LTM

30 m x 530 μm, 2.65 μm HP-5


272 nm & 332 nm

14 GC FPD - 4 Split (10:1) 4mm ID tap GW DB-5MS 30 m x 0.32mm, 0.25 µm5% Phenyl 95%

dimethylpolysiloxaneInitial 45 C - 10 C/min - Final 260 C

15 GC FPD - 5 Split (1:12)spl-2010 with

glass woolDB-5 50 m x 0,32 mm, 1µm 95% methylpolylsiloxane

50°C, initial time 4'. 1st ramp: heating rate

8°C/min, final temp 80°C. 2nd ramp: heating rate

20°C/min, final temp 250°C and stay 1'

16 GC FPD - 1 Splitlessgooseneck with

glass wool

Restek Rtx 1

Pesticide

capillary

30 m x 0.32 mm, 0.25 μm

nonpolar phase;

Crossbond® dimethyl

polysiloxane

Equilibration Time: 1.00 minutes at 50°C;

Initial temperature: 50°C for 0.50 minute;

Ramp: 40°C/minute to 220°C with a 0.5 minute

18 GCMS-QQQ

(EI+)76/78 4 Split (1:7)

Open liner with

carbofrit

Varian-

Chrompack,

capillary

50 m x 0.32 mm CP-Sil 8

Injector temperature: 200 ˚C Oven temperature

program: 40 ˚C (5 min) – 40 ˚C/min -> 250 ˚C (4.75

min)

Spectrophotometer

Spectrophotometer

Spectrophotometer

ColumnInjection

Liner type Temperature profileDetectorIons/wavelengths

monitored

Spectrophotometer

Spectrophotometer

Lab

numberInstrument


APPENDIX 4: Standard Material & Calibration

Table A4.1: Standard and internal materials used in each laboratory

Only Lab 12 used Thiophen as an internal standard, while other laboratories used their standards only.

Lab

numberChemical Supplier

Lot

number

Purity

(%)Certified

Solvent of

stock/working

standard

Internal standard

1 Mancozeb (used for spiking) Dr. Ehrenstorfer 40617 73.5 Yes N/A Not used

2 Carbon disulfide Sigma STBF6705V 100 Yes Isooctane Not used

3Sodium diethyldithiocarbamic acid

sodium salt trihydrate Chem Service 3791900 99.5 Yes Distilled water/N/A Not used

4 Carbon disulfide VWR/Merck ED096610 99.9 Yes Isooctane Not used

5N,N-diethyldithiocarbamate sodium

salt trihydrateDr. Ehrenstorfer 41024 99.5 Yes Water Not used

6Sodium diethyldithiocarbamate

trihydrateBDH Chemicals 10244 98.5 Distilled water Not used

7 Carbon disulfide Kanto 103G1315 99.9 Yes Ethanol/Cyclohexane Not used

8 Carbon disulfide Chem Service 2265500 99 No Isooctane Not used

9 Thiram Fluka SZBB193XV 99.9 Yes Acetone Not used

10Sodium diethyldithiocarbamate

trihydrateSigma-Aldrich MKQB7086 Yes Purified water Not used

11 Carbon disulfide Chem Service 2265500 99 No Isooctane Not used

12 Thiram Dr. Ehrenstorfer 21024 99.5 Yes Thiophen

13 Sodium diethyldithiocarbamate Sigma-Aldrich MKBS4219V 100 No Water Not used

14 Carbon disulfide Sigma-Aldrich SZBE118SV 99.9 Yes Isooctane Not used

15 Carbon disulfide Sigma-Aldrich 11073BD >99.9 Yes Isooctane Not used

16 Carbon disulfide Chem Service 3,924,900 100 µg/mL Yes Isooctane Not used

18 Carbon disulfide Dr. Ehrenstorfer 80714 99.5 Yes Toluene/Isooctane Not used


Table A4.2: Calibration applied by each laboratory

BUR FCV

1 Potassium-O -methyl dithiocarbanate External solvent standards 5 Forced None 0.99 0.99

2 Carbon disulfide Procedural standards 1 - - - -

3 Potassium-O-methyl dithiocarbanate Procedural standards 8 Excluded None 0.9972 0.9972

4 Carbon disulfide External solvent standards 4 Excluded None 1.0 1.0

5 Carbon disulfide Matrix-matched standards 4 Included None 0.999 0.999

6 Potassium-O -methyl dithiocarbanate Procedural standards 5 Included None 0.997 0.993

7 Carbon disulfide Matrix-matched standards 4 Excluded None


9 Carbon disulfide Matrix-matched standards 5 Excluded None 0.999 0.999

10 Potassium-O -methyl dithiocarbanate Procedural standards 12 Included None 0.9979 0.9979


12 Carbon disulfide Matrix-matched standards 3 Forced None 0.9919 0.9977

13 Potassium-O -methyl dithiocarbanate Procedural standards 5 Excluded None 0.9999 0.9999

14 Carbon disulfide External solvent standards 5 Excluded 1/X 0.998 0.998

15 Carbon disulfide External solvent standards 5 Excluded 1/X 0.999 0.999


18 Carbon disulfide External solvent standards 6 Excluded 1/X 0.99

Lab

numberAnalyte

Calibration

TypeNumber of

standardsOrigin

Curve

weighting

r2

0.9982 (using FCV blank)


Several different calibrations of quantitation were used for this JETS in participating

laboratories:

- 8 laboratories; External solvent standards

- 5 laboratories; Procedural standards

- 4 laboratories; Matrix matched standards

Standard chemicals included maneb, carbon disulfide, sodium Sodium diethyldithiocarbamic

acid sodium salt trihydrate, N,N-diethyldithiocarbamate sodium salt trihydrate, sodium

diethyldithiocarbamate trihydrate, thiram, sodium diethyldithiocarbamate. However analyte

chemicals were carbon disulfide and potassium-O-methyl dithiocarbamate. The former

analyte was detected by GC-MS, GM-FPD or GC-ECD, while the latter was all measured by

spectrophotometer.


APPENDIX 5: Questions & Comments

Answers from each laboratory to the following questions are described in Table A5.1:

- How much time did you allow between spiking the recovery and adding the extraction

solvent?

- Is the method used for Dithiocarbamates analysis accredited (ISO 17025)?

Table A5.1: Each laboratory’s answers to the questions

Lab

number

Time between

spiking the recoveries

and

adding the extraction solvent

(min.)

Is the method used for

Dithiocarbamates analysis

accredited (ISO 17025)?

1 0 Yes

2 5 Yes

3 10 Yes

4 15 Yes

5 1 Yes

6 120 No

7 0 No

8 Above 120 No

9 0.3 Yes

10 Approximately 5 Yes

11 Above 120 No

12 5 Yes

13 3 Yes

14 30Yes in fruit and vegetable,

but not in tobacco

15 15Yes in fruit and vegetable,

but not in tobacco

16 5-10 No

18 15 Yes


The comments submitted from participating laboratories are shown below:

Lab 3

Samples refluxed for approximately 40 minutes and analyzed on UV using wavelengths

272nm and 332nm as background points and 302nm for the peak.

Lab 6

UV-vis Spectrophotometer

Spectrophotometric measurements were made at 272, 302 and 332 nm, using a 10 mm quartz

cell, against a reagent blank of 25 mL potassium hydroxide reagent plus 25 mL distilled water.

Calibration

A solution of 59.2 μg/mL sodium diethyl-dithiocarbamate 3H2O, equivalent to 20 μg CS2/mL

was prepared in water. A range of standards, equivalent to 40, 60, 80, 100, 120 and 160 μg

CS2 were prepared by analyzing 2, 3, 4, 5, 6 and 8 mL of this solution under conditions

identical to those used for the analysis of tobacco.

A calibration curve was prepared by plotting amount of CS2 in μg against extinction (ΔE).

equation (1)

ΔE = E302 E272 E332

2

A calibration factor (f) was calculated from the slope of the calibration graph

equation (2)

𝑓 =ΔE

𝜇𝑔 CS2

The amount of CS2 in moisture free tobacco expressed in mg CS2 per kg moisture free

tobacco (ppm) is:

CS2 in mg/kg =∆𝐸 𝐶 100

𝑓 𝐶 𝑀 (100 − 𝑊)

ΔE = extinction, corrected as formula (1)

f = calibration factor calculated as formula (2)

M = tobacco weight (g)

W = moisture content of tobacco (%)


Lab 10

Instrumentation and Standards:

Sample solutions are read against a blank at three ultraviolet wavelengths (332, 302 and 272

nm) on a recording spectrophotometer. Calculation is based on comparison of the sample

potassium o-methyl dithiocarbamate net absorbance to a dithiocarbamate standard curve

processed through the method and application of the method dilution factor. Since it is not

normally known which dithiocarbamate is present, results are expressed in total

dithiocarbamates as carbon disulfide.

Lab 13

I. Instrumental Technique

The instrument that used for determination Dithiocarbamates is UV-Vis Spectrophotometer at

272, 302 and 332 nm

II. Procedural Standard Calibration

Procedural standards are prepared by spiking a series of blank test portions with different

amounts of analyte, prior to extraction. The procedural standards are then analyzed in exactly

the same way as the samples.

The preparation of calibration curve by plotting amount of CS2 in μg/mL against extinction

(ΔE), 1.00; 2.00; 4.00; 8.00 and 10.00 calculate using the following formula:

ΔE = E302 - (E272 + E332)

2

Where:

ΔE = corrected extinction

E272/302/332 = measured extinction at each wavelength

The determination of Dithiocarbamates in sample is express as CS2.

The amount of CS2 in tobacco expressed in mg CS2 per kg (ppm) is:

CS2 in kg/mg = CS2 (μg/mL) x V (mL)

W (gram)

Where:

CS2 = Calculated CS2 by using linear regression of calibration plot (μg/mL)

W = Sample weight (gram)

V = Volume Extraction (mL)


Lab 16

Testing repeated due to low recoveries. Normal recoveries are 60 – 95 %.

Lab 18

We have only performed recovery tests at the 5 ppm level, as requested in the protocol.

It has been noticed with dithiocarbamates, that recoveries might be lower at higher spike

levels, depending on the matrix (type).

This effect should be kept in mind when evaluating the results of the FCV sample for which

lower residue levels were detected. If recovery correction has to applied, this might lead to an

extra error due to the possibly concentration dependent recoveries.

Extra experiments might be necessary to study this.

Also, taking lower sample amounts can help to increase recoveries for those sample types

normally showing relatively lower recoveries. However, this will consequently lead to higher

method-LOQs.

Documents

Joint Experiment Technical Study (JETS) Final Report 16/1 … · 2017-04-06 · AA-075-CTR JETS Final Report 16/1 Dithiocarbamates in Tobacco – April 2017 5/32 4. Participating