Bis[(2-ethyl-1-oxohexyl)oxy]dioctylstannane

Endpoint summary

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

dermal absorption in vivo

Type of information:

experimental study

Adequacy of study:

key study

Study period:

2013

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Reason / purpose for cross-reference:

other: read-across target

Qualifier:

according to guideline

Guideline:

other: OECD 402

Deviations:

yes

Remarks:

Additional determination of dermal absorption of the test item (DOTL)

Principles of method if other than guideline:

Additional determination of dermal absorption of the test item (DOTL), via Sn in plasma

GLP compliance:

yes

Radiolabelling:

no

Species:

rat

Strain:

Wistar

Sex:

male/female

Type of coverage:

semiocclusive

Vehicle:

unchanged (no vehicle)

Duration of exposure:

24h

Doses:

2000 mg/kg bw

No. of animals per group:

5 female, 5 male

Control animals:

no

Signs and symptoms of toxicity:

no effects

Dermal irritation:

no effects

Time point:

3 h

Dose:

2000 mg/kg

Parameter:

percentage

Absorption:

0 %

Time point:

24 h

Dose:

2000 mg/kg

Parameter:

percentage

Absorption:

0 %

Conversion factor human vs. animal skin:

Not relevant, since no absorption was detected

Conclusions:

The study proves, that no test material has been absorbed via the dermal route.

Reference 2

Endpoint:

dermal absorption in vivo

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

supporting study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Study conducted on read-across material

Justification for type of information:

Read-across to structurally similar substance.

Reason / purpose for cross-reference:

read-across source

Time point:

3 h

Dose:

2000 mg/kg

Parameter:

percentage

Absorption:

0 %

Time point:

24 h

Dose:

2000 mg/kg

Parameter:

percentage

Absorption:

0 %

Reference 3

Endpoint:

dermal absorption in vivo

Type of information:

experimental study

Adequacy of study:

key study

Study period:

2013

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

other: see 'Remark'

Remarks:

extension of an guideline study under glp in pharmaceutical institut; Dioctyltin dilaurate and Tin, dioctylbis(2,4-pentanedionato-κO2,κO4)- hydolyses at once in contact with humidity and / or water at once into dioctlytin oxide and lauric acid respepective intto 2,4-Pentadione. For the resulting tin level in the blood the hydrolysis product Dioctlytin oxide is responsible

Reason / purpose for cross-reference:

other: read-across target

Qualifier:

according to guideline

Guideline:

other: OECD 402

Deviations:

yes

Remarks:

additional blood sampling and quantification of Sn in plasma

Principles of method if other than guideline:

Method in OECD 402 was follwoed; additinal bloold sampling for qauntification on Tin in plasma at hour 3 and 24 of treatment

GLP compliance:

yes

Radiolabelling:

no

Species:

rat

Strain:

other: HsdHan:WIST

Sex:

male/female

Type of coverage:

semiocclusive

Vehicle:

unchanged (no vehicle)

Duration of exposure:

24 h

Doses:

2000 mg/kg bw

No. of animals per group:

5 male and female

Control animals:

no

Time point:

3 h

Dose:

2000 mg/kg bw

Parameter:

percentage

Absorption:

ca. 0 %

Remarks on result:

other: LL0Q = 5 ng/mL

Conversion factor human vs. animal skin:

no conversion factor, in due concentration of Sn in plasma of rats < 5 ng/ml (LL0Q)

Conclusions:

The test material is not bioavailable via dermal exposure

Reference 4

Endpoint:

dermal absorption in vivo

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

supporting study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Study conducted on read-across material

Justification for type of information:

Read-across to structurally similar substance.

Reason / purpose for cross-reference:

read-across source

Time point:

3 h

Dose:

2000 mg/kg bw

Parameter:

percentage

Absorption:

ca. 0 %

Remarks on result:

other: LL0Q = 5 ng/mL

Reference 5

Endpoint:

basic toxicokinetics in vitro / ex vivo

Type of information:

experimental study

Adequacy of study:

supporting study

Study period:

Not reported

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: see 'Remark'

Remarks:

The study was performed using a non pre-validated in vitro method. The method was designed in line with good scientific principles and reported to a sufficiently high standard to assess the quality of the data presented. As such the study was considered reliable with restriction.

Reason / purpose for cross-reference:

other: read-across target

Objective of study:

other: biotransformation

Qualifier:

no guideline followed

Principles of method if other than guideline:

A series of hydrolysis tests were performed at low pH (~1-2 in 0.07 N HCl) at 55-60 °C (due to problems with the solubility of the test material) to simulate the mammalian gastric system. The rate and degree of hydrolysis were studied by analysing the amount of DOTC. The concentrations of the hydrolysis products were determined with gas chromatrography mass spectroscopy (GC-MS).

GLP compliance:

no

Radiolabelling:

no

Species:

other: Test material exposed to low pH to simulate gastric hydrolysis

Route of administration:

other: Test material was added to two different media

Vehicle:

unchanged (no vehicle)

Duration and frequency of treatment / exposure:

The actual times for the exposure of the test substance to the test media was not reported

Remarks:

Doses / Concentrations:
The mass of test substance added to the media was not reported

Metabolites identified:

no

Details on metabolites:

The extraction of DOTC from 0.07 N HCl with dichloromethane yielded a recovery of less than 5 %, the procedure was found to be unsuitable. The additional verification experiments carried out to improve the method all yielded comparable results.
Several additional verification experiments were carried out to develop an improved method. Based collectively on prior knowledge of ecotoxicity testing on the substance and the observations in this study, the author concluded (with consultation with the sponsor) that DOTO would not sufficiently hydrolyse under gastric conditions. DOTO appeared to be relatively resistant to the simulated gastric hydrolysis.
Experiment:
A) Approximately 5 % hydrolysis of DOTO to DOTC occurred
B) Approximately 27 % hydrolysis of DOTO to DOTC occurred
C) Approximately 2 % hydrolysis of DOTO to DOTC occurred

During the course of the study it was found that the only way the test substance could be dissolved was hydrolysing to the acetate using acetic acid.
It was noted that the particle size of the test substance influenced the level of DOTC formed and was considered to be the cause of the high variability between replicates. Increased temperatures higher than 37 °C (up to 60 °C) increased the speed of hydrolysis.
The percentage of hydrolysis never exceeded 55 % and in most cases were considerably lower.
A gastric hydrolysis study on the test substance was found not to be possible.

Conclusions:

No meaningful results were obtained, DOTO was extremely insoluble in the test system and only partially hydrolysed
It was not possible to measure the hydrolysis of the test substance due to the difficulties presented with the analytical procedures and the physico-chemical properties of the test substance. It was concluded that DOTO would not fully hydrolyse in the test system.

Executive summary:

A series of hydrolysis tests were performed at low pH (~1-2 in 0.07 N HCl) at 55-60 °C (due to problems with the solubility of the test material) to simulate the mammalian gastric system. The rate and degree of hydrolysis were studied by analysing the amount of DOTC. The concentrations of the hydrolysis products were determined with gas chromatrography mass spectroscopy (GC-MS). Low recovery was noted, and several verification studies were performed in an attempt to improve the method used. It was not possible to measure the hydrolysis of the test substance due to the difficulties presented with the analytical procedures and the physico-chemical properties of the test substance. It was concluded that DOTO would not fully hydrolyse in the test system.

Reference 6

Endpoint:

basic toxicokinetics in vitro / ex vivo

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

supporting study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Study conducted on read-across material

Justification for type of information:

Read-across to structurally similar substance dioctyltin oxide (Public name: DOTO, CAS: 870-08-6, EC: 212-791-1), see attached justification.

Reason / purpose for cross-reference:

read-across source

Metabolites identified:

no

Details on metabolites:

The extraction of DOTC from 0.07 N HCl with dichloromethane yielded a recovery of less than 5 %, the procedure was found to be unsuitable. The additional verification experiments carried out to improve the method all yielded comparable results.
Several additional verification experiments were carried out to develop an improved method. Based collectively on prior knowledge of ecotoxicity testing on the substance and the observations in this study, the author concluded (with consultation with the sponsor) that DOTO would not sufficiently hydrolyse under gastric conditions. DOTO appeared to be relatively resistant to the simulated gastric hydrolysis.
Experiment:
A) Approximately 5 % hydrolysis of DOTO to DOTC occurred
B) Approximately 27 % hydrolysis of DOTO to DOTC occurred
C) Approximately 2 % hydrolysis of DOTO to DOTC occurred

During the course of the study it was found that the only way the test substance could be dissolved was hydrolysing to the acetate using acetic acid.
It was noted that the particle size of the test substance influenced the level of DOTC formed and was considered to be the cause of the high variability between replicates. Increased temperatures higher than 37 °C (up to 60 °C) increased the speed of hydrolysis.
The percentage of hydrolysis never exceeded 55 % and in most cases were considerably lower.
A gastric hydrolysis study on the test substance was found not to be possible.

Reference 7

Endpoint:

basic toxicokinetics in vitro / ex vivo

Type of information:

experimental study

Adequacy of study:

key study

Study period:

21 September 2016 to 07 November 2017

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

guideline study with acceptable restrictions

Qualifier:

according to guideline

Guideline:

other: OECD Guideline 111 (Hydrolysis as a Function of pH)

Version / remarks:

2004

Deviations:

no

Qualifier:

according to guideline

Guideline:

other: EU Method C.7 (Degradation: Abiotic Degradation: Hydrolysis as a Function of pH)

Deviations:

no

GLP compliance:

no

Remarks:

(not required as study is not a toxicological/ ecotoxicological study)

Radiolabelling:

not specified

HYDROLYSIS AT PH 4,7 AND 9

- At pH 4 and 7 the 119Sn-NMR spectra were analogous to the unhydrolised substance.

- The pH 9 spectrum of the extracted hydrolysate shows a significant decrease of the product peak  at -156 ppm and two sharp peaks (of equal intensity) at -224 and -222 ppm.

- Acyl-substituted teraorganodistannoxanes are reported in literature to be formed as hydrolysis products of dialkyltincarboxylates, following an initial formation of a hydroxide. The reported chemical shifts correlate well with the shifts found in the extract of the hydrolysate.

- So the formation of a dimeric teraoctyl distannoxane is most likely breakdown product of the test material at pH 9.

HYDROLYSIS AT PH 1.2

- The 119Sn-NMR spectrum of the organic extract the substance signal broadens significantly and decrease to 39.5 Mol% in intensity.

- An additional broad peak appears at -34 ppm (52 Mol%).

- This behaviour is similar to the simulated gastric breakdown under the same conditions of DBTL and DOTL.

- The broad signal can be attributed to the formation of dioctyltin chloro 2-ethylhexanoate a monochloroester of the test material.

- The broadening of the signals may point to a intermolecular conjugation of the test material itself or with DOTO. However this is speculative.

- In contrast to the mentioned dioctyltin di laurates here no signals are found which indicate the formation of a dimeric tetraoctyl dichloro distannoxane.

- No DOTC was formed under the conditions of the study

MASS BALANCE RECOVERY RATES

pH 4: 95 %

pH 7: 91 %

pH 9: 77 %

pH 1.2 97 %

ATOMIC ABSORPTION SPECTOMETRY

- The aqueous phase of the low pH hydrolysis has been analysed after extraction with hexane by AAS and contained < 5mg/L Sn. So the formation of an organotin substance better soluble in water than in hexane can be excluded.

Conclusions:

Under the conditions of this study, the test material can be considered hydrolytically stable at pH 4 and 7.
After 5 days of hydrolysis at 50 °C at pH 4 and 7 no reaction could be identified based on the 119Sn NMR spectra of the extracted hydrolysates.
At pH 9 under the same conditions the formation of a dimeric teraoctyl distannoxane is the most likely breakdown product.
Under simulated gastric conditions the test material breaks down to its monochloroester. The Formation of DOTC and a dimeric teraoctyl dichloro distannoxane can be excluded. The tin content in the aqueous phase of the hydrolysis remained under the detection limit of < 5 mg/L in AAS. Thus the formation of a water soluble organotin species can be excluded.

Executive summary:

The hydrolysis of the test material as a function of pH was investigated in accordance with the standardised guidelines OECD 111 and EU Method C.7.

The stability of the test material was investigated at pH 4, 7 and 9 and pH 1.2 using NMR spectroscopy.

Under the conditions of this study, the test material can be considered hydrolytically stable at pH 4 and 7.

After 5 days of hydrolysis at 50 °C at pH 4 and 7 no reaction could be identified based on the 119Sn NMR spectra of the extracted hydrolysates.

At pH 9 under the same conditions the formation of a dimeric teraoctyl distannoxane is the most likely breakdown product

Under simulated gastric conditions the test material breaks down to its monochloroester. The Formation of DOTC and a dimeric teraoctyl dichloro distannoxane can be excluded. The tin content in the aqueous phase of the hydrolysis remained under the detection limit of < 5 mg/L in AAS. Thus the formation of a water soluble organotin species can be excluded.

Description of key information

Basic Toxicokinetics

Hydrolysis as a function of pH

Under the conditions of this study, the test material can be considered hydrolytically stable at pH 4 and 7.

After 5 days of hydrolysis at 50 °C at pH 4 and 7 no reaction could be identified based on the 119Sn NMR spectra of the extracted hydrolysates.

At pH 9 under the same conditions the formation of a dimeric teraoctyl distannoxane is the most likely breakdown product.

Under simulated gastric conditions the test material breaks down to its monochloroester. The Formation of DOTC and a dimeric teraoctyl dichloro distannoxane can be excluded. The tin content in the aqueous phase of the hydrolysis remained under the detection limit of < 5 mg/L in AAS. Thus the formation of a water soluble organotin species can be excluded.

 

Read-across to structurally similar substance DOTO (Dioctyltin oxide), CAS 870-08-6

- Schilt & Vondervan-van den Beuken (2004): It was not possible to measure the hydrolysis of the test substance due to the difficulties presented with the analytical procedures and the physico-chemical properties of the test substance. It was concluded that DOTO would not fully hydrolyse in the test system.

 

Dermal Adsorption

Read-across to structurally similar substance dioctyltin dilaurate, CAS 3648-18-8

- Dreher (2013):The study proves, that no test material has been absorbed via the dermal route.

- Blackburn (2013): The test material is not bioavailable via dermal exposure

Key value for chemical safety assessment

Bioaccumulation potential:

no bioaccumulation potential

Absorption rate - oral (%):

50

Absorption rate - dermal (%):

1

Absorption rate - inhalation (%):

100

Additional information

Introduction

The substance is bis[(2-ethyl-1-oxohexyl)oxy]dioctylstannane (EC 246-325-3, CAS 24577-34-2)which is also referenced as DOT(EH)2.

Physico-chemical properties and the results ofin vitroandin vivostudies with DOT(EH)2, DOTO (dioctyltin oxide, EC 212-791-1, CAS 870-08-6) and dioctyltin dilaurate (CAS 3648-18-8, EC 222-883-3) have been used to determine a toxicokinetic profile as detailed below.

Since limited data are available forDOT(EH)2, read-across has been used to the data from two structurally similar substances DOTL and DOTO:

 

	Target Substance	Source Substance 1	Source Substance 2
Chemical Name	bis[(2-ethyl-1-oxohexyl)oxy]dioctylstannane (DOT(EH)2)	Dioctyltin dilaurate (DOTL)	Dioctyltin oxide (DOTO)
CAS No	24577-34-2	3648-18-8	870-08-6
EC No	246-325-3	222-883-3	212-791-1
Structure

 

Physicochemical properties

The substance bis[(2-ethyl-1-oxohexyl)oxy]dioctylstannane (DOT(EH)2) is a colourless turbid solid. Its molecular formula is C32H64O4Sn, with a molecular weight (MW) of 631.57 g/mol. Its water solubility is 1.099 × 10^-7mg/L at 25 °C (practically insoluble) with a calculated (KOWWIN v1.68) log Pow value of 10.489. The vapour pressure was calculated (MPBPVP v1.43) to be 705 Pa at 25 °C, based on the mean of Antoine & Grain methods.

 

Oral absorption

The high molecular weight in combination with the low water-solubility and high octanol/water partition coefficient indicate that DOT(EH)2 is unlikely to be readily absorbed from the gut.

In anin vitrohydrolysis study (Nasshan 2017), DOT(EH)2 was shown to be hydrolytically stable at pH 4, 7 and 9. Under simulated gastric conditions (i.e. pH 1.2), DOT(EH)2 breaks down to its monochloroester, but no DOTC and/or a dimeric teraoctyl dichloro distannoxane are formed.

In anin vitrostudy (Schilt & Zondervan-van den Beuken 2004) simulating gastric conditions to determine the extent of hydrolysis of DOTO to dioctyltin chloride, no meaningful results were obtained, because DOTO was extremely insoluble in the test system and only partially hydrolysed. It was noted that the particle size of the test substance influenced the level of dioctyltin chloride formed and was considered to be the cause of the high variability between replicates. Increased temperatures to higher than 37 °C (up to 60 °C) increased the speed of hydrolysis. The percentage of hydrolysis never exceeded 55% and in most cases was considerably lower (2-5%); it was concluded that DOTO would not hydrolyse to an appreciable extent under gastric conditions. DOTO appeared to be relatively resistant to the simulated gastric hydrolysis. 

The available information on DOT(EH)2 regarding acute oral toxicity supports limited oral absorption, as the reported LD₅₀values are generally higher than a limit dose of 2000 mg/kg bw (Pooles A., 2018).

Some oral absorption is demonstrated by results of a fully compliant OECD 422 study performed with the read-across substance DOTO (Waalkens-Berendsen 2004), using the dietary route of exposure. Treatment at a dietary level of 25 ppm of DOTO resulted in adverse effects on thymus; at a higher dietary concentration of 250 ppm some signs of liver involvement were also noted. 

As a worst-case estimate, for human health risk assessment purposes oral absorption of DOT(EH)2 is considered to be 50%.

Dermal absorption

No information is available on dermal absorption of DOT(EH)2.

In the dermal toxicity study on DOTL (Blackburn et al., 2013), there was no absorption of tin into the blood of the study animals found. In a dermal absorption study in male and female Han Wistar rats (Dreher, 2013) there was no percutaneous absorption after 3 hours at 2000 mg/kg bw.

Based on physicochemical data available for DOT(EH)2, it is concluded that the substance is lipophilic and practically not soluble in water. Therefore the rate of transfer between the stratum corneum and the epidermis is expected to be slow and will limit absorption across the skin.

This is further supported by the lack of any skin irritating or corrosive properties, as determined inin vitrostudies with DOT(EH)2 (Lacey 2018), the lack of any sensitisation potential as determined in a fully compliant Local lymph Node Assay performed with the read-across substance DOTO (Sanders 2012), and the acute dermal LD50 greater than 2000 mg/kg bw also determined with the read-across substance DOTO (Sanders 2012). 

Therefore, for human health risk assessment and using a really conservative approach, dermal absorption is estimated 1%.

 

Inhalation absorption

No information is available on inhalation toxicity of the substance, or on inhalation absorption.

DOT(EH)2 is a colourless turbid solid that is not described as being particulate in nature. It is therefore unlikely that DOT(EH)2 would reach the lower tract of the respiratory system, as the proportion of the test material having an inhalable particle size of <100 μm, a thoracic particle size of <10.0 μm or a respirable particle size of <5.5 μm is expected to be practically zero. Furthermore DOT(EH)2 has a high molecular weight and is practically insoluble in water.

In the absence of any quantitative data, and using a really conservative approach, for human health risk assessment purposes absorption by inhalation of DOT(EH)2 is assumed to be 100%.

 

Distribution

No data are available on DOT(EH)2.

Results of the OECD 422 study on DOTO suggest that once absorbed it is widely distributed within the body, as the target organ was the thymus and, at higher doses,the liver. Given the high estimated log Pow and the extremely low water solubility, micellular solubilisation may play a major role for absorption, and preferential partition to tissues with high lipid content.

Metabolism and Elimination

The substance was determined to be hydrolytically stable at pH 4 and 7; after 5 days of hydrolysis at 50 °C at pH 4 and 7 no reaction could be identified using NMR spectroscopy. After 5 days of hydrolysis at 50 °C at pH 9 the formation of a dimeric teraoctyl distannoxane was considered to be the most likely breakdown product. Under simulated gastric conditions the monochloroester is formed however the extent of hydrolysis was very low (recovery of the unchanged molecule was 97%). The tin content in the aqueous phase of the hydrolysis remained under the limit of detection (< 5 mg/L) and therefore the formation of a water soluble organotin species is excluded.

Unchanged DOT(EH)2 and the monochloroester are expected to be excreted mainlyviathe faeces. A bioaccumulation study in fish with a structurally related substance determined aBCF of >100. No upper limit appears to have been verified hence it is not possible to estimate the extent of elimination of the material from fatty tissues.

Conclusion

Based onin vitroandin vivodata from studies performed with DOT(EH)2, and the source substances DOTO and DOTL, for risk assessment purposes DOT(EH)2 is considered to be absorbed orally, dermally andviainhalation at 50 %, 1 % and 100%, respectively.

The potential for bioaccumulation is low, and absorption of unhydrolyzed DOT(EH)2 is expected to be low and any potentially absorbed amount to be completely excreted in faeces.

Read-across Justification

The target substance bis[(2-ethyl-1-oxohexyl)oxy]dioctylstannane (DOT EHA CAS: 24577-34-2, EC: 246-325-3) is a mono-constituent organotin substance that consists of a tin as central metal element with two octyl-ligands. The source substance dioctyltin oxide (Public name: DOTO, CAS: 870-08-6, EC: 212-791-1) is also an organotin compound and has the identical structure elements as the target substance in respect of the tin-alkyl moiety.

According to WHO IPCS CIRCAD (2006) organotin compounds are characterized by a tin–carbon bond and have the general formula RxSn(L)(4−x), where R is an organic alkyl or aryl group and L is an organic (or sometimes inorganic) ligand. The organotin moiety is significant toxicologically. The anionic ligand influences physicochemical properties but generally has little or no effect on the toxicology.

Since the target substance and the source substances share the identical organotin moiety, and the organotin moiety is generally recognized as the relevant toxophore of organotins and the toxicity estimates (AE) respectively toxicity limits for organotins are expressed as tin, the overall ecotoxicity/systemic toxicity of the target can be interpolated by assessing the (eco-)toxicity of the source (WHO IPCS CIRCAD, 2006, BAUA AGS TRGS 900, 2014, Summer KH, Klein D and Greim H, 2003). As the source has a higher relative amount of tin based on molecular weight (source 32.87%, target 18.80%) the read-across approach to DOTO can be assessed as conservative (worst case).

The purity of the source and target substance are expected to be high, based on the manufacturing method. The impurity profile is not expected to have strong effects on substance properties and any impurity of (eco-)toxicological relevance of the source substances is expected to be present in the target substance. Consequently, the hazard profiles of the source substances, including those of their impurities, are intrinsically covered. Differences in impurities are not expected and thus do not have an impact on the (eco-)toxic properties.

The bioavailability of both substances varies in a predictable manner and is assumed to be dependent on the water solubility. The prediction of the effects of the target substance is based on the relationship between solubility and (eco-)toxicity or on a worst-case basis.

 

References

BAUA (Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (Federal Institute for Occupational Safety and Health)) AGS (Ausschuss für Gefahrstoffe (Committee on Hazardous Substances)) TRGS (Technical Rules for Hazardous Substances) 900 (2014). Begründung zu n-Octylzinnverbindungen, April 2014.

Summer KH, Klein D, Griem H (2003). Ecological and toxicological aspects of mono- and disubstituted methyl-, butyl-, octyl-, and dodecyltin compounds - Update 2002. GSF National Research Center for Environment and Health, Neuherberg, for the Organotin Environmental Programme (ORTEP) Association.

World Health Organization (WHO) International Programme on Chemical Safety (IPCS) Concise International Chemical Assessment Document (CICAD) 73 Mono- and disubstituted methyltin, butyltin, and octyltin compounds (2006). Published under the joint sponsorship of the United Nations Environment Programme, the International Labour Organization, and the World Health Organization, and produced within the framework of the Inter-Organization Programme for the Sound Management of Chemicals. World Health Organization ISBN 978 92 4 153073.