The classification history of EGBE (2-butoxyethanol)
During this period, the classification of substances hazardous to humans and the environment was carried out according to the Dangerous Substances Directive 67/548. Annex I of the directive listed those substances for which a harmonised classification had been agreed at EU level. The entry for EGBE was as follows:
|Classification end point||Classification||Risk phrase||Special concentration limits|
|Inhalation||Xn (Harmful)||R20||2.5% (default 25%)|
|Skin||Xn (Harmful)||R21||12.5% (default 25%)|
|Oral||Xn (Harmful)||R22||12.5% (default 25%)|
Under the Dangerous Preparations Directive 88/379, this resulted in classification of mixtures as follows :
|>=20%||Harmful by inhalation, dermal and oral exposure.||Xn; R20/21/22|
|Irritant to the respiratory system||Xi, R37|
|>=12.5% but <20%||Harmful by inhalation, dermal and oral exposure.||Xn; R20/21/22|
Mixtures containing less than 12.5% EGBE were unclassified.
Change in classification in 2000
Following the publication in 1998 by the National Toxicology Programme of the results of their cancer study of EGBE in rats and mice, the following year the French authorities proposed that the classification of EGBE was updated to include a category 31 classification for cancer. They also proposed changing the classifications for irritancy, adding classifications for skin irritation (R38) and irreversible damage to the eyes (R41)2. Alongside this, they proposed dropping the R37 classification as there did not appear to be any evidence to support it. They also proposed considering adding a reprotoxicity classification category 3.Following the publication in 1998 by the National Toxicology Programme of the results of their cancer study of EGBE in rats and mice, the following year the French authorities proposed that the classification of EGBE was updated to include a category 3 classification for cancer. They also proposed changing the classifications for irritancy, adding classifications for skin irritation (R38) and irreversible damage to the eyes (R41). Alongside this, they proposed dropping the R37 classification as there did not appear to be any evidence to support it. They also proposed considering adding a reprotoxicity classification category 3.
This proposal was discussed by the Commission Working Group on the Classification and Labelling of Dangerous Substances at their October 1999, January and May 2000 meetings. Based on the input from industry and the additional information from existing literature and new studies, the Working Group eventually rejected both the cancer and reproductive toxicity classification proposals. Following a late industry study from BASF, they agreed that an R36 classification for eye irritation was more appropriate than R41. At the request of industry, they also reviewed the existing acute toxicity classification. The issue of choice of species was raised3 but the Working Group were minded to stick to the use of ‘preferred species’ and so the current classification was retained. However, the Working Group accepted that there was no recorded basis for the special concentration limits and nobody could justify why they were required, therefore these were deleted. Formal ratification of this change to the classification took place at a technical progress meeting in January 2001 with the following classification resulting:
|Classification end point||Classification||Risk phrase||Special concentration limits|
Under the Dangerous Preparations Directive 88/379 and its replacement 99/45, this resulted in classification of mixtures as follows :
|>=25%||Harmful by inhalation, dermal and oral exposure.||Xn; R20/21/22|
|Irritant to eyes and skin||Xi, R36/38|
|>=20% but <25%||Irritant to eyes and skin||Xi, R36/38|
This could be regarded as a reduction in the classification requirement for mixtures, as those containing up to 20% EGBE were now unclassified and the harmful classification was not triggered until 25% EGBE.
Review of classification in 2006
As part of the EU risk assessment of EGBE under the Existing Substances Regulation 793/93, the classification of EGBE was again reviewed. The French Competent Authority were again the rapporteurs and the same end points were again discussed. However, this time, there was no proposal to deviate from the prevailing harmonised classification and the risk assessment concluded that the classification agreed in 2000 was still appropriate and was retained going forward.
Migration from Dangerous Substances and Preparations directive to CLP regulation
When the CLP regulation 1272/2008 superseded the old DSD directive, the harmonised classification under the latter was ‘read across’ to the former. The classification limits are not identical between the directive and regulation, which means that the migrated classification was potentially recognised as not correct, with the obligation for those supplying the substance and mixtures containing it to check the data and amend this ‘minimum’ classification accordingly. (In fact, this is the reason why the discussion on the correct classification of EGBE first came about.). Under the listed CLP harmonised classification, mixtures were obliged to be classified at least as follows:
|>=55%||Eye irrit cat 2; Skin irrit cat 2||H315, H319|
|Acute tox inha, cat 4; Acute tox dermal cat 4; Acute tox oral cat 4||H302, H312, H332|
|>=25% but 55%||Eye irrit cat 2; Skin irrit cat 2||H315, H319|
|Acute tox oral, cat 4||H302|
|>=10% but <25%||Eye irrit cat 2; Skin irrit cat 2||H315, H319|
Review of classification in 2018
With the change in the classification regime from the outgoing Dangerous Substances Directive to the CLP regulation, and the change in classification thresholds under the latter, it was inevitable that the acute toxicity classification of EGBE would again come under review eventually. In October 2017, ECHA published for public comment a proposal from Germany to amend the acute toxicity classification of EGBE from harmful to toxic by the inhalation and dermal routes, to change the eye irritation classification from category 2 (severe irritant) to category 1 (eye damage) and to add a specific target organ toxicity by repeated exposure (STOT-RE) category 2 classification due to the haemolysis seen in rats, mice and rabbits. The dossier submitter also proposed to add specific ATE values for acute toxicity (although these were actually the default values in the regulation.)
This proposal was eventually discussed at the Risk Assessment Committee meeting in September 2018. In this case, the committee were prepared to accept that the guinea pig was a more appropriate species to use to model toxicity in humans. For that reason they decided that the classification for dermal toxicity should be deleted altogether and, whilst the classification for oral toxicity was correct, the ATE value should be above the default when based on guinea pig data. Unfortunately, there were only two guinea pig studies considered for the inhalation route. One of these was very old and considered unreliable in terms of the exposure conditions, whilst a modern, GLP study was only for one hour exposure time. The RAC considered neither robust enough for classification purpose, which therefore meant they retained their use of rat data and upheld the toxic classification proposal.
The eye irritation classification proposal was rejected and the existing classification retained on the basis that the dataset was still identical to that considered in 2006 and therefore there was no need to revisit the decision, particularly since the critical issue seen in the available studies is recovery time and this has not changed between the DSD and CLP approaches. Finally, the STOT(RE) classification was rejected since the sub-chronic data (preferred approach) did not support classification, even without considering the issue of haemolysis and relevance to humans.
As of January 2020, the RAC proposal for re-classification had not been accepted in its entirety by the EU Commission. This is due to uncertainty regarding the correct interpretation of the data on acute inhalation toxicity that was presented to the RAC and the availability of new data that further questions the validity of the conclusion of category 3 for the inhalation route. Until such a review of the decision is undertaken the current classification (category 4 for acute toxicity by all routes of exposure) remains in force.
The resulting classification once implemented into legislation via the 15t ATP will therefore be as follows:
|Endpoint||Future Annex VI|
|Acute Tox oral||Cat 4|
ATE: 1200 mg/kg
|Acute Tox inhalation||Cat 4*|
|Acute Tox dermal||No classification|
|Skin||Skin Irrit 2|
|Eye||Eye Irrit 2|
|STOT RE||No classification|
1- Equivalent to category 2 under the GHS system/CLP regulationEquivalent to category 2 under the GHS system/CLP regulation
2- The classification for these end points was wrong, with the data clearly indicating R36/38 as appropriate so a revision of the classification for these end points was clearly justified.
3- It was requested that the guinea pig was used as the species on which to base the LD50 values rather than rodents and rabbit because of the known sensitivity of the latter to haemolysis and the fact that both guinea pigs and humans are not sensitive to this effect.