Registration Dossier

Administrative data

Description of key information

Effects of gallium arsenide on male and female rats and mice were examined in 16-day, 14-week and 2-year inhalation toxicity studies (NTP, 2000).  Pulmonary alveolar proteinosis leading to chronic active inflammation of the lung were seen in rats and mice as a  specific toxic reaction to the inhalation of a GaAs dust (MMAD < 2 µm).

Key value for chemical safety assessment

Repeated dose toxicity: inhalation - systemic effects

Endpoint conclusion
Dose descriptor:
LOAEC
0.01 mg/m³
Study duration:
chronic
Species:
rat

Mode of Action Analysis / Human Relevance Framework

Additional information

There exist 9 studies with repeated oral administration of GaAs to rats. None of them has a sufficient reliability. Therefore these studies are ignored.

In contrast to the studies with oral administration the 4 studies with inhalative exposure of rats and mice to a very fine dust of GaAs are fully reliable.Their results are summarized and completed with the non-neoplastic results of the 2 -year inhalation carcinogenicity studies with rats and mice.

2 further studies of Tanaka et al. (2000 and 2004) with repeated intratracheal instillation of GaAs are again not sufficiently reliable and therefore ignored.

NTP-Studies:

In a 16 -day dose range finding study mice and rats were exposed to particulate aerosols of GaAs at concentrations of 0, 1, 10, 37, 75 and 150 mg/m³.

In the following 14-week study the concentrations of the GaAs aerosol were 0, 0.1, 1, 10, 37 and 75 mg/m³. On the background of the results of these studies the dosages of the 2 -year inhalation carcinogenicity studies were chosen:

rats were exposed to particulate aerosols of GaAs at concentrations of 0, 0.01, 0.1 and 1mg/m³,

mice were exposed to particulate aerosols of GaAs at concentrations of 0, 0.1, 0.5 and 1.0 mg/m³.

 

The NTP studies (NTP, 2000) with rats and mice revealed that after inhalative exposure to a fine dust of GaAs (MMAD < 2µm) a severe local reaction in the lung is triggered. This reaction can be described as pulmonary alveolar proteinosis leading to chronic active inflammation.

 

In the NTP studies the mass median aerodynamic diameter (MMAD) of the gallium arsenide particles ranged from 0.8 to 1.6 µm.

Particles smaller than 2.5 µm diameter are respirable. These particles can enter the alveoli of the lung (DIN/EN 481). This means that nearly 100% of the GaAs dust of the NTP studies can enter the alveoli of the lung. Here the fine GaAs particles caused a local irritation.The primary target of inhalative exposure to gallium arsenide aerosols at particle sizes (MMAD) of 0.8 – 1.0μm in the NTP studies was the lung.

In the 2-year study with male rats analytical levels of Gallium or Arsenic in the blood are difficult to interpret, no significant differences are given. There seem to be higher levels of Gallium and Arsenic in the blood in the highest exposure group, 1 mg/m³ (NTP, 2000).

The irritating effects of GaAs in the lungs are somehow different in mice and rats: rats are more sensitive than mice to the effects of GaAs in the lung.

In the 2-year study with rats 0.01 mg/m³ caused proteinosis with a considerable number of animals. The proteinosis was increased with higher exposure concentrations. With mice a similar effect was seen after 2-year exposure to 0.1 mg GaAs/m³.

In both studies, in the 2-year rat and in the 2-year mouse study, the animals of the lowest dose group showed weaker effects and a smaller number of animals was affected than in the higher dose groups, nevertheless both studies missed to demonstrate a no-effect level.

As seen in the inhalation toxicity studies GaAs has a specific local irritating effect in the lung if inhaled. Therefore, the dust of GaAs is classified as: specific target organ toxicity after repeated exposure:

STOT Rep. Exp Cat 2, H373 - May cause damage to organs through prolonged or repeated exposure, affected organ: lung, route of exposure: inhalation.

The classification according to COMMISSION REGULATION (EU) No 944/2013 of 2 October 2013 with regard to specific target organ toxicity after repeated exposure: STOT RE 1, H372 (respiratory and haematopoietic systems).


Repeated dose toxicity: inhalation - systemic effects (target organ) respiratory: lung

Justification for classification or non-classification

According to CLP-guidance page 31: „Classification (and labelling) is based on intrinsic hazards, i.e. the basic properties of a substance as determined in standard tests or by other means to identify hazards.“

In this context toxicological data are regarded as intrinsic properties. For a toxicological study the substance/test article has to be administered to the body/organism. It is obvious that this is a problem with substances which are produced in massive form or substances which exist only under extreme (for example pH) conditions. Here the standard tests cannot be used.

Even a more detailed view on the „normal“ substances reveals that the results of toxicological studies can be changed by the formulation used. For drugs it is known that the galenic formulation can be relevant for the pharmacological effect. Of course this is also true for possible undesired/toxic effects of industrial chemicals.

As a result it has to be admitted that toxicological effects are not at all an intrinsic property of a substance like a melting point: they are a specific interaction of a substance with an organism.

However, CLP accepts specific exceptions: „when a chemical can be considered as not biologically available such as the derogation not to label a metal in the massive form“ and of course this should also be the case for the classification.

With GaAs the produced „substance“ is a mono-crystalline bloc (ingot) of several kilogrammes which is normally cut into disks (wafer). These wafers are the product, which is placed on the market. In rare cases a complete crystal (ingot) is sold. This situation corresponds to a metal in the massive form as mentioned in the CLP regulation.

For the classification of the environmental hazards „massives will be usually tested as 1 mm particles“ and „Examples of alternative approaches include the use of Water Accomodated Fractions from the massive form...“. These procedures should be also applicable for the identification and classification of toxicological hazards.

The classification (and labelling) shall be based on the smallest particle size marketed or, if the material is massive, on 1 mm particles as default. For GaAs the powder form is not marketed. As GaAs is not a metal but a metallic compound, it is adequate to evaluate its biological properties as if it is a metal in the massive form.

There are no data that GaAs in its massive form is biologically available. Toxic effects that were observed for unrealistic physical forms (e.g. dust of very small particle size of ca. 2 µm in diameter and below) and unrealistic applications (inhalative exposure), are not relevant in the context of the evaluation of GaAs in its massive form.

As a consequence GaAs in its massive form and therefore also GaAs-wafers do not have to be classified and labelled according to the CLP regulation (and of course according to Directive 67/548/EEC) and subsequent regulations.

During production and processing of the blocs dusts can occur containing gallium arsenide particles, which may lead to exposure to aerosols. These particle dusts/aerosols, however, are not part of the registration under the REACH regulation.

The NTP studies (NTP, 2000) with rats and mice revealed that after repeated inhalative exposure to a fine dust of GaAs a local reaction in the lung is triggered. This reaction can be described as pulmonary alveolar proteinosis leading to chronic active inflammation.

Fine GaAs-particles have a specific locally irritating effect in the lung if it is inhaled.

The particle size distribution at the workplace (Weißhaupt, 2011) is fundamentally different from that in the NTP studies (larger particles at the workplace) which may explain that reports on lung effects in production and processing of the ingots and wafers are absent. The relevance of the results of the NTP studies for humans at the workplace is thus questionable (for a detailed evaluation see Bomhard, 2010a,b).