Registration Dossier

Data platform availability banner - registered substances factsheets

Please be aware that this old REACH registration data factsheet is no longer maintained; it remains frozen as of 19th May 2023.

The new ECHA CHEM database has been released by ECHA, and it now contains all REACH registration data. There are more details on the transition of ECHA's published data to ECHA CHEM here.

Diss Factsheets

Administrative data

Description of key information

As ozone is a gas, testing the acute toxicity via oral and dermal route was omitted according to Annex VIII, column 2.

Various data on the acute toxicity via inhalation is available. The reported LC50 value is within a range of 3.6 to 34 ppm, depending on the exposure time, species and ozone generator used. In a weight-of-evidence analysis, 3.6 ppm (equals 7.06 mg/m3) with an exposure time of 4 h was determined as LC50 value used for the risk assessment in a worst-case approach (Svirbely and Saltzman, 1957).

Key value for chemical safety assessment

Acute toxicity: via oral route

Endpoint conclusion
Endpoint conclusion:
no study available

Acute toxicity: via inhalation route

Link to relevant study records

Referenceopen allclose all

Endpoint:
acute toxicity: inhalation
Type of information:
experimental study
Adequacy of study:
weight of evidence
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
study well documented, meets generally accepted scientific principles, acceptable for assessment
Qualifier:
no guideline available
Principles of method if other than guideline:
No guideline available, study performed before guideline publication
GLP compliance:
no
Test type:
other: research study
Limit test:
no
Specific details on test material used for the study:
Ozone was genereated with electric ozonizer from air. Several different ozonizers were used in this study. The air used for dilution was pretreated accoording different protocols before mixing with the ozone.
Species:
rat
Strain:
Wistar
Sex:
male
Details on test animals or test system and environmental conditions:
Male adult Wistar-derived rats, obtained from a commercial breeder and maintained in this laboratory for several weeks on a ration of Rockland Rat Diet, were used in the inhalation studies. The basic diet and water were available to the animals at all times except when the exposures were in progress.
Route of administration:
inhalation: gas
Type of inhalation exposure:
whole body
Vehicle:
air
Details on inhalation exposure:
Routinely, 5 rats were exposed simultaneously for a single four-hour period to varying concentrations of ozone-air mixtures in the chamber.
Two dielectric-type room air ozonizers, a "plastic type" and "Mica type", differing in current density, were used for the studies. Further, ozone was prepared following 6 different protocols; from scrubbed tank oxygen, from scrubbed or unscrubbed compressed air, from a synthetic air mixture of tank oxygen ( 20%) and water-pumped nitrogen (80 %), from scrubbed air which had been further purified by passage through an electric combustion furnace followed, successively, by columns of Drierite, perchlorate, and Ascarite or from scrubbed air and aged for 20 to 30 minutes in a 10 liter bottle.
Infrared analysis of the scrubbed compressed air used for the toxicity studies indicated that no traces of organic impurities could be detected. Tests for possible ozone contaminants, such as oxides of nitrogen, hydrogen peroxide, and free radicals (H02 , OH, HO3 04 etc.), in a specially constructed mass spectrometer failed to reveal significant amounts of these substances, and, consequently, it is improbable that they affect the toxicity of ozone in laboratory animals.
Analytical verification of test atmosphere concentrations:
yes
Remarks:
Ozone determinations were made with use the method of Byers (1956).
Concentrations:
2.4 ppm to 8.2 ppm
Ozone determinations were made with use the method of Byers (1956).
No. of animals per sex per dose:
In general, 5 animals per exposure group.
Control animals:
no
Details on study design:
Routinely, groups of 5 rats were exposed for a single four-hour period to varying concentrations of ozoneair mixtures in the chamber. Two different ozonizers and six different protocols for ozone preparations were used. In total 11 experiments were performed, see results, mortality
The LD50 values were calculated for a 24-hour period starting from the beginning of the exposure (of 4-hour duration), because the lethal effect of the ozone was usually manifest during this period.
Statistics:
The LD50 was calculated by the method of Thompson and WeiJ (1952) which utilizes moving averages and interpolation to estimate the median effective dose.
A addendum to the orginal paper (Siverbely et.al., 1956) states that based on "recently reported analytical work (Byers et.al., 1956)" the values in this paper shouls be increased by a factor 1.5. "This correction was necessary, since there appeared to be a deviation in the stoichiometry of the alkaline iodide procedure at low concentrations of ozone used in this study."
For this IUCLD entry the original published ozone concentrations have been multiplied with a factor 1.5 at entering in IUCLID.
Preliminary study:
No
Sex:
male
Dose descriptor:
LC50
Effect level:
7.2 ppm
Based on:
test mat.
95% CL:
>= 5.4 - <= 9.6
Exp. duration:
4 h
Remarks on result:
other: PT Scrubbed air
Sex:
male
Dose descriptor:
LC50
Effect level:
7.65 ppm
Based on:
test mat.
95% CL:
>= 5.1 - <= 11.4
Exp. duration:
4 h
Remarks on result:
other: PT Tank oxygen
Sex:
male
Dose descriptor:
LC50
Effect level:
7.5 ppm
Based on:
test mat.
95% CL:
>= 5.85 - <= 9.75
Exp. duration:
4 h
Remarks on result:
other: PT Tank oxygen +nitrogen
Sex:
male
Dose descriptor:
LC50
Effect level:
12.3 ppm
Based on:
test mat.
95% CL:
>= 10.5 - <= 14.25
Exp. duration:
4 h
Remarks on result:
other: PT Scrubbed air
Sex:
male
Dose descriptor:
LC50
Effect level:
10.35 ppm
Based on:
test mat.
95% CL:
>= 9.15 - <= 11.85
Exp. duration:
4 h
Remarks on result:
other: PT Scrubbted air furnace treated
Sex:
male
Dose descriptor:
LC50
Effect level:
3.6 ppm
Based on:
test mat.
Exp. duration:
4 h
Remarks on result:
other: MT Scrubbed air
Sex:
male
Dose descriptor:
LC50
Effect level:
9 ppm
Based on:
test mat.
95% CL:
>= 6.45 - <= 12.45
Exp. duration:
4 h
Remarks on result:
other: MT Tank oxygen
Sex:
male
Dose descriptor:
LC50
Effect level:
4.95 ppm
Based on:
test mat.
95% CL:
>= 3.75 - <= 6.45
Exp. duration:
4 h
Remarks on result:
other: MT Tank oxygen+nitrogen
Sex:
male
Dose descriptor:
LC50
Effect level:
6.15 ppm
Based on:
test mat.
95% CL:
>= 4.35 - <= 8.55
Exp. duration:
4 h
Remarks on result:
other: MT unscrubbed air
Sex:
male
Dose descriptor:
LC50
Effect level:
9.3 ppm
Based on:
test mat.
95% CL:
>= 6 - <= 14.25
Exp. duration:
4 h
Remarks on result:
other: MT Scrubbed air
Mortality:
MT Scrubbed air aged O3 for 20-30 min: male rats LC50 12.3 (7.95-18.9) ppm (did not fit in the IUCLID format)

The data indicate that the apparent toxicity of ozone was not altered by either the type of ozonizer employed or the gas used to generate the ozone.

Ozone prepared from scrubbed tank oxygen appeared to be as toxic as that prepared from scrubbed or unscrubbed compressed air, from a synthetic air mixture of tank oxygen ( 20%) and water-pumped nitrogen (80 %), or from scrubbed air which had been further purified by passage through an electric combustion furnace followed, successively, by columns of Drierite, magnesium perchlorate, and Ascarite. Ozone generated from scrubbed air and aged for 20 to 30 minutes in a 10 liter bottle appeared to be as toxic as freshly prepared ozone.
Clinical signs:
other: Other effects of ozone toxicity, in addition to mortality, appeared to vary with the concentrations employed. At low concentrations signs of respiratory distress and transient convulsive seizures were frequently noted during or shortly after the inhalatio
Body weight:
no data
Gross pathology:
Postmortem examination usually showed pulmonary edema, hemorrhagic lungs, congested livers, and frequent occurrence of slight to markedly darkened adrenals. The degree of these pathologic changes depended on the concentration of ozone and the duration of survival.
Other findings:
It was generally observed in this investigation that the signs of intoxication in ozone-treated animals were severer during periods of high summer temperature. The average chamber temperature was between 24 and 27 °C, but it went as high as 31 °C in the summer months.

An unexpected finding in the course of these studies was that rats which had survived a single four-hour exposure to concentrations of ozone around 2 3 or 7 ppm, respectively, developed a tolerance to ozone when reexposed two or four and one-half weeks later to about double these concentrations.
Interpretation of results:
Category 1 based on GHS criteria
Remarks:
Migrated information
Conclusions:
This scientific publication investigated the acute toxicity of a 4 hour ozone exposure in a range of 3.6 to 12.3 ppm in rats as well as potential difference in ozonizers or gases used to generate ozone. The determined LC50 values range from 3.6-12.3 ppm with no significant altereation by either the type of ozonizer employed or the gas used to generate the ozone. These results lead to a classification as Acute Tox. Cat. 1 according to CLP criteria.
Executive summary:

In an acute toxicity study perfomed before the OECD guidance was published, 5 Wistars rats (sex and age not specified) per group were exposed to ozone (3.6 - 19 ppm) for 4 h in 11 independent whole body inhalation experiments using several different ozonizers varying in current density and gases to generate the ozone. The same test protocoll was applied in additional eyperiments on mice and hamsters revealing no significant differences between the species.

At low concentrations signs of respiratory distress and transient convulsive seizures were frequently noted during or shortly after the inhalation period. When the concentrations approached the LD50, the majority of the animals exhibited dyspnea, became comatose, and died within the 24-hour period from the start of the exposure. In the absence of convulsive seizures, some rats shortly after death showed a slight cloudiness or a complete opacity of the cornea at relatively high concentrations of ozone.The obtained LC50 -values ranges from 3.6 to 12.3 ppm showing no significant differences between the O3 generation protocols.

The injurious effects of ozone appear to be lessened by a previous exposure, to relatively low concentrations of ozone for a short period as it was invesitgated in a nadditionla experiment. This tolerance was apparent for at least four and one-half weeks after exposure.

The publication is a well-documented report which meets basic scientific principles, with great attention for ozone generation protocols and analytical methodes applied at that time. The information from the study is relevant for risk-assessment and provides key information on the survival following ozone exposure. In addition the study provides valueable information on the relation between toxicity and various ozone generation protocols . Based on the results of this study, ozone does warrant for classification as Acute Tox. Cat. 1 accroding to CLP criteria.

Endpoint:
acute toxicity: inhalation
Type of information:
experimental study
Adequacy of study:
weight of evidence
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
study well documented, meets generally accepted scientific principles, acceptable for assessment
Qualifier:
no guideline available
Principles of method if other than guideline:
No guideline available, study was done before guidelines were published.
GLP compliance:
no
Test type:
other: Research study
Limit test:
no
Specific details on test material used for the study:
Ozone was made in situ from oxygen (experiment 1) or dry-air (experiment 2) by a water-cooled corona discharge type of ozonizer designed by Armour Research Foundation.
Species:
rat
Strain:
Wistar
Sex:
not specified
Details on test animals or test system and environmental conditions:
No information provided
Route of administration:
inhalation: gas
Type of inhalation exposure:
whole body
Vehicle:
air
Remarks:
1st experiment made from oxygen and 2nd experiment made from dry air with 2%CO2
Details on inhalation exposure:
The animals were exposed to ozone at a flow rate of 10 to 25 l/min. made by a water-cooled corona discharge type of ozonizer designed by Armour Research Foundation. Rats in groups of five were exposed for three hours in a glass chamber.
In experiment 1 the ozone was made and diluted with 100% oxygen.
In experiment 2 ozone was made from filtered dry-air and enriched with carbon dioxide to form a 2% carbon oxide-air-ozone misture
Analytical verification of test atmosphere concentrations:
yes
Remarks:
A modified aluminum chloride-buffered potassium iodide method was used to determine the ozone concentration.
Concentrations:
In experiment 1: A total of 123 albino Wistar strain rats, in groups of five, were exposed for three hours to ozone concentrations ranging from 5.2 to 17.7 ppm.
In experiment 2: A total of 76 albino Wistar strain rats, in groups of five, were exposed for three hours to ozone concentrations ranging from 5.7 to 16.9 ppm.
No. of animals per sex per dose:
Rats were exposed in groups of 5 animals, 5- 10 for each concentration
Control animals:
not specified
Details on study design:
The purpose of the experimental studies reported was to determine whether 100% oxygen or 2% carbon dioxide will influence the toxicity of ozone for various laboratory animals.
Statistics:
The LD50 values were calculated by the Reed and Muench (1938) method.
Preliminary study:
not applicable
Dose descriptor:
LC50
Effect level:
12.2 ppm
Based on:
test mat.
Exp. duration:
3 h
Remarks on result:
other: O3 made from 100% oxygen
Dose descriptor:
LC50
Effect level:
10.2 ppm
Exp. duration:
3 h
Remarks on result:
other: O3 made from air with 2% CO2
Mortality:
See other findings. The LC50 values for ozone in 100% oxgen and carbon dioxide-air-ozone mixture are not significantly different from the LC50 13.1 ppm in dry air. (Mittler et al 1956)
Clinical signs:
other: no data
Body weight:
no data
Gross pathology:
no data
Other findings:
TABLE I: SURVIVAL OF RATS EXPOSED TO VARIOUS CONCENTRATIONS
OF OZONE IN 100% OXYGEN (Three-hour exposure)
________________________________________________________________________________________________________
Cumulative
________________________________________________________________
Sur
Ozone Sur- Sur- Deaths, vivals.
ppm Deaths vivals Death. vivals Total % %
________________________________________________________________________________________________________
6.2 0 5 0 79 79 0 100
6.6 0 5 0 74 74 0 100
6.3 0 5 0 69 69 0 100
7.0 0 5 0 64 64 0 100
7.7 0 5 0 59 59 0 100
7.9 0 5 0 54 54 0 100
8.8 4 1 4 49 53 7.5 92.5
9.4 0 5 4 48 52 7.7 92.3
9.8 1 4 5 43 48 10.4 89.6
10.1 3 7 8 39 47 17.0 83.0
10.7 2 3 10 32 42 23.7 76.3
10.9 0 4 10 29 39 25.6 74.4
11.6 0 5 10 25 35 28.5 71.5
11.9 7 3 17 20 37 45.9 54.1
12.3 1 4 18 17 35 51.4 48.6
12.8 3 1 21 13 34 61.7 38.3
13.0 2 3 23 12 35 65.7 34.3
18.9 4 5 27 9 36 75.0 25.0
14.3 2 3 29 4 33 87.9 12.1
14.8 4 0 33 1 34 97.1 2.9
16.0 4 0 37 1 38 97.4 2.6
16.5 5 0 42 1 43 97.7 2.8
17.7 2 1 44 1 44 97.8 2.2
________________________________________________________________________________________________________



TABLE II: SURVIVAL OF RATS EXPOSED TO OZONE IN AIR ENRICHED
TO 2 % CARBON DIOXIDE (Three-hour exposure)
________________________________________________________________________________________________________
Cumulative
_________________________________________________________________
Sur
Ozone Sur- Sur- Deaths, vivals.
ppm Deaths vivals Death. vivals Total % %
_________________________________________________________________________________________________________
5.7 0 5 0 28 28 0.0 100.0
6.5 4 0 4 23 27 14.8 85.2
7.6 0 5 4 23 27 14.8 86.2
8.5 0 3 4 18 22 18.8 81.7
9.2 2 3 6 15 21 28.2 71.8
9.9 5 4 11 12 23 47.7 62.3
10.9 8 2 13 8 21 62.0 38.0
11.6 8 1 21 6 27 77.8 22.2
12.3 3 2 24 5 29 82.8 17.2
13.2 6 3 30 3 33 90.4 9.6
14.1 4 0 34 0 34 100.0 0.0
15.5 3 0 37 0 37 100.0 0.0
16.9 5 0 42 0 42 100.0 0.0
_________________________________________________________________________________________________________
Interpretation of results:
Category 1 based on GHS criteria
Conclusions:
This study investigated the dose relationship of ozone in a range of 5-17 ppm, and the influence of the carrier-gas, 100% oxygen and 2% carbon dioxide-air mixture on survival igated. The resutls show that the LC 50 values for ozone in 100% oxygen (LC50 12.2 ppm) and carbon dioxide-air-ozone mixture (LC50 10.2 ppm) are not significantly different from the LC50 of 13.1 ppm in dry air. Based on these results ozone is classified as Acute Tox. Cat. 1 according to CLP criteria.

Executive summary:

In an acute toxicity study perfomed before the OECD guidance was published, 5 -10 Wistars rats (sex and age not specified) were exposed to ozone (5 - 18 ppm) for 3 h in two independent whole body inhalation experiments. In experiment one the animals were exposed to ozone made frome pure oxygen while in the second experiment ozone was genereated from carbon dioxide enriched air (2% CO2) to investigate possible differences in toxicitiy. No information on clinical signs nor pathology was given in the publication.

The LC50 cvalues were detemined using the method according to Reed and Muench (1938). The study shows that the LC 50 values for ozone in 100% oxygen (LC50 12.2 ppm) and carbon dioxide-air-ozone mixture (LC50 10.2 ppm) are not significantly different from the LC50 13.1 ppm in dry air as published earlier by this group. Additionally, LC50 values were determined under the same conditions using mice and guiniea pigs. Showing, that O3 made from oxygen was more toxic for mice than ozone in air while the 100% oxygen had no influence on the toxicity of ozone for rats and guinea pigs. Ozone in a 2% carbon dioxide-air mixture was more toxic for guinea pigs than ozone in air, but had no significant effect on the toxicity of ozone for mice and rats.

The study, for its time, is a well-documented report which meets basic scientific principles making it suitable for risk-assessment.

Based on these results ozone does warrant for classification as Acute Tox. Cat. 1 according to CLP criteria.

Endpoint:
acute toxicity: inhalation
Type of information:
experimental study
Adequacy of study:
weight of evidence
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
study well documented, meets generally accepted scientific principles, acceptable for assessment
Qualifier:
no guideline available
Principles of method if other than guideline:
The study was performed before guidelines were published
GLP compliance:
no
Test type:
other: Research study
Limit test:
no
Specific details on test material used for the study:
Ozone gas was generated from prue oxygen in situ by a water-cooled corona discharge-type ozonizer developed by Armour Research Foundation personnel.
Species:
rat
Strain:
Wistar
Sex:
not specified
Details on test animals or test system and environmental conditions:
Two to three-month-old Wistar rats. no further data.
Route of administration:
inhalation: gas
Type of inhalation exposure:
whole body
Vehicle:
air
Details on inhalation exposure:
A large glass aquarium 12.5 in x 30 in x 16 in was converted into an air-tight chamber with an aluminum inlet tube at one end and an aluminum outlet tube at the other. To equalize decomposition of ozone on the surface of the animals, the rats were exposed in groups of 5. The air flow containing
ozone was maintained above 10 l/min.
The water-cooled corona discharge-type ozonizer used was developed by Armour Research Foundation personnel in other research in the field of ozone technology.
Dry filtered air or oxygen is passed through a 1/16 in annular space and approximately 45 ppm of ozone is produced at an input of 4.5 L/min. of air at a current of 0.09 ma using a 6,000-v neon sign transformer. The 2 to 60 ppm ozone production was regulated by controlling the amount of air to the ozonizer, by changing the amount of current, and by controlling dilution.
Analytical verification of test atmosphere concentrations:
yes
Remarks:
A modified aluminum chloride-buffered potassium iodide method was used to determine the ozone concentration.
Concentrations:
88 two to three-month-old Wistar strain albino rats were exposed, usually in groups of five, to concentrations of ozone ranging from
14.3 to 39.4 ppm (w/w) for three hours. In this study, ppm is parts per million by weight.
No. of animals per sex per dose:
groups of 5 -14 rats were used for each ozone concentration.
Control animals:
no
Details on study design:
Each group of rats was exposed at a given concentration for three hours. Usually most of the deaths occurred during and 24 hours after the ozone exposure; the number of survivors was determined 24 hours after the exposure.

In "subacute experiments", the rats were exposed for various periods of time to concentrations which are now known to be non-lethal. This was done to determine the relationships between lenght of exposure, ozone concentration, and first symptoms of ozone poisoning. Lung injury was assessed by the Joffe's modification (1954) of Rahn's method based on the ratio of the weight of the lungs as removed from the animal and after drying. The presence of edema, or blood as a result of hemorrhage, causes a greater wet lung weight and thus a higher ratio. In this method each lung serves as its own control.
Statistics:
The LD50 was calculated by the method of Reed and Muench which involves plotting the percentage of cumulative deaths and cumulative survivals against the concentration. The intersection of the two lines represents the LD50 value. This method tends to equalize chance variations because it utilizes the large total number of animals consisting of small groups of animals each exposed at a different concentration.
Preliminary study:
no
Sex:
not specified
Dose descriptor:
LC50
Effect level:
21.8 ppm
Based on:
test mat.
Exp. duration:
3 h
Mortality:
See other findings.
Clinical signs:
other: Some general observations were made about acute ozone poisoning. The animals become very quiet and seem to doze 15 min after exposure to ozone began. During the second hour of expusure, if the concentration was high, breathing became rapid and labored; if
Body weight:
no data
Gross pathology:
Every autopsy of an animal killed by ozone showed extensive hemorrhage from the capillaries in the lungs and considerable edematous fluid in the respiratory system which may cause death by asphyxiation.

Albino rats have hemorrhagic lungs after 18 hours' exposure at 5 ppm; the data are presented in Table I (other findings)

In albino rats exposed to concentrations below 4 ppm for 20, 22, and 24 hours; no lung damage was found after exposure to ozone concentrations of 3.3 ppm or less. The data from this series of experiments are presented in Table II. (other findings).
Other findings:

TABLE I EFFECTS OF EXPOSING RATS* TO SUBACUTELY TOXIC
CONCENTRATIONS OF OZONE FOR 18 HOURS
Lung
____________________________________________________________
Ozone
ppm Ratio** Gross Appearance
____________________________________________________________
Control 4.53 ------
3.6 4.46 Normal
3.7 4.48 Normal
3.8 4.24 Normal
4.0 4.26 Normal
4.4 4.62 Four slightly hemorrhagic
5.0 4.88 All hemorrhagic
5.1 5.77 All hemorrhagic
5.8 4.85 All hemorrhagic
6.0 5.01 All hemorrhagic and edematous
7.4 5.42 All hemorrhagic and edematous
7.5 5.69 All hemorrhagic and edematous
7.7 All hemorrhagic and edematous
____________________________________________________________
*Five rats were exposed at each concentration.
** Wet/Dry weight.


TABLE II EFFECTS OF EXPOSING RATS TO SUBACUTELY TOXIC
CONCENTRATION OF OZONE
_____________________________________________________________________________________________
Ave.
Expo. Body Lungs
sure. Ozone, No. of Weight _____________________________________________________
br ppm Animals g Ratio* Gross Appearance
_____________________________________________________________________________________________
20 3.0 4 292 4.11 Normal
3.3 4 275 4.23 Normal
3.5 4 144 4.19 Normal
3.8 4 286 4.98 Three slightly hemorrhagic
4.4 4 152 4.24 One slightly hemorrhagic
22 2.8 4 828 4.48 Normal
3.2 5 357 4.1 Normal
3.5 5 172 4.50 Three slightly hemorrhagic
3.5 5 211 4.41 Normal
4.1 5 166 4.61 Five slightly hemorrhagic
24 2.9 5 4.26 Normal
_____________________________________________________________________________________________
*Wet/Dry weight.

Interpretation of results:
Category 1 based on GHS criteria
Conclusions:
Based on the results of this scientific publication meeting basic scientific principles, ozone is classified as Acute Tox. Cat. 1 according to CLP criteria as the LC50-value for inhalative acute toxicitiy for a 3 h exposure of rats was determined to be 21.4 ppm. Furhermore, this study investigated the relationship between length of exposure, ozone concentration and first symptoms of ozone poisoning (lung endema and hemorrhage) and provides information on the severe intoxication of the lung following ozone exposure.
Executive summary:

In an acute toxicity study perfomed before the OECD guidance was published, 5 -13 2 to 3 month old Wistars rats (sex not specified) were exposed to ozone (14 - 40 ppm) for 3 h in a whole body inhalation experiment. The animals become very quiet and seem to doze 15 min after exposure to ozone began. During the second hour of expusure, if the concentration was high, breathing became rapid and labored; if the animal was free to move, it staggered about gasping for air. Death usually occurred in those animals which oozed edematous fluid from the nostrils. Every autopsy of an animal killed by ozone showed extensive hemorrhage from the capillaries in the lungs and considerable edematous fluid in the respiratory system which may cause death by asphyxiation.

The LD50 for a three-hour exposure of ozone was found to be 21.8 ppm for rats. Besides rats, the LD50 values in mice (21 ppm), cats (34.5 ppm), rabbits (36 ppm) and guinea pigs (51.7 ppm) have been determined. In a second experiment reported by this publication, the animals were exposed for various periods of time to concentrations which are now known to be non-lethal. Lung injury was assessed by the ration of the wet and dry weight of the lungs. Edema occurred in the lungs of rats exposed to 6 ppm for 18 hours; no damage was caused by 24-hour exposure to concentrations of less than 3 ppm ozone.

For the following time (h) x concentrations (ppm) combinations;  3x15,  18x4 and 24x 3 the lungs looked normal.

The publication is a well-documented report which meets basic scientific principles and is therefore suitable for risk-assessment.

Based on these results ozone does warrant for classification as Acute Tox. Cat. 1 according to CLP criteria.

Endpoint:
acute toxicity: inhalation
Type of information:
experimental study
Adequacy of study:
weight of evidence
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
study well documented, meets generally accepted scientific principles, acceptable for assessment
Qualifier:
no guideline followed
Principles of method if other than guideline:
Study was published before guidance documents were in place.
GLP compliance:
no
Test type:
other: experimental research study
Limit test:
no
Specific details on test material used for the study:
Ozone was generated by means of a silent discharge ozonizer (Tack Air Conditioning Ltd., Model L03).
Species:
rat
Strain:
not specified
Sex:
male/female
Details on test animals or test system and environmental conditions:
No details on housing and test animals given
Route of administration:
inhalation: gas
Type of inhalation exposure:
whole body
Vehicle:
air
Details on inhalation exposure:
Ozone generated by means of a silent discharge ozonizer (Tack Air Conditioning Ltd., Model L03). Oxygen (99-5%) was passed from a cylinder through the ozonizer at a rate of 1 to 3 l./min. At this rate the ozone output was of the order of 27 mg./min. The ozonized oxygen was allowed to leak at controlled rates in the region of 20 ml./min into a stream of filtered air which had been dried by means of a silica gel column to a humidity of about 10%. Analytical control of the ozone dilution demonstrated that provided the flow rates were maintained constant, the ozone concentration of the atmosphere produced remained constant for the duration of each experiment. The ozone dilution passed through a 5 L mixing vessel and thence to the exposure chamber. The exposure chamber was constructed, as described in our previous paper, from a 20 L glass desiccator. The inlet and outlet tubes were sealed into the chamber by means of polythene bungs and all flexible connexions in the system were made with polyvinyl chloride tubing.
Analytical verification of test atmosphere concentrations:
yes
Remarks:
Trapping in potassium iodide solution buffered and titration with 001 N sodium thiosulphate (Wilska, 1951)
Concentrations:
Concentration range from 3-36 ppm, see mortality
No. of animals per sex per dose:
3-4 rats per group, see mortality
Control animals:
not specified
Details on study design:
Groups of rats were exposed for 4 hour to a serie of experiments with increasing ozone concentrations.
Statistics:
No data
Preliminary study:
Not applicable
Sex:
not specified
Dose descriptor:
LC50
Effect level:
>= 10 - <= 12 ppm
Based on:
test mat.
Exp. duration:
2 h
Mortality:
Table 1: Effects on rats of single four-hour exposure to ozone
Concentration Rats
(ppm) No Sex Effects
3-4 3 F Laboured breathing towards end of exposure complete recovery
3-6 4 M As above
8 4 M Dyspnoea after 2 hours, becoming severe ; 1 death within 24 hours of exposure
9 3 F Dyspnoea after 2 hours, becoming severe ; no deaths
14 3 F Dyspnoea and lethargy after 30 minutes ; all died during exposure
17 4 M Acute dyspnoea after 2 hours ; 3 died during exposure, 1 later
30-36 4 M Acute dyspnoea after 2 hours ; all died during exposure

The results do not permit a statistical calculation of a lethal concentration (LC50), but it is most probably in the region of 10 to 12 p.p.m. Concentrations below this may produce signs of transient respiratory distress during the experiment, above this figure the animals become comatose and develop acute dyspnoea which terminates in death during or within 24 hours after the exposure.
Post-mortem examination revealed the lungs distended with oedematous fluid. Histological examination of pulmonary tissue showed extensive oedema and emphysema, but no haemorrhage or engorgement of alveolar capillaries.
Clinical signs:
other: see mortality
Body weight:
no data
Gross pathology:
see mortality
Interpretation of results:
Category 1 based on GHS criteria
Conclusions:
Based on the results of this scientific publication meeting basic scientific principles, ozone is classified as Acute Tox. Cat. 1 according to CLP criteria as the LC50-value for inhalative acute toxicitiy was determined to be in the range of 10 to 12 ppm .
Executive summary:

In an acute toxicity study perfomed before the OECD guidance was published, 3 - 4 rats or mice of both sexes were exposed to ozone (3 - 36 ppm) for 4h in a whole body inhalation experiment. At the lower concentrations the animals showed laboured breathing or dyspnoea after nor more than 2 h of exposition. From a concentration of 14 ppm, all animals died during exposure or the 24 h observation period.

Post-mortem examination revealed the lungs distended with oedematous fluid. Therefore the cause of death was assumed to be acute pulmonary oedema. No LC50 -value was determined due to the low number of animals, but the concentration of ozone which causes 50% mortality to rats and mice exposed for one period of four hours has been found to be in the region of 10 to 12 ppm.

The publication is a well-documented report which meets basic scientific principles and is therefore suitable for risk-assessment.

Based on these results ozone does warrant for classification as Acute Tox. Cat. 1 according to CLP criteria.

Endpoint conclusion
Endpoint conclusion:
adverse effect observed
Dose descriptor:
LC50
Value:
7.06 mg/m³ air

Acute toxicity: via dermal route

Endpoint conclusion
Endpoint conclusion:
no study available

Additional information

Oral and dermal route

No studies, which were performed according to internationally accepted guidelines (OECD Test Guideline 401, 402, 420, 423 or 425) for evaluation of the acute toxicity via the oral or dermal exposition route of pure ozone, were found in the public domain or in the archives of the applicants. This is likely due to the physical nature of ozone. Ozone is a gas, and hence, standard dermal toxicity studies with the pure substance for classification and labelling purposes are technically not feasible. As neither dermal nor oral exposure is a primary route of exposure, dosing of the pure substance or high concentrations of the substance to laboratory animals via oral or dermal route is omitted in accordance with Annex VIII column 2. This is also in line with the 3R concept.

 

Inhalation route

No studies, which were performed according to internationally accepted guidelines (OECD Test Guideline 403 or 436) for evaluation of the acute toxicity of pure ozone, were found in the public domain or in the archives of the applicants. However, research publications on ozone that are available in the public domain give LC50 (e.g. values from 3 - to 4-hour exposure followed by a 24-hour observation period) in the range from 2.4 to 21.8 ppm in rats, which seem to be the most sensitive species. While the values were obtained from the studies not performed under GLP, all are substantially lower than 100 ppm, which is the upper limit for classification of a gas for acute toxicity in category 1. Corrosion was observed in the airways of living animals when exposed by inhalation to low concentrations of ozone in air. It is therefore concluded that pure ozone is corrosive to tissues of living animals and that it is not ethical to perform additional in vivo experiments for classification and labelling of the pure substance. In a worst-case approach 3.6 ppm was determined as LC50 value (Svirbely and Saltzman, 1957). The original value reported by Svirbely and Saltzman was 2.4 ppm, but it needed to be adjusted by a factor of 1.5 according to an addendum to the publication. This value leads to a classification (REGULATION (EC) No 1272/2008 and amendments) as "Acute toxicity by inhalation - Category 1" according to CLP criteria.

Numerous studies on inhaled ozone have been performed over the last 80 years. A selection of the literature available has been identified in the public domain and reviewed for the acute inhalatory toxic properties of ozone. While the summary below does not provide extensive description of all the existing literature on ozone, but it does cover 12 selected relevant studies, which altogether provide sufficient evidence to draw a firm conclusion on the acute toxic properties of ozone in animals. The majority of these studies were performed using rats and monkeys (considered to be the most relevant species in this respect), but there is also data on rabbits, guinea pigs and mice. The exposure comprised a single acute exposure with a duration up to 24 hours, and most of the studies were conducted using minor ozone concentrations (< = 1 ppm), thereby showing mostly reversible inflammatory effects. As these publications are very specific on certain symptoms or protein changes, they are only used as supporting information. The risk assessment was made using published studies with higher tested/reported ozone concentrations in a weight-of-evidence approach.

Summary

Pulmonary response following inhalation of relative high ozone concentrations

The respiratory tract is the principal target for toxicity by inhalation of ozone in animals. At relatively low concentrations (ppm level), ozone has been shown to be already lethal in animals. Death was consequent of severe lung injury that had resulted in alveolar oedema, congestion, and haemorrhage (Mittler et al., 1956). LC50 values for 3- to 4-h ozone exposure followed by a 24-h observation period in rats have been reported to be in the range from 2.4 to 21.8 ppm (Mittler et al. 1956, Mittler et al. 1957; Diggle and Cage 1955, Svirbely and Saltzman 1957). Furthermore, LC50 values ranging from 12.6 to 51.7 ppm were determined by Mittler and colleagues (1956 & 1957) in guinea pigs, mice, rabbits and cats using whole body or funnel inhalation (using cats and rabbits). Diggle and Cage (1955) exposed mice and rats to ozone concentrations ranging from 3.4 to 36 ppm for 4 h and reported the first deaths at 8 ppm. No LC50 values were calculated in this study due to the low number of animals used, but LC50s were estimated to be in the range from 10 to 12 ppm. This wide range of LC50 values is probably due to the different methods to produce ozone and tools to analyse/measure ozone concentrations. Svirbely and Saltzman (1957) evaluated the difference between several ozonizers as well as the impact of ozone source (e.g. generated from pure O2, normal air or oxygen-nitrogen-mixtures) on lethality and reported LC50 values ranging from 2.4 to 8.2 ppm. According to an addendum to this publication, these LC50 values need to be adjusted by a factor of 1.5.

 

Pulmonary response following inhalation of relative low ozone concentrations

It has been shown in rats exposed to ozone concentrations ranging between 0.12 and 1.5 ppm for 6 h to have transient increases in polymorphonuclear leukocytes (PMNL) and pulmonary alveolar macrophages (AM). At 18 hours after exposure to 0.12 ppm ozone, increased PMNL were recovered from nasal lavage fluid (NLF) of rats, yet there was no change in the number of PMNL recovered from bronchoalveolar lavage fluid (BALF) at any time after exposure. Rats exposed to 0.8 ppm ozone had more PMNL in NLF immediately after exposure but no concomitant increase in PMNL in BALF at that time. However, as the number of PMNL in BALF increased (maximum at 42 h), the number of PMNL in NLF decreased (minimum at 42 h). Finally, rats exposed to 1.5 ppm ozone had no significant increase in PMNL in NLF at any time after exposure but had greatly increased numbers of PMNL in BALF at 3, 18, and 42 h after exposure. The number of PMNL recovered by nasal and bronchoalveolar lavage accurately reflected the concentration dependency of the tissue PMNL response at different sites within the nasal cavity and lung that were injured by acute ozone exposure. Exposure to 0.12 ppm ozone in rats had no measurable effect on the total number, labelling index (LI), mitotic index (MI), or morphology of rat pulmonary AM. However, at 42 and 66 hours after exposure to 0.8 and 1.5 ppm ozone, the number of pulmonary AM was approximately twice of that from the controls. There was also an increase in MI of the pulmonary AM at 42 and 66 h after exposure to 1.5 ppm ozone and at 42 h after exposure to 0.8 ppm ozone. The increase in the number of pulmonary AM in mitosis was preceded by an increase in the LI of the pulmonary AM. The LI of the pulmonary AM at 18 and 42 h after exposure was greater than that of the controls but returned to near normal levels by 66 h after exposure. There was a transient decrease in the mean nuclear/cytoplasmic ratio of the pulmonary AM from rats assessed 18 and 42 h after exposure to 1.5 ppm ozone due to an increase in the mean pulmonary AM cytoplasmic area. Comparison of the pulmonary AM population doubling time (Dt) and cell cycle time (Ct) suggests that cell proliferation plays a significant role in the observed increase in pulmonary AM following exposure to 0.8 and 1.5 ppm ozone (Hotchkiss et al., 1989a).

Studies have been published to investigate the impact of concentration (C) and time (T) in controlling the response of animals to acute and continuous exposure to ozone. Costa et al. (1989) exposed Fischer rats for 2, 4, and 8 h to 0.1, 0.2, 0.4, 0.8 and 1.2 ppm ozone. Lung function parameters were measured immediately after exposure, and BALF was collected immediately as well as 24 h after exposure. Only minor alterations in dynamic lung function were observed at concentrations below 0.8 ppm. The earliest effects were observed at 1.2 ppm ozone for 2 h and 0.8 ppm ozone for 4 h. Decrements in lung function increased with both C and T, levelling off when the product of C and T became >6 ppm*h. Protein leakage into the air space, as measured by the protein level in BALF, increased rapidly when C x T is >4 ppm*h.

Rombout et al. (1989) exposed rats for 1, 2, 4 and 8 h to 0.75, 0.75, 1.25 and 2.0 ppm ozone during daytime conditions and for 4, 8 and 12 h to 0.125, 0.25 to 0.375 ppm ozone during nighttime (active) conditions. Measurement of BALF protein levels at various times post-exposure ranging from 1 to 72 h revealed that the responses varied with the ozone concentration, exposure time, and/or time post-exposure; exposure time per day had a significant effect on BALF protein levels. Animals exposed during nighttime were more sensitive to ozone. At a given ozone concentration, effects elicited during nighttime exposures were generally twice as large as effects elicited during daytime exposures, showing that exercise enhanced the response of the animals to ozone under these temporal conditions. The minimal C x T product required to increase the BALF protein in the lung lumen was 0.125 ppm ozone for 4 h. Multivariate analysis revealed that the influence of time increased with increasing ozone concentration. Even at the lowest tested concentration of 0.125 ppm ozone, time remained an important factor in altered BALF protein levels mediated by ozone.

Exposure of rats to ozone at a concentration of 1.8 ppm for 2 and 4 hours significantly induced pulmonary inflammation, evidenced by increased cell numbers, LDH activity and albumin levels in BALF, with full recovery thereof at 8 days post exposure (Bassett et al., 1988). Bouthillier et al. (1998) reported that exposure of rats to 0.8 ppm ozone for 4 h resulted in a centriacinar injury, with some edema and fibrin deposition in the alveolar duct lumen, as well as a limited intra-alveolar and interstitial infiltration by neutrophils. Histological observations showed that the injury induced by ozone in the central acinus did not translate into significant shifts of the morphometric estimates. Exposure of the animals to ozone for 4 h resulted in higher recoveries of protein, fibronectin, and PMNL in lavage fluids at 20 h post-exposure. The phagocytic activity of macrophages was depressed after exposure to ozone.

 

Non-pulmonary effects of ozone inhalation

Ozone exposure causes non-pulmonary effects involving the central nervous system (CNS) and the cardiovascular system. It is most likely that molecules derived from reactions of ozone with lung tissue mediate these non-pulmonary effects. Inflammatory factors as well as free radicals are hypothesised to be the molecules that flow through the bloodstream to other organs producing a secondary response upon ozone inhalation.

CNS response to single ozone exposure

Rivas-Arancibia et al. (2000) reported increased lipid peroxidation in the frontal cortex, hippocampus, striatum and cerebellum of young rats after ozone exposure of 4 h to 1 ppm. In older rats, lipid peroxidation was observed in the hippocampus and striatum. Three hours after exposure to 1 ppm ozone, rats had increased striatal lipid peroxidation levels that lasted up to 3 days after exposure. Ozone also increased basal dopamine, glutamate and nitric oxide concentrations and decreased those of 5-HT (serotonin). Concentrations of GABA were initially decreased 3 h after ozone but were then increased 3 and 5 days afterwards (Rivas-Arancibia et al., 2003). It has been shown that the brain regions associated with both cognitive (memory) and motor function are vulnerable to ozone exposure. Rats exposed for 4 h to 0.2, 0.5, or 1 ppm ozone showed long-term memory deterioration and decreased motor activity, which was reversed 24 h later. In young (47 days) and aged (900 days) rats, exposure for 4 h to 0.7 ppm ozoneled to short-term and long-term memory alterations. Mature rats (540 days) showed a tendency of impaired memory. Three hours after exposure to 1 ppm ozone, rats had decreased exploratory and increased freezing behaviors (Rivas-Arancibia et al., 2000, 2003). Furthermore, increased lipofuscin, neuronal cytoplasm and dendrite vacuolation and dilation of rough endoplasmic reticulum cisterns and dark cells were observed in striatal medium spiny neurons 3 h after a 4 h exposure to 1 ppm ozone (Rivas-Arancibia et al., 2003). The authors hypothesise that the neurochemical and ultrastructural changes observed after 4 h exposure to 1 ppm ozone reflect a neurodegenerative process caused by oxidative stress, disrupting the redox homeostasis in vivo, resulting in neural cell damage with rapid (by 3 h) concomitant alterations in dopamine, glutamate, GABA and nitric oxide levels, which are still evident 5 days after exposure.

 

Cardiovascular response to single ozone exposure

Farraj et al. (2012) performed electrocardiography (ECG) measurements on rats exposed for 3 h to 0.2 or 0.8 ppm ozone. High ozone concentration (0.8 ppm) caused bradycardia, PR prolongation, ST depression and substantial increases in atrial premature beats, sinoatrial block and atrioventricular block, accompanied by concurrent increases in several heart rate variability parameters that were suggestive of increased parasympathetic tone. Low (0.2 ppm) ozone exposure did not induce any overt changes in autonomic tone, heart rhythm or ECG.

Justification for classification or non-classification

Based on the LC50 values reported for inhalation, ozone qualifies for classification as Acute Tox. Cat. 1 according to CLP criteria.