Chlorine

Endpoint summary

Description of key information

Standard guideline studies to determine the irritation/corrosivity potential of chlorine can not be done because chlorine is a gas at ambient temperature. Furthermore primary references which describe the effect of chlorine on the human skin have not been found. Based on a recently approved international hazard assessment (SIDS Initial Assessment Profile, SIAP), chlorine is corrosive to the skin (OECD, 2003). Furthermore secondary references and also internal documentation of chlorine producers shows that gaseous chlorine (and liquid chlorine) is irritating/corrosive to the skin. Contact with liquid chlorine will cause skin burn and frostbite. However, it is not possible to derive a threshold concentration for skin irritation or corrosion for gaseous chlorine because the respective data are not available and effects on the skin will strongly depend on the duration of exposure. It should be realised that there is no need to determine these threshold concentrations because the target organ of chlorine is the respiratory tract. 
Acute inhalation studies with animals show that gaseous chlorine is irritating to the eyes. Based on human data, irritating effects on the eyes were reported at a chlorine concentration range of 0.2 to 4 ppm (0.6 to 12 mg/m3). According to secondary literature exposure to chlorine can result in injury of the cornea and enduring impaired vision and blindness(corrosive effect). 
Many animal and human studies have been reported about the effect of chlorine on the respiratory tract. A well documented study with human volunteers, done according to Good Clinical Practice, showed no inflammatory response or irritant effects on the nasal epithelium at 0.5 ppm (1.5 mg/m3) and for this reason 0.5 ppm (1.5 mg/m3) is considered a NOAEC (equivalent to OEL) for respiratory irritation. 

Key value for chemical safety assessment

Skin irritation / corrosion

Endpoint conclusion

Endpoint conclusion:

adverse effect observed (irritating)

Eye irritation

Endpoint conclusion

Endpoint conclusion:

adverse effect observed (irritating)

Respiratory irritation

Endpoint conclusion

Endpoint conclusion:

adverse effect observed (irritating)

Additional information

Skin:

Standard guideline studies to determine the irritation/corrosivity potential of chlorine can not be done because chlorine is a gas at ambient temperature. Acute dermal toxicity studies with animals are also not possible/available. Although many animal experiments studied the short term effects of chlorine exposure, skin effects have not been described in most cases because the main target organ is the respiratory tract. Solutions of chlorine in water contain hypochlorous acid and/or hypochlorite. Thus, considering the natural water content of the skin surface data with sodium hypochlorite can be used for read across.

Sodium hypochlorite studies:

A solution of sodium hypochlorite (4.74% available chlorine) in a mixture with other ingredients used as control bleach in a patch study was applied (0.5 ml) under a semiocclusive patch on the dorsal skin of the rabbits for a 24-hour period. The skin was examined for erythema and edema directly after patch removal and 48 hours later. The evaluation of the lesions was carried out according to the FHSA Regulation (1973). A primary irritation index (PII) of 5 or higher indicates a primary irritation response in accordance with FHSA Regulations. The compound was considered to be non irritant to the rabbit skin based on the PII reported as < 5. (Osterberg et al., 1977).

Sodium hypochlorite 5.25 % solution (pH 10.7, 0.5 ml) was applied on rabbit and guinea pig abraded and non-abraded skin in a 4-hour patch test as outlined in the revised FHSA procedure that had been proposed by FDA (Edwards, 1972). The skin was examined at 4, 24 and 48 hours after patch removal. Results showed the compound to be slightly irritant to both rabbits (PII = 1.2) and guinea pigs (PII = 0.8) (Nixon et al., 1975).

0.5 ml of sodium hypochlorite 12.5 % available chlorine was applied to the intact and abraded rabbit skin for 24 h. The skin was examined for effects up to 72 hours. The scoring of the irritation response was carried out according to the Draize classification. Initial solution (12.5 %) and dilution of ½ (6.25 %) were considered as severe irritant (with PII = 5.6 for both concentrations); a dilution of ¼ (3.12 %) was considered to be moderately irritant (PII = 4.0) and a dilution of 1/8 (1.56 %) as slightly irritant (PII = 1.9) (Duprat et al., 1974). It is pointed out that Duprat used a much longer exposure time (24 hours) compared to the standard classification patch test, which recommends patching for 4 hours. This can also explain the higher scores in comparison to the Nixon study, which did use the standard 4 hours exposure time.

0.5 ml of sodium hypochlorite 12.7 % active chlorine was applied to intact and abraded rabbit skin for 24 h. The skin was examined for effects up to 72 h. Scores from intact skin and abraded skin were added and a mean was calculated. Hypochlorite at 12.7 % was considered to be moderately irritant (PII = 4.04) (Colgate-Palmolive, unpublished data-1985).

In a dermal irritancy/corrosion test on 20 compounds in aqueous solution, sodium hypochlorite solution 8-12% available chlorine applied to the dorsal skin of the rabbit, was tested at different concentrations (2, 20, 35, 50% w/v, i.e. 0.24, 2.4, 4.2 and 6 % available chlorine): The study showed slight irritation effects at the lowest concentration, moderate irritation at the other concentrations and corrosive effects at the highest concentration tested probably according to the Draize scale) (Loden et al., 1985). Based on poor reporting, the study protocol followed could not be verified, although the Draize scale appears to have been used.

The primary skin irritation score in the rabbit was found to be 5.08 using 0.5 ml of undiluted liquid. This score was an average of mean scores on intact and abraded skin. Contact time is not specified. The compound was considered as “corrosive” (Industrial Bio-Test Laboratories Inc., 1970). However, this study is considered of very limited value to assess the irritation properties of hypochlorite, as neither the test substance concentration in this study nor the exact protocol used are described.

The studies from Loden and Biotest had been given a validity 4 because only limited details of the design of the studies were reported.

Eye:

Standard guideline studies to determine the eye irritation/corrosivity potential of chlorine can not be done because chlorine is a gas at ambient temperature. However, acute inhalation toxicity studies with chlorine gas clearly showed that chlorine is irritating to the eyes. Solutions of chlorine in water contain hypochlorous acid and/or hypochlorite. These solutions can be irritating or corrosive to the eyes.

Effects in humans

Matt reported already in 1889 that 3.5-4 ppm chlorine produced an immediate burning sensation of the eyes while severe burning in eyes was observed at 2.5 ppm (7.5 mg/m3) after 5-16 minutes. In many cases the reported effects included effects on the eyes, like “burning of conjuctiva” or “irritation of the eyes”. These irritating effects on the eyes were reported at a chlorine concentration range of 0.2 to 4 ppm (0.6 to 12 mg/m3).

Sodium hypochlorite studies:

An eye irritation test was carried out in rabbits by Momma et al., using the Draize method (1986). The ocular irritation score was calculated from examinations up to 21 days post exposure of 5% solution. The eyes were either left unrinsed or rinsed with water after the application of sodium hypochlorite. Scoring at 24, 48 and 72 hours revealed slight – moderate eye irritation potential in both groups. In the group where rinsing was applied, eyes had returned to normal by day 14. More persistent and more pronounced injury to the cornea and the conjunctiva was observed in the group without washing. Day 21 was the end of observation time in this study at which some effects were still noted. The study indicates that rinsing of the eyes either 4 or 30 seconds after instillation significantly reduced the degree of ocular irritation.

Osterberg also performed an eye irritation test using an unofficial ocular irritation classification method (with similarities to the FHSA and Draize methodology). A dose of 0.1ml of an otherwise unspecified mixture containing hypochlorite at a conc. of 4.74 % available chlorine, was placed into the rabbit eyes and the score for alteration was followed up to 7 days post exposure. The compound was found to be severely irritant to the rabbit eye according to the specific grading scale used. Detailed scores observed however are not reported in the study. Recovery was not complete at day 7 (Osterberg et al., 1977). It has to be noted that mixture was specified as laundry bleach by the author and contained other ingredients, which would have contributed to irritancy.

The standard Draize method was applied for the evaluation of eye irritation in the rabbit. In addition to the Draize mean average scores, a non-standard microscopic evaluation was used in the grading. Commercial sodium hypochlorite (12.5% available chlorine) and a ½dilution (6.87%) were considered severe irritants with a Draize MAS score of 60 and 49, respectively. Complete recovery was observed in week 10 for the 12.5% solution and in week 4 for the 6.25% solution. ¼Dilution (3.6%) of the solution was found to be moderately irritant with a Draize MAS of 11 and complete recovery at day 15 while a 1/8 dilution (1.85%) was found to be slightly irritant (Draize MAS 1) with complete recovery at day 4. Rinsing with 20 ml physiological saline was done at 10 sec, 1 min and 5 min after application of the test substance. Washing was also done using higher volumes of physiological saline (300 ml or 600 ml). The immediate rinse (at 10 sec with 20 ml physiological saline) was the most effective in reducing the irritant effect significantly (Duprat et al., 1974).

According to the Draize method, a quantity of 0.1 ml of sodium hypochlorite 12.7 % active chlorine was applied in rabbit eyes. At day 7, Hypochlorite 12.7% was considered to be severely irritant (MAS of 64.75). Most effects had not cleared by the end of the observation period, day 14 (Colgate-Palmolive, unpublished data-1985).

Undiluted solutions at 5.25% or 8% were found to be low to moderate eye irritants in rabbits. Specifically, a 8% concentrated solution resulted in moderate irritant effects in a Draize test, with recovery being completed within 7 days. Low irritant effects were observed when the 8% solution was tested in an LVET study (Low Volume Eye Test, applied dose = 0.01 ml), with recovery being completed in 3 days. 5.25% sodium hypochlorite was tested according to the LVET protocol only and led to low irritant effects. Exact scores are not reported in the study. The authors emphasize that the LVET provides a better correlationwith human eye irritancy experience than the Draize test. Dilutions of 1:10 of the described bleaches (0.55% and 0.8 % sodium hypochlorite) also had only a low eye irritant potential in an LVET study. It was observed that a rinse with water following the eye contact reduced the degree of irritation. (Racioppi et al., 1994).

The instillation of 0.1 ml of undiluted sodium hypochlorite (unspecified concentration) into the rabbit eye gives a MAS score for irritation of 61.3 according to the Draize methodology. The compound was considered as “severe irritant“ (Industrial Bio-Test Laboratories Inc., 1970). However, this study is considered of very limited value to assess the irritation properties of hypochlorite, as the test substance concentration in this study was not described.

Griffith et al. (1980), studied hypochlorite amongst other chemicals using the Draize scale for grading of the effects. Different dose volumes were applied (0.01, 0.03 and 0.1 ml) with the purpose of comparing the observed effects to data from human experience (literature, occupational incidents and consumer accidental exposures). 0.01 ml was considered the volume leading to effects most consistent with human eye reactions. Therefore, application of 0.01 ml (as used in the LVET protocol) was proposed to be a much more realistic test of eye hazard than the Draize test. A 5 % solution of sodium hypochlorite (pH 11.1-11.6) produced only mild transient effects in a Draize test (MAS of 11 after 1 day) when rinsed out with water in the first 30 seconds. When the LVET protocol was used (0.01 ml), similar mild and transient effects were noted. Effects had cleared completely by day 7. If 0.1 ml was applied and the eyes were not rinsed, moderate irritation effects involving the cornea and the conjunctiva occurred (MAS of 31 after 1 day). Recovery was not entirely complete at the end of the observation period (day 21). At the intermediate dosing volume of 0.03 ml applied, also moderate effects were observed (MAS of 28 at day 1) which had cleared completely by day 18 (Griffith et al., 1980).

A 15% solution of sodium hypochlorite caused severe pain and damage and there were indications that healing was not complete 2-3 weeks after exposure (Grant, 1962).

Carter and Griffith (1965) summarized effects that Buehler and Newsman (1964) observed when comparing the eye irritation potential of a sodium hypochlorite aqueous solution 5.5 % in rabbits and monkeys. The eyes were not rinsed. The Draize methodology appears to have been used. No detailed scores apart from the recovery times are reported. The irritant response was much greater in rabbits (recovery between day 7 and day 35) than in monkeys (recovery at day 2). Data were also compared to recovery dates from human exposure in factory eye accidents. The author noted that the irritation response observed in the monkey seemed to be a better indicator of eye irritation following accidental exposure in workers than the response observed in the rabbit.

Pashley et al. (1985) reported rabbit eye experiments following a Draize-related methodology. Detailed scores are not reported. Sodium hypochlorite, applied to rabbit eyes at concentration of 5.25 % NaClO produced moderate to severe conjunctival palpebral edema and hyperemia within 30 minutes of exposure with the maximum severity observed at 2 h. The eyes exposed to 5.25 % revealed corneal pitting but no ulceration. Some conjunctival edema was observed, which had not fully cleared up at day 7 (and of observation period). With a 0.52 % NaClO solution, only moderate effects were observed and the reaction was gone within 24 hours. The author pointed out that “the rabbit eye model tends to exaggerate the toxicity of agents since rabbits blink their eyes at a much lower frequency than humans” (Pashley, 1985).

Other Data (in-vitro)

Solutions of sodium hypochlorite were also tested “in vitro“ using different cell systems. The principal aim of the studies was to compare the results obtained “in vitro“ with the “in vivo“ Draize score methods. The results obtained “in vitro“ basically confirmed the irritation properties on the sodium hypochlorite solution obtained with the standard methodology (Chan, 1985; Borenfreund and Borrero, 1984, Borenfreund and Shopsis 1985; Shopsis and Sate, 1984). However, the use of in-vitro methods for eye irritancy assessment of hypochlorite containing products at this point in time is questionable, as current in-vitro methods are known not to realistically predict the irritant properties of oxidizing substances like hypochlorite.

Respiratory tract

Studies in animals

Mice

To evaluate the degree of sensory irritation of chlorine in mice, the measurement of the percentage decrease in respiratory rate was utilised. Mice were exposed for 10 minutes to concentrations from 0.7 to 38.4 ppm (2.1 to 115.2 mg/m3). At 38.4 ppm (115.2 mg/m3) the maximum response (80-85% decrease) was observed. Recovery from exposure was delayed at levels higher than 10 ppm (30 mg/m3). The RD50 for male mice was calculated to be 9.30 ppm (27.9 mg/m3) for 10 minutes exposure, while at 19 ppm (57 mg/m3) a 50% decrease in respiratory rate was reached within 2 minutes (Barrow et al., 1977; Chang and Barrow, 1984).

Gagnaire et al (1994) observed a 10-minute RD50 (Concentration that caused a 50% decrease in respiratory rate) for chlorine of 8 ppm (24 mg/m3) and a 60- minute RD50 of 3.5 ppm (10.5 mg/m3) in mouse. Recovery was rapid and complete after exposures up to 4.6 ppm (13.8 mg/m3) for 120 minutes.

In mice no sensory irritation response appeared at 0.7 ppm (2.1 mg/m3) (Barrow et al., 1977).

Recently the acute respiratory responses of the mouse to chlorine were studied by Morris et al. (2005). Respiratory physiological responses were measured in female C57Bl/6J mice exposed to 0.8 to 4.0 ppm (2.4 to 12 mg/m3) chlorine gas. Chlorine was a potent sensory irritant with an RD50 of 2.3 ppm (6.9 mg/m3). The gas induced airway obstruction as indicated by a concentration-dependent increase in specific airways resistance (sRaw) during the 15- min exposure. At 0.8 ppm (2.4 mg/m3), chlorine produced only mild sensory irritation (<20% change in breathing frequency) and a 65% increase in sRaw.

Rats

For male rats 10 minutes RD50 values of 10.9 ppm (32.7 mg/m3) (Chang and Barrow, 1984) and 25 ppm (75 mg/m3) (Barrow and Steinhagen, 1982) were reported. Delayed recovery was also observed at exposures above 10 ppm (30 mg/m3).

Exposure of rats to 0.4 ppm (1.2 mg/m3) (4 hours) was the irritation threshold and raised the neutrophil levels in washings from lungs, bronchi and nasal cavity, whereas the levels of polynucleoli were lower (Ponomereva, 1980).

Cats, rabbits and guinea pigs

Cats exposed to 1 ppm (3 mg/m3) chlorine for 7.5 hours showed only sneezing and some salivation. Rabbits and guinea pigs showed to be less sensitive with first effects appearing at 18 ppm (54 mg/m3) during a 5-h exposure (Lehmann, 1887).

Effects in humans

Matt (1889) noted that 3.5-4 ppm (10.5-12 mg/m3) chlorine produced an immediate burning sensation of eyes and nasal congestion while severe burning in eyes, itching in mouth and throat, nasal congestion, heavy coughing and breathing pains were observed at 2.5 ppm after 5-16 minutes.

Joosting and Verberk (1975) reported on an experiment with human volunteers exposed to chlorine for two hours to 0.5, l, 2 and 4 ppm (1.5, 3, 6 and 12 mg/m3). No effects on respiration rate and ventilatory capacity where observed. Subjective phenomena (smell, taste, cough, irritation of eyes, nose, throat) showed a dose-related response. Irritation was clearly detected at 2 ppm (6 mg/m3), while for eye irritation and cough at 1 ppm (3 mg/m3) some subjects turned out more responsive than the others did. The exposure to 0.5 ppm (1.5 mg/m3) revealed only borderline effects. It was concluded that 2 ppm (6 mg/m3) for 2 hours was reasonably bearable for healthy subjects.

A chlorine study performed on 30 students disclosed some sensation of odour, throat irritation and urge to cough in subjects exposed for 4 hours at 0.5 and 1.0 ppm (1.5 and 3 mg/m3). Exposure at 2 ppm (6 mg/m3) was reported to be much more irritating than that at 0.5 and 1.0 ppm (1.5 and 6 mg/m3) for the same period. Neither 8 hour exposures to 0.5 ppm (1.5 mg/m3) nor 15 min exposures to 2 ppm (6 mg/m3) produced significant effects (pulmonary function) other than recognition (Anglen et al., 1980).

Additional studies on respiratory irritation of chlorine were reported by Beck (1959) and by Rupp and Henschler (1967).

Further evidence that 0.5 ppm (1.5 mg/m3) is a NOAEL in humans is provided by a study of Emmen and Hoogendijk (1997), which was published by Schins et al. (2000). The study was well documented and was done according to Good Clinical Practice. The objectives of this study were:

1) to determine if chlorine exposure at low levels induces nasal effects in humans as it does in rodents; and

2) to establish a possible occurrence of respiratory effects in human volunteers exposed to chlorine vapour at concentrations of 0, 0.1, 0.3 and 0.5 ppm (0, 0.3, 0.9 and 1.5 mg/m3).

The human male volunteers were exposed for 6 hour per day on 3 consecutive days to each of the 4 exposure conditions. The total number of volunteers was 8 (data analysis was limited to 7 subjects since one volunteer decided to stop participating for reasons not related to the study) and their age was 20-50 years. Furthermore they were considered healthy, as determined by medical and laboratory examination, and they had a normal lung function. Nasal lavage measurements were conducted during the study and the results did not support an inflammatory response or irritant effects on the nasal epithelium. Furthermore no significant effect on lung function parameters was found. The study did not support an inflammatory effect in the nose nor shows changes in the respiratory function at repeated exposure up to 0.5 ppm (1.5 mg/m3).

Based on respriatory irritating and tissue damaging potential in humans (inhalation studies) a classification and labeling STOT SE cat. 3 as given in Annex VI in CLP is justified.

Effects on skin irritation/corrosion: irritating

Effects on eye irritation: irritating

Effects on respiratory irritation: irritating

Justification for classification or non-classification

Currently solutions of sodium hypochlorite with 5-10 % active chlorine are classified as irritating (R36, Irritating to the eyes; R38, Irritating to skin) in the European Union, while solutions of sodium hypochlorite with 10 % active chlorine or more are classified as corrosive in the EU.

Standard guideline studies to determine the irritation/corrosivity potential of chlorine can not be done because chlorine is a gas at ambient temperature. Furthermore primary references which describe the effect of chlorine on the human skin have not been found. Based on a recently approved international hazard assessment (SIDS Initial Assessment Profile, SIAP), chlorine is corrosive to the skin (OECD, 2003). Furthermore secondary references and also internal documentation of chlorine producers shows that gaseous chlorine (and liquid chlorine) is irritating/corrosive to the skin. Contact with liquid chlorine will cause skin burn and frostbite. However, it is not possible to derive a threshold concentration for skin irritation or corrosion for gaseous chlorine because the data are not available and because the effects on the skin will also depend on the duration of exposure. It should be realised that there is no need to determine these threshold concentrations because the target organ of chlorine is the respiratory tract.

Acute inhalation studies with animals show that gaseous chlorine is irritating to the eyes. Based on human data, irritating effects on the eyes were reported at a chlorine concentration range of 0.2 to 4 ppm (0.6 to 12 mg/m3). According to secondary literature exposure to chlorine can result in injury of the cornea and enduring impaired vision and blindness(corrosive effect).

Many animal and human studies have been reported about the effect of chlorine on the respiratory tract. A well documented study with human volunteers, done according to Good Clinical Practice, showed no inflammatory response or irritant effects on the nasal epithelium at 0.5 ppm (1.5 mg/m3) and for this reason 0.5 ppm (1.5 mg/m3) is considered a NOAEC (equivalent to OEL) for respiratory irritation.

Currently chlorine is classifed irritating to eyes, respiratory system and skin. (Xi, R36/37/38) according to 67/548/EEC and eye irrit. cat 2, skin irrit cat 2 and STOT SE cat. 3 (H319, H335, H315) according to CLP (GHS). Based on the available data for sodium hypochlorite these conclusions are justified as only a limited amount of hypochlorite is formed in naturally occuring water in the skin. Corrosive effects are thus not expected.