Registration Dossier

Administrative data

Description of key information

In vivo irritation studies of boric acid on skin (Weiner et al, 1982), eye (Doyle, 1999) and respiratory tract (Kirkpatrick, 2010) were performed. Eye irritation study of boric oxide was performed (Cuperus et al. 1990).

Key value for chemical safety assessment

Skin irritation / corrosion

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (not irritating)

Eye irritation

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (not irritating)

Respiratory irritation

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (not irritating)

Additional information

Skin Irritation

No acute dermal irritation studies with boric oxide were available.Boric acid is not a skin irritant.

Eye Irritation

Boric oxide induced slight to moderate conjunctivae redness and chemosis in a primary eye irritation study in rabbits. The irritation was reversible after 24 hours with a return to near normal by 72 hours after exposure.Therefore, no classification is indicated.

Respiratory tract

Borates act as mild sensory irritants, indicated by the effects observed in humans (i. e. nose, eye and throat irritation; sneezing) and by the results of the Alarie-tests by Krystofiak & Schaper (1996) and Kirkpatrick (2010), which demonstrated a depression of the respiratory frequency in mice after exposure to sodium borate. This reflex can be triggered by agents that stimulate receptors in the respiratory tract e. g. on the trigeminal nerve (Wegman et al. 1991, Nielsen et al., 2007, Krystofiak & Schaper, 1996, Kirkpatrick, 2010). The actual mechanism, however, has not yet been elucidated. The Kirkpatrick 2010 study of boric acid using the Alarie ASTM mouse RD50 test was unable to reach an exposure concentration of boric that would cause a 50 % depression of respiration and thus did not meet the criteria specified by the test protocol for a respiratory irritant.

Wegman et al. (1991) and Woskie et al. (1998) proposed changes of osmolarity in the lining fluid of the mucous membrane as possible cause for receptor activation. Changes in osmolarity could also act indirectly by stimulating mast cells to secrete histamine or other immune modulators. Histamine is known to be able to mediate the sensory component of irritation. The importance of osmolarity in the case of borate dusts is further substantiated by Cain et al. (2008) where the changes of local osmolality from a desiccating dust may cause sensations of dryness. They also indicated that more acidic dusts, as compared to borate dusts, would lead to a change in nasal pH which might trigger the nasal receptors in a different way.

Acute irritant effects are extensively documented in human workers exposed to sodium borates (EPA, 2004; Wegman et al. 1991; Garabrant 1984, 1985; Woskie et al., 1994, 1998; Cain et al., 2004, 2008). The described symptoms are typical for those which would be produced in the exposed population rather than being an isolated reaction or response triggered only in individuals with hypersensitive airways. Symptoms include nasal and eye irritation, throat irritations, cough, and breathlessness. Boric acid exposure was only studied by Garabrant 1984 and Cain et al. 2008. The Garabrant 1984 study did not distiquish which of the two exposures (boric oxide or boric acid) was associated with reported symptoms. Boric oxide reacts exothermically with water to form boric acid suggesting a possible mechanism for boric oxide irritancy. It is believed that these irritant effects are caused by the exothermic hydration of boric oxide to boric acid. Cain et al. (2008) reported a NOAEL for irritation among human volunteers inhaling boric acid of 1.75 mg B/m3 (10 mg/m3 of boric acid), the highest exposure evaluated for boric acid. The exposures of 2, 5 and 10 mg/m3evaluated in Cain et al. did not reach a level defined by the investigators as being irritating. Furthermore, for any given point in exposure time the dose-response curve had a very low slope, not characteristic of an irritant.

In the Transitional Annex XV Dossier, Poisson regression analysis of the results from Wegman et al. (1991) was used to estimate a NOEC (See Appendix A). For NOEC derivation 15-minute interval exposure data were plotted against the sum of “any symptom” (nose, eye, and throat irritation, sneezing breathlessness, coughing; Table 37, Wegman et al., 1991). The lower limits of the exposure ranges presented in Table 37 were used for the non-linear regression analysis (Poisson-model). Applying the equation derived from the regression analysis, resulted in a predicted rate for effects at background of 0.002, with lower and upper 95 % CI of 0.0002 and 0.016, respectively. The upper 95 % CI of this rate was considered equivalent to “no-observed-effect”. The boron concentration with a lower 95 % CI of the predicted rate of symptoms equal to this value (0.016) was used as the point of departure for DNEL derivation. The corresponding boron concentration equals 0.4 mg B/m3. A correction factor of 2 was then applied for the methodological underestimation of exposure measurements resulting in a NOEC of 0.8 mg B/m3. However, Wegman et al. did not study boric acid or boric oxide but instead an alkaline dust containing unspecified sodium borates. And sodium borate information can not be extrapolated to boric acid or boric oxide for irritant responses.

The Wegman data is based on subjective responses on a severity scale assigned to exposure ranges rather than a specific exposure level and contains no clear dose-response information. There is no way to identify where in this exposure spectrum symptoms occurred. Furthermore, symptoms were also reported in the group of workers not considered to be exposed (office workers), making any estimate of the NOEC unreliable.

Benchmark dose analysis was conducted of the data presented in Table 37 of Wegman et al. (1991) (See Appendix A). Table 37 presents the incidence of “Any Symptom” reported by a participant in the study that was confirmed by both the marker being pressed on the data logger worn by the worker and by a subsequent questionnaire administered by a study technician. The exposure doses used were the calculated mean concentration of each concentration range presented in the table. The identified dose-descriptor for acute irritant effects is the BMDL05 value of 0.94 mg B/m3 based on Wegman et al. (1991). The methods used for exposure measurements in this study were underestimates and a conversion factor of 2.5 was used to correct for the methodological underestimation of exposure measurements. This results in a final BMDL05 of 2.35 mg B/m3 for exposure to sodium borate dusts.

An airway sensory irritation respiratory depression (RD50) study of boric acid and sodium tetraborate pentahydrate was conducted in male Swiss-Webster mice based on the ASTM E981-04 (2004) standard test method of estimating sensory irritancy of airborne chemicals. The ASTM E981-04 sensory iritancy test (Alarie assay) has been demonstrated to be a reliable test for estimating sensory irritancy of airborne irritants and RD50s are a basis, at least partially, for OELs by ACGIH (Kuwabara et al. 2007). ECHA guidance (Chapter R.8) acknowledges the use of the Alarie assay in assessing respiratory irritation.

It was not possible to achieve an aerosol concentration high enough to result in a 50 % respiratory depression in mice for boric acid based on the results in the mouse sensory irritation model. The highest concentration of boric acid that was achievable with acceptable control of the aerosol concentration was 1096 mg/m3 with a %RD of 19%. Based on these results, the RD50 is > 1096 mg/m3 for boric acid. The ASTM standard uses the value of 0.03 x RD50 for estimation of threshold limit values (TLV). Alarie et al. (2001) has established that a value of 0.01 x RD50 as the concentration where no sensory irritation would be seen in humans. Therefore, although the highest achievable concentration was below the RD50 value for boric acid, based on the high aerosol concentrations achieved with %RD values below 50%, it is clear that boric acid is not a respiratory irritant or at worst has extremely low potency as a sensory irritant. The practical side of these results is that occupational exposure limit of 10 mg/m3 total particulate will prevent any sensory irritation in workers.


Justification for selection of skin irritation / corrosion endpoint:
No study is selected since all studies conducted with boric acid are negative.

Justification for selection of eye irritation endpoint:
No study is selected since all studies conducted with boric acid and one study with boric oxide are negative.

Justification for classification or non-classification

Boric oxide was not classified for skin or for eye irritation under criteria defined in directive 67/548/EEC as no irritating effects were observed on application of boric acid to the skin or eyes, or of boric oxide to eyes of test animals. All scores were under the cut-off values triggering classification and labelling. Therefore, boric oxide does not meet criteria under EU CLP Regulation (EC) No. 1272/2008) for classification and labelling as a skin or eye irritant. Boric oxide does not meet the criteria for a respiratory irritant.