Formic acid, manufacture of, by-products from

Endpoint summary

Description of key information

Due to its corrosive properties, the local effects of formic acid prevail over systemic effects following exposures via all routes; no signs of systemic toxicity have been observed in sub-chronic and chronic studies using formate salts. Propylidynetrimethanol and pentaerythritol are of low toxicity following repeated dose, showing only unspecific effects of limited toxicological significance. The primary effect following repeated exposures to the reaction mass of 2,2-bis(formyloxymethyl)propane-1,3-diyl diformate and formic acid is local, site of contact toxicity due to the caustic properties of formic acid. 

Key value for chemical safety assessment

Repeated dose toxicity: via oral route - systemic effects

Endpoint conclusion

Endpoint conclusion:

adverse effect observed

Dose descriptor:

NOAEL

67 mg/kg bw/day

Study duration:

subchronic

Species:

rat

Quality of whole database:

Klimisch score = 1. The key study is an acceptable, well documented 90-day repeated dose study which meets scientific principles and is therefore reliable for the determination of the sub-chronic endpoint.

Repeated dose toxicity: inhalation - systemic effects

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed

Dose descriptor:

NOAEC

122 mg/m³

Study duration:

subchronic

Species:

rat

Quality of whole database:

Klimisch score = 1. The key study is an acceptable, well documented 90-day repeated dose study which meets scientific principles and is therefore reliable for the determination of the sub-chronic endpoint

Repeated dose toxicity: inhalation - local effects

Endpoint conclusion

Endpoint conclusion:

adverse effect observed

Dose descriptor:

NOAEC

62 mg/m³

Study duration:

subchronic

Species:

mouse

Quality of whole database:

Klimisch score = 1. The key study is an acceptable, well documented 90-day repeated dose study which meets scientific principles and is therefore reliable for the determination of the sub-chronic endpoint

Repeated dose toxicity: dermal - systemic effects

Endpoint conclusion

Endpoint conclusion:

no study available

Repeated dose toxicity: dermal - local effects

Endpoint conclusion

Endpoint conclusion:

no study available

Additional information

The reaction mass of 2,2-bis(formyloxymethyl)propane-1,3-diyl diformate and formic acid consists of formic acid, propylidynetrimethanol-esters and pentaerythritol-esters. The toxicity of the propylidynetrimethanol-esters and pentaerythritol-esters is predicted to be comparable to propylidynetrimethanol and pentaerythritol respectively. Due to the corrosivity of formic acid, repeated doses studies using this substance are scientifically unjustified. The repeated dose toxicity of formic acid has been determined by read-across to the formate salts. Repeated dose toxicity data on formate salts, propylidynetrimethanol and pentaerythritol are considered appropriate to meet the REACH Annex VII-X data requirements for the reaction mass of 2,2-bis(formyloxymethyl)propane-1,3-diyl diformate and formic acid. The data requirements for this endpoint have been addressed using a weight of evidence approach, taking into account available information on the components and related studies.

Formic acid – oral

Due to the corrosivity of formic acid, studies requiring repeated oral dose testing are not justified for scientific reasons and on animal welfare grounds. Formate salts have been used in studies requiring repeated oral dosing. NOAEL values obtained in such studies may be used to calculate the NOAEL for the formate anion which may be read across to formic acid, taking into account stoichiometry and formula weights. Read-across to the findings from repeated dose toxicity studies on potassium formate was therefore used to elucidate the repeated oral dose toxicity of the formate ion, and hence of formic acid. Toxicity data frompotassium diformate (formula: KCOO-COOH) are used as read across, because formic acid is liberated by this salt in a 1:1 molar ratio.

 

The repeated dose toxicity of potassium diformatehas been investigated in rats and mice in dietary subchronic 13 -week studies and in combined chronic toxicity and carcinogenicity studies. The effects seen in 13 -week and 52 -week rat studies were comparable, but rats were more susceptible than mice. The NOAEL was 600 mg potassium diformate/kg bw/day in the 13 -week rat study. It should be noted that satellite studies of the 13 -week study were also conducted. It was demonstrated that the effects were largely reversible at the end of a 4 -week recovery study. Formate blood levels remained low, i. e. below the Level of Detection (= approx. 65 μg/L) throughout the 13 -week exposure period except shortly after the nocturnal feeding. Within a few hours the formate levels were below the Limit of Detection. It is therefore assumed that formate is rapidly metabolised and does not accumulate.

 

Subchronic

 

The repeated dose toxicity of potassium diformate has been investigated in rats and in a dietary subchronic 13-week study conducted according to OECD Test Guideline 408 (Ridings, 1998). In the study potassium diformate (1:2) was administered to 10 rats/sex/dose in the diet at dose levels of 0, 600, 1200, and 3000 mg/kg bw/day. The doses were selected based on the results of a 7 day range finding study. Additional animals were used in satellite studies, i.e. a 4-week recovery study (0 and 3000 mg/kg bw/day, 10 rats/sex/dose), and in an absorption study (3000 mg/kg bw/day, 10 rats/sex/dose).

 

In the absorption study, formate blood levels were below the level of detection(LOD; 62.5 µg/mL) in most instances. Slightly increased levels in the range of 66 to 158 µg/mL were occasionally seen in some animals few hours after the nocturnal food uptake, but these declined to below the detection limit within few hours. Blood levels were also below the LOD at the end of the 4-week recovery period. These findings indicate that formate is rapidly metabolised in the rat, and that there was no accumulation over the 13-week treatment period.

 

No mortality or clinical signs of toxicity were seen in the main study. Male body weight was significantly decreased in all treated groups, whereas in females, body weight reduction gained a level of significance only in the high dose group animals. Slightly reduced food consumption was seen in groups with reduced body weights. Test substance consumption was close to the nominal doses, i.e. 0, 590, 1180, and 2880 (males) and 2950 (females) mg/kg bw/day, respectively. There were no effects in ophthalmology. No clear effects were seen in hematology and clinical chemistry, though some parameters gained a level of statistical significance. Organ weights of male and female rats were similar to those of the controls. Pathology revealed no remarkable finding except from the observation of a thickening of the limiting ridge in the stomach in all male groups and in the female high dose group at the end of the 13-week period. Histopathology revealed a dose-related increase in the incidence and severity of squamous cell hyperplasia ain the forestomach of male and female rats. This effect was largely reversible in the 4-week recovery group.

 

No overt systemic or target organ toxicity was observed in the study. A dose-related body weight decrease was seen in all treated males and in the high dose female group, possibly as a result of the reduced palability of the diet. The small but statistically significant changes in hematology and clinical chemistry parameters in the high dose animals were not of toxicological significance, due to the small changes and the lack of target organ toxicity. There was no specific target organ toxicity, except from local irritation effects in the stomach which caused a dose-related thickening of the stomach at all dose levels, and was confirmed to be squamous cell hyperplasia. This was considered to be a response to minor irritation by the test substance, rather than a toxic effect.

 

Overall, dietary administration of potassium diformate, to rats at 600, 1200, and 3000 mg/kg bw and day for 13 weeks did not produce any overt toxicity. Minor irritation had occurred in the forestomach of both sexes, with effects being seen in the males at all dose levels. Recovery was not complete in all animals after the 4-week treatment-free period.

 

The NOAEL was considered to be below 600 mg/kg bw/day, based on the irritation of the forestomach and the squamous cell hyperplasia seen at 600 mg/kg bw and day in both sexes. Systemic toxicity was not seen up to 3000 mg/kg bw/day

 

Based on the findings of the study, the LOAEL was reported to be 600 mg/kg bw in this study, based on local irritation of the forestomach.  The NOAEL for local effects was is <600 mg/kg bw.  A NOAEL or LOAEL for systemic effects was not derived in the study.

 

Chronic 

The chronic repeated dose toxicity of potassium formate (1:2) has been evaluated in a combined chronic toxicity/carcinogenicity study conducted according to OECD Test Guideline 453 (BASF, 1999; Wood, 1999, 2002). In the study rats received potassium formate in the diet at doses of 0, 50, 400 and 2000 mg.kg bw/day for 52 weeks. All dose levels were well-tolerated without effects on clinical condition or survival. Treatment related findings were noted at 2000 mg/kg bw/day and included a statistically significant depression of body weight gain and, at terminal kill, a thickening of the stomach confirmed as basal cell hyperplasia or foveolar epithelium hyperplasia in the majority of the high dose animals. These changes were less pronounced than in a previous 90-day rat study. There was no evidence of systemic target organ toxicity due to test substance administration, including the eyes. The spectrum of tumours observed in the study was generally consistent with that expected in rats of this strain. There were no tumours of an unusual nature or incidence indicative of specific target organ carcinogenicity in the stomach or any other tissue. The NOAEL for local and systemic toxicity associated with 52 weeks of treatment was considered to be 400 mg potassium diformate/kg bw/day,based on the local effects in the stomach and reduced body weight of the high dose rats. Taking the molecular weights and stoichiometry into account, this corresponds to aNOAEL of 142 mg formic acid/kg bw/day, and283 mg formate/kg bw/day

 

Formic acid – dermal toxicity

Dermal testing is not required in accordance with Annex IX, No. 8.6.2, column 2 because there is sufficient subchronic and chronic information for the oral and inhalation routes. These represent the most relevant exposure routes in general population and in workers, respectively. In accordance with Annex XI, No. 1.1, it is sufficiently known from incidential human dermal exposure that formic acid is absorbed through the skin under certain conditions (large area, high concentration) and may cause systemic toxicity. However, dermal contact should be minimised due to the corrosivity of formic acid. Therefore, human dermal exposure is generally low. Additionally, repeated dermal testing of corrosive substances is unjustified due to animal welfare considerations.

 

Formic acid – inhalation

The sub-chronic repeated dose toxicity of formic acid has been investigated in rats in a 90-day inhalation toxicity study conducted according to OECD Test Guideline 413 and under GLP conditions (Leach et al, 1989; NTP 1992). In the study, Fischer 344/N rats (10 rats/sex), were exposed via whole body inhalation to formic acid vapour at 0, 0.015, 0.03, 0.061, 0.122 and 0.244 mg/l (0, 8, 16, 32, 64 and 128 ppm; dose selection based on the results of a range finding study) for 6 hrs/day, 5 days/week for 13 weeks.

 

No mortalities, clinical signs or signs of systemic toxicity were observed in treated rats in any of the dose groups. There were no unusual gross lesions noted during necropsy and organ weights were not affected by treatment. Male and female reproductive parameters (sperm motility, density, and testicular or epididymal weight; length of the estrous cycle) were not affected. Histopathology revealed increased incidences of squamous metaplasia of the respiratory epithelium and degeneration of the olfactory epithelium in the high-dose male and female rat groups where most of the animals were affected. However, the severity was generally minimal to mild. Based on the findings of the study, local respiratory effects were the critical effects observed after sub-chronic repeated inhalation exposures to formic acid. NOAEC and LOAEC values of 64 ppm (0.122 mg/l), and 128 ppm (0.244 mg/l) respectively were determined based on irritation of the upper respiratory tract and degeneration of the olfactory epithelium and squamous metaplasia of the respiratory epithelium seen at 128 ppm (0.244 mg/L). Treatment-related systemic effects were not observed: increased absolute or relative liver weights and decreased lung weights were seen without histopathological correlation. The NOAEC for systemic toxicity was 128 ppm (0.244 mg/l), the highest concentration tested.

 

In the same study, the sub-chronic repeated dose inhalation toxicity of formic acid was investigated in mice (Leach et al, 1989; NTP, 1992). In the study, B6C3F1 mice were exposed to formic acid via inhalation using the same protocol and concentrations as those used in the rat study (10 mice/sex/dose; 0, 0.015, 0.03, 0.061, 0.122 and 0.244 mg/l (0, 8, 16, 32, 64, and 128 ppm); dose selection based on results a range finding study). There were no mortalities or treatment-related signs of toxicity in male and female mice exposed to formic acid at up to 128 ppm (0.244 mg/l) for 13 weeks. Systemic toxicity was generally low, but body weight gain was reduced in both sexes at 128 ppm, resulting in terminal body weights that were 16-20% below those of the controls. A small, but statistically significant increase of liver weight was noted in males treated at 32 ppm (0.061 mg/L) and at 64 ppm (0.122 mg/L). Findings of histopathology were limited to few cases of minimal degeneration of the olfactory epithelium. Sperm motility and estrous cycle length were not affected. Mild degeneration of the olfactory epithelium was seen at the two highest concentrations, typical for irritants. Based on histopathological changes of the respiratory tract and in the olfactory epithelium, the NOAEC and LOAEC for local toxicity were determined to be 0.062mg/L (32 ppm) and 0.122mg/L (64 ppm), respectively. The NOAEC for systemic toxicity was 128 ppm (0.244 mg/l), the highest concentration tested.

 

A LOAEC for systemic toxicity was not observed in these studies. The lack of systemic effects in both 2-week and 13-week NTP inhalation studies may be related to the rapid metabolizing capacity of formate to CO₂in rodents, due to high levels of tetrahydrofolate and 10-formyl tetrahydrofolate dehydrogenase. Levels of tetrahydrofolate and 10-formyl tetrahydrofolate dehydrogenase are much lower in humans suggesting they have a lower capacity to metabolise formate than rodents and are significantly more sensitive to its toxicity. Humans may be significantly more sensitive to the formate toxicity than rodents. Caution should therefore be used in considering the results obtained with rodents in determining potential human risks associated with systemic exposure to formic acid. Nevertheless human experience does not indicate that formic acid represents a significant systemic hazard to humans unless at high concentrations following intended or incidental ingestion or large scale skin contact, where the caustic effect also governs the toxic mode of action.

 

Propylidynetrimethanol

Oral sub-acute [28-day]

The sub-acute, repeated dose oral toxicity of propylidynetrimethanol was investigated in a 28-day range-finding feeding study in which albino rats received the substance at 0, 0.33, 1.0 and 3.0% in the diet for 4 weeks (Wijnands, 1969). Growth, symptomatology, food and water intake, weights of liver and kidney, gross pathology at autopsy and microscopic examination of liver and kidney was conducted. No distinct growth depression occurred at any dose level. Liver and kidney weights were increased in the groups fed on 1.0 and 3.0%. This increase was statistically significant in the group at the 3.0% level. Gross examination did not reveal relevant pathology. Microscopic examination of liver and kidney showed slight hepatic changes in males and females of the two highest dose-groups and minimal renal changes only at the 3% level in either sex. The NOAEL was 0.33 % in diet (approx 220 mg/kg bw/day); The LOAEL was 1 % in diet (approx. 667 mg/kg bw/day).

The sub-acute repeated dose oral toxicity of propylidynetrimethanol has been assessed in rats in a combined repeated dose and reproductive /developmental toxicity screening study conducted according to OECD Test Guideline 422 (MHLW, 1994). In the study, rats were administered trimethylopropane by gavage at doses of 0, 12.5, 50, 200 and 800 mg/kg bw/day. The NOAEL for repeated dose toxicity was 200 mg/kg bw/day based on lowered body weight compared to the controls, and significantly increased absolute and relative liver weights in males and elevated liver weights in females receiving 800 mg/kg bw/day with histopathological changes which could not be unequivocally attributed to treatment.

Oral sub-chronic [90-day]

The sub-chronic repeated oral dose toxicity of propylidynetrimethanol has been investigated in a 90-day feeding in rats (de Knecht - van Eekelen and van der Neulen, 1969, at the request of Perstorp AB). In the study, the administration of propylidynetrimethanol to young male and female Wistar rats via the diet at 0, 0.03, 0.1, 0.3, 1.0 % (corresponding to 20, 67, 200 or 667 mg/kg bw/d) for 3 months resulted in distinct ill effects mainly at the highest feeding level. At a dose level of 1 % in the diet the abnormalities consisted of slightly decreased hemoglobin levels and red blood cell counts and the presence of normoblasts and white blood cell fragments, increased relative weights of kidneys, liver and spleen and microscopical changes in liver and spleen. At the 0.3 % level the only significant differences with the controls consisted of decreased packed cell volume and red cell counts and increased occurrence of normoblasts in the blood of females. These phenomena did not occur at lower levels. Based on these findings, the NOAEL for the sub-chronic repeated dose toxicity of propylidynetrimethanol was determined to be 0.1 % in diet corresponding to 67 mg/kg bw/day.

There are no studies available on the repeated dose dermal or inhalation toxicity of propylidynetrimethanol.

Pentaerythritol

Oral sub-acute

The subacute repeated oral dose toxicity of pentaerythritol was has been assessed in a 28-day study in rodents comparable to OECD Test Guideline 407 (Hayashiet al, 1992). In the study a group of male and female F344 rats was gavaged with pentaerythritol at the limit dose of 1000 mg/kg bw/day. No deaths occurred and there were no signs of toxicity. No effects of clear toxicological significance were seen; minimal statistically significant effects on isolated clinical chemistry and haematology parameters were not considered to be of toxicological significance in the absence of consistency and pathological correlate. The NOAEL for the subacute repeated oral dose toxicity of pentaerythritol was therefore considered to be 1000 mg/kg bw/day.

In a combined reproductive and developmental toxicity screening study, conducted according to OECD Test Guideline 422, rats were administered pentaerythritol via gavage at dose levels of 0, 100, 300 and 1000 mg/kg bw/day (Yahata, 1996). Males were exposed for 46 days (starting 14 days prior to mating), and females were exposed for 14 days prior to mating, drunig mating and gestation until lactation day 3. No mortalities occurred during the study. The only treatment related effects seen in the 300 and 1000 mg/kg bw/day groups were intermittent soft stools and diarrhoea. A tendency towards increased water consumption in the top dose group was not significant and attributed to the occurrence of intermittent diarrhoea. The NOAEL for repeated dose toxicity was considered to be 100 mg/kg bw/day based on these effects; overt signs of systemic toxicity were not observed.

There are no studies on the sub-chronic repeated oral dose toxicity or the repeated dose dermal or inhalation toxicity of pentaerythritol.

The reaction mass of 2,2-bis(formyloxymethyl)propane-1,3-diyl diformate and formic acid

Due to the corrosivity of formic acid and also of the surrogate, potassium diformate, predominately local effects were seen. Systemic target organ toxicity was not observed. Body weight reductions seen in the oral study were attributable to low palability. The NOAEL for the chronic, repeated dose oral toxicity of formic acid is 142 mg formic acid/kg bw/day (read across: rat, chronic 52-week feed study; calculated from potassium diformate results; stomach irritation). This is consistent with a NOAEL of 212 mg formic acid/kg (read across: rat, chronic 13-week feed study; calculated from potassium diformate results). Repeated dose dermal toxicity studies have not been performed due to the corrosive properties of formic acid. An NOAEC of 64 ppm (0.122 mg/L; 122 mg/m³) was derived in rats for the sub-chronic, repeated dose inhalation toxicity of formic acid based on upper respiratory tract irritation). An NOAEL of 67 mg/kg bw/day has been identified for the sub-chronic repeated dose toxicity of propylidynetrimethanol in rats based on effects in the blood, liver, kidneys and spleen. Subacute repeated dose oral toxicity studies using pentaerythritol indicate low toxicity: a 28-day limit dose study reported a NOAEL of 1000 mg/kg bw/d; whereas in a combined reproductive and developmental toxicity screening study, the effects were limited to soft stool and/or diarrhoea at dose levels of 300 mg/kg bw/d and above. Comparable low repeated dose toxicity is predicted for the esters of propylidynetrimethanol and pentaerythritol. The primary effect following repeated exposures to the reaction mass of 2,2-bis(formyloxymethyl)propane-1,3-diyl diformate and formic acid is expected to be local, site of contact toxicity due to the caustic properties of formic acid. 

 

Taking account the available information, the critical NOAEL for systemic repeated dose effects is considered to be 67 mg/kg bw/day based on a sub-chronic study using propylidynetrimethanol and the critical NOAEC for local inhalation effects is 0.062 mg/L based on a sub-chronic study using formic acid in mice. Systemic effects have not been observed in inhalation studies in rodents following exposures up to 0.244 mg/L .

 

Justification for selection of repeated dose toxicity via oral route - systemic effects endpoint:
Read-across from reaction mass of 2,2-bis(formyloxymethyl)propane-1,3-diyl diformate and formic acid to available data on its components is appropriate to meet the REACH Annex VII-IX data requirements. Taking account the available information, the critical NOAEL for systemic repeated dose effects is considered to be 67 mg/kg bw/day based on a sub-chronic study using propylidynetrimethanol

Justification for selection of repeated dose toxicity inhalation - systemic effects endpoint:
Read-across from reaction mass of 2,2-bis(formyloxymethyl)propane-1,3-diyl diformate and formic acid to available data on its components is appropriate to meet the REACH Annex VII-IX data requirements. No signs of systemic toxicity were observed in inhalation studies using formic acid.

Justification for selection of repeated dose toxicity inhalation - local effects endpoint:
Read-across from reaction mass of 2,2-bis(formyloxymethyl)propane-1,3-diyl diformate and formic acid to available data on its components is appropriate to meet the REACH Annex VII-IX data requirements. Local respiratory effects were observed in an sub-chronic inhalation study using formic acid.

Justification for classification or non-classification

Repeated exposures to formic acid are associated with local, site of contact affects mediated by its corrosive properties; no overt signs of systemic toxicity are apparant. Formic acid is classified as corrosive to the skin. In inhalation studies, the LOAEC was 0.244 mg/L in a 13-week rat study, based on histopathological changes in the nose. Like the gastric effects seen in the oral studies, these are considered to be effects secondary to corrosion. A MAK-value of 5 ppm was established as early as in 1966. This value is still valid and has been adopted by many countries after several independent re-evaluations This OEL has therefore proven in practice to protect against inhalation exposure for decades. Moreover, the pungent odour of formic acid would immediately indicate elevated concentrations, thus providing additional protection.The European IOEL is 5 ppm (9 mg/m³) and may be used for the derivation of DNEL values.

Pentaerythritol and propylidynetrimethanol are shown to be of relatively low toxicity following repeated exposures; the results of the studies available do not indicate any severe toxicity and do not show any effects at dose levels of 67 mg/kg bw or lower. Similar low repeated dose toxicity is predicted for esters of these substances.

The available data on its components indicate that the reaction mass of 2,2-bis(formyloxymethyl)propane-1,3-diyl diformate and formic acid does not cause systemic or target organ toxicity after repeated oral or inhalation exposures. The substance does not meet the criteria for classification for repeated dose toxicity according to Regulation 1272/2008/EC.