Registration Dossier

Toxicological information

Additional toxicological data

Currently viewing:

Administrative data

additional toxicological information
Type of information:
migrated information: read-across based on grouping of substances (category approach)
Adequacy of study:
supporting study
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: Accptible publication

Data source

Reference Type:
Chemical behaviour of seven aromatic diisocyanates (toluenediisocyanates and diphenylmethanediisocyanates) under in vitro conditions in relationship to their results in the Salmonella/microsome test
Seel K, Walber U, Herbold B & Kopp R
Bibliographic source:
Mutat. Res. 438: 109-123

Materials and methods

Type of study / information:
Type: other: chemical reactions during biological test

Test material

Constituent 1
Chemical structure
Reference substance name:
4,4'-methylenediphenyl diisocyanate
EC Number:
EC Name:
4,4'-methylenediphenyl diisocyanate
Cas Number:
Molecular formula:
Constituent 2
Reference substance name:
benzene, 1,1'- methylenebis[4-isocyanato-
benzene, 1,1'- methylenebis[4-isocyanato-
Details on test material:
4,4'-diisocyanatodiphenylmethane (4,4'-MDI), Bayer-Germany, Desmodur 44M, CasN° 101-68-8
2,4'-diisocyanatodiphenylmethane (2,4'-MDI), Bayer-Germany, Casn° 5873-54-1
Mixture of isomers of monomeric MDI (4,4'-, 2,4'-, and 2,2'-MDI), Bayer-Germany, CasN° 26447-40-5 Polymeric MDI, Desmodur 44 V 20, Bayer-Hermany, CasN° 9016-87-9
SOLVENTS EGDE (Merck) DMSO for IR spectroscopy (Merck); DMSO samples had been dried and stored over molecular sieves but still contained 0.01-0.1 wt.%/wt.% of water.

Results and discussion

Any other information on results incl. tables


-When 4,4'-MDI was dissolved in EGDE, 4,4'-diaminodiphenylmethane was not formed. When a 3-fold surplus of water was added, there was 98-99% of the original diisocyanate left after 1 hour, and 85% after 4 hours. When the solution contained approximately equimolar amounts of 4,4'-MDI (4mM) and water, >99% of diisocyanate was present after 4 hours. A water: 4,4'-MDI molar ratio of approximatley 7:1 resulted in 93% diisocyanate after 1 hour, and 79% after 4hours. When the various forms of MDI, in EGDE, were added to water so as to simulate the biological test at 500 or 5000 micrograms per plate, 91-98% of the MDI was still available after 1 minute. MDA was not detected in any of the experiments involving 4,4'-MDI, MDI mixed monomers, or pMDI. The experiment with 2,4'-MDI at 500 micrograms per plate gave 0.3% conversion to 2,4'-MDA after 1 minute.

When 4,4'-MDI at 500mg/100ml EGDE was again used to simulate the biological test at 500 micrograms per plate, 96% remained after 45 seconds. 4,4'-MDA was not detected. Repitition in the presence of the test ingredients resulted in 29% remaining, and 0.6% being converted to 4,4'-MDA. -When replacing EGDE by DMSO, and using the test ingredients, the results were none detected, and 2.1%, respectively. The amount of 4,4'-MDA was 2.8% after 5 and 15 seconds.

The reactions of 2,6-TDI and water, or aqueous test ingredients, were markedly quicker than those of 4,4'-MDI, and more diamine was formed, with the exception of the DMSO/test ingredients experiment.

The authors conclude, that predissolved aromatic diisocyanates react with the aqueous biological test media during tests. MDI and TDI are converted to mixtures of compounds similar in structure , but different in composition. They assume, that the Salmonella/microsome test results of MDI and TDI do not reflect the properties of these products, but rather those of the compounds formed under the specific conditions



-IR spectra: Stability in DMSO: The reaction patterns of MDI in „dry" DMSO were comparable with those of TDI; see (B. Herbold 1998); „The N=C=O content of 50-500 mg 4,4'-MDI, dissolved in 100ml relatively dry DMSO (0.03-0.04% water), dropped to 40% or less within the first 15 min of the test. After 2 hr, the NCO absorptions began to disappear completely." Stability in EGDE: compared with the findings in DMSO, homogeneous solutions of aromatic diisocyanates in EGDE can be considered relatively stable. Even after 4 h, more than 98% of N=C=O groups still exist. -HPLC with ultraviolet detection: The advantage of this method is that the unreacted diisocynates, as well as their possible reaction products can be quantified if suitable reference compounds are available. Only 1% of the unmodified 4,4'-MDI was left over after 30 minutes. 4,4'-MDI, with a final concentration of 3%, could be detected in solution almost immediately. Several multiple peaks appeared and created complex chromatograms.

2. THE MODE OF DEGRADATION of 2,4-TDI and 4,4'-MDI, dissolved in EGDE, was different to that in DMSO, the main difference being that neither 2,4TDA nor 4,4'-MDA were found. Furthermore, solutions with a nearly 3-fold surplus of water were comparably stable. After 1 h, between 98 and 99% of the original diisocyanates were still available and after 4 h, more than 85% could be found. In a supplementary experiment, the influence of increased amounts of water on the stability of solutions of 4,4'-MDI in EGDE was monitored. In a nearly equimolar solution (4.03 mM 4,4'-MDI: 3.89 mM water), more than 99% of the diisocyanate was still present after a period of 4 h. Raising the water content to 26.11 mM, which brought the diisocyanate:water ratio to approx. 1:7, left more than 93% of the original diisocyanate after I h, and still 79% after 4 h.

3. FATE OF AROMATIC DIISOCYANATES in simulated Salmonella/microsome tests environments: At 500 and 5000 µg/test tube samples, more than 90% of the MDI was still available after 1 min. During this time, no MDA was detected in any of the 4,4'-MDI, MDI monomer mix or p-MDI runs. However, in one of the 2,4'- MDI isomer concentrations (500 mg/1OO ml EGDE), 0.2 and 0.3% of 2,4'-MDA were found after 45 and 60 s, respectively. This indicates that the 500 g 2,4'-MDI/test tube samples shaken with water for 45 or 60 s led to the deposition of traces of 2,4'-MDA on the plates. In all other MDI series, the incubation begins in the absence of MDA.

4. REACTIONS of 4,4'-MDI and water or aqueous test ingredients were markedly slower than those of 2,6-TDI. No MDA was detected when the predissolved MDI was mixed with water. Similarly, in the presence of the real test ingredients, no MDA was seen during the first 30 s of the test. However, after 45 s, 0,6% 4,4'-MDA appeared. At this time, less than 30% 4,4'-MDI remained intact, whereas in distilled water around 95% of the initial diisocyanate were still available for the bacterial test.

5. At the end of the analytical part of this study, the area concerning the stability of aromatic diisocyanates predissolved in DMSO was revisited to enable a more realistic comparison with EGDE. For this purpose, freshly prepared solutions of 2,6-TDI and 4,4'-MDI in DMSO were mixed with a complete set of the aqueous test ingredients. The chemical composition of the mixtures was monitored over a period of 45 s. The content of 2,6-TDI and 4,4'-MDI disappeared almost completely within the first minute of this test. Only around 3% of the 2,6-TDI and less than 0.7% of 4,4'-MDI were left over in solution. At the same time, up to 9.1% 2,6-TDA and 2,8% 4,4'-MDA appeared. These findings explain the positive Salmonella/microsome test results reported for TDI and four types of MDI, all of which were predissolved in DMSO.

Applicant's summary and conclusion

From the comparison of DMSO and EGDE as solvent for aromatic diiscocyanates in bacterial tests, the conclusion can be drawn that the chosen solvent considerably influences the biochemical fate of reactive chemicals.