ACD/Boiling Point

Planning a distillation in the near future? Hoping to characterize your new compound with measurement of a simple physicochemical property? We suggest you think twice before relying completely on the nearest Catalog Handbook of Fine Chemicals published by a major supplier.

Fused-ring faux pas

From what we can see, there are some discrepancies that need to be addressed. Take 9-vinylanthracene, for example. The boiling point is listed as 61-66°C (at 10 torr). That is questionably low, compared to the analogue 9-methylanthracene, also listed in the Catalog at 196-7°C (at 12 torr). Is it reasonable that a switch from a vinyl to a methyl side group gives such a pronounced difference in boiling point? The value predicted by ACD/Boiling Point for 9-vinylanthracene is 214.6°C (at 10 torr).

9-vinylanthracene		9-methylanthracene
Aldrich: 61-66°C/10 torr		Aldrich: 196-7°C/12 torr
ACD/BP: 214.6°C /10 torr		ACD/BP: 195.6/12 torr

Fluorinated Folly

Or how about 2,4-dichlorobenzotrifluoride? Its boiling point is listed in the Catalog as 117-8°C with no pressure stated, leading one to presume it is a 760 torr measurement. Again, we find this boiling point questionably low; a very near relative to this compound, 2-chlorobenzotrifluoride, for example, has a boiling point at 152°C and another close relative, 1,3-dichlorobenzene has a boiling point of 172-3°C. The value predicted by ACD/Boiling Point for 2,4-dichlorobenzotrifluoride is 193°C.

2,4-dichlorobenzotrifluoride		2-chlorobenzotrifluoride
Aldrich: 117-8°C		Aldrich: 152°C
ACD/BP: 193.2°C		ACD/BP: 155.7°C

Contradictory azetidinones

Catalog values for the azetidinone family appear to be slightly problematic, too. The boiling point for 2-azetidinone is listed as 106°C/15 torr which corresponds to 222°C/760 torr. However, its ester derivative, 4-acetoxy-2-azetidinone, is listed as 80-82°C, with no pressure given so presumably it is at 760 torr. The ester boiling point seems questionably low to us. The value predicted by ACD/Boiling Point for 4-acetoxy-2-azetidinone is 297°C.

4-acetoxy-2-azetidinone		2-azetidinone
Aldrich: 80-82°C		Aldrich: 106°C/15 torr ; 222°C
ACD/BP: 296.5 °C		ACD/BP: 105.6/15 torr; 222.4°C

Dienal Discrepancy

The trans, trans-2,4-dienal set of compounds have Catalog boiling points of 115°C /10 torr and 97.5°C /10 torr for the decadienal and nonadienal forms, respectively. These appear to be in agreement, but not with their heptadienal homologue, which is listed at 84-5°C with no pressure given (therefore, presumed to be 760 torr). The value predicted by ACD/Boiling Point for trans, trans-2,4- heptadienal is 49°C/10 torr, corresponding to 160(9)°C.

trans, trans-2,4-heptadienal		trans, trans-2,4-nonadienal
Aldrich: 84-5°C		Aldrich: 97.5°C /10 torr; 222.4°C
ACD/BP: 177.4°C		ACD/BP: 114.4°C/10 torr; 244.6°C

The Trouble with Phosphorus

Some phosphorus-containing compounds show very wide discrepancies. In certain cases, we found the B.P. values in another database, the Chapman and Hall Chemical Database, at reduced pressures. These values compared very well with those from our database, but served to cast further doubt on the values found in the Aldrich catalog. For example:

Trimethyl phosphonoacetate Aldrich: 118°C ACD/BP: 253.5°C [from DB]		Trimethyl phosphonoacetate Chapman & Hall DB: 87-90°C/0.2 torr ACD/BP: 65.7°C/0.2 torr [from DB]

By analogy, a compound closely related to trimethyl phosphonoacetate--trimethyl 2-phosphonoacrylate-suffers from a similar problem. This compound had an error of 165ºC at 1 atm pressure!

trimethyl 2-phosphonoacrylate Aldrich: 91°C/760 torr ACD/BP: 258.9oC

Hall of Shame

The following compounds show differences of 200°C or greater between the catalog-tabulated values and the ACD/Boiling Point values. Below are listed the values from the Chapman and Hall Chemical Database, which give independent confirmation of the ACD/Labs values at low-pressure measurements:

2,2,2-Trichloroethyl phosphorodichloridate Aldrich: 56-58°C ACD/BP: 269.4°C	2,2,2-Trichloroethyl phosphorodichloridate Chapman & Hall DB: 116-118°C/14 torr ACD/BP: 137.4°C/14 torr
2,2,2-Tribromoethyl phosphorodichloridate Aldrich: 91°C ACD/BP: 298.1°C	2,2,2-Tribromoethyl phosphorodichloridate Chapman & Hall DB: 90-92°C/0.4 torr ACD/BP: 89.3°C/0.4 torr
Diethyl cyanomethylphosphonate Aldrich: 102°C ACD/BP: 305.2oC [from DB]	Diethyl cyanomethylphosphonate Chapman & Hall DB: 88-94°C/0.02 torr ACD/BP: 58oC/0.02 torr [from DB]
Triethyl 2-fluoro-2-phosphonoacetate Aldrich: 75°C ACD/BP: 339.5°C [from DB]	Triethyl 2-fluoro-2-phosphonoacetate Chapman & Hall DB: 125°C/0.7 torr ACD/BP: 133.2°C/0.7 torr [from DB]
Triethyl phosphonoformate Aldrich: 118°C ACD/DB: 267.8°C [from DB]	Triethyl phosphonoformate Chapman & Hall DB: 103-105°C/0.08 torr ACD/BP: 101.0°C/0.08 torr [from DB]

Conclusion

We think these examples are enough to deliver a simple message: the tabulated values have some inconsistencies. We realize that the chemical supplier is not in the business of measuring boiling points, and we don't really want to criticize, because in many cases their listed boiling points are fine. But if you need an estimate you can depend on, we stand behind ACD/Boiling Point.

Our predictive software, ACD/Boiling Point, is more likely to give you a realistic value for the temperature at which you can expect your compound to boil. A survey of some 6,000 experimental boiling points, summarized in the accompanying figure, shows that our predictions normally lie within 5°C of the true value.

We also calculate the uncertainty for the prediction. Our software will show, as a graph or a table, the boiling point dependence over the range of pressures from 0.001 to 7,600 torr. And all this is run from a simple intuitive sketching interface.

ACD/Boiling Point is an empirical additive-constitutive method, which interpolates or extrapolates, based on the compound structure, from several thousand known values. The algorithm was developed from solid approximations made by our world-class team of physical chemists. When it comes to boiling points, we are second to none.