[Bart]

Hi guys,

I have an issue. I performed a reaction of cyanoethylation of ammonia (1 mole) with acrylonitrile (3 moles) in reaction calorimeter in water. The product is N,N,N-tris(cyanoethyl)amine.
The process is relatively easy, it's exothermic and in reaction calorimeter I can measure the changes of the overall energy of the process, as a sum of all chemical events.

Now I would like to compare the experimental results with theoretical ones.
So here is the question.
Is it possible, by means of enthalpy of formation or bonds enthalpy, to calculate the enthalpy of this reaction when I know mostly nothing about N,N,N-tris(cyanoethyl)amine and I cannot find any piece of information in chemical tables?

I would highly appreciate any thoughts about this.

Thank you in advance,
Bart

[mjc123]

You can estimate the reaction enthalpy using bond enthalpies. What bonds are broken, and what are formed?
Bear in mind that tabulated bond enthalpy values are averages from lots of compounds, so the values in your particular compounds might be a bit different (in particular, I suspect N-H in ammonia may be different from values calculated for organic amines). But here are some values for you to try with, to get a first estimate (kJ/mol):
N-H 391
C-H 413
C-C 347
C=C 612
C-N 286
What other contributions to the reaction enthalpy might you expect when performing this reaction in water?

[Bart]

Hey,
you are right, I can calculate this from the bond enthalpies, but only if every substrate and product is in a gaseous state. To calculate "real" values I would need ΔH of vapourization, and form my product there is nothing like that.
Water is used as a reaction medium.

Cheers!

[Enthalpy]

Since ammonia and acrylonitrile have known heats of formation, all you need is ΔHf for liquid tri(propionitrile)amine, which I'm confident can be estimated well.

Bond enthalpies would be too inaccurate here, especially for the heat of a reaction that isn't very exothermal. Software would do it worse. Even for the gas, both would be unsatisfactory, and yes, you need liquids or dissolved species.

Do it by comparison with known molecules in their liquid state. In your case, without polycycles nor interactions, you can hope 6kJ accuracy.

Take the ΔHf of liquid
- Triethylamine (better than trimethylamine)
- Acetonitrile or propane nitrile
- Added -CH₂- (deduce it from octane and decane for instance, both even)
and tinker as needed.

But for instance, hypothetic data from monopropionitrilamine and ammonia would be misleading because the successive C bonded to N bring varied heat of formation, so it's a bit tricky.

The nice aspect is that you keep the interactions of the liquid molecules during the formal transformation. This is often true, an exception would be if a molecule A-B has strong intermolecular interactions between the A and B ends but not among two A not two B, then the comparison with A-A and B-B would underestimate the heat of vaporisation. Here between the tertiary amine and the nitrile, the interaction is reasonable.

It wouldn't work as nicely with solids, whose crystal arrangement depends on every molecular detail, hence doesn't compare with similar molecules.

I'd check on an other molecule with tertiary amine and nitrile groups if one has been measured.

What I haven't grasped: you put "liquid" but also "water as a reaction medium". If you want the heat of reaction in water, then it's a matter of dissolved species, not liquid. At ammonia, the difference is huge. I'd use the same method, but with the corresponding ΔHf as water solutions. If you don't find it tabulated for acrylonitrile, measure yourself the heat of solution.

[Irlanur]

You might want to look into this: https://en.wikipedia.org/wiki/Benson_group_increment_theory

shocking but not too much of a surprise: This is often more accurate than quantum chemistry.