Energetic dendrimers 2: "cyanuric picrates"

Continuing with my theme of energetic dendrimers, I want to propose another. In the review paper attached (1), I noticed that beside for amines, phenolic derivatives could be condensed with a cyanuric chloride-like derivative to yield not melamine centred dendrimers, but cyanurates. This gave me the idea of forming a "cyanuric picrate" or perhaps more correctly, a tris(trinitrophenyl) cyanurate:



This structure would probably be twisted into a propeller shape to minimise the steric hindrance of the nitro groups.

I thought it might hypothetically be prepared by from reacting three equivalents ammonium picrate and cyanuric chloride in ethanol, forming ammonium chloride and the cyanuric picrate?

Is the organic chemistry proposed even plausible?

I don't expect this to have particularly special explosive performance, but it might have interesting explosive properties... possibly making it useful once characterised. As I've stated before, energetic dendrimers may be useful in binary combinations with other energetic materials. I'm also hoping that maybe this version of an energetic dendrimer has a chance of being insensitive.

Reference (attached for your convenience):

(1) Mackay B. Steffensen, et al. Dendrimers Based on [1,3,5]-Triazines. J Polym Sci Part A: Polym Chem 44: 3411–3433, 2006.

Attachment: Dendrimers Based on [1,3,5]-Triazines.pdf (1.2MB)
This file has been downloaded 671 times

[Edited on 6-12-2014 by deltaH]

deltaH,

In 1966, the known explosive chemist Micheal D. COBURN from the Los Alamos National Laboratory (LANL) synthesized several s-triazine based energetic materials [1] similar to yours. Two of these structure along with their experimental density are presented below:



Compound (I) can be prepared via two steps: first the reaction of aniline with cyanuric trichloride and then nitrating the formed product with concentrated HNO₃/H2SO₄ at 70°C for 3 hours.

Further nitration of (I) with a mixture of acetic anhydride/HNO₃ mixture give you compound (II) which is unstable (rapid decomposition at 100°C and slow decomposition at 0°C).

Picrate salt are weak nucleophile so it is unlikely that a picrate salt will react with cyanuric trichloride to give your proposed structure. A multi-step synthesis is required. Triaryl Cyanurates can be easily synthesized by the reaction of cyanuric trichloride with sodium salt of hydroxyaryl compounds (ArONa) in water under microwave irradiation or under other classical conditions (see ref. [2] and references therein). For example, the sodium salt of para-nitrophenol react with cycanuric trichloride in water under microwave condition to yield the corresponding triaryl cyanurate (see picture below) in 85% yield. The reaction time was only 1 minute.



This product can be further nitrated to the corresponding 2,4,6-trinitroderivative.

References:

[1] M. D. Coburn, J. Heterocycl. Chem. 1966, 3, 365-366.

[2] A. D. Sagar, D. S. Patil, B. P. Bandgar, Synth. Commun. 2000, 30, 1719-1723.

Dany.


[Edited on 7-12-2014 by Dany]

There is a good and facile way described by Thurston et al. [1] to synthesize triphenyl cyanurate and this by reacting cyanuric trichloride with phenol in the presence of sodium hydroxide and acetone as solvent. The procedure is as follows:



You can see that the reaction works very well (95% yield) and can be performed on multigram scale. The authors also mentioned that cyanuric trichloride would not react completely in the absence of organic solvent even at 75°C.

References:

[1] J. T. Thurston, F. C. Schaefer, J. R. Dudley, D. Holm-Hansen, J. Am. Chem. Soc. 1951, 73, 2992-2996.

Dany.

deltaH,

A related s-triazine based energetic materials called PL-1, displays important chemical and physical properties. PL-1 was synthesized by Indian scientist [1] at the High Energy Materials Research Laboratory , Pune, India. The structure of PL-1 along with it's experimental crystal density is shown below:

As one can see PL-1 possess astonishing crystal density of 2.02 g/cm³ which is one of the highest reported density for an energetic material.

The synthesis of PL-1 is achieved via 3 steps:

1) reaction of 3,5-dichoroaniline with cyanuric trichloride 2) nitration of the resulting product 3) reaction with ammonia which displays the chlorine atoms and form PL-1. However, the reported overall yield is relatively low (31%).

The authors estimated the detonation velocity of PL-1 by the method described by Rothstein & Petersen [2]. The method is simple and require only the chemical composition and the density of the explosive. Based on d= 2.02 g/cm³ the calculated detonation velocity is 7861 m/s. The detonation pressure is 312 kbar. Detonation pressure can be obtained by the following formula:

P= ρD²/ɣ+1, where D is the detonation velocity (in km/s), ρ is the density, ɣ is the adiabatic coefficient of the detonation gases at the Chapman-Jouguet point (as an approximation ɣ= 3).

However, a more recent theoretical study [3] (done using DFT computational method) on PL-1 reveals that this molecule has a heat of formation of ΔHf^O= +427.6 kJ/mol. Based on this heat of formation and the reported crystal density of 2.02 g/cm³ the authors estimated the detonation velocity and pressure by the Kamlet-Jacobs [4] method which is better than Rothstein & Petersen method cited above.

The calculated D and P are: D= 8500 m/s; P= 355 kbar. Detonation performance close to RDX.

Finally it should be noted that PL-1 is a thermally stable and insensitive explosive. PL-1 display a DTA (Differential Thermal Analysis) exotherm at 335°C.

References:

[1] V. K. Bapat, A. K. Sikder, Mehilal, B. G. Polke, J. P. Agrawal, J. Energ. Mater. 2000, 18, 299-309.

[2] L. R. Rothstein, R. Petersen, Propellants, Explos., Pyrotech.1979, 4, 56-60.

[3] X.-H. Ju, Z.-Y. Wang, J. Energ. Mater. 2008, 27, 51-62.

[4] M. J. Kamlet and S. J. Jacobs, J. Chem. Phys. 48 (1), 23–35 (1968).

Dany.

[Edited on 8-12-2014 by Dany]

I did my final year research project in Pune at the National Chemical Laboratory on an exchange program

PL-1 is impressive and clever chemical design/synthetic methodology.

Thanks again for the valuable literature contribution!

Quote: Originally posted by PHILOU Zrealone

Please be cautious ! Ar-NH2 and Ar-NH-Ar do form upon nitration Ar-NH-NO2 and Ar-N(NO2)-Ar. The nitraniline compound is unstable and free NO2(+) to another place on the ring in exchange of a H(+). So yes Ar-N(NO2)-Ar is "unstable" vs Ar-N(NO2)-Alk especially if both Ar ring are strong electro-withdrawing groups like triazine or picryl. So what is true for Ar-NH-Ar might not be true for Ar-NH-Alk like the TRIS variant you mentionned in another tread. --> Tetryl and Heptryl are known existing molecule with practical explosive applications. Thanks Dany, the PL-1 molecule is one more proof that polymeric material are a good way to increase density, power and VOD and decrease sensitivity. Do you have by chance the datas for the same molecule as PL-1 but with a TATNB (triaminotrinitrobenzene) core instead of a melamine (triamino-triazine) one? Note that the final stage with NH3 in the synthesis of PL-1 could be done with CH3-NH2 to form after nitration a tetryl related PL-1. Also note that PL-1 could react in dendrimer extension reaction with trinitrochlorobenzen to add an extra perimeter of 6 picryl on the 6 external amino groups. The resulting planar structure should display even more interesting properties (d, IS, VOD, Energy of formation and detonation) PL-1 should be acidic by the NH inner core and external NH2...so it might form energetic salts of interesting properties, like by order of increasingly interesting properties: -hydrazinium -hydroxylamonium -H2N-CH2-C(NO2)3 (trinitroethylamonium) -H2N-NH-CH2-C(NO2)3 (trinitroethylhydrazinium) [Edited on 9-12-2014 by PHILOU Zrealone]

Philou,

My search didn't yield and results for the TATNB core instead of a melamine. The high density of PL-1 is NOT the result of the polymerisation but by the presence of multitude of intra- and intermolecular hydrogen bonding between the amino and the nitro groups. If the amino groups were replaced by methyl the density will be far less than 2.02 g/cm³.

To design a dense explosive one should think about dense molecule that yield good crystal packing. Polycyclic molecule are known to be good candidate. Example include TEX (4,10-Dinitro-2,6,8,12-tetraoxadodecane) an explosive [1] with very high density (d= 1.985 g/cm³.

References:

[1] E.-C. Koch, Propellants, Explosives, Pyrotechnics 2014, DOI: 10.1002/prep.201400195

Dany.

[Edited on 9-12-2014 by Dany]

Until now the discussion is focused on the derivatives of 1,3,5-Triazine as energetic materials. However, it should be kept in mind that 1,3,5-Triazine possess two other positional isomers: 1,2,3-Triazine and 1,2,4-Triazine shown below:

A theoretical study [1] was performed on all possible nitro derivatives of the three rings using DFT and ab initio computational methods. The fully nitrated structures are shown below:

As one can notice, structure (III) has the highest calculated crystal density (1.93g/cm³ while structure (II) has the highest Heat Of Formation (HOF) and also the highest detonation performance (calculated using the Kamlet-Jacobs method).

References:

[1] YAKUP ÇAMUR, A COMPUTATIONAL STUDY ON NITROTRIAZINE DERIVATIVES, MIDDLE EAST TECHNICAL UNIVERSITY, 2008, Master thesis.

[2] M. J. Kamlet and S. J. Jacobs, J. Chem. Phys. 48 (1), 23–35 (1968).

Dany.

[Edited on 9-12-2014 by Dany]

So one needs only the gas phase heats of formation, not the solid? I didn't realise that, my bad.

Density from a molecular modelling point of view is hard to accurately predict for hypothetical EM structures... I've tried it several times, but honestly, it still seemed to me like a very woolly science (the density predictions, not molecular modelling per say). Eventually I decided it wasn't worth the effort (or computational wait for that matter )

[Edited on 10-12-2014 by deltaH]

Dany,
There must be something else...for PL-1 to reach 2.02 g/ccm

1)Both free "monomeric" molecules display lower densities than PL-1 while diplaying the same atom-groups and thus the same subsequent NH2-NO2 and NH2-N interractions...
a)1.3.5-Triamino-2.4.6-trinitrobenzene (TATB) density: 1.93g/ccm
b)2.4.6-Triamino-1.3.5-triazine (cyanuric triamide/melamine) density: 1.574 g/ccm
c)a mix of 3/4 TATB and 1/4 melamine would have a density of approx 1.84 g/ccm

2)The free molecules do have more freedom of move than the arms of PL-1 vs its core to ensure best packing for cristal lattice. Also PL-1 should display structural voids.

--> I really think it has to do with the condensation/polymerisation patern.

For obvious reasons CH3 is a bad densophore group vs NH2 and NO2, so your example is a bit tricked .

PL-1 is a polycyclic compound and so it enters your rule...just as my putative 2D or 3D polymer .

TEX is a caged polycyclic compound.

I repeat here that finite molecule are a limitation for the density increase purpose!
As an image, a simple study of the density of dry sand, silice glass and silice cristal (all 3 SiO2)should give you a hint:
-Dry sand (representing finite molecules packing) d=1.6
-Silice glass (representing 3D polymeric) d=2.53
-Silice cristal (representing 3D max cristallinity polymer) d=2.65
Where does this come from? This is due to the contraction of volume resulting from the creation of chemical bonds...those are by definition way shorter than H bondings molecular interactions ...the only interaction comming close to chemical bonding is ionic charge interactions (like in salts...and this explains that CHNO salts display good densities).

Quote: Originally posted by Dany

Quote: Originally posted by PHILOU Zrealone   Thanks Dany, the PL-1 molecule is one more proof that polymeric material are a good way to increase density, power and VOD and decrease sensitivity. Do you have by chance the datas for the same molecule as PL-1 but with a TATNB (triaminotrinitrobenzene) core instead of a melamine (triamino-triazine) one? Also note that PL-1 could react in dendrimer extension reaction with trinitrochlorobenzen to add an extra perimeter of 6 picryl on the 6 external amino groups. The resulting planar structure should display even more interesting properties (d, IS, VOD, Energy of formation and detonation) PL-1 may react further with cyanuric tri-chloride to extand the molecule to a 2D polymer (or 3D polymer) alterning triazine and TNB ring bridged by -NH- groups... The high density of PL-1 is NOT the result of the polymerisation but by the presence of multitude of intra- and intermolecular hydrogen bonding between the amino and the nitro groups. If the amino groups were replaced by methyl the density will be far less than 2.02 g/cm<sup>3</sup>. To design a dense explosive one should think about dense molecule that yield good crystal packing. Polycyclic molecule are known to be good candidate. Example include TEX (4,10-Dinitro-2,6,8,12-tetraoxadodecane) an explosive [1] with very high density (d= 1.985 g/cm<sup>3</sup>.

Hydrogen bonding can even lower density for solids, here's why:

Consider the richly hydrogen-bonded melamine cyanuric acid 1:1 complex, the most extensively/densely hydrogen bonded material I know of, off the top of my head. It has a Specific Gravity of only 1.7.Compare to the SG of pure cyanuric acid (2.5) and melamine (1,6), so the hydrogen bonding actually made the density less than what you would expect (something closer to 2.05).

If you look at the structure of the complex, you see that by alligning all the hydrogen bonds, one forces the solid structure into a packing that has more wasted space.

*****************

I agree with PHILOU, this is about bond lengths. In a way, dendrimers can be thought of as a 2d compromise between the 3d bonding of caged energetics and 1d discretes. The compromise on density is also one of difficulty of synthesis.

As such, I believe they may hold valuable promise

[Edited on 11-12-2014 by deltaH]