Chemical Forums
Chemistry Forums for Students => Organic Chemistry Forum => Organic Chemistry Forum for Graduate Students and Professionals => Topic started by: Enthalpy on May 21, 2018, 05:58:06 AM
-
Dear all,
would you know a trick to guess whether one hydrazine is badly toxic?
Hydrazine, MMH, UDMH are being phased out from satellites (ton amounts) and launchers (hundreds of tons) for being bad carcinogens, toxic, volatile, flammable - yuk. But others have no bad reputation. Tetramethylhydrazine is only classified as flammable. Pimagedine (aminoguanidine) was tested as a drug.
The Osiris applet tries to guess the toxicity of compounds
www.organic-chemistry.org/prog/peo
but I don't understand its criteria, maybe subsequences of N-N-C substituted or not. Not perfect: the applet identifies only a "tumor medium risk fragment" in UDMH.
Methylated hydrazines were preferred for performance, but other hydrazines may be safer, more efficient than equivalent amines and hydrocarbons, and easy to mass-produce. The Osiris applet imagines that diazetidyl and gem-dicyclopropylhydrazine are not very toxic. Both give 7s more specific impulse than RG-1, and gem-dicyclopropylhydrazine isn't quite flammable.
A tentative synthesis of gem-dicyclopropylhydrazine is appended. Variants include CHBr2Cl, CHCl3. Recycling the halogens on site would reduce the risks and costs, producing at an ammonia plant would limit the transports.
Thank you!
Marc Schaefer, aka Enthalpy
-
1). Hydrazine is basic and therefore, irritant and caustic to eyes, skin, as well to respiratory and gastrointestinal tract. Substituted hydrazines are less basic and therefore less irritant.
2). Hydrazine has intercalating properties, meaning that it can penetrate between the DNA chains, as being a small molecule and destroys the "zip-fastener" structure of DNA by forming H-bonds with DNA-bases and consequently, hydrazine is a highly carcinogenic and mutagenic substance. Substituted hydrazines are weaker intercalating agents, depending on their substitution and molecular size and therefore, less carcinogenic and mutagenic.
PS: Indeed, “Osiris” is a very useful application.
-
Thanks pgk!
"Less carcinogenic and mutagenic" sounds discouraging, when considering hundreds of tons...
I had hoped some would be harmless enough, after Osiris (only a soft) ranks them so and pimagedine was swallowed by volunteers who survived over the trial duration.
So it would be a mere matter or size, not of combination of N and C substituted or not, secondary or tertiary...? Osiris tries to make such subtle differences, but I didn't grasp its explanations.
Or rather, some hydrazines are nasty and the others still unknown?
-
Or rather, some hydrazines are nasty and the others still unknown?
I would hazard that this is the best analysis. Anything more than the most basic toxicoligcal info is only collected with exhaustive, time consuming, and highly expensive research. Think of how much work it takes to get a compound through the FDA. I currently am working "with" the EPA to get some of our chemicals approved through TSCA, and we are very nervous because doing tox testing is quite expensive. I think its really hard to give a blanket toxicity statement to any compounds aside from the most obvious (like, say, hydroperoxdies are probably all bad for you).
-
There is no need to be discouraged. Some of them are neither carcinogenic nor mutagenic, such as pimagedine (aminoguanidine) and [2-(dimethylamino)ethyl]hydrazine.
You can search for the MSDS (Material Safety Data Sheet) of various substituted hydrazines and see their toxicity data in sections in Section 2 and Section 11, as well as their environmental impact in Section 12 and see what it fits.
e.g. MSDS Pimagedine Hydrochloride (aminoguanidine HCl),
https://www.caymanchem.com/msdss/81530m.pdf
MSDS [2-(dimethylamino)ethyl]hydrazine
http://datasheets.scbt.com/sds/aghs/en/sc-340263.pdf
Hint 1: Toxicity/ carcinogenicity, mutagenicity subcategories.
Category 1: Mild/ Suspicious to be but not proven in animals and humans
Category 2: Medium/ Proven in animals but not in humans
Category 3: Severe/ Proven in animals and humans
Hint 2: Roughly, carcinogenicity/mutagenicity category 1 classification permits the commercialization for professional use only but not the use in consumer goods. However, there are exceptions depending on the content whether being < 10% or < 1%, in relation with the subcategory.
Hint 3: Some manufactures/suppliers (e.g. Aldrich) refer to SDS (Safety Data Sheet), regarding the Safety document.
PS: “Osiris” is a useful tool to pre-estimate the toxicity of a chemical substance but you cannot trust a software for the safety of hundreds of tons of that substance. Contrary, what is cited in MSDSs and SDSs, is double checked.
-
But attention, if the product is entirely new with unknown toxicological and environmental data, you have to “co-work” with the competent authorities, e.g. EPA, OSHA in USA, ECHA in EEC member countries, etc. and being in accordance with the REACH Legislation for the commercialization/stocking in Europe, quantities > 1 ton per year.
-
**mod wildfyr ** I am sorry enthalpy, I accidentally edited your post instead of quoting it! You will have to repost your response. :-[
-
Never mind, I remember what it was about:
How is the cancer risk measured? Exposing to the compound a group of mice with abnormal tumour tendency? Or can it be done on culture tissues?
And if the test lasts for a few months (does it?), is it a decent indication of long-term effects in humans?
How expensive is such a test: 103, 104, 105 €?
-
I think its in the ballpark of a hundred thousand euros.
-
How is the cancer risk measured? Exposing to the compound a group of mice with abnormal tumour tendency? Or can it be done on culture tissues?
Ames test is among the first (in vitro) steps.
https://en.wikipedia.org/wiki/Ames_test
-
Ouch.
Well, if a fuel is for sure ze very best magic thing, 105€ is cheaper than the sale price for the first flight. But testing a dozen fuels would be the wrong way.
Nice that in vitro tests can provide a first filter step.
-
Ahum. The intermediate dicyclopropylamine seems hard to obtain, since several teams proudly announced their complicated way. The only similar route I saw uses alkoxy-oxysilane-cyclopropane.
I had hoped the increased reactivity of gem-dihalo would ease this step.
Or maybe all these teams want to use only liquids at moderate temperature?
-
In case the route over dicyclopropylamine makes any sense, I've observed the aspect of some molecules in the ArgusLab software.
Di(bromocyclopropyl)amine is badly overcrowded, and chloro too. Starting this step with monohalocyclopropane should be less bad.
Bromodicyclopropylamine is strained, but chlorodicyclopropylamine little so and should be easier to obtain.
At least plain cyclopropane is stable at +400°C, while the usual Olin-Raschig process (ammonia with hypochlorite over chloramine to hydrazine) needs +5°C and +130°C.
-
1,2-Diazetidines too have a N-N bond but the Osiris applet estimates they're harmless. They would be about as efficient as other strained amines.
Alkenes can make cis-1,2-cyclobutanes by photoaddition. I haven't seen the equivalent reaction with imines, so the appended tentative syntheses need luck. At least, they look simple.
With C+N=10, the dicyclopropyl would still catch fire a bit easily if warm (estimated bp=+190°C) and should be liquid at RT for being unsymmetric and deformable. The dimethyl would be badly flammable (estimated bp=+93°C) and the unsubstituted 1,2-diazetidine too (estimated bp=+115°C). Horvitz, Pope and Colgain synthesized the dimethyl in 1968 by a laboratory route
https://archive.org/download/DTIC_AD0042818/DTIC_AD0042818.pdf
and measured bp=+70°C, mp=-90°C, ρ=810kg/m3. The unsubstituted may freeze more readily.
I can't predict the stability at heat.
Marc Schaefer, aka Enthalpy
-
Perhaps you can save a step and react the monobromo cyclopropane with methanimine.
I'm pretty uncomfortable with your notation, are the hydrogens implied in your plain "N"s?
-
Thanks wildfyr!
Err, yes, H are implied - but I had to write them in CH2O to distinguish from carbon monoxide.
-
TheUnassuming hinted (thanks!) at efficient N-N coupling by O2+catalyst or by electrolysis
http://www.chemicalforums.com/index.php?topic=97919.msg345959#msg345959
https://pubs.acs.org/doi/10.1021/jacs.8b05245 and https://pubs.acs.org/doi/10.1021/ja5013323
which achieves molecules as crowded as tetraphenylhydrazine, so cyclopropylmethylhydrazines look accessible. This needs a good path to dicyclopropylamine, as some previous compounds here do.
The dicyclopropyldimethyl would be slightly flammable but a mix with heavier compounds shall improve it and lower the freezing point, the isomers being welcome too.
The Osiris applet estimates that these hydrazines are harmless (but it considers Udmh as slightly carcinogenic).
The seemingly cheap N-N coupling would apply to other compounds here like diazetidyl.
Cyclopropyls are expected to weaken the N-N bond further. Bad for the heat stability, but it may improve the hypergolic ignition with N2O4.
Marc Schaefer, aka Enthalpy
-
Careful hand estimations of the heats of formation, as liquids at 298K, give all cyclopropylmethylhydrazines +7s specific impulse advantage over RG-1 "kerosene" rocket propellant.
The Osiris applet can't give certainty about the health risks of cyclopropylmethylhydrazines, which are 3s short of dangerous Mmh and Udmh but only 2s better than the supposedly safe diazetidylcyclopropane, so alternatives exist.
Marc Schaefer, aka Enthalpy
-
I tried to estimate the heats of formation of cyclopropylated and cyclobutylated diazetidines, but insufficient data lets cumulate many uncertainties.
- Dicyclobutyl was measured by Katorgyn, so 1,2-dicyclobutylcyclobutane is accessible.
- To 1,2-dicyclopropylcyclobutane, I applied the same +46.2kJ/mol as from methylcyclobutane to methylcyclopropane.
- From tertiary C-C to N-N, I applied +251.4kJ/mol because it's +175.1 from propane to Mmh, +200.0 from butane, +203.1 from isobutane.
- And I supposed that the ring strain acts on diazetidine as on cyclobutane.
N,N'-Dicyclopropyl-1,2-diazetidine is then 10s better than RG-1, as efficient as Mmh. The Osiris applet doesn't consider it badly toxic, but that's only software. C+N=10 makes it flammable if lukewarm, similarly to turpentine.
Mixing some cyclobutylcyclopropyldiazetidine raises the flash point and depresses the melting point. Being 8s better than RG-1, it wastes little performance.
While matching the perfomance of Mmh would be an achievement, dicyclopropyldiazetidine can't replace it on existing rockets, as the mix ratios differ too much. The photocoupling synthesis I suggested on June 23, 2018 is hypothetic too.
Marc Schaefer, aka Enthalpy
-
So the diazetidines: typical C-C bond strength is 350 kJ/mol. typical N-N bond strength is much less - 165 kJ/mol. Meanwhile, cyclobutane ring has strain energy of 110 kJ/mol - so the bond strength in the diazetidines will only be about 55 kJ/mol - that is 13 kcal/mol. Bonds this weak can't be isolated at room temperature.
I would be vaery careful with all of these compounds - put me in mind of mustard gases.
-
I dont think they are alkylating but hydrazines are never non-toxic. They could be impossible to synthesize and non-stable.
-
Hi Rolnor, thanks for your interest!
One nice improvement here is that the hydrazines are far less volatile than Mmh or Udmh. If a leak occurs at a satellite or rocket, people won't lick it, but Mmh is as volatile as ethanol, ouch.
At least N, N'-dimethyl-1,2-diazetidine was synthesized, although in small amount using a path unsuitable to rocket amounts: ethylene bromide and sym-dimethylhydrazine, big dilution, low yield. It was stable enough to measure the 70°C boiling point. Appended is a screenshot of page numbered 28 in the report, is 39/123 in the Pdf
https://apps.dtic.mil/dtic/tr/fulltext/u2/093924.pdf
The page 80 of the document uses an estimated heat of formation for dimethyldiazetidine. I get far better figures for the estimated dicyclopropyl.
-
Hi hollytara, nice to read you!
To my understanding, the strain energy spreads on four bonds (N-N may reacts differently to it, but I have no data), so with your figures, N-N would retain 137kJ. Accordingly, 1,2-diazetidines were already synthesized, see page numbered 28 in
https://apps.dtic.mil/dtic/tr/fulltext/u2/093924.pdf
The authors of the report didn't investigate this compound more. They boiled the dimethyl at 1atm and 70°C without detonation, but for instance sensitivity to shocks would be no-no.
-
If they made it, they made it. I read about the cyclic hydrazine with one of the N's in a 4 membered ring - it was pretty reactive, and ring opened quite easily.
-
OK, the 3-membered is reactive I know, its maybe different when the nitrogen in the 4-membered ring is connected to another nitrogen atom. The mustard gas type of compounds are different with the ability to cross-link DNA etc.
-
There is this issue of "neighboring group participation" which is what makes the mustards so reactive.
-
Synthesis of 1,2-diazetidines: a thesis exists
http://wrap.warwick.ac.uk/38106/7/WRAP_THESIS_Brown_2011b.pdf
especially page 18 of the pdf, marked page 8, gets 60% yield with
RNNR+BrCCBr -> R-diazetidine-R
where R is isopropyl, giving me hope for cyclopropyl.
Whether ethylene oxide is better?
I've seen no mention of photoaddition there.
Just below,
"These diazetidines were found to be very stable compounds."
but this is on a chemists' scale. Rockets are more demanding.
-
what if we try TMEDA or also TEMED as hypergolic with N2O4
-
Hi sriram, thanks for your interest!
Most amines are hypergolic with tetroxide, tertiary amines better so, hydrazines even better. A rocket engine a very fast ignition by contact, and this is uncommon.
One reason is that accumulation of propellants in the chamber before ignition is a show stopper. An other reason is that hypergolic propellants don't demand a perfectly stable flame at the beginning of the chamber, since they can catch fire downstream occasionally if needed, and this eases the design of the chamber, and can make it smaller.
Additional choice criteria are (1) safety (2) safety and (3) safety. TMEDA with C+N=8 is as volatile and flammable as gasoline, nothing desired. PMDETA is good for this with C+N=11.
https://www.chemicalforums.com/index.php?topic=79637
Diamines mass-produced for polyamines are a good start too. Permethylating improves the liquid range, important.
Though, my goal is not hypergolicity, because nobody wants to keep the toxic and inefficient tetroxide. The future is liquid oxygen. I have interest in amines because they bring 2-3s better performance than the alkane homologues and are easier to produce. Among these, PMDETA is my standard choice for safety, easy production, liquid range, and some performance improvement over RG-1 "kerosene" (it's nearer to Diesel oil). More exotic compounds, including the hydrazines of doubtful harmlessness, are a speculative attempt at higher performance, exceeding the strained but probably sound amines like diazetidylcyclopropane
https://www.chemicalforums.com/index.php?topic=86972.0
From my ramblings, you can trust my estimates for the heat of formation and the performance. The boiling and flash points are reasonable. The melting points are always doubtful. The toxicity of the hydrazines is essentially unknown, most are strongly undesired, but the present thread exists because aminoguanidine was tested as a drug and the volunteers survived. And from the synthesis paths I propose, don't believe a word, I'm not a chemist.
Whether engines and stages will exist with such propellants? They gain 4-7s while methane gains 11s. I'd prefer a safer strained amine, storable and denser than methane, but SpaceX and Blue Origin have already adopted methane.
https://www.chemicalforums.com/index.php?topic=79637.msg337420#msg337420
The other option is to cool the engine with the liquid oxygen rather than the fuel: this enables better fuels and more pumping cycles where ethylene is a trivial choice, but other fuels can be more interesting.
-
Synthesis of 1,2-diazetidines: [...]
RNNR+BrCCBr -> R-diazetidine-R
Whether ethylene oxide is better? [...]
Ahum. No, ethylene oxide doesn't react with both amines at once, quite the opposite.
However, I disagree with the usual "bromide hence expensive". To produce tons, the plant would catch the by-produced HBr and recycle it to produce the dibromoethane reactant, either by anti-Markovnikov addition on acetylene feed, or over Br2 added to a safer ethylene feed.
-
[...] Alkenes can make cis-1,2-cyclobutanes by photoaddition. I haven't seen the equivalent reaction with imines, so the appended tentative syntheses need luck. [...]
Irradiation knowingly enlarges cyclopropylimines to 5-membered aza-enes rings. Whether some conditions favor the cycloaddition?
-
More ramblings... As a synthesis for N,N'-diisopropyl-diazetidine is known, but C+N=10 is only as reluctant to flame as turpentine, I imagine di(cyclopropylaza)bicyclopentane with one cycle more would bring wow performance and raise the flash point. I fully ignore how toxic and explosive this hydrazine is.
- Cis-dihalocyclopropane exposes the halogens simultaneously to the N-H, which may help this step.
- But the N-N bond is more strained than in diazetidine. Ouch.
The recycling of HBr is displayed too, because many people claim that bromine compounds are expensive. Electrolysis produces HOBr which acts even on cyclopropane.
I didn't see a cis-dihalocyclopropane synthesis. If someone knows, please tell! Cyclise tri- or tetrahalopropane with a metal, hoping it stops before cyclopropene?
Marc Schaefer, aka Enthalpy