Author Topic: CTH compilation, questions, future directions  (Read 3886 times)

0 Members and 1 Guest are viewing this topic.


  • Guest
CTH compilation, questions, future directions
« on: October 10, 2004, 07:58:00 PM »
Catalytic Transfer Hydrogenation has recently gotten my attention as I think it should many other bees searching for a better way to help themselves and others get to the honey.  It is my intention here to compile what we know about CTH with all of its different possible variables (choice of metal catalyst, hydrogen donor, amine source, solvent mix, temp. etc.) because I believe that there are common threads between all of the possible catalysts.  This is not to say that the best hydrogen donor for one metal will be the same for another, but just that getting a better mental picture will make it possible for us to better understand what tweaking of conditions will likely bee the best direction to go.  Then one hits the lab and comes back with the results. 

So, please, post below your thoughts, knowledge, questions, and whatever else pertinent and helpful to the CTH reduction of imines and other stellar hydrogenations.  I'll go first. ;)


  • Guest
« Reply #1 on: October 10, 2004, 08:40:00 PM »
The elements: Ru, Rh, Pd, Pt, Ni, Zn (any others?)

Pd is known to do CTH in both its elemental and 4+ state (PdCl2). 

Ru is able to use IPA as a hydrogen source for CTH of imines.  IPA is oxidized to acetone. The best rate or rxn was found with a solvent mix of benzene or toluene to IPA being 1.7:1.  More polar solvents decreased the rate.  Less polar solvents with a ratio of 24 equivalents of IPA to imine was best.  The rate was further increased by a small amount of water (0.8%), so best rxn mix was found to be IPA:benzene or toluene:H2O = 1:1.7:0.02. Other optimized rxn conditions: a few hours @ 70 °C and 0.3-1 mol % of catalyst.  This gave 93-98% yields.  Ketimines were also found to react faster than aldimines.  (Chem. Eur. J. 2002, 8, No. 13, p.2955)

Formic acid has been used in many papers as the hydrogen source for Pd CTH rxns.  The hydrogens are donated from either the formyl or the carboxyl position on two molecules of formic acid.  Palladium diformate is considered a key intermediate. (Chem. Commun., 1998, p.1935)

Why then is it that so many have discussed Potassium formate and triethylamine formate for the hydrogen source, as well as ammonium formate who's obvious choice is for the imine formation with ammonia? 

In the case of ammonia as the amine source, might one want to adjust the pH to 4.5 with formic acid after adding the ammonium formate since this is supposedly the "magic pH" for imine formation with ammonia?

Although potassium formate and triethylamine formate (formic acid too?) have supposedly been shown to not require as high of Pd catalyst loading (Barium) for decent yields of amine as compared to ammonium formate, what would bee the explanation for this?

Methanol:H2O 9:1 has been shown to bee a good mix for Pd CTH.  Might we try IPA/H2O since it might act as both solvent and hydrogen donor?  Another thought, I believe that using the minimal amount of solvent in this rxn is the key to not needing such a high catalyst loading.  The proof is in the pudding or should I say honey.

More info and thoughts later......


  • Guest
« Reply #2 on: October 10, 2004, 11:27:00 PM »
This could beecome an interesting post, but please read

Post 382254

(Barium: "A great CTH method", Novel Discourse)
this might help us from just copy the old post.
And by the way, I'm much more interested in eksperimental work whitc good results than theoretikal ekplanation's why something might not  ;)


  • Guest
« Reply #3 on: October 11, 2004, 12:51:00 AM »
See also

Post 531624

(indole_amine: "comprehensive catalytic hydrogenation overview", Methods Discourse)
...  ;)

About NH3: "Due to the high reactivity of partially hydrogenated reaction intermediates, imines or Schiff bases, a conventional hydrogenation process leads to a mixture of primary, secondary, and tertiary amines. Promotion by bases of the reaction medium was found to decrease the formation of secondary and tertiary amines, though the primary amine yield was never complete. The promoting effect of NH3 can come from
  • a thermodynamic influence on the reaction between primary imine and amine, leading to secondary imine and NH3
  • a modification of the electronic properties of the hydrogenating metal; and
  • a poisoning of the surface acid sites which would be mainly responsible for the coupling reaction between primary imine and primary amine to secondary amine."
(Journal of Catalysis 167, 142-152 (1997)



  • Guest
« Reply #4 on: October 24, 2004, 01:08:00 AM »
So, at first it was thought that it would bee best to run a full series of rxns with different catalysts, but then she thought that, really, Pd/C is much cheaper and more common than Rh/Barium or Pt/C for the average chemist so that really isn't the way to go unless Pd/C simply can't do the job excellently, which I don't believe is the case.  Raney nickel was also an option but the experimentor was not familiar with the neccessary catalyst loading or safety issues and also she got the impression that it was inferior anywayz.  Maybe somebody might convince her that Ni would bee worth trying.  Those arguements are welcome...

So, 1st trial:

39 g or .219 moles 'tone
2 equivalents of MeNH4Cl
2 equivalents of triethylamine
4+ equivalents of formic
2 equivalents of NaOH
10 g 10% Pd/C
100 mL MeOH
250 mL IPA
40 mL H2O

She added the 'tone last after the temp had cooled down and let stir.  Then she left to get a drink with some friends and then thought "oh no!  Might the 'tone bee reduced to an alcohol b4 it gets a chance to form the imine?"  But a friend said (in a simplified sense) that Pd would much rather reduce an imine than a 'tone and so she needn't worry.  When she returned an hour or two later, she put the rxn flask and a beaker w/ 300 mL water in a regular microwave, set it on medium, and proceeded to microwave for 1 minute increments ten times.  She did this all because of her excitement after reading that microwave paper mentioned by I-forget-who.  Well, the microwave was definitely not beneficial.  Another ~36 hours until there was no more noticable bubbles (even very small ones) and after work-up, the product oil was a bit on the dark side.  Sunlight was absolutely right, heat is not good for this rxn, and microwaves are not a shorter route to product in this rxn.  Yield 55%.  Oh, and another thing realized is that triethylamine is a bitch to get rid of.  It was used instead of K+ ion because she felt it would help drive imine formation and thus require less methylamine.  Not worth it IMO.

2nd trial:

39 g 'tone
3.4 equivalents of MeNH2 (aq)
1.86 equivalents of MeNH3Cl
5 equivalents of KOH
formic acid until pH=6
325 mL MeOH
160 mL IPA
90 mL water (including aq. MeNH2)
10 g 10% Pd/C (used in the previous rxn, stored wet with water)

Notice both my higher ratio of water to solvent and the use of less overall solvent than Sunlight.  I believe the latter is crucial to using less catalyst.  This time the 'tone was added second, after the aq. MeNH2.  This will not bee done again because she felt it is prolly rough on the 'tone with all the base and acid adding and resultant heat.  The solution was cooled in a bath to a bit below r.t. and then the Pd/C was added and added slowly because the resultant release of CO2 was intense.  The rxn ran for the first 12 hours in a mostly closed hood about 8-10 inches from another reaction that was refluxing in a 150 °C oil bath- dummy.  The flask wasn't all that warm upon checking, but still, it is obvious that heat is not desirable.  Results will bee in soon. 

Improvements:  She thinks a more intelligent way would bee adding everything but 'tone and Pd/C, while cooling  to 0 °C in an ice bath, adding the 'tone, stir for an hour (paranoia), add Pd/C (still in ice bath) and let warm to r.t. over rxn period of ~36-48 hours, basically until absolutely no teeny tiny bubbles are noted.

Is storing Pd/C wet with dH2O the best way???

I, of course, give mad props to Sunlight, for any possible improvement would bee based on his/her work....


  • Guest
possible modifications
« Reply #5 on: October 24, 2004, 11:07:00 AM »
IMO a more intelligent way would bee to prepare your methylammonium formate without any Cl- or K+ ions involved, i.e. bubbling CH3NH2 into MeOH/H2O containing the correct amount HCOOH, or using an aeqous methylamine solution to neutralize a HCOOH/MeOH solution (maybe methanolic methylamine to aequous formic acid, even easier??), then bring back to room temp., add the ketone and catalyst and stir (at moderate speed, but with big stirbar) at 18°C for 48 hours. Use a fumehood (or vent the methylamine gas to the outside with a hose, and place the flask in such a position that it won't be warmed up by anything - placing it on a small mag.stirrer situated on a bookshelf is perfect...  :) ). Nothing else.

And please don't use IPA. It becomes reduced by the Pd/C and solvent proportions change. More percentual water will be present after a while, and this isn't good... just use MeOH. It strangely seems to be the only acceptable alcoholic solvent for this purpose.



  • Guest
Interesting about the IPA.
« Reply #6 on: October 24, 2004, 12:59:00 PM »
Interesting about the IPA.  She was putting it in just incase it might donate protons as is the case with some other CTH versions. 

And yes, methanolic or ethanolic methylamine would bee the choice mainly for simplicity's sake (and beecause the salt supply is dwindling but the gas is plentiful.  Since you say MeOH and not EtOH, ok then, but I'm all for the K+ ions. Half methylamine formate, half potassium formate.


  • Guest
« Reply #7 on: October 24, 2004, 04:41:00 PM »
Indole: Why do you think the potassium and chloride ions disturbs the reaction? And no, IPA is not, to my experience anyway, acting as a hydrogen donor here. I have performed many CTH's using IPA and have never noticed any acetone during the removal of the solvent. Do you base your comments on experience or asre you just guessing?

Dopey: I would suggest to try to add formic acid dropwise to the pregenerated imine. Add 2 eq methylamine to your ketone of choice, add the catalyst and then start to add the formic acid dropwise. Check the reaction progress with TLC.

This way a excess methylamine is always present possibly preventing the formed amine to react with any ketone which may form from the hydrolysis of the imine. Doing it this way I don't think the water present from a ordinary 40% aq methylamine solution will cause any problems. Of you could of course make the methylamine in situ from methylamine chloride and KOH in MeOH to reduce the water content even further. Just a suggestion.


  • Guest
I could imagine that a) potassium or chlorine...
« Reply #8 on: October 24, 2004, 05:45:00 PM »
I could imagine that
a) potassium or chlorine ions *could* affect the catalyst reactivity in some way.. (poisoning or maybe even positive effects?)
b) if the acid is added to the ketone/imine/amine, "pockets" of acid could form due to insufficient mixing or too fast addition rate, resulting in temporary local overacidification. Not favorable. Surely nothing to bother with, but can be easily avoided by premixing amine and acid, mixing well (as Dope_amine said, one hour for paranoid bees  :) ) and then adding the ketone, followed by the catalyst.

About the water content: I don't know if reducing the amount of H2O is favorable; think 9:1 MeOH/H2O is the optimal proportion for this kind of red. amination with HCOOH... And I once read about a CTH using solely IPA as the hydrogen donor; but never about MeOH becoming reduced by Pd/C. And I know that this kind of reductive amination doesn't proceed through dehydrating the hemiaminal C(OH)-NHCH3 to the imine C=N-CH3 followed by hydrogen transfer to give amine C-NHCH3, but rather direct reduction of the hemiaminal to the amine being complexed with Pd and making some kind of hydrolysis necessary. With such amounts of water present, the imine formation is greatly disfavoured, in contrast to the following: the palladium(0) inserts into the formate ion, becomes oxidized to Pd(2) and causes a proton to be given off, which in turn causes dehydration of the hemiaminal and the formation of a H-Pd(II)amine complex with CO2 evolution. In easier words: the Pd destroys the formate moiety, and this initiates the reaction and drives it forward. The imine isn't formed by dehydration through water removal/absence alone but rather through the oxidative insertion of Pd into formate. Which happens favorably under slightly basic aequous conditions, as depicted in the article below... :)

Heterogenous Catalytic Transfer Hydrogenation and Its Relation to Other Methods for Reduction of Organic Compounds
(Robert A.W. Johnstone, Anna H. Wilby, Ian D. Entwistle)
Chemical Reviews 85 (1985), 129-170

EDIT: Of course uploading articles being already present is redundant; my fault, sorry. I have changed the link to the file being already present and have deleted the newer one...



  • Guest
« Reply #9 on: October 25, 2004, 01:38:00 AM »
indole_amine: It is completely unnecessary for you to upload articles already posted here - this one could for example be found in

Post 455063

(Rhodium: "The Discovery of Catalytic Transfer Hydrogenation", Tryptamine Chemistry)


  • Guest
« Reply #10 on: October 25, 2004, 11:04:00 AM »
indole:  Thanx for reminding me about that article.  I need to sit down and read that.  Funny that it came out so long ago. 

Results: %60  :(

BUT, I've come to realize that she was recovering only about 50% of her Pd/C back and that means she was only using 5 (from previous rxn) grams this last time.  Also, when she went to do the work-up, she came inside from a cold night with cold hands and felt the flask which she had moved a bit farther away from that refluxing flask and it was definitely warm  :(  :( .  Again, the oil was a maroon color. 

Next time it will bee done as said in my post above, with an ice bath, fresh Pd/C, alcoholic methylamine- not cuz she's worried abot water but just cuz, 'tone stirred for an hour, then Pd/C, and then let ice bath slowly warm to r.t......


  • Guest
a) potassium or chlorine ions *could* affect...
« Reply #11 on: October 25, 2004, 05:09:00 PM »
a) potassium or chlorine ions *could* affect the catalyst reactivity in some way.. (poisoning or maybe even positive effects?)

Which means that you are guessing.

b) if the acid is added to the ketone/imine/amine, "pockets" of acid could form due to insufficient mixing or too fast addition rate, resulting in temporary local overacidification. Not favorable. Surely nothing to bother with, but can be easily avoided by premixing amine and acid, mixing well (as Dope_amine said, one hour for paranoid bees ) and then adding the ketone, followed by the catalyst.

And then you miss the point of having about 1 eq free methylamine present to supress any side reactions. The reducing agent could be added at about the same rate as it is comsumed. The local over acidification would not likely be any problem with that much methylamine present.

About the IPA: Have you any personal experience of IPA as hydrogen donor with Pd/C? I have, and I can tell you it is a bitch to use as a hydrogen donor. With homogenous catalysts or nickel it is another story.


  • Guest
« Reply #12 on: October 26, 2004, 03:39:00 AM »
And then you miss the point of having about 1 eq free methylamine present to supress any side reactions.

I surely don't miss any point here;  I clearly stated that this order of reactands is not necessary in my opinion. Still overacidification *CAN* happen if the ketone is added before the acid. And this can be avoided. Although I don't consider it being important.

Is there any advantage of adding only small amounts of HCOOH while it is consumed? Do you have any experience with it? Are you just guessing?

Further I never said that IPA would be a suitable hydrogen donor in this case; I just said that due to its lack of inertness, it is not the solvent of choice here. As you said, it is a bad hydrogen donor - but it is one! And it is NOT favourable to conduct this reaction with a proportionally increasing amount of water - believe me or not. I don't even know exactly what happens to the IPA (acetone? Propane?) as I never wanted to know - because I already know that it is both a poor hydrogen donor and a poor solvent in this case.



  • Guest
possible hydrogenation mechanism
« Reply #13 on: October 28, 2004, 05:02:00 PM »
Stoichiometric, Catalytic, and Enantioface-Selective Hydrogenation of C=N Bonds by an Ionic Mechanism
(Magee, M. P.; Norton, J. R.)
J. Am. Chem. Soc., 2001; 123(8); 1778-1779


"The hydrogenation of carbon-carbon double bonds is a classic application of homogeneous catalysis, and numerous complexes are known to catalyze it. Furthermore, the use of chiral ligands to produce enantioface selectivity in such catalytic hydrogenations is now well developed.1 The catalytic cycle for such reactions involves an olefin dihydride complex (with the addition of H2, step A, and the coordination of the olefin, step B, occurring in either order), giving the hydrogenated product by insertion and reductive elimination..."




  • Guest
dimethylammonium borohydride (!) as H2 donor..
« Reply #14 on: October 28, 2004, 05:10:00 PM »
Well, that's a pretty wierd one - the title seemed very fascinating to me... :)

(translated to "hive-conform" syntax, it would be an imine formation with subsequent rhodium-catalyzed borohydride reduction i guess..  ;) )

Ambient Temperature, Tandem Catalytic Dehydrocoupling-Hydrogenation Reactions Using Rh Colloids and Me2NH·BH3 as a Stoichiometric H2 Source
(Jaska, C. A.; Manners, I.)
J. Am. Chem. Soc. 2004; 126(9); 2698-2699


Highly active Rh colloids, generated in situ during the catalytic dehydrocoupling of Me2NH·BH3 using [{Rh(cod)(-Cl)}2] as a precatalyst, are capable of efficiently hydrogenating alkenes at 25 C in a one-pot procedure using only the evolved H2 from the initial dehydrocoupling reaction.

Now what about using RuCl2/methylammonium borohydride for a CTH reductive amination like described above? (just an idea)



  • Guest
green hydrogen ;^)
« Reply #15 on: October 28, 2004, 05:14:00 PM »
Looks like a novel "green chem" approach... ;)

Hydrogen-Rich Gas Production from Biomass Catalytic Gasification
(Lv, P.; Chang, J.; Wang, T.; Fu, Y.; Chen, Y.; Zhu, J.)
Energy & Fuels; (Article); 2004; 18(1); 228-233



  • Guest
novel CTH catalyst: Ni on zeolithes...
« Reply #16 on: October 28, 2004, 05:20:00 PM »
Regio- and Chemoselective Catalytic Transfer Hydrogenation of Aromatic Nitro and Carbonyl as Well as Reductive Cleavage of Azo Compounds over Novel Mesoporous NiMCM-41 Molecular Sieves
(Mohapatra, S. K.; Sonavane, S. U.; Jayaram, R. V.; Selvam, P.)
Org. Lett. 2002; 4(24); 4297-4300

Regio- and chemoselective reduction of nitroarenes and carbonyl compounds and reductive cleavage of azo compounds, including bulkier molecules, was achieved by the catalytic transfer hydrogenation method (CTH) using a novel nickel-containing mesoporous silicate (NiMCM-41) molecular sieve catalyst. In addition, the catalyst was also found to behave as a truly heterogeneous catalyst as the yield was practically unaffected.

(also note that they used IPA as the hydrogen donor)



  • Guest
another novel CTH catalyst: Pd on clay...
« Reply #17 on: October 28, 2004, 05:39:00 PM »
Clay is a pretty dang cool thing... :)

Preparation of Ultrafine Palladium Particles on Cationic and Anionic Clays, Mediated by Oppositely Charged Surfactants: Catalytic Probes in Hydrogenations
(Z. Király,* B. Veisz, Á. Mastalir, and Gy. Köfaragó)
Langmuir, 17 (17), 5381 -5387, 2001 (!)

Finely divided Pd particles (2-3 nm in diameter) were synthesized by the reduction of Pd2+ precursor ions in the presence of cationic (myristyltrimethylammonium bromide, MTA+Br-) or anionic (sodium dodecyl sulfate, Na+DS-) surfactants. The protective adsorption layer of the ionic surfactants around the particles ensured the long-term stability of the aqueous dispersions. When the palladium hydrosol stabilized with MTA+Br- was mixed with an aqueous suspension of sodium montmorillonite, Na+MM- (a cationic clay), the cation-exchange reaction between Na+ and MTA+ rendered the montmorillonite surface hydrophobic, in parallel with the incorporation of the released Pd particles into the MTA+MM- organoclay host. In a similar way, after addition of the palladium hydrosol stabilized with Na+DS- to an aqueous suspension of hydrotalcite nitrate, HT+NO3- (an anionic clay), the ion exchange between NO3- and DS- resulted in the formation of a hydrophobic clay, HT+DS-, with simultaneous deposition of the released Pd particles onto the clay lamellae. The low-loaded, highly dispersed Pd-organoclay materials displayed extremely high catalytic activities under mild conditions in the liquid-phase hydrogenations of styrene, hex-1-ene, and cyclohexene. Furthermore, the catalysts exhibited high selectivities for the partial hydrogenation of 1-phenyl-1-pentyne to 1-phenyl-cis-1-pentene. These high activities and selectivities were explained in terms of the high degree of dispersion of the Pd particles and the hydrophobic nature of the catalysts.



  • Guest
designing stereoselective catalysts...
« Reply #18 on: October 28, 2004, 06:03:00 PM »
Asymmetric Catalytic Hydrogenation. Design of New Ru Catalysts and Chiral Ligands: From Laboratory to Industrial Applications
(Jean-Pierre Genet)
Acc. Chem. Res., 36 (12), 908 -918, 2003

10.1021/ar020152u S0001-4842(02)00152-8

This account covers the design of Ru catalysts and ligands. Two classes of chiral phosphine ligands are prepared: the electron-rich trans-2,4-substituted phosphetanes, readily available from optically pure 1,3-diol cyclic sulfates, and atropoisomeric ligands (SYNPHOS, MeO-NAPhePHOS, bearing heterotopic biaryl moieties, and a chiral water-soluble diguanidinium binaphthyl diphosphine, Digm-BINAP). Applications of these ligands to rhodium- and ruthenium-mediated hydrogenation of ketones and olefins have been reported with high enantioselectivities. The recognition abilities of Ru-SYNPHOS for a wide range of ketones is superior to those observed with BINAP, MeO-NAPhePHOS, and MeO-BIPHEP. Several biologically active compounds have been prepared through dynamic kinetic resolution. This work gives access to a number of highly active catalysts of the type [Ru(biphosphane)(H)(6-cot)]BF4. These catalysts have demonstrated their utility in the enantioselective hydrogenation of the tetrasubstituted cyclopentenone "dehydrodione", which leads to the commercially important perfume component Paradisone.

..probably interesting if any bee would like to prepare their customized stereoselective hydrogenation catalyst complex? (who knows?  :) )



  • Guest
Yea, so
« Reply #19 on: November 13, 2004, 12:01:00 AM »
58.9 g 'tone
3x molar Formic
1.8x molar MeNH3Cl
4x KOH
660 mL MeOH
75 mL H2O
18 g Pd/C (fresh)

3 days stirring in ice water.  Got back 54% in amine, 19% back in 'tone.  She is bummed mainly by the return of 73% total back from the extraction.  The rxn mix was close to a liter in volume.  Pd/C filtered, and 1 L of (100mL HCl +H2O) added, and extraction w/ 100 + 100 mL DCM, followed by basifying to 10->11 w/ 6 M NaOH over the course of extraction with 250 + 200 + 150 + 100 mL DCM.  She, at first considered rotovaping b4 filtration of catalyst, but decided she did not want to heat the rxn mix after having kept it cool for so long.  What she should have done instead was filter the catalyst and then rotovap before adding acid.  Doh!

The product oil was much cleaner without tinting of color. 

If one were to assume 100% extraction recovery of reactants/products at the above mentioned ratio of product:reactant, the product yield would bee 74%.  Not too bad with a 1.8 molar excess of MeNH3Cl.  Definitely could be better though...

She is done with this theoretical laboratory research buzzing in her dreams.