Doping the Iron out

The use of recyclable catalysts in chemical reactions seems to be a hot research area! I have looked the a recent issue of Chem. Commun. and have read another nice paper about the immobilization of a ferrocene-containing phosphine onto iron nanoparticles, through the use of dopamine as a linker [1]. The corresponding rhodium and palladium complexes of the magnetic ferrocenyl-phosphine ligands were prepared, and they have been shown to be great and recyclable (hence re-usable) catalysts for hydroformylation (Rh) and Heck coupling reactions (Pd) respectively.

Figure 1. Synthesis of the ligand and preparation of the metal complexes.

Let me show you the design of this modular ligand (Figure 1). First, the ligand was a ferrocenyl phosphine known as BPPFA. This phosphine ligand was first conjugated to dopamine. When the dopamine-linked phosphine was sonicated with Fe₃O₄, the phosphine was thus functionalized with magnetic nanoparticles, and the structure was designated as Fe₃O₄@dop-BPPF. These magnetite nanoparticles were formed as a black precipitate, and it was notable that they could be collected and separated from the reaction mixture through the use of a piece of magnet.

When  Fe₃O₄@dop-BPPF was allowed to react with a slight excess of [Rh(nbd)Cl]₂ or [Pd(C₃H₅)Cl]₂, the resulting rhodium or palladium complexes were afforded respectively.

Why did the researchers use dopamine as a linker? They rationalized that the 2 hydroxyl groups on dopamine, which were arranged in a 1,2 relationship, was great as a bidentate ligand and thus could provide tighter binding to the iron oxide, leading to higher stability for the resulting catalyst.

Before submitting to rounds of catalysis, the chemical and magnetic properties of these compounds were first investigated with a number of techniques. Besides NMR and FTIR spectroscopy, transmission electron microscopy, selected area electron diffraction, and X-ray diffraction studies have also been employed to access the appearances of these ligands and metal complexes.

One aspect I find particularly fascinating is the investigations of the magnetic properties of the nanoparticles. It is related to a phenomenon known as ‘magnetic hysteresis’, and the concept of hysteresis is actually important in the field of Nonlinear Dynamics, something I am also highly fascinated in.

The researchers have found that, the dop-BFFP phosphine’s magnetic properties, namely its coercivity (H_c) and remanence (M_r), were not affected by the functionalization with iron nor complexation to rhodium or palladium. While the saturation magnetization value was decreased slightly upon the iron nanoparticle functionalization and complexation with metals, the bulk magnetization has not been affected, which meant that the magnetic properties were retained even if the iron particles were functionalized with the phosphine and complexed to the metal center, and this would benefit the separation process at the end of a catalytic trial through the use of a magnet. Indeed, if we look at the magnetic hysteresis loops of  Fe₃O₄@dop-BPPF, Fe₃O₄@dop-BPPF-Pd and Fe₃O₄@dop-BPPF-Rh at room temperature were almost overlapping with each other, confirming the above arguments.

Figure 2. Hydroformylation catalysis.

Next, the researchers used their newly-prepared metal complexes to do 2 catalytic reactions – Rh-catalyzed hydroformylation and Pd-catalyzed Heck coupling reaction. Not only they wanted to investigate the conversion and substrate scope, a prime aim was to see whether these nanoparticle catalysts could be recycled and recused efficiently.

For the hydroformylation (Figure 2), they have focused much of their studies on the substrate styrene. They have been able to optimize the reaction conditions and the branch to linear selectivity was reasonably high, with the branched product as the predominant. They have also found that styrene substituted with electron withdrawing groups (such as NO₂ and Br) gave a very high branch to linear selectivity of 99:1. Linear alkenes, such as 1-octene, gave the opposite selectivity, with the normal isomer as the predominant product (b : l = 0 : 100).

The hydroformylation catalyst could be recycled and re-used for 3 further times, and then a gradual loss of reactivity was observed, and the researchers attributed to a higher pressure in the reaction system, leading to catalyst leaching.

Figure 3. Heck coupling catalysis.

The Pd catalyst was subject to Heck coupling reaction, with styrene and iodobenzene as the 2 coupling partners. Reaction temperature, choice of base, and solvent system were optimized. When the less reactive bromobenzene was employed instead of iodobenzene, full conversion could be achieved through a longer reaction time, and this signified the stability of the new catalyst.

With the styrene / iodobenzene system, the Pd catalyst could also be recycled and re-used, by simply using an external magnet to separate the catalyst from the reaction mixture. The clean catalyst could be re-used for at least 9 more times, which was very impressive.

Figure 4. Cobalt-phosphine catalyzed hydroarylation - a potential reaction to explore?

This is great research, and the authors are working on other catalytic reactions to explore the scope of their novel ligands. Since they have been using styrene as a substrate, that fact rings a bell on me as I remember reading an article about a cobalt-catalyzed hydroarylation of styrene from J. Am. Chem. Soc. [2]. That could also be an interesting and potentially useful reaction to try, too!

by Ed Law

19/7/2016

Reference:

1. Magnetic nanoparticle-supported ferrocenylphosphine: a reusable catalyst for hydroformylation of alkene and Mizoroki–Heck olefination

M. Nasiruzzaman Shaikh, Md. Abdul Aziz, Aasif Helal, Mohamed Bououdina, Zain H. Yamani and Tae-Jeong Kim

RSC Adv., 2016, 6, 41687–41695

2. Regioselectivity-Switchable Hydroarylation of Styrenes

Ke Gao and Naohiko Yoshikai

J. Am. Chem. Soc., 2011, 133, 400–402

Fatten up a Few Gear

Ever since my undergraduate years in biochemistry, I have developed an intense fascination with the chemistry and biology of membranes. While most elementary textbooks on biology have often portrayed the cell membrane as a pair of lines (or curves in some cases), this organelle is actually one of the most dynamic components of the cell, and the chemistry and molecular biology behind it serve as the driving force for many wonderful molecular phenomena. 

Now, I am working in the field of bi-phasic catalysis, and it is quite gratifying to see that there are parallels between my field and that of membrane science. If possible in the future, membrane chemical biology is certainly a research field I would love to explore! This week, I have read a great article in Journal of American Chemical Society about lipid membrane and hydrogel formation [1], and I would like to share with you here.

Figure 1. Hydrogel network formation through the catalysis of a lipid membrane.

The paper is about the use of a negative-charged lipid membrane as a catalyst to facilitate the formation of a hydrogel network (Figure 1). A hydrogel is defined as a polymeric, gel-like macromolecular structure, which is made by the cross-linking of polymer chains. Hydrogels have been used in drug delivery, tissue engineering, supramolecular catalysis [2c], and many other fields.  The self-assembly and aggregation of the resulting fibrous hydrogel network has to be designed in a way so that the material properties of the resulting gel fiber can be controlled. On the other hand, the spatial position of hydrogel formation has to be carefully defined, to serve the aim of controlled release or delivery of, for example, drug molecules.

The researchers have found that a negatively-charged lipid membrane can be used as a catalyst to form supramolecular hydrogel networks. The negatively-charged liposomes can be used to catalyze the formation of a gelator molecule 3, and they have been able to achieve spatial control – the gelator molecule 3 is formed near the membrane, so that the resulting hydrogel networks can be formed in this well-defined area. The mechanism involves the generation of a local proton (H⁺) gradient - due to the prevalence of the negative charges. The high proton concentration facilitates the acid-catalyzed formation of hydrazone– the functional group in the gelator molecule 3.

For the chemistry, the first stage to form the hydrogel is a reaction between a hydrazide 1 and an aldehyde 2 to form the hydrazone derivative 3. At neutral pH, this reaction is very slow. Yet, the researchers have established that, when an acid or aniline is added, the reaction rate improves a lot. The structure of the hydrazone enables itself to self-assemble and forms a fiber-like structure, and aggregates towards a cross-linked network, resulting in gelation of the surrounding solvent.

The researchers are interested to see whether a negatively-charged lipid membrane can somehow catalyze the formation of hydrazone 3 and also the formation of the final hydrogel network. Their rationale is that the negative charges on the membrane can induce an increase in the proton concentration, leading to a decrease of pH and renders the chemical environment more acidic. They believe this acidic environment may catalyze the formation of the hydrazone stucture and subsequent hydrogelation.

The reaction is carried out in the following way. The hydrazide 1 and the aldehyde 2 are mixed in a defined ratio in a buffered solution at neutral pH. After that, the negative-charged liposome solution is added, and the reaction is carried out at room condition. As observed, the gelator molecule 3 is formed and eventually gelation of solvent occurs, signifying the complete formation of the hydrogel network.

How did the researchers access whether catalysis occurred from the liposome? They employed a parameter known as minimum gelation concentration (MGC). They have first measured a control value – where no liposome is added to 3 at pH 7. When a liposome from a lipid with negative charges, DPPG (Figure 2), is added to the reaction mixture with 1 and 2, a decrease in MGC is evident. Of course, if an acid or aniline is added to the reaction mixture instead, due to their catalytic effects, the MGC should also decrease. Indeed, the researchers have found that the negatively charged liposomes work even better – they lead to lower MGC as compared to the acid / aniline scenarios. As another control experiment, the addition of a positive charged liposome sample does not lead to a decrease of MGC. With the use of an UV-active hydrazide substrate, the researchers can monitor the formation of hydrazone using absorbance spectroscopy for different reaction conditions. Thus, the logical conclusion from these trials is that a negative charged membrane is essential for catalysis to occur.

Figure 2. Structure of DPPG and DOPG.

Yet, there is one important thing to note before we jump to simple conclusions. The researchers have found that, while negatively charged membrane is likely to catalyze the formation of hydrogelator network, not all negative charged membranes can achieve that. The one missing piece in the puzzle is the melting temperature, T_m, of the liposome. When the researchers added the liposome from the negative charged lipid DOPG (Figure 2, T_m -20^oC, a liquid phase membrane at room temperature), to the reaction mixture, no hydrogelation occurs even at high liposome concentration. Yet, when the researchers increase the rigidity of the DOPG membrane through the addition of cholesterol (this should be familiar to anyone doing biochemistry or membrane biology), the DOPG-cholesterol hybrid membrane can then catalyze hydrogelation, suggesting the T_m of this hybrid liposome has increased.

Thus, the researchers have summarized that, in order for the liposome to catalyze hydrogelation, 2 criteria have to be fulfilled:

(1)   a negatively charged membrane surface

(2)   a solid phase at room temperature

The researchers have also established, from oscillatory rheology, that lipid concentration can control the physical properties of the hydrogel network. They also believe that liposomes are serving as nucleation points for the formation of fibrous network. I can think of a similar analogy in the case of cytoskeleton biology, where accessory proteins can serve as nucleation centers for actin polymerization.

By using confocal microscopy and a fluorescent aldehyde substrate, the researchers can also visualize the formation of the hydrogel network. When no liposomes are present, the resulting structure is very slack and un-connected. In the presence of liposomes, by contrast, the resulting hydrogel network becomes well-organized and dense, and the effect is enhanced when the liposome concentration is increased.

A very interesting aspect of the gel fiber formation occurs from the ‘underdog’, DOPG, which does not meet up to potential at the catalytic tests only until cholesterol comes to help. Rather curiously, because the T_m of DOPG is low, that meant the membrane it forms is more fluid than the membrane from DPPG. If we look at the chemical structure of DOPG and DPPG, it may shed light on this observation, and this concept is also familiar to biochemistry students. DOPG contains carbon-carbon double bonds, while DPPG is totally saturated on the carbon chains. The presence of double bonds will provide kinks and prevent a close-packing of the hydrocarbon chains, which by contrast is facile when only saturated chains are present. Thus, the unsaturated DOPG is more fluid than the saturated DPPG, and this also explains why the T_m of DOPG is lower.

DOPG has a higher affinity for the gel fibers. The affinity of the hydrogel fiber for the lipid membrane is related to the phase behavior of the hydrocarbon chains of the lipids, thus, a more fluid membrane should favor this interaction. Thus, DOPG-derived membrane seems to interact better with the hydrogel fiber than DPPG-type membrane.

What I am particularly impressed is that, by carrying out so many control experiments, the researchers draw together all the clues and provide a coherent explanation for the different performance of the DPPG and DOPG in catalysis. They propose that, because the DPPG membrane has less affinity to the hydrogel fiber, so the fiber is not blocking the way for DPPG to effect catalysis on its surface, therefore an efficient catalysis occurs and it goes on and on. In contrast, DOPG-type membrane, which is fluid and ‘loves’ the hydrogel fiber,  interacts with the gel fiber with such a high affinity that it is literally blocking the way for further rounds of catalysis. The researchers also draw analogy to the product blocking phenomena in heterogeneous catalytic systems, and I find this as an impressive explanation!

At the biological side, the researchers have also generated hydrogel fiber formations on HeLa cell systems.

All in all, this is a wonderful paper on membrane chemical biology. I have learnt a number of new techniques from it, and I am particularly impressed by the mechanistic insights, both in terms of catalytic and material, from all the great experiments they have carried out to arrive at the conclusions. Brilliant!

by Ed Law

13/7/2016

Reference:

1. Negatively Charged Lipid Membranes Catalyze Supramolecular Hydrogel Formation

Frank Versluis, Daphne M. van Elsland, Serhii Mytnyk, Dayinta L. Perrier, Fanny Trausel, Jos M. Poolman, Chandan Maity, Vincent A. A. le Sage, Sander I. van Kasteren, Jan H. van Esch and Rienk Eelkema

J. Am. Chem. Soc., 2016, asap, DOI: 10.1021/jacs.6b03853

2. Originally, I plan to talk about a self-emulsifying system which is used in the hydroformylation of lipid substrates. When I read Ref. [1], I decide to talk about that instead. It is interesting to see there are some connections regarding the two topics. Here are the references:

(a) A self-emulsifying catalytic system for the aqueous biphasic hydroformylation of triglycerides

T. Vanbésien, A. Sayede,   E. Monflier and    F. Hapiot 

Catal. Sci. Technol., 2016,6, 3064-3073

DOI: 10.1039/C5CY01758K

(b) Supramolecular Emulsifiers in Biphasic Catalysis: The Substrate

Drives Its Own Transformation

Théodore Vanbésien, Eric Monflier, and Frédéric Hapiot

ACS Catal. 2015, 5, 4288−4292

DOI: 10.1021/acscatal.5b00861

(c) Thermoresponsive Hydrogels in Catalysis

Frédéric Hapiot, Stéphane Menuel, and Eric Monflier

ACS Catal. 2013, 3, 1006−1010

dx.doi.org/10.1021/cs400118c 

'Rhod' Up the Usual Suspects

Figure 1. The allene molecule.

I have just read an advance article from Angew. Chem. Int. Ed., about an atom-economical, rhodium-catalyzed cyclization reaction [1]. The researchers have found that, with the use of a chiral and modified DIOP ligand, they could carry out asymmetric cyclizations, which resulted in the formation of large-sized cyclic esters known as macrolactones. Certainly, this is a hot research area, yet there are a number of reasons why I am so excited about this article, and I would like to share with you here.

Figure 2. The rhodium-catalyzed asymmetric cyclization.

First, the reaction involves the use of the interesting allene functionality (Figure 1 and 2). This 3-carbon component is not only fascinating in terms of its structure, but it has also found a lot of applications in the areas of transition metal catalysis or phosphine / NHC-type organocatalysis. The allene functionality can form complexes with a number of metals – rhodium, palladium and gold are some examples that immediately come up to my mind. In many cases, the outcome of the catalytic reaction will lead to the generation of an olefin structure. In this paper, it is an exocyclic olefin, and it can be used for further functionalizations. Cross-coupling, metathesis, epoxidation or aziridination followed by organocopper / cuprate coupling or nucleophilic substitutions – all these diverse reactions can be used to build up a side-chain at the cyclic structure, and thus infer interesting chemical or biological properties to the final compound.

Figure 3. Cyclic peptide synthesis.

The target compound class, macrolactone, is an extremely biologically-relevant type of molecules, and they have been intensely studied in the field of natural product chemistry and drug discovery. While the rhodium methodology could provide access to numerous macrolactone targets, what fascinated me even more was the fact that the method could be used in the synthesis of cyclic peptides (Figure 3). Indeed, the researchers have shown their method to be tolerant to a number of sensitive functional groups, and they have also employed their methods in the synthesis of some depsipeptide targets. This novel method should enter the arsenal of methods that can ‘close up’ a peptide chain, such as peptide stapling via metathesis, Pd-catalyzed cycloisomerization, or simply an amide bond formation. In this case, the exocyclic olefin I have mentioned can find further use, because a spacer can be attached through that part, and this should facilitate easy separation if the reaction condition calls for it, for example, if the compound is attached to a solid support for automated synthesis. And, I believe a great research direction is to explore whether this method can be carried out in an aqueous or a more bio-friendly solvent system. This catalytic reaction has the benefit that it can be carried out at room temperature, which is ideal for temperature sensitive bio-molecules. Thus, not only we can make useful cyclic peptides from it, we may even employ this method to generate synthetic protein architectures, which can then be further investigated for biological applications.

All in all, this work is just brilliant and the reaction makes my day!

by Ed Law

7/7/2016

Figure 4. Use of sugar allenes in the synthesis of furanose sugar derivatives. Taken from [2].

P.S. Speaking of the use of allene, I have also read another article from Angew. Chem. Int. Ed. [2], which is about the synthesis of a number of furanose sugars via the use of some ‘sugar allenes’, via a (1) Pd-catalyzed hydroalkoylation, (2) Ru-catalyzed ring closing metathesis, and (3) Os-catalyzed dihydroxylation sequence (Figure 4). Also worth a read.

Reference: 

1. Enantioselective Rhodium-Catalyzed Atom-Economical Macrolactonization

Stephanie Ganss and Bernhard Breit

Angew. Chem. Int. Ed., 2016, asap, DOI: 10.1002/anie.201604301

  

2. De Novo Synthesis of Furanose Sugars: Catalytic Asymmetric Synthesis of Apiose and Apiose Containing Oligosaccharides

Mijin Kim, Soyeong Kang and Young Ho Rhee

Angew. Chem. Int. Ed., 2016, asap, DOI: 10.1002/anie.201604199

Fishing out the Precious

In this recently-published paper, the research team has immobilized a rhodium complex on a solid support, and the resulting catalyst has proved to be an efficient hydroformylation catalyst.

The researchers have designed and synthesized a novel catalyst for hydroformylation, which involved the immobilization of air-stable rhodium nanoparticles on a magnetic support, which was in turn functionalized with intensely branched polymer that contained phosphine groups at the terminal. The researchers have shown that the rhodium phosphine catalyst could be recycled for further runs of catalytic reactions, and the rhodium complex would not be decomposed even if the catalyst recovery process was carried out without using an inert atmosphere, and the leaching of rhodium metal was negligible. These attributes all represent some of the golden standards of a good recyclable catalyst, in particular in biphasic systems.

The catalyst is modular in nature. Because the rhodium metal is incorporated into the system, let me show you the synthesis of the ligand architecture. The iron oxide core and the hyper-branched polymeric section – ‘HYP’ (which contains the key glycine residue), was separated by a long spacer. The phosphorus center is attached onto this architecture through the use of Ph₂PH and paraformaldehyde, resulting in the formation of 2 diphenylphosphine moieties on each glycine residue. The rhodium nanoparticles were then immobilized onto the ligand structure. ICP-OES technique and also transmission electron microscopy have been employed to characterize the resulting phosphine-rhodium complex.

Synthesis of the rhodium hydroformylation catalyst.

The catalyst, designated as Fe₃O₄@SiO₂-HYP-N(CH₂PPh₂)₂Rh (I would abbreviate it as ‘FeRh’ from now on), was tested for hydroformylation reactions. A number of alkene substrates were tested, and some of them were natural-occurring terpenes. Since the compound estragole was the most reactive for the hydroformylation, the researchers chose this as the model compound for their further investigations on catalyst stability and re-usability.

Hydroformylation of estragole as a model reaction.

The researchers have found that, when the new Fe/Rh catalyst was used to perform hydroformylation on estragole, the catalyst loading was lower and the conversion was far higher than their last-generation catalyst, which did not possess the hyperbranched polymeric section (HYP). The researchers proposed that the inclusion of the polymeric structure could open up more active sites for the phosphine-rhodium moieties on the catalyst surface, hence the improvements in the catalyst performance.

The recyclability and re-usability were also investigated. Because the iron nanoparticle could be attracted to a permanent magnet, the Fe/Rh catalyst was thus recovered from the reaction mixture by magnetic separation. The catalyst was re-used for 5 more times, using any compromise in terms of activity, conversion or selectivity, suggesting that the catalyst was not ‘dead’ (or decomposed). ICP-OES data also showed no significant rhodium leaching, and the Fe/Rh catalyst was observed to be stable in air.

They have obtained Raman spectra to show that the diphenylphosphine has been grafted more efficiently on the iron-polymeric backbone of the ligand. Yet, they also cautioned that further investigations were needed because the diphenylphosphine-based ligand, while so far proved air-stable, they could not rule out a likely possibly that partial oxidation could take place on the phosphorus center. Indeed, I feel it makes total sense. Because of the reaction methodology (i.e. HPPh₂ / paraformaldehyde), there would result in the formation of a di-phenylphosphine, and the alkyl CH₂ group could render the phosphine more sensitive to atmospheric oxygen, which is not the case for the indefinitely air-stable triphenylphosphine.

One final thing that fascinates me is the synthesis of rhodium nanoparticles, as I never know how it can be done. The rhodium source is the red-colored rhodium(III) chloride hydrate (RhCl₃.xH₂O), for which I have used before to prepare some [Rh(cod)Cl]₂ for other rhodium complexes. The RhCl₃ and tetraoctylammonium bromide (TOAB) were combined in a 2-phase system, and then NaBH₄ solution was added to the reaction mixture. The organic phase, which was black in color, contained the Rh-TOAB nano-particles.

Wonderful work!

by Ed Law

4/7/2016

  

Reference:

1. Support Functionalization with a Phosphine-Containing Hyperbranched Polymer: A Strategy to Enhance Phosphine Grafting and Metal Loading in a Hydroformylation Catalyst

Marco A. S. Garcia, Rodrigo S. Heyder, Kelley C. B. Oliveira, Jean C. S. Costa,

Paola Corio, Elena V. Gusevskaya, Eduardo N. dos Santos, Reinaldo C. Bazito and

Liane M. Rossi, ChemCatChem,2016, 8, 1951 – 1960.