Download PDF
Synlett 2013; 24(7): 804-812
DOI: 10.1055/s-0032-1318483
account
© Georg Thieme Verlag Stuttgart · New York

New Methods in Organic Synthesis Through Copper-Catalyzed Borylation Reactions: Stereoselective Synthesis of 1,4-Diols and Vinylboronates

Ricardo Alfaro
Departamento de Química Orgánica (módulo 01), Universidad Autónoma de Madrid, Cantoblanco, 28049 Madrid, Spain   Fax: +34(91)4973966   Email: mariola.tortosa@uam.es
,
Alejandro Parra
Departamento de Química Orgánica (módulo 01), Universidad Autónoma de Madrid, Cantoblanco, 28049 Madrid, Spain   Fax: +34(91)4973966   Email: mariola.tortosa@uam.es
,
José Alemán
Departamento de Química Orgánica (módulo 01), Universidad Autónoma de Madrid, Cantoblanco, 28049 Madrid, Spain   Fax: +34(91)4973966   Email: mariola.tortosa@uam.es
,
Mariola Tortosa*
Departamento de Química Orgánica (módulo 01), Universidad Autónoma de Madrid, Cantoblanco, 28049 Madrid, Spain   Fax: +34(91)4973966   Email: mariola.tortosa@uam.es
› Author Affiliations
Further Information

Publication History

Received: 22 January 2013

Accepted after revision: 26 February 2013

Publication Date:
15 March 2013 (online)

 
 PDF Download Permissions and Reprints All articles of this category


In memory of Christian G. Claessens

Abstract

Boronic acid derivatives have become important intermediates in organic and medicinal chemistry and, as a result, the ­development of new methods that permit the efficient creation of C–B bonds under mild conditions has become an active field of research in organic synthesis. In the last ten years, copper-catalyzed borylation reactions have emerged as versatile new tools for introducing boron atoms into organic molecules. Here, we describe our recent research in this field. We have developed two copper-­catalyzed borylation reactions. The first involves the SN2′ addition of bis(pinacolato)diboron to allylic epoxides. This reaction permits the stereoselective synthesis of a wide range of compounds containing a 1,4-diol moiety, a fragment that is present in many natural products with important biological activities. The second reaction is a copper-catalyzed formal carboboration process. In this reaction, a C–B and a C–C bond are created in a single catalytic cycle, and the method provides a new tool for the synthesis of tri- and tetrasubstituted vinylboronates.

1 Introduction

2 Stereoselective Synthesis of 1,4-Diols

3 Synthesis of Tri- and Tetrasubstituted Vinylboronates

4 Outlook


#

Key words

boron - esters - copper - catalysis - diols - alkenes - carbo­borations

Biographical Sketches

Zoom Image

Ricardo Alfaro was born in 1974 in Distrito Federal, México. He received his B.S. degree in chemistry from the Universidad Nacional Autónoma de México in 1997. In 2005, he received his master’s degree in organic chemistry from the Universidad Nacional Autónoma de México, and in 2009, he earned his Ph.D. from the same university under the supervision of Professor Francisco Yuste López. He is currently a postdoctoral fellow in Professor García Ruano’s group at the Universidad Autónoma de Madrid, Spain. His research interests concern asymmetric synthesis and organocatalysis.

Zoom Image

Alejandro Parra was born in 1980 in Toledo, Spain. He received his B.S. degree in chemistry from the Universidad Autónoma of Madrid, Spain, in 2004. In 2006, he spent four months in the laboratory of Professor Andrew Myers at ­Harvard University, USA, working on the chemistry of tetracyclines. He earned his Ph.D. at the Universidad Autónoma de Madrid under the supervision of Professor José Luis García Ruano in 2009. He then carried out post-doctoral research at the Institute of Organic Chemistry, RWTH Aachen, Germany, with Professor Magnus Rueping. In 2011, he returned to the Universidad Autónoma de Madrid to join Professor García ­Ruano’s group, and his research interests concern asymmetric synthesis and organocatalysis.

Zoom Image

José Alemán obtained his Ph.D. working on sulfur chemistry under the supervision of Professor Jose Luis García Ruano in 2005. In 2003, he spent six months in the laboratory of Professor Albert Padwa at Emory University in Atlanta, USA. He then carried out post-doctoral research at the Center for Catalysis at Aarhus, Denmark, with Professor Karl A. Jørgensen. He returned to Spain in 2009, and is currently a Ramón y Cajal researcher at the Universidad Autónoma de Madrid, Spain. His research interests include asymmetric synthesis and catalysis.

Zoom Image

Mariola Tortosa obtained her B.S. in Chemistry from the Universidad Autónoma de Madrid in 1999. She then joined the group of Dr. R. Fernández de la Pradilla at the Instituto de Química Orgánica General, CSIC, Madrid, Spain. After obtaining her Ph.D. in 2005, she moved to The Scripps Research Institute in Florida, USA, to work as a postdoctoral fellow with Professor William R. Roush. Her research in Florida was directed toward the completion of the total synthesis of the antitumor agent superstolide A. In 2011, she moved to the Universidad Autónoma de Madrid, Spain, as a Ramón y Cajal fellow. Her research interests include boron chemistry, catalysis, and the synthesis of natural products.

1

Introduction

Boronic acid derivatives are versatile synthetic intermediates, useful in the preparation of a wide range of organic molecules.[ 1 ] Their unique properties as intermediates in coupling reactions and the possibility of transforming the C–B bond into a C–O or C–N bond makes boronic acids particularly attractive as targets for synthesis. The potential uses of these compounds as saccharide sensors and in boron neutron-capture therapy,[ 2 ] together with the recent approval of the anti-cancer agent bortezomib (Velcade), highlight the importance of boronic acids, not only in synthetic chemistry, but also in medicinal chemistry. Therefore, the development of new methods for creating C–B bonds efficiently, inexpensively, and in an environmentally friendly way is an important challenge in organic synthesis.

The alkylation of trialkylborates or haloborates with organomagnesium or organolithium reagents is an established method for the preparation of boronic acids, especially those containing aryl substituents.[ 1 ] Although this approach has been widely used, it is limited by the range of functional groups that are compatible with the organometallic species. In this context, the development of transition metal-catalyzed cross-coupling reactions of B–B or B–H reagents with organic halides or triflates has permitted the preparation of boronic acids containing a variety of functional groups.[ 3 ] Undoubtedly, these methods represent a pivotal advance in the synthesis of boronic acid derivatives, and they are largely responsible for the growing importance of this group of compounds. In particular, copper-catalyzed borylation reactions have emerged as powerful new tools for the construction of C–B bonds. In 2000, Miyaura and co-workers described the first copper-catalyzed addition of bis(pinacolato)diboron to α,β-unsaturated carbonyl compounds (Scheme [1], equation 1).[ 4 ] The authors proposed that copper–boryl species act as reactive intermediates that might be formed by equilibrium dissociation of boron–ate complexes (Scheme [1], equation 2). At about the same time, Hosomi and co-workers reported the β-borylation of α,β-unsaturated ketones by using catalytic amounts (10 mol%) of a copper(I) salt and tributylphosphine in the absence of a base (Scheme [1], equation 3).[ 5 ] Despite their limitations, these two seminal works have had a marked influence on the development of copper-catalyzed borylation reactions. Importantly, both studies suggested that copper–­boryl complexes generated in situ can react as nucleophilic boryl synthons.[ 6 ]

Zoom Image
Scheme 1

Another significant advance in the field was reported in 2005 by Ito and co-workers, when they published the first account of an SN2′ addition of a copper–boryl complex to an allylic carbonate in the presence of a catalytic amount of copper(I) tert-butoxide and Xantphos [4,5-bis(diphenyl-phosphino)-9,9-dimethylxanthene] as a ligand (Scheme [2]).[ 7 ] These results reinforced the idea that copper–boryl species can undergo reactions characteristic of copper compounds. They also proved that chiral phosphines can be used to perform enantioselective copper-catalyzed borylations, starting from achiral carbonates.[ 8 ]

Zoom Image
Scheme 2

Another major advance in the area of copper-catalyzed borylations was made in 2006 by Yun and co-workers,[ 9 ] who found that addition of alcohol additives dramatically accelerates the copper-catalyzed borylation of a variety of α,β-unsaturated carbonyl compounds (Scheme [3]). They postulated that the alcohol protonates the alkyl copper intermediate formed after the addition, generating the copper alkoxide that is needed to continue the catalytic cycle. These conditions permit the β-borylation, not only of enones, but also of unsaturated esters and nitriles. Lee and Yun also showed that an enantioselective β-borylation can be achieved by using chiral phosphine ligands.[ 10 ]

Zoom Image
Scheme 3

Subsequently, many other groups have contributed to the development of this emerging field. Copper–boryl complexes have been used in reactions with aldehydes,[ 11 ] imines,[ 12 ] and allylic[ 13 ] or propargylic substrates,[ 14 ] as well as with alkyl halides,[ 15 ] activated alkenes,[ 16 ] and alkynes[ 17 ] among others. Additionally, theoretical studies have improved the understanding of the mechanisms of these transformations.[ 18 ]

Our goal was to contribute to this field by the development of new copper-catalyzed borylation reactions, and we were inspired by some unsolved problems in the total synthesis of complex molecules. During the last two years, we have found that copper-catalyzed borylations are valuable tools for the stereoselective synthesis of 1,4-diols[ 19 ] and for the preparation of cis-methyl-branched alkenes,[ 20 ] two fragments that are present in many biologically active compounds.


# 2

Stereoselective Synthesis of 1,4-Diols

The first of our copper-catalyzed borylation projects was inspired by the 1,4-diol subunits present in a number of biologically active natural products (Figure [1]).

Zoom Image
Figure 1

Most efforts made to synthesize the 1,4-diol fragment have focused on the preparation of symmetric diols.[ 21 ] These methods are important for the design of new ligands in asymmetric catalysis, but are very difficult to apply in total syntheses of complex molecules. Olefination reactions,[ 22 ] asymmetric reductions of γ-hydroxy ketones,[ 23 ] epoxidation of β-hydroxy allylsilanes,[ 24 ] and additions of 1-alkyn-3-ols to aldehydes[ 25 ] are among the most important ways of preparing nonsymmetrical, enantiomerically pure 1,4-diols (Scheme [4]). In most cases, two different chiral sources are needed to install the two stereocenters in the 1,4-diol subunit. A logical method that might be used by nature to synthesize the 1,4-diol fragment would be the hydrolysis of a vinyl epoxide. Taking our inspiration from this possible approach, we hypothesized that copper-catalyzed SN2′ addition of diboronates to allylic epoxides might provide a potentially powerful transformation for the synthesis of 1,4-diols via the corresponding 1,4-hydroxyboronates (Scheme [4]). Essentially, this method would constitute a formal stereocontrolled hydrolysis of the vinyl epoxide. This approach seemed particularly attractive because it might permit the synthesis of either syn- or anti-diols by appropriate choices of the geometries of the double bond and the oxirane. Additionally, it should be possible to synthesize primary, secondary, or tertiary diols, as well as to achieve selective introduction of orthogonal protecting groups onto the alcohols.

Zoom Image
Scheme 4

We began by examining the reactions of the racemic allylic epoxide 1 with bis(pinacolato)diboron (B2pin2) in the presence of catalytic amounts of a ligand, copper(I) chloride, and sodium tert-butoxide. Because of the instability of the 1,4-hydroxyboronate 3, we encountered some difficulties in finding the right conditions to carry out the reaction. All our efforts to isolate this intermediate were unsuccessful. However, oxidation of the C–B bond in situ gave the anti-diol 2. Appropriate choices of the ligand and the reaction temperature played key roles in the diastereoselectivity of this reaction. After several experiments, we found that the use of copper(I) chloride (10 mol%), Xantphos (10 mol%), and sodium tert-butoxide (30 mol%) in tetrahydrofuran at –20 ºC, followed by oxidation, gave the anti-diol 2 in high yield and with high diastereoselectivity. The use of a mild base such as potassium bicarbonate was also important, because standard oxidation conditions (sodium hydroxide/hydrogen peroxide or sodium per­borate) did not afford good yields of the 1,4-diols.

Having determined the optimal conditions for the reaction, we were able to synthesize a number of acyclic or cyclic primary, secondary, or tertiary diols in either the syn-form (compounds 5, 8, 9, and 11) or the anti-form (compounds 2, 7, 13, and 14), simply by choosing the appropriate geometry for the epoxide and the double bond (Scheme [5]). In most cases, the 1,4-diols were obtained with excellent yields and diastereoselectivities.

Zoom Image
Scheme 5
Zoom Image
Scheme 6

The observed stereochemical outcome can be explained by an anti-attack of the copper–boryl intermediate on the allylic epoxide in an s-trans conformation (Scheme [6]).[ 26 ] Therefore, anti 1,4-diols can be prepared from trans-E or cis-Z allylic epoxides and syn diols can be prepared from the corresponding cis-E or trans-Z isomers.

A possible mechanism for the copper(I)-catalyzed borylation of allylic epoxides is shown in Scheme [7]. Bis(pinacolato)diboron reacts with copper(I) tert-butoxide through a σ-bond metathesis to form a copper–boryl complex. Formation of a copper–alkene π-complex then occurs, followed by addition of the Cu–B bond across the alkene to give a β-borylalkyl copper intermediate. Elimination, with cleavage of the epoxide ring, gives a copper alkoxide that regenerates the catalyst by reaction with bis(pinacolato)diboron.

Zoom Image
Scheme 7

Despite the positive results that we obtained, we were disappointed by our inability to isolate the 1,4-hydroxyboronate intermediate. We reasoned that protection of the hydroxyl group in situ before oxidation of the C–B bond might increase the stability of the resulting compounds by eliminating the possibility of intermolecular nucleophilic attack of the hydroxyl group on the boron atom and might permit isolation of the intermediates (Scheme [8]; X = Bpin). Addition of triethylsilyl chloride and imidazole after the allylic epoxide had been consumed afforded a series of anti- and syn-1,4-silyloxyboronates 15, 17, and 18 in good yields and high diastereoselectivities. We were delighted to find that, indeed, the protected 1,4-hydroxyboronates were very stable. They could be purified by chromatography on silica gel and stored for several months in a freezer without any observable decomposition. The 1,4-hydroxyboronates are valuable intermediates that can be used in subsequent diastereoselective transformations, such as the conversion of the C–B bond into a C–N group to give 1,4-amino alcohols,[ 27 ] in the allylation of aldehydes and imines,[ 28 ] or in homologation reactions to give 1,5-diols.[ 29 ] Furthermore, we also carried out one-pot copper(I)-catalyzed addition–protection–oxidation sequences to give a series of orthogonally protected 1,4-diols 16, 19, 20 (Scheme [8]; X = OH). We believe that this one-pot process might be useful in total syntheses of complex molecules, where the manipulation of protecting groups is frequently a challenge.

Zoom Image
Scheme 8

# 3

Synthesis of Tri- and Tetrasubstituted Vinylboronates

The aim of our second borylation projects was to synthesize cis-branched methylated fragments, which are widely found in natural products and biologically active compounds (Figure [2]).

Zoom Image
Figure 2

The zirconium-catalyzed methylalumination reaction was the established method of choice for the synthesis of the cis-methylated alkene fragment.[ 30 ] Although this method is powerful, the vinylalane intermediates are sensitive to oxygen and moisture, and they need to be used immediately upon preparation. cis-Methyl branched vinylboronates might be useful alternative intermediates for preparing compounds such as those shown in Figure [2], but there is a paucity of methods available for the stereoselective synthesis of cis-branched methylated alkenylboronates. Methylalumination followed by transmetalation with a trialkoxyborane (Scheme [9], equation 1)[ 31 ] or platinum-catalyzed dehydrogenative borylation of propargylic or homopropargylic alcohols (Scheme [9], equation 2)[ 32 ] are two representative examples of methods for the synthesis of cis-branched methylated alkenylboronates.

Zoom Image
Scheme 9

Methods in which a C–B bond and a C–Me bond are created in a single catalytic cycle are especially rare. To the best of our knowledge, there is only one example, described by Suginome,[ 33 ] which involves the use of a homopropargylic or propargylic alcohols derivative, a palladium catalyst, and a methylzirconium reagent (Scheme [9], equation 3).[ 33 ]

We reasoned that a copper–boryl complex might be used to synthesize cis-branched methylated vinylboronates. Our idea was supported by previous reports of a copper-catalyzed hydroboration of alkynes. In 2008, Yun and co-workers described a copper-catalyzed hydroboration of α,β-acetylenic esters, in which Xantphos was used as the ligand.[ 17a ] The stereoselectivity of the reaction was very high, and the product of syn addition to the triple bond was obtained almost exclusively (Scheme [10]). Recently, the same research group described a copper-catalyzed hydroboration of internal arylacetylenes. In this case, it was necessary to use tri(4-tolyl)phosphine to obtain good conversions and high stereoselectivities.[ 17b ] Unfortunately, simple alkyl acetylenes did not react under these conditions.

Zoom Image
Scheme 10

In 2011, Hoveyda and co-workers reported the hydroboration of alkyl- and aryl-substituted terminal alkynes catalyzed by N-heterocyclic carbene (NHC) complexes of copper (Scheme [11]).[ 17c ] The structural and electronic properties of the NHC ligands play key roles in the regioselectivity of the addition. High α-selectivities were observed with mesityl groups on the NHC ligand, whereas adamantyl substituents gave the corresponding β-boryl alkenes.

Zoom Image
Scheme 11
Zoom Image
Scheme 12

Recently, Sawamura and co-workers described the copper-catalyzed monoborylation of 1,3-enynes (Scheme [12]).[ 17d ] Interestingly, in the case of 1,4-disubstituted enynes (type A), the regioselectivity was controlled by the ligand. The use of triphenylphosphine resulted exclusively in addition to the triple bond, whereas borylation of the alkene moiety was observed with Xantphos. In contrast, with highly substituted enynes (type B), borylation of the triple bond occurred exclusively with either ligand.

The proposed mechanism for these transformations involves the addition of a copper–boryl complex to the alkyne to form a vinylcopper intermediate (Scheme [13]). This intermediate reacts with methanol to give the corresponding vinylboronate and copper methoxide; the latter reacts with bis(pinacolato)diboron to regenerate a copper–boryl complex. Although vinylcopper species are proposed as intermediates, their subsequent reaction with electrophiles other than protons had not been reported at the outset of our project. On the basis of this catalytic cycle, we reasoned that a methylboration reaction would require the presence of iodomethane rather than methanol and the use of at least one equivalent of sodium tert-butoxide instead of the 0.2–0.3 equivalents that are usually required for the hydroboration. Under these conditions, the copper(I) iodide formed by reaction of the vinylcopper intermediate with iodomethane would be transformed into the catalytically active copper tert-butoxide.

Zoom Image
Scheme 13

We began by examining the methylboration of phenylacetylene 21 with bis(pinacolato)diboron to give the vinylborane 22 (Scheme [14]). After examining several variations, we found that the use of copper(I) chloride (10 mol%), Xantphos (10 mol%), iodomethane (4 equivalents), and sodium tert-butoxide (1.1 equivalents) gave the desired compound in a good yield as a single regio- and stereoisomer. The use of four equivalents of iodomethane was necessary to avoid the formation of the hydroboration product. The yield decreased to 26% when we reduced the amount of sodium tert-butoxide to 0.6 from 1.1 equivalents, confirming our initial hypothesis. Additionally, the catalyst loading could be reduced to 5 mol% without affecting the yield.

Zoom Image
Scheme 14

Next, we studied the scope of the reaction with various arylalkynes to give a range of vinylboranes 23–35 (Scheme [14]). Alkyl and methoxy groups, halo groups, base-sensitive substituents, and heterocycles were all compatible with the carboboration conditions described above. Alkynes with strong electron-donating groups proved to be less reactive and needed longer reaction times (as, for example, in the cases of compounds 29 and 30). Compounds with strong electron-withdrawing groups on the aromatic ring did not react, presumably as a result of deprotonation of the alkyne. Nonterminal alkynes required further optimization of the reaction conditions. Whereas the use of Xantphos at room temperature resulted in very low conversions, the use of tri(4-tolyl)phosphine at 65 ºC gave the corresponding carboboration products 33–35. Importantly, in each case a single stereoisomer was obtained as a result of the syn addition of the methyl and the boron groups.

Simple alkyl-terminated alkynes did not react under our standard carboboration conditions. However, propargylic ethers such as 36 and the tertiary alcohol derivative 38 gave good yields of the desired compounds 39 and 41, respectively (Scheme [15]). We were also pleased to find that acetal 37 gave the protected aldehyde 40, a highly versatile intermediate.

Zoom Image
Scheme 15
Zoom Image
Scheme 16

We were also interested in studying the carboboration of enynes. As mentioned above, Ito and co-workers have published a report on the related copper-catalyzed hydroboration of enynes.[ 17d ] Interestingly, they found that monoborylation of 1,3-enynes with substituents in positions 1 or 2 was highly dependent on the ligand (Scheme [16]). Borylation of enyne 42 with Xantphos gave alkynylboronate 43 with excellent regioselectivity, whereas in the presence of triphenylphosphine, dienylboronate 44 was obtained exclusively. On the other hand, the 1,3-enyne 45 gave boronate 46, regardless of the phosphine that was used. We tested our carboboration conditions with 1-substituted enynes 47a and 47b in the hope of achieving regio- and stereoselectivities similar to those found for 42. However, we obtained some surprising results. The use of either Xantphos or triphenylphosphine gave a mixture of regioisomers resulting from carboboration of the triple bond. In no case did we observe borylation of the double bond. These results suggest that the mechanism of the carboboration is significantly different from the generally accepted mechanism for copper-catalyzed hydroboration. The regioselectivity was improved in the case of 2-substituted enynes. Under carboboration conditions, compounds 48c and 48d were obtained, each as a single regioisomer. Additionally, 2,4-disubstituted enyne 47e ­afforded tetrasubstituted vinylboronate 48e along with a small amount of the regioisomer 49e.

At this point, we became intrigued by the possibility of using different electrophiles. Unfortunately, we found that terminal alkynes did not react with other electrophiles (3-bromoprop-1-ene, benzyl bromide, or benzoyl chloride). In some cases, we isolated the product of alkylation at the terminal position of the alkyne, which was never observed with iodomethane. However, we obtained better results with nonterminal alkynes (Scheme [17]). To date, we have been able to introduce benzyl and allyl groups in the position cis to the boryl moiety to give products 50–52. The result obtained with compound 52 was surprising because the same alkyne precursor (R1 = Me) gave a single regioisomer when iodomethane was used as the electrophile. This reinforces the idea that there are mechanistic differences between the hydroboration and carboboration processes that need to be elucidated.

Zoom Image
Scheme 17
Zoom Image
Scheme 18

The vinylboronates that we obtained are useful intermediates for the synthesis of functionalized alkenes through Suzuki cross-coupling reactions. We found that, conveniently, the carboboration–Suzuki sequence can be carried out as a one-pot process (Scheme [18]), thereby permitting the synthesis of trisubstituted alkenes such as 53–55 without isolation of the vinylboronate intermediate.

Within the duration of our project, three reports were published describing reactions of vinylcopper species with electrophiles other than protons. Yoshida et al. reported the copper-catalyzed diboration (Scheme [19], equation 1)[ 17e ] and borylstannylation[ 34 ] of alkynes. More recently, Hou and co-workers[ 35 ] found that vinylcopper species could be trapped with carbon dioxide (Scheme [19], equation 2). These three reports, taken in conjunction with our study, prove that vinylcopper species formed in copper-catalyzed borylation reactions can be trapped by a range of electrophiles. Other classes of electrophiles might provide access to boronic acid derivatives not readily accessible by other routes.

Zoom Image
Scheme 19

# 4

Outlook

In the last ten years, copper-catalyzed borylation reactions have become increasingly important in synthetic organic chemistry. Numerous research groups around the world have contributed to establishing these reactions as new tools for introducing C–B bonds into organic molecules. As discussed in this account, our contributions to this area have focused on two different borylation reactions. We have found that copper–boryl complexes are valuable tools for the stereoselective synthesis of 1,4-diols and for the preparation of cis-methylated branched alkenes; these two fragments are present in many biologically active compounds. The carboboration process is especially interesting because a C–B and a C–Me bond are created in a single catalytic cycle. This transformation has raised some questions that require further investigation, particularly those related to the mechanism of the reaction and the use of various electrophiles. In this context, we predict that additional studies will be performed that will include the extension of the copper-catalyzed carboboration to less-reactive, nonactivated alkenes. New results should stimulate further advances that will facilitate the synthesis of boronic acid derivatives and expand their potential as synthetic tools.


#
#

Acknowledgment

The authors are indebted to Professor García Ruano for his generous support and guidance. Financial support from Spanish Government (CTQ-2009-12168) and CAM (programa AVANCAT CS2009/PPQ-1634) is gratefully acknowledged. J.A. and M.T. thank the MICINN for Ramon y Cajal contracts, R.A. thanks the Consejo Nacional de Ciencia y Tecnología de México for a postdoctoral fellowship.



   Permissions and Reprints All articles of this category
Zoom Image
Zoom Image
Zoom Image
Zoom Image
Zoom Image
Scheme 1
Zoom Image
Scheme 2
Zoom Image
Scheme 3
Zoom Image
Figure 1
Zoom Image
Scheme 4
Zoom Image
Scheme 5
Zoom Image
Scheme 6
Zoom Image
Scheme 7
Zoom Image
Scheme 8
Zoom Image
Figure 2
Zoom Image
Scheme 9
Zoom Image
Scheme 10
Zoom Image
Scheme 11
Zoom Image
Scheme 12
Zoom Image
Scheme 13
Zoom Image
Scheme 14
Zoom Image
Scheme 15
Zoom Image
Scheme 16
Zoom Image
Scheme 17
Zoom Image
Scheme 18
Zoom Image
Scheme 19