催化学报  2016, Vol. 37 Issue (10): 1756-1763   PDF    
扩展功能
加入收藏夹
复制引文信息
加入引用管理器
Email Alert
RSS
本文作者相关文章
Larisha Cisneros
Pedro Serna
Avelino Corma
Selective reductive coupling of nitro aliphatic compounds with aldehydes in hydrogen using gold catalyst
Larisha Cisnerosa, Pedro Sernab, Avelino Cormaa     
a. Instituto de Tecnología Química, Universitat Politècnica de València-Consejo Superior de Investigaciones Científicas, Avda. de los Naranjos s/n, 46022 Valencia, Spain ;
b. Corporate Strategic Research, ExxonMobil Research & Engineering Co. Inc., 1545 Route 22 East, Annandale, New Jersey 08801, United States
Foundation Item: This work was supported by Consolider-Ingenio 2010 (project MULTICAT), Generalitat Valenciana (Prometeo program) and Severo Ochoa Program (SEV-2012-0267) and the CONACyT Program (3115621)
* Corresponding author. Avelino Corma, Tel: +34-96-3877800; Fax: +34-96-3879444; E-mail: acorma@itq.upv.es
Abstract: Nitrones were synthesized in good yields directly from nitro aliphatic compounds, aldehydes, and H2 using highly dispersed gold nanoparticles on titania. The high selectivity for nitrone synthesis contrasts with the platinum supported on carbon and corresponds to an increase from roughly 50% to 90%. The catalytic performance is tuned by precise control of the struc-ture of the active sites, the characteristics of the support and reaction conditions.
© 2016, Dalian Institute of Chemical Physics, Chinese Academy of Sciences.
Published by Elsevier B.V. All rights reserved.
Key words: Gold catalyst     Nitro compound     Cascade-type reaction     Hydrogenation     Nitrone    
金催化剂上脂肪族硝基化合物与醛在氢气中选择还原耦合
Larisha Cisnerosa, Pedro Sernab, Avelino Cormaa     
a. 瓦伦西亚理工大学-西班牙国家研究委员会, 化学工程研究所, 瓦伦西亚 46022, 西班牙 ;
b. 埃克森美孚研究与工程有限公司企业战略研究, 新泽西 08801, 美国
摘要:采用TiO2上高度分散的金纳米粒子做催化剂,从脂肪族硝基化合物、醛和氢气可高选择性合成硝酮.与碳负载Pt催化剂相比,该催化体系上硝酮的选择性从50%增加到90%.其催化性能可有活性位结构、载体特性和反应条件精确调节.
关键词金催化剂     硝基化合物     串联反应     加氢     硝酮    

1 Introduction

Nitrones are organic molecules that found wide applications for the synthesis of important organic molecules such as isoxazolidines, isoxazolines and β-amino alcohols [1] and, in general as precursors for some nitrogen-containing derivatives.Furthermore, their radical trapping abilities have found applications as antioxidants [2], spin-traps [3], and enzymes inhibitors[4].Their reactant abilities have shown to have anti-cancer activity in several experimental cancers models and potential therapeutics in some cancers [5-7].

The most common methods for the synthesis of nitrones involve the synthesis of hydroxylamines or oximes which are then reacted with aldehydes or N-alkylation respectively.The first step of the process named above requires the synthesis of hydroxylamines or oximes by partial hydrogenation with reducing agents.Though Zn dust can act as reducing agent in stoichiometric amounts for producing the hydroxylamine intermediate in the presence of acetic acid or ammonium chlorides, the use of solid acids is desired.Nevertheless, one should also take into account that hydroxylamines are unstable under storage and potentially explosive, and consequently the number of hydroxylamines commercially used is very limited (Scheme 1(a)).

Scheme1. Somerepresentative routes for the synthesis of nitrones.

Oximes can be synthesized by hydrogenation of α, β-unsaturated nitro compounds, with a chemoselective catalyst that avoids the reduction of the C=C bond [8].Nevertheless, the synthesis of nitrones from oximes requires several steps since α, β-unsaturated nitro compounds should be first prepared from nitroalkanes and an aldehyde or ketone, followed by hydrogenation of a conjugated nitro group and N-alkylation of the resultant oxime (Scheme 1(b)).There is a third route, able to produce nitrones by oxidation of imines or secondary amines which also involve a multi-step process, most of the times starting with the reduction of a nitro compound (Scheme 1(c)).

It should be noticed that complexity of the preceding routes increases when the desired nitrones involve groups sensitive to oxidation or reduction, such as aldehydes, alcohols, ciano, olefinic or acetylenic bonds.In those circumstances chemoselectivity has been often achieved in the past by using stoichiometric oxidation or reduction agents, with the corresponding generation of residues.

An efficient way to produce nitrones could be envisaged by considering a cascade-type reaction whereby hydroxylamine generated in situ by partial hydrogenation of the nitro group would be rapidly reacted with a carbonyl compound.It is clear that if one could find a selective catalyst able to produce the above cascade reaction, the new process would avoid problems associated to hydroxylamine storage and will reduce the number of steps.Notice that such catalyst should prevent: (a) a complete reduction of the nitro group into the corresponding amine, (b) hydrogenation of the carbonyl group from the aldehyde, and (c) the degradation of the nitrone via secondary reaction paths (Scheme 2).For doing that, we found that there was one route of hydrogenation of nitrobenzene and substitued nitrobenzenes that goes through the hydrogenation to nitrosobenzene, followed by formation of the corresponding hydroxylamine that can be further hydrogenated to the amine.

Scheme2. Typical reactions in mixtures of H2, nitro and carbonyl compounds.

Moreover, it was also observed that depending on the nature of the support it was possible to stabilize in a higher extend the concentration of the intermediate nitroso compound and hydroxylamine on the catalyst surface [9].If this is so, one could then optimize a catalyst that rapidly reacts the intermediate nitroso compound species with other molecules in a cascade mode to give products other than amines or amino derived products when starting from nitrobenzene.For instance, one could react the intermediates nitrosobenzene to produce the azoxy and azo compounds directly or by reacting with the corresponding hydroxylamine (Scheme 3).

Scheme3. Proposed reaction pathways for the reduction of an aromatic nitro compound to the corresponding aniline [10, 11].

The objective can be achieved by means of an Au/CeO2 catalyst; that gives high conversion and selectivity to azo compounds, starting from substituted nitro aromatics [9].However, if one could react the nitrosobenzene and the corresponding hydroxylamine with an aldehyde, it should be possible to obtain the corresponding nitrone (Scheme 4).

Scheme4. Reductive coupling of nitrobenzene and aldehyde.

Nevertheless, to achieve the nitrone in high yields, the consecutive hydrogenation of the nitrone to the imine and secondary amine should be avoided (Scheme 5).

Scheme5. Consecutive hydrogenation pathway for the reductive coupling of nitrobenzene and aldehyde.

Recently, we showed that a carbon-decorated platinum catalyst was able to synthesize aromatic nitrones in high yields provided by a precise control of the structure of the active sites and the characteristics of the support.Particularly, platinum on carbon is a suitable catalyst due to: (a) it is arranged in the form of nanoparticles that lack metal sites located in cristal terraces accessible to the reactants, (b) the support is a high surface area mesoporous active carbon, and (c) the nitro compound has an aromatic character.Due to this, good yields to nitrones were obtained with the Pt/C catalyst with substituted nitroaromatic compounds and substituted aromatic aldehydes.The concept could also be extended to other metals such as Pd and Ru.We reported, however, that the selectivity was only moderate when the reaction involved a nitroalkane or nitro cycloalkanes as a starting reactant, mainly because then a parallel (undesired) reduction of the aldehyde compound occurred [12].Nevertheless, there is an interest in the design of a selective, efficient and eco-friendly catalyst for the preparation of this type of products not only to expand the scope of synthetic possibilities, but also because some relevant families of nitrones, derive from non-aromatic nitro compounds, are particularly relevant as potential therapeutic molecules for some cancers (e.g.PBN).

Since Au/TiO2 was able to perform the hydrogenation of nitro aromatic compounds in the presence of aldehydes and ketones into the amines through a route that involve the hydroxylamine as intermediate, we thought that, conveniently tuned, Au/TiO2 catalyst could be an adequate catalyst for producing nitrones starting from nitro compounds, aldehydes and H2.Herein we have explored the possibilities of gold catalysts as an alternative for one pot synthesis of the most reluctant aliphatic nitrones.

2 Experimental
2.1 Catalyst preparation

Precipitation of gold nanoparticles on TiO2 (P-25, Degussa) was done from an aqueous solution of HAuCl4 that was neutralized with NaOH (0.2 mol/L) to have a final pH of 7 [13-15].TiO2 and the HAuCl4 solution were contacted at room temperature and the resultant slurry was vigorously stirred for 5 h.The pH of this slurry was maintained at 7 during the process.The catalyst was then filtered and washed thoroughly to remove chlorides from the catalyst surface.After drying the solid powder for 12 h at 373 K, the sample was calcined in a fix bed reactor with air flowing (50 mL/min) at 673 K and 1 bar for 3 h.To get 1 wt% of Au loading, it must be considered that only~70% of the nominal gold is deposited on the surface of TiO2 at a pH of 7.

Deposition of platinum or palladium on gold catalyst was done from aqueous solutions of H2PtCl6·6H2O (Aldrich, > 37.5% as Pt); or PdCl2 (Aldrich, 99%).Gold catalyst was impregnated with the solution containing the desired amount of metal to form a thick paste (incipient wetness impregnation).After a perfect mixing of the wet solid, samples were dried at 373 K for 12 h.The resultant powders were reduced in fix bed reactor using 50 mL/min of H2 at the selected temperature for 3 h before reaction.

Deposition of platinum was done from aqueous solutions of H2PtCl6.Supports (Active carbon Darco KB-B, Aldrich) were impregnated with the solution containing the desired amount of metal to form a thick paste (incipient wetness impregnation).As an example, 1.2 mL of an aqueous solution containing 10.62 mg of H2PtCl6·6H2O were contacted with 2 g of carbon Darco to prepare the 0.2 wt% Pt/C catalyst.After a perfect mixing of the wet solid, samples were dried at 373 K for 12 h.The resultant powders were reduced in fix bed reactor using 50 mL/min of H2 at the selected temperature for 3 h before reaction.

2.2 Catalytic performance

Catalytic reactions were performed in a reinforced glass reactor (2 mL) with temperature and pressure control, and stirred magnetically.The feed composition was (mol): 90.5% solvent, 6% nitro compound, 3% aldehyde, 0.5% o-xylene (internal standard).Typically, 100 mg of catalyst were used for 900 mg of feed.Reaction pressure and temperature are indicated in the corresponding table of the main text.Conversion and selectivities were determined using a gas chromatograph (Varian 3900) equipped with an FID detector and a 30-m HP-5 capillary column.The products were identified by mass spectrometry using a GC/MS device (Agilent MDS-5973) equipped with a quadrupole electron-impact ionization detector.

3 Results and discussion

In this paper, we have designed a catalyst that affords nitrones in higher yields than any preceding process, starting from aliphatic nitro compounds, aldehydes (aromatic or aliphatic) and H2 in a single step process, for doing that the catalyst should hydrogenate selectively nitroaromatic compounds, without causing reduction of other sensitive groups such as carbonyls.This is a necessary condition for obtaining nitrones via a cascade-type reaction.The condition is, however, not sufficient and previous attempts to accomplish this reaction with gold catalysts were fruitless [12].Despite the tolerance of the carbonyl function, the Au/TiO2 catalyst required higher temperatures to dissociate H2 and the catalyst gives lower selectivity to the nitrone 3 when reacting with 5-methylfurfural 2 (Table 1) since, the imine 4 (formed by condensation of aniline and the 5-methylfurfural 2) is obtained in high yields (the corresponding imine forms according to the sequence of reactions in Schemes 2 and 5).

Table 1
Effect of temperature on the selectivity to nitrone 3 formed from nitrobenzene 1, 5-methylfurfural 2 and H2 (5 bar) using the different types of catalysts [12].

A first series of experiments show that the selectivity to nitrone compounds with Au/TiO2 is largely dependent on the reaction temperature, and the nature of solvent.More particularly, higher reaction temperatures imply, in general, lower selectivities to the nitrones, not only with the Au/TiO2catalyst, but also with other metals (Table 1).

It was presented before that while Au/TiO2 was not selective enough for producing nitrones from nitro aromatics compounds since this catalyst must operate at higher temperatures to achieve practical reaction rates; with Pt on carbon the rate of H2 dissociation could be enhanced and the catalyst can work at lower temperature, showing very high selectivity to aromatic substituted nitrones [12].Unfortunately, Pt/C catalyst, used to obtain nitrones in high yields from nitroaromatic compounds and aldehydes, is less selective when nitroalkanes were reacted, due to the concomitant formation of the corresponding alcohol (Table 2).On the other hand, we realized that nitrones coming from nitro aliphatic compounds were more stable to reaction temperature and the Au/TiO2 catalyst (that requires higher reaction temperatures) could be a suitable option.

Table 2
Catalytic performance of various catalysts for the reductive coupling of nitrobutane 7 and 5-methyl furfural 2.

With the objective to analyze the mechanism of interaction between active centers and different functional groups with Pt/C and Au/TiO2 catalysts, activation of nitrobenzene and styrene was studied in a situation of competitive adsorption.Fig. 1(a) shows presence of nitrobenzene avoiding double bond reduction with Au/TiO2 and also with Pt/C catalysts; even though with Pt/C catalyst a higher hydrogenation rate is observed.Competitive adsorption shows an important diminish in styrene activity due to strong adsorption of nitro aromatic compound on the surface avoiding styrene activation on active centers, because of this, there is not an important reduction of double bond until more than 90% of nitrobenzene has been consumed.

Fig. 1. Evolution of the conversion of styrene and nitro compounds in terms of competitive reactivity (equimolecular blend) using Au/TiO2 and Pt/C catalysts. Reaction mixture composition: 4.25% nitro compound, 4.25% styrene, 0.005% o-xilene and 91.495% ethanol.Reaction conditions: Au/TiO2 353 K, 10 bar H2 and 100 mg 1%Au/TiO2; Pt/C: 313 K, 5 bar H2 and 50 mg 0.2%Pt/C.

An equivalent experiment with a nitro aliphatic compound, like nitrobutane, in the presence of styrene (Fig. 1(b)) was done and the results show a much higher rate of disappearance of double bond with Pt/C catalyst, meanwhile good selectivity to nitrobutane compound is preserved with Au/TiO2 catalyst.In conclusion, reduction of double bond on Pt/C catalyst is much faster when there is a nitro aliphatic compound in the reaction media and the highest selectivity of Au/TiO2 catalyst showed in this paper can be explained because the strong activation of -NO2 group in the interface Me-Ti [16].

Taking into account these results, behavior of gold nanoparticles supported on titania in the formation of aliphatic nitrones was studied.When reacting nitrobutane 7 and 5-methyl furfural 2, a high selectivity to nitrone 8 was obtained, and only gives small yields of the imine 9 and amine 10 (Table 2).Notice that the Au/TiO2 catalyst was not forming the 5-methyl furfural alcohol 6 (Fig. 2).Other important information is that changing the solvent to a non-polar (toluene), imine 9 formation was favored, resulting in a slight decrease in the formation of nitrone 8.Meanwhile the 0.2 wt% Pt/C not only produces 5-methyl furfural alcohol 6 but also large amounts of by-products 9 and 10, when working under conditions that was selective for producing aromatic nitrones.A first series of experiments have shown that selectivity to nitrone with Au/TiO2 catalysts is largely dependent on the reaction temperature and the nature of solvent.

Fig. 2. Kinetic curve correspond to conversion of 5-methyl furfural and selectivities to nitrone, imine, amine and tertiary amine (circles) during the reductive coupling of nitrobutane and 5-methyl furfural at 353 K and 10 bar with Au/TiO2 catalyst.

At this point, in an attempt to increase the activity of gold catalyst, a series of bimetallic Me-Au/TiO2 catalysts were prepared with the objective of increasing the rate of hydrogenation while preserving high chemoselectivity.With this purpose, small amounts of Pt or Pd were introduced to the Au/TiO2 catalyst to accelerate the rate of H2 dissociation.The resultant catalysts were tested at 333 K and the results show that it is, indeed, possible to increase the reaction rate in the presence of a polar solvent (ethanol).However, though the rate of H2dissociation was increased, the selectivity to nitrone 8 was not improved due to the formation of a new by product, i.e.tertiary amine 11.When 50 ppm Pt on Au/TiO2 catalyst was tested using ethanol as solvent, the selectivity to nitrone 8 was only 57% (Table 2), and increasing the amount of Pt to 100 ppm resulted in a slight decrease to nitrone 8 (46%).Changing to a non-polar solvent (toluene), imine 9 is the main product with 50 ppm Pt-Au/TiO2 and 50 ppm Pd-Au/TiO2 catalysts.From results of Table 2 we can conclude that the activity towards hydrogenation was increased considerably with Pt-Au/TiO2and Pd-Au/TiO2 catalysts, in order to have good selectivity to nitrone 8 not only a chemoselective catalyst is needed, but a polar solvent is also required.

When a nitro cycloaliphatic is used, like nitro cyclohexane 12, the polymerization reaction to tertiary amine is minimized and using a proper amount of Pt (50 ppm Pt-Au/TiO2) selectivity to nitrone 13 could be increased up to 85%.The bimetallic Pd-Au/TiO2 catalyst is also selective to the nitrone 13, though it is less selective than the equivalent Pt-Au/TiO2 catalyst but both of them are more selective than the commercial Lindlar catalyst.As showed in Table 3, we can conclude that when only small amounts of higly dispersed Pt are added to the catalyst, the activity towards hydrogenation increases considerably while the chemoselectivity of Au/TiO2 catalyst is maintained, kinetic profiles are shown in Fig. 3.

Table 3
Catalytic performance of various catalysts for the reductive coupling of nitro cyclohexane 12 and 5-methyl furfural 2.

Fig. 3. Kinetic curve correspond to conversion ot 5-methyl furfural and selectivities to nitrone, imine, amine and tertiary amine during the reductive coupling of nitro cyclohexane and 5-methyl furfural at 353 K and 10 bar with Au/TiO2 catalyst.

Au/TiO2 catalyst was tested for reacting various nitro aliphatic compounds and aromatic aldehydes in the presence of H2.Nitrones coming from the nitro aliphatic compounds, which are more stable to reaction temperature, were observed.Showing that the gold based catalyst allows obtaining higher selectivities to the corresponding aliphatic nitrone (87%-67%) than Pt/C catalyst (Table 4).Regenerability of Au/TiO2 catalyst was tested during three cycles without losing activity and selectivity, in order to recover catalytic activity after each cycle it should be washed with ethanol and water and then calcined at 673 K.

Table 4
Catalytic performance for the reductive coupling of various nitro aliphatic compounds and aromatic aldehydes.

Special mention should be made to the reaction between an aliphatic nitroalkane and an aliphatic adehyde, i.e.nitrobutane 7 and valeraldehyde 16 (Scheme 6).

Scheme6. Catalytic performance of reductive coupling of nitrobutane 7 and valeraldehyde 16.

In this case, besides the nitrone 17, the hydroxyl secondary amine 18 is also obtained.Results in Table 5 show again that gold catalyst is more selective to the nitrone 17 than the bimetallics and Pt/C catalysts, being Au/TiO2 catalyst the most selective.When different amounts of Pt were added to Au/TiO2 catalyst, the distribution of byproducts changed, but in all cases the selectivity to nitrone 17 was similar.Results can be extended to aldehydes with longer aliphatic chains (i.e.heptaldehyde) indicating that this kind of substrates are not as sensitive to changes in the properties of metal-support interface as the aromatic aldehydes.According with results showed in Table 5, this behavior is not observed with nitro tert-butane and valeraldehyde 16 and we can suggest that the low reactivity of the reactants above are due to the high steric hindrance of nitro compound.

Table 5
Catalytic performance for the reductive coupling of various aliphatic nitro compounds and aliphatic aldehydes.

Due to anti-cancer activity of PBN (a-phenyl-tert- butylnitrone) many efforts have been done in order to synthesis this important nitrone with Au/TiO2 and Pt/C catalysts but were fruitless; only doping Pt/C catalyst with Pb according to Siegrist procedure [17] was possible to obtain a certain selectivity to PBN nitrone (Table 6).

Table 6
Catalytic performance of various catalysts for the reductive coupling of nitro-tert-butyl and benzaldehyde.

4 Conclusions

It is possible to produce nitrones from nitro compounds, aldehydes and H2 in a one-pot system, provided that chemoselective hydrogenation metals are used as catalysts.Selective formation of nitrones from nitroaromatic compounds can be achieved with platinum on carbon catalyst with a high proportion of unsaturated Pt sites, but, when reacting nitroalkanes, Pt on carbon gives low to moderate selectivity to nitrones due to the concomitant formation of the corresponding alcohol.However, Au/TiO2 catalyst is able to afford nitrones from aliphatic nitro compounds, aldehydes (aromatic or aliphatic) and H2 in higher yields than any preceding solution, when reaction temperature is increased.A proper amount of Pt or Pd introduced into the Au/TiO2 catalyst is able to increase the rate of hydrogenation reaction of the nitro groups, while preserving high chemoselectivity.

References
[1] E. G. Janzen, C. A. Evans, E. R. Davis, in:W. A. Pryor ed., Organic Free Radicals, ACS Symp. Ser., Vol. 69, American Chemical Society, Washington, DC, 1978, 433-446.
[2] Scott G., Smith K. V., Eur. Polym. J.,1978, 14 :39–43. DOI:10.1016/0014-3057(78)90149-0
[3] Lee K. J., Kim D. H., Bioorg. Med. Chem. Lett.,1998, 8 :323–326. DOI:10.1016/S0960-894X(98)00023-7
[4] Gothelf K. V., Jorgensen K. A., Chem. Rev.,1998, 98 :863–909. DOI:10.1021/cr970324e
[5] Chen G., Bray T. M., Janzen E. G., McCay P. B., Free Radic. Res. Commun.,1990, 9 :317–23. DOI:10.3109/10715769009145690
[6] Floyd R. A., Aging Cell,2006, 5 :51–57. DOI:10.1111/ace.2006.5.issue-1
[7] Floyd R. A., Kopke R. D., Choi C. H., Foster S. B., Doblas S., Towner R. A., Free Radic. Biol. Med.,2008, 45 :1361–1374. DOI:10.1016/j.freeradbiomed.2008.08.017
[8] Corma A., Serna P., Garcia H., J. Am. Chem. Soc.,2007, 129 :6358–6359. DOI:10.1021/ja0704131
[9] Combita D., Concepcion P., Corma A., J. Catal.,2014, 311 :339–349. DOI:10.1016/j.jcat.2013.12.014
[10] Richner G., van Bokhoven J. A., Neuhold Y. M., Makosch M., Hungerbuehler K., Phys. Chem. Chem. Phys.,2011, 13 :12463–12471. DOI:10.1039/c1cp20238c
[11] Haber F., Zeit. Elektrochem.,1898, 4 :506–513. DOI:10.1002/bbpc.v4:22
[12] Cisneros L., Serna P., Corma A., Angew , Chem ., Int. Ed.,2014, 53 :9306–9310. DOI:10.1002/anie.201402878
[13] Sakurai H., Akita T., Tsubota S., Kiuchi M., Haruta M., Appl. Catal. A,2005, 291 :179–187. DOI:10.1016/j.apcata.2005.02.043
[14] Cha D. Y., Parravano G., J. Catal.,1970, 18 :200–211. DOI:10.1016/0021-9517(70)90178-8
[15] Corma A., Serna P., Concepcion P., Calvino J. J., J. Am. Chem. Soc.,2008, 130 :8748–8753. DOI:10.1021/ja800959g
[16] Corma A., Serna P., Science,2006, 313 :332–334. DOI:10.1126/science.1128383
[17] Siegrist U., Baumeister P., Blaser H. U., Studer M., Chem. Ind. (Dekker),1998, 75 (Catalysis of Organic Reactions):207–219.