Doped microporous graphitic carbons as metal-free catalysts for the selective hydrogenation of alkynes to alkenes

Highlights

• Microporous graphitic carbons (0.64 nm) doped with N or P at various levels have been prepared.

• The process involves pyrolysis ofα-cyclodextrin that self-assembles during the process.

• N- and P-doped graphitic carbons exhibit activity for partial hydrogenation of alkynes to alkenes.

• N- and P-doped carbons appear stable under reaction conditions and exhibit constant TOF values.

• The most active N-doped carbons exhibits chemo and stereo selectivity towards the cis configured alkene.

Abstract

Microporous graphitic carbons obtained by pyrolysis of α-cyclodextrin have been conveniently doped with N and P elements by performing the thermal graphitization adding urea or phosphoric acid as source of these elements. Electron microscopy shows that doping does not influence the structure and dimensions of the micropores, while XPS reveals the distribution of these dopant elements in two main families corresponding to pyridinic and pryrrolic N atoms and to triphenylphosphine and triphenylphosphine oxide for P doping. Thermo-programmed desorption of H_2, CO₂ and NH₃ indicates that the presence of dopant element is responsible for chemisorption of these gases and introduces basic and acid sites. These desorption measurements agree with the catalytic performance of the series of carbons for the chemo and stereo selective hydrogenation of internal aliphatic and aromatic alkynes to cis configured alkenes. N/P doped microporous carbons were found to be stable under the reaction conditions, according to the constant TOF values and XRD and XPS characterization of the materials after their use as catalysts. In addition microporosity appears to play a positive effect by comparing the activity of microporous carbons with the performance of analogous N/P-doped graphenes.

Introduction

Porosity plays a key role in heterogeneous catalysis. On one hand, the presence of pores increases the surface area of a material compared to bulk particles. On the other hand, there are considerable experimental evidence supported by theoretical calculations on the possible operation within the pores of the so-called confinement effect [1], [2], [3], [4], [5]. Confinement effects can increase the reaction rate when the dimensions of the pores are similar to that of reagents [6], [7]. Compared to flat surfaces, location of active sites inside pores can be favorable to overcome activation barriers in a chemical reaction [8], [9]. Curvature of the pore walls around substrates and reagents may result in their activation by a combination of weak van der Waals interactions as well as dipole–dipole forces and internal electrostatic fields [10], [11]. Most of the examples regarding confinement effect correspond to microporous inorganic materials, having pores of dimensions smaller than 2 nm [12], [13], [14]. Confinement effect has been proposed, among other examples, to contribute to hydrocarbon activation in cracking in acid zeolites, explaining the remarkable activity of these aluminosilicates in this reaction in spite of the moderate acid strength measured by conventional techniques [13], [15].

Regarding carbons, there are various procedures to produce porous carbons [16], [17], [18]. However, most of them based on the use of hard or soft templates, rendering carbonaceous materials having meso- or macropores. Preparation of microporous carbons with strictly defined, regular quasi-crystalline pore size has resulted elusive in most of the cases. In this context, we have recently reported that pyrolysis of cyclodextrins renders highly crystalline, microporous graphitic carbon nanoparticles with strictly regular pore size from 0.64 to 1.09 nm depending on the dimensions of the cyclodextrin precursor [19]. It was found that these microporous graphitic carbons were able to promote aerobic oxidation of benzylic hydrocarbons with notable higher activity than active carbons lacking uniform porosity or reduced graphene oxide considered as flat 2D layers [19]. The catalytic activity correlates with the pore size, the microporous graphitic carbon derived with α-cyclodextrin (C_α) and exhibiting the smallest pore size (0.64 nm) being the most active material. DFT calculations indicate that the reason for this higher activity can be attributed to a confinement effect since molecular oxygen inside the channels of C_α receives electron density from the carbon walls in an extent that decreases as the dimensions of the channel increases. This causes an elongation of the OO bond that corresponds in fact to a preactivation of the molecule. Interestingly, the aerobic oxidation occurs on or within the microporous nanoparticles, since removal of the solid carbon stops completely the oxidation reaction. Confirmation of the superior activity as microporous metal-free catalysts of C_α compared to other related materials is an issue of wide general interest in the quest for providing metal-free carbocatalysts with improved performance for general reaction types, some of them supposedly being exclusively promoted by transition metals [20], [21].

Hydrogenations of CC multiple bonds is in fact one general reaction of large industrial importance that are currently performed using transition metal catalysts, in many cases containing noble metals [22]. We have previously reported that defective graphenes are active catalysts for alkene hydrogenation [23], [24], for the selective hydrogenation of alkynes to alkenes [25] and the reduction of nitro to amino groups [26]. To put into context the catalytic activity of C_α materials and determine the influence of microporosity respect to flat reduced graphene layers and related carbons, a step forward would be to assess the activity of microporous C_α as hydrogenation catalysts. In addition, the influence of doping on these microporous graphitic carbons in their catalytic activity still remains unexplored. Specifically, it is of interest to determine in which way dopant elements enhance the activity of microporous graphitic carbons as hydrogenation catalysts.

Herein, it will be shown that while C_α is devoid of any activity as hydrogenation catalyst due to the lack of defects considered as active sites [27], N or P doping renders the resulting doped microporous carbons active as metal–free hydrogenation catalysts as consequence of their H₂ chemisorption ability that is absent in the undoped C_α. The performance of these microporous C_αmaterials compares favorably with that of related non-porous graphenes, illustrating the potential of combining active sites and adequate microporosity as a valid strategy to increase the catalytic activity of these carbocatalysts.