1 Introduction

The selective oxidation of cyclohexane is a key industrial route in the production of cyclohexanol and cyclohexanone (collectively termed KA oil), which are key feedstocks for adipic acid and caprolactam and ultimately Nylon-6,6 and Nylon-6. Conventionally the industrial route to KA, via the high temperature (140–180 °C) aerobic oxidation of cyclohexane utilises homogeneous transition metal catalysts including cobalt [1, 2], chromium [3], or manganese [4], with conversion limited to around 5% to inhibit overoxidation of the desired products. In particular significant concentrations of ring-opened by products including 6-hydroxyhexanoic acid and glutaric acid are produced at even moderate rates of conversion [5, 6]. Typically, KA oil production involves two consecutive steps: (i) the non-catalytic auto-oxidation of cyclohexane to cyclohexyl hydroperoxide (CHHP) and (ii) the catalysed decomposition of CHHP to KA oil, with carboxylic acids and esters also produced in low quantities.

The high energy costs that result from the use of elevated temperatures and the difficulties associated with separating homogeneous catalysts from the product stream has led to growing interest in the use of heterogenous catalysts for aerobic KA oil production, with previous studies reporting the efficacy of transition metal containing molecular sieves [7,8,9,10] and precious metals catalyst [11, 12]. While further studies have reported that the replacement of molecular O2 with either tert-butylhydroperoxide (TBHP) [13, 14] or pre-formed H2O2 [15, 16] allows for the use of significantly lower reaction temperatures. However, typically these oxidants are utilised in excess to improve efficiency [14], with the catalysed and non-catalysed decomposition of H2O2 to H2O a major competitive reaction pathway.

Apart from concerns around atom efficiency there are several economic and environmental drawbacks associated with the use of commercial H2O2 as an oxidant. These are generally associated with the means by which H2O2 is produced on an industrial scale, the in-direct or anthraquinone oxidation (AO) process, which although highly optimized, with high H2 selectivity, is only economically viable when operating on a large scale, often prohibiting production at the site of final use. As such H2O2 is typically transported and stored, at concentrations in excess of 70 wt.% prior to dilution, effectively wasting significant amounts of energy utilised in concentration. Furthermore, the low stability of H2O2 at mild temperatures necessitates the use of acidic stabilizers to prevent decomposition to H2O during transport and storage, these stabilising agents often need to be removed from product streams prior to shipping, with additional costs passed on to the end user [17, 18].

The in-situ production of H2O2 from molecular H2 and O2 would offer an attractive alternative to the use of pre-formed H2O2 as an oxidant for cyclohexane, overcoming the aforementioned issues associated with commercial H2O2. In addition, in-situ H2O2 production would also lead to significantly reduced process costs, compared to the current industrial route to KA oil, associated with lower reaction temperatures. Indeed we have recently demonstrated that limited concentrations of KA oil can be produced via the in-situ production of H2O2 over supported AuPd nanoparticles, prepared via a conventional wet-impregnation procedure, under reaction conditions where no conversion of cyclohexane is observed using O2 alone [19, 20].

With these previous studies in mind we now investigate a range of bi-functional Pd-based catalysts that combine the H2O2 synthesising activity of Pd and the ability of a range of transition metals to selectively catalyse the oxidation of cyclohexane to KA oil.

2 Experimental

2.1 Catalyst Preparation

Mono- and bi-metallic 1%PdX/TiO2 (X = Au, Mn, Fe, V, Co, Ni, Cu, Ce) catalysts, prepared on a weight basis, with a Pd: X ratio of 1: 1 (wt/wt) have been synthesised via a modified impregnation procedure, based on methodology previously reported in the literature [21, 22]. With catalysts produced via an impregnation procedure widely studied for both the direct synthesis of H2O2 [23] and cyclohexane oxidation [24] due to the simplicity and ease with which this approach can be scaled to meet industrial application. The procedure to produce 0.5% Au–0.5% Pd/TiO2 (2 g) is detailed below, with a similar methodology utilized for all mono- and bi-metallic catalysts using chloride based metal precursors in all cases (see Table S.1 for further details). In all cases catalysts have been prepared using PdCl2 (0.58 M HCl, 6 mgmL–1, Sigma-Aldrich).

Aqueous acidified PdCl2 solution (1.667 mL, 0.58 M HCl, 6 mgmL–1, Sigma-Aldrich) and aqueous HAuCl4·3H2O solution (0.8263 mL, 12.25 mgmL–1, Strem Chemicals) were mixed in a 50 mL round-bottom flask and heated to 65 °C with stirring (1000 rpm) in a thermostatically controlled oil bath, with total volume fixed to 16 mL using H2O (HPLC grade). Upon reaching 65 °C, TiO2 (1.98 g, Degussa, P25) was added over the course of 5 min with constant stirring. The resulting slurry was stirred at 65 °C for a further 15 min, following this the temperature was raised to 95 °C for 16 h to allow for complete evaporation of water. The resulting solid was ground prior to a reductive heat treatment (5%H2/Ar, 500 °C, 4 h, 10 °C min–1).

Surface area measurements for key materials, as determined by 5-point N2 adsorption, are reported in Table S2.

2.2 Catalyst Testing

2.2.1 Note 1

Reaction conditions used within this study operate below the flammability limits of gaseous mixtures of H2 and O2 (4–94%).

2.2.2 Note 2

The conditions used within this work for H2O2 synthesis and degradation have previously been investigated, with the presence of CO2 as a diluent for reactant gases and a methanol co-solvent have identified as key to maintaining high catalytic efficacy towards H2O2 production [22, 25].

2.3 Direct Synthesis of H2O2 From H2 and O2

Hydrogen peroxide synthesis was evaluated using a Parr Instruments stainless steel autoclave with a nominal volume of 50 mL, equipped with a PTFE liner and a maximum working pressure of 2000 psi. To test each catalyst for H2O2 synthesis, the autoclave liner was charged with catalyst (0.01 g) and HPLC grade standard solvents (5.6 g methanol and 2.9 g H2O). The charged autoclave was then purged three times with 5% H2 / CO2 (100 psi) before filling with 5% H2 / CO2 to a pressure of 420 psi, followed by the addition of 25% O2 / CO2 (160 psi) to achieve a total pressure of 580 psi. Pressures of 5% H2 / CO2 and 25% O2 / CO2 are given as gauge pressures. The reaction was conducted at a temperature of 20 °C, for 0.5 h with stirring (1200 rpm), with the reactor temperature controlled using a HAAKE K50 bath/circulator using an appropriate coolant.

H2O2 productivity was determined by titrating aliquots of the final solution after reaction with acidified Ce(SO4)2 (0.0085 M) in the presence of ferroin indicator. Catalyst productivities are reported as molH2O2kgcat−1 h−1.

2.4 Degradation of H2O2

Catalytic activity towards H2O2 degradation was determined in a similar manner to that used to measure the direct synthesis activity of a catalyst. The autoclave liner was charged with methanol (5.6 g, HPLC standard), H2O2 (50 wt. %, 0.69 g), H2O (2.21 g, HPLC standard) and catalyst (0.01 g), with the solvent composition equivalent to a 4 wt. % H2O2 solution. From the solution, prior to the addition of the catalyst, two 0.05 g aliquots were removed and titrated with acidified Ce(SO4)2 solution using ferroin as an indicator to determine an accurate concentration of H2O2 at the start of the reaction. The autoclave was purged three times with 5% H2 / CO2 (100 psi) before filling with 5% H2 / CO2 to a gauge pressure of 420 psi. The reaction was conducted at a temperature of 20 °C, for 0.5 h with stirring (1200 rpm). After the reaction was complete the catalyst was removed from the reaction mixture by filtration and two 0.05 g aliquots were titrated against the acidified Ce(SO4)2 solution using ferroin as an indicator. The degradation activity is reported as molH2O2kgcat −1 h−1.

2.5 Cyclohexane Oxidation via the In-situ Production of H2O2

The conditions utilised within this work have previously been optimised for the oxidation of cyclohexane via in-situ H2O2 production [19].

Cyclohexane oxidation has been evaluated using a Parr Instruments stainless steel autoclave with a nominal volume of 50 mL, equipped with a PTFE liner and a maximum working pressure of 2000 psi. To test each catalyst for cyclohexane oxidation, the autoclave was charged with catalyst (0.05 g), t-butanol solvent (6.375 g, Sigma Aldrich) and cyclohexane (2.125 g, 25 mmol, Sigma Aldrich) with mesitylene (0.43 g, 3.6 mmol, Sigma Aldrich) used as an internal standard. The charged autoclave was then purged three times with 5% H2/N2 (100 psi) before filling with 5% H2/N2 (420 psi, 2.5 mmol H2) followed by the addition of 25% O2/N2 (160 psi, 4.8 mmol O2) to achieve a total pressure of 580 psi. With all pressures given as gauge pressure. The temperature was then increased to 80 °C with stirring (500 rpm). Once the desired temperature was reached the reaction mixture was stirred (1200 rpm) for 17 h. After the reaction was complete the reactor was cooled in ice to a temperature of 15 °C, upon which a gas sample was taken for analysis by gas chromatography, using a Varian CP-3380 equipped with a TCD detector and a Porapak Q column. Product yield was determined by gas chromatography using a Varian 3200 GC equipped with a FID and CP Wax 42 column.

Quantification of the intermediate cyclohexyl hydroperoxide (CHHP) is determined by reacting a 2 mL aliquot of the post reaction mixture with an excess of triphenyl phosphine (0.13 g, 0.5 mmol). Reaction of triphenyl phosphine and CHHP produces cyclohexanol and hence comparison of GC analysis for cyclohexanol pre- and post-treatment with triphenyl phosphine can determine the yield of CHHP.

Catalytic conversion of H2 was determined using a Varian 3800 GC fitted with TCD and equipped with a Porapak Q column.

H2 conversion (Eq. 1), cyclohexane conversion (Eq. 2) and selectivity to all C6 products based on H2 (Eq. 3) are defined as follows:

$${\text{H}}_{2} {\text{Conversion }}\left( {\text{\% }} \right) = \frac{{{\text{mmol}}_{{{\text{H}}_{{2}} {\text{(t(0))}}}} - {\text{mmol}}_{{{\text{H}}_{{2}} {\text{(t(1))}}}} }}{{{\text{mmol}}_{{{\text{H}}_{2} {\text{(t(0))}}}} }}{ } \times { }100$$
(1)
$${\text{Cyclohexane Conversion }}\left( {\text{\% }} \right) = \frac{{{\text{mmol}}_{{{\text{Cyclo }}\left( {{\text{t}}\left( 0 \right)} \right)}} - {\text{mmol}}_{{{\text{Cyclo}}\left( {{\text{t}}\left( 1 \right)} \right)}} }}{{{\text{mmol}}_{{{\text{Cyclo }}\left( {{\text{t}}\left( 0 \right)} \right)}} }}{ } \times { }100$$
(2)
$${\text{C}}_{6} {\text{Product Selectivity }}\left( {\text{\% }} \right) = \frac{{{\text{Total product }}\left( {{\text{mmol}}} \right)}}{{{\text{H}}_{2} {\text{ conversion }}\left( {{\text{mmol}}} \right)}}{ } \times { }100{ }$$
(3)

2.5.1 Note 3

Given the relatively low conversion rates observed within this work total product yield (including CHHP, cyclohexanone and cyclohexanol) is used as a substitute for cyclohexane conversion.

2.6 Hot-Filtration Experiments

In order to determine the contribution of leached species towards the oxidation of cyclohexane, a standard reaction is carried out to that described above. After the initial 17 h reaction the heterogeneous catalyst was removed via filtration and the reaction solution returned to the autoclave, for a further 17 h reaction, under the conditions outlined above.

2.7 Catalyst Reusability in the Oxidation of Cyclohexane via In-situ Production of H2O2

In order to determine catalyst reusability, a similar procedure to that outlined above for the oxidation of cyclohexane is followed utilising 0.15 g of catalyst. Following the initial test, the catalyst was recovered by filtration, washed with cyclohexane, and dried (30 °C, 17 h, under vacuum); from the recovered catalyst sample 0.05 g was used to conduct a standard cyclohexane oxidation experiment.

2.8 Cyclohexane Oxidation Under Aerobic Conditions

To evaluate the efficacy of in-situ conditions (H2 + O2) catalysts have also been studied for their activity towards cyclohexane oxidation under aerobic conditions. A procedure similar to that outlined above for the oxidation of cyclohexane is followed using 25%O2/N2 (160 psi) and N2 (420 psi) to maintain total reaction pressure at 580 psi.

2.9 Catalyst Characterisation

A Thermo Scientific K-Alpha+ photoelectron spectrometer was used to collect XP spectra utilising a micro-focused monochromatic Al Kα X-ray source operating at 72 W. Data was collected over an elliptical area of approximately 400 μm2 at pass energies of 40 and 150 eV for high-resolution and survey spectra, respectively. Sample charging effects were minimised through a combination of low energy electrons and Ar+ ions, consequently this resulted in a C(1 s) line at 284.8 eV for all samples. All data was processed using CasaXPS v2.3.24 using a Shirley background, Scofield sensitivity factors [26] and an electron energy dependence of − 0.6 as recommended by the manufacturer.

Scanning Transmission Electron Microscopy (STEM) and X-ray Energy Dispersive Spectroscopy (X-EDS) data was taken using a JEOL JEM-ARM200CF microscope in Diamond Laboratory. The TEM specimens investigated were the as-prepared 0.5%Pd-0.5%V/TiO2 catalyst and the analogous sample after use in the cyclohexane oxidation reaction.

Total metal leaching from the supported catalyst was quantified via inductively coupled plasma mass spectrometry (ICP-MS). Post-reaction solutions were analysed using an Agilent 7900 ICP-MS equipped with I-AS auto-sampler. All samples were diluted by a factor of 10 using HPLC grade H2O (1%HNO3 and 0.5% HCl matrix). All calibrants were matrix matched and measured against a five-point calibration using certified reference materials purchased from Perkin Elmer and certified internal standards acquired from Agilent.

Brunauer Emmett Teller (BET) surface area measurements were conducted using a Quadrasorb surface area analyser. A 5-point isotherm of each material was measured using N2 as the adsorbate gas. Samples were degassed at 250 °C for 2 h prior to the surface area being determined by 5-point N2 adsorption at − 196 °C, and data analysed using the BET method.

3 Results and Discussion

Our initial studies investigated the efficacy of a series of mono-metallic catalysts, prepared by a modified impregnation procedure, towards the direct synthesis and subsequent degradation of H2O2, under conditions favourable towards H2O2 stability (Figure S.1) [22, 25]. The limited activity towards both the formation and subsequent degradation of H2O2 is clear for all mono-metallic catalysts with the exception of the 1% Pd/TiO2 catalyst, which displays rates of H2O2 synthesis (30 molH2O2kgcat−1 h−1) and H2O2 degradation (198 molH2O2kgcat−1 h−1) far greater than those observed for the non-Pd analogues. The greater activity of the 1%Pd/TiO2 catalyst is perhaps unsurprising given the extensive studies into supported Pd catalysts for H2O2 direct synthesis [27, 28]. The alloying of Pd with Au has been widely reported to enhance catalytic activity towards the direct synthesis of H2O2 with the synergistic enhancement observed for AuPd alloys typically attributed to electronic and isolation effects [29,30,31,32]. In keeping with these studies, we report a significant improvement in catalytic activity towards H2O2 synthesis over the 0.5%Au-0.5%Pd/TiO2 catalyst (68 molH2O2kgcat−1 h−1). However, we do not observe a decrease in H2O2 degradation rate, which is typically reported upon the alloying of Au and Pd [33]. With this observation in keeping with the findings of Santos et al. who have recently investigated a series of AuPd catalysts, prepared by an identical procedure, for H2O2 synthesis, under a range of reaction conditions [22]. This is likely due to the use of a reductive heat treatment in the preparation of these catalysts and the resulting formation of homogeneous alloy nanoparticles [21]. With numerous studies reporting the formation of a Au-core PdO-shell nanoparticle morphology, promoted by the oxidative heat treatment of AuPd catalysts, to be crucial to their enhanced selectivity towards H2O2 [25, 34].

In recent years focus has shifted towards alloying Pd with a range of non-precious metals, with numerous studies reporting the enhanced catalytic efficacy achieved through the introduction of Sn [35, 36], Ag [37], Zn [38], Ni [39, 40], In [41, 42], Sb [43], and Te [44] into Pd nanoparticles. Typically, the improved selectivity of the bimetallic catalysts has been attributed to a combination of a reduction of contiguous Pd ensembles and a modification of Pd oxidation state. In keeping with these previous reports our investigations also demonstrate the enhanced catalytic efficacy that can be achieved through the introduction of a range of transition metals into supported Pd catalyst (Fig. 1), with a clear reduction in H2O2 degradation rate observed upon the introduction of all secondary metals, with the exception of Au. This is perhaps unexpected given the ability of a range of these secondary metals, such as Fe, Mn and Cu, to catalyse the decomposition of H2O2 through Fenton or Fenton-like pathways [45, 46]. However, it is possible that the limited degradation activity observed can be related to a combination of two factors (i) the choice of reaction conditions used to evaluate H2O2 synthesis activity, with the in-situ formation of carbonic acid, through the dissolution of the CO2 reactant gas diluent resulting in the stabilisation of H2O2 and (ii) the modification of Pd oxidation states upon introduction of secondary metals, as observed via XPS analysis (Table S.3), with the formation of mixed domains of Pd0-Pd2+ reported to offer enhanced catalytic efficacy in both H2O2 synthesis [47, 48] and aerobic oxidation reactions [49]. It should be noted that our analysis by XPS reveals that the introduction of Au, unlike the majority of the other transition metals, does not promote the formation of these mixed oxidation state domains, which may explain the enhanced rates of H2O2 synthesis and degradation observed over this catalyst [50].

Fig. 1
figure 1

Catalytic activity of bimetallic supported catalysts towards the direct synthesis (Black bars) and degradation of H2O2 (Blue bars). H2O2 direct synthesis reaction conditions: Catalyst (0.01 g), H2O (2.9 g), MeOH (5.6 g), 5% H2/CO2 (420 psi), 25% O2/CO2 (160 psi), 0.5 h, 20° C, 1200 rpm. H2O2 degradation reaction conditions: Catalyst (0.01 g), H2O2 (50 wt.% 0.68 g) H2O (2.22 g), MeOH (5.6 g), 5% H2/CO2 (420 psi), 0.5 h, 20 °C, 1200 rpm

Full size image

Building on these findings we next investigated the efficacy of these Pd-based bi-metallic catalysts towards the selective oxidation of cyclohexane (Fig. 2). It should be noted that under the reaction conditions used within this study no residual H2O2 was measured in post reaction solutions. This is unsurprising given the comparatively high reaction temperatures and long reaction times utilised within this work. However, our previous studies, have elucidated the ability of H2O2 to be synthesised under similar reaction conditions, although this previous study reported complete selectivity towards cyclohexanol, likely as a result of the short reaction times utilised [20]. Interestingly the introduction of a range of secondary metals into supported Pd nanoparticles is seen to lead to a significant increase in catalytic activity towards the oxidation of cyclohexane under in-situ conditions (i.e. in the presence of H2 and O2). As expected, given the extensive studies into AuPd systems for a range of selective oxidation reactions [51,52,53] as well as for the direct synthesis of H2O2 [54, 55], in this study the introduction of Au into a supported Pd catalyst is also seen to result in an enhancement in the oxidation of cyclohexane, with a total C6 product yield of 24.1 µmol observed over the 0.5%Au-0.5%Pd/TiO2 catalyst, far greater than that observed over the Pd-only analogue (3.6 µmol), with the former also displaying a significantly enhanced selectivity to all C6 products compared to the 1%Pd/TiO2 catalyst (Table S.4).

Fig. 2
figure 2

Catalytic activity of bi-metallic Pd based supported catalysts towards the oxidation of cyclohexane via the in-situ production of H2O2. Cyclohexanol; blue, cyclohexanone; red, CHHP; purple. Reaction conditions: Catalyst (0.05 g), cyclohexane (2.13 g, 25 mmol), t-butanol (6.37 g), 5% H2 / N2 (420 psi), 25% O2 / N2 (160 psi), 17 h, 80 OC 1200 rpm. Note: When under aerobic conditions 25% O2 / N2 (160 psi) is used in addition to N2 (420 psi) to maintain total pressure. All other conditions as described above

Full size image

It should also be noted that the total product yield, observed over the 0.5%Au-0.5%Pd/TiO2 catalyst studied within this work (prepared via a modified-impregnation methodology) is far greater than that we recently reported for an analogous catalyst prepared via a conventional wet co-impregnation methodology (13 µmol) [19]. With this ascribed, at least in part, to the greater activity of the catalyst synthesised via a modified-impregnation methodology to synthesise H2O2, as indicated by our testing under conditions conducive towards H2O2 (Table S.5).

Interestingly, despite demonstrating a limited activity towards H2O2 synthesis (40 mol H2O2kgcat−1 h−1), the 0.5%V-0.5%Pd/TiO2 catalyst is observed to display a total C6 product yield (56 µmol) far greater than that of either the 0.5%Au-0.5%Pd/TiO2 (24.1 µmol), or 1%Pd/TiO2 (3.6 µmol) catalysts (Fig. 2). With a corresponding improvement in selectivity towards all C6 products based on H2 also observed over the 0.5%V-0.5%Pd/TiO2 catalyst (Table S.4). Evaluation of apparent turnover frequencies (TOFs) based on total metal content further highlights the improved activity of the 0.5%V-0.5%Pd/TiO2 catalyst compared to all other bi-metallic catalysts studied (Table S.6).

The high activity of V-based catalysts when used in conjunction with pre-formed H2O2 to catalyse the oxidation of alkanes has been widely linked to the V4+/V5+ redox cycle [56,57,58,59]. With the enhanced activity of the VPd supported catalyst in this work attributed to the dual functionality of the catalyst, with Pd catalysing the synthesis of H2O2, which is then subsequently activated by V (Scheme 1). Indeed a similar explanation can be attributed to the enhanced activity observed when alloying Pd with Fe (14.8 µmol) or other Fenton-like metals used in this study such as Cu, (18.7 µmol) with previous studies reporting the high cyclohexane oxidation activity when using a range of Fenton-like metals in conjunction with pre-formed H2O2 [16, 60].

Scheme 1
scheme 1

Proposed reaction pathway for the selective oxidation of cyclohexane via in-situ H2O2 production

Full size image

As expected, given the relatively low reaction temperatures used within this work the activity of all catalysts under in-situ reaction conditions (i.e. in the presence of H2 and O2) greatly outperforms that observed using O2 alone, clearly highlighting the benefits of the in-situ approach (Fig. 2). Indeed, under in-situ reaction conditions the 0.5%V-0.5%Pd/TiO2 catalyst is seen to offer greater product yield (56 µmol) than that observed when using commercial H2O2 (25.9 µmol) (Figure S.2).

Further studies reveal the benefits of immobilising both Pd and V onto the same grain of support (Fig. 3a). The combination of the monometallic 1%Pd/TiO2 and 1%V/TiO2 catalysts as a physical mixture, is seen to result in a marked improvement in total product yield (31.8 µmol) and selectivity to all C6 products based on H2 (Table S.7) compared to the corresponding monometallic catalysts. However, both the total product yield and selectivity towards C6 products observed over the physical mixture system is still far lower than that observed over the bimetallic catalyst. The activity of the 1%V/TiO2 catalyst when used alone (17.8 µmol) is noteworthy and can be ascribed to the catalysed aerobic oxidation pathway, with V catalysts well reported to offer some activity towards cyclohexane oxidation, even at ambient temperatures [61]. It should be noted that in the case of the 1%V/TiO2 catalyst the extent of H2 conversion was so low to be within experimental error, further supporting the role of V to catalyse the aerobic route.

Fig. 3
figure 3

a Catalytic performance of V-Pd catalysts towards the oxidation of cyclohexane via the in-situ production of H2O2. b The activity of 1% Pd/TiO2, 1% VPd/TiO2, V2O5, V2O4, and physical mixtures of 1% Pd/TiO2 and vanadium oxides towards the oxidation of cyclohexane via the in-situ production of H2O2. Cyclohexanol; blue, cyclohexanone; red, CHHP; purple. Reaction conditions: Catalyst (0.05 g), cyclohexane (2.13 g, 25 mmol), t-butanol (6.37 g), 5% H2 / N2 (420 psi), 25% O2 / N2 (160 psi), 17 h, 80 OC 1200 rpm. Note: Where physical mixtures have been used 0.025 g of each catalyst are utilised

Full size image

The excellent efficacy of V based catalysts towards the oxidation of alkanes, either in the form of unsupported vanadium oxides [62] or vanadium oxides supported on secondary oxide supports [63, 64] is well known, with perhaps the greatest interest placed on the vanadium phosphorus oxide (VPO) catalysts [58, 59, 65, 66]. With these studies in mind, we next compared the efficacy of the 0.5%V-0.5%Pd/TiO2 catalyst with a physical mixture of vanadium oxides (V2O4 and V2O5) and a 1%Pd/TiO2 catalyst (Fig. 3b). In keeping with the background aerobic activity observed over the 1%V/TiO2 catalyst, we observe some activity when using either V2O4 (7.9 µmol) or V2O5 (7.4 µmol) alone, again with negligible conversion of H2 observed. When using the 1%Pd/TiO2 catalyst in addition with either vanadium oxide a significant improvement in catalytic activity and selectivity to all C6 products based on H2 is seen, compared to when either component is used in isolation (Table S.8). The activity and H2 selectivity of the physical mixtures of vanadium oxide and 1%Pd/TiO2 are comparable to that observed for the 0.5%V-0.5%Pd/TiO2 catalyst despite the far lower concentrations of both Pd and V present in the bi-metallic catalyst, further supporting the need for both metals to be present on the same grain of support in order to reach maximal efficiency.

With the high catalytic efficacy of Pd-V/TiO2 catalysts established we next investigated the effect of V: Pd ratio, while maintaining total metal loading at 1 wt.% (Fig. 4). Given the separate and distinct roles of Pd and V, with the former primarily catalysing the synthesis of H2O2, and the latter subsequently activating the synthesised H2O2 and catalysing the oxidation of cyclohexane it is imperative to balance both reactions to achieve maximal oxidant efficiency. An optimal catalyst formulation of 0.5%V-0.5%Pd/TiO2 is observed, with this catalyst displaying the greatest total product yield and selectivity to all C6 products (Table S.9) and highest turnover frequency (Table S.10).

Fig. 4
figure 4

Catalytic activity of bi-metallic 1%VPd/TiO2 supported catalysts towards the oxidation of cyclohexane via the in-situ production of H2O2. Cyclohexanol; blue, cyclohexanone; red, CHHP; purple. Reaction conditions: Catalyst (0.05 g), cyclohexane (2.13 g, 25 mmol), t-butanol (6.37 g), 5% H2 / N2 (420 psi), 25% O2 / N2 (160 psi), 17 h, 80 OC 1200 rpm

Full size image

The oxidation state of Pd is well known to dictate catalytic activity towards the direct synthesis and subsequent degradation of H2O2, with the enhanced activity of reduced Pd-species towards both H2O2 formation and degradation to water (via decomposition and hydrogenation pathways) compared to analogous Pd2+ catalysts widely reported [47, 50, 67]. Recently several studies have revealed the improved efficacy of mixed domains of Pd0-Pd2+ towards both H2O2 synthesis [47, 48] and a range of aerobic oxidation transformations [49]. Likewise, the oxidation state of V is a crucial factor in the oxidation of cyclohexane, with a general agreement in the literature that the V4+/V5+ redox cycle is crucial in achieving high efficacy, while the accumulation of V3+ is suggested to lead to deactivation of the catalyst [65, 68].

XPS spectra of the as-prepared 1%VPd/TiO2 catalysts, with varying V: Pd ratios can be seen in Fig. 5. Upon co-immobilisation of both metals onto the same grain of support a stark difference in both Pd and V oxidation states is observed, compared to those present in the corresponding monometallic analogues.

Fig. 5
figure 5

Pd(3d) and V(2p3/2) core levels spectra for: (a) 1%Pd/TiO2, (b & h) 0.25%V-0.75%Pd/TiO2, (c & g) 0.5%V-0.5%Pd/TiO2, (d & f) 075%V-0.25Pd/TiO2 and (e) 1%V/TiO2. For Pd (3d) region: Pd0 (Green) Pd2+ (Blue), for V(2p3/2) region: V5+ (Green), V4+ (Blue)

Full size image

With the introduction of V into a Pd-only catalyst the proportion of Pd2+ present is seen to increase considerably compared to the 1%Pd/TiO2 analogue, which consists almost entirely of Pd0. Indeed, this culminates in a total shift towards Pd2+ in the case of the 0.75%V-0.25%Pd/TiO2 catalyst (Fig. 5d). However, this is not seen to detrimentally affect catalytic performance, and indeed it is likely that the oxidation states of the fresh material will not be representative of that under-reaction conditions.

For all catalysts containing V, two oxidation states of V are noted at 517.2 and 516.1 eV representative of V5+ and V4+ species respectively.[69] Notably, for the 0.25%V-0.75%Pd/TiO2 catalyst, a shift towards a lower binding energy, ca. 515.5 eV, is observed (Fig. 5h) which is typically representative of V3+ oxides and hydroxides. This coincides with a stark loss in catalytic performance (Fig. 4, Table S.9) and is in keeping with previous studies, which report that the presence of V3+ is able to inhibit the activity of V-based catalysts towards cyclohexane oxidation [65, 68].

For any heterogeneous catalyst operating in a three-phase system the possibility of the leaching of the active phase and resulting homogeneous contribution to observed catalytic performance is of great concern, with the activity of homogeneous V and Pd species known to catalyse the oxidation of cyclohexane and the direct synthesis of H2O2 respectively [70, 71]. Analysis of post reaction solutions via ICP-MS (Table S.11) reveals minimal leaching of Pd, with approximately 0.29% (0.001 wt.%) Pd detected in the post-reaction solution, on the other hand a significant amount of leached V is observed, approximately 21% (0.11 wt.% V).

Subsequent experiments, using an identical number of moles of V to that present in the 0.5%V-0.5%Pd/TiO2 catalyst, further identify the ability of homogeneous V species (as VCl3) to catalyse the oxidation of cyclohexane (11.7 µmol), though aerobic pathways as indicated by the negligible H2 conversion observed. In keeping with our earlier studies (Fig. 3a–b) a significant improvement in catalytic performance is seen when combining VCl3 with a heterogeneous Pd catalyst (33 µmol), again so that total moles of metal are identical to that present in the 0.5%V-0.5%Pd/TiO2 catalyst (Figure S.3, Table S.12). However, as in these earlier studies the 0.5%V-0.5%Pd/TiO2 catalyst is seen to offer greater product yield (56 µmol), highlighting the need for both metals to be in close proximity.

Hot filtration experiments (Fig. 6) confirm a contribution to cyclohexane oxidation from leached species (31.8 µmol). However, the contribution to total observed products in the hot filtration experiment from the decomposition of existing CHHP (i.e. CHHP produced in the presence of the heterogeneous catalyst) should be considered, with 14 µmol CHHP detected in the heterogeneously catalysed reaction, prior to the hot filtration experiment. Given the high stability of Pd, with minimal leaching observed in post reaction solutions, it is reasonable to assume that the major route to KA oil in the hot filtration experiments, excluding the decomposition of the pre-formed CHHP (which possibly accounts for more than 40% of the additional observed products), is via V catalysed aerobic oxidation pathways, as previously observed when using homogeneous V species, vanadium oxides or 1%V/TiO2 as catalysts for the cyclohexane oxidation reaction (Fig. 3, Figure S.3).

Fig. 6
figure 6

Reusability of 0.5%V-0.5%Pd/TiO2 in the oxidation of cyclohexane via the in-situ production of H2O2. Cyclohexanol; blue, cyclohexanone; red, CHHP; purple. Reaction conditions: Catalyst (0.05 g), cyclohexane (2.13 g, 25 mmol), t-butanol (6.37 g), 5% H2 / N2 (420 psi), 25% O2 / N2 (160 psi), 17 h, 80 OC 1200 rpm

Full size image

Despite the extensive leaching of V a significant improvement in the activity of the 0.5%V-0.5%Pd/TiO2 catalyst is seen upon re-use, (90.9 µmol) (Fig. 6) with a subsequent improvement in catalytic selectivity based on H2 also observed (Table S.13). While we have established that the activity may in part be attributed to leached V species, with a continual leaching of V observed upon re-use (14%, 0.07 wt.% V), there is clearly a significant heterogenous component to the oxidation reaction.

Evaluation of the fresh and used 0.5%V-0.5%Pd/TiO2 catalysts by XPS (Figure S.4, Table S.14) reveals a shift in both Pd(3d) and V(2p) spectra indicative of an increase in both Pd2+ and V4+ content, with no V3+ observed. This in turn correlates well with a measurable improvement in both catalytic performance and selectivity based on H2 (Fig. 6, Table S.11), which is perhaps unsurprising, with the formation of mixed domains of Pd0-Pd2+ reported to offer enhanced catalytic efficacy in both H2O2 formation and aerobic oxidation reactions [48, 49].

Meanwhile analysis of the as-prepared 0.5%V-0.5%Pd/TiO2 catalyst, (Fig. 7a) in addition to the analogous sample after use in the cyclohexane oxidation reaction (Fig. 7b) by HAADF-STEM can be seen in Fig. 7, with accompanying EDX analysis reported in Figure S.5. From our analysis there is no clear indication of alloyed Pd-V species within the fresh sample, with V observed to be highly dispersed over the entirety of the support, whereas Pd is present as discrete nanoparticles. In keeping with previous investigations these Pd nanoparticles are generally relatively small (2–5 nm) with very few larger agglomerates (> 10 nm) observed [22]. Although it should be noted that it was not possible to count a statistically relevant number of nanoparticles to generate an accurate mean particle size for the fresh and used catalysts. As with the fresh sample, analysis of the 0.5%V-0.5%Pd/TiO2 catalyst after use in the cyclohexane oxidation reaction indicates that V remains well dispersed, although it should be reiterated that our analysis by ICP-MS (Table S.11) does indicate leaching of V during the reaction, with these homogeneous species only responsible for limited catalytic activity.

Fig. 7
figure 7

Representative bright field transmission electron micrographs of 0.5%V-0.5%Pd/TiO2 catalyst a as prepared by modified impregnation methodology and b after use in the cyclohexane oxidation reaction, under in-situ conditions

Full size image

By comparison to V, some agglomeration of the Pd nanoparticles is observed upon use in the cyclohexane oxidation reaction, with an increased proportion of Pd present as larger (> 10 nm) nanoparticles, although the smaller species previously observed in the fresh sample are still present. This is in keeping with our previous investigation into the ability of supported AuPd nanoparticles to catalyse the oxidation of cyclohexane under identical reaction conditions [19]. As in this previous study the apparent increase in particle size does not appear to detrimentally effect catalytic efficacy towards cyclohexane upon re-use (Fig. 6).

4 Conclusion

We have demonstrated that it is possible to achieve appreciable rates of cyclohexane oxidation to KA oil using H2O2 generated in-situ using a 1%VPd/TiO2 catalyst, prepared by a readily scalable modified-impregnation procedure. This is observed using conditions where activity is limited when using molecular O2 alone, with no loss in catalytic activity observed with re-use, although catalyst stability is still of concern. The enhanced performance of the optimal 0.5%V-0.5%Pd/TiO2 catalyst, observed to be an order of magnitude greater than that of the Pd-only analogue, is attributed to the production of domains of mixed Pd oxidation states upon V incorporation and the dual functionality of the catalyst. We consider that these catalysts represent a promising basis for further exploration for the selective oxidation of a range of feed stocks.