alumina catalysts for the selective hydrogenation of propyne, propadiene and propene mixed feeds

Chemical Engineering Journal 285 (2016) 384–391

Contents lists available at ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

Palladium assisted copper/alumina catalysts for the selective hydrogenation of propyne, propadiene and propene mixed feeds Alan J. McCue a,⇑, Andrew Gibson a, James A. Anderson a,b,⇑ a b

Surface Chemistry and Catalysis Group, Department of Chemistry, University of Aberdeen, Aberdeen AB24 3UE, UK Materials and Chemical Engineering Group, School of Engineering, University of Aberdeen, Aberdeen AB24 3UE, UK

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 Pd assists Cu by providing a site for

hydrogen dissociation at low temperature.  Full propyne conversion observed at temperatures below 393 K.  High selectivity achievable with mixed feeds containing propyne/ propadiene/propene.  >85% propene selectivity achieved at 99% conversion at only 383 K.

a r t i c l e

i n f o

Article history: Received 18 June 2015 Received in revised form 14 September 2015 Accepted 29 September 2015

Keywords: Propyne Propadiene Propene Copper Palladium Selective hydrogenation

Propane

Propene

Cu

H

H

Pd

Pd

Pd

Pd

Pd

dissociation on Pd

Cu

Cu

H

H

Pd

Cu

Cu

Cu

Cu

Cu

a b s t r a c t A series of copper rich catalysts with different Cu:Pd atomic ratios were screened for the selective hydrogenation of propyne. Sample with 50:1 Cu:Pd ratio exhibited high propene selectivity, yet could be operated at temperatures far lower than typically observed for Cu only catalyst. It is believed that Pd facilitates reaction at Cu sites by promoting hydrogen dissociation at low temperature, followed by spillover onto Cu where the reaction occurs selectively. Catalyst testing with propyne alone showed that full conversion could be achieved at only 383 K with greater than 70% selectivity to propene. Industrially relevant tests were also conducted with a mixed C3 feed containing propyne, propadiene, propene and propane which is unique given that most literature studies fail to consider that propadiene is also an impurity which must be removed during selective hydrogenation of C3 cuts from naphtha crackers. Under such conditions and at only 383 K, propene selectivity of around 90% was achievable at >99% conversion. The option to operate at such low temperature, in the absence of CO, makes 50-CuPd sample an interesting alterative to current industrial catalysts. Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction Naphtha crackers produce alkene streams with contain alkyne and alkadiene impurities which need to be removed before the ⇑ Corresponding authors at: Surface Chemistry and Catalysis Group, Department of Chemistry, University of Aberdeen, Aberdeen AB24 3UE, UK (J.A. Anderson). Tel.: +44 1224 272905; fax: +44 1224 272901. E-mail addresses: [email protected] (A.J. McCue), [email protected] (J.A. Anderson). http://dx.doi.org/10.1016/j.cej.2015.09.118 1385-8947/Ó 2015 Elsevier B.V. All rights reserved.

H spillover onto Cu

H2

alkene can be utilised for polymerisation processes [1–3]. The preferred purification route is selective hydrogenation since this, in principle, converts an impurity into a valuable product. However avoiding over-hydrogenation is not simple and can rely upon the use of CO as a competitive adsorbate to hinder alkene adsorption which would otherwise lead to alkane formation. Whilst CO is a transient selectivity modifier, too high a concentration can reduce catalyst activity leading to alkyne/alkadiene slip. As such, it is necessary to regulate the amount of CO used in real time as catalyst

385

A.J. McCue et al. / Chemical Engineering Journal 285 (2016) 384–391

activity varies with lifetime, which is far from ideal. A number of studies have explored the reasons why a Pd catalyst may or may not be selective through both surface science and computational approaches. Results indicate that formation of a Pd-hydride phase [4–8] is detrimental to achieving high alkene selectivity and this in turn may be influenced by the role of Pd-carbides [7–10]. Importantly, alloying with a second metal such as Ag may hinder hydride formation leading to enhanced selectivity [11]. Industrial catalysts reflect these results and typically use low loaded PdAg catalysts (supported on 2–6 mm a-Al2O3 pellets) with Pd acting as the active site for hydrogenation at temperatures below 393 K [12]. Several strategies have been explored in an attempt to prepare more selective catalysts but improvements often rely on a compromise between enhanced selectivity and changes to process conditions (i.e., operation at lower pressure or operation/activation at higher temperature). For example, monometallic catalysts such as Cu [13–17], Ni [16,18–20], Ag [21,22] and Au [23–26] exhibit high inherent alkene selectivity but suffer from a significant energy barrier which limits hydrogen dissociation rates meaning that reactions must be performed at elevated temperatures (473–523 K). Similarly, bimetallic and trimetallic catalysts such as PdGa [27,28], AuAg [29], NiZn [30] and CuNiFe [31,32] offer high selectivity but require activation/use at high temperature. Metalfree CeO2 [33–35] has been shown to offer high alkene selectivity but limited activity at temperatures employed industrially unless doped with Ga or In [36]. Interestingly, CeO2 offers best performance when a large excess of H2 is used relative to alkyne (typical front-end conditions) whereas more industrial reactors now operate with a stoichiometric amount of hydrogen (typical back or tailend conditions). The performance of monometallic Pd catalysts has been shown to be greatly improved by the addition of organic sulphur [37–40] and phosphorous [38,40,41] based modifiers through a surface template effect which creates sites that favour alkyne adsorption but hinder alkene adsorption (although limiting hydride formation also appears significant). Such catalysts can operate at low temperature and retain high selectivity at 10 bar pressure [40] but only when CO is co-fed using triphenylphosphine as modifier [41]. Alternative methods of improving performance of Pd catalysts for acetylene hydrogenation have been recently reviewed [42]. Two other catalyst formulations – NiAu and CuPd demonstrate promise since they offer high selectivity and activity at industrially relevant temperatures. Nikolaev and co-workers have shown that low loaded NiAu/Al2O3 catalysts offer high activity/selectivity at temperatures around 357 K but no report of tests at higher pressures have been reported to date [23,43]. CuPd catalysts have been shown to offer excellent performance for acetylene hydrogenation at 373 K [44] under both non-competitive (no alkene co-fed) and competitive (alkene co-fed) conditions, although tests indicated a drop-off in performance was likely at elevated pressure [45]. In the case of CuPd catalysts it is thought that Pd acts to promote hydrogen dissociation with spillover onto Cu where hydrogenation takes place [46–49] – therefore marking these catalysts somewhat unique when compared with most bimetallic catalysts reported for alkyne hydrogenation where Pd acts as both the site for hydrogen activation and reaction (Note: a similar approach has been reported for AuPd catalysts [50]). Importantly both NiAu and CuPd catalysts demonstrated high alkene selectivity in the absence of CO, representing an improvement compared with PdAg catalysts. In the case of CuPd catalysts surface composition/catalytic performance could be improved further by using CO induced surface segregation [51]. In this study, the use of CuPd catalysts for purification of C3 alkyne and alkadiene streams is described. Simple single reagent (propyne) tests are used to discriminate between catalysts of different Cu:Pd atomic ratios. Subsequently, the best catalyst (50:1

Cu:Pd ratio) is tested under competitive conditions using a complex C3 mixture containing propyne, propadiene, propene and propane. This is somewhat unique since almost all studies involving propyne hydrogenation fail to consider the propadiene that is also present as an impurity in C3 streams produced from naphtha cracking.

2. Materials and methods 2.1. Synthesis of catalysts Bimetallic copper–palladium catalysts were prepared by a sequential impregnation method as described previously [44]. Briefly, 10% Cu/Al2O3 was prepared by impregnation using Cu (NO3)2
3H2O and Aeroxide Alu-C Al2O3 (Evonik, 100 m2 g 1) followed by calcination at 673 K. Bimetallic samples were prepared by addition of an appropriate amount of an aqueous solution of Pd(NO3)2 to Cu/Al2O3, followed by calcination at 673 K again, to yield materials with different Cu:Pd atomic ratios (Table 1). A monometallic 1.7% Pd/Al2O3 sample was prepared as a reference and contains the same nominal Pd loading as 10-CuPd sample. 2.2. Catalytic testing All catalyst testing was performed in a Microactivity reference reactor (PID Eng & Tech, supplied by Micromeritics) in a 9 mm inner diameter stainless steel reactor tube following sample reduction (1 h, 30% H2/N2, 323 K for 1.7% Pd/Al2O3, 523 K for Cu/Al2O3 and CuPd samples). Effluent gas from the reactor was sampled and analysed online using a Perkin-Elmer Clarus 580 GC equipped with an FID detector and a 30 m  0.53 mm elite alumina capillary column. In general, tests were conducted at 1 bar pressure, temperature varied in the range 323–398 K and 5 h time on stream was allowed at each temperature. All catalyst testing was conducted with fine powder samples (<250 lm size). Given that the particle size is significantly smaller than the reactor size, the thickness of any boundary layer should be negligible as determined by application of the Mears criterion [52]. The effect of internal mass transfer was considered using the Weisz–Prater criterion [53] which was met assuming the diffusivity of the gas mixture was at least 2.2  10 7 m2 s 1. It should be noted that no literature data on diffusivity was available for a comparable C3 mixture, although the necessary diffusivity falls well below the value reported by Asplund [54] for an acetylene/ethylene mixture (3.3  10 7 m2 s 1). Some data was determined at temperatures in excess of that necessary to achieve 100% conversion. This reaction data is excluded when calculating reaction rates or activation energy barriers since the rate will be limited by reactant availability. However data collected under these conditions is illustrative of the ability of the catalysts to avoid overhydrogenation and also mirrors the industrial process where high selectivity must be obtained at conversion in excess of 99.95%. Non-competitive conditions were used to screen catalysts using a mixture of 0.85% propyne/2.6% hydrogen/balance N2 to give a H2: propyne ratio of 3:1 and a space velocity of 24,000 h 1. Propyne

Table 1 Characteristics of metal loaded catalysts. Sample name

Cu/wt (%)

Pd/wt (%)

Cu:Pd atomic ratio

10% Cu 1.7% Pd

10 –

– 1.7

– –

10-CuPd 25-CuPd 50-CuPd 75-CuPd

10 10 10 10

1.7 0.7 0.3 0.2

10 25 50 75

386

A.J. McCue et al. / Chemical Engineering Journal 285 (2016) 384–391

Table 2 Propyne conversion and product selectivity over monometallic Cu and Pd catalysts under non-competitive conditions at 383 K. Catalyst

10% Cu 1.7% Pd

Propyne conversion (%)

Product selectivity (%)

9.9 100

Propene

Propane

Oligomers

32.3 0.0

6.6 83.9

61.1 16.1

conversion was calculated as the amount reacted divided by the amount introduced. Propene/propane selectivity was calculated as the amount formed divided by the amount of propyne reacted, correcting for traces of propene/propane present in the feed gas. Selectivity to oligomers was calculated based on a carbon balance. Tests under competitive conditions were conducted with a complex C3 mixture of 0.85% propyne/0.65% propadiene/10% propene/1.75% propane/4.5% hydrogen/balance N2 with both propyne

and propadiene considered as reactive components (i.e., the aim was to selectively hydrogenate both). For these tests, space velocity (24,000–96,000 h 1) and H2: reactive component (3–10:1) were varied. Conversion and selectivity were calculated in the same manner as described for non-competitive conditions, although since the feed gas contained a large excess of alkene, it proved more challenging to determine oligomer selectivity based purely on carbon balance. As such, results are presented in terms of ‘C3 product selectivity’.

3. Results and discussion 3.1. Catalyst structure The CuPd bimetallic catalysts used in this study have been extensively characterised with details reported elsewhere

(a) 100

100

90 80 70

Selectivity / %

Propyne conversion / %

80

60

40

60 50 40 30 20

20

10

10-CuPd

25-CuPd

50-CuPd

75-CuPd

10-CuPd

25-CuPd

50-CuPd

75-CuPd

75-CuPd

50-CuPd

25-CuPd

0 10-CuPd

0

(b) 100

100

90 80 70

Selectivity / %

Propyne conversion / %

80

60

40

60 50 40 30 20

20

10

75-CuPd

50-CuPd

25-CuPd

0 10-CuPd

0

Fig. 1. Propyne conversion and product selectivity for CuPd catalysts under non-competitive conditions at (a) 338 K and (b) 383 K. Propyne conversion (grey), propene selectivity (yellow), propane selectivity (cyan) and oligomer selectivity (pink). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

387

A.J. McCue et al. / Chemical Engineering Journal 285 (2016) 384–391 100 90

Conversion/Selectivity / %

[44,45,51]. However, for sake of clarity the results can be summarised as follows. The samples have a copper rich surface which reflects the large excess of Cu relative to Pd with a crystallite size of 22–24 nm estimated from XRD of calcined sample [44]. XPS results suggest that very little Pd exists within the surface layers following reduction, although the presence of Pd is confirmed by temperature programmed reduction [44,45]. For example, additional features are apparent which can be attributed to Pd and there is a decrease in the apparent temperature necessary to reduce Cu, consistent with enhanced hydrogen dissociation. FTIR of adsorbed CO confirms the absence of large Pd ensembles suggesting no monometallic Pd particles exist. Instead, surface Pd is likely to exist as isolated Pd atoms or Pd–Pd dimers with neighbouring Cu surface atoms. The exact distribution of Pd surface atoms depends on Pd loading (higher loadings favour Pd–Pd neighbouring) with a correlation found between the presence of Pd–Pd surface dimers and over-hydrogenation during acetylene testing [51].

80 70 60

5h

50 40 30 20 10 0

323

338

353

368

383

Temperature / K

In order to access the potential of CuPd catalysts for selective hydrogenation of propyne and propadiene, tests were first conducted under non-competitive conditions with propyne as the sole reagent. Using a 3-fold excess of H2 relative to propyne, monometallic Pd is exceptionally active (Table 2), although this leads to extensive over-hydrogenation with oligomers as minor product. Under identical conditions (Table 2), monometallic Cu is considerably less active (<10% conversion) but does produce much more propene than propane. However, the main product produced is oligomers with greater than 60% selectivity. This is not uncommon for Cu catalysts which are prone to extensive oligomer formation, most notably under conditions where hydrogen dissociation is limited (i.e., low temperature). For example, Bridier et al. [16] reported that copper hydrotalcite catalysts produced oligomers with greater than 70% selectivity at 423 K which decreased to ca. 25% at 523 K under otherwise similar conditions. Overall, the results of tests performed over the two monometallic catalysts highlight that Pd is very active but unselective, whereas Cu offers poor activity but significantly less over-hydrogenation. In order to identify the optimum Cu:Pd ratio, samples were screened for performance in the range 323–383 K. Propyne conversion and product selectivity at 338 and 383 K are presented in Fig. 1. 10-CuPd sample is highly active with 100% conversion observed at temperatures as low as 338 K. However, the product distribution strongly favours propane suggesting that the reaction is primarily occurring on a ‘Pd like’ surface (compare with Table 2). The 25-CuPd sample also shows high activity but with a reduced tendency to form propane. As temperature is increased for this sample, propane selectivity increases which implies the reaction is taking place on a surface strongly influenced by Pd. 50-CuPd catalyst attains 66% conversion at 338 K with 64% selectivity to propene and 33% selectivity to oligomers. The combination of reduced activity and higher alkene/oligomer selectivity is consistent with the reaction taking place on a Cu surface. Importantly, the activity is considerably higher than on 10% Cu sample which suggests that hydrogen dissociation is more facile on 50-CuPd sample. Similarly, reduced oligomer formation (33% for 50-CuPd vs 61% for Cu) also suggests hydrogen dissociation is more favoured on the bimetallic sample. This is interpreted in terms of Pd acting as a site for hydrogen dissociation with spillover onto neighbouring Cu sites, allowing the reaction to occur on a Cu surface at lower temperature than a Cu only surface (i.e., Pd assists Cu) – consistent with recent surface science [46–48] and DFT studies [49]. Similar trends are apparent with 75-CuPd sample, although the maximum conversion achieved at 383 K is only 64%. At 338 K 50-

Fig. 2. Time on stream data for 50-CuPd under non-competitive conditions at temperatures between 323 and 383 K. Propyne conversion (grey squares), propene selectivity (yellow circles), propane selectivity (cyan triangles) and oligomer selectivity (pink diamonds). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

CuPd and 75-CuPd produce propylene yields (defined as: conversion  propylene selectivity) of 42% and 9%, respectively. The apparent activation energy barrier over 10% Cu was calculated from an Arrhenius plot as 81 kJ mol 1, assuming propyne hydrogenation to be first order in hydrogen and zero in propyne [55]. This figure compares reasonably well with the barrier reported by Bridier et al. (75 kJ mol 1) for a Cu (111) slab calculated by DFT (Note: this barrier corresponds to addition of the first hydrogen atom to an adsorbed propyne molecule) [56]. The apparent activation energy barrier over 50-CuPd and 75-CuPd were determined to be 44 and 59 kJ mol 1 respectively, confirming that the reaction should proceed at lower temperatures over bimetallic samples. Given that the aim of the work was to identify a catalyst which offered high alkene selectivity at as low a temperature as possible, it is considered that 50-CuPd sample offers the greatest promise.

100 90

Conversion/Selectivity / %

3.2. Catalytic performance with propyne

80 70 60 50 40 30 20 10 0

323

338

353

368

383

Temperature / K Fig. 3. Conversion and C3 product selectivity for 50-CuPd under competitive conditions at temperatures between 323 and 383 K, SV = 24,000 h 1 and 3:1 H2: (propyne + propadiene) ratio. Propyne conversion (grey squares + line), propadiene conversion (grey stars + line), propene selectivity (yellow bar) and propane selectivity (cyan bar). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

388

A.J. McCue et al. / Chemical Engineering Journal 285 (2016) 384–391

be achieved at higher temperatures. This result is somewhat counter-intuitive but may be rationalised as follows. At low temperature the reaction is thought to occur on Cu sites with hydrogen dissociation occurring on Pd sites. As temperature is increased, hydrogen can dissociate on Cu as well as via Pd sites resulting in increased hydrogen availability and a reduced tendency to form oligomers. 3.3. Catalyst testing with propyne, propadiene, propene and propane mixtures Since 50-CuPd sample demonstrated promising performance for low temperature propyne hydrogenation under simple conditions, the study was extended to a complex C3 mixture much more

100

100

90

90

80

80

70

70

Selectivity / %

Conversion / %

Time on stream analysis was performed for propyne hydrogenation over 50-CuPd sample (Fig. 2) in order to access stability. Generally, Cu catalysts are prone to extensive oligomer formation leading to deactivation [14], however activity over 50-CuPd appears to be relatively stable. The only evidence of minor deactivation occurs at 368 K with conversion decreasing from 99.7 to 98.8% over 5 h on stream. Increased stability with respect to pure copper is likely caused by decreased rate of oligomer formation associated with higher hydrogen availability. Product selectivity appears to be fairly constant, irrespective of temperature with propane selectivity remaining remarkably low (4% at 323 K vs 6% at 383 K). Interestingly, as temperature is increased propene selectivity actually increases suggesting that whilst it is desirably to operate at as low a temperature as possible, optimum alkene yield may

60 50 40

60 50 40

30

30

20

20

10

10

0

1.5

3

6

0

10

1.5

3

6

10

Hydrogen equivalents

100

100

90

90

80

80

70

70

Selectivity / %

Conversion / %

Fig. 4. Conversion and C3 product selectivity for 50-CuPd under competitive conditions at H2:(propyne + propadiene) ratios between 1.5 and 10, SV = 48,000 h 1 and 338 K. Propyne conversion (light grey bar), propadiene conversion (dark grey bar), propene selectivity (yellow bar) and propane selectivity (cyan bar). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

60 50 40

60 50 40

30

30

20

20

10

10

0

24000

48000

96000

0

24000

48000

96000

Space velocity / h-1 Fig. 5. Conversion and C3 product selectivity for 50-CuPd under competitive conditions at space velocities between 24,000 and 96,000 h 1, 353 K and 3:1 H2:(propyne + propadiene) ratio. Propyne conversion (light grey bar), propadiene conversion (dark grey bar), propene selectivity (yellow bar) and propane selectivity (cyan bar). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

389

A.J. McCue et al. / Chemical Engineering Journal 285 (2016) 384–391 100 90 80

Conversion/Selectivity / %

akin to that encountered in industry (i.e., alkyne, alkene and alkane all present in the feed gas). Importantly, propadiene was also included in the reactant mixture. Propadiene is an additional impurity present in C3 streams which needs to be removed although this additional complexity is generally neglected in academic literature. It should be noted that the feed gas contains a large excess of propene (10%) which meant that oligomer selectivity could not confidently be determined by carbon balance. Instead, results are presented in terms of ‘C3 product selectivity’ since these values could be determined with certainty. Fig. 3 displays propyne/ propadiene conversion and propene/propane selectivity for 50-CuPd under equivalent conditions to those used for ‘propyne only’ tests (i.e., SV = 24,000 h 1, 3 equiv. H2 relative to propyne + propadiene). Conversion of both components increased as temperature increased, although propyne conversion always exceeded propadiene conversion. Interestingly, 100% conversion of both impurities was first observed at 383 K which correlates well with the single reagent experiment with propyne (Fig. 2). In terms of selectivity, propene selectivity is exceptionally good (88%) at low temperature when only moderate conversion (38% propyne, 18% propadiene) is achieved. As conversion/temperature increased, more propane was observed although at 383 K propene was still the major product (58% selectivity). In agreement with propyne only reactions, alkene selectivity actually increased at higher temperature (63% at 398 K – not shown). Overall, this test suggests that the promising performance of 50-CuPd can be extended to mixed C3 streams. Figs. 4 and 5 show the effect of hydrogen partial pressure and space velocity, respectively, on conversion and selectivity at fixed temperature. Increasing hydrogen partial pressure results in higher conversion of both components (Fig. 4). This is not surprising for a reaction which is generally thought to be first order with respect to hydrogen partial pressure [55]. Increasing the amount of hydrogen from 1.5 to 6 equivalents has a relatively small impact on selectivity with propene selectivity decreasing from 94% to 86%. Operation with a 10-fold excess of H2 results in high conversion at low temperature but extensive over-hydrogenation. As a result it can be concluded that the operating window of 50-CuPd is limited to H2: impurity ratios of <10. Changing the space velocity from 24,000 to 96,000 h 1, results in decreased conversion but increased propene selectivity (Fig. 5). For example, propylene selectivity increases from 65% to 90% as space velocity is increased from 24,000 to 96,000 h 1. In order to verify if enhanced selectivity at higher space velocity was simply an effect of decreased conversion, a reaction was performed using a space velocity of 96,000 h 1 but over a wider range of temperatures (Fig. 6). Interestingly, propene/ propane selectivity remains remarkably constant as temperature/conversion is increased suggesting that alkene selectivity is enhanced by increasing space velocity. For example, at 323 K, propyne + propadiene conversion is only 8% but with 90% selectivity to propene whereas at 383 K, conversion is 98.5% but with alkene selectivity of 86%. Recent literature studies have reported that the use of higher space velocities/shorter contact times diminish oligomer formation [29,50,57]. It is possible that a similar effect is observed here, although under the conditions employed it was not possible to quantify oligomer selectivity under competitive conditions. A subtle but important point to note is that at 383 K conversion decreases from 99.3% to 98.5% over 5 h TOS stream suggesting that the catalyst suffers from a small degree of deactivation. Using the data shown in Fig. 6 it was again possible to calculate an apparent activation energy barrier from an Arrhenius plot. The barrier limiting propyne hydrogenation was calculated to be 53 kJ mol 1 which was higher than under non-competitive conditions (44 kJ mol 1). This is perhaps not surprising given that propadiene and propene (to a lesser extent) will compete for adsorption

70 60

5h

50 40 30 20 10 0

323

338

353

368

383

Temperature / K Fig. 6. Time on stream data for 50-CuPd under competitive conditions at temperatures between 323 and 383 K, SV = 96,000 h 1 and 3:1 H2:(propyne + propadiene) ratio. Combined propyne and propadiene conversion (grey squares), propene selectivity (yellow circles), propane selectivity (cyan triangles) and oligomer selectivity (pink diamonds). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

sites. The equivalent activation energy barrier for propadiene hydrogenation was found to be 46 kJ mol 1. Finally it was possible to estimate the turnover frequency (TOF) of 50-CuPd for both propyne and propadiene hydrogenation. It is thought that the active reaction site is Cu – the surface area of which can be determined from N2O decomposition and this been previously determined for 10-CuPd sample [51]. Since both 10-CuPd and 50-CuPd were prepared from the same parent batch of 10% Cu/Al2O3 and the crystallite size for both samples as judged by XRD [44] is equivalent it seems reasonable to assume the Cu surface is similar for both samples. Therefore, on this basis the TOF for propyne and propadiene were estimated as 0.050 and 0.019 s 1, respectively for 50-CuPd.

3.4. CuPd catalysts vs other catalysts Table 3 summarises the performance of catalysts reported in literature to be effective for the selective hydrogenation of propyne under non-competitive conditions. The group of Pérez-Ramírez have demonstrated that a number of catalysts exhibit greater than 90% selectivity to propene (Table 3: entries 1–4). CeO2 and Ag operate most efficiently with a large excess of hydrogen since this limits oligomer formation and enhances rate (H2 dissociation is the rate limiting step). Whilst high propene selectivity at higher hydrogen concentrations is impressive, this deviates from the majority of industrial reactors which operate with a stoichiometric amount of hydrogen. Similarly, operation at higher temperature is costly and may require modification to industrial reactors limiting widespread application. In this regards Ga promoted ceria offers the greatest promise (Table 3: entry 2) since it can operate at relatively low temperatures. Out of the catalysts which demonstrate optimum performance using a small excess of hydrogen, only the 50CuPd catalyst described in this work can operate at industrially relevant temperatures – albeit with a slightly decreased propene selectivity compared with other literature reports. It should be noted that the majority of the most selective catalysts from noncompetitive tests (Table 3: entries 1–6) have not been tested under conditions where both propene and propadiene are present. This is in contrast to 50-CuPd which has been shown to be selective under complex conditions more akin to those encountered industrially. In addition, it is important to note that excellent alkene selectivity

390

A.J. McCue et al. / Chemical Engineering Journal 285 (2016) 384–391

Table 3 Operating conditions and catalytic performance of catalysts reported in literature for propyne hydrogenation under non-competitive conditions. Entry

Catalyst

1 2 3 4 5 6 7 8

Cu–Ni–Fe Ga promoted CeO2 Ag CeO2 Ni–Al hydrotalcite Cu–Al hydrotalcite Cu 50-CuPd

Optimum operating conditions H2: propyne ratio

Temperature (K)

3 30 25 30 3 3 4 3

523 373 473 523 473 523 473 383

can be obtained without the use of CO as a competitive adsorbate – therefore representing a significant advantage over traditional PdAg catalysts. 4. Conclusions By optimising the Cu:Pd ratio of Cu rich catalysts, it is possible to prepare a bimetallic catalyst which offers high propene selectivity during the hydrogenation of propyne/propadiene. It is thought that Cu acts as the active site (offering high inherent selectivity) whereas Pd assists by providing a site for hydrogen dissociation, therefore allowing for operation at industrially useful temperatures. Under non-competitive conditions propene selectivity of 70% could be achieved at 383 K with propane selectivity of only 6%. Performance extended to mixed feeds containing propyne, propadiene and propene with 98.5% conversion of propyne and propadiene achievable at 383 K with 86% propene selectivity. It is thought that 50-CuPd offers an interesting alternative to industrial catalysts if longer term stability was demonstrated. Acknowledgement We thank the University of Aberdeen for financial support. References [1] A. Borodzin´ski, G.C. Bond, Selective hydrogenation of ethyne in ethene-rich streams on palladium catalysts. Part 1. Effect of changes to the catalyst during reaction, Catal. Rev. 48 (2006) 91–144. [2] A. Borodzin´ski, G.C. Bond, Selective hydrogenation of ethyne in ethene-rich streams on palladium catalysts, part 2: steady-state kinetics and effects of palladium particle size, carbon monoxide, and promoters, Catal. Rev. 50 (2008) 379–469. [3] S.A. Nikolaev, I.L.N. Zanaveskin, V.V. Smirnov, V.A. Averyanov, K.I. Zanaveskin, Catalytic hydrogenation of alkyne and alkadiene impurities from alkenes. Practical and theoretical aspects, Russ. Chem. Rev. 78 (2009) 231–247. [4] D. Teschner, J. Borsodi, A. Wootsch, Z. Révay, M. Hävecker, A. Knop-Gericke, S. D. Jackson, R. Schlögl, The roles of subsurface carbon and hydrogen in palladium catalyzed alkyne hydrogenation, Science 320 (2008) 86–89. [5] J. Sá, G.D. Arteaga, R.A. Daley, J. Bernardi, J.A. Anderson, Factors influencing hydride formation in a Pd/TiO2 catalyst, J. Phys. Chem. B 110 (2006) 17090– 17095. [6] M. Armbrüster, M. Behrens, F. Cinquini, K. Föttinger, Y. Grin, A. Haghofer, B. Klötzer, A. Knop-Gericke, H. Lorenz, A. Ota, S. Penner, J. Prinz, C. Rameshan, Z. Révay, D. Rosenthal, G. Rupprechter, D. Teschner, D. Torres, R. Wagner, R. Widmer, G. Wowsnick, How to control the selectivity of palladium-based catalysts in hydrogenation reactions: the role of subsurface chemistry, ChemCatChem 4 (2012) 1048–1063. [7] M. García-Mota, B. Bridier, J. Pérez-Ramírez, N. López, Interplay between carbon monoxide, hydrides, and carbides in selective alkyne hydrogenation on palladium, J. Catal. 273 (2010) 92–102. [8] B. Yang, R. Burch, C. Hardacre, G. Headdock, P. Hu, Influence of surface structures, subsurface carbon and hydrogen, and surface alloying on the activity and selectivity of acetylene hydrogenation on Pd surfaces: a density functional theory study, J. Catal. 305 (2013) 264–276. [9] W. Ludwig, A. Savara, K.H. Dostert, S. Schauermann, Olefin hydrogenation on Pd model supported catalysts: new mechanistic insights, J. Catal. 284 (2011) 148–156.

Propyne conversion (%)

Propene selectivity (%)

References

100 – 92 96 80 100

>95 95 93 91 80 75 70–82 70

[31] [36] [21] [33] [20] [16] [14] This work

100

[10] W. Ludwig, A. Savara, R.J. Madix, S. Schauermann, H.J. Freund, Subsurface hydrogen diffusion into Pd nanoparticles: role of low-coordinated surface sites and facilitation by carbon, J. Phys. Chem. C 116 (2012) 3539–3544. [11] N.A. Khan, S. Shaikhutdinov, H.J. Freund, Acetylene and ethylene hydrogenation on alumina supported Pd–Ag model catalysts, Catal. Lett. 108 (2006) 159–164. [12] D.B. Tiedtke, T.T.P. Cheung, L. Leger, S.A. Zisman, J.J. Bergmeister, G.A. Delzer, Chemicals influencing the activity of palladium-based catalysts for the selective hydrogenation of acetylene to ethylene in acetylene converters, in: 13th Ethylene Producers Conference, 10 (2001) 1–21. [13] J.T. Wehrli, D.J. Thomas, M.S. Wainwright, D.L. Trimm, N.W. Cant, Selective hydrogenation of propyne over an ion-exchanged copper on silica catalyst, Appl. Catal. 66 (1990) 199–208. [14] J.T. Wehrli, D.J. Thomas, M.S. Wainwright, D.L. Trimm, N.W. Cant, Selective hydrogenation of propyne over supported copper catalysts: influence of support, Appl. Catal. 70 (1991) 253–262. [15] J. Ossipoff, N.W. Cant, The hydrogenation and oligomerisation of propyne over an ion-exchanged copper on silica catalyst, J. Catal. 148 (1994) 123–133. [16] B. Bridier, N. López, J. Pérez-Ramírez, Partial hydrogenation of propyne over copper-based catalysts and comparison with nickel-based analogues, J. Catal. 269 (2010) 80–92. [17] B. Bridier, M.A.G. Hevia, N. López, J. Pérez-Ramírez, Permanent alkene selectivity enhancement in copper-catalyzed propyne hydrogenation by temporary CO supply, J. Catal. 278 (2011) 167–172. [18] D.L. Trimm, I.O.Y. Liu, N.W. Cant, The selective hydrogenation of acetylene over a Ni/SiO2 catalyst in the presence and absence of carbon monoxide, Appl. Catal. A Gen. 374 (2010) 58–64. [19] D.L. Trimm, N.W. Cant, I.O.Y. Liu, The selective hydrogenation of acetylene in the presence of carbon monoxide over Ni and Ni–Zn supported on MgAl2O4, Catal. Today 178 (2011) 181–186. [20] S. Abelló, D. Verboekend, B. Bridier, J. Pérez-Ramírez, Activated takovite catalysts for partial hydrogenation of ethyne, propyne, and propadiene, J. Catal. 259 (2008) 85–95. [21] G. Vilé, D. Baudouin, I.N. Remediakis, C. Copéret, N. López, J. Pérez-Ramírez, Silver nanoparticles for olefin production: new insights into the mechanistic description of propyne hydrogenation, ChemCatChem 5 (2013) 3750–3759. [22] E. Oakton, G. Vilé, D.S. Levine, E. Zocher, D. Baudouin, J. Pérez-Ramírez, C. Copéret, Silver nanoparticles supported on passivated silica: preparation and catalytic performance in alkyne semi-hydrogenation, Dalton Trans. 43 (2014) 15138–15142. [23] S.A. Nikolaev, V.V. Smirnov, Synergistic and size effects in selective hydrogenation of alkynes on gold nanocomposites, Catal. Today 147S (2009) S336–S341. [24] J. Jia, K. Haraki, J.N. Kondo, K. Domen, K. Tamaru, Selective hydrogenation of acetylene over a MoP catalyst, J. Phys. Chem. B 104 (2000) 11153–11156. [25] J.A. Lopez-Sanchez, D. Lennon, The use of titania- and iron oxide-supported gold catalysts for the hydrogenation of propyne, Appl. Catal. A Gen. 291 (2005) 230–237. [26] Y. Segura, N. López, J. Pérez-Ramírez, Origin of the superior hydrogenation selectivity of gold nanoparticles in alkyne + alkene mixtures: triple-versus double-bond activation, J. Catal. 247 (2007) 383–386. [27] J. Osswald, R. Giedigkeit, R.E. Jentoft, M. Armbrüster, F. Girgsdies, K. Kovnir, T. Ressler, Y. Grin, R. Schlögl, Palladium–gallium intermetallic compounds for the selective hydrogenation of acetylene, J. Catal. 258 (2008) 210–218. [28] J. Osswald, K. Kovnir, M. Armbrüster, R. Giedigkeit, R.E. Jentoft, U. Wild, Y. Grin, R. Schlögl, Palladium–gallium intermetallic compounds for the selective hydrogenation of acetylene. Part II: surface characterization and catalytic performance, J. Catal. 258 (2008) 219–227. [29] X. Liu, Y. Li, J.W. Lee, C.Y. Hong, C.Y. Mou, B.W.L. Jang, Selective hydrogenation of acetylene in excess ethylene over SiO2 supported Au–Ag bimetallic catalyst, Appl. Catal. A Gen. 439–440 (2012) 8–14. [30] F. Studt, F. Abild-Pedersen, T. Bligaard, R.Z. Sørensen, C.H. Christensen, J.K. Nørskov, On the role of surface modifications of palladium catalysts in the selective hydrogenation of acetylene, Angew. Chem. Int. Ed. 47 (2008) 9299– 9302. [31] B. Bridier, J. Pérez-Ramírez, Cooperative effects in ternary Cu Ni Fe catalysts lead to enhanced alkene selectivity in alkyne hydrogenation.pdf, J. Am. Chem. Soc. 132 (2010) 4321–4327.

A.J. McCue et al. / Chemical Engineering Journal 285 (2016) 384–391 [32] B. Bridier, J. Pérez-Ramírez, A. Knop-Gericke, R. Schlögl, D. Teschner, Surface state during activation and reaction of high-performing multi-metallic alkyne hydrogenation catalysts, Chem. Sci. 2 (2011) 1379–1383. [33] G. Vilé, B. Bridier, J. Wichert, J. Pérez-Ramírez, Ceria in hydrogenation catalysis: high selectivity in the conversion of alkynes to olefins, Angew. Chem. Int. Ed. 51 (2012) 8620–8623. [34] G. Vilé, S. Colussi, F. Krumeich, A. Trovarelli, J. Pérez-Ramírez, Opposite face sensitivity of CeO2 in hydrogenation and oxidation catalysis, Angew. Chem. Int. Ed. 53 (2014) 12069–12072. [35] J. Carrasco, G. Vilé, D. Fernández-Torre, R. Pérez, J. Pérez-Ramírez, Molecularlevel understanding of CeO2 as a catalyst for partial alkyne hydrogenation, J. Phys. Chem. C 118 (2014) 5352–5360. [36] G. Vilé, P. Dähler, J. Vecchietti, M. Baltanás, S. Collins, M. Calatayud, A. Bonivardi, J. Pérez-Ramírez, Promoted ceria catalysts for alkyne semihydrogenation, J. Catal. 324 (2015) 69–78. [37] A.J. McCue, J.A. Anderson, Sulfur as a catalyst promoter or selectivity modifier in heterogeneous catalysis, Catal. Sci. Technol. 4 (2014) 272–294. [38] F.M. McKenna, R.P.K. Wells, J.A. Anderson, Enhanced selectivity in acetylene hydrogenation by ligand modified Pd/TiO2 catalysts, Chem. Commun. 47 (2011) 2351–2353. [39] F.M. McKenna, J.A. Anderson, Selectivity enhancement in acetylene hydrogenation over diphenyl sulphide-modified Pd/TiO2 catalysts, J. Catal. 281 (2011) 231–240. [40] F.M. McKenna, L. Mantarosie, R.P.K. Wells, C. Hardacre, J.A. Anderson, Selective hydrogenation of acetylene in ethylene rich feed streams at high pressure over ligand modified Pd/TiO2, Catal. Sci. Technol. 2 (2012) 632–638. [41] A.J. McCue, F.M. McKenna, J.A. Anderson, Triphenylphosphine: a ligand for heterogeneous catalysis too? Selectivity enhancement in acetylene hydrogenation over modified Pd/TiO2 catalyst, Catal. Sci. Technol. 5 (2015) 2449–2459. [42] A.J. McCue, J.A. Anderson, Recent advances in selective acetylene hydrogenation using palladium containing catalysts, Front. Chem. Sci. Eng. 9 (2015) 142–153. [43] S.A. Nikolaev, V.V. Smirnov, A.Y. Vasil’kov, V.L. Podshibikhin, Synergism of the catalytic effect of nanosized gold–nickel catalysts in the reaction of selective acetylene hydrogenation to ethylene, Kinet. Catal. 51 (2010) 375–379. [44] A.J. McCue, C.J. McRitchie, A.M. Shepherd, J.A. Anderson, Cu/Al2O3 catalysts modified with Pd for selective acetylene hydrogenation, J. Catal. 319 (2014) 127–135. [45] A.J. McCue, A.M. Shepherd, J.A. Anderson, Optimisation of preparation method for Pd doped Cu/Al2O3 catalysts for selective acetylene hydrogenation, Catal. Sci. Technol. 5 (2015) 2880–2890.

391

[46] H.L. Tierney, A.E. Baber, J.R. Kitchin, E.C.H. Sykes, Hydrogen dissociation and spillover on individual isolated palladium atoms, Phys. Rev. Lett. 103 (2009). 246102–1–246102–4. [47] G. Kyriakou, M.B. Boucher, A.D. Jewell, E.A. Lewis, T.J. Lawton, A.E. Baber, H.L. Tierney, M. Flytzani-Stephanopoulos, E.C.H. Sykes, Isolated metal atom geometries as a strategy for selective heterogeneous hydrogenations, Science 335 (2012) 1209–1212. [48] M.B. Boucher, B. Zugic, G. Cladaras, J. Kammert, M.D. Marcinkowski, T.J. Lawton, E.C.H. Sykes, M. Flytzani-Stephanopoulos, Single atom alloy surface analogs in Pd0.18Cu15 nanoparticles for selective hydrogenation reactions, Phys. Chem. Chem. Phys. 15 (2013) 12187–12196. [49] Q. Fu, Y. Luo, Active sites of Pd-doped flat and stepped Cu(111) surfaces for H2 dissociation in heterogeneous catalytic hydrogenation, ACS Catal. 3 (2013) 1245–1252. [50] G.X. Pei, X.Y. Liu, A. Wang, L. Li, Y. Huang, T. Zhang, J.W. Lee, B.W.L. Jang, C.-Y. Mou, Promotional effect of Pd single atoms on Au nanoparticles supported on silica for the selective hydrogenation of acetylene in excess ethylene, New J. Chem. 38 (2014) 2043–2051. [51] A.J. McCue, J.A. Anderson, CO induced surface segregation as a means of improving surface composition and enhancing performance of CuPd bimetallic catalysts, J. Catal. 329 (2015) 538–546. [52] D.E. Mears, Tests for transport limitations in experimental catalytic reactors, Ind. Eng. Chem. Process Des. Dev. 10 (1971) 541–547. [53] P.B. Weisz, C.D. Prater, Interpretation of measurements in experimental catalysis, Adv. Catal. 6 (1954) 143–196. [54] S. Asplund, Coke formation and its effect on internal mass transfer and selectivity in Pd-catalysed acetylene hydrogenation, J. Catal. 158 (1996) 267– 278. [55] J.C. Fajardo, C. Godínez, A.L. Cabanes, G. Villora, Kinetic analysis of rate data from propylene and methylacetylene hydrogenation, Chem. Eng. Process. 35 (1996) 203–211. [56] B. Bridier, N. Lopez, J. Pérez-Ramírez, Molecular understanding of alkyne hydrogenation for the design of selective catalysts, Dalton Trans. 39 (2010) 8412–8419. [57] X.Y. Liu, C.Y. Mou, S. Lee, Y. Li, J. Secrest, B.J.W. Jang, Room temperature O2 plasma treatment of SiO2 supported Au catalysts for selective hydrogenation of acetylene in the presence of large excess of ethylene, J. Catal. 285 (2012) 152–159.