Sage Journals: Discover world-class research

Abstract

Neodymium (Nd)- or yttrium (Y)- modified bulk Ni₂P catalysts (Nd-Ni₂P or Y-Ni₂P) have been successfully prepared and their catalytic performance in benzofuran hydrodeoxygenation have been investigated. The as-prepared catalysts were characterised by X-ray diffraction, N₂ adsorption–desorption, CO uptake and X-ray photoelectron spectroscopy. The addition of Nd or Y, especially Nd, can increase the surface area of the catalysts and promote the formation of smaller and more highly dispersed Ni₂P particles. The Nd-Ni₂P catalyst showed the highest benzofuran hydrodeoxygenation activity of 95.3% and the O-free products yield of 74.6%, which gives an increase of 25.3% and 35.4% when compared with that found for Ni₂P.

Keywords

hydrodeoxygenation bio-oil neodymium (Nd)yttrium (Y)

Introduction

Bio-oil is an abundant and inexpensive renewable resource, which has attracted much attention as an alternative to fossil fuels.¹ However, highly oxygenated compounds degrade the chemical and physical stability of bio-oil, so a hydrodeoxygenation (HDO) process is needed.² A number of catalysts have been developed and investigated for HDO,³ among which, Ni₂P was the most promising catalyst with unique chemical and physical properties in the HDO reaction.⁴

The activity of Ni₂P catalysts has been found to be influenced by promoters. Aluminium (Al),⁵ citric acid (CA),⁶ boron (B),⁷ titanium (Ti)⁸ and platinum (Pt)⁹ have been reported as efficient promoters. The rare-earth metals with unfilled 4f electron shells exhibit outstanding optical, chemical, physical and electronic properties.¹⁰ Hence, the rare-earth metals have been adopted and studied as promoters to enhance catalysts.¹¹ To our knowledge, the study of rare-earth metals, such as Nd and Y, to modify bulk Ni₂P for the HDO reaction has not been reported. Therefore, it was meaningful to explore the effect of Nd and Y on the HDO performance over Ni₂P catalysts.

In this article, Nd or Y were successfully incorporated into bulk Ni₂P catalysts, and the effects of Nd or Y on the catalytic properties for benzofuran (BF) HDO were investigated.

Experimental

Preparation of bulk Ni₂P, Nd-Ni₂P and Y-Ni₂P catalysts

A solution of Ni(NO₃)₂ together with (NH₄)₂·HPO₄ was stirred and evaporated (initial P/Ni molar ratio of 1). The solid obtained was calcined at 500°C for 3 h to an oxidic Ni₂P precursor.

Nd or Y were incorporated by impregnating the oxidic precursor with Nd(NO₃)₃ or Y(NO₃)₃ solution, respectively, and the molar fraction of Nd or Y to Ni₂P was 0.1. The resulting solid was dried and calcined at 500°C for 3 h to obtain Nd-Ni₂P or Y-Ni P catalyst precursors.

The bulk Ni₂P, Nd-Ni₂P and Y-Ni₂P catalysts were prepared by reducing the precursors at 500°C for 2 h in flowing H₂, then cooling naturally to room temperature and passivating in an O₂/N₂ mixture (0.5 vol% O₂) for 1 h. The bulk catalysts obtained are denoted as Ni₂P, Nd-Ni₂P and Y-Ni₂P, respectively.

Characterisation of catalysts

The crystal structures were determined by X-ray diffraction (XRD) analysis, which was carried out with a D/max-2200PC-X-ray diffractometer using CuKα radiation. The textural properties were analysed using Micromeritics adsorption equipment type NOVA2000e. Carbon monoxide (CO) uptakes were performed with a Micromeritics ASAP 2010 installation under static volumetric conditions. X-ray photoelectron spectroscopy (XPS) patterns were obtained with an ESCALAB MKII spectrometer with Mg radiation (E = 1253.6 eV) and a hemispherical analyzer.

Catalytic activities

The BF HDO was performed in a fixed-bed reactor using a feed consisting of a decalin solution of BF (2 wt%). The HDO conditions were 3.0 MPa, 300°C, weight hourly space velocity (WHSV) = 4.0 h^–1 and an H₂/Oil ratio of 500 (V/V). The feed and reaction products were analysed by flame ionisation detector (FID) – gas chromatography with a GC-14C-60 column.

Results and discussion

X-Ray Diffraction

The XRD patterns of the catalysts are exhibited in Figure 1. For all catalysts, diffraction peaks at 2θ = 40.6°, 44.5°, 47.1° and 54.1° (powder diffraction file: 03-0953) were assigned to the Ni₂P phase. For bulk Ni₂P catalyst, diffraction peaks at 2θ = 28.8°, 30.2°, 31.6°, 43.9°, 47.8° and 53.0° (PDF: 18-0883) can be ascribed to the Ni₅P₄ phase. No diffraction peaks could be seen at the corresponding positions for Nd-Ni₂P and Y-Ni₂P, which indicated that the addition of Nd or Y can restrain the formation of the Ni₅P₄ phase. The average sizes of Ni₂P crystallites were calculated by the Scherrer equation (column 5 of Table 1). For Ni₂P catalyst, the average size of Ni₂P crystallites was 54 nm. Compared with Ni₂P catalyst, the average Ni₂P crystallite sizes for Nd-Ni₂P (39 nm) and Y-Ni₂P (21 nm) were smaller. This result may be caused by the strong interaction between nickel metal and certain reducible oxides, which constrain the growth of the nickel particle size.¹²

Figure 1.

XRD patterns of the Ni₂P, Nd-Ni₂P and Y-Ni₂P catalysts.

Table 1.

Textural characterisation of Ni₂P, Nd-Ni₂P and Y-Ni₂P catalysts.

Sample	S_BET (m² g^–1)	V_p (cm³ g^–1)	d (nm)	D_c (nm)	CO uptake (μmol g^–1)
Ni₂P	11.2	0.060	42.1	54	171
Nd-Ni₂P	25.6	0.129	18.7	39	326
Y-Ni₂P	16.0	0.086	20.1	21	314

Textural properties

Table 1 shows the textural properties of the catalysts. The surface area of bulk Ni₂P was 11.2 m² g⁻¹ with a pore volume of 0.060 cm³ g⁻¹. Upon addition of Nd and Y, the surface area and pore volume were increased. What is worth noting is that the Nd-Ni₂P catalyst displayed twofold higher surface area (25.6 m² g⁻¹) and pore volume (0.129 cm³ g⁻¹) compared with bulk Ni₂P catalyst. The result can be caused by the diminution in enrichment of phosphorus on the surface of the catalyst. This is discussed in X-ray photoelectron spectroscopy analysis.

CO uptake

CO uptakes were used to quantify the nickel atoms on the catalyst surface.¹³ The measured CO uptakes are listed in column 6 of Table 1. The CO uptake of bulk Ni₂P was 171 μmol g⁻¹, which is very low. This may be caused by the lower surface area and the accumulation of phosphate. Upon addition of Nd or Y, the CO uptakes were 326 μmol g⁻¹ for Nd-Ni₂P and 314 μmol g⁻¹ for Y-Ni₂P. The CO uptake of Nd-Ni₂P was larger than that of Y-Ni₂P, which is possibly because Nd-Ni₂P possesses a large surface area (Table 1, BET results) and lower surface accumulation of phosphate (see discussion of the XPS results) than those of Y-Ni₂P.

X-Ray Photoelectron Spectroscopy

The XPS spectra are shown in Figure 2. Table 2 gives the binding energies and the Ni/P atomic ratios. As depicted in Figure 2(a), for all catalysts, the bands centred at 852.5–852.7 eV and 856.4–856.7 eV belong to the Ni^δ+ species in the Ni₂P phase and Ni²⁺ species interacting with phosphate as a consequence of superficial oxidation, respectively.¹⁴ In Figure 2(b), the peaks centred at 128.8–130.3 eV can be assigned to a P^δ− species in the Ni₂P phase¹⁵ and that at 134.6–134.8 eV can be attributed to phosphate (P⁵⁺) due to superficial oxidation.¹⁶

Figure 2.

XPS spectra of the Ni₂P, Nd-Ni₂P and Y-Ni₂P catalysts. (a) Ni 2p core level spectra, (b) P 2p core level spectra.

Table 2.

Spectral parameters obtained by XPS analysis.

Sample	Binding energy (eV)
	Ni 2p_3/2			P 2p_3/2		Superficial atomic ratio
	Ni^δ+	Ni²⁺	Satellite	P⁵⁺	P^δ–	R _Ni ^δ+	P/Ni
Ni₂P	852.7	856.7	861.2	134.8	128.8	0.041	3.0
Nd-Ni₂P	852.5	856.4	861.7	134.6	129.9	0.055	2.2
Y-Ni₂P	852.6	856.4	860.9	134.6	130.3	0.054	2.4

The ratios of the Ni^δ+ species to total Ni of the Ni₂P phase (RNi^δ+) are shown in Table 2. The RNi^δ+ of bulk Ni₂P is only 0.041, whereas those of the Nd-Ni₂P and Y-Ni₂P catalysts are 0.055 and 0.054, respectively. This result indicates that the addition of a suitable amount of Y or Nd promotes the formation of active Ni₂P particles.

XPS data were applied to calculate the surface Ni/P atomic ratios and the results are shown in column 8 in Table 2. All samples showed larger P/Ni values than the initial P/Ni molar ratio (1.0) caused by an enrichment of phosphorus on the surfaces. The surface P/Ni atomic ratios were 2.2 and 2.4 for Nd-Ni₂P and Y-Ni₂P, respectively, which was smaller than that for bulk Ni₂P (3.0). This result suggests that the addition of Y or Nd to bulk Ni₂P can suppress the enrichment of P, which would lead to more nickel sites being exposed on the surface.¹⁷

Catalytic results and reaction pathways

Figure 3 depicts the BF conversions over samples as a function of the reaction times. The BF conversion gradually increased and then remained stable with time. The BF conversion over bulk Ni₂P was 70.0% after 8 h. Upon adding Nd or Y to the bulk Ni₂P, the BF conversion was increased. The BF conversions over Nd-Ni₂P and Y-Ni₂P were 95.3% and 89.1%, respectively. This was attributed to Nd-Ni₂P and Y-Ni₂P having smaller sized particles and a better dispersion of the active Ni₂P phase, together with the higher surface area and lower enrichment of phosphorus when compared with the bulk Ni₂P catalyst.

Figure 3.

BF HDO conversion over Ni₂P, Nd-Ni₂P and Y-Ni₂P catalysts.

Figure 4 shows the selectivity to reaction products of the prepared catalysts. On the basis of the products and previous work,¹⁷ the reaction route is shown in Scheme 1. The first step involves hydrogenation of the furan ring, which led to the formation of 2,3-dihydrobenzofuran (2,3-DHBF), then 2,3-DHBF is further converted into 2-ethylphenol (2-EtPh) by a C-O bond cleavage of the heterocyclic ring through a hydrogenolysis. Ethylbenzene (EB) is obtained from 2-EtPh by dehydration, then converted into ethylcyclohexane (ECH) by hydrogenation of the benzene ring, followed by demethylation of ECH to give methylcyclohexane (MCH).

Figure 4.

Product selectivities to BF HDO over Ni₂P, Y-Ni₂P and Nd-Ni₂P catalysts.

Scheme 1.

Reaction pathways for BF HDO over Ni₂P, Y-Ni₂P and Nd-Ni₂P catalysts.

For Ni₂P catalyst, the order of product selectivity was 2,3-DHBF (37.9%) > ECH (26.9%) > 2-EtPh (20.9%) > MCH (9.7%) > EB (4.6%). The selectivity over Nd-Ni₂P decreased in the order ECH (62.0%) > 2-EtPh (15.1%) > MCH (12.3%) > 2,3-DHBF (10.3%) > EB (0.3%). As mentioned above, the conversion of BF over Nd-Ni₂P catalyst has increased to 25.3% when compared to that of the Ni₂P catalyst, but the selectivity to 2,3-DHBF has decreased to 27.6%. This showed that the addition of Nd could accelerate the rate of C-O bond cleavage of the heterocyclic ring on 2,3-DHBF via hydrogenolysis and lead to the formation of more 2-EtPh. The selectivity to 2-EtPh over Nd-Ni₂P has a decrease of 5.8% when compared with that found for Ni₂P. This result indicated that the 2-EtPh dehydration reaction was also accelerated by addition of Nd and generated much more EB than did the Ni₂P catalyst. However, for Nd-Ni₂P, the EB could hardly be detected (0.3%), which showed that the addition of Nd could efficiently accelerate the hydrogenation of the benzene ring on EB to generate ECH. In accordance with the efficient EB hydrogenation, the selectivity to ECH over Nd-Ni₂P (62.0%) has an evident increase of 35.1% when compared with that found for Ni₂P (26.9%). The selectivity to MCH over Nd-Ni₂P (12.3%) had an increase of 2.6% when compared with that found for Ni₂P (9.7%). In consideration of the selectivity to ECH over Nd-Ni₂P being much higher than that of Ni₂P, whether the addition of Nd was beneficial to the ECH demethylation to MCH still needs more research.

Similar results were obtained by addition of Y to Ni₂P. The selectivity over Y-Ni₂P decreased in the order ECH (56.4%) > 2-EtPh (17.6%) > MCH (12.0%) > 2,3-DHBF (11.7%) > EB (2.4%). However, the HDO performance of Nd-Ni₂P was superior to that of Y-Ni₂P.

Figure 5 shows the yields of O-free compounds over the catalysts. The yield of O-free compounds over bulk Ni₂P catalyst was only 41.2%, while over Nd-Ni₂P and Y-Ni₂P catalysts they were 74.6% and 67.7%, respectively. These data showed that the benefit of addition of Y was inferior to the addition of Nd as regards the yields of O-free compounds, which corresponded with the characterization of the catalysts.

Figure 5.

Yields of O-free compounds of BF HDO over Ni₂P, Nd-Ni₂P and Y-Ni₂P catalysts.

Conclusion

In this work, bulk Ni₂P, Nd-Ni₂P and Y-Ni₂P catalysts have been successfully prepared. XRD results indicated that the addition of Nd or Y can restrain the formation of Ni₅P₄ and thereby facilitate the generation of the Ni₂P phase. Nd or Y increased the surface area of catalysts greatly, which was attributed to better dispersion of the Ni₂P particles. On the contrary, the enrichment of P on the catalyst surface decreased, leading to the number of exposed nickel atoms being increased and a higher CO uptake. The Nd-Ni₂P and Y-Ni₂P gave high conversions for BF HDO, and the Nd-Ni₂P catalyst reached 95.3%, some 25.3% higher than that of the bulk Ni₂P. The yields of O-free compounds over Nd-Ni₂P catalyst reached 74.6%, which was an increase of 33.4% when compared with that found for bulk Ni₂P and was an increase of 6.9% when compared with that found for Y-Ni₂P. The selectivity to the main O-free product MCH over Nd-Ni₂P catalyst was 66.0%, indicating that the addition of Nd was highly effective for promoting the BF HDO reaction.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) received financial support for the research, authorship, and/or publication of this article: The authors acknowledge the financial support from the National Natural Science Foundation of China (21276048) and Innovative scientific research projects of Innovative Practice Base for Postgraduate Training in Northeast Petroleum University (YJSCX2017-016NEPU).

Author biographies

Hua Song received her PhD degree in Oil and Gas Field Chemical Engineering from Northeast Petroleum University, China. She is currently a Distinguished Professor of Longjiang Scholar and the Director of provincial key laboratory of petrochemical engineering at College of Chemistry & Chemical Engineering, Northeast Petroleum University, China. Her laboratory is focused on catalytic materials and catalyst engineering.

Xueya Dai is a Master Candidate working under Professor Song at College of Chemistry & Chemical Engineering, Northeast Petroleum University, China. Her research interest is hydrogenation theory and catalysts.

Nan Jiang received her MS degree in Chemical Engineering from Clausthal University of Technology, Germany. She is currently a PhD candidate in Chemical Engineering at College of Chemistry & Chemical Engineering, Northeast Petroleum University, China.

Zijin Yan, a Master candidate under Prof Song at College of Chemistry & Chemical Engineering, Northeast Petroleum University, China. Research interest is hydrogenation theory and catalysts.

Tianhan Zhu, a Ph.D. candidate in chemical engineering under prof. Song at College of Chemistry & Chemical Engineering, Northeast Petroleum University, China.

Feng Li received his PhD degree in Chemical Technology from Northeast Petroleum University, China. He is currently a Professor in Chemical Engineering at College of Chemistry & Chemical Engineering, Northeast Petroleum University, China. His research is focused on catalytic materials and catalyst engineering.

References

Lup

ANK

Abnisa

Wan

MAWD

et al . Appl Catal A: Gen 2017; 541: 87–106.

Bie

Lehtonen

Kanervo

Appl Catal A: Gen 2016; 526: 183–190.

Shetty

Murugappan

Prasomsri

et al . J Catal 2015; 331: 86–97.

Liu

Wang

Liu

et al . Catal Today 2017; 292: 133–142.

Zhu

Song

Dai

et al . Chinese J Chem Eng 2017; 25: 1789–1790.

Zhang

et al . J Catal 2017; 345: 295–307.

Lan

et al . Chinese J Chem Eng 2014; 22: 903–908.

Song

Wang

et al . J Catal 2014; 311: 257–265.

Yao

Tian

et al . Fuel Process Technol 2015; 133: 146–151.

10.

Prodius

Mudring

AV.

Coordin Chem Rev 2018; 363: 1–16.

11.

Phonthammachai

Rumruangwong

Gulari

et al . Colloid Surface A 2004; 247(1): 61–68.

12.

Sun

Wang

et al . Top Catal 2012; 55: 1010–1021.

13.

Layman

Bussell

ME.

J Phys Chem B 2004; 108: 10930–10941.

14.

Korányi

Vít

Poduval

et al . J Catal 2008; 253: 119–131.

15.

Yang

Wang

et al . J Catal 2015; 330: 330–343.

16.

Song

Dai

Guo

et al . Fuel Process Technol 2012; 96(3): 228–236.

17.

Song

Gong

Song

et al . Appl Catal A: Gen 2015; 505: 267–275.

The effect of neodymium and yttrium on benzofuran hydrodeoxygenation performance over a bulk Ni 2 P catalyst

Abstract

Keywords

Introduction

Experimental

Preparation of bulk Ni2P, Nd-Ni2P and Y-Ni2P catalysts

Characterisation of catalysts

Catalytic activities

Results and discussion

X-Ray Diffraction

Textural properties

CO uptake

X-Ray Photoelectron Spectroscopy

Catalytic results and reaction pathways

Conclusion

Footnotes

Declaration of conflicting interests

Funding

Author biographies

References

Preparation of bulk Ni₂P, Nd-Ni₂P and Y-Ni₂P catalysts