Study of the Reaction Pathways for the Hydrogenation of Quinoline over Nickel Phosphide Catalysts

Abstract

Nickel phosphide catalysts (Ni₂P) were prepared using mesoporous molecular sieves as supports during isobaric co-impregnation. Ni₂P catalysts with different loading values were characterized, showing that the active phase on the surface of the catalysts was mainly Ni₂P and the catalysts still retained the mesoporous structural characteristics of the supports. The catalysts were evaluated using a 10 mL fixed-bed hydrogenation unit. The results showed that the nickel phosphide catalysts had a higher hydrogenation capacity than the sulfide catalysts and were able to preferentially hydrogenate and saturate most of the quinolines to decahydroquinolines, reduce the conversion of 1,2,3,4-tetrahydroquinoline to o-propylaniline, and reduce the inhibition of reactivity due to competitive adsorption. The effect of the catalyst on the path selectivity of quinoline hydrogenation was investigated, and the products of quinoline hydrogenation and denitrogenation consisted mainly of propylbenzene and propylcyclohexane, with propylcyclohexane accounting for 91.7% of the product and propylbenzene for 4.8%, under the conditions of nickel phosphide catalysts. Furthermore, the 25 wt% Ni₂P/SBA-15 catalyst exhibited no significant loss of catalytic activity during a 72 h stability evaluation conducted at 360 °C.

1. Introduction

The removal of nitrogen compounds from oil is more challenging than the removal of sulfur from sulfur-containing compounds. As alkaline nitrogen compounds increase the production of soluble gums, especially aniline and quinoline, they have a significant effect on the color stability of gasoline and kerosene [1,2,3], and alkaline nitrogen compounds inhibit the hydrodesulfurization of liquefied oils [4]. Quinoline, quinoline-type compounds, and their derivatives account for a large proportion of the alkaline nitrogen compounds in oils, and they are difficult to remove, so it is difficult to meet the requirements of nitrogen removal in coal liquefaction oil hydrofinishing by the traditional conventional sulfided hydrofining catalysts.

In recent years, transition metal phosphides represented by Ni₂P have attracted attention for their excellent hydrodenitrogenation (HDN) and hydrodesulfurization (HDS) performance, as well as high thermal and chemical stability [5,6]. The triangular prismatic coordination unit in phosphides is structurally analogous to that in sulfides. However, unlike the layered structure of sulfides, phosphides exhibit a three-dimensional, non-layered architecture. This structural feature exposes a higher density of coordinatively unsaturated sites on the surface compared to sulfides, resulting in greater active site density and enhanced catalytic activity. Moreover, phosphides do not require in situ activation, exhibit minimal leaching, and possess strong resistance to poisoning, making them highly promising for industrial hydrotreating applications. Consequently, research in this area remains highly active [7,8,9,10,11,12]. SBA-15, a mesoporous silica molecular sieve structurally similar to SBA-3, stands out owing to its larger pore diameter compared to both SBA-3 and MCM-41. Combined with thicker pore walls and higher pore volume, this feature makes SBA-15 particularly suitable for processing coal-derived oils containing bulky molecules. Furthermore, its superior hydrothermal stability enables application under high-temperature conditions [13,14].

In this work, phosphide catalysts based on SBA-15 were prepared to investigate the conversion pathways of the typical basic nitrogen compound quinoline (Q) in the hydrodenitrogenation (HDN) process. The results showed that the nickel phosphide catalyst exhibited significantly higher hydrogenation activity than the sulfided catalyst. It was able to selectively hydrogenate and saturate most of the quinoline to decahydroquinoline while suppressing the formation ratio of 1,2,3,4-tetrahydroquinoline (THQ1) to o-propylaniline (OPA). This characteristic effectively mitigated the reaction inhibition caused by competitive adsorption in the OPA pathway, thereby enhancing the quinoline HDN activity and accelerating the reaction kinetics. The HDN performance of the catalyst was further investigated. Under the influence of Ni₂P, the primary HDN product derived from quinoline was saturated propylcyclohexane, which accounted for 91.6%, whereas propylbenzene constituted only 4.9%.

2. Results

2.1. Carrier and Catalyst XRD Analysis

As shown in Figure 1a, the low-angle XRD patterns for SBA-15 and various catalysts are presented. The peaks at 0.86°, 1.55°, and 1.75° corresponding to the (100), (110), and (200) diffraction planes of the SBA-15 crystals, respectively, confirm the remarkable well-ordered hexagonal mesoporous structure of the material, particularly evident in the pristine SBA-15 support. Diffraction peaks associated with the (100) plane remain observable in the Ni-containing phosphide catalysts, indicating that the mesostructure is preserved at lower Ni loadings. However, the intensities of the (110) and (200) reflections are significantly reduced at 5% Ni loading and gradually disappear with further increases in Ni content. This indicates progressive degradation of the long-range order of SBA-15 as the Ni loading increases. The growth of nickel phosphide species within the pores likely causes pore blockage, leading to partial collapse of the mesoporous framework. As a result, the intrinsic structural order of the molecular sieve is compromised, which impairs its catalytic performance.

The high-angle XRD patterns for SBA-15 and N₂P/SBA-15 catalysts with varying Ni contents are shown in Figure 1b. From the figure, it can be seen that no obvious Ni₂P characteristic peaks appear in the XRD spectrum at a Ni content of 5%. This is due to the low concentration of the Ni₂P component at this stage. When the Ni content was 15%, weak diffraction peaks appeared at 2θ angles of 41°, 44.7°, 47.2° and 54°, which correspond to the characteristic diffraction peaks of Ni₂P. The intensity of these Ni₂P characteristic diffraction peaks increased with increased Ni₂P loading and reached a maximum when the loading reached 40%. The ICP analysis results confirmed the successful synthesis of Ni₂P/SBA-15 catalysts with varying loadings (

Table 1. ICP test results for catalysts with different Ni₂P contents.

Catalyst	Ni (wt%)	P (wt%)	Ni2P (wt%)	Ni/P(at%)
5wt%Ni2P/SBA-15	3.5	1.4	4.9	1.3
15wt%Ni2P/SBA-15	11.5	3.2	14.7	1.9
25wt%Ni2P/SBA-15	19.9	5.5	25.4	1.9
40wt%Ni2P/SBA-15	31.5	8.3	39.8	2

). Notably, the Ni/P atomic ratio of the high-loading catalyst was found to be closer to the ideal stoichiometry of Ni₂P (2:1) compared to that of the low-loading counterpart. This observation is consistent with the XRD findings, where the high-loading catalyst exhibited better crystallinity of the Ni₂P phase.

The pore size distributions of the SBA-15 support and Ni₂P/SBA-15 catalysts with different Ni contents are shown in Figure 1c, and their surface structural parameters are shown in

Table 2. Physico-chemical properties of catalysts with different Ni contents.

Catalyst	Ni (wt%)	Specific Surface (m2 g−1)	Pore Volume (cm3 g−1)	Pore Size (nm)
SBA-15	0	715	1.16	6.8
Ni2P/SBA-15	5	431	0.65	6.3
	15	342	0.49	5.9
	25	255	0.40	5.4
	40	167	0.33	4.9

. This study prepared an SBA-15 carrier with a specific surface area of 715 m²/g, a pore volume of 1.16 cm³/g, and a pore diameter of 6.8 nm; with the increase in the stretcher amount, the Ni₂P component entered the pores of the SBA-15 support, and the increasing loading caused the specific surface area and pore diameter of the catalyst to decrease progressively. When the Ni content reached 40%, these values dropped to 167 m²/g for specific surface area, 0.33 cm³/g for pore volume, and 4.9 nm for pore diameter. The trend indicates that the pore structure of the catalyst deteriorates with increasing Ni₂P concentration.

2.2. Study of the Denitrogenation Reaction of Quinoline

2.2.1. Hydrogenation Pathway of Quinoline

As shown in Figure 2, the general HDN reaction pathway of quinoline, the main intermediates in the catalytic hydrogenation of quinoline are THQ1, 5,6,7,8-tetrahydroquinoline (THQ5), OPA, and decahydroquinoline (DHQ). There are two routes for the nitrogen removal from quinoline: one proceeds via OPA and the other via DHQ [15], and the selectivity of the reaction pathways varies considerably with different catalysts. The bond energies of the C=N double bonds are much higher than those of the C-N single bond (615 kJ/mol for C=N double bonds, 305 kJ/mol for C-N bonds), and the nitrogen removal requires hydrogenation saturation of the unsaturated bond before C-N bond cleavage can occur. Satterfield et al. [16] concluded that quinoline and TH Q1 can easily reach thermodynamic equilibrium under certain reaction conditions, and that quinoline is essentially converted to THQ1. As shown in Figure 3 and Figure 4, 15 wt% Ni₂P/SBA-15 exhibits the highest nitrogen removal rate and PCH/PB ratio, indicating promotion of the hydrogenation saturation pathway (Pathway I). This may be attributed to the fact that Ni₂P provides more active hydrogenation centers, thereby enhancing the hydrogenation conversion rate in Pathway I [17].

2.2.2. Effect of Different Temperatures on Conversion and Denitrogenation Rates

The role of temperature in quinoline hydrogenation conversion under different catalyst conditions is shown in Figure 3a. The conversion of quinoline during hydrogenation was strongly influenced by temperature. Increasing the reaction temperature from 280 °C to 340 °C led to a substantial increase in quinoline conversion under all tested catalyst conditions. At 340 °C, quinoline conversion exceeded 93% for all catalysts. As shown in Figure 3a, the Ni₂P catalyst exhibits better low-temperature hydrogenation performance compared to the commercial sulfided catalyst FT1, and higher Ni content facilitates the reaction at lower temperatures, However, in terms of final conversion, the phosphide catalysts offer no advantage over the commercial catalyst FT1.

The effect of Ni loading on the catalytic activity was also significant, and the conversion of quinolines increased with increasing Ni loading over the temperature range of 280–340 °C. This is because the number of active metal sites that can be accessible to the quinoline hydrotreating reactants increases with higher Ni₂P loading. However, increasing the temperature to 360 °C and the loading to 40% did not significantly improve conversion, and even tended to decrease it, as excessive loading can block active pores and reduce the catalyst’s specific surface area, which negatively affects catalytic activity. Additionally, raising the temperature to 360 °C affected the conversion of quinoline to tetrahydroquinoline due to the exothermic nature of the reaction. The higher temperature accelerated the reverse reaction rate of tetrahydroquinoline hydrogenation, shifting the equilibrium away from complete conversion and thereby negatively impacting quinoline conversion.

As shown in Figure 3b, the relationship between temperature and the hydrodenitrogen removal rate of quinoline is presented under various catalyst conditions. A comparison between curves in Figure 3 shows that the denitrogenation rate tends to increase in parallel with the conversion rate as temperature goes up, and the phosphide catalyst has a higher denitrogenation rate relative to the sulfide catalyst FT1 across different temperature conditions, reaching over 92% at 320 °C—significantly higher than that of the sulfided catalyst. As shown in

Table 3. Summary of reported results in HDN of quinoline (Q).

Catalysts	Feed	Reaction Conditions	HDN/%	Reference
Ni2P/meso-beta@MAS	1.0 wt% Q in cyclohexane	0.5 g cat, T = 340–400 °C, P = 4.0 MPa, WHSV = 12.8 h−1, H2/Oil = 600.	78.3	[5]
Ni2P/H-beta@SBA-16-2	1.0 wt% Q in cyclohexane	0.5 g cat, P = 4 MPa, H2/feed = 400, WHSV = 12.8 h−1, 400 °C	81.6	[6]
Ni2P/3DOMZSM-5	1.0 wt% Q in cyclohexane	0.5 g cat, P = 4 MPa, H2/feed = 400, WHSV P = 12.8 h−1, 400 °C	91	[17]
NiMoW/BK-3	0.2 wt% Q in cyclohexane	P = 4 MPa, H2/feed = 400, WHSV = 10 h−1, 280 °C	52.4	[18]
Ni2P/SBA-15	1.0 wt% Q in decahydronaphthalene	P = 6.0 MPa, T = 280–360 °C, WHSV = 20 h−1.	92	This work

, Ni₂P/SBA-15 exhibits a higher HDN% than the other catalysts listed.

2.2.3. Selectivity Study of Quinoline Hydrogenation Products

In each performance evaluation test, the same mass of catalyst was used to investigate how catalysts with different loadings affect performance under varying operating conditions. The distribution of quinoline hydrodenitrogenation products with reaction temperature under phosphide catalyst (15%Ni₂P/SBA-15) and the sulfided commercial catalyst (FT1) is shown in Figure 4. The product distribution reveals that the intermediate species are the same under both catalysts: THQ1, THQ5, and DHQ. The final products PB and PCH were detected, while propylcyclohexylamine (PCHA) and propylcyclohexene (PCHE) were not observed in the product assay. This indicates that the corresponding hydrogenation and ring-opening reactions proceed readily under these conditions. The nitrogen removal rates of both catalysts were very low at low temperatures. On the nickel phosphide catalyst, the main product was fully saturated DHQ, whereas a significant amount of THQ1 remained unconverted over the sulfided commercial catalyst FT1. This suggests that the hydrogenation capacity of the nickel phosphide catalyst is significantly higher than that of the sulfided catalyst at low temperatures, which is one of the reasons for its superior denitrogenation activity. Satterfield [16] demonstrated that quinoline addition and ring-opening reactions were not detected on sulfide catalysts, further indicating that hydrogenation proceeds readily under these conditions. He also showed that quinoline HDN on sulfide catalysts occurs primarily via two pathways—OPA and DHQ—with the OPA pathway accounting for approximately 40% of the total. However, Perot and coworkers [19] reported that the OPA route is significantly inhibited in the presence of quinoline-type molecules (Q, THQ, DHQ) due to competitive adsorption, resulting in a slower reaction rate for this pathway. In contrast, under phosphide catalyst conditions, more THQ1 is hydrogenated and saturated to DHQ (71.4% of intermediate products at 280 °C) compared to the sulfided catalyst. This shifts the HDN reaction predominantly toward the DHQ pathway and reduces the proportion of THQ1 converted via the OPA route, thereby indirectly avoiding the inhibition associated with the OPA pathway and enhancing the overall denitrogenation reactivity of quinoline, thus accelerating the reaction.

As shown in Figure 4b, the OPA content remains low throughout the temperature ramp, indicating that the rapid increase in PCH does not originate from OPA. In addition, as the temperature increases, the selectivity toward DHQ decreases, whereas that toward PCH increases. The denitrogenation rate profile closely resembles the mole fraction curve of PCH, further suggesting that the HDN of quinoline over the phosphide catalyst proceeds predominantly via the Q-DHQ-PCHA-PCH pathway. This is the main reaction route under these conditions. Since PCHA is readily deaminated, the hydrogenolysis of the C-N bond in DHQ is likely the rate-determining step of the total HDN reaction rate. At 360 °C, the final quinoline HDN products are primarily PB and PCH. Over the sulfided commercial catalyst, the product distribution is 50.4% PB and 31.6% PCH, whereas over the nickel phosphide catalyst, PCH accounts for 91.6% of the products with only 4.9% PB.

Ni₂P has a unique crystal structure with numerous active sites on its surface, which facilitates the adsorption and activation of quinoline. The P atom not only modulates the electronic properties of Ni but also acts as a weak acid site, helping to stabilize the intermediates of quinoline hydrogenation. SBA-15 is an ordered mesoporous silica-based support whose structure effectively regulates the dispersion and particle size of Ni₂P, thereby influencing the number of active sites. The ordered mesoporous structure of SBA-15—featuring pore sizes significantly larger than those of quinoline molecules—reduces diffusion resistance for reactants, enabling quinoline to rapidly access the active sites of Ni₂P and significantly enhancing the hydrogenation reaction rate.

2.3. Analysis of Catalyst Durability

Catalyst durability is a key criterion for industrial applications; therefore, the 25%wt%Ni₂P/SBA-15 catalyst was further evaluated for stability. As shown in Figure 5a, the catalyst maintained stable activity during a 72 h stability test at 360 °C. To investigate the origin of this stability, the catalyst was characterized by XRD, XPS and TEM before and after the reaction. As shown in Figure 5b, the XRD patterns of the catalyst before and after the reaction are nearly identical, indicating that the active species remained Ni₂P. XPS analysis revealed the valence changes of the catalyst surface species before and after the reaction. In the Ni 2p XPS spectrum (Figure 5c), Ni²⁺ is evidenced by two fitted peaks at 862.1 eV and 856.3 eV, which are assigned to Ni²⁺ and satellite peaks, respectively. The peak at binding energies of 852.8 eV is attributed to Ni^δ+ (0 < δ < 1) species in the Ni₂P crystal. Similarly, the P 2p XPS spectra (Figure 5d) showed two peaks at 128.9 eV and 134.7 eV, which were attributed to reduced phosphorus in the form of phosphorylated phase (P-Ni) and phosphate species due to surface oxidation, respectively. TEM analysis was performed to assess structural changes before and after the reaction. As shown in Figure 5e,f, the catalyst morphology remains largely unchanged. Both HRTEM images reveal a lattice spacing of 0.22 nm, corresponding to the (111) crystal planes of Ni₂P. Structural stability is crucial for maintaining catalytic activity, and the 25 wt% Ni₂P/SBA-15 catalyst exhibits excellent stability—supporting its potential for continuous operation in industrial applications. In this study, the comparative evaluation of catalysts with 15% and 25% loadings constitutes a hierarchical investigation centered on the catalyst performance and application scenario. Specifically, testing of the 15% loading catalyst (Figure 4) focuses on exploring the “activity limit” of the catalyst system, while the stability assessment of the 25% loading catalyst (Figure 5) emphasizes its “practical application potential”. This approach provides more comprehensive insights for the design of Ni₂P catalysts in future studies.

3. Materials and Methods

3.1. Catalyst Preparation

SBA-15 was synthesized by many methods, and the present experiment was carried out with reference to the methods described in the literature. Dissolve 2.0 g P123 (EO₂₀PO₇₀EO₂₀, average Mn ~5800, Sigma-Aldrich, Beijing, China) in 20 mL water, adding 4.5 mL TEOS (Si (OC₂H₅)₄ 99%, Aladdin, Shanghai, China) and 2 mL HCl (4 mol/L, AR, Sinopharm Chemical Reagent, Beijing, China). The mixture was stirred for one day before being transferred to a Teflon bottle and kept at 100 °C for crystallization for another day. The resultant product was filtered through an 8 μm membrane, washed with deionized water, dried in a drying oven (DZF-6050, Yiheng, Shanghai, China) at 110 °C for 2 h, and calcined in a tube furnace (OTF-1200X-S, KJMTI, Hefei, China) at 550 °C for 5 h. The powder of SBA-15 molecular sieve was obtained [14,20].

The catalyst was loaded using the equal volume impregnation method. The final Ni₂P loading was controlled by controlling the mass ratio of SBA molecular sieve to nickel. The ratio of Ni to P is a molar ratio. It is added by molar ratio during equal volume impregnation. At room temperature, a 10 mL solution was first prepared by dissolving 23.12 mmol of Ni (NO₃)₂·6H₂O (99%, Macklin, Shanghai, China). Subsequently, 23.12 mmol of (NH₄)₂·HPO₄ (99%, Macklin, Shanghai, China) was added to this solution. The pH of the resulting impregnating solution was adjusted to the range of 3.0~4.0 using dilute HNO₃ (AR, Sinopharm Chemical Reagent, Beijing, China), and the solution was further diluted to a total volume of 44 mL. Finally, 20 g of pre-prepared SBA-15 was added to the above-mentioned impregnating solution and mixed thoroughly, leaving it for more than 10 h, drying at 110 °C, and putting it into a muffle furnace (KSL-1500X-S, KJMTI, Hefei, China) for 3 h at 550 °C to obtain a metal phosphate precursor of the metal phosphide (Ni₃(PO₄)₂). Oyama S T, Korányi T I et al. [21,22] showed that the active phase was pure Ni₂P when P/Ni ≥ 0.75. In the present experiments, phosphide catalysts with different Ni actual loadings were prepared under the loading condition of Ni/P = 2. As a comparison, the commercial catalyst FT1 (carrier γ-Al₂O₃, Fushun Research Institute of Petroleum and Petrochemicals, Fushun, Liaoning, China), which has a better nitrogen removal effect in practice, was chosen. The characteristics are shown in

Table 4. Properties of commercial catalyst.

Pore Volume (mL/g)	Specific Surface (m2/g)	Average Pore Size (nm)	Intensity (N/cm)	Effective Component/(wt/%)
				WO3	MoO3	NiO	P
≮0.25	≮150	8.13	≮182	≮24.0	≮2.4	≮2.5	0.3–1.7

.

Nickel phosphide precursors need to be reduced to produce nickel phosphide catalysts. After roasting, the catalytic precursor samples of nickel phosphide were ground, crushed, and sieved through a 50-mesh sieve. The sieved precursor particles with >50 mesh were placed in a stainless steel high-pressure reduction reactor, hydrogen was introduced at 200 mL/min, and the system’s temperature was gradually increased to 650 °C at a rate of 10 °C/min and then maintained at this temperature for 4 h. Then, we continued to introduce cold hydrogen until it cooled down to room temperature. Since the phosphide catalyst easily undergoes a violent oxidation reaction in direct contact with the atmosphere, in order to passivate the surface of the catalyst for easy characterization, the catalyst was passivated for 2 h by continuing to pass 50 mL/min of Ar gas containing 1% (v/v) pure O₂, and the passivated samples were taken out for the structural characterization of the catalyst and the catalytic reaction.

The synthesis mechanism of Ni₂P involves two key sequential steps: the formation of a nickel phosphate precursor and its subsequent hydrogen reduction [21,22]. Initially, the nickel phosphate precursor is prepared through the reaction between Ni (NO₃)₂·6H₂O and (NH₄)₂·HPO₄, followed by a calcination process; this calcination step converts the reactants into the primary precursor component, Ni₃(PO₄)₂. In the subsequent reduction step, the Ni₃(PO₄)₂ precursor is treated in a H₂ atmosphere at 650 °C, which drives the formation of Ni₂P along with characteristic byproducts, as described in Equation (1). During this reduction process, the Ni species in the precursor undergoes a valence state transition: Ni²⁺ ions in Ni₃(PO₄)₂ are reduced to the mixed valence state (approximated as Ni^δ+, where δ < 2) that is inherent to the Ni₂P product. Concurrently, the P species in the PO₄³⁻ of the precursor undergoes partial reduction: the majority of reduced P combines with Ni to form the target Ni₂P phase, while a minor fraction of P is further reduced to PH₃ as a byproduct. Additionally, H₂O is generated as the primary oxygen-containing byproduct, originating from the reduction of oxygen atoms in the PO₄³⁻ groups of the precursor.

The reaction governing the synthesis of Ni₂P is given in Equation (1):

$$8{\mathrm{N}\mathrm{i}}_3(\mathrm{P}{\mathrm{O}}_4{)}_2+38{\mathrm{H}}_2\to 12{\mathrm{N}\mathrm{i}}_2\mathrm{P}+2\mathrm{P}{\mathrm{H}}_3+32{\mathrm{H}}_2\mathrm{O}$$

(1)

3.2. HDN Activity Evaluation Device

The HDN reaction was carried out in a 10 mL stainless steel high-pressure fixed-bed reactor (Baikal, Shanghai, China), and the flow is shown in Figure 6 (Drawn using Autodesk AutoCAD 2023). The whole device mainly includes a gas supply system, a feed system, a reaction system, a separation system, and a tail gas metering system. Hydrogen is pressurized by the hydrogen compressor and then passes through the system pressure setting valve, and then through the hydrogen mass flow meter. In the upper part of the reactor, hydrogen is pumped over the raw material oil through the check valve, and the two are mixed, and together they enter the reactor from the upper part of the reactor. The temperature control of the reactor is divided into three sections. The catalyst is loaded in the middle of the thermostatic section, and the upper and lower sections are protected by ceramic ball support. After the reaction products come out from the bottom of the reactor, they pass through two separators, which can be separated from each other to realize a pressure-less discharge of the system. The gas phase product from the separators is metered by a gas flow meter.

3.3. Reaction Materials and Conditions

Decahydronaphthalene (AR) solution containing 1.0 wt % quinoline (Q) was used as the reaction feedstock at a reaction pressure of 6.0 MPa, a reaction temperature of 280–360 °C, and a gravimetric space-vacuum velocity (WHSV) of 20 h⁻¹.

3.4. Catalyst Characterization

3.4.1. X-Ray Diffractometer

The analysis was performed using a Rigaku D/MAX 2500VB 2+/PC X-ray diffractometer (Rigaku Corporation, Akishima, Japan), with a light source of Cu Kα radiation and a tube voltage of 40 kV, and a scanning range of 0.5–10° in the low-angle region and 10–90° in the high-angle region. The high-angle region characterizes the physical phase of the catalyst, and the low-angle region characterizes the mesoporous structure as well as its ordering.

3.4.2. Nitrogen Adsorption and Desorption Apparatus

The pore size was analyzed using a fully automated physical and chemical adsorption apparatus Sorptomatic 1990 (Thermo Fisher Scientific, Waltham, MA, USA). The catalyst needed to be degassed at 300 °C for 4 h before testing. To determine the specific surface area, the BET method was employed, whereas the BJH method was used to calculate the pore size distribution.

3.4.3. X-Ray Photoelectron Spectroscopy

X-ray photoelectron spectroscopy (XPS) was employed to characterize the surface elemental composition and chemical states of the catalysts. Measurements were performed using a Perkin-Elmer PHI 5000C ESCA system (Ulvac-Phi, Chigasaki, Japan) equipped with Mg Kα radiation (hν = 1253.6 eV). All binding energies were referenced to the adventitious carbon C 1s peak set at 284.8 eV.

3.4.4. High-Resolution Transmission Electron Microscopy

High-resolution transmission electron microscopy (HRTEM) was performed to characterize the microstructure and particle size of the catalysts using a JEOL JEM-2100F microscope (JEOL, Akishima, Japan) operated at 200 kV.

3.4.5. Inductively Coupled Plasma

Inductively coupled plasma was used for qualitative and quantitative analyses of elements in the samples. The composition of Ni₂P was analyzed with an ICP-OES 5800 (Agilent Technologies, Santa Clara, CA, USA).

3.5. Hydrogenation Product Analysis

The analysis of the quinoline hydrogenation products was performed on an Agilent 7890-II (Agilent Technologies, USA) gas chromatograph with the following chromatographic conditions: 0.53 mm × 50 m capillary column, FID detector, 280 °C injector, 280 °C detector, and 100–280 °C column temperature, and the internal standard method was used to quantitatively calculate the composition of the products and the selectivity of the hydrogenation products.

The nitrogen content of the products was analyzed with an Antek 9000 micro nitrogen and sulfur analyzer (Antek by PAC, Troy, MI, USA), and the measuring principle for nitrogen was chemiluminescence; the measuring range was 20~170,000 μg/kg.

The conversion rate of quinoline was calculated as per Equation (2):

$${\eta }_{\mathrm{Q}}={\displaystyle \frac{{\mathrm{C}}_{\mathrm{Q}0}-{\mathrm{C}}_{\mathrm{Q}\mathrm{t}}}{{\mathrm{C}}_{\mathrm{Q}0}}}\times 100\mathrm{\%}$$

(2)

where C_Q0 is the concentration of feedstock quinoline, and C_Qt is the concentration of product quinoline after hydrogenation.

There are a large number of intermediates in quinoline hydrogenation, so the nitrogen removal rate is not equivalent to the conversion rate. The denitrogenation rate is calculated as in Equation (3):

$${\eta }_{\mathrm{N}}={\displaystyle \frac{{\mathrm{C}}_{\mathrm{N}0}-{\mathrm{C}}_{\mathrm{N}\mathrm{t}}}{{\mathrm{C}}_{\mathrm{N}0}}}\times 100\mathrm{\%}$$

(3)

where ${\eta }_{\mathrm{N}}$ is the nitrogen removal rate, C_N0 is the concentration of feedstock nitrogen, and C_Nt is the concentration of product nitrogen after hydrogenation.

4. Conclusions

The impregnation method was employed to synthesize the Ni₂P catalyst, using mesoporous molecular sieve SBA-15 as the support. The results indicated that the active phase of the catalysts was primarily Ni₂P, and that catalysts with different loadings retained their mesoporous structures. However, with increasing Ni₂P loading, particle growth led to partial blockage of the SBA-15 pore structure. Additionally, as the loading increased, a steady decrease was observed in the specific surface area, pore volume, and pore diameter of the samples. Activity evaluation results showed that the nickel phosphide catalyst exhibited higher hydrogenation capacity compared to the sulfided catalyst. It preferentially hydrogenated and saturated most quinoline to DHQ, thereby reducing the ratio of THQ1 to OPA. This shift indirectly avoids the reactivity inhibition caused by competitive adsorption in the OPA pathway, thus enhancing the denitrogenation reactivity of quinoline and accelerating the overall reaction. Under nickel phosphide catalysis, the main final hydrodenitrogenation product of quinoline was saturated propylcyclohexane, accounting for 91.6%, with propylbenzene representing only 4.9%. Importantly, the 25 wt% Ni₂P/SBA-15 catalyst showed no significant deactivation after 72 h of operation at 360 °C, demonstrating excellent catalytic stability.