Enhanced catalytic hydrodechlorination of 2,4-dichlorophenol over Pd catalysts supported on nitrogen-doped graphene

Abstract

Pd catalysts supported on graphene and N-doped graphene (GN-1, GN-2 and GN-3) with varied N-doping amounts were prepared using the deposition–precipitation method, and liquid phase catalytic hydrodechlorination (HDC) of 2,4-dichlorophenol (2,4-DCP) was investigated over these catalysts. The catalysts were characterized by X-ray diffraction, elementary analysis, N₂ adsorption–desorption isotherms, transmission electron microscopy, and X-ray photoelectron spectroscopy. Characterization results showed that graphene could be successfully doped by N using the heat treatment method with melamine as precursor, and N doping amounts were determined to be 5.7, 8.6 and 11.3% for GN-1, GN-2 and GN-3, respectively. Additionally, Pd²⁺/Pd⁰ ratios and Pd dispersions in the Pd/GN catalysts were much higher than those in Pd/graphene. For a similar Pd loading, the Pd dispersion of Pd/GN first increased and then decreased with the increase of N-doping amount, and the highest Pd dispersion was observed on Pd(2.9)/GN-2. Accordingly, GN supported Pd catalysts exhibited much higher catalytic activities than Pd/graphene, the catalytic reaction first increased and then decreased slightly in activity with the increase of nitrogen doping amount, and the highest activity was identified on Pd(2.9)/GN-2. Moreover, the dechlorination of 2,4-DCP over supported Pd catalysts proceeded via both a stepwise and concerted pathway, and the concerted pathway became predominant upon N doping.

---

1. Introduction

Chlorophenols are widely used chemical raw materials and intermediate products in the manufacture of pesticides, dyes, leather and disinfectants.^1–3 These chemicals have very strong toxicity and low biodegradability. The discharge of chlorophenols due to industrial production and human activities could cause severe environmental pollution. Hence, effective treatment techniques are highly demanded to eliminate chlorophenol pollution in water.

At present, a variety of techniques are available to treat chlorophenols in water, such as advanced oxidation process,^4,5 incineration,⁶ biological methods^7,8etc. Recently, the liquid phase catalytic hydrodechlorination (HDC) method is considered as a promising method for the treatment of chlorinated organic pollutants.^9–11 Supported noble metal catalysts are usually employed in the HDC reactions, and carbon materials are the most commonly used supports because of their high surface areas and strong resistance to acid, alkalis and chlorine poisoning during the HDC reaction.¹² For example, Pedro et al.¹³ compared the HDC of 4-chlorophenol over Pd/Al₂O₃ and Pd/AC and concluded that the AC supported catalyst was less susceptible to the chloride poisoning during the reaction. Diaz et al.¹⁴ also found that Pd/AC catalyst maintained a constant activity in the HDC of 2,4-DCP.

For the liquid phase catalytic HDC of 2,4-DCP, smaller Pd particles usually intrinsically higher specific activities.¹⁵ Moreover, strong metal–support interaction between Pd particles and the carrier can form electron-deficient Pd active sites to favor the HDC reaction.¹⁶ Carbon materials normally have weak interactions with the supported noble metal due to their strong hydrophobicity, resulting in aggregated metal particles on the support, and easy leaching during the liquid phase reaction.^17,18 Hence, many researchers attempted to modify carbonaceous supports to enhance the noble metal dispersion/stability as well as the catalytic activities. Heteroatom doping (e.g. N, B, or P) has been recognized as an effective method to modify the carbon surface and introduce chemically reactive sites for anchoring metal nanoparticles^19–22 on carbonaceous supports. Nitrogen is one of the mostly used elements to improve the catalytic activity and stability. For example, Liu et al.²³ found that N-doped mesoporous carbon supported Pt catalyst with uniform Pt dispersion exhibited a higher catalytic activity for p-nitrophenol reduction than mesoporous carbon supported Pt catalyst. Chizari et al.²⁴ investigated the liquid-phase hydrogenation of cinnamaldehyde over nitrogen-doped carbon nanotubes supported Pd catalysts, and concluded that the introduction of nitrogen atoms into the carbon matrix effectively modified the chemical properties of the support, which substantially enhanced the hydrogenation activities. In parallel, our previous study found that Pd catalysts supported on B-doped mesoporous carbon had a higher HDC activity than that on mesoporous carbon.²⁵

As a new carbon material graphene consisted of monoatomic layer structure with sp² hybridization, and processed excellent electrical conductivity, high chemical and thermal stability, and large surface area, which could be used as a common catalyst support for the liquid phase catalytic HDC operating under very strong alkaline conditions.^26–28 Up to now, however, graphene and doped graphene were mostly applied in electrochemical fields,^29–31 and few studies on HDC of chlorinated phenols had been conducted over catalysts supported on graphene or heteroatoms doped graphene.

In the present study, we prepared a series of Pd catalysts supported on graphene and N-doping graphene with varied nitrogen and Pd contents, and investigated the liquid phase catalytic HDC of 2,4-DCP over these catalysts. The results showed that Pd catalysts supported on N-doping graphene had smaller Pd particle sizes and higher Pd²⁺ concentrations, exhibiting enhanced catalytic activities for the liquid phase catalytic HDC of 2,4-DCP.

2. Experimental

2.1. Catalyst preparation

Graphite oxide (BET surface area of 4.5 m² g⁻¹) was purchased from Xianfeng Nanomaterials company (Nanjing, China). Nitrogen-doped graphene (GN) were prepared using melamine as the N-containing precursor.³² Briefly, 0.6 g of graphite oxide was suspended in 2 ml of alcohol solution containing a certain amount of melamine and stirred at room temperature for 5 h, then the mixture was dried by slowing vaporizing alcohol at 80 °C. The resulting sample was placed in a tubular oven, which was heated to 950 °C in an inert N₂ atmosphere with a ramping rate of 10 °C min⁻¹, and hold at 950 °C for 30 min. After cooling down to room temperature, the material was washed with boiling water several times to remove excess melamine decomposition products. Samples with varied N contents were recorded as GN-1, GN-2 and GN-3, respectively. The N-doping contents determined using an elemental analyzer (CHN-O-Rapid, Germany) were 5.7, 8.6 and 11.3% for GN-1, GN-2 and GN-3, respectively. Graphene was prepared according to the same process without adding melamine.

Supported Pd catalysts with different supports (graphene and GN) and varied Pd loading amounts were prepared by the deposition–precipitation method. Briefly, the support was suspended in PdCl₂ solution and heated at 45 °C for 2 h under vigorously stirring. Then, 1.0 M Na₂CO₃ solution was added dropwise to the suspension to adjust the solution pH to 10.5. After stirring for 1 h, supported Pd catalysts with different supports and Pd loading amounts were collected by filtration, washed thoroughly with distilled water, followed by drying at 105 °C for 6 h and finally reducing under a H₂ stream (40 ml min⁻¹) at 300 °C for 2 h. The resulting catalysts are referred to as Pd(x)/graphene or Pd(x)/GN, where x is the Pd loading amount (wt%). All catalysts were ground to pass through a 400-mesh sieve (<37 μm) to avoid the potential intraparticle diffusion before catalytic test.³³

2.2. Catalyst characterization

Surface areas of the samples were measured using the Brunauer–Emmett–Teller (BET) method on a Micromeritics ASAP 2020 (Micromeritics Instrument Co., Norcross, GA) instrument. The samples were degassed under N₂ flow at 100 °C for 4 h prior to analyze at −196 °C (77 K). Powder X-ray diffraction (XRD) patterns of the samples were obtained using a Rigaku D/max-RA powder diffraction-meter (Rigaku, Tokyo, Japan) equipped with a Cu Kα radiation. Elemental analysis was conducted on an elemental analyzer (CHN-O-Rapid, Germany). The Pd contents in the catalysts were determined on an inductive coupled plasma emission spectrometer (ICP) (J-A1100, Jarrell-Ash, USA). Transmission electron microscopy (TEM) images of the samples were conducted on a JEM-200CX electron microscope (JEOL Co., Tokyo, Japan). X-ray photoelectron spectroscopy (XPS) was performed on a PHI-5000 Versa Probe using a monochromatized Al Kα excitation source (hν = 1486.6 eV) (ULVAC-PHI, Japan). The C 1s peak (284.6 eV) was used for the calibration of binding energy.

2.3. Liquid phase catalytic HDC of 2,4-DCP

The liquid phase catalytic HDC reactions of 2,4-DCP were carried out in a 250 ml of four-necked flask reactor with a sample port, pH-stat, H₂/N₂ inlet and outlet under atmospheric pressure of hydrogen/nitrogen. The reaction temperature was stabilized at 25 ± 0.5 °C with a water-bath (SDC-6, Scientz Co., China). Briefly, 50 mg of catalyst was suspended in 200 ml of 3.0 mM 2,4-DCP solution with pH pre-adjusted to 12 using 1.0 M NaOH aqueous solution. The mixture was purged with a N₂ flow (50 ml min⁻¹) for 30 min, and then the N₂ flow was switched to a H₂ flow (250 ml min⁻¹) under continuously vigorous stirring (1400 rpm). Samples were taken at selected time intervals and the catalyst particles were separated by fast filtration. The concentrations of the reactant, intermediate and product in the filtrate were determined by a high-performance liquid chromatography with an ultraviolet (UV) detector at 270 nm using a 4.6 × 150 mm HC-C18 column (Agilent) as stationary phase and a mixture of acetonitrile and water (60 : 40, v/v) as mobile phase. Prior to liquid chromatography analysis, the filtrate was neutralized using 1.0 M HCl. The initial activity of the catalyst was calculated based on the first-order rate law at the conversion of 2,4-DCP below 25%. Two separate runs of the HDC of 2,4-DCP on Pd(2.8)/GN-3 were conducted, and the results showed raw data reproducibility that was better than ±7% (see Fig. 3S, ESI†). To test the possible mass transfer limitation in the reaction system, the HDC of 2,4-DCP catalyzed by Pd(2.8)/GN-3 with varied catalyst dosages was carried out (results presented in Fig. 4S, ESI†). The catalyst dosage normalized activities remained nearly constant for the HDC of 2,4-DCP, indicative of the absence of mass transfer limitation under our experimental conditions.³⁴ The fractional conversion of 2,4-DCP (X_2,4-DCP) is defined as

X_2,4-DCP = (C_2,4-DCP,0 − C_2,4-DCP)/C_2,4-DCP,0 (1)

and selectivity, with respect to 2-CP (S_2-CP), is calculated from

S_2-CP = C_2-CP/(C_2,4-DCP,0 − C_2,4-DCP) (2)

3. Results and discussion

3.1. Catalyst characterization

The XRD patterns of graphite, Pd(2.7)/graphene, Pd(2.9)/GN-1, Pd(2.9)/GN-2 and Pd(2.8)/GN-3 are shown in Fig. 1. For all supported catalysts, diffraction peaks assigned to Pd or PdO were not identified, and thus the diffraction peaks in the spectra can be only attributed to the supports, e.g. graphene, GN-1, GN-2 and GN-3. For graphite, diffraction peaks were observed at 26.5°, 44° and 54°, assigned to hexagonal crystalline graphitic carbon structure.³¹ As for graphene, the diffraction peaks characteristic of graphite disappeared, and a broad and much low diffraction peak was identified around 23.4°, reflecting that the interlayer spacing increased and laminated graphene structure was formed by weakening the van der Waals interactions between graphitic layers.³⁵ The XRD patterns of GN-1, GN-2 and GN-3 were similar to graphene, while the peak intensity was stronger, might be due to the interlayer spacing was shrinked when nitrogen source was added.

Fig. 1 XRD patterns of graphite, Pd(2.7)/graphene, Pd(2.9)/GN-1, Pd(2.9)/GN-2 and Pd(2.8)/GN-3.

The XRD patterns of Pd/GN-3 catalysts with varied Pd loading amounts are compiled in Fig. 1S, ESI.† The diffraction peaks characteristic of metallic Pd was only observed around 40.2° in the catalyst with the highest Pd loading (Pd(4.1)/GN-3). For other Pd catalysts, diffraction peaks attributed to PdO or Pd was not identified, likely due to the high Pd dispersions in the catalysts.

Some structural properties of the catalysts, e.g. BET surface areas and nitrogen contents of GN are shown in Table 1. The nitrogen contents of Pd(2.9)/GN-1, Pd(2.9)/GN-2 and Pd(2.8)/GN-3 were 5.7, 8.6 and 11.3%, respectively. In comparison with Pd/graphene, N doping led to decreased surface area, and the surface areas of Pd/GN reduced slightly with the increase of N-doping amount, probably because nitrogen doping suppressed the expanding of graphite lamella, as also indicated by the XRD results.

Table 1 Properties of graphene and GN supported Pd catalysts

Catalyst	BET surface area (m^2 g^−1)	N-doping amount^a (%)	Pd content^b (wt%)	Pd particle sizes^c (nm)
a Determined by elementary analysis. b Determined by ICP. c Calculated from TEM.
Pd(2.7)/graphene	120.5	0	2.7	3.9
Pd(2.9)/GN-1	103.5	5.7	2.9	3.5
Pd(2.9)/GN-2	92.6	8.6	2.9	3
Pd(2.8)/GN-3	85.5	11.3	2.8	3.2
Pd(0.51)/GN-3	86.7	11.3	0.51	1.7
Pd(1.47)/GN-3	90.9	11.3	1.47	2.5
Pd(4.1)/GN-3	82.4	11.3	4.1	3.9

---

The surface properties of graphene and GN were further studied using XPS, and the results are shown in Fig. 2S.† The carbon 1s signals of graphene and GN-3 were found to be centered at 284.6 eV, and the broadened carbon peak could be deconvoluted into three peaks around 284.6, 286 and 289 eV, ascribed to lattice carbon, carbon atoms singly and doubly coordinated with oxygen atoms (C–O, C=O), respectively.³⁶ The N 1s XPS spectra of GN-1, GN-2 and GN-3 are displayed in Fig. 2a. All samples gave two peaks around 398.6 and 401 eV, assigned to the nitrogen species in the form of pyridinic and graphitic type.³⁷ The pyridinic N species were bonded with two carbon atoms locating on the edges of graphite planes, and graphitic nitrogen species were associated with three carbon atoms in the graphite plane.³⁸ Notably, the content of pyridinic-N increased with N-doping amount in the catalysts.

Fig. 2 XPS profiles of (a) GN in the N 1s region, (b) Pd(2.7)/graphene, Pd(2.9)/GN-1, Pd(2.9)/GN-2 and Pd(2.8)/GN-3 in Pd 3d region, and (c) GN-3 supported Pd catalysts in the Pd 3d region.

The binding energy of Pd in the 3d region for Pd(2.7)/graphene, Pd(2.9)/GN-1, Pd(2.9)/GN-2 and Pd(2.8)/GN-3 is compared in Fig. 2b. The Pd 3d peaks of the catalysts differed in the peak width, implying the coexistence of different Pd species. Hence, the XPS peaks were also deconvoluted and the fitting parameters are listed in Table 2. The Pd²⁺/Pd⁰ ratios of Pd(2.9)/GN-1, Pd(2.9)/GN-2 and Pd(2.8)/GN-3 were determined to be 0.65, 1.12 and 0.69, respectively, much higher than that of Pd(2.7)/graphene (0.39). Notably, N doping led to the formation of surface N containing species, which could serve as the anchoring sites for Pd particles. Hence, such a higher Pd²⁺/Pd⁰ ratio in Pd/GN is likely due to a strong metal–support interaction between Pd particles and N-containing anchoring sites from GN support, resulting in the transfer of electrons from metallic Pd to support, and forming electron deficient Pd particles as well as a high Pd dispersion. Similarly, Chetty et al.³⁹ found that nitrogen species in multiwalled carbon nanotubes acted as the anchoring sites for the deposition of Pt Ru particles. In addition, XPS results showed that the Pd²⁺/Pd⁰ ratio first increased then decreased with N doping, and the maximum value was observed on Pd(2.9)/GN-2(1.12), implying that the Pd/GN catalyst had the highest metal–support interaction when N-doping amount was 8.6%. The chemical states of Pd in the Pd/GN-3 catalysts with varied Pd loadings were also analyzed using XPS spectra. The fitting lines are presented in Fig. 2c and the resulting parameters are listed in Table 2. For the Pd/GN-3 catalysts, the peak area in Pd 3d region increased with Pd loading amount. Additionally, increasing Pd loading from 0.71 to 4.1 wt% resulted in a decrease in the Pd²⁺/Pd⁰ ratio from 0.79 to 0.63, reflecting an attenuated interaction between Pd particle and the support with Pd loading amount.

Table 2 Parameters of XPS spectra of Pd catalysts supported on graphene and GN

Catalysts	Pd^2+/Pd^0	N/C
Pd(2.7)/graphene	0.39	0
Pd(2.9)/GN-1	0.65	0.046
Pd(2.9)/GN-2	1.12	0.055
Pd(2.8)/GN-3	0.69	0.087
Pd(0.51)/GN-3	0.79	0.089
Pd(1.47)/GN-3	0.74	0.09
Pd(4.1)/GN-3	0.63	0.088

---

The TEM images and histograms of Pd particle size distributions of graphene and GN supported catalysts are presented in Fig. 3. For all the catalysts, flakes-like graphene sheets decorated with fine Pd particles were clearly observed. The average Pd particle sizes of the catalysts were further calculated using eqn (1):¹⁶

where n_i is the number of Pd particles counted with diameter of d_i, and the total counted number of Pd particles is larger than 300.

Fig. 3 (a) TEM images and (b) histograms of Pd particle size distributions of the catalysts.

The distributions of Pd particles varied from the supports and Pd loading amounts (see Fig. 3b). For Pd(2.9)/GN-1, Pd(2.9)/GN-2 and Pd(2.8)/GN-3, the average Pd particle sizes were 3.5, 3.0 and 3.2 nm, respectively, smaller than that of Pd(2.7)/graphene (3.9 nm), reflecting that N-doping improved the dispersion of Pd particles. For Pd/GN catalysts, Pd particle size first decreased and then increased slightly when N doping content increased from 5.7 to 11.3%, confirming a positive impact of the metal–support interaction on Pd dispersion. Notably, Pd(2.9)/GN-2 had smaller Pd particles than Pd(2.8)/GN-3, implying that overabound nitrogen species (pyridinic or/and graphitic type) on the support surface might inhibit the dispersion of Pd particles. For the Pd/GN-3 catalysts with varied Pd loading amounts, Pd particle sizes were 1.7, 2.5, 3.2 and 4.1 nm for Pd(0.71)/GN-3, Pd(1.47)/GN-3, Pd(2.8)/GN-3 and Pd(4.1)/GN-3, respectively, suggesting that Pd particles tended to aggregate on the support surface when Pd loading increased.

3.2. Liquid phase catalytic HDC of 2,4-DCP

3.2.1. Effect of N-doping. The catalytic HDC of 2,4-DCP over Pd(2.8)/GN-3 with varied catalyst dosages was carried out to check the potential mass transfer limitation, and the results are shown in Fig. 4S, ESI.† The HDC efficiency increased with catalyst dosage, while the initial activities remained nearly identical, confirming the absence of mass transfer limitation in the reaction system.⁴⁰

The catalytic HDC curves of 2,4-DCP over Pd(2.7)/graphene and Pd(2.8)/GN-3 are presented in Fig. 5S.† For the catalytic HDC of 2,4-DCP, only 2-chlorophenol (2-CP) was identified as the partially dechlorinated product. Further hydrogenation products from phenol did not appear throughout the reaction process. The absence of 4-chlorophenol could be ascribed to the steric hindrance, by which para-substituted Cl was easier to be attacked by the active H.⁴¹ For Pd(2.7)/graphene, 2,4-DCP was removed by 77% within 120 min and 2-CP still remained in the reaction system, while complete removal of 2,4-DCP was achieved within 50 min for Pd(2.8)/GN-3, and 2-CP disappeared after reaction for 120 min. The catalytic HDC of 2,4-DCP over Pd/graphene and Pd/GN catalysts with varied N doping amounts are further compared in Fig. 4a. Notably, the catalytic activity was closely related to the supports, and all the Pd/GN catalysts exhibited much higher activities than Pd(2.7)/graphene. As evidenced by TEM and XPS analysis, GN supported Pd catalysts had smaller Pd particle sizes and higher Pd²⁺ contents, due to the presence of effective anchoring sites in N doping graphene. At an identical Pd loading amount, smaller Pd particle in supported catalyst may provide more exposed Pd sites. Additionally, the presence of Pd²⁺ facilitated the activation of C–Cl bonds,⁴² giving rise to much higher catalytic activities of Pd(2.9)/GN-1, Pd(2.9)/GN-2 and Pd(2.8)/GN-3 than that of Pd(2.7)/graphene. In addition, though the reaction rate of Pd/GN-2 is similar to the Pd catalysts reported previously under the same reaction conditions,²⁵ the catalysts preparation in this paper is much simple and practicable.

Fig. 4 (a) Catalytic HDC of 2,4-DCP over Pd(2.7)/graphene, Pd(2.9)/GN-1, Pd(2.9)/GN-2 and Pd(2.8)/GN-3. Reaction conditions: pH 12. Catalyst dosage: 0.25 g l⁻¹, 2,4-DCP concentration: 3.0 mM. (b) Initial activities of Pd catalysts supported on graphene and GN. (c) 2-CP selectivity (S_2-CP), as a function of 2,4-DCP conversion (X_2,4-DCP) for reaction over Pd catalysts supported on graphene and GN.

The role of N doping was also verified by the influence of N doping amount on the catalytic activity, and the results are compiled in Fig. 4b. The initial activities first increased and then decreased slightly with the increase of N-doping amount, for example, the initial activities were 259.8, 654.3, 820.5 and 781.1 mM g_cat⁻¹ h⁻¹ for Pd(2.7)/graphene, Pd(2.9)/GN-1, Pd(2.9)/GN-2 and Pd(2.8)/GN-3, respectively, exhibiting an activity order of Pd(2.9)/GN-2 > Pd(2.8)/GN-3 > Pd(2.9)/GN-1 > Pd(2.7)/graphene, suggesting that the catalytic activity for HDC of 2,4-DCP had a positive relationship with the metal–support interaction and Pd dispersion. In addition, this result showed that superfluous nitrogen species on the support surface are not favorable to the catalyst activity, similar results were also obtained by Chen et al.⁴³

The dependence of the HDC pathway on the N doping amounts can be demonstrated from a consideration of 2-CP selectivity (S_2-CP) as a function of 2,4-DCP consumption (X_2,4-DCP); the corresponding data for different nitrogen amounts are presented in Fig. 4c. At lower N doping amounts, the activity/selectivity profiles for Pd(2.7)/graphene, Pd(2.9)/GN-1 and Pd(2.9)/GN-2 overlapped, which suggests that for catalysts with lower nitrogen concentration (<8.6%), the reaction pathway is independent of N-doping. However, the S_2-CPvs. X_2,4-DCP profiles for Pd(2.9)/GN-2 and Pd(2.8)/GN-3 do not coincide, indicating that higher N doping amount can impact on the reaction network. The lower values of S_2-CP at a given X_2,4-DCP for Pd(2.8)/GN-3 may be taken into account based on the abstraction of the nucleophilic chloride anion in the 2-chlorophenolate species by Pd²⁺ as well as nitrogen species on the catalyst.

3.2.2. Effect of Pd loading amount. The catalytic activity of Pd/GN-3 may strongly depend on its Pd loading amount, and the catalytic HDC of 2,4-DCP over Pd/GN-3 with varied Pd loading amounts is presented in Fig. 5. When Pd loading increased from 0.71 to 1.47 wt%, the normalized initial catalytic activity increased from 226.0 to 285.6, while further increasing Pd loading level to 4.1 wt% led to decreased catalytic activity. Characterization results from TEM and XPS analysis showed that the Pd/GN-3 catalyst with a higher Pd loading had larger Pd particle size and lower Pd²⁺/Pd⁰ ratio. It was previously concluded that C–Cl bond activation and H₂ activation were two crucial steps in the liquid phase HDC reaction,⁴⁴ H₂ can be chemisorbed and dissociated on Pd⁰ to form active H, while the C–Cl bonds of 2,4-DCP can be activated dissociatively on Pdⁿ⁺ sites by abstraction of the nucleophilic chloride anion (Cl⁻), thus Pd²⁺ and Pd⁰ were respective active sites. Hence, smaller Pd particles with higher Pd²⁺ content favored C–Cl bond activation, while larger Pd particles with higher Pd⁰ content facilitated H₂ activation by enhancing H₂ solubility and forming β-PdH phase.⁴⁵ Accordingly, for the Pd/GN-3 catalyst increasing Pd loading led to enhanced H₂ activation, but to weakened C–Cl bond activation for the catalytic HDC of 2,4-DCP, thus giving a volcano-type dependence of catalytic activity on Pd loading amount.

Fig. 5 (a) The catalytic HDC of 2,4-DCP on GN-3 supported catalysts with varied Pd loading amounts. (b) Dependence of normalized initial activity on Pd particle size. Reaction conditions: pH 12. Catalyst dosage: 0.25 g l⁻¹.

3.2.3. Effect of initial 2,4-DCP concentration. For a heterogeneous catalytic reaction, catalysis is expected to occur on catalyst surface, and thus the adsorption of reactant is believed to be a prerequisite step. To gain the insight into the reaction mechanism, catalytic HDC of 2,4-DCP with varied initial concentrations on Pd(2.8)/GN-3 was conducted and the results are shown in Fig. 6a. The initial activity increased gradually with the initial concentration, suggesting that the reaction efficiency was positively correlated with 2,4-DCP concentration absorbed on the catalyst surface. The experimental data were further fitted using the Langmuir–Hinshelwood model:⁴⁶where r₀ is the initial reaction rate, C₀ is the initial concentration of 2,4-DCP, θ_s is the coverage of 2,4-DCP on the catalyst surface, k is the reaction rate constant, and b is the adsorption constant.

Fig. 6 (a) The catalytic HDC of 2,4-DCP at varied initial 2,4-DCP concentrations. (b) Linear plot of 1/r₀versus 1/C₀. Reaction conditions: pH 12. Catalyst: Pd(2.8)/GN-3. Catalyst dosage: 0.25 g l⁻¹. Line represents the fitting curve using the Langmuir–Hinshelwood model.

The plot of 1/r₀ as a function of 1/C₀ is presented in Fig. 6b. A good liner relationship between 1/r₀ and 1/C₀ was observed with R² higher than 0.98, reflecting that the HDC of 2,4-DCP obeyed the Langmuir–Hinshelwood model and the surface adsorption was the rate-controlling step.⁴⁷

3.2.4. Reaction kinetics. The catalytic HDC of polychlorinated organic pollutants could be achieved through the stepwise or/and concerted pathway,⁴⁸ and the possible dechlorination pathways are shown in Scheme 1. The reaction rate equations for the liquid phase catalytic HDC of 2,4-DCP could be depicted as follows:

Scheme 1 Reaction pathways of the liquid phase catalytic HDC of 2,4-DCP.

Accordingly, the rate constants could be calculated according to eqn (9)–(11):

C_2,4-DCP = C⁰_2,4-DCP exp(−(k₁ + k₂)t) (9)

where C⁰_2,4-DCP is the initial 2,4-DCP concentration, C_2,4-DCP, C_2-CP and C_phenol are the concentrations of 2,4-DCP, 2-CP and phenol at reaction time t, k₁, k₂ and k₃ are the rate constants of 2,4-DCP to 2-CP, 2,4-DCP to phenol, and 2-CP to phenol, respectively.

The rate constants were obtained by integrating and data fitting, and the fitting parameters are listed in Table 3. All the fitting correlation coefficients are very high, and the rate constants are greater than zero, indicating that the HDC of 2,4-DCP was completed via both stepwise and concerted pathways. The reaction rates of 2,4-DCP over Pd(2.7)/graphene, Pd(2.9)/GN-1, Pd(2.9)/GN-2 and Pd(2.8)/GN-3 were 0.0224, 0.0873, 0.2012 and 0.1666 min⁻¹ respectively, in agreement with the activity order of Pd(2.9)/GN-2 > Pd(2.8)/GN-3 > Pd(2.9)/GN-1 > Pd(2.7)/graphene. Additionally, k₂/k₁ ratios were 0.35, 0.7, 1.01 and 1.41 for Pd(2.7)/graphene, Pd(2.9)/GN-1, Pd(2.9)/GN-2 and Pd(2.8)/GN-3 respectively, clearly indicating that the concerted dechlorination pathway became predominant with the increase of N-doping amount in the catalyst. The higher contents of cationic Pd and nitrogen species may facilitate the simultaneous activation of ortho- and para-substitutional groups, favoring enhanced dechlorination by the concerted pathway.

Table 3 Fitting rate constants of the catalytic HDC of 2,4-DCP over Pd catalysts supported on graphene and GN

Catalyst	k 1 (min^−1)	k 2 (min^−1)	k 3 (min^−1)	k 1 + k2 (min^−1)	k 2/k1
Pd(2.7)/graphene	0.0166	0.0058	0.0038	0.0224	0.35
Pd(2.9)/GN-1	0.0514	0.0359	0.0117	0.0873	0.70
Pd(2.9)/GN-2	0.1	0.1012	0.0263	0.2012	1.01
Pd(2.8)/GN-3	0.069	0.0976	0.0304	0.1666	1.41

---

To verify possible catalyst deactivation, catalyst reuse was conducted without re-activation and the results are compiled in Fig. 7. For Pd(2.8)/GN-3, gradual deactivation was observed with catalyst recycle, and after 5 cycles the catalytic activity decreased by 23%.

Fig. 7 Catalytic HDC of 2,4-dichlorophenol on Pd(2.8)/GN-3 over five cycles.

4. Conclusions

In the present study, a series of Pd catalysts supported on graphene and GN with varied nitrogen doping amounts were prepared and the liquid phase catalytic HDC of 2,4-DCP was investigated. At similar Pd loadings, the catalysts with N-doped graphene as the supports exhibited much higher catalytic activities than Pd/graphene due to their higher Pd²⁺ contents and Pd dispersions. Additionally, increasing N doping amount led to first increased and then decreased catalytic activity of Pd/GN. The catalytic HDC of 2,4-DCP followed the Langmuir–Hinshelwood model, reflecting an adsorption controlled reaction mechanism. For all catalysts, HDC of 2,4-DCP proceeded via both stepwise and concerted pathways, and the concerted pathway became more pronounced with the increase of N-doping amount in the catalyst support. Furthermore, for the Pd/GN-3 catalysts a volcano type dependency of catalytic activity on Pd particle size was observed. The present findings highlight the superior role of N doping in the enhancement of the activity of Pd/graphene for the liquid phase catalytic HDC of chlorinated organic pollutants.

Acknowledgements

The financial support from National Key Basic Research Program of China (2014CB441103), the National Natural Science Foundation of China (no. 21277066, 21407074 and 51508080), and the National Natural Science Foundation of Jiangsu (BK20140598) is gratefully acknowledged. We are indebted to the Modern Analytical Center, Donghua University for the catalyst characterization.

References

1. G. Q. Chen, S. Guan, G. M. Zeng, X. D. Li, A. W. Chen, C. Shang, Y. Zhou, H. K. Li and J. M. He, Appl. Microbiol. Biotechnol., 2013, 97, 3149–3157.
2. Y. Q. Peng and G. H. Song, Green Chem., 2002, 4, 349–351.
3. R. Andreozzi, I. D. Somma, R. Marotta, G. Pinto, A. Pollio and D. Spasiano, Water Res., 2011, 45, 2038–2048.
4. P. T. Marc, G. M. Veronica and A. B. Miguel, Appl. Catal., B, 2004, 47, 219–246.
5. S. Dhanus and H. Kiyoshi, J. Hazard. Mater., 2010, 183, 490–496.
6. R. Weber, T. Sakurai and H. Hagenmaier, Appl. Catal., B, 1999, 20, 249–256.
7. M. Henry, A. Farrell and Q. Brid, J. Environ. Sci. Health, Part A, 2006, 41, 763–777.
8. J. H. Kim and S. T. Lee, Process Biochem., 2002, 37, 1367–1373.
9. J. Zhou, K. Wu, W. J. Wang, Y. X. Han, Z. Y. Xu, H. Q. Wan, S. R. Zheng and D. Q. Zhu, Appl. Catal., B, 2015, 162, 85–92.
10. C. Menini, C. Park, E. J. Shin, G. Tavoularis and M. A. Keane, Catal. Today, 2000, 62, 355–366.
11. C. B. Molina, A. H. Pizarro, J. A. Casas and J. J. Rodriguez, Appl. Catal., B, 2014, 148–149, 330–338.
12. S. Ordonez, B. P. Vivas and F. V. Diez, Appl. Catal., B, 2010, 95, 288–296.
13. Z. M. Pedro, E. Diaz, A. F. Mohedano, J. A. Casas and J. J. Rodriguez, Appl. Catal., B, 2011, 103, 128–135.
14. E. Diaz, A. F. Mohedano, J. A. Casas, L. Calvo, M. A. Gilarranz and J. J. Rodriguez, Catal. Today, 2015, 241, 86–91.
15. S. Gómez-Sainero, F. Cardenas-Lizana and M. A. Keane, Ind. Eng. Chem. Res., 2008, 47, 6841–6853.
16. G. Yuan and M. A. Keane, Appl. Catal., B, 2004, 52, 301–314.
17. Y. Shao, Z. Y. Xu, H. Q. Wan, Y. Q. Wan, H. Chen, S. R. Zheng and D. Q. Zhu, Catal. Commun., 2011, 12, 1405–1409.
18. C. E. Chan-Thaw, A. Villa, P. Katekomol, D. S. Su, A. Thomas and L. Prati, Nano Lett., 2010, 10, 537–541.
19. S. H. Liu, M. T. Wu, Y. H. Lai, C. C. Chiang, N. Yu and S. B. Liu, J. Mater. Chem., 2011, 21, 12489–12496.
20. E. Rangel and E. Sansores, Int. J. Hydrogen Energy, 2014, 39, 6558–6566.
21. V. B. Parambhath, R. Nagar and S. Ramaprabhu, Langmuir, 2012, 28, 7826–7833.
22. Y. A. Lv, Y. H. Cui, Y. Z. Xiang, J. G. Wang and X. N. Li, Comput. Mater. Sci., 2010, 48, 621–625.
23. Z. Y. Liu, C. L. Zhang, L. Luo, Z. Chang and X. M. Sun, J. Mater. Chem., 2012, 22, 12149–12154.
24. K. Chizari, I. Janowska, M. Houlle, I. Florea, O. Ersen, T. Romero, P. Bernhardt, M. J. Ledoux and C. Pham-Huu, Appl. Catal., A, 2010, 380, 72–80.
25. J. Zhou, K. Wu, W. J. Wang, Z. Y. Xu, H. Q. Wan and S. R. Zheng, Appl. Catal., A, 2014, 470, 336–343.
26. L. F. Dong, R. R. S. Gari, Z. Li, M. M. Craig and S. F. Hou, Carbon, 2010, 48, 781–787.
27. L. H. Tang, Y. Wang, Y. M. Li, H. B. Feng, J. Lu and J. H. Li, Adv. Funct. Mater., 2009, 19, 2782–2789.
28. M. J. Lopez, I. Cabria and J. A. Alonso, J. Phys. Chem. C, 2014, 118, 5081–5090.
29. D. Chen, H. B. Feng and J. H. Li, Chem. Rev., 2012, 112, 6027–6053.
30. Q. Z. Sun and S. Kim, Electrochim. Acta, 2015, 153, 566–573.
31. Y. M. Li, L. H. Tang and J. H. Li, Electrochem. Commun., 2009, 11, 846–849.
32. D. Hulicova-Jurcakova, M. Seredych, G. Q. Lu and T. J. Bandosz, Adv. Funct. Mater., 2009, 19, 438–447.
33. G. Yuan and M. A. Keane, Chem. Eng. Sci., 2003, 58, 257–267.
34. G. Yuan and M. A. Keane, Ind. Eng. Chem. Res., 2007, 46, 705–715.
35. P. Wu, Y. Y. Huang, L. Q. Zhou, Y. B. Wang, Y. K. Bu and J. N. Yao, Electrochim. Acta, 2015, 152, 68–74.
36. S. L. Ding, S. J. Zheng, M. J. Xie, L. M. Peng, X. F. Guo and W. P. Ding, Microporous Mesoporous Mater., 2011, 142, 609–613.
37. X. W. Wang, G. Z. Sun, P. Routh, D. H. Kim, W. Huang and P. Chen, Chem. Soc. Rev., 2014, 43, 7067–7098.
38. L. S. Zhang, X. Q. Liang, W. G. Song and Z. Y. Wu, Phys. Chem. Chem. Phys., 2010, 12, 12055–12059.
39. R. Chetty, S. Kundu, W. Xia, M. Bron, W. Schuhmann, V. Chirila, W. Brandl, T. Reinecke and M. Muhler, Electrochim. Acta, 2009, 54, 4208–4215.
40. O. M. Ilinitch, F. P. Cuperus, L. V. Nosova and E. N. Gribov, Catal. Today, 2000, 56, 137–145.
41. C. H. Xia, Y. Liu, J. Xu, J. B. Yu, W. Qin and X. M. Liang, Catal. Commun., 2009, 10, 456–458.
42. S. Arulmozhiraja and M. Morita, J. Phys. Chem. A, 2004, 108, 3499–3508.
43. H. Chen, P. Zhang, W. H. Tan, F. Jiang and R. Tang, RSC Adv., 2014, 4, 38743–38749.
44. L. M. Gómez-Sainero, X. L. Seoane, J. L. G. Fierro and A. Arcoya, J. Catal., 2002, 209, 279–288.
45. M. A. Aramendia, V. Borau, I. M. Garcia, C. Jimenez, J. M. Marinas and F. J. Urbano, Appl. Catal., B, 1999, 20, 101–110.
46. A. Pintar, J. Batista, J. Levec and T. Kajiuchi, Appl. Catal., B, 1996, 11, 81–98.
47. X. F. Xue, K. Hanna, M. Abdelmoula and N. S. Deng, Appl. Catal., B, 2009, 89, 432–440.
48. J. Zhou, Y. X. Han, W. J. Wang, Z. Y. Xu, H. Q. Wan, D. Q. Yin, S. R. Zheng and D. Q. Zhu, Appl. Catal., B, 2013, 134–135, 222–230.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra17946g

---