稀有金属(英文版) 2019,38(01),1-13

Recovery and regeneration of Al₂O₃ with a high specific surface area from spent hydrodesulfurization catalyst CoMo/Al₂O₃

Qi Liu Wen-Qiang Wang Yue Yang Xue-Gang Liu Sheng-Ming Xu

Institute of Nuclear and New Energy Technology, Tsinghua University

School of Minerals Processing and Bioengineering, Central South University

Beijing Key Laboratory of Fine Ceramics,Tsinghua University

作者简介：*Sheng-Ming Xu e-mail:smxu@tsinghua.edu.cn;

收稿日期：11 January 2018

基金：financially supported by the National Natural Science Commission-Yunnan Joint Fund Project (No.U1402274);

Recovery and regeneration of Al₂O₃ with a high specific surface area from spent hydrodesulfurization catalyst CoMo/Al₂O₃

Qi Liu Wen-Qiang Wang Yue Yang Xue-Gang Liu Sheng-Ming Xu

Institute of Nuclear and New Energy Technology, Tsinghua University

School of Minerals Processing and Bioengineering, Central South University

Beijing Key Laboratory of Fine Ceramics,Tsinghua University

Abstract：

Aluminum recovery is a key issue for the overall recycling of valuable metals from spent catalysts. This paper focuses on the recovery and regeneration of alumina with high additional value from the spent hydrodesulfurization catalyst CoMo/Al₂O₃. The results indicate that 98.13% alumina is successfully leached from the treated spent catalysts by an alkaline leaching process under the conditions of 5 mol·L^-1 sodium hydroxide,a liquid/solid ratio of 20 ml·g^-1,a temperature of 160 0 C and a reaction time of 4 h. In the leaching residue, no difficult leaching compound is found and cobalt and nickel are enriched,both of which are conducive to the subsequent metal recovery step. The reaction order of aluminum leaching is 0.99. This reaction fits well with the interfacial chemical reaction model, and its apparent activation energy is calculated as 45.50 kJ mol^-1. Subsequently, y-Al₂O₃ with a high specific surface area of 278.3 m²·g^-1, a mean size of 2.2 μm and an average pore size of 3.10 nm is then regenerated from the lixivium, indicating its suitability for use as a catalyst carrier. The recovery and regeneration of alumina from spent catalysts can not only significantly contribute to the total recycling of such hazardous spent catalysts but also provide a new approach for the preparation of y-Al₂O₃ with a high specific surface area using spent catalysts as the aluminum sources.

Keyword：

Spent catalysts; CoMo/Al₂O₃; Aluminum leaching; Regeneration; γ-Al₂O₃;

Received： 11 January 2018

1 Introduction

The consumption of oil products in China continues to increase with the rapid growth of car ownership.The sulfur contents in crude oil result in equipment corrosion and vehicle exhaust pollution if they are not efficiently removed [ 1, 2, 3] .Hydrodesulfurization (HDS) is one of the most essential procedures in the petroleum refining industry to remove undesirable impurities,such as sulfur and nitrogen [ 4, 5, 6] ,in which CoMo/Al₂O₃ catalysts,as the most commonly used HDS catalysts,are widely used.These catalysts usually lose catalytic activity and turn into waste after a certain period of use due to coking,sintering or depositing of heavy metals [ 7, 8] .The spent HDS catalysts contain approximately 15%-30%Al,4%-12%Mo and 0%-4%Co or Ni [ 9] .These metal concentrations are much higher than those in primary ores.In addition,some toxic elements (Ni,Co,etc.) within the spent HDS catalysts are considered as hazardous materials by the environmental protection agency (EPA) [ 8] .Thus,the proper safe treatment and disposal are essential.To this end,the recovery and recycling of valuable metals from spent HDS catalysts are promising solutions to relieve economic and environmental concerns.

There are many comprehensive studies focused on recovering valuable metals from spent HDS catalysts.Hydrometallurgical,pyrometallurgical and biohydrometallurgical methods are the three main approaches to deal with spent HDS catalysts [ 10, 11, 12, 13, 14] .Among these,the hydrometallurgical method has attracted particular attention because of its unique advantages,such as controllable process scale,easier operation,lower gas emissions,higher metal recovery and smaller amounts of secondary waste [ 9, 15] .The recovery of metals from spent HDS catalysts by hydrometallurgy usually consists of leaching,valuable metal separation and products preparation [ 12, 16, 17, 18] .Final products such as CaMoO₄ [ 19] ,MoO₃ [ 20] ,PbMoO₄ [ 21] ,V₂O₅ [ 22, 23, 24] ,CoC₂O₄ [ 25] ,CoSO₄ and NiSO₄ [ 26] can be successfully produced.However,this method generates hard-to-recycle impurities and toxic gases because of the uncontrollable temperature resulting from improper pretreatment.Alumina,acting as the carrier of HDS catalysts,has been reclaimed from spent catalysts as fused alumina [ 27] or aluminum salts,such as NH₄Al(-SO₄)2.12H₂O [ 28] and Al₂(SO₄)₃ [ 29] ,which are low value-added products.Impurities are introduced because most of the above-mentioned substances are recovered by acid leaching.Thus,the recycling value and the utilization efficiency are not high.Surprisingly,little effort was made in the previous studies to recover the alumina by an alkaline leaching process to produceγ-Al₂O₃,which is the main majority of the catalyst carrier and has very high specific surface area,good adsorption properties and thermal stability [ 30, 31, 32] .In conclusion,the main bottleneck impeding the industrial recovery of spent HDS catalysts is the lack of an overall recycling procedure to extract valuable metals,especially aluminum,from spent catalysts.

Our research group has been working on the overall recycling of spent catalysts for many years.A de-oiling technology was developed to ensure precise temperature control and avoid the generation of difficult leaching impurities during the recovering process [ 33] .It laid a good foundation for the subsequent recovery and regeneration of alumina in spent HDS catalysts.This work aims to establish an economical and sustainable process for the recovery of alumina and preparation ofγ-Al₂O₃ with excellent properties from spent HDS catalysts.The effects of the liquid/solid (L/S) ratio,concentration of leaching agent,temperature and leaching time were investigated.Orthogonal experiments were conducted to determine the optimized conditions for aluminum leaching.Leaching kinetics experiments were performed to obtain the macro-kinetics equation for the aluminum leaching process.γ-Al₂O₃ was prepared from lixivium and systematically characterized by scanning electron microscopy (SEM),energy dispersive X-ray spectroscopy (EDS),X-ray diffraction (XRD),X-ray photoelectron spectroscopy (XPS) and Brunauer-EmmettTeller (BET) N₂ gas absorption-desorption analysis.The results indicate that alumina can be efficiently recovered and the product shows excellent performance.The recovery process can not only promote the total recycling of metals from spent HDS catalysts but also create a new approach for the preparation ofγ-Al₂O₃ with a high specific surface area using spent catalysts as the aluminum source.

2 Experimental

2.1 Materials and chemicals

The spent HDS catalyst CoMo/Al₂O₃ studied in this work was provided by SINOPEC Beijing Yanshan Company,China.It was de-oiled,roasted and leached according to the methods illustrated in a previous report [ 33] .Hydrochloric acid (HCl),nitric acid (HNO₃),sodium hydroxide (NaOH),urea (NH₂CONH₂) and polyethylene glycol (PEG-1000)were purchased from Aladdin Industrial Corporation.All reagents were of analytical grade,and all solutions in their specified concentrations were prepared using deionized water.

2.2 Aluminum leaching

The treated spent HDS catalyst was dissolved by aqua regia and then analyzed by inductively coupled plasma optical emission spectrometry (ICP-OES,VARIAN,VISTA-MAPX CCD) to obtain the mass of aluminum in the treated spent HDS catalysts.Then,10 g treated spent catalyst CoMo/Al₂O₃ was removed using sodium hydroxide in a500-ml autoclave.The autoclave coupled with a stirrer(Easy Chem Technology Development Co.,Ltd.,AC-500)was heated by a thermocouple with automatic temperature control.The leaching process was conducted at an agitation speed of 300 r·min^-1.The trends in the aluminum leaching efficiency with the concentration of sodium hydroxide,liquid-solid (L/S) ratio,reaction time and temperature were obtained by means of single-factor aluminum leaching experiments.Orthogonal experimental design is one of the most effective methods involving multiple variables to study which and how factors can affect the target properties of a product [ 34] .In this paper,orthogonal experiments were conducted to determine the optimized conditions for aluminum leaching.

To explore the kinetics of aluminum leaching,the dependence of reaction time on leaching efficiency was investigated at different concentrations of sodium hydroxide,temperatures and L/S ratios.Once the experiments reached the preset reaction time,a certain amount of leaching solution was extracted and then filtered and cooled for analysis.

The concentration of aluminum in the leaching solution was determined by ICP-OES.The aluminum leaching efficiency can be calculated according to Eq.(1).

whereηis aluminum leaching efficiency;m₁ is mass of aluminum in the leaching solution,g;m₂ is mass of aluminum in the treated spent catalysts,g.

2.3 Regeneration ofγ-Al2O3

The extracted solution,as the source of aluminum,was diluted to 0.25 mol·L^-1.Then,urea (NH₂CONH₂) was used as the precipitant,and the molar ratio of the aluminum source to the precipitant was adjusted to 4.Polyethylene glycol PEG-1000 with a concentration of 0.15 mol·L^-1was used as the surface active agent.The reaction generated in the reactor was incubated at 80℃for 6 h.After washing,centrifugation and drying,the product was incubated for 4 h at 500℃in a muffle furnace under air atmosphere.

2.4 Measurement and characterization

The compositions and contents of the leaching residue under the optimized leaching conditions were analyzed by ICP-OES (VARIAN,VISTA-MAPX CCD).The phase and structure of the leaching residue under the optimized leaching conditions were analyzed by XRD (Rigaku D/Max 2500 PC,Japan) using Cu Kαradiation(λ=0.15406 nm).The elemental maps of the treated spent HDS catalyst and the leaching residue under the optimized leaching conditions were characterized by EDS (Thermo).The surface morphology of obtained products was characterized by SEM (MERLIN).The elemental composition of the obtained products was determined by EDS(Thermo).The phase and structure of the products were analyzed by XRD (Rigaku D/Max 2500 PC,Japan) using Cu Kαradiation (λ=0.15406 nm).The chemical and electronic states of the elements were analyzed by XPS with an Axis Ultra-Kratos spectrometer using monochromatic Al Kαradiation (hv=1486.6 eV) equipped with a hemispherical analyzer operating at fixed pass energy of40 eV.The specific surface area and pore size distribution of the products were examined by BET N₂ gas absorptiondesorption analysis at 77 K (Micromeritics ASAP 2020,USA).

3 Results and discussion

3.1 Aluminum leaching

ICP-OES analytical results of the treated spent HDS catalysts are shown in Table 1.As seen,the main metal elements include Al,Ni and Co.The recovery of aluminum as well as the enrichment of cobalt and nickel can be achieved by alkaline leaching [ 35] .However,the alumina leaching efficiency in sodium hydroxide solution is very low at normal temperature and pressure.To improve the alumina leaching efficiency and reduce the consumption of sodium hydroxide,the ideal pressure and temperature for alkaline leaching need to be found.The primary reaction occurring in this process is shown in Eq.(2).

  下载原图

Table 1 Components of treated spent HDS catalysts (wt%)

Single-factor aluminum leaching experiments were conducted to investigate the impacts on aluminum leaching efficiency resulted from the NaOH concentration,L/S ratio,reaction time and reaction temperature,and the results are demonstrated in Fig.1.The effect of the NaOH concentration on the aluminum leaching efficiency was tested in the range of 1.5 to6.0 mol·L^-1 NaOH for 3 h by maintaining the temperature at 160℃and L/S ratio at 10 ml·g^-1 (Fig.1a).As NaOH concentration increases,the aluminum leaching efficiency keeps increasing at a decreasing rate.When the concentration of sodium hydroxide reaches 5 mol·L^-1,the aluminum leaching efficiency is almost unchanged.So,the appropriate concentration of sodium hydroxide is5 mol·L^-1.Figure 1b shows the aluminum leaching efficiency under different L/S ratios after 3 h with5 mol·L^-1 NaOH at 160℃.It can be seen that L/S ratio of 20 ml·g^-1 is optimal,as change in aluminum leaching efficiency becomes insignificant with L/S ratio further increasing.When L/S ratio increases,the leaching solution tends to have a higher degree of unsaturation,which is to be beneficial for the leaching reaction as well as its mass transfer and heat transfer [ 36] .The effect of reaction time on aluminum leaching efficiency was investigated under the condition of 20 ml·g^-1 L/S ratio,5 mol·L^-1 NaOH and160℃(Fig.lc).The result shows that the aluminum leaching efficiency increases with reaction time for the first4 h and then remains unchanged after that.This is probably because the maximum amount of alumina in the treated spent catalysts was dissolved after 4 h,and thus,the aluminum leaching efficiency is not further promoted with leaching time prolonging.Figure 1d shows the effect of reaction temperature on aluminum leaching efficiency after4 h with a 20 ml·g^-1 L/S ratio and 5 mol·L^-1 NaOH.It can be seen that the aluminum leaching efficiency scales up before 160℃and then remains unchanged as temperature goes up.The increase in temperature is conducive increasing the diffusion rate and chemical reaction rate.However,when the temperature is too high,the evaporation of water leads to the increase in solution viscosity and the decrease in the amount of effective alkali reacting with the spent catalysts.At this time,the effect of the further increase in the temperature on the aluminum leaching efficiency is little [ 37] .

Fig.1 Effect of a NaOH concentration,b liquid-solid (L/S) ratio,c reaction time and d reaction temperature on aluminum leaching efficiency

The L9 orthogonal array of the Taguchi method [ 34] was chosen in this paper to investigate the sequence of factors affecting the aluminum leaching efficiency and the optimum leaching process conditions.The factors and levels of the designed orthogonal experiments are listed in Table 2.Four factors were considered as follows:NaOH concentration,L/S ratio,reaction time and temperature.For each factor,three levels were selected based on previous single-factor experiments,and the aluminum leaching efficiency was taken as the measured indicator.Table 3shows nine experimental runs derived from the L9orthogonal array design and the results of the measured indicator.It can be seen that R(D)>R(B)>R(A)>R(C),which indicates that the sequence of every factor for aluminum leaching efficiency is reaction temperature,L/S ratio,NaOH concentration and reaction time.Figure 2shows the trend in the experimental indicator based on the experimental factors.It can be seen that to achieve the maximum aluminum leaching efficiency,the optimized leaching condition should be A₃B₂C₂D₃,which means5 mol·L^-1 NaOH,L/S ratio of 20 ml·g^-1,reaction time of4 h and reaction temperature of 160℃.Three parallel experiments were conducted under the optimized leaching conditions.The results show that the average aluminum leaching efficiency is 98.13%.

Figure 3 shows XRD patterns of the samples before leaching and the residues under the optimized leaching conditions.The results indicate that the alumina and residual molybdenum in the treated spent catalysts were substantially leached,and the difficult leaching compound of cobalt and nickel is not found in the leaching residues,which is favorable for the subsequent acid leaching of cobalt and nickel.Figures 4 and 5 show SEM images and EDS results of the treated spent HDS catalyst CoMo/Al₂O₃and the leaching residues under the optimized conditions,respectively.Figure 5 shows that there is still a small amount of aluminum in residue,which is probably due to the formation of poorly solubleα-Al₂O₃ during the roasting pretreatment of the spent catalysts.By comparing Fig.4with Fig.5,it is clear that the aluminum is well leached and cobalt and nickel are enriched.The leaching residues were analyzed by ICP-OES to determine the composition,and the results are shown in Table 4.It can be seen that the leaching experiments not only achieve satisfied leaching performance for aluminum but also realize the enrichment of cobalt and nickel,which corresponds to the results in Figs.4 and 5.Cobalt and nickel in the leaching residue can be easily separated and recycled in consequent procedures.According to the procedure reported in Ref. [ 38] ,the leaching residue is first dissolved in 30%sulfuric acid.The recovery rates of 98.2%and 98.5%are obtained,respectively,for nickel and cobalt at 80℃with a L/S ratio of8 ml·g^-1,reaction time of 4 h and stirring rate of 800r·min^-1.This process provides the basis for subsequent alumina regeneration.

  下载原图

Table 2 Factors and levels of designed orthogonal experiments

  下载原图

Table 3 Design matrix based on an L9 orthogonal array and results of measured indicator

Fig.2 Trend of experimental indicator based on experimental factors

Fig.3 XRD patterns of samples before leaching and residues after leaching

Fig.4 a SEM image,elemental mappings (b O,c Al,d Co and e Ni) and f EDS results of treated spent HDS catalyst CoMo/Al₂O₃

Fig.5 a SEM image,elemental mappings (b O,c Al,d Co and e Ni) and f EDS results of leaching residue (leaching conditions of 5 moI·L^-1NaOH,160℃,L/S ratio of 20 ml.g^-1 and 4h)

  下载原图

Table 4 Components of residues after aluminum leaching under optimal conditions (wt%)

3.2 Kinetic analysis of aluminum leaching

According to Eq.(2),the aluminum leaching process can be considered as a liquid-solid non-catalytic reaction.It starts on the solid surface of the spent catalyst and gradually penetrates into the inside of solid.No solid layer is generated during the chemical reaction.Therefore,the shrinkage particle model with a reduced unreacted core is chosen to describe the reaction.The reaction steps include(1) diffusion of the liquid reactants or products through the liquid boundary layer and (2) the interfacial reaction.The spent catalyst (solid reactant) is assumed to be a sphere,and the interfacial chemical reaction is assumed to be an irreversible reaction.Thus,the two kinetic equations are as follows:

The diffusion rate of the liquid reactant A through the liquid boundary layer is:

(2) The interfacial reaction rate is:

where n_A is the amount of substance of the liquid reactant A,mol;t is the reaction time,h;r₀ is the initial reaction radius of the solid particles,mm;D is the effective diffusion coefficient of reactant A through the liquid boundary layer,mm²·s^-1;δis the thickness of the liquid boundary layer,mm;C_A is the concentration of liquid reactant A at the liquid boundary layer,mol·L^-1;C_AS is the concentration of liquid reactant A at the interface of the unreacted core,mol·L^-1;r is the reaction radius of the unreacted solid particles,mm;and k_r is the reaction rate constant of the interfacial chemical reaction [ 39] .

At quasi-steady state,assuming that the concentration of the liquid reactant is kept constant,two control step equations can be derived from Eqs.(3) and (4) by integration,i.e.,the liquid film diffusion control and the interfacial chemical reaction control.The equations are as follows:

where x is the aluminum leaching efficiency,k_i is the constant of the apparent diffusion rate and k_e is the constant of the apparent reaction rate.

Figure 6 shows the relationship between the aluminum leaching efficiency and the reaction time with different concentrations of sodium hydroxide at 160℃with a L/S ratio of 20 ml·g^-1.The experimental results were substituted into Eqs.(5) and (6).As presented in Fig.7,a linear relationship between 1-(1-x)^1/3 and t is found.The fitted lines do not pass through the origin.The possible reason is that some of alumina is dissolved before reaching the set temperature.Figure 7 indicates that the leaching process conforms to the model of interfacial chemical reaction control.In fact,the slope of the straight line is the reaction rate (v) at the corresponding concentration.According to (where k is the reaction rate constan,C_Ab is the concentration of sodium hydroxide and n is the reaction order),a plot of lnv versus lnC_Ab is obtained and presented in Fig.8.Linear regression was performed on the straight line in Fig.8,and the slope is0.99.This indicates that the aluminum leaching reaction is a first-order reaction.

Fig.6 Relationship between aluminum leaching efficiency and reaction time with different concentrations of sodium hydroxide

Fig.7 Leaching dynamics curves for different concentrations of sodium hydroxide

Fig.8 Relationship between lnv and ln C_Ab during leaching

Figure 9 shows the relationship between the aluminum leaching efficiency and the reaction time at different temperatures with 5 mol·L^-1 NaOH and a L/S ratio of20 ml·g^-1.The experimental data at different leaching temperatures were substituted into Eqs.(5) and (6).The results show a linear relationship between 1-(1-x)^1/3and t,as presented in Fig.10.Therefore,the interfacial chemical reaction control model is proven to be applicable for the aluminum leaching process.

Figure 1 1 shows the relationship between the aluminum leaching efficiency and the reaction time at different L/S ratios and specific conditions,160℃and 5 mol·L^-1NaOH.The experimental data were substituted into Eq.(6),and it can be seen that the relationship between1-(1-x)^1/3 and t is still linear,as presented in Fig.12.This result confirms our previous conclusion that the aluminum leaching process is controlled by an interfacial chemical reaction.

The Arrhenius formula [ 40] can be used to determine the macro-kinetics equation.Taken the Napierian logarithm of both sides of the formula into account,Eq.(7) can be derived:

where k is the reaction rate constant,s^-1;A is the preexponential factor,s^-1;E is the apparent activation energy,kJ·mol^-1;R is the gas constant,J.mol^-1·K^-1;and T is the thermodynamic temperature,K.Linear regressions were performed on the straight line in Fig.10,and the reaction rate constants (k) under different temperatures were calculated as the slope of the straight line.The relationship between ln k and T^-1 is shown in Fig.13.It can be seen from Fig.13 that the slope of the line is fitted as-5472.79and the intercept is 10.922.The results were substituted into Eq.(7),resulting in E_a=45.50 kJ·mol^-1 and A=5.538×10³ s^-1.The above apparent activation energy is within the typical apparent activation energy range of the chemical reaction,which further proves that the aluminum leaching process is controlled by the interfacial chemical reaction.

Fig.9 Relationship between aluminum leaching efficiency and reaction time at different temperatures

Fig.10 Relationship between 1-(1-x)^1/3 and reaction time at different temperatures

Fig.11 Relationship between aluminum leaching efficiency and reaction time with different L/S ratios

Fig.12 Relationship between 1-(1-x)^1/3 and reaction time with different L/S ratios

Fig.13 Relationship between ln k and T^-1 during leaching

The above experimental results were substituted into Eq.(6) to give the concrete expression (Eq.8) of the macro-kinetics equation of the aluminum leaching process controlled by an interfacial chemical reaction.

3.3 Regeneration and characterization of y-Al2O3

As a carrier for many industrial catalysts,γ-Al₂O₃ is usually prepared from the precursor,boehmite by calcination.Boehmite is often produced by precipitation from sodium aluminate or aluminum salts under hydrothermal conditions.It has multiple shapes or a variety of crystallite sizes,which can be obtained under different hydrothermal conditions from aluminum nitrate or sodium aluminate [ 41, 42, 43, 44] .Boehmites can be precipitated hydro thermally from different inorganic aluminum salt solutions,such as Al(NO₃)₃.9H₂O [ 45] ,AlCl₃ and Al₂(SO₄)₃ [ 46] ,using urea as the precipitating agent,and then,γ-Al₂O₃ can be prepared through the present hydrothermal route.However,there are very few studies in the public literature on the recovery of the alumina in the form ofγ-Al₂O₃ from spent HDS catalysts.

In the present study,aluminum monohydroxide(AlOOH) was obtained from the leaching solution of treated spent catalyst CoMo/Al₂O₃ using urea as the precipitant and polyethylene glycol as the surfactant.The main chemical reactions occurred according to Eqs.(9)-(12).After washing,centrifugation and drying,the AlOOH was incubated for 4 h at 500℃in a muffle furnace under air atmosphere.

Figure 14a,b illustrates that the regenerated products are near-spherical aggregates with uniform particle size,excellent dispersibility and no agglomeration.At the same time,the bounders among the primary particles existing in the regenerated products are clear,implying the high crystallinity.EDS was carried out to analyze the elementary composition of the regenerated products.As seen from Fig.14c,the regenerated products contain only two elements,A1 and O,and the atomic ratio of Al and O is nearly 3:2,which is consistent with that in Al₂O₃.This shows that the regenerated product is Al₂O₃ with little impurities.The results also show that the particle size distribution of the regenerated products is in a small range,with the largest size of 6μm and a mean size of 2.2μm,as presented in Fig.14d.XRD was conducted to reconfirm the chemical structure of the regenerated product,as shown in Fig.15.It can be seen that the characteristic peaks of the regenerated product completely match the standard graph ofγ-Al₂O₃ (PDF No.47-1308) fairly well.The nonexistence of impurity peak also proves the generation of pure-phase y-Al₂O₃ product.The crystallite size of the regeneratedγ-Al₂O₃ is approximately 10.4 nm,as calculated by Debye-Scherrer equation,which is consistent with the reports in Ref. [ 47] .

The regenerated products were characterized by XPS to determine the chemical states.The results are illustrated in Fig.16.The characteristic peaks at a binding energy of74.1 and 530.9 eV for Al 2p and O 1s,respectively,are consistent with the characteristic peaks of A1 and O in yAl₂O₃ according to Ref. [ 48] .This result proves once again that y-Al₂O₃ is successfully regenerated.

The specific surface area ofγ-Al₂O₃ is one of the most important parameters to evaluate whether the material can be used as a catalyst carrier.Figure 17a shows N₂adsorption isotherm curve of regeneratedγ-Al₂O₃,the specific surface area of which is 278.3 m²·g^-1.Compared with y-Al₂O₃ prepared by other methods (Table 5 [ 49, 50, 51, 52, 53] ),the regenerated y-Al₂O₃ has a larger specific surface area.The N₂ adsorption-desorption isotherm and pore size distribution of regenerated y-Al₂O₃ are shown in Fig.17b.The pore size ofγ-Al₂O₃ shows a normal distribution,indicating a uniform pore size distribution.The average pore diameter ofγ-Al₂O₃ is 3.10 nm,which implies excellent surface properties.The regenerated products are dissolved in aqua regia and then analyzed by ICP-OES.The results indicate that the regenerated product contains impurities of 0.03 wt%Ni,0.07 wt%Mo and 0.02wt%S,which are within the standard range [ 54] .In conclusion,the analyses show that the performance of regeneratedγ-Al₂O₃ can meet the requirements of industrial HDS catalysts.

Fig.14 a,b SEM images,c EDS spectrum and d particle size distribution of regeneratedγ-Al₂O₃

Fig.15 XRD pattern of regeneratedγ-Al₂O₃

4 Conclusion

This study focused on the recovery and regeneration of alumina from spent HDS catalyst CoMo/Al₂O₃.Using an alkaline leaching process,98.13%alumina was successfully leached from the treated spent catalysts under the conditions as 5 mol·L^-1 sodium hydroxide,L/S ratio of20 ml·g^-1,160℃and reaction time of 4 h.Cobalt and nickel were enriched both of which are conducive to subsequent metal recovery,and no difficult leaching compound was formed in the leaching residues.The reaction order of aluminum leaching is 0.99.This reaction can fit well with the interfacial chemical reaction model,and the apparent activation energy is calculated as 45.50 kJ·mol^-1.Meanwhile,γ-Al₂O₃ was regenerated by a hydrothermal precipitation process using lixivium as the aluminum source with a mean size of 2.2μm,an average pore size of3.10 nm and a specific surface area of 278.3 m²·g^-1.Theγ-Al₂O₃ is suitable to be used as catalyst carrier.The recovery and regeneration of alumina from spent catalysts not only contribute to the overall recycling of such hazardous spent catalysts but also provide a new concept for the preparation ofγ-Al₂O₃ with a high specific surface area using spent catalysts as the aluminum source.

Fig.16 XPS spectra of regeneratedγ-Al₂O₃:a Al 2p and b O 1s

Fig.17 a N₂ adsorption isotherms and b N₂ adsorption/desorption isotherms and BJH pore size distribution plot (inset) of regeneratedγ-Al₂O₃

  下载原图

Table 5 Comparison of specific surface area ofγ-Al₂O₃ prepared by different methods

参考文献

[1] Ramirez S, Schacht P, Quintana-Solorzano R, Aguilar J.Leaching of heavy metals under ambient resembling conditions from hydrotreating spent catalysts. Fuel. 2013;110(10):286.

[2] Song CS. An overview of new approaches to deep desulfurization for ultra-clean gasoline, diesel fuel and jet fuel. Catal Today. 2003;86(1-4):211.

[3] Villarreal A, Ramirez J, Caero LC, Villalon PC,Gutiérrez-Alejandre A. Importance of the sulfidation step in the preparation of highly active NiMo/Si02/Al_2O_3 hydrodesulfurization catalysts. Catal Today. 2015;250:60.

[4] Dai XP, Du KL, Li ZZ, Liu MZ, Ma YD, Sun H, Zhang X, Yang Y. Co-doped MoS_2 nanosheets with the dominant CoMoS phase coated on carbon as an excellent electrocatalyst for hydrogen evolution. ACS Appl Mater Interfaces. 2015;7(49):27242.

[5] Layan Savithra GH, Bowker RH, Carrillo BA, Bussell ME,Brock SL. Mesoporous matrix encapsulation for the synthesis of monodisperse Pd5P2 nanoparticle hydrodesulfurization catalysts.ACS Appl Mater Interfaces. 2013;5(12):5403.

[6] van Haandel L, Bremmer M, Kooyman PJ, van Veen JAR,Weber T, Hensen EJM. Structure-activity correlations in hydrodesulfurization reactions over Ni-promoted MoxW_((1-x))S2/Al_2O_3 catalysts. ACS Catal. 2015;5(12):7276.

[7] Furimsky E, Massoth FE. Deactivation of hydroprocessing catalysts. Catal Today. 1999;52(4):381.

[8] Ruiz V, Meux E, Schneider M, Georgeaud V. Hydrometallurgical treatment for valuable metals recovery from spent CoMo/Al_2O_3 catalyst. 2. Oxidative leaching of an unroasted catalyst using H2O_2. Ind Eng Chem Res. 2011;50(9):5307.

[9] Akcil A, Veglio F, Ferella F, Okudan MD, Tuncuk A. A review of metal recovery from spent petroleum catalysts and ash. Waste Manag. 2015;45:420.

[10] Asghari I, Mousavi S, Amiri F, Tavassoli S. Bioleaching of spent refinery catalysts:a review. J Ind Eng Chem. 2013;19(4):1069.

[11] Barik S, Park K-H, Parhi P, Park J. Direct leaching of molybdenum and cobalt from spent hydrodesulphurization catalyst with sulphuric acid. Hydrometallurgy. 2012;111(1):46.

[12] Zeng L, Cheng CY. A literature review of the recovery of molybdenum and vanadium from spent hydrodesulphurisation catalysts:Part I:metallurgical processes. Hydrometallurgy.2009;98(1):1.

[13] Huang XW, Long ZQ, Wang LS, Feng ZY. Technology development for rare earth cleaner hydrometallurgy in China.Rare Met. 2015;34(4):215.

[14] Tian L, Liu Y, Zhang TA, Lv GZ, Zhou S, Zhang GQ. Kinetics of indium dissolution from marmatite with high indium content in pressure acid leaching. Rare Met. 2017;36(1):69.

[15] Chauhan G, Pant KK, Nigam KD. Metal recovery from hydroprocessing spent catalyst:a green chemical engineering approach. Ind Eng Chem Res. 2013;52(47):16724.

[16] Zeng L, Cheng CY. A literature review of the recovery of molybdenum and vanadium from spent hydrodesulphurisation catalysts:Part II:separation and purification. Hydrometallurgy.2009;98(1):10.

[17] Cortes-Torres R, Nolasco-Terron EY, Olea-Mejia O,Varela-Guerrero V, Barrera-Diaz CE, Cuevas-Yanez E. Solvent mediated impurity removal process for a spent hydroprocessing catalyst and its use in alcohol oxidations. Catal Today. 2018;305:126.

[18] Khalid M, Athraa B. Experimental study on factors affecting the recovery of nickel from spent catalyst. J Powder Metall Min.2017;6(1):1.

[19] Zhang Y, Lin H, Dong YB, Xu XF, Wang X, Gao YJ. Comprehensive recovery of iron, niobium rare earth and fluorite in Bayan Obo tailings. Chin J Rare Met. 2017;41(7):799.

[20] Sahu K, Agrawal A, Mishra D. Hazardous waste to materials:Recovery of molybdenum and vanadium from acidic leach liquor of spent hydroprocessing catalyst using alamine 308.J Environ Manage. 2013;125(1):68.

[21] Pinto IS, Soares HM. Recovery of molybdates from an alkaline leachate of spent hydrodesulphurisation catalyst-proposal of a nearly-closed process. J Clean Prod. 2013;52:481.

[22] Erust C, Akcil A, Bedelova Z, Anarbekov K, Baikonurova A,Tuncuk A. Recovery of vanadium from spent catalysts of sulfuric acid plant by using inorganic and organic acids:laboratory and semi-pilot tests. Waste Manag. 2016;49:455.

[23] Zhan C, Zhang YM, Bao SX, Huang J, Yang X. Separation and enrichment of vanadium from stone coal acidic leach solution using tertiary amine N235. Chin J Rare Met. 2017;41(4):422.

[24] Li Z, Chen M, Zhang QW, Liu XZ, Saito F. Mechanochemical processing of molybdenum and vanadium sulfides for metal recovery from spent catalysts wastes. Waste Manag. 2017;60:734.

[25] Nguyen TH, Lee MS. Development of a hydrometallurgical process for the recovery of calcium molybdate and cobalt oxalate powders from spent hydrodesulphurization(HDS)catalyst.J Clean Prod. 2015;90(10):388.

[26] Cibati A, Cheng KY, Morris C, Ginige MP, Sahinkaya E,Pagnanelli F, Kaksonen AH. Selective precipitation of metals from synthetic spent refinery catalyst leach liquor with biogenic H2S produced in a lactate-fed anaerobic baffled reactor.Hydrometallurgy. 2013; 139(3):154.

[27] Feng ZY, Huang XW, Wang M, Zhang GC. Progress and trend of green chemistry in extraction and separation of typical rare earth resources. Chin J Rare Met. 2017;41(5):604.

[28] Wang B, Meng Y, Duan CS. New process for complete recycling of metals from spent Co, Mo, and-Al_2O_3 catalyst. Mod Chem Ind. 2005;25(S1):204.

[29] Busnardo RG, Busnardo NG, Salvato GN, Afonso JC. Processing of spent NiMo and CoMo/A1_2O_3 catalysts via fusion with KHS04. J Hazard Mater. 2007;139(2):391.

[30] Huang SB, Zhao ZW, Chen XY, Li F. Alkali extraction of valuable metals from spent Mo-Ni/Al_2O_3 catalyst. Int J Refract Metal Hard Mater. 2014;46(9):109.

[31] Tabesh S, Davar F, Loghman-Estarki MR. Preparation of y-Al_2O_3 nanoparticles using modified sol-gel method and its use for the adsorption of lead and cadmium ions. J Alloy Compd.2018;730:441.

[32] Svetlichnyi VA, Stadnichenko AI, Lapin IN. Preparation ofγ-Al(OH)3 andγ-Al_2O_3 nanoparticles by the method of pulsed laser ablation of metal aluminum in water. Russ Phys J. 2017;60(2):377.

[33] Yang Y, Xu SM, Li Z, Wang JL, Zhao ZW, Xu ZH. Oil removal of spent hydrotreating catalyst CoMo/A12O_3 via a facile method with enhanced metal recovery. J Hazard Mater. 2016;318:723.

[34] Jin HX, Wu FZ, Mao XH, Wang ML, Xie HY. Leaching isomorphism rare earths from phosphorite ore by sulfuric acid and phosphoric acid. Rare Met. 2017;36(10):840.

[35] Wang JY, Xu Y, Wang LS, Zhao LS, Wang Q, Cui DL, Long ZQ, Huang XW. Recovery of rare earths and aluminum from FCC catalysts manufacturing slag by stepwise leaching and selective precipitation. J Environ Chem Eng. 2017;5(4):3711.

[36] Whittington BI, Fletcher BL, Talbot C. The effect of reaction conditions on the composition of desilication product(DSP)formed under simulated Bayer conditions. Hydrometallurgy.1998;49(1-2):1.

[37] Liu CL, Xia JP, Zhang YB. Optimization and kinetics on extraction of alumina from coal gangue by acid leaching. Chin J Process Eng. 2015;15(04):579.

[38] Chen Y, Feng QM, Shao YH, Zhang GF, Ou LM, Lu YP.Research on the recycling of valuable metals in spent Al_2O_3--based catalyst. Miner Eng. 2006;19(1):94.

[39] Wang B, Hao YL, Chu WQ, Rong S, Sun HL. Kinetics of leaching of 20CaO13Al_2O_33MgO_3SiO_2. Miner Process Extr Metall IMM Trans C. 2016;126(4):1.

[40] Nayl AEAA, Aly HF. Extraction equilibria and kinetics of Ti(IV)from leached chloride liquors of ilmenite. Rare Met.2017;36(8):676.

[41] Amini M, Mirzaee M. Effect of solution chemistry on preparation of boehmite by hydrothermal assisted sol-gel processing of aluminum alkoxides. J Sol-Gel Sci Technol. 2005;36(1):19.

[42] Okada K, Nagashima T, Kameshima Y, Yasumori A, Tsukada T. Relationship between formation conditions, properties, and crystallite size of boehmite. J Colloid Interface Sci. 2002;253(2):308.

[43] Santos PdS, Coelho ACV, Santos HdS, Kiyohara PK.Hydrothermal synthesis of well-crystallised boehmite crystals of various shapes. Mater Res. 2009;12(4):437.

[44] Wang JQ, Liu JL, Liu XY, Qiao MH, Pei Y, Fan KN.Hydrothermal transformation of bayerite to boehmite. Sci Adv Mater. 2009;1(1):77.

[45] Mishra D, Anand S, Panda R, Das R. Hydrothermal preparation and characterization of boehmites. Mater Lett. 2000;42(1):38.

[46] Mishra D, Anand S, Panda R, Das R. Effect of anions during hydrothermal preparation of boehmites. Mater Lett. 2002;53(3):133.

[47] Wang WW, Zhou JB, Zhang Z, Yu JG, Cai WQ. Different surfactants-assisted hydrothermal synthesis of hierarchical y-Al_2O_3 and its adsorption performances for parachlorophenol.Chem Eng J. 2013;233(2):168.

[48] Wang Q, Cheng XH, Zheng L, Shen LY, Li JJ, Zhang DL, Qian R, Yu YH. Interface engineering of an AlNO/AlGaN/GaN MIS diode induced by PEALD alternate insertion of AIN in Al_2O_3.RSC Adv. 2017;7(19):11745.

[49] Cai WQ, Yu JG, Jaroniec M. Template-free synthesis of hierarchical spindle-like y-Al_2O_3 materials and their adsorptionaffinity towards organic and inorganic pollutants in water.J Mater Chem. 2010;20(22):4587.

[50] Inoue M, Kominami H, Inui T. Synthesis of large pore-size and large pore-volume aluminas by glycothermal treatment of aluminium alkoxide and subsequent calcination. J Mater Sci. 1994;29(9):2459.

[51] Cao JM, Hou HT, Ma XJ, Ji GB, Zheng MB, Lu HX.Solvothermal synthesis of nanoporous gamma aluminum oxide.Chin J Inorg Chem. 2005;9(21):1379.

[52] Meng XH, Duan LH, Xie XH, Wang Q, Wang HY. Synthesis of macro-mesostructured y-Al_2O_3 with large pore volume and high surface area by a facile secondary reforming method. China Pet Process Petrochem Technol. 2014;16(2):20.

[53] Potdar HS, Jun K, Bae JW, Kim S, Lee Y. Synthesis of nano-sized porous y-alumina powder via a precipitation/digestion route. Appl Catal A-Gen. 2007;321(2):109.

[54] Li GM, Guo FX, Fan GQ. Standard of activated inorganic chemical products in China. Inorg Chem Ind. 2009;41(1):60.