宋 涛，廖景明，2，肖 军，沈来宏

(1.东南大学 能源热转换及其过程测控教育部重点实验室，江苏 南京 210096;2.福建省电力勘探设计院，福建 福州 350003)

1 Introduction

Apart from those returned to the field as fertilizer or used as fuels in rural households，forage，and industrial raw materials，etc.，biomass materials available for energy production annually amounted to 365 million tons of coal equivalent (Mtce)in 2005 in China［1］.This amount is projected to be increased further.Being a proven effective and highly efficient technology for using biomass on a large scale，direct combustion for electricity generation is playing a crucial role in biomass energy conversion in China［1］.Besides，biomass thermal chemical process also has been widely developed and applied in some areas focusing on the pyrolysis，gasification or liquification.

Among biomass chemical thermal processes，pyrolysis is one form of energy recovery process，which has a potential to obtain char，oil and gas products.The char as a solid product has a different property with the original biomass feed stock.It usually becomes more porous due to the fast disappearance of volatile matter or moisture content during pyrolysis.These changes in the properties usually result in high reactivity，and hence，an alternative usage of char as a sorbent becomes attractive.After activated by some treatments，the char can be produced as activated carbon (AC)，which is being widely used for air cleaning or wastewater treatment，and other areas［2，3］.

CO2capture by AC adsorption seems to be an interesting and attractive method.It is due to the low energy needed and low primary investment for performing this process［4，5］.Maroto-Valer et al.［6］investigated the CO2adsorption performance by AC produced from anthracite.Results showed that the AC after steam activation had a BET surface area of 540 m2/g，and the maximum CO2adsorption capacity was 65.7 mg-CO2/g.Also，they used the high carbon fly ashes as the original materials to obtain a maximum CO2adsorption capacity of 68.6 mg-CO2/g［7］.An et al.［8］investigated the CO2adsorption performance by activated carbon fibre-phenolic resin composites.It was found that a maximum CO2adsorption capacity of 130 mg-CO2/g.Also，the desorption performance of this AC was investigated［9］.

Some kinds of biomass residues have been used as raw materials for AC production.A detailed summary and illustration as a review was given by Ioannidou［10］.The characteristics of ACs can be affected by some control factors，such as BET surface area，micropore surface area，and volume properties.The pyrolysis parameters，physical activation parameters，chemical activation parameters could also have a significant influence on the characteristics of ACs and their CO2adsorption performance.Corn is an important crop planted in a large area in the northern China.At present，the traditional treatment for the corn stalks is to return them to the field as fertilizers.As a potential material with abundant resources and low cost for the AC production，it is necessary to give an investigation to assess its characteristics for AC and for CO2adsorption.Herein，the corn stalks were used as the raw material.The effect of physical activation parameters，chemical activation agents and post thermal treatments were evaluated.Further，the relationship between CO2adsorption and the structural characterization of ACs was also illustrated.

2 Experimental

2.1 Pyrolysis/physical activation

Corn stalk was used as feedstock for the production of AC.The proximate and ultimate analysis are presented in Table 1.As for the corn stalk used，it had a low ash content of only 3.15%，and a high fixed carbon content of 16.75% and volatile content of 70.74%.It is suitable as a raw material for AC production.Basically，there are two main steps for the production of AC.It consists of the carbonization in the absence of oxygen，and the activation of the produced char.In this work，the two step processes were combined，and it was carried out in a horizon tubular reactor (inside diameter of 80 mm，length of 340 mm)［11］.Briefly，the experimental apparatus mainly consisted of a tubular reactor，a temperature control device，a heat exchanger，an oilpot.The pyrolysis temperature was designated as 800 ℃.In each experiment，after the reactor reached the designated temperature，the gas mixture of N2and CO2(as activation gas)with a total flow rate of 500 mL/min was introduced into the reactor to maintain an atmosphere in the absence of oxygen.Afterward，each sample with a certain weight was placed in a porcelain boat，which was rapidly pushed into the reactor.The sample was quickly heated in the furnace and reached the furnace temperature.When the designated time for the pyrolysis was reached，the gas was switched to pure nitrogen until the products were cooled.The AC production conditions are summarized in Table 2.The combination of pyrolysis times and activation gas concentrations is different for each run.

Table 1 Proximate and ultimate analysis of the corn stalk.

Table 2 Pyrolysis/physical activation conditions during AC preparation.

2.2 Chemical activation and post thermal treatment

Some samples of AC which were produced following the above procedure were further activated with KOH，HNO3，or CH3COOH solution.These chemical solutions were used to remove the ash contents of as-produced AC samples.

KOH activation.1 mol/L and 4 mol/L KOH solutions were used for the chemical activation.It was found that high concentration of KOH can cause destruction of carbon structure of AC，which was not beneficial for the following adsorption.Thus，the results obtained using 1 mol/L KOH activation were shown in this work.Two methods were carried out for this chemical activation.The first one was to put the AC sample in the KOH solution for 48 h at 25 ℃.Another one was to put the AC samples in a balloon flask filling of 1 mol/L KOH solution.Then，it was heated the in the water bath by using a magnetic mixer at a temperature of 100 ℃for 1 h.After that，the KOH washed samples were further washed with distilled water several times until neutral pH value.Both of the produced samples were further dried at 110 ℃for 3 h.

HNO3activation.4 mol/L HNO3solution was used for AC activation.The same two treatments were carried out following the above procedure.

CH3COOH activation.4 mol/L CH3COOH solution was used.The CH3COOH washed samples were put in the solution for 48 h.Then they were further washed and dried.

Some samples before and after acid (HNO3or CH3COOH)or alkali (KOH)activation were further processed by post thermal treatment.The horizon tubular reactor was employed by performing the post thermal treatment.It was carried out at two different temperatures (300 or 600 ℃)for 1 or 2 h under the pure nitrogen flow.14 samples (4 samples without acid or alkali activation，10 samples with chemical activation)was produced.The CO2adsorption using theses ACs was investigated in detail in this work.Meanwhile，a commercial AC from the market was used for comparison.Table 3 gives a summary of the producing process for the 10 samples used.

Table 3 Chemical activation conditions and further treatments for the ACs.

2.3 Characterization of ACs

The pore structure of the samples was measured by nitrogen adsorption at 77 K with a Micromeritics instrument ASAP 2020.The surface area was calculated from the Brunauer-Emmett-Teller (BET)equations.The micropore area and micropore volume were calculated from the t-plot method.The mesopore volume and average pore diameter were calculated by Barrett-Joyner-Halenda (BJH) method.Table 4 shows the structure parameters of all the ACs samples.

Thermogravimetric analysis (TGA)was carried out with Cahn TG-131 to determine the CO2adsorption property of the ACs.The sample (10 mg)was first dried at 25 ℃ under a pure N2flow rate of 20 mL/min to remove the influence of some moisture on the adsorption.Then，10% CO2equilibrated by N2or 100% CO2gas were introduced to the furnace.The CO2adsorption efficiency (ηt，%)was defined as:

Table 4 Structural characterization of all the AC samples.

Where W0means the original weight of the AC samples，and Wtpresents the weight of the AC samples at the adsorption time of t.For some samples，100%CO2gas was used as the adsorption gas.In this condition，120 min experiments were carried out.η120was defined as the adsorption efficiency at an equilibrium state.

3 Results and discussion

3.1 N2adsorption

Fig.1 presents the N2adsorption isotherms of ACs used in this work.The shape of the adsorption isotherm can provide preliminary qualitative information on the adsorption mechanism and on the porous structure of the ACs［12］.All isotherms at very low relative pressures (p/p0＜0.2)presents a sharp increase typical of adsorption taking place by micropore filling.When the relative pressure is larger than 0.2，the adsorption volume kept nearly constant，indicating that the end of adsorption by micropores.Then，the adsorption on the surface area became dominant.When the pressure is close to the saturated vapor pressure，the adsorption volume increases again due to the presence of mesopore or macropores.According to the classification of BDDT［13］，the adsorption isotherms for the samples investigated could be classified as I-B resembling microporous structured materials.

Fig.1 N2adsorption isotherms of activated carbons prepared from corn stalks after(a)physical activation with CO2and (b)further chemical activation.

3.2 Effect of physical activation time

The CO2adsorption efficiency for CSC1#，CSC2#，CSC3 # with different CO2activation times is shown in Fig.2.The reacting gas concentration of CO2introduced to the furnace was 20% with N2as balancing gas.As shown in Fig.2，it could be seen that the maximum adsorption efficiency of 1.42%，1.86% and 1.51%，corresponding to different activation times of 20，30，and 60 min respectively.That is，the efficiency increases from the initial activation time from 20 to 30 min and then decreases.The optimal activation for the corn stalk is at an activation time of 30 min with a maximum adsorption efficiency of 1.86%.

Fig.2 Effect of activation time onadsorption efficiency.Reacting gas:10% CO2.

As shown in Table 4，the micropore area and micropore volume of the physical activated ACs (CSC1#，CSC2# and CSC3#)decrease from 297.7 to 277.4 m2/g，0.138 to 0.129 cm3/g respectively with activation time.Meanwhile，the mesopore volume increased nearly twice，from 0.036 to 0.086 cm3/g.The total pore volume increased from 0.175 to 0.215 cm3/g.These indicate that during this activation from 20 to 30 min，some micropores are connected，thus leading to an increase of mesopore volume.The decrease of micropore area may be due to the block effect of some tar or volatile matter in the channel of micropores.However，after the activation time is increased to 60 min，for CSC3#，the BET surface area and micropore area increase to 425.4 m2/g and 315.1 m2/g respectively.Therefore，it could be deduced that the volatile matter continuously disappears with activation time from 30 to 60 min.Also，the activation gas of CO2is beginning to react with the produced carbon.This breaks the structure of activated carbon，which isnot beneficial for CO2adsorption.

It is much clear that the micropore adsorption and the effect of CO2diffusion in the mesopore are the main factors affecting the CO2adsorption.The objective to carry out these experiments is to detect which is the key factor for CO2adsorption.However，as illustrated above，there is no direct trend for the relationship of the CO2adsorption and BET surface area，micropore area and micropore volume.Thus，one primary assumption could be that the effect of CO2diffusion in the mesopore is more important than the adsorption in micropores，which makes the sample CSC2# a maximum adsorption efficiency and adsorption rate.

3.3 Effect of CO2concentration

Fig.3 gives a comparison for the adsorption efficiency of the two samples (CSC2# and CSC4#)with two CO2activation concentrations of 20% and 50%.Both the adsorption efficiency and the adsorption rate for the CSC2# are much higher than that for CSC4#.The adsorption efficiency decreased from 1.86% with a CO2activation concentration of 20% to 1.33% of that of 50%，which indicates that for the AC production in the initial physical activation，high CO2activation concentration is not beneficial for adsorption efficiency.

Fig.3 Effect of CO2concentrations on adsorption efficiency(CSC 2#:20% CO2;CSC 4#:50%CO2).

The increase of the concentration of CO2can improve the activation rate.It could make a remarkable development of surface area and volume.As shown in Table 4，the sample of CSC4# has a higher BET surface area and micropore area than that for the sample of CSC2#.When the concentration increases from 20 to 50%，the BET surface area and micropore area increase by 24.3 and 20.3% respectively.Also，the micropore volume and total pore volume increases by 19.9 and 4.0% respectively.However，the mesopore volume decreases from 0.086 to 0.069 cm3/g.As shown in Fig.3，the increase of BET surface area and micropore area has no influence on the CO2adsorption.The enhancing CO2adsorption performance is more favored by the increasing mesopore volume.

3.4 Effect of chemical agents

The samples of CSC-K，CSC-H and CSC-A were used for discussing the effect of chemical agents on the CO2adsorption.These three samples were further activated by HNO3，KOH or CH3COOH respectively.The reacting gas concentration of CO2was 100%.As shown in Fig.4，in comparison with the AC produced without chemical activation (CSC2#)，the ones after chemical activation shows an improved CO2adsorption performance，not only for the adsorption efficiency but for the adsorption rate.The adsorption performance for ACs after HNO3and KOH activation shows similar results.Also，the performance increase is the highest for the AC further activated by CH3COOH.

Fig.4 Effect of chemical agents on adsorption efficiency.

The precipitous isotherms for the ACs after chemical activation and an improved nitrogen adsorption are observed in Fig.1b.Also，the BET surface area and micropore area for the ACs after chemical activation significantly increase at different degrees，as shown in Table 4.Further，the BET surface area or micropore area is nearly the same as those of the commercial AC.The surface area and pore volume are significantly enhanced for the ACs after acid or alkali activation.

Generally，under this condition，the effect of CO2diffusion in the mesopore would becomes weakened，whereas the micropore adsorption progressively plays the dominant role for CO2adsorption.Two results could support this conclusion.The first one is that the adsorption performance for AC after HNO3activation is nearly similar to the AC after KOH activation，though it has a larger mesopore volume.Another one is that due to a maximum micropore area among the three samples，the best CO2adsorption performance is obtained for the sample of CSC-A after CH3COOH activation (Fig.4).Therefore，it could be confirmed that under this condition，the micropore adsorption becomes the main factor affecting the adsorption.

The influencing factors in the CO2adsorption efficiency have been progressively changed from the effect of CO2diffusion in the mesopores to the micropore adsorption after the further chemical activation.However，the critical value for switching the roles in CO2adsorption from the mesopore-diffusion dominant to the micropore-adsorption dominant should be given to make this work reliable and meaningful.

3.5 Effect of post thermal treatment

The CO2adsorption with some samples of CSCK，CSC-KC，CSC-KB，and CSC-KBC are investigated to discuss the effect of post thermal treatment.The samples of CSC-H，CSC-HC，CSC-HB，CSCHBC and CSC-HBC+are also tested with the same objective.

Fig.5 shows the effect of post treatment on the CO2adsorption efficiency for all samples.

Fig.5 Effect of post treatment condition on adsorption efficiency after (a)KOH activation and (b)HNO3activation.

As shown in Fig.5a，it could be observed that the samples after heating in the water bath plus the post thermal treatment shows significantly enhanced adsorption performance.For the sample of CSC-KBC(heating in water bath 1 h plus post thermal treatment 1 h at 300 ℃)，the CO2adsorption efficiency could reach to 5.97%.The improved performance for the ACs after heating in the water bath plus post thermal treatment is also seen in Fig.5b.Also，a longer post thermal treatment time is beneficial to obtain a better adsorption performance，as seen for the samples of CSC-HBC and CSC-HBC +.However，there is one exception found for the samples of CSC-H and CSCHB.The adsorption efficiency with the sample of CSC-HB which was heated in the water bath for 1h is lower than that for CSC-H without heating.Therefore，from this point，the relation of CO2adsorption to the effect of heating in the water bath is not clear.However，the post thermal treatment seems to be a good method to improve CO2adsorption.

Among all the samples investigated，the sample of CSC-HBC+(HNO3activation+water bath 1 h+post thermal treatment 2 h at 600 ℃)exhibits a maximum CO2adsorption efficiency of 7.33%，which is higher than that for the commercial AC from the market with a CO2adsorption efficiency of 6.55%，as shown in Fig.6.

3.6 Effect of BET surface area and micropore area

As illustrated above，the switching roles for the effect of CO2diffusion in the mesopores to BET surface area or micropore area are observed.In this section，the relationship between CO2adsorption and its factors is revealed.The adsorption capacity (q，mg/g)was defined to evaluate the adsorbed amount for CO2per g AC.

Fig.6 A comparison of the maximum CO2adsorption efficiency among the ACs produced in this work with a commercial AC.

Fig.7 shows the relationship of adsorption capacity (q)with the BET surface area and micropore area.A similar trend could be found from the two figures.The ACs without chemical activation produced by varying physical activation concentrations or activation times have BET surface areas smaller than 500 m2/g.There is no well-defined relationship between the adsorption capacity and BET surface area as well as micropore area.However，the BET surface areas of the samples activated by chemical activation or post thermal treatment are larger than 500 m2/g.The CO2adsorption capacity increase with the BET surface area and micropore area.The effect of CO2diffusion in the mesopores becomes weakened and the CO2adsorption is dominantly dependent on the micropores under this condition.

As shown in Table 4，all the four samples of CSC-K，CSC-KC，CSC-KB，and CSC-KBC have BET surface areas larger than 500 m2/g.The best adsorption performance is found for the sample of CSCKBC among the four samples，as observed in Fig.5a.For this sample，it has a largest micropore area of 430.2 m2/g，leading to a largest adsorption efficiency.In a word，the adsorption in micropores is dominant for the samples with BET surface areas larger than 500 m2/g.

The ACs after HNO3activation and post thermal treatment show similarly enhanced performance for CO2adsorption，especially for the samples obtained with CSC-HBC and CSC-HBC+(Fig.5b).Both the two samples have BET surface areas larger than 590 m2/g and micropore areas larger than 410 m2/g.Especially for CSC-HBC +，the BET surface area and micropore area are 639.8 and 457.0 m2/g respectively.This results in higher CO2adsorption efficiencies of 7.33% for CSC-HBC+and 6.49% for CSCHBC.Also，the BET surface area is increased by 8.1% and the micropore area by over 9% when the thermal treatment time increases from 1 to 2 h.As a result，a higher CO2adsorption efficiency is obtained.

Fig.7 Effect of (a)BET surface area and (b)t-plot micropore area on CO2adsorption capacity.

3.7 Adsorption kinetics

3.7.1 Simulation method by Bangham model

The apparent gas adsorption model has been developed to give a quantitative description of the adsorption and detect the adsorption mechanism，such as Bangham model［14］，Elovich model［15］，Lagergren model［16］，and Ho model［17］.Among the models developed，the Elovich model and Ho model are usually used for describing the adsorption by chemical adsorption.The Lagergren model is better to describe the adsorption controlled by diffusion.The Bangham model has a high accuracy in the prediction of the adsorption controlled by physical adsorption.In this work，the apparent CO2adsorption kinetic was investigated by means of Bangham model，since the CO2adsorption by ACs is believed to be a physical adsorption.It is supposed that the adsorption pressure kept constant.The Bangham model could be shown as:

Where qtpresents the adsorption amount at a certain time of t，and q0is the adsorption amount at an equilibrium condition，mg/g.k is the Bangham rate coefficient，min-n.

Based on the MATLAB software，the Bangham model for CO2adsorption was built.The genetic algorithm (GA)was used to optimize the simulation results.In order to evaluate the deviation between the experimental data and calculated data，the mean residual (δ)is defined as:

Where R2is the residual sum of squares，and N is the group number of calculated data.

Table 5 lists the simulation results using the Bangham model for the CO2adsorption for different ACs used in this work.qeand q0presents the equilibrium adsorption amount of the experimental data and simulation results respectively.It could be seen that the simulation results for all the ACs fitted well with the experimental data (Fig.8).It is confirmed that the CO2adsorption is a physical adsorption.Compared with the three figures in Fig.8，it could be observed that when a higher CO2reacting gas concentration of 100% is used，a higher CO2adsorption rate is found.Since a driving force is enhanced with a high concentration reacting gas，thus resulting in a faster diffusion rate in mesopores，which could supply more CO2gas to be adsorbed in the micropores.3.7.2 Primary Bangham model description

Under the same adsorption mechanism for the ACs used in this work，the value of n should be same.As shown in Table 5，for the ACs adsorption at a 100% CO2as a reacting gas，the value of n is in the range of 0.798 to 1.065.An average value of n as 0.940 is selected and kept constant.Then，another two parameters of q0and k are re-optimized using GA.Table 6 shows the simulation results for the different ACs at a constant n value of 0.940.Fig.9 shows the curves of the simulation results and experimental data.Results shows that the predicted results fit well with the experimental ones.

Fig.8 CO2adsorption capacity as a function of time for different AC samples:(a)reacting gas 20% CO2and (b，c)reacting gas 100% CO2.

From Table 6，it could be seen that there is no significant change for the Bangham rate coefficient of k.Following the same method，an average value of k of 0.0275 is selected and kept constant.In this step，the two parameters of n and k are identified.The final step is to identify the last parameter of q0using the same method.Table 7 shows the simulation results for the different ACs at a constant value of n=0.940 and k=0.0275.The simulation results with all the ACs fit well with the experimental data (Fig.10).

Table 5 Bangham kinetic model calculation results.

Table 6 Bangham kinetic model calculation results，n=0.940.

Fig.9 n=0.940.CO2adsorption capacity as a function of time for different AC samples under 100% CO2.

Then，the Bangham model for the adsorption of ACs used in this work could be described as:

n and k are equal to 0.940 and 0.0275 respectively.This model is suitable to predict CO2adsorption kinetics when the CO2reacting gas concentration is 100%.

Table 7 Bangham kinetic model calculation results，n=0.940 and k=0.0275.

Fig.10 n=0.940 and k=0.0275.CO2adsorption capacity as a function of time for different AC samples under 100% CO2.

3.7.3 Developed Bangham model description

Fig.11 shows the relationship between the CO2adsorption capacity and BET surface area as well as micropore area for the ACs after chemical activation.The arithmetic expression of the CO2adsorption capacity with the two factors could be obtained as:

Fig.11 CO2adsorption capacity as functions of (a)BET surface area and (b)t-plot micropore area.

3.7.4 Model prediction

Fig.12 Summarizes the prediction results using the BET Bangham model as well as t-plot Bangham model as compared with the experimental data.

Fig.12 The maximum CO2adsorption (q)as a function of (a)BET surface area and (b)t-plot micropore area.

The adsorption capacity for all the samples with a BET surface area larger than 500 m2/g is shown in Fig.12.It could be observed that the t-plot Bangham model shows a good prediction than that for the BET Bangham model.The equilibrium adsorption amount for the prediction of t-plot Bangham model is approximately equal to the experimental data.

The CO2adsorption by ACs produced from biomass residues is a complex.Further consideration should be performed to comprehensively evaluate the effect of the BET surface area and micropore area to improve the prediction accuracy of the Bangham model.

4 Conclusions

During the physical activation，the optimal condition for AC production from corn stalk was at 800℃with 20% CO2flow for 30 min.The chemical agents and post thermal treatment could expand the pore structure of ACs，thus resulting in an improvement in CO2adsorption performance.The suitable treatments of ACs from corn stalks is at 100 ℃with water bath heating for 60 min plus post thermal treatment.In this case，the BET surface area of the sample (HNO3activation+water bath 1 h+post thermal treatment 2 h at 600 ℃)is 639.8 m2/g.A maximum CO2adsorption efficiency of 7.33% is obtained，which is higher than that of 6.55% for the commercial AC.When the BET surface area of AC is smaller than 500m2/g，the CO2adsorption property is dominantly dependent on the mesopore volume of the ACs.However，the CO2adsorption property is significantly associated with micropore area when the BET surface area of AC is larger than 500 m2/g.The t-plot Bangham model shows a good prediction for the CO2adsorption with the ACs used.

［1］Wang G L，Shen L H，Sheng C D.Characterization of biomass ashes from power plants firing agricultural residues［J］.Energy ＆Fuels，2012，26(1):102-111.

［2］Yang J，Qiu K Q.Preparation of activated carbons from walnut shells via vacuum chemical activation and their application for methylene blue removal［J］.Chemical Engineering Journal，2010，165(1):209-217.

［3］Mohan D，Sarswat A，Singh V K，et al.Development of magnetic activated carbon from almond shells for trinitrophenol removal from water［J］.Chemical Engineering Journal，2011，172(2-3):1111-1125.

［4］Plaza M G，Pevida C，Arenillas A，et al.CO2capture by adsorption with nitrogen enriched carbons［J］.Fuel，2007，86(14):2204-2212.

［5］Plaza M G，Pevida C，Arias B，et al.Development of low-cost biomass-based adsorbents for post combustion CO2capture［J］.Fuel，2009，88(12):2442-2447.

［6］Maroto-Valer M M，Tang Z，Zhang Y.CO2capture by activated and impregnated anthracites［J］.Fuel Processing Technology，2005，86(14-15):1487-1502.

［7］Maroto-Valer M M，Lu Z，Zhang Y，et al.Sorbents for CO2capture from high carbon fly ashes［J］.Waste Management，2008，28(11):2320-2328.

［8］An H，Feng B，Su S.CO2capture capacities of activated carbon fibre-phenolic resin composites［J］.Carbon，2009，47(10):2396-2405.

［9］An H，Feng B.Desorption of CO2from activated carbon fibrephenolic resin composite by electrothermal effect［J］.International Journal of Greenhouse Gas Control，2010，4(1):57-63.

［10］Ioannidou O，Zabaniotou A.Agricultural residues as precursors for activated carbon production-A review［J］.Renewable ＆Sustainable Energy Reviews，2007，11(9):1966-2005.

［11］Qu T，Guo W，Shen L，et al.Experimental study of biomass pyrolysis based on three major components:hemicellulose，cellulose，and lignin［J］.Industrial ＆ Engineering Chemistry Research，2011，50(18):10424-10433.

［12］Stavropoulos G G，Zabaniotou A.Production and characterization of activated carbons from olive-seed waste residue［J］.Microporous and Mesoporous Materials，2005，82(1-2):79-85.

［13］Brunauer S，Deming L S，Deming W E，et al.On a theory of the van der Waals adsorption of gases［J］.Journal of the American Chemical Society，1940，62(7):1723-1732.

［14］Low M J D.Kinetics of chemisorption of gases on solids［J］.Chemical Reviews，1960，60(3):267-312.

［15］Zhang J S，Stanforth R.Slow adsorption reaction between arsenic species and Goethite (α-FeOOH):diffusion or heterogeneous surface reaction control［J］.Langmuir，2005，21(7):2895-2901.

［16］Ho Y S，Ng J C，McKay G.Kinetics of pollutant sorption by biosorbents:Review［J］.Seper Purif Method，2000，29(2):189-232.

［17］Ho Y S，Mckay G.Pseudo-second order model for sorption processes［J］.Process Biochemistry，1999，34(5):451-465.