柳旭升;马军;杨晶;史风梅
【摘 要】采用离子交换法制备了铁掺杂羟基磷灰石(Fe-HAp)介孔材料,并以非定域密度函数理论(NLDFT)计算了不同浓度铁离子掺杂羟基磷灰石的孔径分布曲线.NLDFT模型结果表明,羟基磷灰石以集合体间孔隙为主,当铁离子浓度大于0.01 mol·L-1时,Fe-HAp的集合体内孔隙逐渐增加.在铁离子浓度为0.05 mol· L-1时制备的Fe-HAp具有最大量的集合体内孔隙.当铁离子浓度大于0.05 mol·L-1时,Fe-HAp粒子尺寸急剧减小,导致集合体内孔隙相应地减少.透射电子显微镜很好地证实了所选择NLDFT模型的结果,相比传统BJH模型,NLDFT模型更为准确. 【期刊名称】《黑龙江大学自然科学学报》 【年(卷),期】2014(031)005 【总页数】7页(P639-645)
【关键词】羟基磷灰石;离子交换;非定域密度函数理论;孔径分布 【作 者】柳旭升;马军;杨晶;史风梅
【作者单位】哈尔滨工业大学市政环境工程学院,哈尔滨150090;哈尔滨工业大学水资源与环境国家重点实验室,哈尔滨150090;哈尔滨工业大学市政环境工程学院,哈尔滨150090;哈尔滨工业大学水资源与环境国家重点实验室,哈尔滨150090;哈尔滨工业大学建筑设计研究院,哈尔滨150090;黑龙江省农业科学院农村能源研究所,哈尔滨150086 【正文语种】中 文
【中图分类】O647
0 Introduction
Hydroxyapatite(HAp)is a promising material because of its good biocompatibility and low cost［1－3］．Therefore，HAp has been successfully used in biomedical，biotechnological and environmental fields［4］．Recently，the design and development of metal-doped HAp mesoporous materials as drug carriers have drawn a great attention because mesoporous structure can be used for drug storage and release［5－6］．Among metal dopants，iron has a prominent advantage in cost in comparison with other metals such as strontium or
europium．Furthermore，Fe(III)-doped HAp(Fe-HAp)is a blood compatible biomaterial［7］．However，most of Fe-HAp was prepared by a co-precipitation technique，which confines the iron content by the differences of ionic radii and charges between Ca2+and Fe3+ions［8］．On the contrary，the iron content of Fe-HAp is no longer limited by an ion-exchange method until an amorphous FePO4phase is formed［9－10］．To the best of our knowledge，previous investigation on Fe-HAp prepared by an ion-exchange method was focused on local geometry and distribution of Fe3+ions in the HAp lattice．The relationship between the mesopore size distribution and the concentration of substituted Fe3+ions solution was not investigated［11］．Meanwhile，because the pore sizes provided by a traditional Barrett-Joyner-Halenda(BJH)method are smaller than the actual sizes，it is interesting to investigate the pore size
distribution(PSD)of Fe-HAp by a modern non-local density functional theory(NLDFT)method which supplies accurate consistency between calculations and experiments basing on the molecular level models［12－13］．Therefore，study on mesopore size distribution of Fe-HAp based on the NLDFT method would supply guides for application of Fe-HAp mesoporous material．
In this work，Fe-HAp was prepared by an ion-exchange method with different concentrations of Fe(NO3)3solution．The mesopore size distribution of Fe-HAp was analyzed by the N2absorption technique and the NLDFT method．The relationship between the mesopore size distribution of Fe-HAp and the concentration of substituted Fe3+ions solution was investigated by a selected NLDFT model．The transmission electron microscopy(TEM)was used to compare the results between the selected NLDFT model and the traditional BJH model． 1 Experimental section 1.1 Materials
Powder HAp was obtained from Ningbo Kingo Advanced Materials R＆D Co．，Ltd．，China．Iron(III)nitrate nonahydrate was supplied by Xilong Chemical Factory，Guangdong，China．All chemicals were used as Received without purification and deionized water was used to prepare all solutions．
1.2 Experimental procedure
Fe-HAp was prepared as follows:1.00 g of HAp was dispersed in 100 mL of Fe(NO3)3solution with concentration in a range of 0.01～0.1 mol·L－
1．The mixture was vigorously stirred for 15 min at 25 ℃．Then the suspension particles were washed three times with 1 L deionized water and filtered with filter paper．The filtered particles were dried at 105 ℃ in air for 8 h to obtain yellow powder，Fe-HAp． 1.3 Characterization
The surface area and porosity were analyzed by an ASAP 2020 surface area and pore size analyzer(Micromeritics)，using N2adsorption-desorption isotherms at－196℃．The samples before analysis were degassed at 150℃for 6 h．The specific surface area was calculated by the Brunauer-Emmett-Teller(BET)method．The Quantachrome Quadra Win software was used to perform calculation．The TEM measurements were performed with a Tec-nai G2 F30 microscope(FEI Corporation)．The fresh prepared Fe-HAp suspension was dropped on a carbon-coated copper grid，and then the carbon-coated copper grid with the Fe-HAp particles loaded was dried at 105℃ for 30 min in an oven in air． 2 Results and discussion
2.1 Appearance of Fe-HAp in preparation
Yellow gel-like flocs appeared when the concentration of Fe(NO3)3solution was higher than 0.01 mol·L－1 and the maximum amount of these flocs was obtained for 0.05 mol·L －1Fe(NO3)3solution．However，such flocs were not observed when the concentration of Fe(NO3)3solution was higher than 0.05 mol·L －1．Therefore，three concentrations(0.01，0.05 and 0.1 mol·L－1，respectively)of Fe(NO3)3solution were prepared．These samples were noted as Fe-HAp0.01，Fe-HAp0.05and Fe-HAp0.1，
respectively with the footnotes representing the concentration of Fe(NO3)3solution．
2.2 Pore structure analysis of N2adsorption-desorption isotherms Pore structures of the HAp and Fe-HAp samples were investigated by a low temperature N2adsorption-desorption technique．As shown in Fig．1，the isotherms of the HAp and three Fe-HAp samples were identified as a type IV isotherm corresponding to mesoporou materials．Meanwhile，typical H3 hysteresis loops in a relative pressure(P/P0)range of 0.8～1 were characterized by isotherms of the HAp and three Fe-HAp samples．These H3 hysteresis loops did not have any limiting adsorption at high P/P0，indicating the presence of slit-shaped pores caused by aggregation of particles［14］．However，a typical H2 hysteresis loop was observed for the Fe-HAp0.05particles in addition to the H3 hysteresis loop at high P/P0(see Fig．1(c))．Although the H2 hysteresis loop nominally
corresponds to pores with narrow necks and wider bodies(ink bottle pores)，the network effect of inorganic oxide gels should be considered for the gel-like appearance of Fe-HAp0.05in the preparation［15］．It is interesting to note that such H2 hysteresis loop was not observed for the Fe-HAp0.1particles which were prepared at a much high
Fe(NO3)3concentration(see Fig．1(d))．Since intra-aggregate pores were characterized by hysteresis in a P/P0 range of 0.5～0.7，the H2 hysteresis loop indicates the possible existence of predominant intra-aggregate pores in the Fe-HAp0.05particles［16］．On the contrary，such intra-aggregate pores decreased in the Fe-HAp0.1particles with
Fig．1 Adsorption-desorption isotherms of HAp(a)，Fe-HAp0.01(b)，Fe-HAp0.05(c)and Fe-HAp0.1(d)
smaller particle sizes．This deduction is in according with the fact we observed in the preparation of Fe-HAp．When concentration of the Fe(NO3)3solution was higher than 0.05 mol·L－1，some yellow powder-like particles appeared in contrast to the yellow flocs obtained at 0.05 mol·L－1Fe(NO3)3solution．Meanwhile，the hysteresis in a P/P0range of 0.5～0.7 was very weak for the Fe-HAp0.01particles whereas such range hysteresis did not appear for the HAp particles．Therefore，we deduced that there were no intra-aggregate pores in HAp particles and the amount of intra-aggregate pores in Fe-HAp0.01particles was less than either that of Fe-HAp0.05particles or that of Fe-HAp0.1particles． 2.3 Specific surface area analysis
The specific surface area of the HAp and Fe-HAp samples were calculated according to the BET method．As illustrated in Fig．2，the specific surface area of Fe-HAp does not increase with concentration of
Fe(NO3)3solution．The Fe-HAp0.05has the largest specific surface area(204 m2·
g－1)but it was not prepared with the highest concentration of Fe(NO3)3solution．These BET results are in agreement with what we obtained in the pore structure analysis．Therefore，the largest specific surface area of Fe-HAp0.05is related to its special pore structure as discussed previously．
2.4 Pore size distribution analysis by BJH model
As seen in Fig．3，multimodal pore-size distributions were obtained using the BJH model．Sharp peaks at 3.8 nm were identified for the HAp，Fe-HAp0.01and Fe-HAp0.1particles in addition to those broad peaks related to large size inter-aggregate pores．These sharp peaks correspond to slit-shaped pores caused by aggregation of the particles as discussed previously．However，a predominant peak at 4.3 nm was observed in Fig．3(c)，suggesting the major mesopore sizes of Fe-HAp0.05particles were larger than that of any other sample．Meanwhile，the weak peaks related to inter-aggregate pores indicate the uniform distribution of 4.3 nm mesopores in Fe-HAp0.05particles．It should be noted that a shoulder around 4.3 nm in Fig．3(b)suggests existence of mesopores with similar sizes as Fe-HAp0.01．
Fig．2 Specific surface areas of HAp，Fe-HAp0.01，Fe-HAp0.05and Fe-HAp0.1
Fig．3 PSDs of HAp(a)，Fe-HAp0.01(b)，Fe-HAp0.05(c)and Fe-HAp0.1(d)obtained by the BJH model
2.5 Pore size distribution analysis by NLDFT model
As introduced previously，the traditional BJH model cannot provide an accurate PSD as the NLDFT model;therefore，a N2at 77 K on silica(cylinder，pore，adsorption branch)NLDFT model was selected to calculate the PSDs．As illustrated in Fig．4，good fittings with errors less than 0.85%were obtained for all the samples．However，the PSDs obtained from the selected NLDFT model were quite different from those obtained by the BJH model．The sharp peak at 3.8 nm did not appear and the pore
volume of HAp was in a range of 26 ～30 nm(see Fig．5(a))．The PSD obtained from the NLDFT model suggests that the major pores of HAp particles are inter-aggregate pores．On the contrary，prominent peaks at 6.3 nm were identified in the PSDs of all three Fe-HAp samples，which suggest that intra-aggregate pores play a big role in these Fe-HAp particles．The maximum pore volume of 6.3 nm peak was obtained in the Fe-HAp0.05particles，indicating the uniform size distribution of
mesopores(see Fig．5(c))．On the contrary，the 6.3 nm peak only takes a small part of total pore volume in either the Fe-HAp0.01or the Fe-HAp0.1particles(see Fig．5(b)and 5(d))．The different distribution of the 6.3 nm peak could be attributed to variation of intra-aggregate pores in the total pore volume as discussing previously．
Fig．4 The NLDFT model fitting of HAp(a)，Fe-HAp0.01(b)，Fe-HAp0.05(c)and Fe-HAp0.1(d) 2.6 Pore structure analysis by TEM
As seen from Fig．6(a)，the rod-like HAp particles aggregate in plate-like structures with inter-aggregate pores，which demonstrates that the PSD obtained from the NLDFT model is more accurate than that obtained from the BJH model．The Fe-HAp0.01particles were in the same aggregation as that of HAp particles in addition to a small amount of sphere-like particles(see Fig．6(b))．These sphere-like particles evolved from the gel-like flocs to the sphere-like ones in order to reduce surface energy．The Fe-HAp0.05particles were almost all sphere-like with obvious intra-aggregate pores(see Fig．6(c))．It should be noted that these intra-
aggregate pores would be hard to recognize when lattice or coordinated water in the Fe-HAp0.05particles was destroyed by heating due to the electron beam．On the contrary，the particle size(around 5 nm)of Fe-HAp0.1is much smaller than that(around 50 nm)of Fe-HAp0.05，which decrease the amount of intra-aggregate pores as suggested in the discussion of isotherms．The TEM results demonstrated deduction from the PSD of NLDFT model and confirmed the explanation of H2 hysteresis loop in analysis of isotherms previously．The pore sizes measured in the TEM images are closed to the results obtained from PSD of the NLDFT model，which demonstrates the NLDFT model is more accurate than the BJH model．
Fig．5 PSDs of HAp(a)，Fe-HAp0.01(b)，Fe-HAp0.05(c)，and Fe-HAp0.1(d)calculated with the NLDFT model
Fig．6 TEM images of HAp(a)，Fe-HAp0.01(b)，Fe-HAp0.05(c)and Fe-HAp0.1(d) 3 Conclusions
The mesopore size distribution of Fe-HAp prepared in Fe3+solution was successfully calculated by the selected NLDFT method．The PSDs obtained with the selected NLDFT model suggest that inter-aggregate pores are predominant in HAp particles while intra-aggregate pores increase gradually when the concentration of Fe(NO3)3solution is higher than 0.01 mol·L －1．The maximum amount of intra-aggregate pores of Fe-HAp was obtained at 0.05 mol·L－1Fe(NO3)3solution．When concentration of Fe(NO3)3solution was higher than 0.05 mol·L－1，the particle size of Fe-
HAp decreases drastically，which leads to intra-aggregate pore size decreasing．The results from the selected NLDFT model were confirmed by the TEM investigation，which demonstrates that the selected NLDFT model is more accurate than the traditional BJH model． References
［1］ ZUO K H，ZENG Y，JIANG D．Effect of cooling rate and polyvinyl alcohol on the morphology of porous hydroxyapatite ceramics［J］．Mater Des，2010，31:3090－3094．
［2］ GUO Y J，LONG T，CHEN W，et al．Bactericidal property and biocompatibility of gentamicin-loaded mesoporous carbonated hydroxyapatite microspheres［J］．Mater Sci Eng C，2013，33:3583 －3591．
［3］ IBRAHIM A R，WEI W，ZHANG D，et al．Conversion of waste eggshells to mesoporous hydroxyapatite nanoparticles with high surface area［J］．Mater Lett，2013，110:195－197．
［4］ TRECCANI L，YVONNE KLEIN T，MEDER F，et al．Functionalized ceramics for biomedical，biotechnological and environmental applications［J］．Acta Biomater，2013，9:7115 －7150．
［5］ YANG P，QUAN Z，LI C，et al．Bioactive，luminescent and mesoporous europium-doped hydroxyapatite as a drug carrier［J］．Biomater，2008，29:4341－4347．
［6］ LIN K，LIU P，WEI L，et al．Strontium substituted hydroxyapatite porous microspheres:surfactant-free hydrothermal synthesis，enhanced biological response and sustained drug release［J］．Chem Eng J，2013，
222:49 －59．
［7］ SARATH C V，BASKAR G，SUGANTHI R V，et al．Blood
compatibility of iron-doped nanosize hydroxyapatite and its drug release［J］．ACS Appl Mater Interfaces，2012，4:1200 －1210．
［8］ SALVIULO G，BETTINELLI M，RUSSO U，et al．Synthesis and structural characterization of Fe3+-doped calcium hydroxyapatites:role of precursors and synthesis method［J］．J Mater Sci，2011，46:910 －922． ［9］ WAKAMURA M，KANDORI K，ISHIKAWA T．Surface structure and composition of calcium hydroxyapatites substituted with Al(III)，La(III)and Fe(III)ions［J］．Colloids Surf A，2000，164:297 －305．
［10］ TANIZAWA Y，SAWAMURA K，SUZUKI T．Reaction characteristics of dental and synthetic apatites with Fe2+and Fe3+ions［J］．J Chem Soc Faraday Trans，1990，86:1071 －1075．
［11］ KHACHANI M，KACIMI M，ENSUQUE A，et al．Iron-calcium-hydroxyapatite catalysts:iron speciation and comparative performances in butan-2-ol conversion and propane oxidative dehydrogenation［J］．Appl Catal A，2010，388:113 －123．
［12］ OJEDA M L，ESPARZA J M，CAMPERO A，et al．On comparing BJH and NLDFT pore-size distributions determined from N2sorption on SBA-15 substrata［J］．Phys Chem Chem Phys，2003，5:1859 － 1866． ［13］ RAVIKOVITCH P I，NEIMARK A V．Calculations of pore size distributions in nanoporous materials from adsorption and desorption isotherms［J］．Stud Surf Sci Catal，2000，129:597 －606． ［14］ VERMA G，BARICK K C，MANOJ N，et al．Rod-like micelle
templated synthesis of porous hydroxyapatite［J］．Ceram Int，2013，39:8995 －9002．
［15］ SING K S W，EVERETT D H，HAUL R A V，et al．Reporting physisorption data for gas/solid systems with Special Reference to the Determination of Surface Area and Porosity［J］．Pure Appl Chem，1985，57:603 －619．
［16］ KUMAR K N P，KUMAR J，KEIZER K．Effect of peptization on densification and phase-transformation behavior of sol-gel-derived nanostructured titania［J］．J Am Chem Soc，1994，77:1396 －1400．
因篇幅问题不能全部显示，请点此查看更多更全内容