Nucleophilic substitution on silica surfaces: Comparisonof the reactivity of ¡- versus £-chlorosubstituted silanesin the reaction with sodium azide

Miriam KEPPELER,* Jürgen HOLZBOCK,* Johanna AKBARZADEH,**

Herwig PETERLIK** and Nicola HÜSING*,***,³

*Institute of Inorganic Chemistry I, Ulm University, Albert-Einstein-Allee 11, D-89069, Germany**Faculty of Physics, University of Vienna, Strudlhofgasse 4, A-1090 Vienna, Austria***Materials Chemistry, Paris-Lodron University Salzburg, Hellbrunner Str. 34, A-5020 Salzburg, Austria

Azido-group functionalized, hierarchically organized meso-/macroporous silica gels have been prepared through co-condensationof tetrakis(2-hydroxyethyl)orthosilicate with chloromethyltrimethoxysilane or 3-(chloropropyl)-triethoxysilane and subsequentconversion of the chloro groups. Azido functionalities have been obtained by nucleophilic substitution of the surface-bound chloromoieties with NaN3 in N,N-dimethylformamide. A strong dependence of the later azide density (N3 groups nm¹2) in the finalmaterial on the reaction conditions, such as temperature and time, but also on the spacer length (methylene versus propyl) of theorganofunctional silane has been observed. In principle, the nucleophilic substitution benefits from higher reaction temperaturesand times. However, while £-azido groups seem to be stable over a wide range of reaction temperatures and longer reaction times,¡-azido moieties tend to decompose at temperatures above 60°C. The structural features of the monolithic gels and the azidefunctionalities on the surface were determined by IR-ATR-spectroscopy, elemental analyses, N2-sorption analyses, small angleX-ray scattering and transmission electron microscopy.©2015 The Ceramic Society of Japan. All rights reserved.

Key-words : Sol–gel processing, Hybrid materials, Nucleophilic substitution, ¡-silane, £-silane, Azides

[Received April 21, 2015; Accepted June 25, 2015]

1. Introduction

Tailored porosity in inorganic­organic hybrid materials is ofvital importance for a wide variety of applications, such as sepa-ration, adsorption, catalysis, energy storage, etc.1)­3) The combi-nation of the structural characteristics of porous materials, such ashigh specific surface areas, uniform pore sizes, long range orderof the pores or even tailored porosity on several length scales(micro-, meso- and macroporous size regime) with the possibilityto deliberately and selectively functionalize the network to giveinorganic­organic hybrid materials opens up almost unlimitedpossibilities.3) As the most prominent examples for hybrids,silica-based hybrid materials have been extensively investigatedin the last decades.3)­10) In this context, post synthetic approaches(grafting) or alternatively co-condensation reactions of alkoxy-silanes modified with organic groups are often applied.8) The postsynthetic approach can have several disadvantages such as a lossof control of the loading, inhomogeneous distribution of thefunctional groups on the inner pore surfaces or pore blocking.Therefore, an often applied alternative is the co-condensationof a tetraalkoxysilane [Si(OR)4] and an organotrialkoxysilane[RBSi(OR)3], which is however, only applicable to give a homo-geneous distribution of the organic groups in the final network,if the group RB is not too bulky, and when homocondensation ofeither one of the silanes is not favored.11) The organic moiety onthe silica surface is even capable to undergo further chemical

reactions, i.e. by click reactions.12) Despite the fact that manyreports can be found on follow-up reactions on surface-boundorganofunctional silanes, the number of studies dealing withmore detailed insights on the reaction conditions or structuralimplications is rather limited.Based on our previous work on the synthesis of hierarchically

organized silica monoliths starting from tetrakis(2-hydroxyethyl)-orthosilicate in the presence of a structure-directing agent(Pluronic P123μ) in aqueous HCl, we recently reported thesuitability of click chemistry, specifically the Cu(I)-catalyzedazide-alkyne cycloaddition to functionalize these macro-/meso-porous silica monoliths with various organic substrates.13)­16)

Starting from chloroalkyl-modified silica gels, which were ob-tained from a co-condensation of tetrakis(2-hydroxyethyl)ortho-silicate with chloromethyltrimethoxysilane or 3-(chloropropyl)-triethoxysilane, azide functionalities are introduced by nucleo-philic substitution of the chloro groups. The structure of the silicamonoliths is not only completely preserved during the introduc-tion of the azides, but is even changed for the better with respectto an enlargement of the lattice constant and the periodicrepeating unit as well as an increase of the pore volume and thepore diameter with simultaneous reduction of the pore wallthickness. During the subsequent click reaction the attractivestructural features of the material could be fully maintained. Inanother work, Alauzun et al. reported a direct synthesis ofordered (hexagonal phase) mesoporous silica with iodopropylgroups located in the pore channels. However, it was not possi-ble to introduce iodopropyl groups by a co-condensation of aiodopropyl containing trialkoxysilane in the presence of an acidic(HCl(aq)) solution of the structure-directing agent P123. The

³ Corresponding author: N. Hüsing; E-mail: nicola.huesing@sbg.ac.at

‡ Preface for this article: DOI http://dx.doi.org/10.2109/jcersj2.123.P9-1

Journal of the Ceramic Society of Japan 123 [9] 764-769 2015 Paper

©2015 The Ceramic Society of Japan

DOI http://dx.doi.org/10.2109/jcersj2.123.764

764

iodopropyl-functionalities were converted to chloropropyl groupsin such reaction conditions.17)

To get a versatile click-able material, the number and acces-sibility of introduced azides that belong to the most importantfunctionalities for click reactions, is of great interest. Especiallywhen a mesoporous material with pore diameters in the nano-meter range is used, the spacer length between the silica surfaceand the azide functionality should be taken into account. Veryshort spacers are favoured for further modifications and reactionsin such narrowly dimensioned pores. However, it has been shownthat the spacer length might have a decisive influence on thechemical reactivity of a silane. The term ¡-effect was introducedto describe the unique reactivity of ¡-silanes at both the siliconatom and the functional group X, however the ¡-effect is notsharply defined and strongly depends on the nature of the sub-stituents on the silicon atom.18)­20)

In this study, we compare the reactivity of chloro-functional-ized, porous and monolithic silica materials with a specific focuson the reactivity of the respective ¡-chloro-functionalized silses-quioxane of the formula type SiO2/O1,5SiCH2Cl compared to theanalogous £-silane based silsesquioxane SiO2/O1,5Si(CH2)3Cltowards the nucleophilic substitution reaction in the presence ofsodium azide. The influence of reaction temperature, reactiontime, and surface chemistry (hydrophilic versus hydrophobic)will be investigated in detail.

2. Experimental Section

2.1 MaterialsAs silanes, tetraethyl orthosilicate (TEOS, Fluka), 3-chloropro-

pyl(triethoxy)silane (CPES, Aldrich), chloromethyl(trimethoxy)-silane (CMTMS, Wacker Chemie AG), and chloro(trimethyl)-silane (TMCS, Merck) were used without further purification.Ethylene glycol (EG, Aldrich) was purified by drying withNa2SO4 and distillation from Mg. Pluronic P123 (Mav = 5800),EO20PO70EO20 (BASF), N,N-dimethylformamide (DMF, VWR)and NaN3 (Merck) were applied without purification.

2.2 Preparation of hexagonally organized meso-porous chloroalkyl-modified silica gels

Tetrakis(2-hydroxyethyl) orthosilicate (EGMS) was synthe-sized according to Brandhuber et al.21),22) Chloroalkyl-modifiedsilica gels were prepared by co-condensation of EGMS andCMTMS or CPES in an aqueous reaction mixture containingPluronic P123 as the structure-directing agent and HCl(aq) as thecatalyst. The percentage by weight ratio was SiO2/P123/1MHCl:18/30/70wt% in which EGMS and CMTMS were used ina nominal molar ratio of either 9:1 (3.0mmol CMTMS), 6.75:1(4.5mmol CMTMS) or 4:1 (6.0mmol CMTMS) and EGMS andCPES in a nominal molar ratio of 9:1 (3mmol CPES) accordingto Keppeler et al.13) The reaction mixture was homogenized for1min using a vortex stirrer to obtain a viscous white mixture,which was allowed to gel in a closed PE cylinder at 40°C andkept for 7 d at the same temperature for ageing. Removal ofPluronic P123, water and ethylene glycol was either conductedby immersion of the wet gels in a solution of 10wt-% of TMCSin petroleum ether (PE) for 24 h and subsequent washing withPE and ethanol; or simply by soaking in PE (3 times in 24 h),followed by washing with ethanol (3 times in 24 h).

2.3 Nucleophilic substitutionThe wet chloroalkyl-modified silica gels were immersed into

a saturated solution of NaN3 in DMF at different temperaturesand several exposure times, and were subsequently washed by

repeated immersion in water (five times within 24 h) and ethanol(three times within 48 h) to remove non-reacted NaN3. Drying ofthe wet silica gels was performed by simple evaporation of thesolvent under reduced pressure.22)

2.4 CharacterizationSpecific surface areas were determined with N2-sorption analy-

ses on a NOVA 4000e (Quantachrome). Prior to analysis thesamples were degassed at 60°C for 3 h. BET surface areas wereevaluated using adsorption data in a relative pressure range from0.05­0.30 with a 5-point-analysis. The amount of C, H and N ona percentage basis was determined by elemental analysis usingan ELEMENTAR Varino. FT-IR-ATR (Fourier transform infaredspectroscopy in the attenuated total reflection mode) spectra wererecorded on a Bruker Tensor 27. Small angle X-ray scattering(SAXS) measurements were performed with a Bruker Nano-Starusing a pinhole camera with an X-ray generator (CuK¡ radiationmonochromatized by cross-coupled Goebel mirrors, U = 40 kV,I = 35mA). TEM investigations were performed on a Philips EM400 (80 kV). The azide density [N3 nm¹2] on the surface wasestimated by the specific surface area (SBET) and the percentageof nitrogen (assuming that the azide functionality is the onlynitrogen-containing group in the sample) according to:

µðN3Þ ¼mN � NA

MN � 100 � 3 � SBET � 1018

Where mN is the mass of nitrogen in 100 g of the silica gel,NA is Avogadro’s constant, MN is the molar mass of nitrogenand SBET is the specific surface area according to the Brunauer­Emmett­Teller (BET) model. The divisor 3 reflects that one azidegroup consists of three nitrogen atoms.

3. Results and discussion

As the silica source tetrakis(2-hydroxyethyl)orthosilicate wasapplied for two reasons: on one hand, the released glycol showsa better compatibility to the liquid crystalline phases of thestructure-directing agent P123 compared to monoalcohols, suchas ethanol from e.g. tetraethyl orthosilicate, and on the other nocompatibilizing solvent is necessary and hydrolysis and con-densation reactions can take place in a purely aqueous solu-tion.13),14),21) For a better visualization, the resulting monolithsare presented in the ESI, Fig. 1 for a pure silica example. Forthe investigation of the influence of the spacer chain length onthe nucleophilic substitution reactions, chloro-substituted silicagels were prepared by a co-condensation approach.15) In principleseveral pathways could lead to azido-functionalized silica sur-faces: 1) the surface grafting agent (3-azidopropyl)trialkoxysilanecould be synthesized prior to sol­gel processing from 3-chloropropyltrialkoxysilanes.23) However, the ¡-analogue hasnot been reported so far and a stronger influence of the azidomoiety on the phase separation process and thus on the hier-archically organized network structure is expected; 2) Sol­gelprocesses could be catalyzed by HI instead of HCl to facilitate theformation of the azido functionalized surface in the nucleophilichalide-azido exchange. In this case a strong influence on the finalnetwork structure can be expected as has been shown recently byusing HBr or HNO3 as the sol­gel catalyst.24) 3) The iodopropyl-derivatized silanes could be applied, but again in the presence ofHCl halide exchange reactions are expected.17)

In our previous work, we already observed during sol­gelprocessing significant differences between the respective chloro-functionalized ¡-silane, (MeO)3SiCH2Cl (CMTMS), and theanalogous £-silane, (EtO)3Si(CH2)3Cl (CPES), reflected in the

Journal of the Ceramic Society of Japan 123 [9] 764-769 2015 JCS-Japan

765

structural characteristics of the gel. Gels modified with chloro-methyl groups typically exhibit higher specific surface areas(SBET) and larger mesopores with a narrower size distribution thansamples prepared from chloropropyl-modified gels. Gels preparedwith the £-silane (EtO)3Si(CH2)3Cl show a drastically decreasedpore volume and a lower specific surface area compared to theCMTMS analogues. This can probably be attributed to a) thedifferent reactivity towards hydrolysis and condensation, butb) also to the different polarities of the molecules. Nucleophilicsubstitution of chloride into azide in saturated solutions of NaN3

in DMF either at 65°C for methylene-spacered samples or at 80°Cfor propylene-spacered samples have been conducted success-fully.15) For the following discussion all data are taken from sam-ples prepared from gels with a starting EGMS/CMTMS ratio of6.75:1 and EGMS/CPES ratio of 9:1, except otherwise men-tioned. This implies that the number of chloro-functionalities isintrinsically lower for the gels containing the £-spacered function.In our previous report we also observed a high tendency of

the ¡-azide to degrade.15) In this study a more detailed investi-gation of the influence of reaction temperature (20, 40, 60, 80and 100°C) and time (1, 2, 3, 4, 5 and 6 d) on the yield of thenucleophilic substitution reaction (SN-reaction) is presented. Onehas to keep in mind that in our case, the reaction with the azide isperformed directly on the functionalized, highly porous mono-lithic body, thus limitations due to diffusion or accessibility of theactive sites are expected and are accounted for by longer reactiontimes and higher temperatures than typically used. The samplesthat have been obtained after nucleophilic substitution with NaN3

in DMF are azidomethyl- and azidopropyl-modified hierarchi-cally organized silica gels comprising meso- and macropores(schematically shown in Fig. 1). These gels have been treatedwith trimethylchlorosilane (TMCS) in order to remove reactivesilanol groups, extract the structure-directing agent (P123), andfacilitate drying of the monoliths. Opaque, crack-free monolithicmaterials have been obtained as shown in Fig. 1, right.The first part of the manuscript is devoted to the influence of

the reaction temperature. Compared to a iodo functionality thechloro moiety is a suboptimal nucleofuge, requiring higher tem-peratures for the nucleophilic substitution. The reaction temper-ature was optimized with respect to the highest azide yields forboth precursor gel systems. Lower reaction temperatures resultedin incomplete reaction, while higher reaction temperatures ledto degradation of the functional units and thus, again to loweryields. In principle, the successful transformation of the chloridesinto azides has been confirmed for all samples by IR-ATR spec-troscopy (exemplarily shown for two samples in Fig. 2). A newvibration band at approximately 2100 cm¹1 is detected that isattributed to the antisymmetric stretching vibration of the cumu-lated double bond of the terminal azido groups.25)

As previously described, methylene-spacered samples tend toadsorb higher amounts of nitrogen during nitrogen sorption, havelarger pore diameters (DBHJ,Des) and slightly higher specific sur-

face areas (SBET) compared to the propyl-spacered samples.15)

The structural parameters for azido-modified silica gels obtainedafter exposure to DMF/NaN3 for 2 d at different temperaturesare exemplarily given in Table 1 (a complete overview of thestructural parameters of azidoalkyl-modified samples synthesizedwith different exposure temperatures and times are listed inTable 1/ESI).All samples exhibit isotherms with H1-hysteresis loop (ESI,

Fig. 2) according to the classification of Sing, which is character-istic for mesoporous materials with a well pronounced long rangeordering of the mesopores.26)

These data indicate that the reaction temperature of thenucleophilic substitution reaction up to 60°C has no significantinfluence on the structural features of the hybrid silica gels. Above80°C, the pore volumes and specific surface areas areas areslightly increasing. The hexagonal arrangement of the mesopores(especially for the CMTMS samples) is still unchanged asdetected by TEM and SAXS measurements. TEM images illus-trate the typical honeycomb structure of the pore entries or thelong-ranged 2-dimensional pore channels. The SAXS patternshow higher order reflections with the characteristic sequence fora 2-dimensional hexagonal ordering of 1:31/2:2 + (Fig. 3).27)

The final functionalization density with azido groups wasdetermined by elemental analysis (Table 2). For the monolithsample treated for 2 days two different trends can be observed:Whereas the number of inserted azides per nm2 for propyl-spacered samples increases with increasing SN-reaction temper-ature, a clear dependence on temperature for methylene-spaceredsamples is observed. Samples from CPES show an azide densityof about 0.15N3 nm¹2 at 20°C that is increased to 0.94N3 nm¹2

by a SN-reaction temperature of 100°C. For samples fromCMTMS the obtained azide density for a reaction temperature of20°C is in the range of 0.18N3 nm¹2, reaches a maximum at

Fig. 1. Schematic presentation of the nucleophilic substitution conducted on functionalized, porous hybrid silica gels (left);photograph of an azidoalkyl-modified silica gel monoliths (right).

Fig. 2. IR-ATR-spectra of azidoalkyl-modified silica gels and a chloro-containing precursor silica gel.

Keppeler et al.: Nucleophilic substitution on silica surfaces: Comparison of the reactivity of ¡- versus £-chlorosubstituted silanes in the

reaction with sodium azideJCS-Japan

766

60°C with 1.04N3 nm¹2 and lowers to 0.39N3 nm¹2 at 100°C.More information on samples treated with other exposure tem-peratures and times are listed in Table 2/ESI).To obtain a better understanding of the influence of the reac-

tion time, the number of treatment days for the SN reaction wasvaried from 1 to 6 days (see Table 1, ESI). Again, very cleartrends, supporting the data obtained for a reaction time of 2 days,are obtained. For the £-azides, the number of inserted £-azidesexhibits a saturation behavior at approximately 80°C (Fig. 4, left).Whereas for methylene-spacered samples a maximum of the¡-azide density is obtained at the range of 60°C and 5 d, where an¡-azide density of 1.18N3 nm¹2 is reached and a decrease of the¡-azides density is observed at higher SN-reaction temperatures(Fig. 4, right). In addition, for the series at 100°C, the ¡-azidedensity continuously decreases with prolonged reaction time inthe order of 0.39 (2 d), 0.27 (3 d), 0.19 (4 d), 0.15 (5 d) and 0.11(6 d) N3 nm¹2. To exclude an influence of the reaction medium(sodium azide in DMF), it has been exchanged to a fresh solutionof saturated NaN3 in DMF in one sample after two days. However,no influence on the yield of azides was observed. This indicates

that £-azides on nanostructured silica surfaces are more stablecompared to ¡-azides; the latter ones tend to decompose, probablyfacilitated through higher temperatures and presumably due to adestabilization caused by the close proximity to the silicon center(¡-effect). It can be expected that in analogy to carbene formationfrom diazoalkanes, nitrenes are formed via thermolysis of theazide function, followed by further reactions.28)

However, there is a great interest to synthesize azidomethyl-modified silica gels with a sufficient ¡-azide density, becausespacers as short as possible are favoured for further modificationsand reaction in such narrowly dimensioned pore systems. The ¡-azide density can also be increased by simply applying a largerCMTMS concentration in the co-condensation reaction and thusprovide a larger number of chloro-moieties for the reaction. Thehexagonal arrangement of the mesopores and the high degreeof order can be fully maintained by extending the CMTMS con-centration up to 6,0mmol.Whereas for propyl-spacered samples a strong influence on the

structure of the gel network is observed with increasing amount ofCPES (the limit is around 3.0mmol), this is not the case for the

Fig. 3. SAXS pattern (l) and TEM-image (r) of azidomethyl (CMTMS)-modified meso-/macroporous silica gel monolith(EGMS and CMTMS in a nominal ratio of 4.5; 6.0mmol CMTMS).

Table 2. Amount of nitrogen on a percentage base and azide density depending on SN-reaction temperature and the CH2-spacer length(reaction time for SN: 2 d)

SN-time [d] SN-temperature [°C]methylene-spacer propyl-spacer

N [%] ¡-azide density [N3 nm¹2] N [%] £-azide density [nm¹2]

2

20 0.83 0.18 0.63 0.1540 3.42 0.73 0.63 0.1460 5.25 1.04 2.56 0.5280 4.43 0.99 4.37 0.85100 2.01 0.39 4.74 0.94

Table 1. Structural parameters of azidoalkyl-modified hierarchically organized silica gels (reaction time for SN: 2 d)

SN-temperature [°C]methylene-spacer [¡-azides] propyl-spacer [£-azides]

Vmax [cm3 g¹1] DBJH,Desa) [nm] SBETb) [m2 g¹1] Vmax [cm3 g¹1] DBJH,Des

a) [nm] SBETb) [m2 g¹1]

20 626.9 5.38 645 415.1 3.93 61940 652.2 5.76 670 441.2 3.68 65160 665.0 5.38 726 478.6 3.92 70980 690.0 5.78 642 566.2 4.70 736100 852.5 5.76 736 595.4 4.72 725

a) Calculated from the desorption isotherm by using the BJH model. b) Calculated by using the Brunauer-Emmett-Teller (BET) model.

Journal of the Ceramic Society of Japan 123 [9] 764-769 2015 JCS-Japan

767

CMTMS modified silica samples. An increase of the CMTMSconcentration in sol­gel processing from 3.0, 4.5 to 6.0mmolCMTMS leads to the expected increase of the ¡-azide density,indicating, that there is a good accessibility of the chloro groupsin the filigree nanostructured network (Fig. 5). Exemplarily, forsilica gels prepared from a silica precursor solution containing 3.0mmol CMTMS and subsequent SN-reaction at 60°C an ¡-azidedensity of 0.66N3 nm¹2 has be obtained. This value is increasedto 1.34N3 nm¹2 by use of a silica precursor solution containing6.0mmol CMTMS and keeping the temperature of the SN-reactionconstant at 60°C (Table 3). An interesting additional trend isthat with increasing amount of CMTMS on the silica surface, thedegradation of the ¡-azide is not as pronounced anymore, whichis reflected by the absence of the maximum that was found for thesamples with 3.0mmol CMTMS at 60°C reaction temperature.The latter point let us to have a closer look on the surface

chemistry of the silica gels. Typically the gels are treated withTMCS, which reacts with free surface silanols to give a tri-

methylsilylated and hydrophobic surface. Simultaneously, thestructure-directing agent is extracted from the gel body anddrying upon preservation of the filigree structure is facilitated.29)

However, it can be assumed that the inserted trimethylsilylgroups reduce the azide density, because of steric hindranceduring the conversion of the chlorides into azides, which isespecially true for methylene-spacered samples.Therefore, a close look has been given to samples, in which the

template has been removed by extraction with ethanol and DMF.Here, a polar surface carrying a large amount of silanol groupsand less sterically hindered ¡-chloro-functionalities is obtained.For these samples a slightly higher ¡-azide density has beenfound after nucleophilic substitution compared to the TMCStreated samples supporting the above made assumptions. Uponoptimized synthesis conditions it is possible to generate a hexa-gonally organized mesoporous silica network modified with an¡-azide density of more than 2.0N3 groups nm¹2 (6.0mmolCMTMS, 60°C, template removal: extraction, Table 3).

Fig. 4. Increase of nitrogen on a percentage base depending on SN-reaction time and temperature for propyl- (left) andmethylene- (right) spacered silica gels.

Fig. 5. ¡-Azide density depending on CMTMS concentration, temperature of SN-reaction and method of template removal(TMCS or extraction).

Keppeler et al.: Nucleophilic substitution on silica surfaces: Comparison of the reactivity of ¡- versus £-chlorosubstituted silanes in the

reaction with sodium azideJCS-Japan

768

4. Conclusion

In summary we could successfully demonstrate the synthesisof hexagonally organized mesoporous, azidoalkyl-functionalizedsilica gels (alkyl = methylene and propyl) via a co-condensationapproach containing chloroalkyl-precursors and subsequent con-version of the chlorides into azides through SN-reaction. Thereactivity of the chloro-, but also the stability of the inserted azidefunctionalities on the nanostructured surface strongly depends onthe CH2-spacer length, which was varied between methylene andpropylene-based spacers. While azidopropyl groups are stableover 100°C, the azidomethyl groups tend to decompose above60°C depending on reaction time and surface loading. The bestSN-reaction conditions in order to achieve the highest possibleloading with azide functionalities, has been found for the reac-tion at 60°C and 5 d. Furthermore the ¡-azide density can beincreased with higher concentrations of the chloromethyl silaneprecursor that is used to insert the chloromethyl groups via co-condensation. In addition to that a slightly increased ¡-azidedensity could be found for samples, in which the structure-directing agent P123 has been removed via extraction insteadof surface silylation since the steric hindrance caused by thetrimethylsilyl groups is avoided.

Acknowledgements We acknowledge the financial support ofthe Deutsche Forschungsgemeinschaft DFG within the collaborativeresearch network SFB 569. We thank the Wacker Chemie AG forproviding CMTMS and we also gratefully acknowledge C. Eggerfor the N2-sorption measurements and M. Lang for the elementalanalyses.

References1) B.-L. Su, C. Sanchez and X.-Y. Yang, “Hierarchically

Structured Porous Materials”, Wiley-VCH Verlag &Co. KGaA,Weinheim, (2012).

2) A. Inayat, B. Reinhardt, H. Uhlig, W. D. Einicke and D. Enke,Chem. Soc. Rev., 42, 3753­3764 (2012).

3) S. Hartmann and N. Hüsing, “Hybrid Nanocomposites forNanotechnology - Electronic, Optical, Magnetic and Biomed-ical Applications”, Ed. by L. Merhari, Springer, New York(2009) pp 131­171.

4) M. P. Conley, C. Coperet and C. Thieuleux, ACS Catalysis, 4,1458­1469 (2014).

5) J. E. Lofgreen and G. A. Ozin, Chem. Soc. Rev., 43, 911­933(2014).

6) P. Van Der Voort, D. Esquivel, E. De Canck, F. Goethals, I.Van Driessche and F. J. Romero-Salguero, Chem. Soc. Rev., 42,3913­3955 (2013).

7) D. J. Macquarrie, D. B. Jackson, J. E. J. Mdoe and J. H. Clark,New J. Chem., 1, 539­544 (1999).

8) C. Walcarius and C. Delacôte, Chem. Mater., 15, 4181­4192(2013).

9) A. Feinle and N. Hüsing, “Hybrid Materials - Synthesis,Characterization, Applications”; Ed. by G. Kickelbick, Wiley-VCH, Weinheim (2015) in press.

10) A. Mehdi, C. Reyé, S. Brandès, R. Guilard and R. J. P. Corriu,New J. Chem., 29, 965­968 (2005).

11) U. Schubert and N. Hüsing, “Synthesis of Inorganic Materials”,3. Edition, Wiley-VCH, Weinheim (2012).

12) X. Cattoën, A. Noureddine, J. Croissant, N. Moitra, K.Bürglová, J. Hodačová, O. de los Cobos, M. Lejeune, F.Rossignol, D. Toulemon, S. Bégin-Colin, B. P. Pichon, L.Raehm, J.-O. Durand and M. Wong Chi Man, J. Sol-Gel Sci.Technol., 70, 245­253 (2014).

13) C. Triantafillidis, M. Elsaesser and N. Hüsing, Chem. Soc. Rev.,42, 3833­3846 (2013).

14) S. Hartmann, D. Brandhuber and N. Hüsing, Acc. Chem. Res.,40, 885­894 (2007).

15) M. Keppeler and N. Hüsing, New J. Chem., 35, 681­690(2011).

16) M. Keppeler, J. Holzbock, J. Akbarzadeh, H. Peterlik and N.Hüsing, Beilstein J. Nanotechnol., 2, 486­498 (2011).

17) J. Alauzun, A. Mehdi, C. Reyé and R. J. P. Corriu, New J.Chem., 31, 911­915 (2007).

18) N. W. Mitzel, K. Vojinović, T. Foerster, H. E. Robertson, K. B.Borisenko and D. W. H. Rankin, Chem.-Eur. J., 11, 5114­5125(2005).

19) N. W. Mitzel, K. Vojinović, R. Fröhlich, R. T. Foerster, H. E.Robertson, K. B. Borisenko and D. W. H. Rankin, J. Am.Chem. Soc., 127, 13705­13713 (2005).

20) A. Berkefeld, C. Fonseca Guerra, R. Bertermann, D. Troegel,J. O. Daiß, J. Stohrer, F. M. Bickelhaupt and R. Tacke,Organometallics, 33, 2721­2737 (2014).

21) D. Brandhuber, V. Torma, C. Raab, H. Peterlik, A. Kulak andN. Huesing, Chem. Mater., 17, 4262­4271 (2005).

22) D. Brandhuber, N. Huesing, C. K. Raab, V. Torma and H.Peterlik, J. Mater. Chem., 15, 1801­1806 (2005).

23) H. Hettegger, I. Sumerskii, S. Sortino, A. Potthast and T.Rosenau, ChemSusChem, 8, 680­687 (2015).

24) S. Flaig, J. Akbarzahdeh, H. Peterlik and N. Hüsing, J. Sol-GelSci. Technol., 73, 103­111 (2015).

25) D. A. Flemming, C. J. Thode and M. E. Williams, Chem.Mater., 18, 2327­2334 (2006).

26) K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A.Pierotti, J. Rouquérol and T. Siemieniewska, Pure Appl. Chem.,57, 603­619 (1985).

27) S. S. Soni, G. Botons, M. Bellour, T. Narayanan and A.Giboud, J. Phys. Chem. B, 110, 15157­15165 (2006).

28) H. G. O. Becker, Organikum, 21. Edition, Wiley-VCH,Weinheim (2001) p. 286.

29) D. Brandhuber, H. Peterlik and N. Hüsing, J. Mater. Chem., 15,3896­3902 (2005).

Table 3. ¡-Azide density depending on CMTMS concentration and the method of template removal (TMCS or extraction)

SN-temperature [°C]

¡-azide density [N3 nm¹2]

3.0mmol CMTMS 4.5mmol CMTMS 6.0mmol CMTMS

TMCS extraction TMCS extraction TMCS extraction

45 0.59 0.61 0.69 0.80 1.12 1.1650 0.63 0.70 0.85 0.94 1.24 1.3855 0.69 0.76 1.11 1.16 1.35 1.4560 0.66 0.77 1.20 1.19 1.34 2.0665 0.88 1.06 1.30 1.20 1.55 1.3970 0.74 0.92 1.20 1.30 1.78 1.88

Journal of the Ceramic Society of Japan 123 [9] 764-769 2015 JCS-Japan

769