高手话不多that for zeolite Y syntheses very similar to
T2.4,Fig.3.TEM images of zeolite Y nanocrystals synthesized in
batch T2.4(A)and in batch T3.6(B and C).B.A.Holmberg et al./Microporous and Mesoporous Materials 59(2003)13–2819
crystallization follows a gel particle periphery nu-cleation mechanism.XRD analyses clearly indi-cate that at42h there was both crystalline and amorphous suspended solid material in the syn-thesis solution.However,it is possible that our suspended material was a mixture of crystals and amorphous gel particles,which would have arisen from a nucleation mechanism other than the gel particle periphery nucleation mechanism.From our data,it is not clear whether the crystalline material at42h is contained within gel particles(as suggested by Mintova et al.[22])or there were separate zeolite nanocrystals that nucleated in the solution outside the amorphous gel particles.
Between66and215h,when almost all of the suspended material in the synthesis solution was crystalline(Fig.2A)we observed an order of magnitude drop in T2.4crystal number concen-tration.This drop in crystal number concentra-tion,accompanied by growing crystal size and roughly constant suspended solid concentrations with longer synthesis durations lead us to believe that Ostwald ripening of the crystals in T2.4occurs after66h(Fig.5).
Crystallization in T3.6with TMABr was much di?erent than T2.4.From Fig.2B,the particles in T3.6at42h of crystallization possess no crystal-linity,and at54h they have reached almost full crystallinity.The number concentration of nano-crystals at54h(2.2?1014#/g sol)was twice that of amorphous gel particles at42h(1.1?1014#/g sol).From these data we can conclude that in T3.6, solution phase zeolite crystal nucleation is a sig-ni?cant part of the overall nucleation process.The gel periphery nucleation mechanism proposed by Mintova et al.[22]simply could not be the only contributing nucleation mechanism in T3.6be-cause there were not enough gel precursor particles at42h for all the zeolite nanocrystals at54h to originate from.After54h,the crystal number concentration only decreases a small amount,in-dicating no Ostwald ripening e?ect.
The mechanism behind the particle size stabili-zation observed in T3.6is not fully understood. Clearly the crystal size in T3.6has been stabi-lized when compared to T2.4;no appreciable growth or dissolution occurs after138h.This occurs despite the fact that the initial amount of the mineralization agent OHàcharged in both T2.4and T3.6was identical.Two di?er-ent stabilization mechanisms could be a?ecting the T3.6synthesis.The?rst mechanism involves the stabilization of the zeolite Y framework,in par-ticular the sodalite cages,by the occlusion of
an
ion.The TG analyses(Table3)indicate that T3.6 on a weight basis contains about13%more TMA than T2.4.Though we have not veri?ed the loca-tion of this extra template(be it in the sodalite cage or the supercage),it is reasonable to hypothesize that at least a portion of it resides within the sodalite structure of the zeolite.This extra template in the sodalite cage would help stabilize the sodalite structure,making it less susceptible to dissolution.TMA may not be the only ionic species helping to stabilize the zeolite structure;Cocks and Pope[38] reported an average of0.38?0.10halide anions were trapped within each zeolite X sodalite cage synthesized with10–50%of the synthesis alkali earth metals(Na and K)in the halide form.Un-fortunately we have no elemental analyses data for bromide,so its presence must be left to speculation for
now.
The second proposed mechanism behind this stabilization involves the idea that the extra TMABr in the T3.6synthesis could modify the surface charge of the colloidal zeolite,making it less susceptible to dissolution or growth by reac-
tive species such as OHàand Al(OH)à
4.OHàis
well known as a mineralizing agent capable of
dissolving crystals,and Al(OH)à
4is one of the re-
active species instrumental in the growth of alu-minated zeolites.Growth and dissolution are both processes that occur at the interface between the zeolite and the bulksolution.If the surface charge of the colloidal zeolite nanocrystal is modi?ed in such a way that transport of these reactive species to the surface of the nanocrystal was limited,then crystal growth or dissolution could not take place and crystal size would be stabilized in a fashion similar to what we have observed.In past litera-ture[39,40],it has been experimentally shown that salt anions preferentially adsorb on the external surface of large-pore zeolite nanocrystals like zeo-lite Y rather than imbibe within the pore structure. It is thought that preferential external adsorption occurs due to the Donnan exclusion e?ect in the pore structure and that more e?ective ion hydra-tion that can be achieved on the zeolite surface. Therefore it is possible that a TMABr salt solution could a?ect the surface properties of colloidal zeo-lite crystals.However,the presence of external adsorption of salt ions under during zeolite syn-thesis conditions has not been experimentally veri?ed in our study.The e?ects on zeolite crys-tallization that would arise due to salt solution surface charge modi?cation are also unknown;therefore ideas on its role in the particle size sta-bilization of the T3.6synthesis are speculative.
The factors contributing to the improved nu-cleation observed in T3.6also are not fully un-derstood.Increasing template concentration has been shown in the past to increase crystal nucle-ation,and we believe it is the primary explanation for the increase in nucleation observed in T3.6 versus T2.4.However,it is possible that anion concentrations could also be a contributing factor to the nucleation di?erences in the T2.4and T3.6. If additional TMABr were able to modify the surface properties of nanocrystals and help to stabilize them,the same e?ect could stabilize newly formed nuclei and promote their survival past the early stages of nucleation.
3.3.E?ect of OHàconcentration and Brà/OHàratios
In the second investigation of anion concen-tration on zeolite Y nanocrystal synthesis,the (TMA)2O/Al2O3ratio was held constant,based on T3.6,while the TMABr/TMAOHeSTratio was varied(Table2).These batches are denoted ac-cording to their TMABr/TMAOHeSTratio(batch T3.6doubles as batch S0.50).This allowed the separate investigation of the e?ect of two impor-tant synthesis parameters,tetramethylammonium concentration and anion concentration.XRD patterns showed that crystalline zeolite Y products were obtained for syntheses with OHàcharging concentrations between1.4and0.86mol/L and TMABr=TMAOH60:88(Fig.6).Batches with
Table2
Synthesis conditions for batches with varying OHàconcentration and Brà/OHàratios
Batch S0.15S0.30S0.50(T3.6)S0.67S0.88S1.16S1.69 Q,(TMA)2O{2OHà}/Al2O3 3.13 2.76 2.39 2.15 1.91 1.65 1.40 R,(TMA)2O{2Brà}/Al2O30.460.83 1.20 1.44 1.69 1.94 2.39 (TMA)2O/Al2O3 3.59 3.58 3.59 3.59 3.60 3.59 3.60 S,TMABr/TMAOH0.150.300.500.670.88 1.16 1.69 OHàcharged(mol/L) 1.40 1.24 1.080.970.860.750.62 High crystallinity after$90h
crystallization
Yes Yes Yes Yes Yes None None
Mass median diameter after $90h crystallization 52.540.437.136.235.460.217.0
22 B.A.Holmberg et al./Microporous and Mesoporous Materials59(2003)13–28
OHàcharging concentrations less than0.86and TMABr=TMAOH P1:16did not yield zeolite Y, even with extended crystallization times.It has been shown in literature that higher alkalinities are required to nucleate faujasite structures,while faujasite crystallization can be carried out at lower alkalinities[37].Therefore,the lack of crystalline material generated by the above conditions is be-lieved to be due to the OHàconcentration(Table 2)being lower than the minimum level required for zeolite Y nucleation.
When total(TMA)2O/Al2O3was held constant at3.60,nanocrystal mass median diameter after $90h crystallization decreased from53nm for batch S0.15(OHàcharged?1.40mol/L,TMABr/ TMAOH?0.15),to33nm for batch S0.88(OHàcharged?0.86mol/L,TMABr/TMAOH?0.88) (Table2,Fig.7).Batches S1.16and S1.69,(OHàcharged<0.86mol/L,TMABr=TMAOH>0:88) did not yield zeolite Y.
The role of OHàas a strong mineralizing agent in zeolite synthesis is well known.Increasing OHàconcentration increases the solubility of the media that the zeolite nucleates from at reaction tem-peratures.Some of this media is likely dissolved before getting the chance to nucleate and grow into a stable crystal,and this leads to fewer nuclei and more nutrients in solution available for crystal growth.This explains why increased OHàcon-centrations yield larger crystals(Table2,Fig.7). Using lower OHàconcentrations should allow for the synthesis of smaller crystals,but only to a certain point,due to the fact that zeolite Y can only nucleate in a certain alkalinity range[37].The e?ect is similar to what is observed when identical zeolite Y syntheses are performed at di?erent tempera-tures[23].The higher temperature synthesis yields fewer large crystals in a shorter amount of time, while the lower temperature synthesis yields more small crystals after a longer period of time.The di?erent crystallization behaviors are due to the alumosilicate precursor media being more soluble at higher temperatures.
Zeolite Y nanocrystal synthesis has been shown very recently to have overlapping nucleation and crystallization periods by Li et al.[23].Though the data in the present study were not collected with this e?ect in mind,Fig.7clearly shows that no particle growth was observed for batches S0.67 and S0.88with low OHàcharging concentration and high TMABr/TMAOH ratios during the nu-cleation period.Rapid crystallization to the?nal crystal size was also observed.These observations indicate that we may have separated the nucleation and crystallization steps for these batches,similar to that observed for silicalite-1by Li et al.
[23].
3.4.Structural formula of zeolite Y nanocrystals The structural formulae of nanocrystal samples were determined using ICP OES analyses on Si,Al and Na.The assumption was made that Al was counterbalanced by either Na or TMA cations.The resulting Al 2O 3,SiO 2,Na 2O and TMA 2O contents and unit cell composition for zeolite Y nanocrystals from batches T2.4and T3.6are shown in Table 3.The Si/Al ratio of batch T3.6was further veri?ed using 29Si MAS NMR.The 29
