Microporous and Mesoporous Materials 79 (2005) 712

www.elsevier.com/locate/micromeso

Eect of the silica sources on the crystallinity

of nanosized ZSM-5 zeolite
a,*
Reda M. Mohamed , Hisham M. Aly b, Mohamed F. El-Shahat c, Ibrahim A. Ibrahim a

a
Central Metallurgical R&D Institute, Department of Nanostructured Material, CMRDI, P.O. Box 87, Helwan, Cairo 11421, Egypt
b
Faculty of Science, Benha Branch, Zagazig University, Benha 13518, Egypt
c
Faculty of Science, Ain Shams University, Cairo, Abbasia 11566, Egypt

Received 31 August 2004; received in revised form 20 October 2004; accepted 20 October 2004
Available online 8 December 2004

Abstract

Silica sources can inuence dierent aspects of ZSM-5 crystallization and it leads to change in the properties of the nal product.
The crystallinity of nanosized ZSM-5 zeolite from precursors mixtures containing dierent silica sources, e.g. tetraethylorthosilicate
(TEOS), colloidal silica (Ludox LS 30), sodium metasilicate and fumed silica have been studied. The produced samples are inves-
tigated using XRD, SEM, and carbon % in the samples after calcination in air and BET surface area. The product obtained by
TEOS, fumed silica and colloidal silica is ZSM-5 phase, but the product obtained by sodium metasilicate is the quartz phase.
The crystallinity, and % of the nanosized ZSM-5 zeolite increased in the following order: sodium metasilicate < colloidal sil-
ica < TEOS < fumed silica. The average crystal size increased in the following order: fumed silica < TEOS < colloidal sil-
ica < sodium metasilicate. The surface area increased in the following order: sodium metasilicate < colloidal
silica < TEOS < fumed silica. The template was almost completely removed from the ZSM-5 by calcination at 550
C for 8 h in air.
 2004 Elsevier Inc. All rights reserved.

Keywords: Nanosized-ZSM-5; Silica sources; Crystallinity

1. Introduction dissolution, which play an important role in zeolite for-

mation, are also inuenced by the nature of the silica
ZSM-5 catalysts are important from industrial and used. Besides the eects on the formation of a particular
academic point of view of their unique structure, ther- zeolite, the silica source can inuence the specic fea-
mal stability, acidity, shape selective property and appli- tures of the zeolite material [10]. The impurities intro-
cations in oil renery, petro-chemical industry and duced by the silica source in the starting system can
environmental catalysis [15]. The synthesis conditions, also aect the morphology [11] and chemical properties
such as temperature, time, sources of silica and alumina [12,13] of the zeolite. The synthesis of zeolite crystals
inuenced the physico-chemical and catalytic properties with equal particle radius requires homogeneous distri-
of zeolite catalysts [69]. Some of important parameters bution of the viable nuclei in the system. Therefore,
describing the process of zeolite crystallization like the the homogeneity of the starting system and simultaneity
nucleation and crystallization rates depend on the disso- of the events leading to the formation of precursor gel
lution of the silica precursors. The fragile silicate inter- particles and their transformation into crystalline zeo-
mediates released during the process of the silica litic material are of primary importance. In order to ob-
tain such homogeneous starting systems abundant
*
Corresponding author. Tel.: +20 2 5010643; +20 2 5010639. amounts of tetraalkylammonium hydroxides and water
E-mail address: redama123@yahoo.com (R.M. Mohamed). were employed [1418]. On the other hand, the content

1387-1811/$ - see front matter  2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.micromeso.2004.10.031
8 R.M. Mohamed et al. / Microporous and Mesoporous Materials 79 (2005) 712

of alkaline cations is very limited. All these factors to- a rate of 100 cm3/min; 5.039 mg from the synthesized
gether with careful choice of the reactants allow the sta- zeolite was put in platinum cell. Surface areas were re-
bilization of clear starting mixtures where only corded using Nova 2000 series, Chromatech. The FT-
discrete gel particles are present [1922]. Most of the IR spectra were recorded using Jasco FT-IR-460 plus,
investigations were devoted to the synthesis of nano- Japan; % of carbon content in the sample after removal
sized zeolites with tetraethylorthosilicate as a silica of the template was recorded by using Strohlein. The
source. A limited number of articles discuss the forma- zeolite samples were also characterized by scanning elec-
tion of nanosized zeolites by using other silica sources tron microscopy using a Jeol scanning microscope mod-
[2329]. The aim of the present study was to investigate el JSM5410. The amount of silica and alumina in the
the eect of silica sources on the crystallinity of nano- crystalline samples were obtained by EDAX technique.
sized ZSM-5.

3. Results and discussion

2. Experimental
Changing silica sources as fumed silica, TEOS, colloi-
2.1. Materials and procedure dal silica and sodium metasilicate under the following
conditions has carried out a series of experiments:
The silica sources employed were tetraethylorthosili-
cate (TEOS, 98%, Fluka), colloidal silica, sodium alumina source: sodium aluminate,
metasilicate and fumed silica (Aldrich). The alkaline silica/alumina mole ratio: 40,
source was sodium hydroxide pellets (A.R 98%) and template oxide/silica mole ratio: 0.215,
template source employed was tetrapropylammonium type of template: tetrapropylammonium hydroxide,
hydroxide (TPAOH, 20% Fluka). The produced hydro- H2O/SiO2 mole ratio: 9.5,
gel was prepared by mixing the reagents in the following Na2O/SiO2 mole ratio: 0.133,
order: 0.212 g of NaOH was dissolved in 19.6 g of dis- CH3OH/SiO2 mole ratio: 22.
tilled water, 10 ml of 20% tetrapropylammonium
hydroxide was added drop wise to the sodium hydroxide The samples were crystallized at 230
C for 45 h, then
solution with stirring, the mixture was stirred for 30 min; washed and centrifuged until the pH of the liquid
then 13.75 g of silica was added drop wise, the resulting reached 9, then the solid material was dried overnight
mixture (I) was stirred for 30 min. 0.212 g of NaOH was at 110
C. The obtained samples were characterized
dissolved in the 19.6 g of distilled water, 1.1 g of sodium using X-ray, FT-IR, surface area, SEM, % of carbon,
aluminate and 196 ml of methanol, as structure directing DTA and TGA.
agent, were added to the sodium hydroxide solution
with stirring for 1 h; the resulting mixture (II) was added 3.1. X-ray diraction and surface area
dropwise to mixture (I) with stirring and the resulting
gel was stirred for 1 h. Finally, the creamy gel was put Fig. 1 shows the XRD patterns of samples that were
into 500 ml parr autoclave at 230
C for 45 h, and then crystallized by using dierent sources from silica as
the autoclave was quenched immediately with cold TEOS, fumed silica, colloidal silica and sodium metasi-
water. The pH of slurry was about 11, the solid product licate. The patterns of the product synthesized by TEOS,
was ltered and washed with distilled water until the pH fumed silica and colloidal silica indicate that ZSM-5
is dropped to 9, and the product was dried at 110
C phase existed, but the pattern of the product synthesized
overnight. The unlined autoclaves were cleaned with by sodium metasilicate shows that quartz silica existed.
10% NaOH at 160
C prior to each experiment to avoid The pattern of the product synthesized by fumed silica
adventitious seeding. indicates that it is the highest in crystallinity. This is
attributed to when a monomeric silica source is used,
2.2. Characterization nucleation and crystal growth occur more rapidly than
when a highly condensed source of silica is used. Table
X-ray diraction (XRD) patterns and average crystal 1 shows the eect of silica sources on structural param-
size were collected with Bruker axs, D8 Advance. The eters of the obtained samples together with those of
crystallinity was determined from the peak area between standard sample. The results show that the structural
2h = 2225
using a highly crystalline ZSM-5 sample parameters of as-synthesized ZSM-5 samples, which
(ZSM-5 820NAA supplied by Mobil company Japan) were prepared by using fumed silica, colloidal silica
as reference. Thermal analyses were carried out by using and TEOS as a source of silica are near those of stan-
Shimadzu DTA-50 and Shimadzu TGA-50H. The tem- dard samples. But the structural parameters of the sam-
perature of the sample was raised at a uniform rate of ple, which is prepared by using sodium metasilicate as
10
C/min from ambient to 1000
C in an air owing at silica source, are dierent from those of standard sample
 R.M. Mohamed et al. / Microporous and Mesoporous Materials 79 (2005) 712 9

(113.6) < sodium metasilicate (135.5). The surface area

a
was decreased in the following order: fumed sil-
ica > TEOS > colloidal silica > sodium metasilicate due
to decreasing crystallinity [30].

3.2. FT-IR

Fig. 2 shows IR spectra of as-synthesized samples ob-

b tained by using dierent silica sources. The results show
Intensity

that the bands near (1219 and 542 cm 1 characteristic

of the presence of double ve member ring) [31],
(1080 cm 1 characteristic of internal asymmetric
c stretching vibration of SiOT linkage), (790 cm 1
characteristic of the symmetric stretching of the exter-
nal linkage) and (450 cm 1 characteristic of TO
bending vibration of the SiO4 and AlO4 internal tetrahe-
dral) are present for the samples, which were prepared
by using TEOS, fumed silica and colloidal silica as silica
d
sources, but bands around 542 and 450 cm 1 character-
istic of the ZSM-5 crystalline structure [32] are not
0 10 20 30 40 50 60 70 80
present for the sample which was prepared by using so-
2- Theta
dium metasilicate due to formation of quartz silica. Ta-
ble 3 shows the eect of silica sources on the ratio of the
Fig. 1. Eect of silica sources on crystallinity of ZSM-5 formed where intensities of the peaks near (A) 542 and (B) 450 cm 1,
a = fumed silica, b = TEOS, c = colloidal silica and d = sodium
which provides an approximate estimate of the degree
metasilicate.
of crystallinity [33]. The results show that the ratio

Table 1
Eect of silica sources on structural parameters of as-synthesized
samples
Silica sources Structural parameters (A) Unit cell
V (A)3
a b c
Standard ZSM-5 20.02 19.89 13.38 5332.0249
Fumed silica 19.92 19.92 13.36 5301.3330
TEOS 19.92 19.92 13.36 5301.3330
Colloidal silica 20.16 19.90 13.44 5391.9120
Sodium metasilicate 4.24 3.01 5.43 69.2990

due to formation of quartz silica. Table 2 shows the ef-

fect of silica sources on surface area, crystallinity, % and
average crystal size of as-synthesized ZSM-5. The aver-
age crystal size was increased in the following order:
fume silica (55.8) < TEOS (90.3) < colloidal silica

Table 2
Eect of silica sources on crystallinity, surfaces area and average
crystal size of as-synthesized ZSM-5
Silica Crystallinity Surface Average Phase
sources of ZSM-5, % area, crystal formed
m2/g size, nm
Fumed silica 97.3 358.21 55.8 ZSM-5
TEOS 77.1 351.16 90.3 ZSM-5
Colloidal silica 46.7 285.00 113.6 ZSM-5 Fig. 2. IR spectra of as-synthesized ZSM-5 zeolite samples obtained
Sodium metasilicate 9.17 135.7 Quartz by using dierent silica sources where a = colloidal silica, b = TEOS,
silica c = fumed silica and d = sodium metasilicate.
10 R.M. Mohamed et al. / Microporous and Mesoporous Materials 79 (2005) 712

Table 3 similar and hence the samples are homogeneous. From

Eect of silica sources on the ratio of the intensities of the peaks near the crystallinity view, the optimum conditions for silica
(A) 542 and (B) 450 cm 1 and also correlation between XRD
crystallinity and relative crystallinity estimated from IR
source are fumed silica at 55.8 nm crystal size, 358.21 m2/
g surface area and 97.3% crystallinity.
Silica sources A/B ratio Relative XRD
crystallinity crystallinity, %
estimated 3.4. Thermal analysis
from IR, %
Fumed silica 0.967 96.7 97.3 The thermal analysis was carried out to determine the
TEOS 0.730 73.0 77.1 thermal stability of the optimum ZSM-5 zeolite and to
Colloidal silica 0.450 45.0 46.7
measure water and template loss. Fig. 4 shows the TG/
Sodium metasilicate
DTA curves for the optimum ZSM-5. The observed exo-
therms (444.26
C) are due to the oxidative decomposi-
and consequently relative crystallinity increased in the tion of the template. Endotherm (58.72
C) represents
following order: colloidal silica > TEOS > fumed silica water loss. Total weight losses in the temperature range
that is in close agreement with that calculated from from 30 to 600
C represent water loss as well as loss of
XRD. template are about 11.5%. The TG/DTA curves for the
optimum ZSM-5 revealed that the temperature of the
3.3. SEM framework decomposition of zeolite is above 1000
C,
which is in agreement with the published literature
Fig. 3 shows SEM data for the prepared samples by [33]. These data suggest that although ZSM-5 crystalli-
using dierent silica sources as fumed silica, TEOS, col- zation methods are dierent their framework rigidity
loidal silica and sodium metasilicate. The results show or structural stability is almost the same.
that the sample which was prepared by fumed silica
crystallized in cubical shape crystals, the samples which 3.5. The eect of calcination time on the % of carbon
were prepared by TEOS and colloidal silica crystallized remaining
in twinned and intergrown shape crystals, but the sam-
ple which was prepared by sodium metasilicate crystal- To study eects of calcination time on the % of car-
lized in needle shape crystals. The chemical bon remaining, the optimum as-synthesized ZSM-5 zeo-
composition of the prepared samples was measured by lite was calcined in air at 550
C for dierent times (2
EDX techniques summarized in Table 4. The results 8 h) and the percentage of carbon remaining in the sam-
show that the Si/Al atomic % at three areas is almost ples was measured. The results show that the carbon

Fig. 3. SEM images of sample prepared by using dierent silica sources, where in a = fumed silica, b = colloidal silica, c = sodium metasilicate and
d = TEOS.
 R.M. Mohamed et al. / Microporous and Mesoporous Materials 79 (2005) 712 11

Table 4
Si/Al atomic % ratio of the produced samples measured by EDX
Technique at three areas
Silica sources Si/Al atomic % Average Si/Al
atomic %
A1 A2 A3
Fumed silica 27.8 28.78 27.9 28.16
TEOS 23.46 22.33 23.70 23.16
Colloidal silica 21.31 20.08 19.98 20.46
Sodium metasilicate 46.27a 48.93a 37.77a 44.32a
a
Only Si is present and no Al present.

6
0

-2
4
-4
DTA ( uV)

TG ( Wt. loss
-6

2
-8

-10

0 TG
-12
DTA
-14

-2 -16
0 200 400 600 800 1000 1200
Fig. 5. FT-IR spectra of optimum ZSM-5 in its dierent forms: (a) as-
synthesized (ZUC), (b) calcined zeolite (ZC), (c) ion-exchange zeolite
Temperature, C
(ZCI) and (d) calcined after ion-exchanging (ZCIC).
Fig. 4. DTA/TG curves for as-synthesized fully crystalline ZSM-5
zeolite. eect of removal of template on structural parameters.
The results show that the unit cell parameters of the
ZSM-5 sample are greater than that of ZSM-5-tetrapro-
Table 5 pyl sample. This is attributed to removal of template
The eect of calcination time on the % of carbon remaining from ZSM-5 pores; this is conrmed by increasing sur-
Calcination time, h % of carbon remaining face area of ZSM-5 sample from 358.21 to 381 m2/g.
0 3.02 Fig. 5 shows the XRD patterns of ZSM-5-tetrapropyl
2 1.25 sample and ZSM-5. The results demonstrate that the
4 0.8 crystallinity of ZSM-5-tetrapropyl sample is almost sim-
6 0.75 ilar to that of ZSM-5, so that the removal of template
8 0.26
has no signicant eect on the crystallinity.

remaining in the sample was decreased from 3.02% to 3.6. The formation of H-ZSM-5 form
0.26% by increasing calcination time from 2 to 8 h,
respectively, as shown in Table 5. Hence, the template The optimum ZSM-5 sample (ZUC) was calcined at
was almost completely removed from the ZSM-5 by cal- 550
C in air for 8 h. After the rst calcination, the zeo-
cination of the sample, which was prepared by using lite sample was noted as calcined (ZC). This sample was
fumed silica at 550
C for 8 h in air. Table 6 shows the ion-exchanged to replace sodium ion for hydrogen ion
present on the active sites. A 1.0 M aqueous solution
of ammonium nitrate was used at 80
C for carrying
Table 6 out ion-exchange, the zeolite to solution ratio was 1 g
Eect of removal of template on structural parameters to 150 cm3 and the stirring was carried out for 1 h. After
ZSM-5 Structural parameters Unit cell ion-exchange, the zeolite was ltered, washed and dried
(A) V (A)3 at 120
C overnight. After ion-exchange, the zeolite was
a b c noted as (ZCI). The ZCI was calcined at 550
C for 3 h in
Standard ZSM-5-tetrapropyl 20.022 19.899 13.383 5332.0249 airow to decompose the ammonium ion to produce
ZSM-5-tetrapropyl sample 19.920 19.920 13.360 5301.3330 hydrogen form (ZCIC). Fig. 6 shows the infrared spec-
Standard ZSM-5 19.870 20.107 13.369 5343.6836 tra of zeolites (ZUC, ZC, ZCI and ZCIC). These spectra
ZSM-5 sample 19.910 20.030 13.400 5343.8800
show the appearance and disappearance of dierent
12 R.M. Mohamed et al. / Microporous and Mesoporous Materials 79 (2005) 712

References

[1] C. Naccache, Y.B. Tarrit, in: F.R. Ribeiro, A.E. Rodrigues, L.D.
Rollman, C. Naccache (Eds.), Zeolite Science and Technology,
ZSM-5 tetrapropyl The Hague, 1984, p. 373.
[2] T. Inui, G. Takeuchi, Y. Takegnami, Appl. Catal. 4 (1982) 201.
[3] E.G. Derouane, J.P. Gilson, Z. Gabelica, Appl. Catal. 1 (1981)
201.
Intensity

[4] N. Kumar, L.E. Lindfors, R. Byggningsbacka, Appl. Catal. A:

Gen. 139 (1996) 189.
[5] R. Byggningsbacka, N. Kumar, L.E. Lindfors, J. Catal. 178
(1998) 611.
[6] J.P. Feijen, J.A. Martens, P.A. Jacobs, Stud. Surf. Sci. Catal. 84
ZSM-5
(1994) 3.
[7] K.J. Chao, T.C. Tasi, M.S. Chen, J. Chem. Soc. Faraday 77
(1981) 465.
0 10 20 30 40 50 60 70 80 [8] R.M. Barrer, Hydrothermal Chemistry of Zeolites, Academic
2- Theta Press, London, 1982.
[9] R.M. Barrer, Zeolites 1 (1981) 130.
Fig. 6. XRD patterns of ZSM-5 tetrapropyl and ZSM-5. [10] S. Mintova, V. Voltchev, Micropor. Mesopor. Mater. 55 (2002)
171.
[11] J.C. Jansen, C.W.R. Engelen, H. van Bekkum, in: M.L. Occelli,
absorption bands during the pretreatment procedure. H.E. Robson (Eds.), Zeolite Syntheses, ASC Symposium Series,
The spectra of as-synthesized zeolite (ZUC) show weak vol. 398, Washington, DC, 1988, p. 257.
[12] C.T.W. Chu, G.H. Kuehl, R.M. Lago, D.D. Chang, J. Catal. 93
CH stretching absorption bands in the region 2980
(1999) 1639.
2880 cm 1 due to presence of the template together with [13] P.-P.E.A. de Moor, T.P.M. Beelen, R.A. van Santen, J. Phys.
two broad bands at 3670 and 3460 cm 1 assigned to tet- Chem. B 103 (1999) 1639.
raalkylammonium OH stretching modes. Calcination [14] A.E. Persson, B.J. Schoeman, J. Sterte, J.-E. Otterstedt, Zeolites
results in disappearance of absorption bands assigned 14 (1994) 557.
[15] C.E.A. Kirschhock, R. Ravishankar, L. van Loovern, P.A.
to the template. Following ion-exchange with aqueous
Jacobs, J.A. Martens, J. Phys. Chem. B 103 (1999) 4972.
solution of ammonium nitrate, new bands appeared at [16] O. Regev, Y. Cohen, E. Kehat, Y. Talmon, Zeolites 14 (1994)
3170 and 1400 cm 1 due to NH stretching and bending 314.
modes. When the ammoniated zeolite was calcined to [17] B.J. Schoeman, Micropor. Mesopor. Mater. 22 (1998) 9.
produce the hydrogen form of the zeolite, the bands at [18] S.L. Burkett, M.E. Davis, Chem. Mater. 7 (1995) 920.
3170 and 1400 cm 1 disappeared and the intensities of [19] Q. Li, D. Creaser, J. Sterte, Micropor. Mesopor. Mater. 31 (1999)
141.
OH stretching bands increased [34]. [20] W.H. Dokter, H.F. Garderen, T.P.M. Beelen, R.A. Santen, W.
Bras, Angew. Chem. Int. Ed. Engl. 34 (1995) 73.
[21] M. Tsapatsis, M. Lovallo, M.E. Davis, Micropor. Mater. 5 (1996)
4. Conclusions 381.
[22] B.J. Schoeman, E. Babouchkina, S. Mintova, V. Valtchev, J.
Sterte, J. Por. Mater. 8 (2001) 3201.
The silica sources inuenced on zeolite crystallization [23] S. Mintova, N.H. Olson, V. Valtchev, T. Bein, Science 283 (1999)
and lead to changes in the properties of nal product. 958.
Fumed silica, TEOS and colloidal silica give ZSM-5 [24] S. Mintova, N.H. Olson, V. Valtchev, T. Bein, Angew. Chem.
phase, but sodium metasilicate gives quartz phase. The Int. Ed. 38 (1999) 3201.
[25] Q. Li, B. Mihailova, D. Creaser, J. Sterte, Micropor. Mesopor.
crystallinity % of ZSM-5 produced increased in the fol-
Mater. 43 (2001) 51.
lowing order: colloidal silica < TEOS < fumed silica. [26] B.O. Hincapie, L.J. Garces, Q. Zhang, A. Sacco, S.L. Suib,
SEM data show that the sample which was prepared Micropor. Mesopor. Mater. 67 (2004) 19.
by fumed silica crystallized in cubical shape crystals, [27] S. Ferchiche, M. Valcheva-Traykova, D.E. Vaughan, J. War-
the samples which were prepared by TEOS and colloidal zywoda, A. Sacco, J. Cryst. Growth 222 (2001) 801.
[28] I. Guray, J. Warzwoda, N. Bac, A. Sacco, Micropor. Mesopor.
silica crystallized in twinned and intergrown shape crys-
Mater. 31 (1999) 241.
tals, but the sample which was prepared by sodium [29] K.E. Hamilton, E.N. Coker, A. Sacco, A.G. Dixon, R.W.
metasilicate crystallized in needle shape crystals. The Thompson, Zeolites 13 (1993) 645.
synthesized zeolite was found to be thermally stable up [30] W.J. Kim, M.C. Lee, D.T. Hayhurst, Micropor. Mesopor. Mater.
to 1000
C. Calcination of zeolites to decompose and re- 26 (1998) 133.
[31] P.A. Jacobs, H.K. Beyer, J. Valyon, Zeolites 1 (1981) 161.
move templates eectively was achieved by heating the
[32] D. Trong, S. Kaliaguine, L. Benneviot, J. Catal. 157 (1995) 235.
zeolite in airow for 8 h at 550
C. Calcination of zeolite [33] O.G. Somani, A.L. Choudhari, B.S. Rao, S.P. Mirajkar, Mater.
has no signicant on crystallinity %. Unit cell and sur- Chem. Phys. 82 (2003) 538.
face area are increased by calcination and ion-exchange [34] M.A. Ali, B. Brisdon, W.J. Thomas, Appl. Catal. A: Gen. 252
zeolite material. (2003) 149.