close

Вход

Забыли?

вход по аккаунту

?

Microporous Silica Films.

код для вставкиСкачать
Microporous Silica Films
50
By Wilhelm E Maier,* Michael Wiedorn,
and Herbert 0. Schramm
LO
30
Thin ceramic films are useful for optical or protective
coatings. Microporous ceramic membranes with pore diameters from 5-10 A are of interest for separations in the
food industry, gas separation, nonlinear optics, and for microelectronic sensors. The present activity in the field of microporous ceramic films focuses on zeolites, due to its welldefined pore structure. However, because of the limited
crystallite size of zeolites, the preparation of homogeneous,
pinhole-free films is still very difficult. We now report the
pore size distribution of new, microporous, homogeneous
silica films, which can be prepared in any thickness by electron beam evaporation. Such films may be used for the preparation of microporous ceramic membranes.
Our interest in the study of the microporosity of thin silica
films originates in heterogeneous catalysis. In studies on the
mechanism of hydrogenation and H/D-exchange reactions
on hydrocarbons we have used thin silica films (50-200 nm)
that covered the surface of the catalyst"' to shield the catalytically active metal film (Pt film on a Si wafer) from direct
interaction with the organic reagent. The silica layer is impervious to the organic reagent, while hydrogen apparently
diffuses through the oxide layer and is activated at the PtSiO, interface. The activated hydrogen diffuses as spill-over
hydrogen back to the surface (a steady-state concentration
gradient is set up), where it reacts with the organic reagent.
However, the mechanism of diffusion remained unclear,
since no cracks or holes in the silica film were evident. Because of the small amounts of film material it was not possible to check for micropores, which remained as a possible
explanation for our data. Various published discussions document the broad interest in our results.[*]
We have now determined the pore size distribution of the
silica films, prepared by electron-beam evaporation, through
nitrogen adsorption. Instead of Pt/Si wafers as basis for the
silica coatings, commercial aluminum foil was used as substrate for silica deposition, because the silica film could be
separated from the substrate mechanically, by immersion in
pure water, or by dissolving the aluminum foil in 2 N HCl. It
has been shown that the films are amorphous, and that a
silica film on platinum can differ from that on alumina only
in the first few atomic layers. Their structure should thus be
independent of the substrate, especially at thicknesses larger
than 50 nm. From the nitrogen adsorption isotherm the
specific surface and pore volume was determined according
to BET[31and to D~binin[~].
The pore radius distribution
(3-300 A) was determined from the adsorption data by applying the Kelvin equation[5]as well as by the micropore
method.[61
The N,-adsorption isotherm of the silica (Fig. 1) documents the microporosity of the material. Figure 2 shows the
associated pore radius distribution determined by the Kelvin
equation. Although the equation is not valid for pore radii
smaller 15 A,['] the analysis reveals a very narrow pore size
distribution with a maximum clearly below 15 A and no
pores of larger diameter. To analyze the micropores with
radii smaller than 10 A, we relied on the MP method, using
the t-plot method of de Boer.[81As a criterium for the quality
[*I
[**I
Prof. Dr. W. F. Maier, Dipl.-Chem. M. Wiedorn,
Dipl.-Chem. H. 0. Schramm
Institut fur Technische Chemie der Universitat-Gesamthochschule
Schiitzenbahn 70, W-4300 Essen 1 (FRG)
This study was supported by the Bundesministerium fur Forschung und
Technologie.
Angew. Chem. Inr. Ed. Engl. 30 f1991) No. 11
20
10
1
0.02
0
0.04
0.06
P/P,
-
0.1
0.08
0.12
Fig. 1 . Section of the nitrogen adsorption isotherm of silica, prepared by electron-beam evaporation. At pipo > 0.1 (not shown) there was no further N,
desorption. Adsorption volume V,,, in cm2g-'
3.0 T
I t
2.0
t
O
L c.
0
20
10
30
riAl
LO
50
Fig. 2. Pore radii distribution of the silica film from Fig. 1, calculated by applying the Kelvin equation. dVp/dr in mLA-'.
of the data produced by the MP method, the surface and
pore volume were compared with the data obtained by the
BET and Dubinin treatment. Isotherms for the t-plot calculated from the Halsey equationtg1did not result in reasonable
surfaces and pore volumes. Good results were obtained by
the use of finely ground quartz wool as nonporous reference
material. The t-plot16] shows a decrease in slope at a layer
thickness of about 3 A, which indicates a pore maximum at
3.1 8, (see Fig. 3). This procedure was tested on various
molecular sieves. We thus estimate the error of our analysis
to better than
1 A, which points to pore diameters of
for the microporous silica.
6 2
+
T
0.03.
I
0.02:
-d VP .
d r 0.01.
0
1
0
:
:
2
:
:
:
:
4
:
:
rIAl
:
-
6
8
I
10
Fig. 3. Pore radius distribution of the silica film of Fig. 1, calculated by applying the MP method.
To test whether the determined pore volume is due to inner
pore structure or to surface structure, the adsorption
isotherm was obtained from 10 films of different thickness
(between 250 and 4000 nm). The total surface of the material
varied between 155 and 259 m2g- ' and the pore volume
between 0.055 and 0.09cm3/g-', independent of the film
thickness. The maximum of the pore radius distribution was
the same in all cases. The measured pore volume is thus not
due to surface characteristics of the films, but only to inner
pore and channel structures.
0 VCH VerlugsgesellschuftmbH, W-6940 Weinheim. 1991
0570-0833~91/11ll-1509
$3.50+.25/0
1509
Table 1. BET analysis of nitrogen adsorption isotherms.
pore volume [cm3gsurface e [m'g-']
'1
SiO,
film
38,
zeolite
48,
58,
0.07
192
0.003
9.3
0.013
37.2
0.23
661.7
NaY
0.34
960
Because the size of the pores of our materials was clearly
at the limits of the methods used, we were interested in a
seperate method to estimate the actual pore size, in particular for pores too small to be determined by nitrogen adsorption. The adsorption isotherms of three molecular sieves and
a NaY zeolite were determined (Table 1). While the NaYzeolite and the 5 A molecular sieve adsorbed nitrogen readily, the 4 A and 3 A molecular sieve showed little N, adsorption, confirming that these pores are to small for nitrogen as
adsorbate (rotational diameter 4.3 A).["]
To test whether the silica films contain similarly small
pores we had to use an adsorbate smaller than nitrogen. One
of the few suitable molecules is water (rotational diameter
2.6 A). Because of technical problems with recording the
adsorption isotherm of water on our equipment, we have
used thermogravimetry (TGA). Two silica films and three
molecular sieves were saturated with water, dried for one
hour at 120°C (for the 4000 nm film, 80°C), and analyzed in
a SETARAM TGA 92. The temperature-dependent water
loss was registered (Table 2). The temperature of maximal
Table 2. Thermogravimetric study of water adsorption by zeolites and silica
films
Mass used [mg]
AM [mgl la1
AM [w%] [a]
T,,, ["C] [c]
38,
zeolite
4A
5A
thickness of SiO, film
30000A
40000A
85.0
11.7
13.8
211.9
86.7
11.9
13.7
206.0
85.6
11.9
13.9
201.9
41.7
54.9
2.0
3.9
4.7 (6.6) [b]
7.1
118.1
121.2
[a] AM = change in mass. [b] The value in parentheses includes the weight loss
during the predrying stage (see text). [c] T,,,,=temperature of maximal water
loss.
water loss increases with decreasing pore size of the molecular sieves. The relatively low temperatures of maximal water
loss for the silica films shows that they adsorb water less
strongly. In contrast to the pore volume from nitrogen adsorption, but as expected from the identical porosities of the
molecular sieves, the pore volume obtained from water loss
for the three molecular sieves is virtually identical. In the case
of the 5 A molecular sieve the pore volume obtained from
TGA is about 40% below the BET volume (0.23 mLg-'),
which corresponds to the known water adsorption capacity
of the molecular sieve. The difference can be traced to the
predrying procedure of the molecular sieves, because the
volume obtained from TGA for the more mildly predried
4000 nm film was identical with its BET pore volume. Therefore, all micropores in the silica films are larger than the
critical diameter of nitrogen.
From these studies we conclude that silica films prepared
and
by electron-beam evaporation have a porosity of 14 YO,
a very narrow micropore distribution comparable to that of
a 5 A molecular sieve. Silica films are therefore potential
materials for the preparation of microporous membranes
and surface coatings. The simple preparation procedure and
the homogeneity of the films make these materials very
promising alternatives to zeolite films.
1510
0 VCH
Verlugsgesellschafl mhH, W-6940 Weinheim, 1991
The extreme microporosity of the silica films supports the
interpretation of our previous catalytic studies on silica coated Pt/Si-catalysts.['] The narrow pore diameter of about 6 A
is readily penetrable by hydrogen, while diffusion of the organic reactant through the silica layer is highly hindered.
Also formation of a concentration gradient of spill-over
hydrogen as explanation for the observed catalytic phenomenon['] seems plausible.
Experimental procedure
The films were prepared in a Balzers (BAE 250) evaporation unit, equipped
with a 3 kW electron gun (Edwards, E036-15). The rate of evaporation was
2-20nmmin-' at a pressure smaller than 10-5torr. The silica film was
deposited onto aluminum foil, from which it was subsequently removed. After
washing and drying of the film, the nitrogen adsorption isotherm was recorded
on a modified Sorptomatic 1900 from Carlo Erba (smaller sample volumes, a
dosage unit for smaller dosage pressures ( p = 0-100 torr), and a pressure
gange for the range 0- 100 torr). The modifications allowed a reduction of the
sample size needed and the recording of more data in the micropore range. For
data treatment"" a makro. programmed for the spreadsheet Excel". was developed. Repeated measurements showed only small deviations in surface area and
pore volume.
Received: June 7, 1991 [Z 4685 IE]
German version: Angew. Chem. 103 (1991) 1523
CAS Registry number: SO,, 7631-86-9
[I] A. B. McEwen, R. H. Fleming, S . Baumann, W. F. Maier, Nature 329
(1987) 531; J. M. Cogen, K. Ezaz-Nikpay, R. H. Fleming, S. Baumann,
W. F. Maier. Angew. Chem. 99(1987) 1222;Angew. Chem. Int. Ed. Engl. 26
(1987) 1182; W. F, Maier, ihid. f01 (1989) 135 and 28 (1989) 135.
[2] K. Seshan, Appl. C u d . 50 (1989) N 14; S. J. Teichner, ihid. 51 (1989) N 18;
CHEMTECH 1988, 518.
[3] S. Brunauer, P. H. Emmett, E. Teller, J. Am. Chem. SOC.60 (1938) 309;
S . Brunauer, L. S. Deming, W.F. Deming, E. Teller, ihid. 62 (1940) 1723.
[4] M. M. Dubinin, J. Colloid Interface Sci. 23 (1967) 487.
[5] D. Dollimore, G. R. Heal, 1 Appl. Chern. I4 (1964) 109.
[6] R. S. Mikhail, S. Brunauer, E. E. Bodor, J. Colloid Interfuce Sci. 26 (1968)
45.
[7] J. Seifert, G. Emig, Chem. Ing. Tech. 59 (1987) 475.
[8] B. C. Lippens, J. H. de Boer, J. Curd. 4 (1965) 319.
[9] G. D. Halsey, J. Chem. Phys. 16 (1948) 931.
[lo] G. Horvath, K . Kawazoe, J. Chem. Eng. Jpn. 16 (1983) 470.
[Ill J. Schroder, GIT Fuchz. Lab. 30 (1986) 978; ihid. 30 (1986) 1095.
[I21 S. Lowell, J. E. Shields: Powder Surface AreaandPorosit,v. 3rd ed., Chapman and Hall, London 1991.
(131 Note added in proof (30.10.91): We thank Micrometrics for the determination of the micropore distribution of our SiO, film with their new ASAP
2000 M, carried out after submission of this manuscript. Data treatment
by the method of Horvdth and Kuwazoe [lo] gave a maximum for the pore
diameter distribution at 5.5 8,. confirming our data.
Crystal Structure of Chlorine Dioxide **
By Anette Rehr and Martin Jansen *
By virtue of their small number, stable radicals or radical
ions play a subordinate role in the chemistry of the nonmetals. In this setting, the characteristically increasing number of examples found in the case of the triatomic 19-electron
systems is striking. These can be homoatomic such as S;,[']
O;,[2] or P$-[31or heteroatomic such as CIO,, NFZt4]
and
SO; J51 with the more electronegative atom in the terminal
position. A gradual change in tendency to dimerize is ob[*I Prof. Dr. M. Jansen, DipLChem. A. Rehr
Institut fur Anorganische Chemie der Universitat
Gerhard-Domagk-Stra6e 1, 5300 Bonn 1 (FRG)
[**I This work was supported by the Deutsche Forschungsgemeinschaft (Leibniz Program) and the Fonds der Chemischen Industrie.
0570-0833~91jllll-i510~
3.50f.2510
Angew. Chem. Int. Ed. Engl. 30 (1991) No. 11
Документ
Категория
Без категории
Просмотров
2
Размер файла
255 Кб
Теги
microporous, films, silica
1/--страниц
Пожаловаться на содержимое документа