Documente Academic
Documente Profesional
Documente Cultură
Abstract--A saponite pillared with a single (AlzO3) or a mixed (SiO~-TiO:) oxide exhibited basal spacings
of 16-19 and 30-40/~, respectively. The pillared structures were found to be stable up to 700~ Water,
nitrogen, and high resolution argon adsorption were used to study the effect of thermal treatments on
surface chemistry, pore structure, and surface area of these pillared clays. The pillared saponites exhibited
a hydrophobic behavior at temperatures > 500~ whereas such behavior was observed at >300~ for
montmorillonite. Most of the micropores in the A1203 pillared clays were < 10 A, whereas the SiO2-TiO2
pillared clays showed a broad distribution of pores in both micropore and mesopore regions. The SiO2-
TiO2 pillared samples possessed higher surface area compared with A1203 pillared clays. The percent
decrease in surface area was smaller for pillared saponites compared with pillared montmorillonites when
calcined from 300 ~ to 700~ indicating a higher thermal stability of the former. The pillared clays were
also characterized by solid state 27A1and 295i magic-angle spinning nuclear magnetic resonance (MAS/
NMR) spectroscopy. There was no direct evidence of cross-linking (covalent bonding between the clay
layer and pillar) in montmorillonite irrespective of the types of pillars. In saponite, however, a significant
structural modification took place. 27A1 spectra of A1203 pillared saponite heated at >300~ appear to
indicate an increase in A1vl as a result, at least in part, of initiation of hydrolytic splitting of Si-O-AI
bonds. The actual release of A1 from the tetrahedral sheet probably occurred at a temperature > 500~
and completed around 7000C with the formation of Si-O-Si linkages. The decreased intensity of peak due
to Si(1A1) in 29Si spectra of the sample heated at 700~ corroborates the 27A1 MAS/NMR results. Ad-
ditionally, the 295i spectra indicated a cross-linking between SiO4 (clay sheet) with A1203 pillars, which
could be achieved by inverting some silica tetrahedra into the interlayer. 27A1and 295i spectra of SiOz-
TiO2 pillared saponite also showed the trend similar to that exhibited by A1203 pillared saponite, indicating
that the crystal chemistry of the host may be more important than the nature of pillars in the structural
modification and cross-linking behavior of thermally treated pillared clays.
Key Words--Adsorption, Clays, Cross-linking, MASNMR, Montmorillonite, Pillared clay, Saponite,
Smectite.
differences in the high-resolution N M R spectra of un- from Ballarat, California, was procured from the Source
calcined and calcined pillared beidellite were inter- Clay Repository of the Clay Minerals Society. Less than
preted as the consequence of inversion of some A104 2.0 gm sized particles were separated by centrifugation
tetrahedra (tetrahedral sheet), which cross-link with the (Jackson, 1969). The cation exchange capacity (CEC)
A1 of the pillars (Plee et al., 1985). The catalytic activ- was determined to be 79 meq/100 g. The N a - m o n t -
ities and selectivities of pillared beidellite were found morillonite was supplied by K u n i m i n e Industrial Co.,
to be higher than that of pillared montmorillonite due Japan. The cation exchange capacity (CEC) was deter-
to the formation of Si-OH acid sites in the former mined to be 113 meq/100 g. The chemical analyses
(Schutz et al., 1987). Besides catalytic activities, Malla and structural formulas of the raw materials and pil-
et al. (1989) and Malla and K o m a r n e n i (1990a, 1990b) lared clays are presented in Table 1.
observed a definitive variation in water sorption iso-
therms for pillared montmorillonite, nontronite, hec-
Pillaring agents and pillaring process
torite, and saponite. Hydroxy-Al pillaring solution and pillaring process. A
Although cross-linking was not observed in minerals 0.4 M AIC13-6H20 solution was titrated slowly with
having a negative charge in the octahedral sheet, Pin- 0.4 M N a O H under constant and vigorous stirring us-
navaia et al. (1985b) observed an inversion of Si tet- ing a magnetic stirrer. The OH/A1 molar ratio of 2 was
rahedra and cross-linking between the A1 in the pillars maintained. The resulting hydroxy-aluminum solution
and the tetrahedral Si in calcined pillared fluorohec- was aged at 60~ for 18 hr in a stoppered polypropylene
torite. This was attributed to the presence of F in the bottle. An excess of pillaring solution (>30 times the
octahedral sheet of this material. Because of the high CEC of the clay) was mixed with about 2-3 wt. % clay
thermal stability of pillars, such cross-linking behavior suspension and stirred at room temperature for 4 hr.
was also assumed in hydroxy-Si/A1 pillared fluorohec- A previous study has shown that a dilute clay-water
torite by Sterte and Shabtai (1987). Although cross- slurry is not necessary for pillaring and that the pillar-
linking behavior in pillared montmorillonite, hector- ing solution may be directly added to dry clay (Malla
ire, beidellite, and nontronite have been studied, sap- and Komarneni, 1990a). After reaction, each product
onite has not been studied in detail. The objective of was washed four times with deionized water and fil-
this study was, therefore, to characterize the adsorption tered. The washed and filtered samples were then dried
and cross-linking properties of A1203 and SiO2-TiO2 under a stream of dry air.
pillared saponite. Relevant properties of pillared mont-
Silica-titania (Si02- Ti02) pillaring solution-sol and pil-
morillonite are presented for comparison.
laring. Silica-titania mixed sol was prepared following
MATERIALS A N D M E T H O D S the method described by Yamanaka et al. (1988) with
m i n o r modification as described below. Typically for
Host materials 1 g of clay (CEC = 1 meq), silica sol was prepared from
Natural saponite and montmorillonite were used as 6.25 g tetraethylorthosilicate [ S i ( O C 2 H 5 ) 4 ] , 1.8 ml eth-
host materials in this study. The saponite (SapCa-1) anol, and 1.5 ml 2 M HC1. The resulting solution was
474 Malla and Komarneni Clays and Clay Minerals
stirred for 1 hr. For the same amount of clay, 0.85 g Magic-angle spinning nuclear magnetic resonance
titanium (IV) isopropoxide [Ti(OCH[CH3]2)4] was spectroscopy (MASNMR)
mixed with 3.4 g ethanol, resulting in a thick white 27A1 and 29Si M A S N M R spectra were recorded on
slurry. N o acid was a d d e d at this stage. The titania either a Brucker (Model HX-360) or a ChemMagnetics
slurry was stirred for 5 rnin and then added to the above (Model CMC 300A) instrument. The 27A1 chemical
silica sol. The resulting mixture was stirred further for shift values were assigned relative to the 1M aqueous
a period o f 1 hr, giving rise to a yellowish, clear sol. A1C13 reference solution, whereas the 29Si chemical shift
The mixed sol thus prepared was mixed with 1% clay values were assigned with respect to the external stan-
suspension and stirred at 50~ for 3 hr followed by dard tetramethylsilane (TMS).
washing five times with deionized water. The washed
sample was dried at 60~ RESULTS
Chemical analysis Chemical analysis
The samples were dissolved using the lithium meta- Chemical analyses o f the pillared and the original
borate fusion technique (Ingamells, 1970; Medlin et unpillared samples are presented in Table 1. The data
al., 1969). K and N a were determined by flame emis- illustrate the presence of higher amounts of A1203 or
sion spectrometry, while all other elements were de- SiO2 and TiO2 in the pillared clays and show that al-
termined by direct-current plasma emission spectros- most all the original charge balancing cations (Na
copy (DCP). Ca 2+ or K + ) have been replaced by the pillaring spe-
cies. The amount of the pillared species per O20(OH)4
X-ray powder diffraction analysis unit can be estimated to be 1.66Alp, 10.98Sip, and
A few drops of clay suspension were allowed to dry 2.4Tip in montmorillonite and 1.60Alp, 7.38Sip, and
on glass slides at room temperature. Thermal stability 1.72Tip in saponite. The Si/Ti ratios o f the pillared
o f the pillared clays was studied after heating the slide species were calculated to be 4.58 and 4.28 in mont-
mounts for at least 4 hr at 300 ~ 400 ~ 500 ~ 600 ~ and morillonite and saponite, respectively, compared with
700~ X R D analyses o f alumina pillared samples were the ratio of 10 for the original pillaring solution-sol.
performed with a Scintag diffractometer (4~ using
Ni-filtered CuKa radiation. The SiO2-TiO2 pillared X-ray powder diffraction analysis
samples were analyzed by Philips A P D 1700 diffrac- X R D patterns o f alumina and silica-titania pillared
tometer (2~ using graphite monochromated C u K a clays are presented in Figures 1 and 2. The pillars were
radiation. found to be stable up to 700~ The alumina pillared
saponite exhibited peaks of 19 ~ at r o o m temperature
Water and nitrogen adsorption and 16.9 ~ at 700~ indicating that pillar heights (or
Water adsorption and desorption isotherms were the slit widths) are 9.4 ~ and 7.3 ]k, respectively, after
measured by a volumetric method at 25~ using a com- subtracting the thickness o f the aluminosilicate layer
puter-interfaced sorption apparatus (Yamanaka and (9.6 ~). The d-values of A1203 pillared montmorillon-
K o m a m e n i , 1991). Samples were degassed at 200~ ites are also very similar to those of saponite, except
after calcining at 400 ~ for 20 hr or 7000C for 4 hr. that the montmorillonite heated at 700~ gave a 16.0
Nitrogen adsorption and desorption isotherms were ~, spacing. The SiO2-TiO2 pillared clays exhibited much
measured using the Quantachrorne Autosorb-1 at liq- higher d(001)-values (Figure 2) compared with A1203
uid N2 temperature after degassing at 200~ The sam- pillared samples. Typically, the d-values were 39/~, at
ples were precalcined at 300 ~ 400 ~ and 500 ~ for 20 hr r o o m temperature but decreased to 32 /~ at 500~
and 700~ for 4 hr. Surface areas o f A1203 pillared These d-values are much higher than the values (17-
samples were estimated using both the BET and the 19/~) obtained by Occelli (1987) and Sterte and Shabtai
Langmuir fit. With the BET equation, relative pres- (1987) for SIO2-A1203 pillared clays. The d(002) and
sures <0.1 were used to achieve a positive slope. A d(003) in these samples are not exactly multiple inte-
relative pressure range o f 0.05 to 0.30 was used with grals o f d(001), indicating that the layer expansion is
the Langmuir equation. Surface areas o f SiO2-TiO2 pil- apparently not uniform.
lared clays were also estimated from the BET fit in the
relative pressure range o f less than 0.1. Pore size dis- Adsorption properties
tributions were calculated from adsorption isotherms Adsorption-desorption isotherms. Water adsorption
using the BJH method (Barrett et al., 1951). Micropore isotherms are presented in Figure 3. The alumina pil-
analysis of the SiO2-TiO2 pillared clays was performed lared saponite calcined at 400~ exhibited an enhanced
at liquid A r temperature using Ar as the probe molecule adsorption at low relative pressure, whereas the ad-
(ASAP 2000M, Micromeritics Instrument Corpora- sorption at low relative pressure was depressed in the
tion, One Micromeritics Drive, Norcross, Georgia). sample calcined at 700~ The A1203 pillared mont-
Micropore size distribution was calculated using the morillonite calcined at 400~ showed a water adsorp-
method developed by H o r v a t h and Kawazoe (1983). tion isotherm that did not fit either the BET or the
Vol. 41, No. 4, 1993 Properties of pillared saponite 475
o. RT o ~
~. ~ -
/ \ _~.~\ o. RT
%6
"~ k b. 5oo*c
~, I , I , I , I i I , I
_c tl 5.0 I0.0 15.0 20.0 25.0 :30.0
" 17.81 b. 500"C
' c. RT
d. 4 0 0 %
e. 500"C
li 1619A c. 700 *C
i I A I m I ~ 1 ~ I
0.00 6.00 12.00 18.00 24.00 50.00
II , , A. , , ..L~...... , ~ .... Degrees 2e, CuKa radiation
Figure 2. X-ray powder diffraction patterns of SiO2-TiO2
4 14 24 34 pillared smectites as a function of calcination temperature: a
Degrees 20, CuKa radiation and b = saponite; c, d and e = montmorillonite; RT = room
temperature.
Figure 1. X-ray powder diffraction patterns of A1203(APS)
as a function of calcinationtemperature: a) room temperature
(RT), b) 500~ and c) 700~ large enough to show capillary condensation. The
montmorillonite also exhibited a similar isotherm, ex-
cept that the part of the isotherm near saturation was
Langmuir equation and that has been described earlier
relatively flat compared with the saponite. As in the
(Malla et al., 1989) as an "unusual" isotherm shape
A1203 pillared clays, the desorption branch did not
(data not shown). The unusual shape has been attrib-
meet the adsorption branch even at low relative pres-
uted to development of hydrophobic sites after calci-
sure, indicating a chemical reaction of H 2 0 molecules
nation (Yamanaka et al., 1990; Malla et al., 1989).
(hydroxylation) with pore walls. The N2 adsorption
Unlike water isotherms, N2 adsorption in both A1203
isotherms of these clays are slightly different than those
pillared montmorillonite and saponite showed type I
ofA1203 pillared clays, that is, they exhibited isotherms
isotherms (Langmuir type) typical of microporous sol-
transitional between type I and type IV (Figure 4a).
ids (Figure 4a). Both H20 and N2 isotherms exhibited
hysteresis; and in the case of H20, the desorption branch Surface area, pore volume and pore size. Nitrogen sur-
did not meet the adsorption branch, even at a very low face areas, pore volumes and pore sizes of the pillared
relative pressure, indicating an irreversible adsorption clays are summarized in Table 2. Both alumina and
of water (possibly chemical reaction) on the surface of silica-titania piUared clays exhibited large surface area
pillars and aluminosilicate layers (pore walls). The hys- and pore volume up to 700~ although they decreased
teresis loops in N2 isotherms are of type H4 and in- with increasing temperature. The pillared montmoril-
dicate the presence of narrow slit-like micropores (Sing, lonites showed higher values compared to their sapo-
1985). nite counterparts. The extent of decrease in surface area
The SiOe-TiO2 pillared saponite exhibited a type IV (Table 2) and micropore volume (data not shown) with
isotherm (see Gregg and Sing, 1982) with H20 (Figure increasing c a l c i n a t i o n t e m p e r a t u r e was, however,
3), which indicates that the pores in these samples are smaller for saponite than for montmorillonite. Addi-
476 Malla and Komameni Clays and Clay Minerals
~" 0.430 r ~ , i .i I
o
"
=
0
E
0.344
O. 2 5 8
~
// 400 oc i
tionally, the SiO2-TiO 2 pillared clays showed smaller
decrease in surface area compared with alumina pil-
lared clays when heated from 300 ~to 700~ indicating
higher stability of the former pillars.
Pore size distributions in both saponite and mont-
m~176 pillared with A1203 (Figure 4b) are very
O. 17 2 similar, with most of the pores having a diameter <20
with maxima at < 10/k. The SiO2-TiO2 pillared clays
exhibited two maxima, one at < 10 /~ and another
0.0 86 around 20 ~. The latter was broad and continued into
the mesopore region before leveling off around 50/~.
The mesopores were minimized in A1203 pillared clays.
0 .2 .4 .6 .8
Magic-angle spinning (MAS) nuclear magnetic
Relotive pressure (P/Po) resonance (NMR) spectroscopy
Figure 3. Water adsorption (open) and desorption (closed 27A1 spectrum of the original Na-saponite is given
symbol) isotherms of the A1203pillared saponite, (D, II), APS in Figure 5e. The chemical shift at 64.9 ppm can be
at 400* and 700~ as well as the SiO2-TiO2pillared saponite
(/x, A), STPS at 400*(2. attributed to A1TMof the tetrahedral sheet. 27A1 spectra
of A1203 pillared saponite are presented in Figures 5a-
5d. The uncalcined sample (Figure 5a) exhibited the
spectrum similar to Na-saponite (Figure 5e), except
that a broad peak around - 4 ppm due to A1w of the
2
o I18
"~ 94
:g
235"
212 i
188
165
141
71
I 1 I
9
1 I I I I I /
~'(:3
-~
o~
p
3oo!
5.00 4
4.501
4.ool
3.501
2.50 -
d 2.001 -
-~ 1.50 -
1
b
I ' ' ' ' 1
47] T - ~ 1.001 -
24
OnJ
~o
_1
0.1 0.2
I I I
0.3 0.4. 0.5 0.6 0.7
I I 1 I
0.8
I
0.9 1.0
~ I0 2 5 I00 2
Relolive pressure (P/Po) Pore diameter (/~)
Figure 4. Nitrogen adsorption (open) and desorption (closed symbol) isotherms of the pillared smectites: In a) ((3, O), APM;
(n, I), APS; (0, . ) STPM; and (a, A) STPS. Abbreviations explained in Table 1. All samples were calcined at 400~ for
20 hr except the APS sample indicated by an arrow, which was calcined at 700~ for 4 hr. In b) pore size distribution of the
pillared smectites calcined at 4000C calculated on the basis of adsorption isotherms shown in a).
Vol. 41, No. 4, 1993 Properties of pillared saponite 477
Ld
(.D tD
O. e.
<5
I ' ' ' ' 1 ' "v' I ' ''- I . . . . I ' ' ' ~ I
150 IO0 50 0 -50 -I00 150 I00 50 0 -50 ppm
tD
b.
I ' r ' I ' '' ' I '' ' I ' ' ' ' I '-' ' ' ' 'l'''l'''l''' I'''l'''I'''l'''l''
150 IO0 50 0 -50 -t00 120 I 0 0 80 60 40 20 0 -20 ppm
u..)/
N/
ou
to ,.6 to
C. g.
l , f , , l , , , , l , , , , l , , , , l , , , , l ~ , t , , , , F~ , , , i , , , , i , , , , i , , , '
150 I00 50 O -50 -tO0 tSO 100 50 O - 5 0 ppm
ppm from [AI(H20) 613+
(M
to
d.
Figure 5. 27A1MASNMR spectra of a) uncalcined APS, and calcined APS at b) 300 ~ c) 500 ~ and d) 700~ respectively, e)
NaS, f) STPS (uncalcined), and g) STPS calcined at 500~ Abbreviations explained in Table 1.
478 Malla and Komarneni Clays and Clay Minerals
O~
rd
0'1
d
a
tD
__j
- 50;00
[ ' I
- I00.00
'
-t50.00
I '
ppm
/
'
-60
I ' '
/
'
-80
I ' '
h i l l
'
-I00
I ' '
-120
I ' I
140 p p m
~dd
I I
\
\
I ' I ' I ' I ' I ' I ' ' ~ I ' ~ ' I ' ' I ' ' ~ I
0.00 - 50.O0 - I O 0 . O 0 -150.00 - 2 0 0 . 0 0 ppm -60 - 80 -120 --140 p p m - I00
ppm from Tetromethylsilone ppm from Tetramethylsilone
Figure 7. 295i MASNMR spectra of a) NaM, b) STPM (uncalcined), c) STPM (400~ d) NaS, e) STPS (uncalcined), and
f) STPS (500~ Abbreviations explained in Table 1.
Table 3. (AI/Si)Iv ratio, 29Si chemical shift (-b, ppm) and relative contribution (%) of Si-nA1 components and full width at
half maximum (fwhm, ppm) of saponite.
(AI/Si)~v 4Si-0A1 4Si-1A1
~Sample 2Chem 3NMR ~ % fwhm ~ % fwhm
NaS 0.102 0.092 95.6 72.5 4.38 90.6 27.5 3.97
APSRT 0.102 0.111 95.5 66.8 3.63 90.8 33.2 4.33
APS300 0.102 0.137 96.6 58.1 5.36 91.4 40.4 6.88
(59) (41)
APS500 0.102 0.097 95.9 67.4 5.43 91.1 29.4 5.97
97.3 3.1 1.86
APS700 0.102 0.066 94.5 20.9 5.36 91.1 17.1 7.29
(24.2) (19.9)
97.4 48.2 4.45
(55.9)
STPSRT 0.102 0.109 95.4 24.9 4.81 90.5 12.11 4.33
(67.3) (32.7)
STPS500 0.102 0.130 96.3 25.3 5.28 91.0 16.1 7.29
(61.1) (38.9)
' From chemical analysis,
2 See Table 1 for identification of samples; RT = room temperature; numbers following the samples are calcination
temperature.
3 Obtained from the following equation (Plee et al., 1984): (A1/Si) Iv = E8 nl(Si-n A1)/3.
4 Values in parentheses are normalized to 100% for structural Si in Q3 environment.
i i xl i i i I i I
0.018 - /
c, 0.016 /
o< 0.014 - a _ i~'!i~i"%.-:-:.--:
.......
j z
u 0.012-
Q.I ~ k
J 0.010- k x =30,
E= 0 . 0 0 8 \
o 0.006 ~x,
~ 0.004
a. 0 . 0 0 2 -
_:
I I Xl l I I I I I
0"0000 2 4 6 8 I0 12 14 16 18 2 0
o
Pore diameter, A
Figure 8. a) Micropore size distribution (Horvath and Kawazoe method) of STPS (400"C), and b) structural model showing
arrangement of SiO2-TiO= particles in STPS and STPM (after Yamanaka et aL, 1988).
Vol. 41, No. 4, 1993 Properties of pillared saponite 48 l
positively charged TiO2 particles must have been in- the electric field gradient effects in saponite interlayers
tercalated as suggested by Yamanaka et al. (1988). are large compared to other clays and that part of A1vl
In the A1203 pillared clays, the X R D slit widths (7- of pillars escaped detection; and 4) no separate reso-
9 A) are approximately equal to the micropores (< 10 nances for A1TMof pillars and tetrahedral sheet because
~ ) estimated from the gas adsorption isotherms (Figure of the overlapping of both resonances at 63-65 ppm.
4; Michot and Pinnavaia, 1992). This is, however, The first two features are discussed in detail below.
not the case in SiO2-TiO2 pillared samples. For ex- The large increase in intensity of A1v~ resonance rel-
ample, X R D showed the slit widths to be about 24 ative to A1TMwith increased calcination temperature
after calcination at 400~ whereas the pore size dis- can be attributed, in addition to the possible effect of
tribution curves using the Kelvin equation indicated reduced quadrupole interaction after thermal treat-
that most of the pore diameters are to be < 20 A, or ment, to a removal of A1 from the tetrahedral sheets
even < 10 ~.. Since the Kelvin equation is not quite of saponite. In fact, such de-alumination has been de-
valid for pore diameters < 2 0 ~ , Ar adsorption over liberately carried out in NH4-exchanged zeolite Y at
the relative pressure range 10 -6 < P/Po < 0.99 was >400~ in the presence of steam in order to make the
performed and micropore size distribution was deter- zeolite ultrastable without significantly affecting its
mined using the Horvath-Kawazoe model (Horvath crystallinity (McDaniel and Maher, 1976). In alumina
and Kawazoe, 1983). In fact most of the pores were pillared clays, protons are released during the dehy-
found to be <20 ~ with a broad distribution (Figure droxylation of All3 pillars,
8a). Despite a high basal spacing, the preponderance
of micropores in the SiO2-TiO2 pillared samples can [Al1304(OH)24(HaO)t2] TM ~ 6.5 A1203 + 8.5 H 2 0
be visualized from the multilayer intercalation of SiO2- + 7H +
TiO2 sol particles and the formation of micropores in
interstices between intercalated particles and silicate and these protons are apparently responsible for hy-
layers (Figure 8b). The SiO2-TiOz pillared clays also drolytic splitting of SiOA1 bonds followed by the re-
consisted of a significant a m o u n t of mesopores (Figuremoval of part of the tetrahedral AI in saponite. The
4b) whose pore size distribution leveled off around 50 vacant sites created after dealumination can be thought
A. The mesopores probably formed either at the frayed to have been occupied by Si that has migrated from
edges or in between the crystallites. The presence of the part of the collapsed clay structure to form Si-O-
both micropores and mesopores with broad distribu- Si bonds at temperatures -> 500~ This thesis is also
tions may have contributed to the transitional type of supported by the fact that the water adsorption iso-
isotherm shape (Figure 4) of these pillared clays. A therm of saponite heated at 700~ was depressed at
similar isotherm shape was observed by Gregg and Sing low relative pressure and did not fit either the BET or
(1982) for a sample having two types of micropores: the Langmuir equation. This can be attributed to the
small (< 7 Zk) and large (< 18 ~). development of hydrophobic sites resulted from the
loss of protons and the formation of Si-O-Si bonds.
Cross-linking No apparent change in pore size was observed. The
Both 27A1and 298i M A S - N M R results in Al203- and samples heated at <--400~ however, did not exhibit
SiO2-TiO2 pillared montmorillonites indicated that no the unusual shape of isotherm due to presence of hy-
direct cross-linking between the pillars and the struc- drophilic sites, such as H + and/or OH groups. The
ture occurs at the calcination temperature studied. These sample heated at 500~ exhibited an isotherm shape
results are in agreement with the previous 27A1and 298i intermediate of samples heated at 400 ~ and 700~ in-
M A S N M R studies of alumina pillared clays having dicating that the hydrolytic splitting of the SiOA1 bonds
octahedral location of charge (montmorillonites and was completed somewhere in between 500 ~and 700~
hectorites) (Tennakoon et aL, 1986; Plee et al., 1985). The de-alumination process also renders zeolites hy-
Unlike montmorillonite, pillared saponite exhibited drophobic due to the formation of Si-O-Si bonds
m a n y differences before and after calcination in the (Chen, 1976). Although a similar mechanism of cleav-
27A1 and 298i M A S - N M R spectra, indicating consid- age of SiOA1 bonds by proton attack in pillared bei-
erable structural modifications. The striking features dellite has been proposed by Plee et al. (1985) and
of the A1203 pillared saponite are: 1) increased intensity Schutz et al. (1987), they did not observe the increase
of the peak due to A1w (2-6 ppm) in relation to A1TM of A1w in N M R spectra as has been observed in this
as the temperature of calcination (300 ~ to 700~ in- study for saponite. As in A1203 pillared saponite, 27A1
creased; 2) more negative shift and broadening of Si(0A1) spectrum of SiO2-TiO2 pillared saponite heated at
and Si(1A1) resonances in 298i as a function of increased 500~ in addition to A1TM,also exhibited a small but
calcination temperature; 3) lower than the stoichio- broad peak around 0 ppm (Figure 5g), indicating the
metric amount of A1 (63% of the total, as in Table 1) presence of A1v~. Such peak was not observed in the
due to A1vI in 27A1spectrum (Figure 5a), indicating that uncalcined sample (Figure 50. Since the only source
482 Malla and Komarneni Clays and Clay Minerals
Michot, L. J. and Pinnavaia, T. J. (1992) Improved syn- Sterte, J. and Shabtai, J. (1987) Cross-linked smectites. V.
thesis of alumina-pillared montmorillonite by surfactant Synthesis and properties ofhydroxy-siticoaluminum mont-
modification: Chem. Mater. 4, 1433-1437. morillonites and fluorohectorites: Clays & Clay Minerals
Mortland, M. M. and Raman, K. V. (1968) Surface acidity 35, 429-439.
of smectites in relation to hydration, exchangeable cation, Tennakoon, D. T. B., Jones, W., and Thomas, J. M. (1986)
and structure: Clays & Clay Minerals 16, 393-398. Structural aspects of metal-oxide-pillared sheet silicates: J.
Occelli, M. L. (1987) Surface and catalytic properties of Chem. Soc., Farad. Trans. 1 82, 3081-3095.
some pillared clays: in Proc. Int. Clay Conf., Denver, 1985, Tennakoon, D. T. B., Jones, W., Thomas, J. M., Ballantine,
L. G. Schultz, H. van Olphen, and F. A. Mumpton, eds., J. H., and PurneU, J. H. (1987) Characterization of clay
The Clay Minerals Society, Bloomington, Indiana, 319- and pillared clay catalysts: Solid State Ionics 24, 205-212.
323. Van Olphen, H. (1966) Collapse of potassium montmoril-
Pinnavaia, T. J., Tzou, M., Landau, S. D., and Raythatha, H. lonite clays upon heating--"Potassium fixation": Clays &
(1984) On the pillaring and delamination of smectite clay Clay Minerals 14, 393-405.
catalysts by polyoxo cations of aluminum: J. Mol. Catal. Vaughan, D. E. W. (1988) Pillared clays--A historical per-
27, 195-212. spective: Catal. Today 2, 187-198.
Pinnavaia, T. J., Tzou, M., and Landau, S- D. (1985a) New Weir, J. I. and White, J. L. (1951) Potassium fixation in
chromia pillared clay catalysts: J. Amer. Chem. Soc. 107, clay minerals as related to crystal structure: Soil Sci. 71, 1-
4783-4785. 14.
Pinnavaia, T. J., Landau, S. D., Tzou, M., and Johnson, I. Yamanaka, S. and Brindley, G.W. (1979) High surface area
D. (1985b) Layer cross-linking in pillared clays: J. Amer. solids obtained by reaction in montmorillonite with zir-
Chem. Soc. 107, 7222-7224. conyl chloride: Clays & Clay Minerals 27, 119-124.
Plee, D., Borg, F., Gatineau, L., and Fripiat, J. J. (1985) Yamanaka, S., Nishihara, T., and Hattori, M. (1987) Prep-
High resolution solid state 27A1 and 298i nuclear magnetic aration and properties of titania pillared clays: Materials
resonance study of pillared clays: J. Amer. Chem. Soc. 107, Chemistry Physics 17, 87-101.
2362-2369. Yamanaka, S., Nishihara, T., and Hattori, M. (1988) Ad-
Plee, D., Borg, F., Gatineau, L., and Fripiat, J. J. (1987) sorption and acidic properties of clays pillared with oxide
Pillaring processes of smectites with and without tetrahe- sols: in Microstructure and Properties of Catalysts, Proc.
dral substitution: Clays & Clay Minerals 35, 81-88. Materials Research Society, Boston, Massachusetts, 1987,
Robert, M. (1973) The experimental transformation ofmica 111, M. M. J. Treacy, J. M. Thomas, and J. M. White, eds.,
toward smectite. Relative importance of total charge and Materials Research Society, Pittsburgh, Pennsylvania, 283-
tetrahedral substitution: Clays & Clay Minerals 21, 167- 288.
174. Yamanaka, S., Malla, P. B., and Komarneni, S. (1990) Wa-
Schutz, A., Stone, W. E. E., Poncelet, G., and Fripiat, J. J. ter adsorption properties of alumina-pillared clays: J. Col-
(1987) Preparation and characterization ofbidimensional loid Interface Sci. 134, 51-58.
zeolitic and hydroxy-aluminum solutions: Clays & Clay Yamanaka, S. andKomarneni, S. (1991) Apparatus for mea-
Minerals 35, 251-261. suring liquid vapor adsorption and desorption character-
Sing, K. S.W. (1985) Reporting physisorption data for gas/ istics of a sample: U.S. Patent 5,058,442, October 22~ 1991.
solid systems: Pure AppL Chem. 57, 603-619. (Received 19 June 1992; accepted 7 June 1993; Ms. 2236)
Sterte, J. (1986) Synthesis and properties of titanium oxide
cross-linked montmorillonite: Clays & Clay Minerals 35,
658-664.