American Mineralogist. Volume 74, pages224-229. 1989
A hydrated potassium layer silicate and its crystalline silicic acid
K. BrNrxr' G. L,lcnr.v
Institut fiir anorganischeChemie der Universit.iit Kiel, Olshausenstrasse
40, D-2300 Kiel, F.R.G.
Ansrru.cr
A potassium silicate has been prepared from dispersionsof SiO, in aqueoussolutions
of KOH and from potassium water-glasssolutions at 60-100'C. This silicate is related
to magadiite, but differs from the potassium form of magadiite by reversibleinterlamellar
hydration between a state with a water monolayer (basal spacing l5.l A) and a bilayer
arrangement(basalspacing 17.3 A). Removal of the interlayer water decreasesthe spacing
to 12.8 A. In contrast to magadiite and its potassium analogue,the silicate intercalatesa
variety of organic guest molecules.
Cation exchangewith protons (leachingin acids)transformsthe silicate into a crystalline
silicic acid (basal spacing:about tS A in water, 13.4 A A;ea at room temperature, ll.6
A heated above 300 "C). This acid is characterizedby very pronounced intracrystalline
reactivity. It intercalatesa largenumber of organiccompounds including alkanols,nitriles,
ketones,and amino acids.
fxrnonucrroN
Synthesisofhydrated alkali silicatesis governedby the
nature ofthe alkali ion. In dispersionsofsilica in sodium
hydroxide solutions or from sodium water-glasssolutions
at or above 100'C, magadiite(NarSi,oOrn.xHrO)
forms
preferentially (Iagaly et al., 1975; Schwiegeret al., 1987)
whereasKrSi2.O4r.xHrO is easily precipitated from dispersionsin potassium hydroxide solutions (Benekeet al.,
1984). This potassium silicate is closely related to kenyaite (NarSiroOo,.xHrO,Eugster, 1967; McAtee et al.,
1968;Fletcherand Bibby, 1987),exceptfor an increased
intracrystalline reactivity (Benekeand l-agaly, I 9 83). Evidently, potassiumions exert an oppositeinfluenceon the
reactivity ofhydrated layer silicatesand three-layer clay
minerals. Potassiumions reduceor even block the intracrystalline reactions (cation exchange,swelling) of more
highly chargedclay minerals, but increasethe reactivity
of hydrated layer silicates.This observation is confirmed
by a new layered potassium silicate that, below 100 "C,
forms from silica dispersedin potassium hydroxide solution or from potassium water-glasssolutions.
PnocBnunns
Aqueousdispersions
of SiO,andKOH wereallowedto stand
at 60-120"C (Table1).A gel-like,but crystallinematerial(here
definedas K-SH) with a basalspacingof 17.3A was formed
afler sometime. Longerperiodsof reactiontransformedK-SH
into the potassiumsilicateKrSiroOn,.xHrO.
In reproducing
identicalnrns,tracesofpotassiumsilicatefirst
appearedafter different time periodsrangingbetweenI and 3
monthsat 70 "C. Formationof the silicateis probablydelayed
by nucleationproblems.At 60'C, the reactionrequiredabout6
months.K-SH wasformedmost rapidlyat 90 qCfrom dispersionscontaining3 mol SiOr, I mol KOH and 30 mol water
(Tablel). The potassiumsilicatealsoformedrapidly(in 20 d at
0003-004x/89/ 0t024224s02.00
zzl
70 "C) from a solution ofpotassium water glass.The water glass
used here was type 28130 from Henkel Comp. (Germany) with
2l wto/oSiOr, 8.1 wto/oKrO, and was partially neutralized by
adding 3 mL conc. HrSOnto 100 mL, which adjusted the molar
ratio SiO,/KOH to about 4/1.
The presenceofcarbonate ions retards or even impedes formation of the potassium silicate K-SH.
The precipitated products were separatedfrom the mother
liquid by filtration. Extensive washing with water was avoided,
becausethis transformed the silicate into the silicic acid. The
sampleswere air- or freeze-dried.The chemical composition of
the air-dried sampleswas obtained from weight loss at 1200'C
before and after extraction of potassium by lM hydrochloric
acid. The extractedpotassium was determined by a DC plasma
emission spectrometer(SpectraSpanIV, BeckmanInstruments).
The crystalline silicic acid H-SH was preparedby leachingthe
silicate in an excessof lMhydrochloric acid. The sampleswere
washedwith distilled water and air-dried.
Weight-loss curves were obtained by stepwise heating of a
sample for one or more days at the desired temperatures (Fig.
1). Recording in a usual rclore device led to lessreliable values,
becausethe releaseofwater proceedsvery slowly (for instance,
weight loss of the acid at 300 'C: 8.90loafter 24 h, 9.3o/oafter 3
d, 9.5o/oafter 7 d). Smaller weight losseswere observed when
sampleswere heateddirectly to higher temperatures,presumably
becausesome amounts of water remain trapped between collapsing layers. Isothermal dehydration and rehydration \ilere
studiedby equilibrating 100-mgsamplesover standardsolutions
for humidity control.
The new potassium silicate hydrate K-SH and the potassium
silicate hydrate xlSiroOo,.xHrO were identified by their X-ray
powder diagrams,which were obtained in Debye-Scherrercameras (diameter 11.46 cm, CuKa radiation, Ni filter, intensities
from visual inspection).
To study intracrystallinereactions,the air-dried potassiumsilicatesor silicic acids were mixed with a small excessof the liquid
guestcompound. After standingsome days at room temperature
or at 65 'C, the mixture was transferred into Lindemann glass
225
BENEKE AND LAGALY: HYDRATED POTASSIUM LAYER SILICATE
TABLE1. Preparationof the new potassiumsilicatehydrateK-SH
t6
Starting materials(mol)
sio,
HrO
3
2
Temperature of reaction: 60 eC
1
40
1 6 0d : a ;
Temperature of reaction: 70 qC
1
40
35d:a;
1
40
1 0 0d : a :
135
'l
25
1
40
95d:a;
2
Temperature of reaction: 90 eC
135
3
4
3
e
e
e
3
a
Time of reaction (days) and type
of oroducts'
30
25
20
1 3d : a ;
9d:a;
200 d: K-SH
48 d: K-SH
120 d: K-SH
62 d: K-SH
17 d: K-SH
125 d: K-SH
13d:
13d:
20 d:
27 d:
K-SH
K-SH
K-SH
K-SH
15-o
..6
14o
o
?
;o
o
I
rJd
c
q)
o
12
3
oooooo-Eio
' a : product amorphous.K-SH identified the X-ray powder pattern
by
Some amorphous material may also be present.
tubes and X-rayed. Several solid guest compounds (urea and
derivatives, amino acids,etc.) were applied in saturatedaqueous
solutions.
11,^
RBsur,rs
Potassiumsilicate
The potassiumsilicateK-SH forms predominantly from
dispersions containing 24 mol SiO, and 20-50 mol water
per mol KOH. Evidently, it is an intermediate phaseduring formation of KrSiroOo,.xHrO.At temperatures> 100
"C, formation of K2Sir.O4r.xHrO is relatively fast, and
the intermediate product is difficult to detect. This product was observedat early statesof reaction in some runs
at 95 "C. The presenceof sodium ions impedesformation
of potassium silicatesand promotes formation of magadiite.
The irregularly shapedthin platesof K-SH, as observed
by scanning electron microscopy (Fig. 2), are quite different from the spherical aggregatesof magadiite and
K,SiroOo,.xH rO (I-agaly, I 979; Benekeand Lagaly, I 983,
1984;Schwiegeret al. 1985).The chemicalcomposition
of the air-dried precipitatesis about KrO.l4sior.8HrO.
The products were washedwith a small amount of water
to avoid removal of lattice potassium.Repeatedwashing
with ethanol removed only small amounts of KOH, and
the amount of adherent mother liquid was probably small.
When an aqueousdispersion of the potassium silicate
was titrated with hydrochloric acid (Fig. 3), the changes
below pH : 7.5 were similar to those for other alkali
silicates(cf. Fig. 3, in Lagaly, 1979). The upward deviation of the titration curve above pH : 7.5 is related to
dissolution and hydrolysis of some amounts of silicate.
For the same reason, pH of the aqueous dispersion increasesslowly from about 8.7 to 9.5 within 2 h under
conditions as in Figure 3.
In contact with the mother liquid, the basal spacingof
K-SH is 17.3A. The powder pattern of the air-dried sample displays only a few reflections (Table 2a) besidesthe
typical basal reflection at d : l5.l A (air-dried samples)
t3E
,a -d
o
!
o
11
--o-o-^
u-
b
of (a) the potassiumsilicateK-SH
Fig. l. Thermalchanges
and (b) the derivedcrystallinesilicic acid H-SH betweenroom
temperature
and 1200'C, O basalspacing,O weightloss.Solid
below400
lines: samplesheatedseveraldaysat temperatures
.C; dashedlines: samplesheatedfor 24 h; a : amorphous;
cr
: cristobalite;tr : tridymite.
and a reflection at d : 3.43 A. Rehydration in water
increasesthe spacingagain to 17.3 A. Drying above 200
oC decreasesthe spacingto 12.8 A
Gt. 2a). These samples do not rehydrate to 17.3A when dispersedin water.
Samplesdried at 300'C and rehydratedgive a spacingof
13.3 A. After heating to 400 'C, rehydration ceases,and
the spacingremains unchangedat 12.8 A. At about 500
oC, the material becomes amorphous, but recrystallizes
at 600 "C. The crystalline products are low cristobalite
and, in distinctly smaller amounts, ot-quartz.Well-crystallized tridymite is formed at 800 "C and is stableat least
up to 1200oC.Above 1000'C, considerableamountsof
K"O are desorbed.
226
BENEKE AND LAGALY: HYDRATED POTASSIUM LAYER SILICATE
t!A
;.1
'; .t(!, ,
i'r r,iij,
Fig.2. Scanningelectron micrographs of (a) the potassium silicate hydrates KrSiroOo,.xHrO and (b) K-SH, prepared from 3
mol SiO, and I mol KOH in 40 mol H,O at 70'C, after 120 d.
In contrast to magadiite, the potassium silicate K-SH
intercalatesa variety of organic compounds (Table 3). A
pronounced ability to intercalateguest moleculesis very
typical ofthis silicate and also ofthe derived silicic acid
H-SH (seebelow).
The crystalline silicic acid
The potassium silicate K-SH is easily transformed into
the silicic acid H-SH by leaching in diluted acids. Type
and concentration of the acids are not decisive. For in-
227
BENEKE AND LAGALY: HYDRATED POTASSIUM LAYER SILICATE
TAaLE28. Powder-diffractionpattern of the crystallinesilicicacid
H-SH
Air-dried
Dried over P4O,o
d (A)
d(A)
13.0
7.22
6.50
J.O I
3.36
1.83
100
50
50
50
80
10
meq HCI/ g
13.4
7.20
6.63
5.31
4.44',
3.98'
3.s-3.4t
1.83
Fig. 3. Titration of the potassiumsilicateK-SH with 0.1M
potassiumsilicatedispersed
hydrochloricacid(145.34-mg
in 50
mL water;watercontentof the potassiumsilicate13.4o/o).
stance,0.lN and 2N HCl, HNO3, and HrSO4led to identical products. The shape of the platelike particles remains nearly unchanged.A spacingof 13.0 A is typical
ofthe air-dried acid (Table 2b). It increasesto about 15
A when the acid is dispersed in water. The broadened
basal reflection indicates that the degreeof order of the
rehydrated form is lower than in the 13.4-A phase.
Leaching in acids at 100 'C produces samplesthat, after
air-drying, reveal the samebasal spacingbut only expand
ro 14.2-14.4A in water.
The water content of the silicic acid H-SH is insensitive
to the relative humidity between 40 and 600/0,but changes
strongly at lower or higher humidity (Fig. a). The ratio
HrO/SiO, increasesfrom about 0.2 at 0o/oto about 0.8 at
87olorelative humidity and is accompaniedby basal-spacing changesfrom 13.0 to 14.0 A. Capillar condensation
increaseswater adsorption continuously above 70o/orelative humidity, so that the stepwiseincreaseof the basal
spacingfrom 13.6 to 14 A is not reflected.
The spacing of the air-dried acid remains nearly unchangedup to about l7O "C, then decreasesto ll.6 A
and remainsconstantup to 900 'C (Fig. 2). At 1000'C,
the material becomes amorphous. Formation of cristobalite starts at about I 100 "C. In spite of the constant
basal spacing,water is still desorbedbetween400'C and
1200lC. The weight lossincreasescontinuouslyfrom 100/o
Io 13.30/o.
After heating to 1200 oC,the HrO/SiO, ratio is
<0.01. The compositionof the I1.6-A phasevaries between 0. 12 and 0.06 HrO/SiOr, and the air-dried acid has
the compositionSiOr.0.5HrO.
100
40
50
20
10
10
20
' Reflectionbroadened.
i Broadenedband at 3.4-3.5 A.
Samplesheatedto 170 .rCrehydrateinto the 15-A form.
Dehydration at higher temperatures decreasesthe proportion of expandableinterlayer spaces.After rehydration, the spacingremainsbelow l5 A. The interlayerspace
of samples heated to an excessof 300 oC is no longer
expanded by rehydration. Very likely, siloxane bridges
betweenadjacent silicate layers prevent separationofthe
layers.
The surface acidity of H-SH (betweenHammett surto
faceacidity H0: -3.0 and Ho: +1.5) corresponds
the acidity of the acid from XlSiroO4r'xH2O (Werner et
al., 1980).Titration of the acid with KOH in the presence
of KCI indicatesa K*/SiO, ratio of aboul2/14, very close
to the Na*/SiO, ratio in magadiite. Generally, exact values are difficult to obtain from such titration curves (cf.
Benekeet al., 1984) so that this value may be considered
as a first approximation.
Intercalation capabilities
Most striking is the ability of the acid to intercalate a
very large number of guest molecules.The basal spacing
ofthe acid after intercalation ofa few selectedguestcompounds is reported in Table 4. In contrast to other crystalline silicic acids that intercalate only a few types of
compounds, the acid H-SH reactswith alcohols; glycols;
Tmr-e3. Basalspacingof K-SHand H-SHin the presenceof
guestmolecules
Basal spacing (A)
Guest comoound
patternof potassium
TABLE
2A. Powder-diffraction
silicateK-SH
Air dried
Under water
d(A)
15.1
5.00"
3.43
3.14.
2.96
2.85
1.82
d(A)
100
20
50
30
10
10
20
' Reflectionbroadened.
17.2
7.08
5.00
3.s8
3.43
3.32
3.14
1.83
100
10
10
20
40
30
30
30
air-driedsample
unclerwater
15.1
17.3
13.4
15.0
acetamide
N-ethyl urea
piperazine
2.5-dimethyl
prperazlne
ephedrine
imidazole
trimethylamine-N-oxide
ethyleneglycol
glycerol
methanol
butanol
17.7
15.8
19.6
15.4
17.0
19.8
20.3
24.2
21.4
21.3
17.6
17.2
16.0
14.8
19.8
24.2
16.4
15.8
17.1
16.5
16.5
15.9
228
BENEKE AND LAGALY: HYDRATED POTASSIUM LAYER SILICATE
TABLE5. Basal spacingsof H-SH and the acid form of magadiite
Basal spacing d (A)
Guest compound
Acid
form of
magadiite
iUmethylformamide
propionamide
N-ethylformamide
M tv-dimethylformamide
/V-methylacetamide
/V,Mdimethylacetamide
Mw-diethyl formamide
N-methylurea
Methyl urea
N,M-dimethylurea
dimethyl sulfoxide
ethyleneglycol
LdL + 2.8'
H-SH
(A)
DI (A)
15.9
14.5
14.0
16.7
17.4
16.5
16.5
1 7. 0
17.7
17.'l
16.8
194
'17.2
1 7. 5
6.4
7.'l
6.5
6.2
8.8
6.6
6.9
5.7
6.1
6.3
6.3
6.2
6.7
7.O
15.5
14.9
16.2
15.8
15.4
15.7
1 7. 7
16.7
5.1
7.1
6.1
5.3
5.5
6.2
16.6
1 7. 1
6.0
6.5
6.1
5.6
' ad,:
4 - 13.4(A),Ior S-SH.
t D: diameterof the guestmolecule.
(broadenedreflections)by potassium acetate.Potassium
Fig. 4. Watercontentof the crystallinesilicicacid H-SH at acetateis a highly effectiveintercalation agent for kaolindifferentrelativehumidities:(a) weightchanges
of an air-dried ite, but does not react with almost all other types of laysample,O) H,O/SiO,ratio x, (c) basalspacingd,.
ered materials.
The intracrystalline reactivity toward many guestcompounds is lost above 200 or 230 "C. Severalguest molenitrils; acid amides and derivatives; urea and derivatives;
cules,for instance,hydrazine hydrate and piperazine,can
amine oxides, sulfoxides, and phosphine oxides; several
penetratebetweenthe layers and increasethe basal spacamino acids; aliphatic and aromatic amines; aromatic
ing, even if the acid has beenheated at260"Cfor a period
heterocycles;and ketones. The reactivity with alcohols,
of7 d. One may conclude that condensationofadjacent
nitriles, ketones, and several amino acids distinguishes
layers starts above this temperature.
this acid from other intracrystalline reactive layered materials.
DrscussroN
Noteworthy is the lattice expansion to about 19.6 A
The variation of the basal spacing of the potassium
silicate K-SH and the derived crystalline silicic acid suggests that these compounds are intracrystalline reactive
TABLE
4. Basalspacingsof the silicicacid H-SHafter intercalationof guest molecules(basalspacingof air-dried layered silicates.The alkali/SiO, ratio is close to that of
magadiite. However, K-SH is not simply the potassium
acid:13.4(A)
analogueof magadiite NarSi,oOrr.xHrO. The potassium
Group of
analogue of magadiite prepared from the acid form of
compounds
Guestmolecules
Basalspacing(A)
magadiite and KOH reacts quite differently. The most
methanol
16.5
striking difference is the reversible hydration and dehybutanol
15.9
dration of K-SH that is accompanied by basal spacing
octanol
15.6
ethyleneglycol
changesbetween
l5.l and 173 A. The l5.l-Aformcor17.1
glycerol
16.5
responds to an intercalation of a water monolayer behexanediol-1,6
15.3
tween the silicate layers. The spacingof 17.3 A indicates
ketones
acetone
15.7
an intercalation of a bilayer of water molecules. Maganitriles
acetonitrile
15.9
diite
and the potassium analoguedo not expand beyond
butyronitrile
16.2
15.6
A
in contact with water. If potassium silicate K-SH
amrnes
ethylenediamine
15.6
'17.9
and magadiite had an identical layer structure, the cryshexamethylenediamine
triethylenediamine
21.4
talline acids would be identical. This is not observed.The
piperazine
19.8
striking property of the acid H-SH is the interlamellar
ephedrine
24.2
hydration in water, which increasesthe basal spacingfrom
decylamine
40.4
8.4 A to about tS A. fne acid form of magadiite does
aniline
15.9
not intercalate a second layer of water molecules when
trimethylamine-/\Loxide
15.8
pyridine-N-oxide
dispersedin water. Samplesof H-SH heated to an excess
16.5
dimethyl sulfoxide
16.6
of400 oCreveal higher basalspacings(11.6 A than the
acid amides
see Table 5
acid form of magadiite (11.2 A). A further difference is
urea derivatives
see Table 5
the intracrystalline reactivity toward several classesof
BENEKEAND LAGALY: HYDRATED POTASSIUMLAYER SILICATE
guest compounds. For instance,the acid form of magadiite does not intercalate alcohols, nitriles, and ketones.
The different basal spacingsafter intercalation of guest
molecules(i.e., Table 5) also reveal a structural difference
between both the crystalline silicic acids.
In most casesthe guest molecules displace interlayer
water molecules.In a first approximation, the sum A d,
+ 2.8 (A d, representsthe increaseof the basal spacing
in angstroms;2.8 A is the diameter of a water molecule)
should be close to the diameter D of the guest molecule.
This condition is often fulfilled, but deviations of some
tenths of angstromsoccur, as the guestmoleculesseldom
adopt spherical conformation. On the other hand, the
deviations are not very large becausethe mean diameter
of small molecules (calculated from the molecular volume) is often not very different from the largest dimension of the extendedmolecules.This condition is certainly not fulfilled for longer chain molecules.The interlayer
separationby propionamide is distinctly larger than the
diameter of the hypothetical sphere.Dimethyl urea moleculesare compact molecules,and the interlayer separation approximately correspondsto the spherical diameter. In contrast, N-ethyl urea adopts a more extended
conformation, which producesa higher basal spacing.The
largeinterlayer separationafter intercalation of N-methyl
acetamideis also observedfor the acid form of magadiite.
In most cases,the interlayer separation of H-SH by
small guestmoleculesis distinctly higher than for the acid
form of magadiite and suggestsa closer packing of the
guestmoleculesin the interlamellar monolayer. The basal
spacingsof the alkanol complexesgive values of A d, 't
2.8 between4.7 and5.3 A, which correpondsto the thicknessof an alkyl chain. The alkanol moleculesseemtherefore to be intercalatedwith their long axesparallel to the
silicate layers.
CoNcr,usroN
The hydrated potassium silicatesK-SH and l(rsiroo4r.
xH,O only form in alkali solutions with high potassium
content, whereasthe presenceof sodium ions promotes
formation of magadiite. Thus, hydrated potassium silicates should be formed in nature under very particular
conditions only. A further cause,which should restrict
thesematerials, is the low stability of the hydrated alkali
silicates.Leachingin water, for instance,transforms them
into crystalline silicic acids. Silicic acids can also be formed
from several other minerals (such as gillespite, apophyllite, carletonite, sanbornite,litidionite: Pabst, 1958;Guth
et al., 1978;Frondel, 19791'L.ag;aly
arrdMatouschek,1980).
229
The silicic acids may be overlooked in nature becauseof
identification difficulties. These difficulties might arise
from the variable interlayer spacingand the ability ofthe
acids to intercalateguestmolecules.Silhydrite (Gude and
Sheppard, 1972) is the only natural silicic acid that has
been described.It representsthe acid form ofmagadiite,
but its properties are affectedby intercalatedorganic material (Benekeand Lagaly,1977).
RsrnnnNcns crrno
Beneke,K., and kgaly, G. (1977) Peculiaritiesof intercalation reactions
of the natural silicic acid silhydrite. Clay Minerals, 12, 363-365.
( I 983) Kenyaite-Synthesis and properties. American Mineralogist, 68, 8 18-826.
(l 984) Kristalline Kieselsiiuren.GlT-Fachzeitschrift fiir das laboratorium, 516-527.
Beneke,K., Kruse, H.-H., and Iagaly, G. (1984) Eine kristalline Kieselsaure mit hoher Einlagerungsfiihigkeit. Zeitschrift fiir anorganischeund
allgemeineChemie,518, 65-76.
Eugster,H.P. (1967) Hydrous sodium silicatesfrom Lake Magadi, Kenya:
Precursorsof bedded chert. Science,157, 1177-1180.
Fletcher, R.A., and Bibby, D.M. (1987) Synthesisof kenyaite and magadiite in the presenceofvarious anions. Clays and Clay Minerals, 35,
3 l 8-320.
Frondel, C. (1979) Crystalline silica hydrates from leached silicates.
American Mineralogist, 64, 799-804.
Gude, A.L, and Sheppard,R.A. (1972) Silhydrite, 3SiO,H,O, a new mineral from Trinity County, Cdifomia. American Mineralogist, 57, 105!
r065.
Guth, J.-L., Hub€rt, Y., Kalt, A., Perati, B., and Wey, M.R. (1978) Un
nouveau type de silice hydrat6e cristallisee. Academie des Sciences,
Comptes Rendus Hebdomadairesdes Seances,Sdrie D, SciencesNaturelles Paris, 286, 5-8 (in French).
Lagaly, G. (1979) Crystalline silicic acids and their interface reactions.
Advances in Colloid Interface Science,70, 46-55.
Lagaly, G., and Matouschek,R. (1980) Die kristallinen Kieselseurenaus
Apophyllit, Carletonit und Gillespit. Neues Jahrbuch fia MineralogischeAbhandlungen, 138, 8l-93.
Lagaty,G., Beneke,K., and Weiss,A. (1975) Magadiite and H-magadiite:
I. Sodium magadiite and some of its derivatives. American Mineralogtst,60,642449.
McAtee, J.L., House,R., and Eugster,H.P. (1968) Mapdiite from Trinity
County, California. American Mineralogist, 53, 2061-2069.
Pabst,A. (l 958) The crystal structureofleached gillespite,a sheetsilicate.
American Mineralogist, 43, 970-980.
Schwieger,W., Heidemann, D., and Bergk, K.-H. (1985) High resolution
solid-state silicon-29 nuclear magnetic resonancespectroscopic studies
of synthetic sodium silicate hydrates. Revue de Chimie Minerale, 22,
639450.
Schwieger,W., Heyer, W., Wolf, F., and Bergk,K.-H. (1987)Zur Synthese
von kristallinen Metallsilicathydraten mit Schichtstruktur. Zeitschrift
fiir anorganischeund allgemeineChemie, 548,204-216.
Werner, H.-J., Beneke,K., and Laealy, G. (1980) Die Aciditiit kristalliner
Kieselsluren. Zeitschrift fiir anorganischeund allgemeine Chemie, 470,
l 18-l 30.
MeNuscnrrr REcEIvEDFrnnuexv 24, 1988
MeNuscnrrr AccEprEDSrrrrNrsER 13, 1988