Structure of amorphous alumina tubes studied by positron annihilation lifetime spectroscopy

Article abstract of DOI:10.1007/s10789-005-0119-8

Positron annihilation lifetime spectroscopy (PALS) is used to investigate the detailed structure of amorphous alumina tubes prepared via thermohydrolysis of anhydrous AlCl₃ . The results demonstrate that PALS is a sensitive probe of the structure of amorphous materials, capable of distinguishing between the structures of amorphous alumina tubes and amorphous alumina prepared by a standard procedure, which are shown to differ slightly in the concentration and size of free-volume elements. The observed distinctions can be understood in terms of the anisotropic aggregation of primary alumina particles during hydrolysis in aqueous media or thermohydrolysis in water vapor.

Full text of DOI:10.1007/s10789-005-0119-8

Inorganic Materials, Vol. 41, No. 3, 2005, pp. 255–260. Translated from Neorganicheskie Materialy, Vol. 41, No. 3, 2005, pp. 314–320.
Original Russian Text Copyright © 2005 by Shantarovich, Kevdina, Miron, Baronova, Berdonosov, Baronov, Melikhov.
Structure of Amorphous Alumina Tubes
Studied by Positron Annihilation Lifetime Spectroscopy
V. P. Shantarovich*, I. B. Kevdina*, N. F. Miron*, Yu. V. Baronova**,
S. S. Berdonosov**, S. B. Baronov**, and I. V. Melikhov**
*
Semenov Institute of Chemical Physics, Russian Academy of Sciences,
ul. Kosygina 4, Moscow, 119991 Russia
*
* Moscow State University, Moscow, 119992 Russia
e-mail: berd@radio.chem.msu.ru
Received July 8, 2004
Abstract—Positron annihilation lifetime spectroscopy (PALS) is used to investigate the detailed structure of
amorphous alumina tubes prepared via thermohydrolysis of anhydrous AlCl . The results demonstrate that
3
PALS is a sensitive probe of the structure of amorphous materials, capable of distinguishing between the struc-
tures of amorphous alumina tubes and amorphous alumina prepared by a standard procedure, which are shown
to differ slightly in the concentration and size of free-volume elements. The observed distinctions can be under-
stood in terms of the anisotropic aggregation of primary alumina particles during hydrolysis in aqueous media
or thermohydrolysis in water vapor.
INTRODUCTION
and intensity I . Two components are due to singlet
i
positronium and free positron annihilation. The third
component and, occasionally, the fourth and fifth com-
ponents are due to orthopositronium annihilation and
carry information about the FVEs in the material.
The purpose of this work was to evaluate the aver-
age size of free-volume elements (FVEs) in amorphous
materials using positron annihilation lifetime spectro-
scopy (PALS). This method was used earlier in many
studies to probe the defect structure of solids [1–6], in
Most PALS studies of metal oxides were carried out
particular that of metal oxides, including alumina [7–9]. in the 1970s. In recent years, considerable research
effort has been concentrated on nanoparticulate amor-
phous oxide materials. In this context, it is of interest to
assess the potentialities of PALS in probing the fine
structure of such systems.
In PALS, the sample to be studied is irradiated with
positrons, and analysis of positron annihilation charac-
teristics provides information about the structure of the
material. The annihilation of positrons (via interaction
with electrons) in solids may be preceded by several
processes, including the formation of positronium Ps, a
hydrogen-like system. Owing to exchange interaction,
positronium is repelled from host atoms and has time to
localize in FVEs of the material (in lattice defects,
pores of amorphous materials, and intercrystalline
spaces). It is well known that positronium may exist in
In this work, we focus on amorphous alumina tubes.
It has been shown very recently [10–13] that the ther-
mohydrolysis (150–230°C) of anhydrous aluminum
chloride yields alumina tubes 80–100 µm to 3–7 cm in
length and 20 to 300 µm in diameter (Fig. 1) owing to
the self-organization of primary particles. The process
is run in a heated reactor. Anhydrous AlCl is spread
3
t
two forms: long-lived orthopositronium ( Ps) and short-
over an inert material (e.g., in a porcelain boat). The
surface layer of aluminum chloride is first partially
p
lived parapositronium ( Ps).
The orthopositronium annihilation radiation is more hydrolyzed by exposing the sample to water vapor.
informative in studies of FVEs since the relatively long Next, the sample is heated in a gas stream (hydrogen,
t
helium, air, or CO ), which leads to rapid (up to
2
lifetime of Ps depends on the effective size of FVEs in
which it is localized. In some cases, the intensity of the 0.5 mm/s) growth of tubes.
long-lived orthopositronium component in the PALS
As shown by x-ray fluorescence analysis and radio-
spectrum may depend on the size and concentration of
FVEs in the material. The intrinsic lifetime of para-
tracer measurements, the x-ray amorphous tubes thus
prepared consist mainly of oxygen and aluminum, with
–
10
p
positronium is much shorter (1.25 × 10 s), and Ps
27
3
.9–4.3 at % Cl impurity. According to Al NMR
annihilation is not very informative.
results, the Al atoms in the tubes are in octahedral oxy-
t
Taking into account the processes preceding Ps gen coordination, in contrast to conventional amor-
annihilation and using application programs, one can phous alumina, in which Al atoms are in tetrahedral or
distinguish PALS components differing in lifetime τi mixed (tetrahedral + octahedral) coordination.
0
020-1685/05/4103-0255 © 2005 Pleiades Publishing, Inc.

2
56
SHANTAROVICH et al.
in a gas flow. The forming tubes were withdrawn from
the boat with forceps and were then stored in a vial at
room temperature. The residual powder was also with-
drawn from the boat for characterization.
Amorphous alumina was prepared by a standard
procedure, by dehydrating aluminum hydroxide [14].
The starting solution (10% AlCl ) was cooled to 0°C,
3
and 25% aqueous ammonia, also cooled to 0°C, was
added. The resulting precipitate was collected on a filter
and washed with water to neutral pH. Next, it was
placed in a porcelain dish and dried in a desiccator over
P O for 2 weeks. α-Al O (corundum) was prepared by
2
5
2
3
calcining aluminum hydroxide at 900°C for 4 h.
3
0 µm
All of the samples were characterized by SEM and
x-ray diffraction (XRD).
(
a)
Positron annihilation studies. PALS spectra were
measured on a standard EGG&ORTEC fast–fast spec-
trometer (pulse height analysis in the fast channel). The
time resolution (full width at half maximum of the
instant-coincidence curve) was 250 ps. As the positron
2
2
source, we used Ni-sheathed NaCl. The PALS spec-
trum was stored in a multichannel analyzer (1024 40-ps
channels). The total statistics (net count) was above
6
several times 10 coincidences. The spectra were ana-
lyzed using the POSITRONFIT code, which takes into
account the chance coincidence background, resolution
fluctuations, and annihilation in the source material.
The PALS spectra of the samples studied are shown
in Fig. 2. The lifetimes τ and intensities I extracted
i
i
3
0 µm
(
b)
from these spectra are listed in Table 1. Analysis of
these data indicates that the components τ (I ) and τ
3
3
4
Fig. 1. SEM micrographs of alumina tubes.
(I ) arise from orthopositronium annihilation in FVEs
4
and, hence, can be used to evaluate their parameters.
The effective FVE radius was determined using the
Tao–Eldrup–Jean semiempirical formula [2, 3], cor-
rected (which makes sense at positronium lifetimes
over 10 ns) for the intrinsic lifetime of orthopositron-
Scanning electron microscopy (SEM) examination
demonstrates that the tubes consist of particles 150 to
00 nm in size (in accordance with nitrogen BET sur-
2
face area measurements), firmly bonded to each other.
Such particles are assumed to be composed of primary
particles 5 to 40 nm in size [12, 13]. It is of interest to
use PALS in order to gain further insight into the struc-
ture of the tubes and the processes underlying their for-
mation. Note that systematic comparative PALS studies
of amorphous and crystalline Al O materials have not
t
t
–
1
ium, τ = (λ ) [9]:
0
0
1
t
0
λ = - -- = λ
i
τ
i
(1)
R
1
2πR
 
i
–1
2
3
+
2 1 – ------------------ + ---- -- sin ------------------ ns .
 
R + ∆R
yet been reported.
R + ∆R 2π
i
i
In this work, PALS was used to characterize FVEs
in amorphous Al O and, for comparison, in aluminum
For components 3 and 4, i = 3 and 4, respectively.
∆R, which characterizes the penetration of positron-
2
3
hydroxide (intermediate product of AlCl thermohy-
3
drolysis), thermohydrolysis by-products, and polycrys- ium into FVEs, was taken to be 0.166 nm [2, 3] based
talline α-Al O (corundum).
on calibration results for different substances.
2
3
In determining the FVE concentration, we took into
account that an orthopositronium atom may appear in
an FVE (pore) as a result of several processes, of which
the following are the most important:
EXPERIMENTAL
Sample preparation and characterization. To
prepare tubular particles, anhydrous AlCl (2 g) was
3
I. A positronium atom forms, with probability Q, in
spread over a porcelain boat and exposed for 3–4 h to the material being studied and then diffuses until it is
humid air. Next, the sample was heated to 220–250°C captured by an FVE. Further, two situations are possi-
INORGANIC MATERIALS Vol. 41 No. 3 2005

STRUCTURE OF AMORPHOUS ALUMINA TUBES
257
lnN(t)
1
0
9
8
7
6
5
4
3
2
1
1
2
3
4
0
50
100
150
200
250
300
Channel
Fig. 2. PALS spectra of (1) α-Al O (corundum), (2) amorphousAl O tubes, (3) thermohydrolysis by-products, and (4) amorphous
2
3
2 3
Al O .
2
3
ble. (Ia) If the annihilation rate of free (nonlocalized) material. In this case, the FVE concentration cannot be
positronium is comparable to the capture rate, these evaluated from PALS data.
processes compete with one another. An important
II. Positronium forms in a free volume after a
point is that the amount of orthopositronium (the inten- positron has found its way into it. The rate of positron-
sities of the long-lived components I and I ) is deter- ium formation is then limited by the FVE concentration
3
4
(
and possibly by the electron density in the track).
mined by the FVE concentration, which can, therefore,
be inferred from the annihilation characteristics. (Ib) If
the FVE concentration is so high that the positron cap- ized positronium are a nonlocalized positronium atom
ture rate far exceeds the rate of positron annihilation, I3 and a positron, respectively.
Thus, in processes I and II, the precursors to local-
and I are determined not by the FVE concentration but
by the probability Q of positronium formation in the tions of the form (2) [4, 5], which take into account
4
Processes I and II are described by similar rate equa-
Table 1. Annihilation characteristics of Al O samples
2
3
Amorphous
alumina tubes
Amorphous
alumina
Amorphous
by-product
Sample
Al(OH)3
α-Al O (corundum)
2 3
τ1, ns
I1, %
τ2, ns
I2, %
τ3, ns
I3, %
τ4, ns
I4, %
0.2562 ± 0.0044
59.3483 ± 2.3807
0.5245 ± 0.0215
27.087 ± 2.126
1.6811 ± 0.0283
10.161 ± 0.357
35.0336 ± 1.6429
3.404 ± 0.076
0.2793 ± 0.0039
69.5131 ± 1.9203
0.6434 ± 0.0392
20.5043 ± 1.3435
1.4737 ± 0.0338
9.9827 ± 0.8173
0.2853 ± 0.0033
0.2753 ± 0.0039
0.1967 ± 0.0017
72.7119 ± 1.0166
0.4273 ± 0.0078
22.4938 ± 1.0365
2.5205 ± 0.3697
0.5890 ± 0.0434
42.3578 ± 4.7024
4.2054 ± 0.2415
69.9829 ± 1.7346 69.5723 ± 2.4763
0.6223 ± 0.0338 0.5805 ± 0.0436
16.8671 ± 1.4716 21.7022 ± 1.4573
1.7614 ± 0.0171
1.2109 ± 0.0573
8.3491 ± 1.3894
9.6733 ± 0.8037
0.3764 ± 0.0185
13.1500 ± 0.3611
–
–
INORGANIC MATERIALS Vol. 41 No. 3 2005

2
58
SHANTAROVICH et al.
+
+
Ps
either the behavior of the positrons (k = e ) forming where D and D are the respective diffusion coeffi-
orthopositronium upon localization or the behavior of cients. It is commonly believed that, at least in amor-
orthopositronium (k = Ps) localizing in a free volume,
phous materials, these coefficients differ substantially:
+
–3
–1
2
Ps
–5
D
ꢀ 10 to 10 cm /s [15, 16] and D = 10 to
−
4
2





k
f
k
f
k
f
10 cm /s [9, 17, 18]. For this reason, the concentra-
tions estimated for different cases may differ markedly.
dP /dt = – λ P –
v P ,
f
∑
i
(2)
i
k
k
k
dP /dt = (3/4)v P – γ P ,
i
i
f
i
i
RESULTS AND DISCUSSION
k
f
+
The annihilation characteristics of Al O tubes and
with boundary conditions P (0) = 1 at k = e and
k
2
3
conventional amorphous Al O are listed in Table 1 in
2
3
P (0) = Q at k = Ps. Here, P and P are the probabilities
of finding a precursor to localized orthopositronium in features of these data warrant attention.
f
i
f
comparison with those of references samples. A few
a localized microvolume of type i or in a free state (sub-
script f). The factor 3/4 appears because we consider
annihilation of triplet positronium. The parameter Q
represents the fraction of positrons involved in positro-
nium formation.
1
. The spectra of tubular and amorphousAl O show
2 3
only one positronium component (τ , I ), with a rela-
tively short lifetime of 1.47–1.76 nm.
3
3
2
. The spectrum of α-Al O shows two orthop-
2
3
ositronium components (τ , I and τ , I ), but the inten-
sity of the former is very low, which is attributable to
crystallization of amorphous microparticles.
3
3
4
4
Occasionally, annihilation spectra showed two
positronium components, indicating that the sample
contained two types of free volumes. The following
equations, relating the positron capture rates v and v
3
. According to earlier results [12, 13], the mini-
3
4
mum particle size of amorphous Al O does not exceed
to annihilation characteristics, were obtained for two
types of free volumes:
2
3
1
00 nm. This value is very close to the diffusion length of
+
+
2
positrons in solids, λ = 6Dτ , at D = 0.001–0.1 cm /s
[15, 17] and τ = 0.2–0.3 ns and far exceeds the diffusion
I
A
3
4
3


4
v = --- (λ – γ ) --Q – I + (λ – γ )I ,
3
f
3
f
4
4


Ps
–5
length of nonlocalized positronium at D ꢀ 10 to
−4
2
1
0 cm /s [15, 17] and the same τ [4, 17, 18].
I
A
3
4
4


3
v = --- (λ – γ ) --Q – I + (λ – γ )I ,
(3)
4
f
4
f
3
3
In going from amorphous Al O microparticles to
microcrystalline corundum, the short-lived positronium
component (τ , I ) essentially disappears (I ꢀ 0.4%). It
is, therefore, reasonable to assume that this component,
present in amorphous microparticles, is due to internal
microvolumes (to the presence of FVEs, which are very
few and far between in the microcrystallites). Since the
long-lived component τ , attributable to interparticle
free volumes, is missing in amorphous Al O , we con-
clude that positronium does not reach the interparticle
regions. At the same time, as pointed out by Bartenev
et al. [7], some of the particles in question are no
greater than hundreds of nanometers in size, and, in
2
3


3
3
3
3
4
3
4

 

4
A = --Q – I --Q – I – I I .
3
3 4

 

If there is only one type of defect, Eqs. (3) take the
form
4
λ – γ
Q – (4/3)I
f
3
2
3
v = (4/3)I -------------------------- ,
(4)
3
3
3
where λ is the annihilation rate of free (nonlocalized)
f
positrons (positronium) and γ = 1/τ . Process II corre-
i
i
+
2
sponds to Q = 1. For situation Ia, we take, as a rough
approximation, Q = 1 – I (nonlocalized positronium is
view of the large diffusion coefficient, D = 0.1 cm /s,
a positron might reach the microparticle surface and
2
assumed to make an insignificant contribution to this form positronium (mechanism II). Since this is not the
component).
case, we are led to assume that positronium is formed
in the bulk of amorphous grains (this is possible if the
diffusion coefficient of nonlocalized positronium is
In the cases under consideration, the relationship
between the capture rates v and the FVE concentration
i
Ps
–5
–4
2
D = 10 to 10 cm /s [4, 17–19]).
is of importance. For the capture of nonlocalized
positronium (Ia), we have
Thus, it seems likely that mechanism Ia prevails.
Therefore, the FVE concentration in microparticles can
v
i
Ps
N = ------------------ -- ,
(5) be evaluated by Eq. (4). For definiteness, we take D =
i
Ps
–
4
2
4
πD R
10 cm /s. The estimated FVE concentrations in amor-
phous tubes and conventional amorphous Al O are
i
2
3
for the capture of a nonlocalized positron (II),
listed in Table 2.
v
i
It is clear from Table 2 that, even though the amor-
N = ------------------,
(6)
i
+
4
πD R
phous materials are rather close in N and R , they
i
3
3
INORGANIC MATERIALS Vol. 41 No. 3 2005

STRUCTURE OF AMORPHOUS ALUMINA TUBES
259
Table 2. Average radius R and concentration N of FVEs in of these materials differ not very strongly, indicating
amorphous and crystalline aluminas
that they are similar in structure.
Comparison of the PALS data for α-Al O (corun-
2
3
Amorphous
Al O
Amorphous
Material
α-Al O
2 3
dum), amorphous alumina, and alumina tubes provides
some insight into the mechanism of positron annihila-
tion in alumina tubes and allows one to evaluate the
Al O tubes
2
3
2
3
R3, nm
0.263
0.23
0.33
2
0
–3
concentration (about 10 cm ) and effective radius
–
3
3
12.437 × 1019 9.68 × 1019
0.568 × 1016
1.34
N , cm
3
(R ꢀ 2.3 Å) of FVEs in the material.
3
R4, nm
–
–
–
–
Remarkably, we were able to reveal a very slight dif-
ference in the concentration and effective size of FVEs
between alumina tubes and conventional amorphous
alumina, which seems to be related to the difference in
the preparation procedure.
–
1.138 × 1016
N , cm
4
have a number of marked distinctions, which are prob-
ably associated with the fact that the materials prepared
by vapor-phase hydrolysis and vapor-phase topochem-
ical processes differ in structural parameters [12].
ACKNOWLEDGMENTS
This work was supported by the Russian Foundation
In the case of α-Al O , the long-lived component for Basic Research (grant nos. 03-02-32061 and
2
3
0
3-03-32918) and the President’s Grants Council for
with τ > 40 ns, usually missing in molecular crystals,
4
Support to Leading Scientific Schools (grant
no. NSh-1818-2003.3).
warrants attention. It seems likely that this component
arises from the annihilation of the positronium formed
at crystallite boundaries. This mechanism was dis-
cussed repeatedly in the literature [9, 20, 21]. More-
over, the positronium mobility in crystalline phases is
rather high [22]. Since positrons (positronium) may
appear in intercrystalline free volumes, the parameters
of component τ (I ) must depend on the particle size of
REFERENCES
1. Melikhov, I.V., Shantarovich, I.P., Kitova, E.N., et al.,
Microscopic Radionuclidic Characterization of Disperse
Solid Phases, Zh. Fiz. Khim., 1993, vol. 67, no. 1,
pp. 75–79.
. Tao, S.J., Positronium Annihilation in Molecular Sub-
stances, J. Chem. Phys., 1972, vol. 56, no. 11, p. 5499.
. Eldrup, M., Lightbody, D., and Sherwood, L.N., The
Temperature Dependence of Positron Life Times in
Solid Pivalic Acid, Chem. Phys., 1981, vol. 36, no. 1,
p. 51.
4
4
the material, as pointed out in earlier studies of metal
oxides [7]. This effect may be significant at crystallite
sizes below several hundred nanometers.
2
3
Thus, using PALS results and Eq. (5), and taking
+
2
D = 0.1 cm /s, one can estimate the concentration N
4
and effective size R of intercrystalline regions. Similar
4
reasoning applies to the low concentration N of inter-
3
4. Shantarovich, V.P. and Goldanskii, V.I., On the Role of
Free Volume in Positron Annihilation in Polymer Struc-
tures, Hyperfine Interact., 1998, vol. 116, no. 1, p. 67.
5. Shantarovich, V.P., Suzuki, T., He, C., et al., Positron
Annihilation Study of Hyper-Cross-Linked Polystyrene
Networks, Macromolecules, 2002, vol. 35, no. 26,
p. 9723.
crystalline defects, which are responsible for the weak
component τ (I ). The results for this component are
3
3
also presented in Table 2.
Note that component τ (I ) is also prominent in the
4
4
spectrum of aluminum hydroxide (Table 1). The micro-
structural features of the hydroxide responsible for this
component may influence the size of α-Al O microc-
6
. Shantarovich, V.P., Gustov, V.V., Kevdina, I.B., et al.,
2
3
Study of Structural Defects in CaSO · 2H O Crystals by
4
2
rystallites prepared by calcining the hydroxide at
Positron Annihilation and Radiothermoluminescence,
Neorg. Mater., 1997, vol. 33, no. 8, pp. 1007–1011
9
00°C for 4 h.
[
Inorg. Mater. (Engl. Transl.), vol. 33, no. 8,
Component τ , though very weak (0.37%), also
4
pp. 853−856].
appears in the spectrum of the residual powder. Pre-
sumably, this component is due to a small (about 10%)
fraction of oxide crystallites or hydroxide microparti-
7
8
9
. Bartenev, G.M., Varisov, A.Z., Goldanskii, V.I., et al.,
Particle Size Effect on Positron Annihilation in Some
Metal Oxides, Fiz. Tverd. Tela (Leningrad), 1969,
vol. 11, no. 11, p. 3177.
. Paulin, R. and Ambrosino, J., Free Annihilation of the
Ortho-positronium Formed in Some Powders with a
High Specific Surface, J. Phys. (Paris), 1968, vol. 29,
no. 2, p. 263.
. Kusmiss, J.H and Stewart, A.T, Positron Annihilation
in Solid and Liquid Metals, Positron Annihilation,
Stewart, A.T. and Roellig, L.O., Eds., New York: Aca-
demic, 1967, p. 341.
cles that were formed during AlCl thermohydrolysis.
3
CONCLUSIONS
Our results demonstrate that PALS makes it possible
to reveal fine details of the structure of amorphous alu-
mina, in particular to distinguish between the structures
of fine-particle amorphous alumina and amorphous
tubes. At the same time, the annihilation characteristics
INORGANIC MATERIALS Vol. 41 No. 3 2005

2
60
SHANTAROVICH et al.
1
0. Berdonosov, S.S., Baronov, S.B., Kuz’micheva, Yu.V., 16. Weber, M.H and Lynn, K.G, Positron Porosimetry,
et al., Hollow Alumina Macrotubes, Neorg. Mater.,
001, vol. 37, no. 10, pp. 1219–1223 [Inorg. Mater.
Engl. Transl.), vol. 37, no. 10, pp. 1037–1040].
Positron and Positronium Chemistry, Jean, Y.C. et al.,
2
(
eds., New York: World Scientific, 2003, p. 174.
1
7. Waran, K.V., Cheng, K.L., and Jean,Y.C., Application of
Positron Annihilation To Study the Surface Properties of
Porous Resins, J. Phys. Chem., 1984, vol. 88, no. 12,
p. 2465.
8. Shantarovich, V.P., Novikov,Yu.A., and Azamatova, Z.K.,
Positron Annihilation Study of Free Volume Elements in
Polymeric Gas-Separation Membranes, Fiz. Tverd. Tela
(S.-Peterburg), 1998, vol. 40, no. 1, pp. 164–167.
1
1. Berdonosov, S.S., Baronov, S.B., Kuz’micheva, Yu.V.,
et al., Hollow Macrotubes: A New, Elegantly Textured
Form ofAmorphousAlumina, Vestn. Mosk. Univ., Ser. 2:
Khim., 2002, vol. 43, no. 1, pp. 64–67.
1
1
2. Berdonosov, S.S., Melikhov, I.V., Baronov, S.B., et al.,
Thermohydrolysis Route to Tubular Alumina, Dokl.
Akad. Nauk, 2002, vol. 383, no. 2, pp. 211–214.
1
3. Baronov, S.B., Berdonosov, S.S., Kuz’micheva, Yu.V., 19. Brandt, W. and Paulin, R., Positron Diffusion in Solids,
and Melikhov, I.V., Self-organization of Alumina: From
Phys. Rev. Lett., 1968, vol. 21, p. 193.
Nanoparticles to Tubular Forms, Izv. Akad. Nauk, Ser. 20. Perkal, M.B. and Walters, W.B., Positron Annihilation in
Fiz., 2003, vol. 67, no. 7, pp. 912–914.
4. Handbuch der präparativen anorganischen Chemie in
Synthetic Zeolites 4A and 13X, J. Chem. Phys., 1970,
vol. 53, no. 1, pp. 190–198.
1
1
drei Bänden, von Brauer, G., Ed., Stuttgart: Ferdinand 21. Goldanskii, V.I., Mokrushin, A.D., Tatur, A.O., and
Enke, 1978, 3rd ed. Translated under the title Rukovod-
stvo po neorganicheskomu sintezu v 6 tomakh, Moscow:
Mir, 1985, vol. 3.
Shantarovich, V.P., Positronium–Gas Interaction in
Pores of Silica Gel, Kinet. Katal., 1972, vol. 13, no. 4,
p. 961.
5. Nieminen, R.M and Manninen, M.J, Positron in Imper- 22. Greenberger,A., Mills,A.P., Thompson,A., and Berko, S.,
fect Solids: Theory, Positron in Solids, Hautojarvi, P.,
Ed., Berlin: Springer, 1979, p. 148.
Evidence for Positron-like Bloch States in Quartz Single
Crystals, Phys. Rev. Lett., 1970, vol. 32A, no. 1, p. 72.
INORGANIC MATERIALS Vol. 41 No. 3 2005