Large potential of Sb100-xTex films for optical storage

Article abstract of DOI:10.1016/S0025-5408(99)00004-5

As-grown microcrystalline stoichiometric Sb₄₀Te₆₀ films were examined by heat-treatment and laser-irradiation experiments for their potential use in optical storage. Compositional, structural, and optical studies were carried out for films heat-treated at various temperatures in the range 0-260 °C. They revealed a stable phase (Sb₆₀Te₄₀) in the range 110-185 °C and a monotonic decrease of Te below and above this range. The irradiation experiment showed the possibility of both reversible and irreversible phase changes depending on the laser power. Comparing the two experiments, we believe the phase responsible for reversible changes is Sb_100-xTe_x (x = 38-45). The irreversible changes were found to be due to large deviation in this stoichiometry.

Full text of DOI:10.1016/S0025-5408(99)00004-5

Materials Research Bulletin, Vol. 34, No. 2, pp. 203–216, 1999
Copyright © 1999 Elsevier Science Ltd
Printed in the USA. All rights reserved
0
025-5408/99/$–see front matter
PII S0025-5408(99)00004-5
LARGE POTENTIAL OF Sb100؊xTe FILMS FOR OPTICAL STORAGE
x
P. Arun and A.G. Vedeshwar*
Department of Physics and Astrophysics, University of Delhi, Delhi 110 007, India
(Refereed)
(Received January 9, 1998; Accepted June 5, 1998)
ABSTRACT
As-grown microcrystalline stoichiometric Sb Te films were examined by
4
0
60
heat-treatment and laser-irradiation experiments for their potential use in
optical storage. Compositional, structural, and optical studies were carried out
for films heat-treated at various temperatures in the range 0–260°C. They
revealed a stable phase (Sb Te ) in the range 110–185°C and a monotonic
6
0
40
decrease of Te below and above this range. The irradiation experiment
showed the possibility of both reversible and irreversible phase changes
depending on the laser power. Comparing the two experiments, we believe the
phase responsible for reversible changes is Sb100ϪxTe (x ϭ 38–45). The
x
irreversible changes were found to be due to large deviation in this stoichi-
ometry. © 1999 Elsevier Science Ltd
KEYWORDS: A. chalcogenides, A. thin films, B. vapor deposition, C. pho-
toelectron spectroscopy, D. optical properties
INTRODUCTION
The search for materials suitable for optical data storage has increased over the past decade
or so. The initial effort was to store information in binary form by photo-thermally created
holes in metal or alloy films, but the focus was soon shifted to laser-induced phase change
process. Chalcogenide compound films have been found to be a source for such purposes.
The write-once-and-read-many-times (WORM) as well as the erasable kind of storage has
been demonstrated [1]. Most of the erasable storage media developed are chalcogenide films
containing Te. Among these, the Sb–Te system has been investigated to a lesser extent [2,3].
*
To whom correspondence should be addressed.
2
03

2
04
P. ARUN and A.G. VEDESHWAR
Vol. 34, No. 2
In ref. 2, the WORM kind of storage was demonstrated on thermally evaporated Sb Te films
2
3
based on amorphous-to-crystalline phase change. Two other works have demonstrated
erasing based on amorphous-to-crystalline transformation on sputtered Sb Te [4] and Sb–Te
2
3
alloy [3] films. Therefore, we have investigated thermally evaporated Sb Te films by two
2
3
ϩ
different approaches, viz., heat treatment and irradiation using cw Ar laser, to have better
insight into the basic process responsible for their potential as a storage medium.
EXPERIMENTAL
Sb Te films were grown on glass substrates at room temperature, by thermal evaporation at
2
–
3
6
1
0
Torr by placing crystalline starting material in a molybdenum boat. The starting material
was 99.99% pure and was supplied by Aldrich (USA). The thickness and uniformity of the
films were measured using a Dektek IIA surface profiler. The glass substrates used were the
2
usual microscope slides, which were cut into small pieces (5 ϫ 5 mm in size) for various
measurements. The pieces were placed on a copper plate maintained at the desired temper-
ature for a few seconds (30–45 s). Such instantaneously heat-treated films were analyzed,
and in the following passages we report the results of structural, compositional, and optical
studies carried out on these samples. In the final section we compare these heat-treated films
with laser-irradiated films, in an effort to understand photo-thermal changes induced in
Sb Te films upon irradiation.
2
3
RESULTS AND DISCUSSION
The films were instantaneously heat-treated in air at various temperatures ranging from 60 to
60°C. The results are divided into two sections, namely, (a) results of heat treatment and (b)
2
results of laser irradiation. The temperatures mentioned in the first section below are those at
which the films were subjected to instantaneous heat treatment.
Results of Heat Treatment
Compositional analysis. The chemical compositions of the heat-treated films were analyzed
by electron spectroscopy for chemical analysis (ESCA) (Shimadzu ESCA 750). The as-
grown film was confirmed to be stoichiometric. Since heat treatment was done in air, we
suspect the gradual decomposition and subsequent oxidation of the film. Figure 1 shows Te
3
d3/2 and 3d5/2 peaks [5] at two different temperatures. A shift of 3 eV is evident between the
peaks of as-grown film and that treated at 220°C. This is due to change from Sb to oxygen
environment around Te [6]. Also, it was observed that the ratio of areas of the 3d3/2 and 3d5/2
peaks of Sb increased, accompanied by a shift of 2 eV as the function of treatment
temperature. The increase in the ratio of areas of these two peaks can be understood to be due
to the O1s peak (531 eV) overlapping with 3d5/2 peak of Sb [7]. Thus we can confirm the
incorporation of oxygen during treatment.
The ratio of Te (in bonding with Sb) and Sb decreased with treatment temperature, as
shown in Figure 2. A systematic study of the chemical compositions of the films treated at
different temperatures revealed the existence of different phases of TeO (x ϭ 1, 2, 3)
x
between treatment temperatures 180 and 220°C. This fact is further supported by the
broadening of Te peaks in the given treatment temperature range. Estimates of TeO, TeO2,
and TeO were made by deconvolution of these broadened peaks. However, films treated
3

Vol. 34, No. 2
ANTIMONY TELLURIUM FILMS
205
FIG. 1
X-ray photoelectron spectra for Te of 2300 Å thick Sb Te film: (a) as-grown (at room
2
3
temperature) and (b) heat-treated at 220°C.
beyond 254°C showed a sharp Te peak, suggesting the existence of TeO only. As can be
2
seen, at 260°C and above, Te/Sb was equal to zero. This suggests a complete oxidation of the
films by the formation of oxides of both Sb and Te. This was confirmed by X-ray diffraction
analysis, which is discussed next. The nature of the curve in Figure 2 is like that reported in
a similar study of Sb2S3 [8]. This suggests that the nature of the curve is dependent on the
kinetics of the antimony compounds’ reaction with oxygen. The variation of the Te/Sb ratio
with treatment temperature shows two plateaus: one up to 80°C and the other in the
temperature range 110–185°C. The first plateau suggests the gradual loss of Te from bonding
with Sb above 80°C and formation of a non-stoichiometric compound. The second plateau
shown is of interest. The mean composition of the films heat-treated between 110 and 185°C
is Sb Te and, from the boxed area in Figure 2, seems to be stable.
6
0
40
Energy dispersive X-ray analysis (EDAX) was performed (ISIS200 EDS, Oxford) on all
these samples, to confirm the chemical composition of the various films. Unlike ESCA,
which is essentially a surface analysis method, EDAX gives the composition of the bulk, due
to greater penetration depth. Figure 3 shows the spectra of (a) the as-grown film and (b) the
film heat-treated at 120°C. In this figure, for clarity, we have shown only two spectra. The
spectrum in (b) is the same for all of the samples heat-treated in the range 110–185°C,

2
06
P. ARUN and A.G. VEDESHWAR
Vol. 34, No. 2
FIG. 2
Tellurium content in the film as a function of treatment temperature as determined from
X-ray photoelectron spectra. The portion of the curve marked by the box shows the region
of constant composition (nominally Sb Te ).
6
0
40
confirming the constant composition shown in Figure 2. Since film thickness was less than
␮m (d ϭ 2350 Å), the spectra have various peaks (2–2.4 KeV) of Si (silicon) and S (sulfur)
1
contributed from the substrate. The composition of the films was computed by accompanying
software, based on the technique explained in ref. 9. The results of the EDAX analysis gave
the composition of the as-grown film as Sb Te . The composition of the heat-treated films
2
3
was given as Sb0.62Te0.38 (Sb Te ). This is essentially the constant composition of the films
3
2
treated in the range 110–185°C; hence, the ESCA and EDAX results are in good agreement.
In the temperature range bounded by the box in Figure 2, the change in composition was
negligible and, therefore, we can assume a constant film composition in this temperature
range. Similar observations were reported in ref. 4 for an even larger variation of Sb content
in the film. Therefore, we can expect a reversible transformation of either amorphous-to-
crystalline or crystalline-to-crystalline phases in this temperature range.
Structural analysis. The structural studies were carried out using a Philips PW1840 X-ray
diffractometer. Figures 4 and 5 show diffractograms of heat-treated Sb Te film thickness of
2
3

Vol. 34, No. 2
ANTIMONY TELLURIUM FILMS
207
FIG. 3
EDAX spectra of (a) as-grown Sb Te film and (b) film heat-treated at 120°C.
2
3
2
350 Å. The diffractogram in Figure 4a of the as-grown film shows no sharp peaks,
characterizing the film as amorphous, in agreement with previous reports [2]. TEM analysis
of the as-grown film shows the existence of microcrystallinity (see further discussion under
“
Optical Properties”). Therefore, we refer to the as-grown film as being amorphous. The
nature of diffractograms remains the same until the treatment temperature 108°C, at which
a single peak appears. However, well-resolved peaks appear only at 130°C. Diffractograms
(
b) to (f) correspond to crystalline phases of Sb100ϪxTe , which were indexed using a
x
program in Turbo Basic. All of the phases showed orthorhombic structures. The cell
dimensions (a ϭ 8.8, b ϭ 8.0, and c ϭ 4.6 Å) determined from diffractograms (b) to (f) are
almost the same, with a maximum variation of 12% at the extreme temperatures in the range
1
30–190°C. This supports the constant film composition in this temperature range, as
discussed earlier. The crystallization temperature T (temperature at which the peaks first
c
appear in the X-ray diffractogram) shows film thickness dependence, as shown in Figure 6.
The crystallization temperatures are in good agreement with previous reports [4,10].

2
08
P. ARUN and A.G. VEDESHWAR
Vol. 34, No. 2
FIG. 4
X-ray diffraction patterns of (a) as-grown 2300 Å thick Sb Te film and 2300 Å thick Sb Te
3
2
3
2
film heat-treated at (b) 130, (c) 146, (d) 158, and (e) 170°C.
We tried to identify the phases present in our samples by comparing their peaks as well as
their relative intensities with those given in standard ASTM cards. Diffractograms (g) and (h)
of Figure 5 are of interest. The majority of the peaks for sample (g), which was treated at
1
92°C, matched with data for Sb O , TeO , and Sb as given in ASTM 11-689, 20-1240, and
2 3 3

Vol. 34, No. 2
ANTIMONY TELLURIUM FILMS
209
FIG. 5
X-ray diffraction patterns of 2300 Å thick Sb Te film heat-treated at (f) 180, (g) 192, and
2
3
(h) 254°C.
3
5-732, respectively. The remaining peaks belong to the crystalline phase of Sb100ϪxTe ; for
x
clarity, we did not include their Miller indices in the figure. The relative concentrations,
which were determined by the relative intensities of the four phases, suggest that the amount
of Sb present was minute. Diffractogram (h) corresponds to the sample heat-treated at 254°C
and confirms the presence of Sb O and TeO , as given in ASTM 11-693 (TeO ). Thus it
2
3
2
2

2
10
P. ARUN and A.G. VEDESHWAR
Vol. 34, No. 2
FIG. 6
Temperature at which peaks first appear in X-ray diffractogram (T ) as a function of film
c
thickness.
seems that the oxidation process leads to the formation of TeO at lower temperatures,
followed by the formation of TeO and finally that of TeO at higher temperatures. Peaks of
3
2
TeO are, however, not seen in any of the diffractograms between (b) to (f). This suggests the
absence of crystalline phase of TeO at these temperatures, as was observed by Takenaya et
al. [11].
We used SEM to investigate the morphological changes with treatment temperatures. The
results are shown in Figure 7. Even though the grain structure is not seen in (a) or (b) of this figure,
X-ray diffraction showed well-resolved peaks, confirming the crystalline phase. A well-developed
grain structure can be seen in films treated above 150°C as seen in (c) and (d). However, the
morphology of the films remains the same in the range 150–190°C. This may be due to the
constant film composition and the same structure in this range, as discussed above.
Optical studies. We have studied the absorption spectra of various heat-treated films in the
UV-vis and near-IR range, using a Shimadzu 260 photospectrometer. The optical properties
show thickness dependence, in agreement with previous work [2]. The as-grown films
showed increasing reflectivity with film thickness. The variation of optical absorption with
treatment temperatures for three different thickness is illustrated in Figure 8. The region
marked by the broken lines refers to the temperature range 110–185°C discussed in earlier
analyses. There is more than 25% change in absorbance between the limits of this range. The
contrast is better for films of thickness greater than 1000 Å. It should be noted that the change
in reflectivity appears more pronounced in treated samples, which can even be seen by the
naked eye. One can easily distinguish the sample treated at different temperatures at an
interval of 20°C in this range, by visual examination. Since the optical contrast required for

Vol. 34, No. 2
ANTIMONY TELLURIUM FILMS
211
FIG. 7
Morphological changes with treatment temperatures as depicted by SEM for 2300 Å thick
Sb Te film treated at (a) 118, (b) 145, (c) 158, and (d) 170°C.
2
3
storage applications can be achieved in this temperature range, the film with composition
Sb Te (or Sb Te ) shows great potentiality for reversible phase change type of storage
6
0
40
3
2
applications. On the basis of all of the analyses together, we can say that the transition
between various crystalline phases of different reflectivity in the range 110–185°C, inducible
by a suitable laser, can be exploited for the erasable kind of storage applications. However,
the as-grown film can be used for the write-once-and-read-many-times kind of application
also, if it is heated to the irreversible region to obtain the desired optical contrast. The optical
band gap of the as-grown and treated films can be determined using absorbance data. The
absorption coefficient ␣ was calculated as a function of photon energy h␯, following standard
procedure [12,13]. The calculated ␣h␯ was fitted to a general relation for optical transitions
[12,14]:
n
␣
h␯ ϭ (h␯ Ϫ Eg)
(1)
where E is the optical gap of the material. The value of n determines the nature of E of the
g
g
material [14]. The best fit of our data to eq. 1 yields n ϭ 1/2, indicating the allowed direct
transition of the polycrystalline material. In the case of amorphous state, n ϭ 2. It should be
noted that microcrystallinity in the as-grown films was shown in TEM, but was not detectable
in X-ray diffraction. E shows thickness dependence and decreases almost exponentially with
g
thickness. E ϭ 2 eV for a 750 Å thick film decreases to 0.21 eV for a 3000 Å thick film.
g
Eg decreases linearly with treatment temperature up to 110°C, following a relation
E (T) ϭ E (0) Ϫ ␣T
(2)
g
g

2
12
P. ARUN and A.G. VEDESHWAR
Vol. 34, No. 2
FIG. 8
The optical absorbance as a function of heat-treatment temperatures ␭ ϭ 540 nm for (a) 750,
b) 1500, and (c) 2300 Å thick Sb Te films.
(
2
3
Ϫ3
where E (0) (ϭ 2 eV) is the value of as-grown film, ␣ (ϭ 7.5 ϫ 10 eV/°C) is the
g
temperature coefficient of E , and T is the treatment temperature for 750 Å thick film. E is
g
g
constant within error of 0.1 eV in the range 110–180°C. This trend is similar for thicker films
up to 3000 Å.
Results of laser irradiation. As-grown Sb Te films were irradiated using a cw Arϩ
2
3
laser (Coherent’s INNOVA 70-4) of ␭ ϭ 514 nm, having a beam diameter of 400 ␮m.
The effect of laser irradiation was studied as a function of film thickness, irradiation time,
and laser output power. Figure 9 displays SEM micrographs of laser-irradiated (30 s)
areas of as-grown 6700 Å thick films. Uniformly developed crystallites can be seen in a
spot of 178 ␮m diameter (Fig. 9a) where the laser power (p) used was 110 mW. The
enlarged view of the boundary region of the spot is shown in Figure 9b to highlight the
crystallization within the spot more clearly. On increasing the laser power, the spot
diameter increases, accompanied by radially varying grain size and morphology. This is
an indication of laser-induced photo-thermal process in which the laser acts as a heat
source, increasing the temperature at the site of irradiation. Therefore, it is necessary to
know the temperature rise and its profile at these spots. A precise value of temperature
rise upon laser irradiation can be calculated by numerically solving the inhomogeneous

Vol. 34, No. 2
ANTIMONY TELLURIUM FILMS
213
FIG. 9
SEM pictures of laser-irradiated as-grown 6700 Å thick Sb Te films at laser power of 110
2
3
mW (a and b), 220 mW (c), and 330 mW (d and e). Micrographs (b) and (e) show the
enlarged view of the boundary region of the spots shown in (a) and (d), respectively, while
(c) shows the central region of a spot not shown in the figure.
partial-differential heat equation [15]. The main difficulty is to incorporate into the
calculation the functional dependence of film parameters, such as thermal conductivity
(
␬), absorbance (A), or reflectance (R), on changing phases of the film as a function of
temperature, as realized in heat-treatment experiment. However, a qualitative estimate of
the temperature rise can be made using the equation [16]
Ϫ(r/ro)2 Ϯ(␣z)
AP(1 Ϫ R)e
e
Ϯ␬⌬t/c
Tr,z,t
ϭ
(1 Ϫ e
)
(3)
2
␲
r d␬⌬
o

2
14
P. ARUN and A.G. VEDESHWAR
Vol. 34, No. 2
where
5
.784
1
⌬
ϭ
ϩ
(4)
2
2
ro
d
However, it is more accurate for films of low thermal conductivity. Eq. 3 shows temperature
rise T is directly proportional to p and A, while it is inversely proportional to ␬.
The crystalline grains develop at T , which is film thickness dependent, as shown in Figure
c
5
6
. From this figure, T ϭ 60°C for a film of 6700 Å thickness and, hence, we assume T ϭ
0°C at the edge of the spot in Figures 9a and b. Therefore, no crystalline grains are seen
c
beyond the edge. A Gaussian profile of temperature rise gives a temperature at the center
T(0,z,t) ϭ 70°C. A temperature rise directly proportional to P would mean a temperature of
1
40 and 210°C at the center of spots irradiated at 220 mW (Fig. 9c) and 330 mW (Fig. 9d),
respectively. However, the temperature would be quite lower than that estimated above due
to a large ␬. Temperature rise will increase with laser power and therefore the temperature
outside the spot will also increase, due to large ␬, because of which crystalline grains develop
up to the region where T ϭ 60°C, as can be seen in Figures 9d and 9e. Large thermal
conductivity leads to melt-quenching when the laser is switched off [16] and might have led
to a crystalline-to-microcrystalline transition within an area of 150 m radius. Figure 9e shows
the enlarged boundary of this region and the grains present outside the boundary. These
grains have a morphology similar to that of those at the center of the spot irradiated with p ϭ
2
20 mW (Fig. 9c).
We varied the irradiation time for various laser output powers. For powers below p ϭ 280
mW, there was no irradiation time dependence and the results were similar to those shown
in Figure 9. Well-developed grains similar to those shown in Figure 7c or d were seen in films
irradiated for 15 s at laser power ranging from 280 to 415 mW, as shown in Figure 10a. When
we increased the irradiation time to greater than 15 s, the crystalline phase was transformed
to microcrystalline phase irrespective of laser power in the range 280–415 mW. This
indicates the possible photo-induced transformation of a certain photosensitive phase devel-
oped during the initial stages of irradiation (t Յ15 s). The similarity between the grain
structure shown in Figures 7c, 7d, and 10a leads us to believe that the phase developed
initially and transformed to that shown in Figures 6b and 9b–d is Sb Te . The observed
6
0
40
transformation of the as-grown microcrystalline phase to a crystalline phase of bigger grains
and back to microcrystalline phase suggests the possibility of reversibility. However, a film
of Sb Te composition has to be examined to resolve such a possibility and prove it to be
6
0
40
a useful erasable storage medium. Such new phases (i.e., Sb Se1Ϫx) suitable for reversible
x
storage have been reported for the Sb–Se system [17]. However, Sb Se has been found [18]
2
to be the most suitable composition, because of its short crystallization time [18].
The essential requirements of reversible phase change optical storage is that over a large
temperature range there is no chemical compositional change, as represented by a plateau
region in a graph such as that shown in Figure 2. The phase transition, which should bring
about an appreciable optical contrast, should lie on this plateau. Of the investigated and
reported Sb–C systems (C ϭ S, Se, and Te), only Se and Te fulfill these requirements.
However, the Sb–Te system allows for more control, since the transition temperature is
thickness dependent. In actual practice, for optical storage, a laser diameter of a few
micrometers and a time of irradiation of a few nanoseconds pulse are used. Although in our
experiment, we maintained a laser diameter and irradiation time far greater than those used

Vol. 34, No. 2
ANTIMONY TELLURIUM FILMS
215
FIG. 10
Surface morphology dependence on irradiation time for a 6700 Å thick Sb Te film. SEM
2
3
pictures of the central region of the spots created on the film irradiated at (a) 326 mW for 15 s,
b) 326 mW for 60 s, (c) 380 mW for 60 s, and (d) 412 mW for 60 s.
(
in actual practice, the changes occurring in both cases would be similar. By decreasing the
diameter for a given laser power, one can achieve higher temperature rise at the irradiated
site. Oxidation of the film will take place even for short pulses, such as 10–100 ns [19].
CONCLUSIONS
On the basis of both experiments, we find that Sb100ϪxTe films may be used for reversible
x
(for x ϭ 38–45) and irreversible (remaining nonstoichiometric phases) phase change optical
storage applications.
ACKNOWLEDGMENTS
We are thankful to Drs. N.C. Mehra and S. Shukla, USIC, Delhi University, for their
cooperation in the use of analytical techniques. The financial assistance of UGC, India, in the
form of SRF to P. Arun is also gratefully acknowledged.
REFERENCES
1
2
.
.
K. Schwartz, The Physics of Optical Recording, Springer, Berlin (1993).
K. Watanabe, N. Sato, and S. Miyaoka, J. Appl. Phys. 54, 1256 (1983).

2
16
P. ARUN and A.G. VEDESHWAR
Vol. 34, No. 2
3
4
5
.
.
.
S. Fujimori, S. Yagi, H. Yamazaki, and N. Funakoshi, J. Appl. Phys. 64, 1000 (1988).
T. Iijjima, T. Tanabe, and N. Funakoshi, Jpn. J. Appl. Phys. 28, L1985 (1985).
C.D. Wagner, W.M. Riggs, L.E. Davis, J.F. Moulder, and G.E. Muilenberg, Handbook of X-ray
Photo-electron Spectroscopy, Perkin-Elmer Corp., Physical Electronics Division (1978).
D. Briggs, Handbook of X-ray and Ultra-Violet Photoelectron Spectroscopy, Heyden, London
6
.
(1977).
7.
8.
9.
W.E. Morgan, W.J. Stec, and J.R. Van Wazer, Inorg. Chem. 12, 953 (1973).
P. Arun and A.G. Vedeshwar, J. Appl. Phys. 79, 4029 (1996).
J.I. Goldstein, D.E. Newbury, P. Echlin, D.C. Joy, C. Fiori, and E. Lifshin, Scanning Electron
Microscopy and X-ray Microanalysis, Plenum Press, New York (1984).
1
1
0. V. Damodara Das, N. Soundararajan, and M. Paltabi, J. Mater. Sci. 3522 (1987).
1. M. Takenaya, N. Yamada, K. Nishiuchi, N. Akahira, T. Ohta, S. Nakamura, and T. Yamashita,
J. Appl. Phys. 54, 5376 (1983).
1
1
2. K. Seeger, Semiconductor Physics, Springer-Verlag, New York (1973).
3. N.F. Mott and E.A. Davis, Electronic Processes in Non-Crystalline Materials, Clarendon, Oxford
(
1971).
4. A.H. Clark, Polycrystalline and Amorphous Thin Films and Devices, Academic Press, New York
1980).
5. H.S. Carlsaw and J.C. Jaeger, Conduction of Heat in Solids, Oxford University Press, Oxford
1954).
1
1
(
(
1
1
1
1
6. P. Arun and A.G. Vedeshwar, Physica B 229, 409 (1997).
7. R. Stradins, K. Shvarts, and J. Teteris, J. Non-Cryst. Solids 114, 79 (1989).
8. R. Barton, Ch.R. Davis, K. Rubin, and G. Lim, Appl. Phys. Lett. 48, 1255 (1986).
9. N. Rykalin, A. Uglov, I. Zuev, and A. Kokora, Laser and Electron Beam Material Processing
Handbook, Mir, Moscow (1988).