HAL Id: jpa-00221300
https://hal.archives-ouvertes.fr/jpa-00221300
Submitted on 1 Jan 1981
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
PRESSURE DEPENDENCE OF IMPURITY INDUCED RAMAN SCATTERING SPECTRA I N
ZnS CRYSTALS
M. Zigone, M. Vandevyver, D. Talwar
To cite this version:
M. Zigone, M. Vandevyver, D. Talwar. PRESSURE DEPENDENCE OF IMPURITY INDUCED RAMAN SCATTERING SPECTRA I N ZnS CRYSTALS. Journal de Physique Colloques, 1981, 42 (C6), pp.C6-743-C6-745. �10.1051/jphyscol:19816218�. �jpa-00221300�
JOURNAL DE PHYSIQUE
CoZZoque C6, suppze'ment au n o 12, Tome 42, acembre 1981 page C6-743
PRESSURE DEPENDENCE OF I M P U R I T Y INDUCED RAMAN SCATTERING SPECTRA I N Z n S CRYSTALS
M. Zigone, M. vandevyverYand D .I$. ~alwar*
Laboratoire de Physique des Solides, associe' au C.N.R.S., Universite' Pierre e t Marie Curie, 4, Place Jussieu, 75230 Paris Cedes 05, France
"c. E.A., C.E.N.S., BoCte PostaZe n o 2, 91 190 Gif-,sup-Yvette, France
Abstract. - Raman scattering from zinc sulphide crystals containing a transi- tion metal substitutional impurity (Mn and Co) is reported at different pressures up to 40 kbar. Numerical calculations of these impurity induced Ra- man intensities are also made on the basis of Green's function theory. It is shown that the features observed in the Raman spectra of the above crystals may be interpreted as a series of resonant modes due to the weakening of the impurity-sulfur bonds by about 20 %.
1. Introduction. - When impurity atoms are introduced into a crystal, new phonons are induced which were either inactive or inexistent in the perfect crystal. There may arise two possibilities that a) all the modified modes lie within the band mode region, or b) some new modes occur at frequencies in between or greater than the bands of the allowed frequencies of the host system.
In the recent years, these impurity induced modes have intensively been investigated in numerous crystals using both infrared (i.r.) absorption and Raman scattering methods1. Most of the works have concerned isolated or complex defects in crystals at atmospheric pressure. Rather sparse are the studies of the influence of a rever- sible perturbation as hydrostatic pressure on the defect modes.
The present communication is devoted to the effect of hydrostatic pressure up to M 40 kbar on the Raman spectra of ZnS containing transition-metal atomic-substitu- tions (Mn and Co). It is shown that the pressure dependence of the vibrational modes is a useful tool in the assignement of the impurity-induced modes.
A calculation of the impurity induced first-order Raman scattering (IFOR) is also performed here for the above crystals, at 40 kbar pressure. As a result, our calculation reproduces very well the experimental results by assuming a weakening of ~ 2 % 0of the impurity-host bonding.
2. Experimental. - The Raman spectra were measured using a "Spectra Physics" F i r + laser and a "Coderg" double monochromator. The experimental resolution was usually of W 1 cm-l. The high pressure cell was a gasketed diamond-anvil cell. The pressure determination is accurate to f 1 kbar at 40 kbar.
The effect of hydrostatic pressure on the impurity induced Raman spectra has been studied for sS:Mn and ES:Co systems. The results are shown in figs. 1 and 2 for ZnS:Mn only. The measurements have been performed up to* 40 kbar, at room tempera-
-
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19816218
C6-744 JOURNAL DE PHYSIQUE
hypothesis of"true defect modes" cannot be
excluded. First, the phonon density shows a Fig. 1. Pressure denpendence of phonon frequencies in z S : M n ( 3 %) .
narrow gap in the optical region which - The notation 1, 2, 3 correspond to
L'
40
ture. The important result concerns the defect phonons, all of them exhibit the same blue shift - 370
under pressure application without any other g -
modifications. Moreover, the blue shift is very 350
similar to that of the LO (T) phonon. b
0
The frequencies of peaks found in the defect $ 330
spectra (see Table 1) fall into the range where the one-phonon density of states is high or 2 , 4 ' 310
rapidly varying .
Then, we might conclude that we do see the IFOR
290
scattering which is due to the relaxation of the wave-vector conservation rule in the
presence of the defect ("defect' activated 270
. .
suggests the possibility of gap modesL. the defect phonon peaks that are listed in Table 1.
Secondly, resonant modes characterized by a
Pressure (kbar) phonon density"). Nevertheless, the alternative
I I I
- -
&S:Mn , 0
- LO 0 0 -
4 o 0
c+
A . 4 7'
3 A A
1 . ?
.- ' -
2 .*+ - "
4c
- ( & ' + '
.m . :I
1 + m m
- i 1 00 P :
0
- 0
T," o 9
- P -
0
I I I
10 20 30
strong enhanc-ent of amplitude of vibration in the vicinity of the defect, could also be responsible for the observed defect spectra. We expect that, upon pressure application, the Raman spectrum of the "defect activated phonon density" should reflect the variations with pressure of the dispersion curves at zone edges, and thns present a modification of its structure because each of the high symmetry points of the Brillouin zone has a different pressure dependence coef ficient4. We have seen that, in fact, it is not what we have observed in the experiments under
TABLE 1
-
Comparison of the experimental pcrcccr crystal and impurity frequencies vith tllase o t the modes prcdictod by ~alcularion at tuo values of tho prersurc : 1 bar and 4 0 kbar. The measured and e a l c u l ~ ~ e d frequency shifts A 0 between 1 bar and 4 0 kbar are also indicated. All the frequencies are ~ i v e n in em-'. A numerical agreement of frequencies a t 4 0 kbar is achieved for the same value of the defect parameter L(P) I (A'-A(P))IA(P) that has been found ac armspheric pressure. The calculation reproduces also very well the measured frequency shifts do .
1 bar 4 0 kbar I A o = ( 1 ~ ~ - 0 ~
calc.
+ 1 9 4 0 kbar
I
no = o 40-0,
exp.
+ 3 1
1 bar
calc.
for t=+O.ZO 2 7 4
exp.
3 0 5
calc.
+ 3 1 exp.
2 7 4 features
(see figs.
4 and 5 ) TO(r) +
calc.
tor t=+0.20 3 0 5
calc.
for t=+0.10 2 7 4 exp.
2 7 4
compression (c. f . Figs. 1 and 2) .
This behaviour under pressure is an argument in favor of our second hypothesis of "true defect modes".
3. Calculation. - In this section we attempt to calculate the IFOR spectra due to the above impurities on the basis of the Green's function technique, by using a rigid ion model of 11 parameters (RiM 11) for description of the perfect crystals2. Among the possi- ble one-parameter defect models, we consider the one proposed by Grimm et a1. 5. In this model, the defect can be described by a single dimensionless parameter t
as : t = - Af/f. Details on numerical computations of 250 330 350 the impurity induced Raman intensity are given in ref. wave number (cm-1) 3. In an earlier paper4 we have reported a lattice
Fig. 2. Comparison between dynamical study of zinc-chalcogenides under compres- the measured and calculated
Raman intensities of zS:Mn sion. Therefore, a set of model parameters is avail-
(3 %) at 40 kbar pressure.
able at any pressure, and this allow us to calculate (a) represents the experimen- the IFOR scattering at high pressure. tal spectrum obtained at 40
kbar ; whereas in b) t The calculation of the IFOR spectra of the above calculated sum A +E+F " is
plotted. 1 2
crystals at atmospheric pressure has already been
given in a previous work3. We have shown that the calculated results provide an excellent agreement with the experimental structure for the value t = 0.2 of the defect parameter, i.e. a weakening of the impurity-host bonding by a factor 0.8.
This calculation predicts also no gap modes for the- selected value of t and in the frequency range under study. Therefore, the only "true defect modes" which could be responsible for the observed structure are the resonant modes.
In order to clarify this question furthermore, we have calculated the IFOR spectra from Mn and Co impurities at 40 kbar ; the shift of defect modes with pressure is in a very good agreement with the one observed experimentally (cf. fig. 2 and Table 1) for the value of t determined in the calculation at 1 atm. pressure' : t W 0 . 2 . This constitutes a supplementary test for the chosen models of the perfect lattice and of the defect.
References
1. For review see e.g. A.S. Barker Jr. and A.J. Sievers, Rev. Mod. Phys. 47, suppl.
no 2 (1975).
2. K. Kunc, Ann. Phys. (Paris) 8, 319 (1973).
3. D.N. Talwar, M. Vandevyver and M. Zigone, J. Phys. C 13, 3775 (1980) ; Phys. R w . B23, - 1743 (1981) ; M. Zigone, M. Vandevyver and D.N. Talwar, to appear in Phys.
Rev. B.
4. D.N. Talwar, M. Vandevyver, K. Kunc and M. Zigone, to appear in Phys. Rev. B, (July issue).
5. A . Grimm, A.A. Maradudin, I.P. Ipatova, and A.V. Subashiev, J. Phys. Chem.
Solids 33, 775 (1972).
