TIME RESOLVED REFLECTION MEASUREMENTS OF CuCl UNDER HIGH EXCITATION

(1)

HAL Id: jpa-00225064

https://hal.archives-ouvertes.fr/jpa-00225064

Submitted on 1 Jan 1985

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.

TIME RESOLVED REFLECTION MEASUREMENTS

OF CuCl UNDER HIGH EXCITATION

M. Frindi, F. Tomasini, B. Hönerlage, R. Lévy, J. Grun

To cite this version:

(2)

TIME RESOLVED REFLECTION MEASUREMENTS OF CuCl UNDER HIGH EXCITATION

M. F r i n d i , F. T o m a s i n i , B. H o n e r l a g e , R. Levy and J . B . Grun

Laboratoire de Speotrosoopie et d'Optique du Corps Solide , 5, rue de l'Université, F-67084 Strasbourg Cedex, Franoe

Résumé - La r é f l e c t i o n des é c h a n t i l l o n s de CuCl e x c i t é s p a r un f a i s c e a u pompe i n t e n s e au v o i s i n a g e de l a résonance e x c i t o n i q u e Z3, e s t é t u d i é e en u t i l i s a n t deux s o u r c e s l a s e r p u i s é . L ' é v o l u t i o n t e m p o r e l l e des s p e c t r e s montre que l e s p o l a r i t o n s e x c i t o n i -ques c r é é s en grande d e n s i t é a i n s i que l e s p r o c e s s u s de c o l l i s i o n s avec l e s phonons a c o u s t i q u e s t h e r m a l i s é s e t non t h e r m a l i s é s p r o -d u i s e n t -d ' i m p o r t a n t e s m o -d i f i c a t i o n s -des s p e c t r e s -de r é f l e c t i o n .

Abstract - We study the reflection of CuCl samples, excited by an intense pump beam near the Z3 excitonic resonance, using a pump and probe technique. Its time evolution leads us to conclude that the density of excitonic polaritons as well as scattering proces-ses with thermalized and non-thermalized acoustical phonons lead to important modifications of the spectra.

I. INTRODUCTION

CuCl has T<i point group symmetry and a direct gap at the center of the Brillouin zone. The two uppermost valence bands have T7 and Ts symme-try, and the lowest conduction band has T6 symmetry. This band structu-re gives rise to two exciton series Z3 and Z12, their ground states ha-ve the symmetries (?21 ?5) and {V3, ?4, T5) , respectively. They are split by different exchange interactions. In both cases transverse ex-citons with Ts symmetry are dipole-active. Two Z3 exex-citons couple toge-ther to form a biexciton with Pi symmetry in its ground state.

In this work, we report some optical non linearities observed near the Z3 excitonic resonance by reflection measurements on CuCl. The sam-ple is excited by an intense pump laser, whose photon energy was kept close to the Z3 resonance, too.

II. EXPERIMENTAL SET-UP

Figure 1 shows the experimental set-up. A pulsed excimer laser (EMG 101 Lambda Physik) pumps simultaneously two dye lasers. The first one, cal-led pump beam, is a tunable dye laser having a grating working at gra-zing incidence. The laser emission has a spectral width of 0.05 meV, and the pulses have a total duration of about 10ns. They are asymmetric with a half-width of 5ns. The second laser, called test beam, has a spectrally broad emission (20 mV FMWH). Its temporal shape is also asym-metric with a half-width of 3ns. Both beams pass through diaphragms, a variable density filter, and are linearly polarized before being focu-sed onto the crystal surface. The test beam is focufocu-sed into a spot of 80 pm diameter inside the spot of the pump beam, having a diameter of 100 pm. The intensity of the pump beam can be varied up to l|Jax=60 MW/

cm2. The test beam passes through a variable optic delay, its

intensi-ty is kept constant at 200 KW/cm2.

The light of the test beam, reflected normally by the sample, is fo-cused onto the entrance slit of a 3/4m spectrograph, and recorded by an OSA system (WP3 BM Spectronik). Different kinds of CuCl samples were studied : cleaved samples with prismatic form in order to avoid the re-flection from the rear surface, and crystal platelets with different

+Associe au C.N.R.S. : UA 232

(3)

C7-216 JOURNAL DE PHYSIQUE

thicknesses below 40 pm. All measurements havekbeen performed at 1.9 K.

111. EXPERIMENTAL RESULTS

111 1)

-

Reflection spectra at tempo- c

ral-coi~n_cid_esce-of

f t : ~ t t ~ ~ C a ~ ~ ~ e - ~ ~ L -

se

.

FIgure 2a shows the reflection spectrum of a test beam using a prismatic cleaved

sample (I = 0). Then, the studied surface is excfted simultaneously by the pump beam ( I ~ / = I0.45) at different pho- ~ ~ ~ ~ ton energies, which are indicated by arrows on Fig.2 (b to h)

.

We observe a strong decrease (70%) of the reflection amplitude (Fig.ze)

,

and a shift of the maximum of the reflection towards higher enerqies. We notice also the appearance

!sample

Q

C r y o s t a t

NF L SPE

-~-6b%

of a-new reflection structure -(indicated

Fig.l

-

Experimental set-up

by double arrows)above the energy of the

for reflection experiment. longitudinal exciton.

The dotted curves in Fig.2 are obtained from a theoretical model taking into

account the two exciton~ogcillators Z3 and 212. The dielectric function

E (w) is then given2by 11i:

E ~ 2

-

E ~ l

-

+

E E ~

-

~ $ 2

E E ~

-

~ $ 2

rX

E + ~

-

E $ ~ E;~-TI~W~-~TI~WI'~ I

EG2

-

E$l E$1-h2~2-ih w

1

where E T ~ (Ip)

,

EL^

(Ip) and ET2 (Ip)

,

EL2 (Ip) are the energies of the transverse and longitudinal Z3 and 212 excitons, respectively. rx(Ip) and Txl(I ) are their damping constants. All these quantities may de- pend on tKe intensity Ip of the pump beam. Since spatial dispersion is not included explicitly in the model, the dampings account also for its influence on the reflection coefficient R.

At Ip = 0, the following parameters are used for CuCl :

- -

ET2(0) = ET1(0)

+

6 6 . 5 ~ 1 0 - ~ eV

[J.c.

MERLE, Private Communication] EL2(0) = ET2 (0)

+

8 8 ~ 1 0 - ~ eV

R = 0.16 at 3.184 eV, r;(O) = 10 meV 13

1

A good agreement between theoretical and experimental spectrum (Fig. 2a) is obtained for ~b = 3.8 and T,(O) = 1.5 meV. The reflection spectra of excited samples are then fitted by adjusting the parameters E ~ ~ ! I ~ ) and Tx(Ip). Agreement is namely obtained for the spectral po- sltlon and the amplitude of the reflection maximum as well as for the variation of the reflection coefficient on its low energy side. 111 2)

-

Time dependence of t h e - ~ e f L e _ ~ Z i o _ n _ - s ~ e c t r a

Fisure 3 represents the reflection spectra,at different time delays of the test puise with respect to the p;mp pulse. The intensity and photdn energy of the pump beam are fixed (I / I ~ - 0.45 %wp = 3.2118 eV)

.

The fit of these spectra is made is $iscussed above. In Fig.4, we give the variation of ATx(t) = Tx(t)

-

Tx(Ie = 0) as a function of the time delay t. Dots (circles) refer to a pho on energy hwp = 3.2062 eV (hwp)=3.2118 eV).Two different behaviours are observed. A rapid decrease of ATx occurs within the temporal overlap of the laser pulses. Af- terwards, ATx increases again and then decreases very slowly. The variation of A E T ~ (t) = E T ~ (t)

-

E T ~ (Ip = 0)

,

not presented here, shows approximately the same behaviour as AT,.

(4)

pulse. The relative magnitude of

bp

= 3.2118 eV. the curves are indicated. E T ~ and

ELI are given at Ip = 0. Fig.2a: Ip = 0. Fig.2 b to h : Ip = 27 ~ W / c m ~ .

ting the theoretical reflection spectra from the experimental ones of Fig.3. We notice that the new structure, situated above the energy of the longitudinal exciton, is observed only during the overlap of pump and test pulse.

These measurements enable us to specify two points : the sample has not recovered its initial state within a time delay of 60ns. This time is much longer than the lifetime of electronic excitations; therefore, they cannot be responsible for these effects. On the contrary, the new structure, visible only during the overlap of the pump and test pulse, can be attributed to this type of quasi-particles.

111 3) C _ c ~ & e m e n t a r y - ~ e a s u r e m e n t s

(5)

C7-218 JOURNAL DE PHYSIQUE

Fig.4

-

Changes of the damping exciton Arx as function of the Ip = 27 MW/C~'

of the delay.

Fig.5

-

Shape of the new reflection structure at a time delay of 0, 5 , 10ns.

A complementary investigation by means of back-scattered Hyper-Ra- man diffusion (HRS)[2], enables us to measure the energies of the longitudinal and transverse Z 3 exciton. They show no variation of these energies. It is important to notice that HRS takes place in the crystal volume. From these measurements, we conclude that the new structure is a pure surface effect. The shift of the maximum of reflection is effective only within a few pm behind the excited surface of the sample, but the damping of excitonic resonance is felt deeper inside the crystal volume.

IV. DISCUSSION

In CuC1, the maximum of the excitonic reflection shifts towards higher energies, and its amplitude decreases when the temperature of the sample is raised [ 4 ] . In our measurements, however, these two ef fects cannot be explained simultaneously by a local heating of the sample due to the pump beam. Indeed, the maximum shift of the exciton energy shown in Fig.2e would correspond to an increase of the lattice temperature of about 25'K. At this tem erature, however, the reflection amplitude is only diminished by 10% T l ]

,

whereas our spectra show a decrease of 70%. Since the heating is not sufficient to explain the high damping of the excitonic resonance observed, other effects have to be considered. One contribution can be due to collisions between excitons and biexcitons

(6)

ce effect on the excitonic level. These carriers would then relax by emitting cascades of L.O. phonons [7]. The latter having small wave- vectors can only recombine by emitting two acoustical phonons, with large wave-vector. These acoustical phonons cannot easily thermalize with conservation of energy and momentum. Having a large velocity (2 to 3.10' ms-'1, they will fill a large region of the crystal. The pump beam can therefore create a large density of acoustical phonons which are out of thermal equilibrium, and have a much longer lifetime than electronic excitations. This can explain the damping of the excitonic reflection [8] observed after the excitation. The shift of the excitonic line can be attributed to a local heating of the sample by the pump beam, which corresponds to a large number of thermalized acoustical phonons. The new reflection structure, visible only during the overlap of the two pulses, may be explained in the following way : the pump beam creates excitons which thermalize on the lower polariton branch which has a spatial dispersion since the exciton effective mass is finite. The excitons induce a polarization on this branch which depends on the exciton density, and can become important even above EL(I~). If we as- sume Pekar's ABC to be valid, the polarization of the lower polariton branch changes that of the upper branch near the crystal surface. Its polarization is then observed in the reflection measurement, giving rise to the new structure during the lifetime of the excitons.

ACKNOWLEDGEMENTS

The authors are grateful to J.Y. Bigot, H. Haug, G. Mahler, J.C. Merle and M. Sieskind for helpful discussions and their interest in our work. It has been supported by a contract with the 'Ministsre des PTT" of France, 'Direction G&n&rale des T&l&communicationsl', "Direction des Af- faires Industrielles et Internationales".

It has also been carried out in the framework of an operation laun- ched by the Commission of the European Communities under the experimental phase of the European Community Stimulation Action (1983-1985). REFERENCES

/1/ F. Lagois,-Phys. Rev B 16, 1699 (1977).

/2/ B. Hbnerlage, A. Bivas and Vu Duy Phach, Phys. Rev. Letters

41,

49 (1978).

/3/ J. Ringeissen, S. Nikitine, S. Lewonczuk and C. Jung in Polari-

tons,

E. Burstein and F. De Martini eds., Pergamon Press, P.203 (1973).

/4/ Y. Kaifu and T. Komatsu, Phys. Stat. Sol. (b) 48, K 125 (1971)

.

/5/ A. Bivas, Vu Duy Phach, B. Henerlage and J.B.G~u~, Phys. Stat.

Sol. (b)E, 235 (1977)

.

/6/ R. Lsvy, C. Klingshirn, E. Ostertag, Vu Duy Phach and J.B. Grun, Phys. Stat. Sol. (b)E, 381 (1976).

/7/ J. Ringeissen, A. Coret and S. Nikitine, in Localized Excita

-

tions in Solids, Plenum Press, P. 27 (1968).

A. Coret, J. Rinseissen and S. Nikitine, idem, P. 306.