Low-frequency optical phonons in silica

t

Solid State Communications, Vol.34, pp.305—308. Pergamon Press Ltd. 1980. Printed in Great Britain.

LOW—FREQUENCY OPTICAL PHONONS IN SILICA Felipe R. Barbosa and Ramakant Srivastava Instituto de F~sica “Gleb Wataghin”,

Universidade Estadual de Campinas

Campinas l3lOO—S.P._Brasil (Received 27 August 1979; and in revised form 22 February 1980 by A. R. Verma) Measurements under

of low_frequency

longitudinal tensions

Raman spectra of silica

fibers

as a function of temperature have

been correlated with the measurements

of FIR spectra of bulk

silica in reflection and transmission. The results indicate existence of an LO—TO pair of a low—frequency optical phonon branch

in amorphous silica.

1. Introduction The nature silica

and the spectra reported here are for numeri—

of collective vibrations

in

cal aperture NA

and related glasses hss been subject

of study of many authors tigate

the extent

Efforts to inves—

~.

0.15. Fibers of 50—80 meter

length and attenuation of about 50 db/km at 5200

to which a phonon—like des—

were used. Measurements

were carried Out

in forward

cription of the excitations

in glasses is ap

and backward scattering configurations.

propriate have not yielded,

so far, any sa—

temperature-dependence

tisfactory model for the explanation presence

of all

sorption peaks. far—infrared region

the Raman

of the

scattering

and infrared ab

We report here a Raman

study of silica

in 10—150

inside

and

l~ a small

cm~

fiber was

in an attempt to understand the ori_

configurati3n,

a dewar

For

studies, we used forward with the fiber drum

filled with liquid nitrogen. On—

fraction of

the total length of the

left outside

the container for opti_

cal coupling. The spectra obtained at 295 K and

gin of a broad maximum in the region of 50—60 1 in Raman and infrared data. This maximum cm 2,3 has been attributed to acoustic phonons 4,5-. optical phonons , phonon assisted tranai— tions and/or tunneling between equivalent si—

77 K are shown in Figure 1. Measurements on fi— bers under longitudinal tension required back_ ward—scattering configuration, and smaller fi—

tea 6 and to an essentially

of signal—to—noise

effect

7,8

bers (2.5 meters) had to be employed due to space limitations, with consequent deterioration

thermal factor

On the other hand,

observation

of longitudinal in Raman

optical vibrations in glasses 8,9 and i,r. data suggests that glas—

sea appear The

to present

new results: measurement

tra in silica optical fibers nal

Of 295kgf/cm2)

mechanical

flectivity and transmiaaivity in the 33—120

two re—

pure fused quartz

sory.

of re

in bulk silica

spectra of silica fibers

T

Ar

+

ion laser radiation,

to

tem.

Fibers

and D.C.

(l—R)2 —

e —ad

R1 e~

where R and T are frequency—dependent

reflecti—

vity and transmissivity, respectively, and a is the thickness of the sample. Transmission data (Ge doped

were taken on a 1,2mm thick

sample. The data are

shown in Figure 3 and the calculated

a Spex

absorption

coefficient a is plotted in Figure 4.

14018 double monochromator, a refrigerated RCA 31034 photomultiplier

=

1

core) were obtained with approximately 200 my of 5145

a was calculated

using the relation 10

an optical phonon in silica. 2. Experimental and Results Raman

equipped with FIR acces—

Absorption coefficient

cm1 region. Our data indicate

existence of an LO—TO pair corresponding

streSs

2.

sample was used in a Perkin—

Elmer 180 Spectrometer

longitudi—

tension and measurement

(longitudinal

are shown in Figure

lica samples. For reflectivity measurements~ a

of Raman spec._

under

The results obtained

The FIR spectra were measured on bulk si~

to have a long—range Coulomb force.

object of our study is

levant

ratio.

with applied tension of l2Og

3

detection aye—

Discussion

First we discuss

used were made in our laboratory,

peak at 50 cm1

305

the behavior of the Raman

as a function of temperature and

306

LOW—FREQUENCY OPTICAL PHONONS IN SILICA

tension applied. An analysis of the data in Fi gure 1 shows that the intensity in this peak Va— nec as (n+l) is consistent

on the Stokes side. This behavior with one—phonon process. Similar

Vol. 34, No. 5

More relevant

is the fact that when we

apply longitudinal

tensions approaching the me— 2 chanical strength of the fibers (—300 kgf/cm ), we do not notice any change in the shape of the low—frequency Raman scattering.

This insensitivi-

ty of the Raman data to applied pressure as shown in Figure

2 suggests that the peaks Stu-

died are not due equivalent

I

to tunnelling

sites, as

effects between 6 . Since

proposed befo’~

potencial sites are energy expected barriers to be between strongly two pressure equivalent de—

I\

>-

77K

I—

effects, our pendent, if any, results must show be limited that the to tunneling very—low frequency (-

~40-

-40~

‘0.~

5

So

—

S.

I—

~

U) S

30

-

U, S.

-

30

N

~20

—20~

-

10

~

-

30

—

40

50

60

70

80

90

100

110

120

(crri’l Figure

3

—

Room temperature IR reflectivity and tranamissivity

spec-

tra of bulk silica.

ao

I j

1.5-

I Error ‘0 x LO

-

8

.5.

.5

-

0 0

I

I

50

100 w

Figure 4

—

(cn’1

Frequency—dependence in bulk silica rature.

150

1) of a/u2

at room tempe-

10

5

308

LOW—FREQUENCY OPTICAL PHONONS IN SILICA 4.

Conclusion

quency mode of vibration in the theoretical

We interpret our results in the spirit of suggestions brought up by ner et al 8,9

Vol. 34, No. 5

model of Bell & Dean 1,6

the work of Galee—

the

It appears that the two peaks

limited number

model

.

is certainly due

(600)

We definitely

believe that it is neces—

are caused by splitting of a phonon

in an LO—

sary to include some long—range

TO pair as a result of a long—range

Coulomb

ses

force; there

such a long—range is a strong

molecules which

force implies

correlation between

that

of Si0

4. The lack of prediction

silica

Acknowledgement

—

Raman and

We acknowledge TELEBRAS S/A

and CNPq for financial support,

tetrahedra of a

order in glas—

for the purpose of explaining

infrared data.

causing some kind of long—range order

extends to several hundred

Katiyar

low—fre

for many helpul

and Dr. R.S.

discussions

during

the work.

References 1. 1. Simon, Modern vol. I.

Aspects of the Vitreous State,

Edited by .7.D. Mackenzie;

Spectroscopy, Marcel

J.Wong

Dekker, New York

Butterworths,

and C.A. Angell,

(1976), Chapter VII;

London

(1960),

Glass Structure by and references

therein. 2. A.J.Leadbetter,

J.Chem.Phys.

3. P.1. Wong and E.Whalley, 4,

P. Flubacher, Solids

A.L. Leadbetter,

C.Morlet, X,Gerbaux,

Paris 2”60,

4973

(1969).

J.A.Morrison,

(1970).

and G.P.Stoicheff; J.Phys. Chem.

Rev.

J. Phys. Chem.

8. F.L.Galeener

812, 2432

Acad.Sci.

(1970).

Structure and Conference

Solids, AlP, Williamsburg, Va.

9. F.L.Galeener

C.R.

(1975).

Solids3l, 415

and G.Lucovsky,

of Amorphous

D.Canal, F.Brehat and P.Stnimer,

(1965).

6. G.Winterling, Phys. M.Hass,

779

Disc. Faraday Soc. 50, 94

12, 53 (1959).

5. A.Hadni,

7.

51,

and C.Lucovaky, Rev.

Phys. Rev. 165,

Jr., Phys.

11.

R.H.Stolen,

Phys. Chem.

12.

J.F.Scott and

13.

J.B.Bates and A.S. Quiet, J.Chem.

14.

J.B.Bates,

R.W.Hendricks and L.B.Shaffer,

15.

E.Whalley,

J.Chem.

16.

R.J.Bell and P.

S.P.S.Porto,

917

(1976).

(1967).

11, 83 (1970).

Phys. Rev. 161, Phys.

Soc. Faraday Trans.

Dean, Disc.

Proceedings Excitations

(1976).

Lett. 37, 1474

10. A.S.Barker,

Glass.

to

of atoms used in the

903

(1967).

56, 1528

(2)

(1972).

J.Chem. Phys. 61, 4163 (1974). 68, 662

Faraday Soc. 50,

(1972).

55 (1970).