MÖSSBAUER AND CHANNELING EXPERIMENTS ON [MATH]Si AND [MATH]Si

(1)

HAL Id: jpa-00219658

https://hal.archives-ouvertes.fr/jpa-00219658

Submitted on 1 Jan 1980

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.

MÖSSBAUER AND CHANNELING EXPERIMENTS ON [MATH]Si AND [MATH]Si

G. Kemerink, D. Boerma, H. de Waard, J. de Wit, S. Drentje

To cite this version:

G. Kemerink, D. Boerma, H. de Waard, J. de Wit, S. Drentje. MÖSSBAUER AND CHANNELING

EXPERIMENTS ON [MATH]Si AND [MATH]Si. Journal de Physique Colloques, 1980, 41 (C1),

pp.C1-435-C1-438. �10.1051/jphyscol:19801170�. �jpa-00219658�

(2)

JOURNAL DE PHYSIQUE Colloque C1, supplkment au n ^O1, Tome 41, janvier 1980, page C1-435

M~SBAUER AND CHANNELING EXPERIMENTS ON ? e ~ i

AND

EmSi

G.J. Kemerink, D.O. Boerma, H. de Waard, J.C. de Wit and S.A. Drentje

Luboratoriwn voor AZgemene Natuurkunde, University of Groningen, the N e t h e r b d s .

Considerable e f f o r t is made to obtain an insight in the structural a d electronic p r o r O p r t i e s of ion inplanted elemental semiconductors. This research is strongly stimulated by the many appli- cations of semi-conductor devices. We report here

_.f d

on ~ s s b a u e r studies of '29m~eSi and 53SmSi, us- the 27.8 keV transition i n 1291 and the 103.2 keV transition i n 1 5 3 ~ ~ , r e s p e d i v e l ~ , and on chan-

r _3

neling experiments on 1 2 * ~ e s i and 1529n~i with a 2 MeV a-beam £ran the' Groningen Van de Graaff generator. In the bBssbauer expe.riments we used

c ~

and E u F ~ .

~ ~

)zH20 as absorber materials. Sowoe

~ I

ardl absorber were held a t 4.2 K. The W l a n t a t i o n s were generBLly done a t roan tanperatwe with an i q l a n t a t i o n en- of 100-115 keV. For the I6ssbauer and channeling wasumnents we applied similar S i single crystals and the same implan- tation and annealing conditions. Crystals with law doses could only be irnrestigated with the

~ s s b a u e r effect.

3

The s y s t m TeSi.- This system has been studied before by us /1,2/ and others /3/. The reason

for s M y i n g this system in mre detail was an apparent inconsistency in the interpretation of the I6ssbaue.r and the channeling results.

High dose.- A

<loo>

silicon crystal, m l a n t e d w i t h 1 x 1015 at/an2 and laser annealed with an

energy above the threshold f o r epitaxial re-

-

crystallization gives angular yield curves f o r the

<loo>,

<110> and <111> strings with minim of approximately 20% f o r Te and 4-7% f o r Si The curves for Te a r e slightly narrcrwed with respect w those

for Si. These r e s u l t s indicate that 85% of the Fe is on o r close t o substitutional l a t t i c e sites. A similar r e s u l t has been published by Foti et al.

141.

I6ssbauer a- of crystals with doses between 0.7 and 2 x

l o i 5

at/an2, after the sans annealing, can be f i t t e d quite w e l l with one can- p e n t with a large quadruple s p l i t t i n g

(q

= 498 (10) MHz, ismw s h i f t = -2.0 mn/s with respect t o (fig. l a ) . Also, _ifa f i t

Fig. 1': Mkksbauer s p c t n i m of a

<loo>

S i crystal w i t h a . 2 x 1015 Te at/cm2 after laser m m d i n g

with 1.2 J/C$

b. 5 x 1013 e at/an2 after laser - d i n g with 1.1 J/&

a.

¹^x 1013 Te at/an2 as -1anted.

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19801170

(3)

c1-436 JOURNAL DE PHYSIQUE

with m r e ccmpnents is attempted,, we always have to assme the presence of appreciable quadrapole splitting for a t least one of these canpnents. This r e s u l t can not be reconciled with a cubic for the Iodine nor for the implanted Te. Thus,we have to do with s i t e s where the

-1antedTe is close to substitutional, yet the cubic symnetry is strongly perturbed. The exact nature of this s i t e is not yet clear.

Lawer doses.- The spectrum mawred directly after Wlantation of a low dose (1 or 2 x 1013 at/cm2) of 12%e consists of an asymetric dcublet, the intensity of the l i n e a t negative velocity being the largest (fig. l c ) . After laser annealing we observe for crystals implanted w i t h doses between

loi3

and 2 x 1014 at/an2 spectra that can be f i t t e d with a quadrupole s p l i t ccmpnent as al- ready described, plus a single l i n e of width 1.45 mn/s a t -2.4 mn/s (fig. lb). c3anneling experiments must still be performed to locate the Te that gives r i s e to this single line.

High energy laser annealing (= 2 an^) i n a l l cases chywes the spectrum. For crystals w i t h high hplanted doses there is indication of Te-cluster f o m t i o n .

Oven annealing a t 9 0 0 ~ ~ of laser annealed crystals (= 1.3 an*) results in spectra that are similar to spectra cbtained after oven annealing cmly. These spectra differ appreciably f r a n the laser annealed spectra a t a l l doses. They can

not

be f i t t e d by two s h p l e spectrdL cmrpnents, in- d i c a t h g the presence of T e a t a m with a n- of different erwbmmmts.

charmeling experiments on a S i crystal im planted with 2 x 10'5 at/an2 and w e n annealed a t 900'~ shaw Te yield curves that are relatively m e narrawed than after laser annealing, the minimum yield being about 55%. The s u b s t i t u t i o q

or nearly substitutional, f r a c t i o n i n this case is then 50%, i n agr-twith results of Gplai e t al, /3/. The channeling measurements on the laser annealed and on the wen annealed crystals give no evidence that a part of the Te is situated in a tetrahedxal sumo- i n t e r s t i t i a l position in the Si-lattice, a s previously assumed by us /2/.

The systan Mi.- 153Sm has been implanted a t in

<loo>

and <Ill> S i single crystals with doses between 8 x 1012 and 3 x 1014 at/an2. Mter

*

plantation the MZissbauer s p c b X K ~ wnsists of single l i n e w i t h an i s m s h i f t that i s c?haradxz- i s t i c of E U ~ + (- -12 mn/s with resped to an W.

%20 absorkr) (fig, 2a). The area of this line decreases w i t h dose (fig. 3a) and does not deperad on substrate orientation.

A f t e r annealing a t 650% for 20 minutes the spectmn both depends on implanted dose and sub- s t r a t e orientation ( f i g 3b): for doses up to 3 x 1013 at/& we observe.an increase in the intensity of the 2+ oanponent, for doses of

- .

Fig. 2: Mijssbauer H < l L l > S i Crysal with a. 3 x

lo1'+

at/an2, as implanted,

b. 3 x 10'4 152~11 at/cm2, after 20 min

(4)

8 x 1013 a t / m and higher there also appears a l i n e i n the spectnnn t h a t is characteristic f o r

m3+

( a t zero velocity), while the relative inten- s i t y f a r the 3+ cunponent is highest f o r < I l l > S i crystals. Figure 2b shm a typical M5ssbauer s

- .

F'ran measwmmts on annealed sources as a M i o n of temperature it is found that the 2+

and 3+ ccnpnents have characteristic Miissbauer t a p e r a t u r e s O M = 264 and 228 K, respectively.

That this difference is 4 1is consistent with fig. 3b, £ran which it can be deduced that the recoilless fractions f o r Eu2+ and

m3+

do not d i f f e r very mch.

Pirich e t . al. /4/ observed in t h e i r e periments on cold mrked Eu

-

^{Mg alloys}^trans-

formations £ran

m2+

t o

m3+.

~ h q r suggest t h a t this transformation might be due to strain.

The values fourd by u s f o r OM for both a m p n e n t s make such an &lanation unlikely in our case.

Fran channeling experiments p e r f d on

<loo>

S i crystals with 3 1014 a t 152~m/an2, annealed during 20 minutes a t 650°c, we conclude t h a t a part of the inplanted Sm is i n an inter- stitial lattice site, as has been discussed f o r Yb i n S i by Anderson et. al. /5/. The r a a h h g part seems to occupy randan positions. With .<Ill>

S i crystals we have only &served rardan t3n so f a r . These r e s u l t s might be explained by assuming t h a t the i3u2+ mnponent is connected to the i n t e r s t i t i a l 3n while the Eu3+ c-nent belongs t o the randan 3 n

-

^fraction.^{This is}the m r e likely as it is knawn t h a t

<loo>

S i recrystallises better epitaxially than <Ill> Si., I n this case possibly m e Sm associated crystal defects result a f t e r

annealing the <Ill> Si crystal.

Glancing angle RuthaTord backscattering

w i r n m t s with high depth resolution on a S i crystal with a dose of 1015 a t w a n 2 . indicate

&I i n such a crystal strorigly tends to migrate to the surface during annealing, where it acclanulates belaw the oxide layer. This could also happen to scme extent a t lawer doses. This fraction then w i l l be seen as rardan with channeling, and it might well contribute to the

m3+

c a p n e n t i n the 6ssbauer spxtnm. On the other hard, however,

the 3+ canpnent certainly needs not to be due only to 3 n that has accunmlated a t the surface, since directly after '53Sm inplantation i n S i a t 80 K the spectrum consists of both a 2+ snd a 3+ ccnr ponent of about equal i n t a i t y . Only a 3+ can- ponent w a s observed directly a f t e r inplantation of 153Sm i n diamond and a

-

^tin.

VELOCITY ICHISEC)

Fig. 3: a. Line intensity of

m2+

ccnpnent a s a function of dose after hplantation b. Line intensities of Eu2+ and Eu3+

aanpnents in

<loo>

and < I l l > S i crystals a f t e r 20 min annealing a t 650'~ as a function of dose.

(5)

JD. RNAL DE PHYSIQUE

References

/1/ Haf6wi~ter~D.W. arid de W a d , H., Phys. Rev B7 (1973) 3014.

7

/2/ de W a d , H., EUkshpan, S. and Kemerink, G.J.

Hyp. Int.

2

(1977) 45.

/3/ Gyulai, J., Meyex, O., Poshley R.D. and Mayer, J.W., Rad. Eff.

2

(1971) 17.

/4/ Pirich, R.G., Burr, C.R., Shenoy, G.K., Dunlap, B.D., Suits, B. and Phillips, J.D., Phys. Rev.

L e t t .

3

(1977) 1142.