TIME DEPENDENCE OF THE NUCLEATION OF SLIP DISLOCATIONS DURING LASER ANNEALING OF SILICON

HAL Id: jpa-00219929

https://hal.archives-ouvertes.fr/jpa-00219929

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.

TIME DEPENDENCE OF THE NUCLEATION OF SLIP DISLOCATIONS DURING LASER ANNEALING

OF SILICON

G. Rozgonyi, H. Baumgart

To cite this version:

G. Rozgonyi, H. Baumgart. TIME DEPENDENCE OF THE NUCLEATION OF SLIP DISLOCA-

TIONS DURING LASER ANNEALING OF SILICON. Journal de Physique Colloques, 1980, 41 (C4),

pp.C4-85-C4-88. �10.1051/jphyscol:1980414�. �jpa-00219929�

JOURNAL DE PHYSIQUE Colloque C4, suppkment au n " 5, Tome 41, mai 1980, page (24-85

T I M E DEPENDENCE OF THE NUCLEATION OF S L I P D I S L O C A T I O N S DURING LASER ANNEALING OF S I L I C O N

G.A. Rozgonyi and H. Baumgart +

Mm-PZanck-Institut fiir Festk5rperforschwzg 7 Stuttgart, FRG.

Abstract.- We report new observations on cw laser annealing of ion implanted sili- con which have enabled us to determine the time necessary to nucleate a slip dis- location in silicon. Optical microscopy and x-ray topography of 50 to 80 pm,mldiame- ter spots annealed from 10 to 1000 ms indicate that pulses shorter than

50 ms will not introduce slip dislocations even for those cases where the surface regrow via a liquid phase epitaxial process.

It i s well known /1,3/ that pulsed laser annealing of ion implantation damaged sili- con is able to reconstruct an amorphous sur- face, via liquid phase epitaxy, to a single crystal which is totally free of d i s l o c ~ o n s and stacking faults. Presumably, the heating, melting and regrowth of the amorphous layer, all of which occurs in less than a microse- cond, is faster than the time necessary'for a dislocation to nucleate and propagate through the annealed layer. However, cw la- ser annealing which will recrystallize an amorphous surface via a solid phase epitaxy process /4,6/ in times greater than 10 mil- liseconds, is often accompanied by plastic flow in the form of slip dislocations. It occured to us that using a cw laser with progressively faster shutter times from 1 s to less than a millisecond would bridge the gap between slip-free pulsed annealing and slip-inducing cw annealing. The result of this simple experiment would be a determina- tion of the time necessary to nucleate a slip dislocation. This is a fundamental pro- perty of a crystal which may prove important in studies of laser processing of materials, as well as in adding to our basic knowledge about plastic flow phenomena in silicon.

Although we have not yet performed ac- curate time dependent exposures we have reexamined samples previously described /4/

and found that a threshold time for disloca- tion nucleation does, in fact, exist between 1 0 and 100 ms. The samples were annealed with an 18 Watt cw argon-ion laser operating

+, Permanent address : Bell Telephone Labora- tories, Murray Hill, N.J. 07974, U.S.A.

such that all the blue-green lines from 458 to 514,5 nm were simultaneously oscil- lating. The beam was focused with a 10 cm focal length lens providing a diffraction limited spot of about 40 pm, wlth an inci- dent power between 7 and 12 W. The power density was also varied by moving the sam- ple out of the focal plane of the lens, gi- ving spot diameters from 40 to 100 pm. The single crystal (100) oriented surfaces we- amorphized, prior to laser annealing, by

implantation with arsenic ions at 50 kev to doses in the range 1x10'~ to 3x10'~ ~ m - ~ . We illustrate the differences between so- lid phase epitaxy (SPE) and liquid phase epitaxy (LPE) in the series of optical mi- crographs in figure l, as well as the pre- sence or absence of slip lines. Figure la shows a SPE laser annealed spot free of slip lines, while figure lb contains a cross grid network of slip lines. The rou*

textured region in the lower center of the spot in figure lb indicates that melting has occured. We have previously reported /8/ that all samples annealed with a cw laser that showed signs of melting also had slip lines. However, we can now show in figure lc a spot with a large m~lted zone surrounded by SPE which contains no slip lines. Finally, a SPE plus LPE spot with slip is shown in figure Id. We there- fore have four types of spots, which are shown together in the lower magnification photo of figure2., S l and S 2 for SPE without and with slip, respectively ; and M I and M 2 for LPE w'ithout and with slip.

Note tha> slip in SPE samples can be

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1980414

C4-86 JOURNAL DE PHYSIQUE

Fig. 1

Optical Micrographs i l l u s t r a t i n d four types Fig. 1 : b) Type S2 : s o l i d phase e p i t a x y with s l i p of l a s e r annealed s p o t s ( a l s o c o n t a i n s small melted region.

a ) Type SI : s o l i d phase e p i t a x y without s l i p ( s p o t diam -80 pm).

Fig. 1. : Optical Micrographs i l l u s t r a t i n g four types Fig. 1 : d ) Type

: s o l i d phase p l u s l i q u i d phase of l a s e r annealed s p o t s . e p i t a x y with s l i p .

C )

Type M I : s o l i d phase p l u s l i q u i d phase e p l t a x y without s l i p .

suppressed by heating the substrate and re- /6,7/. It is also possible to eliminate

ducing the slip-inducing thermal gradients slip by reducing the laser exposure time,

between annealed and unannealed silicon as mentioned above. Because of inaccurate

control of shutter times we can only estima- te that 25 ms is too short, figure lc, for slip dislocations to form, while 200 ms is a sufficiently long enough time, figure Id.

Fig. 2 : Low magnification optical micrograph showing locations of four types of laser annealed spots defined in figure 1.

Although absence of slip in optical micrographs is a necessary condition for proving no slip dislocations exist in dots of the type M 1 , it is not sufficient proof.

We have therefore obtained x-ray transmis- sion and reflection topographs, shown in figures 3 and 4 respectively, of the same field of view as figure 2. The x-ray images dramatically illustrate, not only the ab- sence of slip dislocations but also a sur- prisingly small amount of lattice strain.

Careful examination of spots of the type S I and MI in the reflection x-ray topographs reveals strain contrast only at the boundary between the single crystal spots and the surrounding amorphous field. This strain contrast is of the same magnitude as the contrast which exists at the boundary of the ion implanted and non-implanted substra- te regions.

so rough. This is in contrast to the ex- tremely smooth surfaces obtained with a properly exposed pulsed laser, where the exposure times are less than a microsecond.

However, the occurance of ripples at the top of the melted zone with -500nm perio- dicity, previously attributed /2/ to an

Mo ka ( 2 2 0 ) t r a n s m i s s i o n t o p o g r a o h

Although there is no slip and the

strain is very low it is surprising that the Fig. 3

Transmission x-ray topograph of same

surface morphology of the melted regime is Of view as figure 2 *

C4-88 ^JOURNAL DE PHYSIQUE

interference with the incident laser beam wevelength, is another indication that chopping a cw laser will yield results si- milar to short pulsed lasers.

Summarizing the above observations., we have determined :

i) Slip dislocations in the arsemic doped silicon surfaces examined in this report re- quire approximately 50 ms of heating time before they nucleate,

ii) Lattice dilatations a the LPE/SPE/a-Si boundaries are extremely small for those laser annealed regions where slip does not occur.

iii) A rough surface morphology, due to LPE regrowth,does not necessarily correlate with high lattice strains, as long as slip dislocations are suppressed.

/3/ Rozgonyi, G.A., Leamy, H.J., Sheng, T.T., and Celler, G.K., in :

Semiconductor characterization Techni- ques, Barnes, P.A., and Rozgonyi,G.A.,

(Eds.), The Electrochemical Society, _- Princeton, N.J. (1978), p.492

/4/ Williams, J.S., Brown, W.L., Leamy, H.J., Poate, J.M., Rodgers, J.W., Rousseau, D., Rozgonyi, G.A., Shelnutt, J.A., and Sheng, T.T., Appli. Phys.

Lett. 2, (1978) 542.

/5/ Auston, D.H., ~olovchenko, J.A., Smith, P.R., Surko, C.M., and Venkatesan, T.N

C . , Appl. Phys. Lett. 32, (1978) 539.

/6/ Gat, A., Lietoila, A., and Gibbons, J.F., J. Appl. Phys. 50, (1979) 2926.

/7/ Celler, G.K., Borutta, R., Brown, W.L., Poate, J.M., Rozgonyi, G.A., and Sheng, T.T., see ref./l/, p. 381

/8/ Rozgonyi, G.A., Leamy, H.J. Sheng, T.T., and Cellar, G.K., see ref./l/, p. 457.

Fig. 4 : Reflection x-ray topograph of same field of view as figures 2 and 3.

The authors are indebted to W.L. Brown, J.W. Rodgers, D. Rousseau and J.A. Shelnutt who carried out the ion implantation and

laser annealing of the samples described here.

References

/1/ Laser-Solid interactions and Laser Pro- cessinq, Ferris, S.D, Leamy, H.J., and Poate, J.M. (Eds.). AIP Proc. Vol. 50

(1979).

/2/ Leamy, H.J., Rozgonyi, G.A., Sheng, T.T., and Celler, G.K., Appl. Phys. Lett.2,

(1978) 535.