Si nanowires grown from silicon oxide

Đây là một bài báo khoa học về dây nano silic trong lĩnh vực nghiên cứu công nghệ nano dành cho những người nghiên cứu sâu về vật lý và khoa học vật liệu.Tài liệu có thể dùng tham khảo cho sinh viên các nghành vật lý và công nghệ có đam mê về khoa học

Ž

Chemical Physics Letters 299 1999 237–242

Si nanowires grown from silicon oxide

Center of Super Diamond and AdÕanced Films, Department of Physics and Materials Science, The City UniÕersity of Hong Kong,

Hong Kong, China

Received 10 August 1998

Abstract

Bulk-quantity Si nanowires have been synthesized by thermal evaporation of a powder mixture of silicon and SiO 2 Transmission electron microscopy showed that, at the initial nucleation stage, silicon monoxide vapor was generated from the powder mixture and condensed on the substrate Si nanoparticles were precipitated and surrounded by shells of silicon oxide The Si nanowire nucleus consisted of a polycrystalline Si core with a high density of defects and a silicon oxide shell The growth mechanism was proposed to be closely related to the defect structure and silicon monoxide q 1999 Elsevier Science B.V All rights reserved.

Nanometer-wide silicon wires have attracted much

attention in recent years because of their potential for

applications in the field of microelectronics One of

the challenging issues has been the synthesis of this

one-dimensional form of nanowires on large scales

Since the successful growth of Si whiskers by the

vapor–liquid–solid VLS method 1,2 , many

ef-forts have been made to improve the synthesis of Si

nanowires by employing different techniques, such

as the photolithography technique combined with

etching 3–5 and scanning tunneling microscopy

w6,7 For the VLS method, Au had to be used andx

this caused contamination The diameters of Si

whiskers obtained from VLS were determined by the

size of Au particles Other techniques were

compli-cated and could not produce bulk quantities of Si

nanowires

)

Corresponding author E-mail: apannale@cityu.edu.hk

Recently, Si nanowires have been successfully synthesized by a novel method of laser ablation of

metal-containing Si targets 8–11 Previous

gations 8,9 have shown that metal or metal-silicide nanoparticles acted as the critical catalyst during the deposition assisted by laser ablation For example,

Fe could form Fe-silicides at high temperatures of 12008C A growth mechanism of Si wires has been

ascribed to the VLS reaction 8,9 However, a dif-ferent model has been proposed which is supported

by the experiment which showed that metal catalyst were not observed in Si nanowires even when metals

w x

were mixed in the target 10 Moreover, it was discovered that metal was not necessary for Si nanowire synthesis by laser ablation Instead, SiO2 was the special and effective catalyst which largely

w x

enhanced Si nanowire growth 12 High-resolution

transmission electron microscopy HRTEM investi-gations have shown that high-density defects and silicon oxide outer layers play important roles for

w x

nanowire growth 12 In this Letter, we report that 0009-2614r99r$ - see front matter q 1999 Elsevier Science B.V All rights reserved.

PII: S 0 0 0 9 - 2 6 1 4 9 8 0 1 2 2 8 - 7

( )

N Wang et al.r Chemical Physics Letters 299 1999 237–242

238

bulk-quantity Si nanowires were synthesized by

ther-mal evaporation of a highly pure Si powder mixed

with SiO Observations of Si nanowire nucleation2

and growth morphology by transmission electron

microscopy TEM are documented By combining

these observations with the results of a Raman study,

we discuss the growth mechanisms

Si nanowires can be synthesized by laser ablation

of a powder mixture of silicon and SiO2 in an

Ž

evacuated quartz tube in an Ar atmosphere 500

w x

Torr 12 However, in the present work, without the assistance of laser ablation, Si nanowires were syn-thesized by simple thermal evaporation at 12008C The solid source was highly pure Si powder mixed

Ž

with about 70 wt% SiO2 all materials were from

Goodfellow, purity 99.99% The temperature around the quartz tube where the nanowire grew was about 9308C After 12 h of thermal evaporation, Si

Fig 1 a TEM image showing the morphology of Si nanowires synthesized by the evaporation method b – d Nucleation stage of the Si nanowires.

Ž

Fig 1 continued

nanowire product sponge-like, dark red in color

formed on the inside wall of the quartz tube To

collect Si nanowire nuclei, a Mo grid was placed in

the region of the quartz tube where the nanowires

grew Some Si nanowires nucleated and grew on the

grid The Mo grid was directly observed in Philips

CM200FEG transmission electron microscope

work-ing under 200 kV Raman measurements were

car-ried out using with a Renishaw 2000 micro-Raman

system

Fig 1a shows the typical morphology of as-grown

Ž

Si nanowires The nanowires major component in

the product are extremely long ) 10 mm with

uniform diameters and smooth surfaces Si

nanopar-ticles are found to coexist with the nanowires A

striking feature is that Si nanoparticles appear in the

form of chain Si nanowire nucleation on the Mo grid is shown in Fig 1b In initial stage, Si nanopar-ticles were formed as identified by electron diffrac-tion Most nanoparticles piled up on the substrate

Ž

Notably, some favorable particles nuclei of

nanowires stood alone and underwent faster growth since their preferable growth direction was normal to

the surface of the substrate see Fig 1b–d There was no detectable metal catalyst or impurity formed

on the tips of the nanowire nuclei Each nucleus simply consisted of a crystalline Si core and an amorphous outer layer The chemical composition of the nuclei was determined by electron energy

sive spectroscopy EDS Only silicon and oxygen were detected which indicated that the amorphous outer layer should have been silicon oxide The Si

( )

N Wang et al.r Chemical Physics Letters 299 1999 237–242

240

crystalline core contained a high density of defects

Most of the defects showed their contrast along the

growth axis of the nucleus These defects were quite

Ž

similar to the planar defects stacking faults and

micro-twins along the axis of Si nanowire in 112

w x

observed in Si nanowires in our previous work 10

It is believed that silicon oxide plays an important

role in nanowire growth We investigated the native

silicon oxide on single Si crystal surfaces The oxide

thickness was only 2–3 monolayers However, the

oxide shells of nanowires were quite thick We

observed that the shell thickness up to 3 nm

gener-w x

ally depended on the diameter of the nanowire 10

In the present experiment, the vapor materials

gener-ated from the mixture of silicon and SiO at 12008C2

consisted mainly of SiO, with little silicon This was

supported by the observation that the material

con-densed on the water-cooled Cu finger was Si Ox y

Žx s 0.51, y s 0.49 as determined by EDS This

chemical composition was reliable since the vapor

phase was quenched on the cool finger Silicon

monoxide SiO is an amorphous semiconductor of

high resistivity which can easily be generated from

powder mixtures especially in equimolar mixtures

of silicon and SiO by heating 13–15 TEM inves-2

tigations confirmed the amorphous structure of the

SiO deposited on the Cu finger surface By heating

the SiO sample in TEM, silicon precipitation was

observed see Fig 2a Such precipitation of Si

nanoparticles from annealed SiO is quite well known

w15 x

According to the above observations, we propose

that the growth mechanism is silicon oxide assisted

The vapor phase of Si O x ) 1 generated by ther- x

mal evaporation is the key factor The nucleation of

nanoparticles is assumed to occur at the substrate by

different decompositions of silicon oxide at the

rela-tively low temperature of 9308C as shown below

Si O ™ Six xy1qSiO Žx ) 1.

and

2SiO ™ Si q SiO 2

These decompositions result in the precipitation of

silicon nanoparticles, i.e the nuclei of Si nanowires,

clad by shells of silicon oxide as observed in Fig 1b

The growth process may involve the following

factors The relatively thick Si O on nanowire tips

w12 acts as a catalyst The SiO component of thex 2

shell, which could be formed during decomposition

of SiO in nanowire growth, retards the sideways growth of the nanowire Defects, such as stacking faults in the nucleus tips, enhance the

sional growth The 111 surface, which has the lowest surface energy among the surfaces in silicon, plays an important role during nanowire growth Since surface energy is more important when the crystal size is reduced to the nanometer scale, the

appearance of 111 surfaces of the Si crystals paral-lel to the axes of the nanowires reduces the system energy Combined, these factors determine the

growth direction of Si nanowires to be 112 This proposed growth mechanism is supported by the results of Raman study as shown in Fig 3a The peak at 521 cmy 1 is broad and strongly asymmetric compared to that from a single Si crystal Such a feature could be due to the small size effect of Si

nanocrystals or defects 11,16 since there were many nanoparticles in the product, as well as Si nanowires

containing a high-density of defects 10,11 In addi-tion, the presence of SiO shells also contributes to the asymmetry of the Raman peak As shown in Fig

Ž

3a, the spectrum taken from SiO deposited on the

Cu finger contains a broad peak located at about

480 cmy 1 For comparison, Si nanowires which

Ž

were fully oxidized by annealing in the air white in

color were studied No Raman scattering was

tected see Fig 3a According to EDS measurement, the fully oxidized nanowires consisted mainly of SiO 2

Fig 3b shows strong photoluminescence PL of SiO at about 740 nm The fully oxidized nanowire gives a weak PL peak at about 600 nm The PL from

Si nanowire product is weak and complicated A typical PL spectrum from Si nanowires covers the range of 600–800 nm range Clearly, the SiO and SiO2 components of the nanowires are the main contributors to this spectrum

The proposed mechanism for nucleation and growth can predict some of the morphology of nanowires For example, during the evaporation,

Si O vapor was continually generated and nucleationx could occur with different crystalline orientation ei-ther on the side surfaces or tips of the nanowires The former resulted in the forking of the nanowires

Žobserved frequently and the latter caused re-nuclea-

Fig 2 a Nanoparticles precipitated by heating the SiO thin film b HRTEM image of the Si nanoparticle chain.

( )

N Wang et al.r Chemical Physics Letters 299 1999 237–242

242

Ž

Fig 3 a Raman spectra taken from the as-grown Si nanowires,

Ž

SiO and fully oxidized Si nanowires b PL spectra taken from

the as-grown Si nanowires, SiO and fully oxidized Si nanowires.

tion The nuclei formed on the tips in an unfavorable

growth direction could not grow fast and

re-nuclea-tion occurred again Such re-nucleare-nuclea-tion resulted in

the formation of nanoparticle chains see Fig 1

HRTEM image taken from one of the chains

pro-vided proof for this growth mechanism As shown in

Fig 2b, the silicon particles in the chain have

differ-ent oridiffer-entations and most of the particles are not

aligned with their 112 orientations parallel to the growth direction

In conclusion, bulk-quantity Si nanowires have been synthesized by thermal evaporation of mixture

of silicon and SiO2 powder Si oxide vapor gener-ated from the powder mixture condensed on the substrate and then decomposed, forming Si

ticles nuclei of nanowires A Si nanowire nucleus consisted of a polycrystalline Si core with a high density of defects and a silicon oxide shell The growth mechanism was proposed to be closely re-lated to the defect structure of Si crystal cores and SiO

Acknowledgements

This work was financially supported in part by the Research Grants Council of Hong Kong and the Strategic Research Grants of the City University of Hong Kong

References

w x 1 R.S Wagner, W.C Ellis, Appl Phys Lett 4 1964 89.Ž .

w x 2 E.I Givargizov, J Cryst Growth 32 1975 20.Ž .

w x 3 H.I Liu, N.I Maluf, R.F.W Pease, J Vac Sci Technol B

10 1992 2846.

w x 4 H Namatsu, S Horiguchi, M Nagase, K Kurihara, J Vac.

Sci Technol B 15 1997 1688.

w x 5 Y Wada, T Kure, T Yoshimura, Y Sudou, T Kobayashi,

Y Gotou, S Kondo, J Vac Sci Technol B 12 1994 48.

w x 6 T Ono, H Saitoh, M Esashi, Appl Phys Lett 70 1997Ž . 1852.

w x 7 R Hasunuma, T Komeda, H Mukaida, H Tokumoto, J.

Vac Sci Technol B 15 1997 1437.

w x 8 A.M Morales, C.M Lieber, ACS meeting 1997, Vol 213, pp651-INOR.

w x 9 A.M Morales, C.M Lieber, Science 279 1998 208.Ž .

w 10 N Wang, Y.H Tang, Y.F Zhang, D.P Yu, C.S Lee, I x

Bello, S.T Lee, Chem Phys Lett 283 1998 368.

w 11 Y.F Zhang, Y.H Zhang, N Wang, D.P Yu, C.S Lee, I x

Bello, S.T Lee, Appl Phys Lett 72 1998 1835.

w 12 N Wang, Y.H Tang, Y.F Zhang, C.S Lee, S.T Lee, not x

published.

w 13 S.W Roberts, G.J Parker, M Hempstead, Opt Mater 6 x

Ž 1996 99 .

w 14 U Setiowati, S Kimura, J Am Ceramic Soc 80 1997 757 x Ž .

w 15 G Hass, C.D Salzberg, J Opt Soc Am 44 1954 181 x Ž .

w 16 G Nolsson, G Nelin, Phys Rev B 6 1972 3777 x