• 沒有找到結果。

Effects of plasma treatment on the precipitation of fluorine-doped silicon oxide

N/A
N/A
Protected

Academic year: 2021

Share "Effects of plasma treatment on the precipitation of fluorine-doped silicon oxide"

Copied!
6
0
0

加載中.... (立即查看全文)

全文

(1)

Journal of Physics and Chemistry of Solids 69 (2008) 555–560

Effects of plasma treatment on the precipitation

of fluorine-doped silicon oxide

Jun Wu

a,



, Ying-Lang Wang

b,c,



, Cheng-Tzu Kuo

a

aDepartment of Materials Science and Engineering, National Chiao-Tung University, Hsin-chu, Taiwan bDepartment of Material Science, National Tainan University, Tainan, Taiwan

c

Department of Applied Physics, National Chiayi University, Chiayi, Taiwan

Abstract

Precipitates were observed on the surface of fluorine-doped silicon oxide (SiOF) films. These precipitates are flake-type and hexagonal in shape, showing up rapidly after initiation, and clustered at the wafer center. Post-deposition N2O plasma treatment (post-plasma

treatment) was found to be most effective in inhibiting the appearance of precipitates. In this paper, effects of post-deposition N2O

plasma treatment on the suppression of precipitates and stabilities of SiOF film were studied. X-ray photoelectron spectroscopy (XPS) analyses were conducted to investigate the changes in surface composition of SiOF films after N2O plasma treatment. The surface

morphology of the film was characterized by atomic force microscopy (AFM). Cross-sectional transmission electron microscopy (TEM) images showed that a surface layer of 150 A˚ was generated after N2O plasma treatment. The changes on surface structures of SiOF films

caused by N2O plasma treatment and the consequent inhibition of precipitate formation were discussed.

r2007 Elsevier Ltd. All rights reserved.

Keywords: A. Electronic materials; A. Semiconductor; A. Thin film; B. Vapor deposition; D. Defect

1. Introduction

With the advancement of ultra-large-scale integrated circuits (ULSI) technology, it is essential to reduce wiring capacitance and minimize the signal delay as the space between metal lines decreases to the sub-half-micron level. It is well accepted that low-k materials are inevitable at the sub-0.13 mm technology nodes [1]. Fluorine-doped silicon oxide film (SiOF) has been found to be very effective in the reduction of dielectric constants and has become most desirable because of its good thermal stability, similarity to silicon oxide, and easy integration into existing silicon processing. The SiOF film is easily deposited by simply introducing a fluorine source gas into the plasma-enhanced chemical vapor deposition system for silicon oxide. The incorporation of highly electronegative fluorine causes changes in the Si–O network to a less-polarizable geometry

and results in the reduction of the dielectric constant

[2]. The dielectric constant decreases by increasing the fluorine concentration in the SiOF film. However, it also reduces film stability. The SiOF film is sus-ceptible to moisture absorption. Films with high fluorine contents become interactive with water from air, and consequently result in degradation of film quality and low-k characteristics. Many researches have focused on the detrimental effects of water absorption on the properties of SiOF films[3–5]. In addition to the structural instability of SiOF bonding reported elsewhere, one specific feature— precipitation on the surface of SiOF film—was also observed.

Fig. 1(a) shows the SEM image of the precipitates on

SiOF films. The wafer mapping of precipitates distribution is shown in Fig. 1(b). As investigated in our earlier report

[6], precipitation initiated spontaneously forming hexago-nal-shaped structure and clustered at the wafer center while the SiOF film was exposed to moist atmosphere.Fig. 2(a)is the cross-sectional transmission electron microscopy (TEM) image of precipitates on SiOF film. On high

www.elsevier.com/locate/jpcs

0022-3697/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.jpcs.2007.11.020

Corresponding author.

E-mail addresses:junwu2668@yahoo.com.tw (J. Wu),

(2)

magnification the precipitate shows loosely connected porous structure.Fig. 2(b)demonstrates the EDX analysis spectra of the precipitate. Mainly Si and O are detected (F is supposed to be beyond tool detection limit). Such defects generated in semiconductor processing could induce film delamination and metal bridges, resulting in detrimental effects on device performance. A surface modification technique by employing the in situ plasma treatment method post film deposition was proposed and has been validated to be an effective method to eliminate the precipitate formation [6]. In this study, effects of post-plasma treatment on the surface character-istics and dielectric properties of SiOF thin films were investigated.

2. Experimental

SiOF films were deposited on n-type silicon substrates in a remote plasma-enhanced chemical vapor deposition

system using SiF4/O2 gas mixture. The reaction chamber

was equipped with an inductively coupled plasma source. The plasma discharge was maintained using 13.56 MHz rf power at 200 W to decompose the gas mixture and obtain SiOF films of 4K A˚ thickness. After deposition, SiOF films were exposed to the moist atmosphere at 25 1C, 45% relative humidity. The fluorine concentration of the as-deposited SiOF film is 9 at%, expressed as the fluorine atomic percentage from FTIR measurement. The post-plasma treatment was applied on N2O gas and was

maintained using 13.56 MHz rf power at 150 W. The surface roughness of SiOF films with and without post-plasma treatment was measured by atomic force micro-scopy (AFM). The scanning area of AFM surface images for each substrate is 10  10 mm2. X-ray photoelectron

Fig. 1. (a) SEM image of precipitates on the surface of SiOF films and (b) wafer mapping of the precipitates distribution.

Fig. 2. (a) Cross-sectional TEM image of precipitates on the surface of SiOF films and (b) EDX analysis of precipitates.

(3)

spectroscopy (XPS) was employed to investigate the difference in the distribution of surface elements that occurred at the exposed surface of plasma-treated SiOF films. The dielectric constants of SiOF films were investi-gated by measuring capacitance–voltage (C–V) character-istics using a metal–insulator–semiconductor (MIS) capacitor structure at 1 MHz. TEM analyses were per-formed to investigate the cross-sectional microstructures of the plasma-treated film surface. Throughout the exp-eriment, the precipitate counts were measured by a wafer surface defect measurement tool (Tool-type KLA SP1). An abrupt increase in defect count (from 101 to 4103) in 1 time interval was considered as the onset of precipitation.

3. Results and discussion

Fig. 3illustrates the variation of precipitate counts with

deposition time for non-plasma-treated and post-plasma-treated SiOF films with 6 at% of fluorine content. Non-plasma-treated (without post-plasma treatment) and in-situ post-plasma-treated SiOF films were subjected to an exposure to moist air. The results indicated that precipi-tates showed up on the non-plasma-treated SiOF films after 2 h, whereas no precipitation was observed for the post-plasma-treated SiOF films. In the post-plasma treat-ment, ion irradiation on the as-deposited SiOF films modified both topography and structure of the surface layer of the SiOF film. This modification is by the combined effects of sputtering, implantation, diffusion, and re-deposition processes [7]. Fig. 4 shows the TEM cross-sectional images of non-plasma-treated and post-plasma-treated SiOF films. A surface layer of 150 A˚ was formed after the post-plasma treatment. The AFM images

of surface topographies for non-plasma-treated and post-plasma-treated SiOF films are shown in Fig. 5. The measured root mean squares of surface roughness, taken from 10  10 mm2 scans, for the as-deposited and

Fig. 4. TEM cross sectional images of (a) non-plasma-treated and (b) post-plasma-treated SiOF films. A surface layer of 150 A˚ was formed after the plasma treatment.

2 6 10 12 14 0 500 1000 1500 2000 2500 3000 Precipitate count(ea)

Post deposition time(h) Postplasma-treated SiOF SiOF

4 8

Fig. 3. Variation of precipitate counts with post-deposition time for SiOF films with 6 at% fluorine content. Non-plasma-treated and in-situ post-plasma-treated SiOF films were subjected to an exposure to moist air.

(4)

the post-plasma-treated SiOF film were around 1.21 and 2.23 A˚, respectively. The surface roughness of the SiOF film increased as a result of the post-plasma treatment. This may be ascribed to the sputtering effects of N2O plasma on

the surface layer of the SiOF film during post-plasma treatment.

Fig. 6 shows the XPS spectra of post-plasma-treated

and non-plasma-treated SiOF films. There is no signi-ficant difference between those with and without post-plasma treatment. The main bonding species were still kept after the plasma treatment. In Figs. 6(b) and (c), the binding energies of O 1s and F 1s after treatment lowered down to about 0.3 eV compared with the non-plasma-treated film. Table 1 summarizes the content of key elements. The concentration of oxygen and silicon increased whereas that of fluorine decreased

in the post-plasma-treated film. XPS result indicates that the chemical composition of the SiOF film at the surface changed from that of the as-deposited SiOF film. Oxidation and surface fluorine depletion occurred during the N2O post-plasma treatment.

Some fluorine atoms on the surface layer of SiOF films were replaced by oxygen atoms after post-plasma treatment. It is likely that the active radicals (e.g. O+) and ion-bombardment effects in the post-plasma treatment changed both the chemical composition and the density at the top surface of SiOF films [8], resulting in both depletion of fluorine concentration and oxidation on the film surface. In our previous study we reported that fluorine in the surface layer of SiOF films participated in the formation of precipitates [6], and fluorine species in the bulk of SiOF film could diffuse to film surface and form precipitates. The surface oxidation and densification due to plasma treatment plays an important role in blocking the diffusion path of fluorine from bulk to the surface, thus inhibiting the formation of precipitates.

Comparison of film properties for the as-deposited and the post-plasma-treated SiOF films is summarized in

Table 2. More pronounced increase in the compressive

stress of SiOF films was observed as a result of post-plasma treatment. Nevertheless, there is no noticeable difference in the overall film density and dielectric constant between the as-deposited and the post-plasma-treated SiOF films, even though the post-plasma treatment leads to the depletion of fluorine on the surface and densification of the film by ion bombardment. The post-plasma treatment is quite an efficient method not only in blocking the moisture[9]but also in inhibiting the precipitation formation. The dielectric properties and stabilities of SiOF films can be improved without degrading its low-k characteristics by N2O

post-plasma treatment.

4. Conclusions

Precipitates were observed on the surface of fluorine-doped silicon oxide films. The N2O post-plasma

treat-ment can effectively suppress the formation of preci-pitates. A surface layer of 150 A˚ was formed after the post-plasma treatment. Post-deposition treatment affects both chemical composition and film density of the surface layer of SiOF film. The post-plasma treatment will lead to oxidation, depletion of fluorine, and densification of the surface layer through active radicals (e.g. Oþd d) and ion-bombardment effects in N2O

plasma. It is quite an efficient method to block the diffusion path of fluorine from bulk to film surface, thus inhibiting the formation of precipitates. The dielectric properties and stabilities of SiOF films can be improved without degrading its low-k characteristics by N2O

post-plasma treatment.

Fig. 5. AFM images of the surface topography for the non-plasma-treated and the post-plasma-treated SiOF films.

(5)
(6)

Acknowledgment

The authors gratefully acknowledge the financial sup-port of the National Science Council (NSC) of Taiwan for this research project.

References

[1] M. Yoshimaru, S. Koizumi, K. Shimokawa, J. Vac. Sci. Technol. A 15 (1997) 2915.

[2] H. Yang, G. Lucovsky, J. Vac. Sci. Technol. A 15 (836) (1997). [3] M. Yoshimaru, S. Koizumi, K. Shimokawa, J. Vac. Sci. Technol. A 15

(2915) (1997).

[4] T. Fukada, T. Akahori, Extended Abstracts of Conference on Solid State Devices and Materials, 1993, p. 158.

[5] V. Pankov, J.C. Alonso, A. Ortiz, J. Appl. Phys. 86 (1) (1999). [6] J. Wu, Y.L. Wang, C.T. Kuo, Thin Solid Films 447 (2004) 599–604. [7] A. Stanishevsky, L. Khriachtchev, J. Appl. Phys. 86 (12) (1999). [8] S. Lee, K. Oh, Y.T. Kim, et al., J. Appl. Phys. 85 (1) (1999). [9] S. Lee, J.W. Park, J. Electrochem. Soc. 146 (2) (1999) 697–701. Table 2

Comparison of film properties for the as-deposited and the post-plasma-treated SiOF films Conditions Dielectric constant (k) Refractive index (n) Densitya(g/ cm3) Stress (MPa) Non-plasma-treated SiOF 3.522 1.435 2.0995 204.1 Post-plasma-treated SiOF 3.523 1.434 2.0953 211.2 a

The density is determined from the n value using Lorentz–Lorentz relationship r ¼ K(n21)/(n2+2), where K ¼ 8.046.

Table 1

Percentage of elements from XPS quantification (unit: %)

Conditions Take-off F O N C Si

No treatment 0 3.8 55.5 0.5 8.8 31.4

60 3.2 53.9 0.2 12.9 29.7

N2O treatment 0 3.1 57.1 0.1 7.4 32.2

數據

Fig. 1. (a) SEM image of precipitates on the surface of SiOF films and (b) wafer mapping of the precipitates distribution.
Fig. 4. TEM cross sectional images of (a) non-plasma-treated and (b) post-plasma-treated SiOF films
Fig. 6 shows the XPS spectra of post-plasma-treated
Fig. 6. XPS spectra of (a) wide scan (MgKa) (b) O 1s, and (c) F 1s for SiOF films with and without post-plasma treatment.

參考文獻

相關文件

For ex- ample, if every element in the image has the same colour, we expect the colour constancy sampler to pro- duce a very wide spread of samples for the surface

You are given the wavelength and total energy of a light pulse and asked to find the number of photons it

好了既然 Z[x] 中的 ideal 不一定是 principle ideal 那麼我們就不能學 Proposition 7.2.11 的方法得到 Z[x] 中的 irreducible element 就是 prime element 了..

Wang, Solving pseudomonotone variational inequalities and pseudocon- vex optimization problems using the projection neural network, IEEE Transactions on Neural Networks 17

volume suppressed mass: (TeV) 2 /M P ∼ 10 −4 eV → mm range can be experimentally tested for any number of extra dimensions - Light U(1) gauge bosons: no derivative couplings. =>

Define instead the imaginary.. potential, magnetic field, lattice…) Dirac-BdG Hamiltonian:. with small, and matrix

incapable to extract any quantities from QCD, nor to tackle the most interesting physics, namely, the spontaneously chiral symmetry breaking and the color confinement.. 

• Formation of massive primordial stars as origin of objects in the early universe. • Supernova explosions might be visible to the most