• 沒有找到結果。

Si

1-x

Ge

x

(x~0.35)

Various Implant splits E=15keV, 50keV 75keV

RTA

Fixed Ge content Si

1-x

Ge

x

(x~0.35)

MSA

Various soak anneal

time 800°C 60s to 1200s

Flow (a) Flow (b) Flow (c)

RTA

MSA Fixed Ge content

Si

1-x

Ge

x

(x~0.35)

Various Implant splits E=15keV, 50keV 75keV

RTA

Fixed Ge content Si

1-x

Ge

x

(x~0.35)

MSA

Various soak anneal

time 800°C 60s to 1200s

Flow (a) Flow (b) Flow (c)

RTA

MSA

Figure 3.3.1 Process flow of the relaxation of strained SiGe associated to (a) various implant conditions, (b) various Ge contents SiGe, and (c) various soak anneal time

75

3. 4 Results and Discussion

3.4.1 Implant effect on relaxation of strained SiGe

In figure 3.4.1, the bow height change of stained SiGe wafer with subsequent implant, RTA and laser MSA annealing process is revealed. At initial stage a 100nm-thick layer strained-Si1-xGex (x nearly~0.35) was deposited and it introduced a compressive film stress.

Then the subsequent implantation process, by using Arsenic (As) with medium projection range (Rp) near 40nm with an energy of 50keV and a dose of 3E13cm-2 , damaged an upper portion of strained SiGe layer and caused partial relaxation, in which bow height changed to less compressive state. Following spike RTA treatment could not fully recover the damaged strained-SiGe and still kept the wafer in compressive state near -20um. The subsequent MSA caused significant SiGe wafer bow height from compressive to highly tensile state near 150um. The TEM image showed significant defect formation in the underlying Si substrate.

For comparison, a strained-SiGe sample directly underwent MSA process and it did not cause significant bow height change and still maintained SiGe wafer at compressive state. The TEM image shows that no defect is formed in strained SiGe film and underlying Si substrate. That indicates upon MSA only certain level of relaxation of strained SiGe wafer results in significant bow height change from compressive to high tensile state and defect formation in the underlying Si substrate. In contrast, non-relaxed strained SiGe upon MSA did not show

76

obvious bow height change.

Fig. 3.4.2 plots a correlation between the relaxation of implanted strained-SiGe during MSA and the bow height change of the SiGe wafer. The three SiGe samples with low, medium and high Rp implantation conditions were followed by RTA, causing 11%, 51% and 75% relaxation of strained-SiGe, respectively. For the case of 11% relaxation of strained-SiGe, the MSA did not significantly degrade the lightly relaxed strained-SiGe keeping bow height in compressive state and did not produce defects in the underlying Si substrate shown as in the TEM image. The medium implantation Rp, subsequent RTA process caused a 51% relaxation of strained-SiGe. MSA at such remarkably relaxed strained-SiGe caused significant bow height change from compressive to tensile state and formed many defects in the underlying Si, as shown in the TEM image. Finally, a more deeply implanted Rp, the RTA process caused an almost 75% relaxation of strained-SiGe, and produced numerous dislocations in the SiGe film.

MSA applying to a high relaxation of strained-SiGe wafer resulted in bow height at less tensile state and did not form defects in the underlying Si, as presented in the TEM image. In designing a high performance of CMOS devices with strained-SiGe and MSA, pseudomorphic relaxation of strain SiGe of less than 11% must be maintained to ensure favorable channel stress to boost

77

SiGe Implant RTA MSA

Blanket wafer,BowHeight (um) SiGe + Implant + RTA + MSA

Thickness=100nm, Ge 35 at.%

Tensile

SiGe Implant RTA MSA

Blanket wafer,BowHeight (um) SiGe + Implant + RTA + MSA

Thickness=100nm, Ge 35 at.%

Tensile

Strain Relaxation of SiGe (%)

W a fe r Bow Hei ght ( um )

Slightly

Strain Relaxation of SiGe (%)

Slightly

Relaxation of strain SiGe (%) -100

Strain Relaxation of SiGe (%)

Slightly

Strain Relaxation of SiGe (%)

Slightly

Relaxation of strain SiGe (%)

-100

Strain Relaxation of SiGe (%)

W a fe r Bow Hei ght ( um )

Slightly

Strain Relaxation of SiGe (%)

Slightly

Relaxation of strain SiGe (%) -100

Strain Relaxation of SiGe (%)

Slightly

Strain Relaxation of SiGe (%)

Slightly

Relaxation of strain SiGe (%)

SiGe

Strain Relaxation of SiGe (%)

W a fe r Bow Hei ght ( um )

Slightly

Strain Relaxation of SiGe (%)

Slightly

Relaxation of strain SiGe (%) -100

Strain Relaxation of SiGe (%)

Slightly

Strain Relaxation of SiGe (%)

Slightly

Relaxation of strain SiGe (%)

-100

Strain Relaxation of SiGe (%)

W a fe r Bow Hei ght ( um )

Slightly

Strain Relaxation of SiGe (%)

Slightly

Relaxation of strain SiGe (%) -100

Strain Relaxation of SiGe (%)

Slightly

Strain Relaxation of SiGe (%)

Slightly

Relaxation of strain SiGe (%)

SiGe

Figure 3.4.1 Bow height change of strained-SiGe wafer associated with following implant, RTA and subsequent MSA, causing large SiGe wafer bending. A fully strained-SiGe sample directly underwent MSA, exhibited no wafer bending

Figure 3.4.2 Correlation between the relaxation of strained SiGe and injection of defects, determined from wafer bow height.

78

the drive current and an absence of defect formation in the Si to ensure a low junction leakage current.

3.4.2 In-situ doped Ge content effect on relaxation of strained SiGe

To investigate how high Ge at.% of SiGe can be achieved to avoid strain relaxation and defect formation during MSA process in advanced nano devices, four strained Si1-xGex

samples (x=0.2, 0.27, 0.33, and 0.44) and their interaction with MSA are explored. Fig.3.4.3 plots step-by-step bow height change of four SiGe wafers with different Ge content, subsequent implant, RTA and laser MSA annealing process flow. Higher Ge content showed higher compressive bow height than that of lower Ge content. After the subsequent medium Rp of Arsenic at energy of approximately 50keV and a dose of 3E13cm-2, then high Ge content of strained SiGe wafer lose most of initial compressive stress. RTA is unable to recover back to original as-deposited bow height state and the strain loss is 64%. In contrast, the other three Ge 20 at.%, 27 at.% and 33 at.% of SiGe wafers loss their strain of 10%, 29%

and 50%, respectively. After MSA step, Ge27 at.% of strained SiGe wafer revealed the largest bow height change among those conditions. Fig.3.4.4 shows the same phenomena of various relaxation degree of strained SiGe with different Ge content and their interaction with MSA causing large wafer bow change. It indicates Ge content near 27 at.% on blanket

79

Medium Ge, 27 at.%

High Ge, 33 at.%

Ultra high Ge, 40 at.%

SiGe

Medium Ge, 27 at.%

High Ge, 33 at.%

Ultra high Ge, 40 at.%

-75

SiGe

Medium Ge, 27 at.%

High Ge, 33 at.%

Ultra high Ge, 40 at.%

SiGe

Relaxation of strained SiGe (%)

Slightly

Relaxation of strained SiGe (%)

Slightly

Relaxation of strained SiGe (%)

Slightly

Relaxation of strained SiGe (%)

Slightly

Relaxation of strained SiGe (%)

Slightly

Relaxation of strained SiGe (%)

Slightly

Figure 3.4.3 Bow height change of four Si wafers with following Si1-xGex

deposition (x=0.2, 0.27, 0.33 and 0.44), implant of Arsenic, spike rapid anneal (RTA), and laser MSA annealing.

Figure 3.4.4 Correlation between the relaxation of strained SiGe and wafer bow height after laser MSA annealing.