Films minces à base de Si nanostructuré pour des cellules ...
Films minces à base de Si nanostructuré pour des cellules ... Films minces à base de Si nanostructuré pour des cellules ...
T d (°C) ν T O3 x = 0/Si from FTIR Si excess (at.%) from FTIR (unbonded Si) x = 0/Si from ellipsometry Si excess (at. %) from ellipsometry (agglomerated Si) 200 1059.6 1.77 4.15 1.92 2.15 300 1061 1.79 3.8 1.92 2.15 400 1064 1.82 3.15 1.91 2.37 500 1067 1.86 2.539 1.76 4.98 Table 3.2: Si excess estimation by FTIR and refractive index (Bruggeman method- analysis with regard to T d . (d) Growth mechanism tel-00916300, version 1 - 10 Dec 2013 It has been reported that r d increases with increasing T d [Gates 89, Kaiser 98]. Our results indicate a steadily decreasing r d which can be related to a continuous increase in the etch rates. But, n 1.95eV increases linearly with T d indicating an increase in the Si content. This necessitates an understanding of the temperature dependent growth mechanism. The following mechanism is proposed based on the kinetic theory of gases and temperature dependence of SiH x etch products. A diagrammatic representation of the proposed explanations are shown in gure 3.3. According to classical mechanics, the kinetic energy is a function of the particle mass (m) and velocity (v) expressed as, E k = (1/2)mv 2 ⋍ (3/2)kT Eqn. (3.9) where (3/2)kT is the averaged kinetic energy of a classical ideal gas per degree of freedom related to temperature T, and k is the Boltzmann constant. Considering our sputter gases, Ar is monoatomic having three degrees of freedom and hydrogen being a diatomic gas has 6 degrees of freedom. It can be seen from equation 3.9 that an increase in temperature increases the average kinetic energy and therefore the velocity of the species in the sputter chamber. When the substrate is heated, it cannot be denied that a certain amount of energy is introduced in the plasma leading to the plasma temperature. Thus it can be said that with increasing T d , E k(avg) and velocity of species in the plasma increases leading to an enhancement in the kinetics of sputtering process. Moreover, in a mixture of particles with dierent masses, the heavier atom will have a lower velocity than the lighter one but have the same average kinetic energy. Hence, at a given T d hydrogen having the lowest mass acquires the greatest velocity. 64
tel-00916300, version 1 - 10 Dec 2013 Figure 3.3: Proposed mechanism of temperature dependent reactive sputtering. As a consequence of an increase in hydrogen velocity with increasing T d as compared to other species, Si deposition is favoured. This is attributed to the increased reaction of hydrogen in reducing the sputtered oxygen species which move towards the substrate at a lower velocity. This explains the continuous rise in the n 1.95eV as the Si content in SRSO layer increases with T d . The role of hydrogen is not restricted to deposition, but also towards an etching mechanism since the weak and strained Si-Si bonds are broken leading to removal of Si atoms from the surface [Tsai 89, Akasaka 95]. Hence, with increased velocity, it is also probable that hydrogen reaches the lm growing surface faster than the other Si or O atoms thereby forming Si-H bonds faster than Si-Si bonds. This leads to a continuous etching process resulting in a decreased r d . Thus, the competing etching mechanism seems to be predominant at T d ranging between 200°C to 400°C. At elevated temperatures, between T d = 425°C to 600°C, in addition to the process described above, the temperature allows the formation of Si-Si bonds since hydrogen from Si-H dissociates into Si and H 2 [Veprek 79]. This explains the decrease of r d and concomitant increase of refractive index. Increasing time and temperature of deposition facilitates the reorganization of the Si-Si bond into a network which could also lead to the agglomeration of few 65
- Page 31 and 32: tel-00916300, version 1 - 10 Dec 20
- Page 33 and 34: Shockley-Queisser limit of 31% [Sho
- Page 35 and 36: tel-00916300, version 1 - 10 Dec 20
- Page 37 and 38: tel-00916300, version 1 - 10 Dec 20
- Page 39 and 40: tel-00916300, version 1 - 10 Dec 20
- Page 41 and 42: Figure 1.12: materials. Energy diag
- Page 43 and 44: tel-00916300, version 1 - 10 Dec 20
- Page 45 and 46: Background of this thesis: A new me
- Page 47 and 48: Chapter 2 Experimental techniques a
- Page 49 and 50: maximum power into the plasma. (a)
- Page 51 and 52: Figure 2.2: Illustration of sample
- Page 53 and 54: Ray Diraction, X-Ray Reectivity, El
- Page 55 and 56: (a) Normal Incidence. (b) Oblique (
- Page 57 and 58: to equation 2.4, when X-ray beam st
- Page 59 and 60: investigation. This value is obtain
- Page 61 and 62: e seen that large θ (here, θ is u
- Page 63 and 64: Figure 2.12: Raman spectrometer-Sch
- Page 65 and 66: Experimental set-up and working tel
- Page 67 and 68: ˆ The presence of Si from SiO 2 or
- Page 69 and 70: 2.2.7 Spectroscopic Ellipsometry Pr
- Page 71 and 72: k(E) = f j(ω − ω g ) 2 (ω −
- Page 73 and 74: tel-00916300, version 1 - 10 Dec 20
- Page 75 and 76: Figure 2.20: Schematic diagram of t
- Page 77 and 78: Chapter 3 A study on RF sputtered S
- Page 79 and 80: (a) Deposition rate. (b) Refractive
- Page 81: Thus, by knowing the refractive ind
- Page 85 and 86: P Ar (mTorr) P H2 (mTorr) r H (%) 1
- Page 87 and 88: such a peak was witnessed in [Quiro
- Page 89 and 90: the host SiO 2 matrix leading to an
- Page 91 and 92: initiated in this thesis, for the g
- Page 93 and 94: P Si (W/cm 2 ) x = 0/Si Bruggemann
- Page 95 and 96: tel-00916300, version 1 - 10 Dec 20
- Page 97 and 98: Figure 3.15: Absorption coecient cu
- Page 99 and 100: Aim of the study To see the inuence
- Page 101 and 102: tel-00916300, version 1 - 10 Dec 20
- Page 103 and 104: It can be seen that there is a sign
- Page 105 and 106: tel-00916300, version 1 - 10 Dec 20
- Page 107 and 108: 3.8.4 Inuence of SiO 2 barrier thic
- Page 109 and 110: Chapter 4 A study on RF sputtered S
- Page 111 and 112: 4.2.2 Structural analysis (a) Fouri
- Page 113 and 114: tel-00916300, version 1 - 10 Dec 20
- Page 115 and 116: tel-00916300, version 1 - 10 Dec 20
- Page 117 and 118: tel-00916300, version 1 - 10 Dec 20
- Page 119 and 120: (a) FTIR spectra of NRSN, Si 3 N 4
- Page 121 and 122: tel-00916300, version 1 - 10 Dec 20
- Page 123 and 124: Figure 4.15: Absorption coecient sp
- Page 125 and 126: A multilayer composed of 100 patter
- Page 127 and 128: multilayered conguration. Therefore
- Page 129 and 130: around 1250 cm −1 . The blueshift
- Page 131 and 132: tel-00916300, version 1 - 10 Dec 20
T d (°C) ν T O3 x = 0/<strong>Si</strong><br />
from FTIR<br />
<strong>Si</strong> excess (at.%)<br />
from FTIR<br />
(unbon<strong>de</strong>d <strong>Si</strong>)<br />
x = 0/<strong>Si</strong><br />
from ellipsometry<br />
<strong>Si</strong> excess (at.<br />
%)<br />
from<br />
ellipsometry<br />
(agglomerated<br />
<strong>Si</strong>)<br />
200 1059.6 1.77 4.15 1.92 2.15<br />
300 1061 1.79 3.8 1.92 2.15<br />
400 1064 1.82 3.15 1.91 2.37<br />
500 1067 1.86 2.539 1.76 4.98<br />
Table 3.2: <strong>Si</strong> excess estimation by FTIR and refractive in<strong>de</strong>x (Bruggeman method- analysis<br />
with regard to T d .<br />
(d) Growth mechanism<br />
tel-00916300, version 1 - 10 Dec 2013<br />
It has been reported that r d increases with increasing T d [Gates 89, Kaiser 98]. Our<br />
results indicate a steadily <strong>de</strong>creasing r d which can be related to a continuous increase<br />
in the etch rates. But, n 1.95eV increases linearly with T d indicating an increase in<br />
the <strong>Si</strong> content. This necessitates an un<strong>de</strong>rstanding of the temperature <strong>de</strong>pen<strong>de</strong>nt<br />
growth mechanism. The following mechanism is proposed <strong>base</strong>d on the kinetic<br />
theory of gases and temperature <strong>de</strong>pen<strong>de</strong>nce of <strong>Si</strong>H x etch products. A diagrammatic<br />
representation of the proposed explanations are shown in gure 3.3.<br />
According to classical mechanics, the kinetic energy is a function of the particle<br />
mass (m) and velocity (v) expressed as,<br />
E k = (1/2)mv 2 ⋍ (3/2)kT Eqn. (3.9)<br />
where (3/2)kT is the averaged kinetic energy of a classical i<strong>de</strong>al gas per <strong>de</strong>gree<br />
of freedom related to temperature T, and k is the Boltzmann constant. Consi<strong>de</strong>ring<br />
our sputter gases, Ar is monoatomic having three <strong>de</strong>grees of freedom and hydrogen<br />
being a diatomic gas has 6 <strong>de</strong>grees of freedom.<br />
It can be seen from equation 3.9 that an increase in temperature increases the<br />
average kinetic energy and therefore the velocity of the species in the sputter chamber.<br />
When the substrate is heated, it cannot be <strong>de</strong>nied that a certain amount of<br />
energy is introduced in the plasma leading to the plasma temperature. Thus it can<br />
be said that with increasing T d , E k(avg) and velocity of species in the plasma increases<br />
leading to an enhancement in the kinetics of sputtering process. Moreover,<br />
in a mixture of particles with dierent masses, the heavier atom will have a lower<br />
velocity than the lighter one but have the same average kinetic energy. Hence, at a<br />
given T d hydrogen having the lowest mass acquires the greatest velocity.<br />
64