Thicknesses of the films were measured in a surface profiler KLA Tenkor P-7, while their surface morphologies were measured in an AFM Nanosurf easyScan DFM. The quantitative results are listed in the Table 1. According to these results, the hydrogen dilution affects significantly the deposition rate, almost duplicating its value in some cases, going from 6.30 to 11.95 nm/min. A similar effect is appreciated when the X_CH4 is decreased, but in this case the change occurs in much lower proportion. In contrast, the roughness is mainly affected by the power density of deposition. At power density of 25 mW/cm² all the films have a root-mean square roughness less to 1 nm, except the aSiC-14 process, whose roughness is affected by the doping level. None temperature tendency is evident in these results.

The structural properties of the a-Si_xC_1-x:H films were studied using a FTIR spectrometer Bruker Vector 22, operating in absorbance mode in middle infrared (500 - 3500 cm^-1). The absorption spectra were normalized by the corresponding thicknesses and the assignment of peaks in the spectra are listed in Table 2. According to the results in Figure 1, the peaks at 1000 cm^-1, 1245 cm^-1 and inside the band from 2860 to 2960 cm^-1 decreases, while the peak in the band from 630 to 650 cm^-1 increases, when Z_H2 goes from 2.7 to 9. Furthermore, the peak in the band from 740 to 800 cm^-1 becomes more pronounced than the peak at 640 cm^-1 when the X_CH4 goes from 0.70 to 0.85. From the viewpoint of bonding, significant changes are not observed about power density (see also Figure 2). These statements may be corroborated based on densities of Si-C, Si-H and C-H bonds (N_SiC, N_SiH and N_CH) presented in Table 1, which were calculated following the procedure used in ^[13].

Table 2

Assignment	Wavenumber (cm-1)	References
Si-Hn wagging	630-650	[10, 11]
Si-C stretching, Si-CHn wagging	740-800	[2, 10-13]
Si-H2 rocking, scissors	845, 895	[11]
CHn wagging, Si-CH3 bending	1000, 1245	[10, 11, 13]
C-CH3, CHn bending and scissors	1370, 1450	[11]
C=C stretching	1540, 1630	[11, 12]
Si-H2 stretching	2080	[10- 13]
CHn sp3 stretching	2860-2960	[10- 13]

Figure 1

Figure 2

The FTIR spectrum of the film corresponding to the aSiC-14 process in Figure 2 includes two peaks at 975 and 1150 cm^-1, which are assigned to the wagging and bending vibration modes of B-H bonds, respectively ^[14]. In the other spectra of the B-doped films only one shoulder at the left of the peak at 1000 cm^-1 is observed, while no peak at 1150 cm^-1 is displayed. The peaks in the band 2350 -2650 cm^-1 corresponding to B-H_n bonds do not appear ^[15].

The UV-visible spectroscopy was carried out in a spectrophotometer Perkin Elmer Lambda 3 from 190 to 900 nm, in order to obtain some optical properties of the films and correlate them with deposition parameters. Using the transmittance spectra and the Tauc equation given by ^[16]:

α h v 1 / 2 = B h v - E g o p t , (1)

where h is the Planck's constant and v is the radiation frequency, the optical band gap (E _gopt ) and the Tauc parameter (B factor) were found. B factor is a parameter that indicates the amount of disorder in a film ^[17].

The data fitting procedure using (1) resulted in the values summarized in Table 3. E _gopt decreases around of 0.16 eV and 0.24 eV for X_CH4 of 0.7 and 0.85, respectively, when Z_H2 changes from 2.7 to 9.0. In undoped films, a trend is observed in B factor versus temperature under the same other deposition parameters, its value is higher at 200 °C than at 150 °C in the majority of cases. Regarding other deposition parameters (X_CH4, Z_H2 and power density), according to the results, their effect on B factor depends of the state of the remaining parameters. In B-doped films, the doping level decreases both E _gopt and B factor.

Table 3

PECVD process	Z H2	T(°C)	E gopt (eV)	B (eV cm) -1/2	E σ (eV)	σ 0 (S/cm)	σ r t (S/cm)
Undoped a-SixC1-x:H films: XCH4 = 0.7, 50 mW/cm2
aSiC-01	2.7	150	2.05	497.75	0.73	954.90	1.05 × 10-9
aSiC-02	9.0	150	1.90	489.34	0.73	1225.10	1.47 × 10-9
aSiC-05	2.7	200	2.01	514.58	0.67	205.35	3.98 × 10-9
aSiC-06	9.0	200	1.84	498.28	0.73	1166.05	1.08 × 10-9
Undoped a-SixC1-x:H films: XCH4 = 0.85, 50 mW/cm2
aSiC-03	2.7	150	2.47	426.24	0.70	369.57	5.55 × 10-10
aSiC-04	9.0	150	2.23	523.64	0.74	2024.49	9.57 × 10-10
aSiC-07	2.7	200	2.43	466.99	0.73	975.36	6.71 × 10-10
aSiC-08	9.0	200	2.19	513.39	0.75	3757.39	1.17 × 10-9
Undoped a-SixC1-x:H films: XCH4 = 0.85, 25 mW/cm2
aSiC-09	2.7	200	2.42	534.74	--	--	--
aSiC-10	9.0	200	2.17	491.02	0.71	573.36	1.02 × 10-9
PECVD process	Yb (%)		E gopt (eV)	B (eV cm) -1/2	E σ (eV)	σ 0 (S/cm)	σ r t (S/cm)
B-doped a-SixC1-x:H films: XCH4 = 0.85, ZH2=9.0, 25 mW/cm2, 200oC
aSiC-11	0.5		1.96	449.68	0.41	0.81	1.72 × 10-7
aSiC-12	1.0		1.81	385.99	0.35	0.29	3.95 × 10-7
aSiC-13	2.0		1.80	395.70	0.33	0.73	2.72 × 10-6
aSiC-14	2.5		1.77	379.61	0.33	0.35	1.26 × 10-6

Electrical dark conductivity measurements were carried out in the range of 300 - 440 K with a step of 10 K within the chamber of a Janis Research cryostat maintaining a pressure of 70 mTorr. Current-voltage measurements were made by using a Keithley 6517A and a LakeShore 331 temperature controller. Before measurements, the structures of undoped and B-doped a-Si_xC_1-x:H/titanium on glass were annealed at 140 °C in nitrogen atmosphere for 2 hours to improve the contact between the material and titanium. In the Figure 3 can be seen that the curves are shifted upward by the change in hydrogen dilution from 2.7 to 9.0. This effect is more apparent in the films deposited at X_CH4 = 0.85 than X_CH4 = 0.70. In contrast, the B-doping not only shifts upward the curve but also changes its slope.

Figure 3

The values of conductivity for undoped and B-doped films calculated using the current-voltage measurements fall on one straight line; this straight line represents a thermally activated conduction regime and can be described by the following equation ^[18]:

σ D = σ 0 exp ⁡ - E σ / k B T , (2)

where σ D is the dark electrical conductivity, σ 0 is the prefactor, E σ is the activation energy, k _B is the Boltzmann constant, and T is the temperature. The Table 3 shows the results obtained by fitting the measurement data to (2). The results of this procedure show minor changes in E σ and major changes in σ 0 by changes in hydrogen dilution, and major changes in E σ and minor changes in σ 0 by B-doping.