Influence of Forced Diffusion of Boron on Electrical Conductivity of Diamond Films

  • PDF / 264,642 Bytes
  • 5 Pages / 414.72 x 648 pts Page_size
  • 103 Downloads / 176 Views

DOWNLOAD

REPORT


deionized water to eliminate the non-diamond carbon on the surface. Then, the samples were cleaned in a hot HNO 3 + HCl + H20 solution for 20 minutes and rinsed with deionized water to purge the surface of metal impurities. Boron powder was used for dopant source and mixed with

carbon paste to assure good electric contact. These samples were placed on a graphite support which was heated by tungsten wire. The temperature of support was monitored by a K-type thermal couple. The general arrangement for forced diffusion with optical and thermal ionization is presented in Figure 1. The chamber was flushed with hydrogen gas at one atmosphere. After diffusion, the samples were cleaned again by the same cleaning procedure described above. The resistance measurements were performed from room temperature to 350 'C in air.

The two-electrode scheme was used for measurements. Two parallel carbon paste strips, 8 mm in length, 1 mm in width and separated by 1.5-2 mm, were applied on the surface of the samples. Figure 2 presents the apparatus of conductance measurements. 649

Mat. Res. Soc. Symp. Proc. Vol. 423 ©1996 Materials Research Society

Diamond film

Laser Beamn Diamond film

1Graphite Heater

Boron powder mixed with carbon paste

Figure 1. The arrangement of forced diffusion with optical and thermal ionization.

V Carbon paste contacts

HP 4140B

A

Diamond film

Figure 2. The arrangement of conductance measurements A computerized Hewlett-Parkard picoammeter with build-in dc voltage source, model 4140B, was used. In the range of bias from -I V to 1 V, the contact had an ohmic behavior over all temperature range. The conductivity measurements were performed while cooling the sample, in order to minimize the influence of contact change with temperature. RESULTS AND DISCUSSIONS: Raman spectrum of all these samples shows the diamond peaks at 1341 cm' and the broad graphite peaks at 1530 cm'. The summary of diffusion conditions and electrical measurements are shown in Table 1. The sign of bias is on the surface which has the dopant source. Figure 3. presents the plots of conductance vs. temperature for 4 samples. It is found that the activation energy of the samples before diffusion is about 1.0 eV and the resistance is larger than 4x10" Q (limit of the measurement system). These values are characteristic of undoped diamond 78, . The activation energy AE was determined using the formula G = Go exp(-AE/kT), where G is conductance; Go is a constant; k is Boltzman constant; T is temperature. The diffusion of boron due to concentration gradient (for simplicity, named "diffusion" later on) does not produce any change in activation energy. After diffusion in the sample 1 and 2 (step 1) the activation energy remained 1.0 eV, the same as the non-diffusion sample. The change of the electrical conductivity approximately one order of magnitude might be due to thermal annealing of the samples during the diffusion process. We assume that boron did not enter substitutionally the diamond lattice during the diffusion process. The change in