A Thin Ti Capping Layer without Pre-treatment

The fabrication procedures of patterned CNT emitters were shown schematically in Figures 2-3(a) ~ 2-3(e). As shown in Figure 2-3(a), about 1-μm-thick photoresist was

spin-coated on an n-type silicon wafer (100) with low resistance and square cells with lengths of 10 μm, 100 μm, and 1000 μm were defined by photolithography. Then a Ti layer with 50-nm in thickness was deposited by the E-beam evaporation system as a buffer layer and a Fe layer with 5-nm in thickness was deposited subsequently as catalytic metal in the same system, as shown in Figure 2-3(b). The Fe/Ti (5 nm/50 nm) patterns were formed after the photoresist was removed by lift-off method as depicted in Figure 2-3(c). With different thicknesses (0, 0.5, 1, 2, and 4 nm) of a thin Ti capping layer was deposited in the sputtering system shown in Figure 2-3(d). Finally, CNTs were grown selectively by thermal CVD system to investigate the growth phenomenon and the field emission properties of CNTs that was shown in Figure 2-3(e). The samples with different Ti capping layers were loaded into the thermal CVD chamber (Figure 2-4(a) and (b)) to grow CNTs in atmospheric pressure at 700

with C

℃ 2H4 (5 sccm) for 10 min as depicted in Figure 2-4(c).

Scanning electron microscopy (SEM) was performed to discover the density and the morphology of CNTs. A high-vacuum measurement environment with a base pressure of 5x10^-6 Torr was set up to characterize the field emission properties of CNTs (Figure 2-5).

Cathode contact was made directly on the wafer. A glass plate coated with indium-tin-oxide (ITO) was positioned 120 μm to 160 μm above the sample surface as an anode. All cables were shielded except for the ground return path to the power source. The emission current densities of CNTs were measured as a function of applied electric field, using Keithley 237 high voltage units as DC source and Keithley 238 high current units as ground source. The measurement instruments were auto-controlled by the computer with IEEE 488 interface.

2.3.1.2 Results and Discussion

The corresponding thicknesses of the thin Ti capping layer were 0 nm, 0.5 nm, 1 nm, 2 nm, and 4 nm in SEM images from a 45° viewing angle as shown in Figures 2-6 (a), (b), (c), (d), and (e), respectively. From the SEM observation, the density of CNTs decreased if the

thickness of the thin Ti capping layer increased. As a speculation, the thin Ti capping layer could hinder the growth of CNTs as a diffusion barrier. With thicker Ti capping layer, the probability of carbon radicals to diffuse into the catalyst would be reduced therefore reduced the growth of CNTs. To estimate the density of CNTs, the SEM images were taken from top view and were shown in Figures 2-7 (a) ~ (e). To calculate the quantity conveniently, the pictures were divided into one fourth and then estimated the number of CNTs. Afterwards, the number which was multiplied by four was approximate density value. The approximate densities of CNTs could be counted and listed in Table 2-1(a). For the case of 0-nm-thick Ti capping layer, the density of CNTs was about 8.8x10⁸ emitter/cm². With the increase of thickness of Ti capping layer, the density of CNTs was getting low. When the thickness of the thin Ti capping layer was increased to 4 nm, the density of CNTs had already dropped to 3x10⁶ emitter/cm².

The field emission properties of CNTs with different thicknesses of the thin Ti capping layer were shown in Figure 2-8(a) which showed the relationship between emission current density and electric field (J-E). The corresponding Fowler-Nordheim plots for CNTs were depicted in Figure 2-8(b) and the linearity slope of the plots confirmed the field emission phenomena. From Table 2-1, the turn-on field defined at 10 μA/cm² of CNTs decreased from 4.071 V/μm to 3.545 V/μm with the thickness of Ti capping layer was increased from 0 nm to 2 nm. From the SEM images, the density of CNTs also dropped from 8.8x10⁸ to 8x10⁶ emitter/cm². This phenomenon may results from the suppression of the screening effects which can increase the field enhancement factor (β) hence reduce the turn-on field. However, the turn-on field did not always decrease with the increase in the thickness of the thin Ti capping layer. When the thickness of the thin Ti capping layer was added to 4 nm, the turn-on field would rise to 4 V/μm resulting from the decrease in the length of CNTs which would cause the reduction of aspect ratio and make the decline in the field enhancement factor (β).

The field emission current density was improved from 1.568 mA/cm² to 10.2 mA/cm² at

electric field of 6 V/μm when the thickness of the thin Ti capping layer was from 0 nm to 2 nm, respectively. With a 4-nm Ti capping layer, the field emission current density dropped greatly as compared to the 2-nm one. It might result from the great reduction of emitting sites which could reduce the field emission current density effectively. With changing the thicknesses of the thin Ti capping layer from 0 nm to 4 nm, the field emission current density of CNTs was achieved to threshold field, which was defined at 10 mA/cm², as the thickness of the thin Ti capping layer was 2 nm. Table 2-1(b) showed the summarized field emission properties of CNTs grown with different thicknesses of the thin Ti capping layer. The simulations of literature [2.23] predict that an inter-tube distance of about 2 times the height of CNTs optimizes the emitted current per unit area. In this experiment, the length of CNTs was probably 1 μm and this would correspond to an ideal density of 2.5x10⁷ emitter/cm² by simulation. However, the optimum thin Ti capping layer thickness, 2 nm, was found out in this experiment and the density of CNTs of this sample was about 8x10⁶ emitter/cm² which was similar to paper survey [2.23].

2.3.1.3 Summary

The density of CNTs could be controlled with different thicknesses of the thin Ti capping layer without pre-treatment and reduced the turn-on field and improved the field emission current density due to the suppression of screening effects. Low density could reduce the turn-on field, however, the number of the emitting sites would also be reduced hence cause the reduction in field emission current density. So it needed an optimized the thickness of the Ti capping layer to gain a reasonable field emission current density. In this experiment, the optimum thickness of thin Ti capping layer was found out to be 2 nm and the density of CNTs was altered from 8.8x10⁸ emitter/cm² to 8x10⁶ emitter/cm². The field emission current density was improved from 1.568 mA/cm² to 10.2 mA/cm² at the electric field of 6 V/μm and the turn-on field was decreased from 4.071 V/μm to 3.545 V/μm with the thickness of the Ti

capping layer increased from 0 nm to 2 nm. As a consequence, the experimental results revealed that the improvement of the field emission properties could be achieved by optimizing the density of CNTs with proper thin Ti capping layer thickness.

2.3.2 A Thin Ti Capping Layer with Pre-treatment