We study the field emission characteristics of large-area films of crystalline microbelt grown on silicon substrate by thermal evaporation in air using a commercial infrared sintering furnace. It is found that their turn-on field, threshold field, resistance to microdischarge and field emission current stability are better than nanowires, nanobelts and nanoflower. In addition, good uniform distribution of field emission sites can be observed. The physical reasons are explained responsible for such improvements on field emission characteristics of material. These results indicate that large-area microbelts may be suitable for cold-cathode electron source application.

1. Introduction

Molybdenum trioxide, a wide-bandgap n-type semiconductor, has been studied as an interesting photochromic and electrochromic material with potential application in information display, sensor device, optoelectronic storage device, and smart window [112]. Recently, various forms of molybdenum trioxide nanostructure have been reported to exhibit attractive field characteristics, indicating their potential application as cold cathode electron emission material [1316]. Molybdenum trioxide nanostructures can be produced under some conditions including solution method [17], hydrothermal method [13], and thermal evaporation [1416]. The synthesis and the growth mechanism of nanostructures have been investigated [1321]. Recently, we developed an approach to prepare a large-area film of nano/microstructures. With some degree of control, we were able to prepare large-area films of microbelts in air and on ITO glasses or silicon substrates at low temperature without catalyst. In particular, the microbelt films have been found to exhibit improved field emission characteristics as compared to those reported for various nanostructures [1316]. Here we report these findings and the possible physical reasons.

2. Experimental

Our sample preparation procedure is as below. ITO glasses or silicon substrates were washed with acetone and then with alcohol in an ultrasonic cleaner. The whole growth process was carried out in a commercial infrared sintering furnace, which comprises 6 temperature zones and their temperatures can be set to have a value in a range from room temperature to C. In the present study, the temperature in different zones was set at C, C, 350°C, C, C, and C, respectively. When the temperatures reached the set values, the substrates (size: ) and the quartz boat containing powders were carried to the set zones by the transmission belt. With reference to Figure 1, the evaporation source was placed at the right-hand side of the C zone and the three substrates were located at the positions with the separations between substrate and source of 27 cm, 32 cm, and 37 cm, respectively. In this arrangement, Sample 1 with separation 27 cm was in the 350°C zone, Sample 2 with separation of 32 cm in right hand side of the C zone, and Sample 3 with separation of 37 cm in the C zone. Thus, with reference to Figure 1, Samples 2 and 3 were in the regions where temperatures were below C. The growth time was 60 minutes. Finally, all the temperature zones were allowed to decrease gradually to room temperature. The substrates appeared white after deposition. The combination of the large area of each temperature zone () and uniformity in the temperature inside the zone may be very important to the development of a large-area film for device application.

3. Results and Discussion

The microbelts were characterized by using X-ray diffraction spectroscopy (D/max 2200 vpc apparatus with Cu radiation), scanning electron microscopy (SEM: Quanta 400 F), and high-resolution transmission electron microscopy (HRTEM: JEM-2010HR). The SEM image (Figure 2(a)) shows that the average width of microbelts is ~1 m and their length is up to 10 m. The microbelts are straight and their upper ends may be in the rectangular flat shape or have irregular one with sharp corners and nanotips as shown in the high resolution SEM images (insets of Figure 2(a)). The typical XRD pattern (Figure 2(b)) shows the diffraction peaks can be indexed to the orthorhombic structure with the lattice constants  nm, nm,  nm (JCPDS: 5-0508). Figure 2(c) is the low-magnification TEM image of the microbelts. The high resolution TEM (HRTEM) image (Figure 2(d)) shows the two sets of parallel fringes with a spacing of 0.396 nm and 0.37 nm corresponding to the (100) and (001) planes, respectively. The SAED patterns (inset of Figure 2(d)) shows that the entire microbelt is single crystalline with a growth direction of [].

The field emission measurements of microbelts were carried out in a vacuum chamber of  torr at room temperature. The substrate with microbelts was first adhered to the surface of an oxygen-free, high-conductivity copper disc. A transparent anode, consisting of a quartz plate 4 cm in diameter and coated with conducting indium doped tin oxide film, was placed in front of, and parallel to, the surface of the sample cathode. First, we investigated how the field emission characteristics of the microbelt film may be affected by the arrangement of substrate and evaporation source. Figure 3 shows J-E characteristics and the corresponding F-N plots of the three samples. Sample 1 has poor emission performance and Samples 2 and 3 show attractive characteristics. The turn-on fields (required to obtain field emission current of 10 A/cm2) of Sample 1, 2, and 3 are 4.6, 2.6, and 3.0 V/m, respectively, and the threshold fields (required to obtain field emission current of 10 mA/cm2) of Samples 2 and 3 are 6.4 and 6.5 V/m, respectively. Compared to the reports earlier, the best turn-on field of the film of microbelts (2.6 V/m) is lower than that of the nanowires (3.5 V/m) [14, 15], nanobelt (8.7 V/m) [16], and nanoflower (4.3 V/m) [13]. This result indicates that the microbelt is a good candidate as cold-cathode material. The good field emission properties of the samples may be attributed to a number of factors, and the first may be the field enhancement effect. The values of field enhancement factor of the microbelt samples were calculated by using the F-N formulation [22] where is the work function of the molybdenum trioxide, is the slope of the F-N plot. The value of the factor may be derived by using the above expression provided that is constant. In the calculation, the work function of MoO3,  eV [23], is used. From the F-N plots, we obtain that the values of field enhancement factor for Samples 1, 2, and 3 are 1822, 4165 and 3861, respectively. The -value of Sample 1 is much smaller than that of the other two samples. Figure 4 shows the optical images of emission site distribution recorded using transparent anode, and it is found that the emission sites of Samples 2 and 3 are distributed over the entire sample surface, while for Sample 1 the number of emission sites and their intensity are relatively low. These differences and similarity may be explained by the observed morphology features of the samples as seen under scanning electron microscope. The top-down SEM images (Figures 5(a) to 5(f)) show that the averaged length of the microbelts of Sample 1 (Figures 5(a) and 5(d)) is longest compared to that of the other two samples (Figures 5(b), 5(c), 5(e) and 5(f)), but the corresponding cross-section image (Figure 5(g)) shows that it grows leaning to the substrate, which is not useful to the field enhancement. On contrary, the orientation of the MoO3 microbelts of Samples 2 and 3 (Figures 5(h) and 5(i)) is almost vertical to the substrate surface; this can enhance the field at the top ends of these vertically aligned microbelts, and thus strong emission from the apexes of these microbelts may be expected. Thus, these two samples have large -values. This observation is in consistence with earlier findings that have revealed that vertically aligned nanowires have better field emission properties because of higher field enhancement at the end of each nanowire [14, 24].

As to the difference between the orientations of the microbelts of Sample 1 and Samples 2 and 3, a brief explanation is given below. Because of thermal radiation, the closer the substrate to the evaporation source is, the higher temperature the substrate is. When the temperature of the substrate is higher, the nucleation for microbelt growth is more difficult in comparison with a lower temperature one. Thus, the density of microbelts is lower for high temperature substrate, and when is too low, microbelts cannot support each other, so they cannot grow vertically to the substrate surface. This can explain why microbelts of Sample 1 grew leaning to the substrate, which was placed at the position less than 27 cm to the source, and was much closer to the source compared to Samples 2 and 3.

Additionally, we should like to explain why the calculated -values are so large. We believe that this is not all contributed by the geometrical field enhancement. Other factors may be responsible to this, such as reduction in surface potential barrier for field emission or hot electron emission [22]. These are highly possible for field emission from semiconductors such as MoO3. If these are the cases, we may have to replace the value of with a more realistic value that reflects the lowering of the surface potential barrier in the calculation using the F-N formulation. This reduction in the value of can significantly decrease the -value. But all these have to be confirmed by further experiments.

Second, using samples grown under the similar conditions described above with separation 32 cm, we found that the films of MoO3 microbelts are resistant to the effects of local microdischarge event. Figure 6 shows three optical images of emission site distributions of one sample taken under three gap field conditions. One may see that after the local microdischarge events (Figure 6(b)), the emission sites distribution was not obviously changed, by comparing Figures 6(a) and 6(c).

Finally, we observed stable emission of the films over 3 hours time duration. The typical field emission I-t curve is shown in Figure 7. No significant decrease of current was observed and the fluctuation is about 2% at a constant electric field of 4.47 V/m for an average emission current of 330  A.

4. Conclusion

In summary, we have demonstrated that large-area films of crystalline MoO3 microbelts may be grown on ITO glasses or silicon substrates by thermal evaporation in air using a commercial infrared sintering furnace. Their typical turn-on field as low as 2.6 V/m and threshold field of 6.4 V/m are superior to the values reported for MoO3 nanostructures. Their field emission performance may be affected by the orientation of the microbelts; the microbelts vertical to the substrate have a better field emission property. In addition, the good uniform distribution of emission sites on large-area samples is also observed. Finally, the MoO3 microbelts show good field emission current stability over time.


The authors gratefully acknowledge the financial support of the project from the National Natural Science Foundation of China (Grant nos. U0634002, 50725206, 60571035, and 50672135), Science and Technology Ministry of China (National Basic Research Program of China: Grant nos. 2003CB314701, 2007CB935501 and 2008AA03A314), the Science and Technology Department of Guangdong Province, the Department of Information Industry of Guangdong Province, and the Science and Technology Department of Guangzhou City.