2.   Model Description

3.   Results

3.1       Cell Geometry and Driving Waveform

The geometry of the PDP cell used in the simulations is shown in Figure 1a. The relative permittivity of the dielectrics is 10. The gas mixture filling the region between the dielectrics is a 4% Xe-Ne mixture at a pressure of 500 Torr. The secondary electron emission coefficients for Ne and Xe ions on MgO are taken to be 0.5 and 0.01 respectively. The width of the cell L is 1260 mm, the gas gap length D is 150 mm, and the thickness of the dielectric layers d is 30 mm. The width w of the X and Y electrodes is set to 300 mm.

In order to compare our results with the results reported in [3], we only apply sustain pulses between the X and Y electrodes, as is shown in Figure 1b. We do not include the address electrode in the simulation results reported here. The driving voltage waveform consists of a sequence of 4 ms pulses (Figure 1b). The rise and fall times of all pulses are 100 ns. A pulse of amplitude V_i is first applied on the X electrode to cause breakdown and form an initial charge distribution on the surface of the upper dielectric layer. This is followed by a sequence of alternating sustaining voltage pulses of amplitude V_s.

electrode acts as the cathode. In both cases, snapshot 1, corresponding to the initial stage of the discharge, shows that the spatial distribution of the Xe excitation is mostly confined at the interface between the expanding plasma and the ion sheath formed below the dielectric surface on the cathode side. In snapshot 2, corresponding to the maximum in power dissipation during the discharge, we observe that the Xe excitation power density is high both in the ion sheath–plasma interface and in the bulk of the plasma. The local maximum of excitation power in the bulk plasma region is close to the anode. Snapshot 3, corresponding to a later stage of the discharge, shows decrease of the excitation power density in both cases. The decrease is greater in the bulk of the plasma. The different behavior of the discharge for the two cases is illustrated in snapshots 4 and 5. In the standard case, the decrease of the excitation power density continues and the discharge power pulse smoothly decreases, as we observe in snapshot 4. The region of high Xe excitation is mostly confined at the ion sheath–expanding plasma interface. In the new structure case, a quite different behavior is observed. When the sheath–plasma interface reaches the cell area below the floating electrode, the discharge power increases. Snapshot 4 corresponds to the second maximum in power dissipation. We observe the spatial expansion of the high Xe excitation area towards the anode region with respect to the previous snapshot. Snapshot 5 corresponds to the final stage of the discharge in both cases. We observe that the region of high Xe excitation at the sheath–plasma boundary has shrank in both cases. In the standard case, the plasma expands until all the dielectric surface area below the cathode (X electrode) is covered with charge to screen the electric field of the 4^th voltage pulse. In the new structure case, the plasma essentially expands only until it covers with charge the dielectric surface below the cathode before reaching the area below the floating electrode. In this area, the electric field of the voltage pulse is screened by charges on the floating electrode so that the corresponding dielectric surface does not have to be covered with charge. This also explains the shorter overall duration of the power pulse in the new structure case. 

In Figures 3a and 3b, we show the dissipated energy density during the application of the 4 pulses (Figure 1b) for the standard and new structure cases respectively. In both cases, high dissipated energy density is observed in the regions directly below the dielectric layer covering the X and Y electrodes. This is expected since the local maxima of power density are always located in these regions. As we noted above, during each discharge there are two regions of high power dissipation. The first is in the ion sheath–plasma interface below the dielectric surface on the cathode side and the second is in the bulk of the plasma exhibiting a local maximum in the area below the dielectric surface on the anode side. We also observe that the high dissipated energy density region is more localized in the new structure case. This result can be explained by the partial screening of the voltage pulse electric field by the floating electrodes described above in detail (snapshot 5 in Figure 2d).

In Figures 3c and 3d, we show the total UV emission energy density integrated over all wavelengths considered (147 nm, 150 nm, and 173 nm) during the application of the 4 pulses (Figure 1b) for the standard and new structure cases respectively. We observe that UV emission is confined in the regions below the dielectric surface covering the X and Y electrodes. However, in both cases the region of high UV emission is wider when compared to the region of high dissipated energy density (shown in Figures 3a and 3b), extending towards the lower dielectric layer. This result is due to diffusion of some of the excited states of Xe which have long lifetimes. In addition, UV emission is more localized in the new structure case, since the region where power is spent by electrons in Xe excitation is more localized, as explained above (Figures 3a and 3b). The discharge current and UV emission confinement of the discharge by the new structure is expected to reduce cross talk between adjacent cells in agreement with the experimental findings reported in [3].

3.3       Parametric Studies

We now use the two-dimensional model to perform some parametric studies for the standard and new structure cases. We calculate the breakdown voltage for the geometry shown in Figure 1a. The breakdown voltage is directly related to the firing voltage V_f of the cell. We also calculate the UV efficiency, defined as the ratio of total UV energy spatially integrated over the gap to the total dissipated power.

In Figures 4a and 4b, we present results for the standard geometry, without inclusion of the floating electrodes shown in Figure 1a. In Figure 4a, we show the dependence of the breakdown voltage V_br on the coplanar electrode gap length g. V_br is an increasing function of g, as expected for the range of values of g studied. In Figure 4b, we show the dependence of the UV efficiency on the coplanar electrode gap length g. In all three cases (g= 80, 100, 140 mm), the amplitude of the initial pulse V_i (Figure 1b) is set to 25 V above the corresponding breakdown voltage V_br, and the amplitude of the sustain pulses V_s is set to 35 V below the corresponding breakdown voltage. The efficiency is an increasing function of g. However, increasing the gap length g is not a practical way of increasing the efficiency, since the operating voltages also increase.

In Figures 4c, 4d, 4e, and 4f we show results for the new geometry case (with floating electrodes). The coplanar electrode gap length g (Figure 1a) is set equal to 100 mm. In all cases, efficiency is measured by setting the amplitude of the initial pulse V_i (Figure 1b) to 25 V above the corresponding breakdown voltage V_br, and the amplitude of the sustain pulses V_s to 35 V below the corresponding breakdown voltage, as was also done in the standard geometry case.

In Figures 4c and 4d, we show the effect of the variation of the vertical distance d_f between the sustain (X and Y) electrodes and the floating electrodes on the breakdown voltage and the efficiency. We choose the floating electrodes’ width w_f to be equal to 150 mm. We observe that the breakdown voltage exhibits essentially no variation with d_f. In addition, the breakdown voltage is almost equal to the corresponding one of the standard geometry. This result is in agreement with the experimental results reported in [3]. The UV efficiency slightly decreases with d_f; however, when compared with the UV efficiency of the standard geometry for the same electrode gap length g, it is found to be almost the same. Thus, the substantial increase of efficiency reported to be measured in [3] is not reproduced by the numerical simulation. 

4.                  Summary

5.    References

[1]     Punset, C., S. Cany, and J.-P. Boeuf, Addressing and sustaining in alternating current coplanar plasma display panels, J. Appl. Phys., 86, 124, 1999.

[2]     Rauf, S., and M. J. Kushner, Dynamics of a coplanar-electrode plasma display panel cell. II. Cell optimization,  J. Appl. Phys., 85, 3471, 1999.

[3]     Kim J. S., C. H. Jeon, E. C. Lee, Y. J. Ahn, S. D. Kang, S. Y. Ahn, Y. K. Shin, J. H. Ryu, J. D. Schemerhorn, Application of a new panel structure for high luminous efficiency in AC-PDPs, SID Digest, 00, 102, 2000.

[4]     Morgan W. L., and B. M. Penetrate, ELENDIF: A time-dependent Boltzmann solver for partially ionized plasmas, Comput. Phys. Commun., 58, 127, 1990.

[5]     Ventzek P. L. G., R. J. Hoekstra, and M. J. Kushner, Two-dimensional modeling of high plasma density inductively coupled sources for materials processing, J. Vac. Sci. Technol. B, 12, 461, 1994.