7.10 Development of a Compact NH_3 Jet - Conclusions

It was demonstrated how difficult it is to compare experimental fluxes obtained from the jet to theoretical predicted fluxes. It proved nearly impossible to obtain good fitting parameters as discussed earlier which could be used to predict jet performance with different initial conditions like nozzle temperature. It was also very hard to reproduce data between runs. Those difficulties are believed to have several origins, both in the way the theoretical values are derived as well as the jet design itself. The theoretical values are derived from numerous assumptions and several models, which are very sensitive to the jet design and are very idealized. The jet performance itself is very sensitive to the operating conditions. Slight variations between runs seem to have huge effects. A prime example is the pumping speed based on the croyshroud, which in turn is very sensitive to the liquid nitrogen flow. It was found that bubbles tend to get trapped in the cryoshroud, having a large effect on the pumping speed and therefore the background pressure P_b, which in turn is a very important parameter in determining centerline intensities and fluxes due to that background pressure.

It was also shown that the dissociation fraction depends strongly on the operating conditions of the jet, owing largely to the low reaction rate at low temperatures, and it is hard to measure an exact fraction with the current system. Also, dissociation not only due to the nozzle and jet expansion has to be considered, but also NH_3 dissociating due to particles making up the background pressure P_b hitting the heating coils in the region between nozzle and skimmer. Unknown fractions of dissociated NH_3 are also believed to contribute to the characterization difficulties since N_2 and H_2 are not pumped by the cryoshroud and poorly pumped by the turbo pump-duct combination.

It was found that the jet has no big advantage in terms of reducing the gas load on the system, being only a factor of 4 for typical growth conditions. Future experiments are planned comparing the growth rate and crystal quality of jet and leak valve grown material, as well as the dependence on jet parameters, especially the nozzle temperature. These experiments will be performed as soon as the jet independent growth rates are determines and a more fundamental understanding of the growth process is achieved.

Last a new simplified jet is being proposed placed closer to the substrate and eliminating the skimmer and cryoshroud. As discussed in the corresponding section such a design is expected to have definite advantages compared to the current NH_3 jet design, both in terms of growth, as well as characterization. Growth experiments exploring the effect of jet parameters on the growth are easier to analyze due to the elimination of attenuation due to the background pressure P_b, and the characterization is greatly simplified due to the same reason.

Table of Contents