| 
                        
                 | 
            
            
                
                        
                        
                                
                                        
                                                1999 Philips S1000
                                         | 
                                 
                                
                                        
                                                 
In 1992, Fusion Systems (USA) presented their sulfur lamp technology at the 6th international symposium on the science and technology of light sources, held in Budapest, Hungary. The system was unlike anything else in the lighting market, it consisted of a spherical quartz bulb containing a superatmospheric (~5 bar) thermal plasma burning in an atmosphere of argon and sulfur and powered with 3400 W of microwave energy at 2.45 GHz. This caused quite a stir as the new technology was shown to radiate 410 klm of white light at 6500 K with a CRI of 86 Ra8, corresponding to an efficacy of 120 lm/W at lamp level. Another benefit of such system is a potentially long lamp service life with a virtually constant flux output due to the lack of electrodes. Fusion was quick to develop the technology further, with support from the US department of Energy, which resulted in a spinoff company, Fusion Lighting, established in 1994, and in various demonstrators installed in the field. This sudden move pushed lighting companies to investigate the technology during the second half of the 1990s. 
 
The Philips S1000 shown here was made at the company’s central lighting lab in Eindhoven, the Netherlands, for scientific studies done both at the Technical University of Eindhoven and at Philips (in Eindhoven and Aachen, Germany). This light source was made according to the published specifications of Fusion’s second-generation sulfur lamp, optimined in the mid-1990s. Compared to the original sulfur lamp, the bulb diameter was increased by 43 % to 40 mm while the input power density was decreased by 88 % to about 30 W/cc. These changes, combined with a slightly reduced sulfur vapor pressure, resulted in a record-high lumen efficacy of 172 lumen per microwave watt, measured at Fusion for a lamp power input of 820 W (the light output was found to be linearly proportional to the input power coupled to the plasma). This improvement arises from reduced thermal losses due to a more favorable surface to volume ratio of the sulfur plasma. Measurements done at Eindhoven University confirmed this performance with a lumen efficacy of 175 lm/W at 1 kW of microwave input power. 
 
This technology, and chalcogenide light sources in general, are interesting for the fact that sulfur atoms combines into molecules at pressures above 1 atm. The dimer S2 is the main source of radiation in the plasma, this molecule converts energy into ultraviolet light very efficiently. Its spectral emission consists in a multitude of ro-vibrational lines which are broadened by the high gas temperature, giving a wide continuum spectrum. At the high operating pressure of sulfur lamp most of the shortwave S2 radiation is absorbed by S3 molecules which are formed in cooler parts of the discharge, near the burner wall. The resulting spectrum is a continuum whose peak emission is shifted to the green, between 500 and 540 nm, near the maximum of the eye’s sensitivity, while the amount of UV and IR is reduced to less than 15 % of the total emitted light output. Such characteristics thus confer a very high spectral efficacy to this source (273 lm/W) but the resulting light color has a noticeable greenish hue (see there). 
 
Unfortunately, this technology did not prove successful. While the light color could be easily corrected with additives, the limited efficiency of high-wattage microwave sources (Magnetrons, 65-70 % efficiency max) and their relatively short service life (~20 kh) resulted in a system not better than state-of-the-art HID systems. Fusion Lighting eventually closed shop in 2003 and other lighting companies focused their attention on other technologies like ceramic metal halide lamps and the then nascent LED systems, which eventually proved much more effective. 
 
                                         | 
                                 
                         
                 | 
        
Tuopeek - You certainly need more than one bar of vapor pressure if you want to produce white light with sulfur, which calls for a cold-spot temperature above ~450 °C. Borosilicate glass could be used, but the main issue would be the management of the discharge tube's hot spot temperature caused by the plasma. Alternatively, one could operate the burner in an oven and run a low-power discharge through the sulfur vapor. Otherwise, you can still try low-pressure discharges in sulfur vapor using standard glass tubing (with outer capacitive electrodes), but that will be a source of UV light primarily (most of it blocked off by the glass).
Another option is to fill the tube with gases which generate more heat (i.e., prone to elastic losses in the plasma), such as helium and neon. Don't go for a too low pressure as you want to maximize the voltage drop across your electrodes for maximum discharge power. This way your can heat the sulfur effectively, provided your cold spot is not too far away from the plasma (which I think is an issue in the experiment shown in your link). Also, you can keep the discharge current relatively low and rely on an external source of heat to vaporize sulfur, such as an oven or a heat gun. If you want to use the discharge's heat losses only for that purpose, then you can also add a glass sleeve around your discharge tube in order to reduce heat losses. There are many options possible there!