Updated 6-VII-2003
Mercury Vapour
Mercury Pressure
Mercury Spectrum
Lamp Nomenclature
Timeline of Developments
Mercury Vapour
J.T. Way
Küch and Retschinsky
MA Medium Pressure
MB High Pressure
MC Low Pressure
MD Water-Cooled
ME Super Pressure
UHP Ultra High Pressure
Mercury Vapour
Fluorescent Coated Lamps
Tungsten Ballasted Lamps
Lamp Electrodes
Additives to the Arc
Electrodeless Designs
Future Developments
Mercury Vapour
High Pressure Circuits
Low Pressure Circuits
Electronic Operation

Sulphide Phosphors

The first application of a phosphor coating to the outer bulb of a mercury lamp was performed in England, and was the development of the GEC Hirst Research Centre in Wembley.  This was most appropriate in view of the fact that the same laboratory created both the earliest fluorescent tubes and the first practical high pressure mercury lamp just a few years earlier.  The goal of the development was to find a phosphor which could generate some reddish colour light to complement the predominately blue-green light of the mercury discharge, thereby increasing its colour rendering properties.

  The 'MAF Mercury Fluorescent Lamp

The year was 1937, and the first lamp to receive the new phosphor treatment was the biggest seller of the day - the 400W MA medium-pressure source with a hard glass arc tube.  The luminescent material was zinc cadmium sulphide with a copper impurity as the activator, (Zn,Cd)S: Cu. This type of coating is sensitive to the longwave UV which escapes from MA arc tubes in copious quantities. It efficiently down-converts this UV to an orange-red colour, the precise wavelength depending on both the Zn to Cd ratio and the physical temperature of the bulb wall onto which the fluorescent layer is applied.  The development resulted in the so-called MAF lamp, the additional letter F signifying the use of a fluorescent bulb.  A typical Osram 400W MAF lamp is shown alongside the clear MA lamp to the right.

Figure XX - Osira 400W MAF and MA Lamps

Fig.XX - MAF Isotherms

It will at once be noticed that the shape of this lamp is very different than the conventional slim tubular outer bulb of normal MA lamps.  It was necessary to increase bulb diameter from 51mm to 165mm simply to ensure that the phosphor was sufficiently far removed from the arc tube that its temperature was low enough to allow it to function correctly. 

The first lamps had a cylindrical shape, but it was very quickly modified to the conical isothermal form which is illustrated both above and to the left.  The shape of the isothermal bulb matches the temperature distribution of the 400W MA arc tube when run in the cap up position, and the efficacy of the lamp was raised slightly by this approach which ensured that the phosphor temperature was as uniform as possible over the greater part of the bulb surface area.  In practice its temperature was maintained fairly accurately in the region of 100-150°C by this method, the precise thermal distribution being revealed in Figure XX.

It is interesting to note that despite the extra orange emission of the phosphor, the total luminous flux of the new fluorescent lamps was no greater than for clear types.  Firstly the phosphors radiated at the red end of the spectrum, where the sensitivity of the human eye is so low as to deprive the additional light of any significant importance in its contribution to luminous flux.  But secondly, the coating also absorbed about as much visible light from the arc tube as it generated from UV down-conversion, effectively nullifying the gain.  It's quantum efficiency was fairly low by present day standards, achieving 50% efficiency in converting the 365nm longwave mercury line into visible light, but only 25% efficiency at the shorter 253.7nm wavelength.

Colour Quality from MAF Lamps

The new coating was very valuable in that it boosted the red ratio from 1% to around 4%, a particularly important increase - the spectral power distribution of the phosphor used in these first lamps can be seen in Figure XX.  However the chosen material had such a high cadmium content that it showed a natural yellowish colouration.  Consequently its transmission of visible light was rather low, its yellowish hue indicating that it was a strong absorber of blue light (an extension of its UV absorption band).  The effect was to filter out much of the blue generated by the arc tube, thus increasing the relative intensity of the remaining green lines.

Fig. XX - Spectral Distribution of ZCS

Although the orange light from the phosphor was of course very useful, the coating filtered out more blue light than the orange it generated.   Thus the net light output was distinctly greenish and most unpleasant to work under, its spectral power distribution being shown in Figure XX.

The problem was partially overcome in a new MACF lamp, which augmented the mercury charge of the arc tube with a small dose of cadmium, often a trace of zinc as well, two metals which radiate very strongly in the blue.  The effect of this was to partially shift the colour point back towards white again by reducing the relative strength of the mercury green lines.  The approximate spectral power distribution of the MACF lamp may be compared with the ordinary MAF and MA lamps in Figure XX.  Cadmium makes a somewhat inefficient arc tube fill material though, and it had a strong negative impact on lamp efficacy which fell by some 12%.  Although the fluorescent mercury-cadmium MACF lamps did deliver better colour rendering with a more pleasant overall white appearance, they were somewhat inefficient creations.  The first coated lamps also carried a hefty price premium - costing almost twice as much as a clear lamp even in 1947, ten years after their introduction.  Nevertheless, there proved to be a niche market for this kind of mercury lamp and they became successful in areas where the higher purchase and operating cost could be tolerated.  Notably, this occurred only in Europe and in particular the UK.  Elsewhere the loss in efficacy was felt to be too significant, and the idea of a coated mercury lamp was rejected until very much later.

Figure XX - Relative Spectral Power Distributions of MA, MAF and MACF Lamps

The MBF Mercury Fluorescent Lamp

In Europe however, developments with phosphors continued.  The next step came in 1937 and involved the application of similar coatings to the low wattage 80W and 125W high pressure MB type lamps employing quartz arc tubes - the development this time being led by Philips in the Netherlands, with the GEC following in 1939.  In view of the fact that the quartz arc tube operates at higher pressure and naturally generates more red light than the glass versions, the degree of required colour correction was not so great and thinner, more efficient coatings could be used.  In addition quartz transmits the shortwave UV to which the glass arc tubes are opaque, and the presence of this radiation meant that some phosphors which were developed for the fluorescent tube could be employed, providing they had the required high-temperature stability.

A new phosphor blend was thus developed for the first MBF lamps.  It was based on a thinner layer of the same (Zn,Cd)S:Cu as used in MA lamps, augmented by a blue-luminescing component which made up for the yellowish colour of the original phosphor.  The new component was also based on zinc sulphide but with an activator of silver instead of copper, ZnS:Ag, this substitution being responsible for the change in colour of fluorescence.  The cadmium addition which was required in MA arc tubes could therefore be dispensed with and consequently MBF lamps were able to achieve very nearly the same luminous efficacy as the uncoated MB types.  For every lumen which was absorbed by the coating, it re-radiated almost one lumen of orange + blue light, making it an efficient light source with rather good colour rendering for its time.

Incidentally as was the case for the MAF lamp, the size of the outer bulbs of MBF lamps also had to be increased to keep the phosphor coating cool enough for good performance.  The new silver-activated phosphor was especially sensitive to temperature and whereas a maximum of 175°C could be tolerated in MAF lamps, 100-125°C or so was the practical limit for MBF types.  The 80W lamps were therefore increased from 80mm to 110mm in diameter, while the 125W size grew from 90mm to 130mm.  Traditionally these lamps had all been fitted with standard 3-pin bayonet or E27s Edison Screw caps, but from the mechanical standpoint, it was not sensible to fit such a small cap onto so large a bulb otherwise many breakages would have occurred.  It was therefore decided to offer the inflated 125W size lamps with a large E40s Goliath Edison Screw cap for new installations.

The new MBF lamps were tremendously successful in the UK, in fact so successful that evidence of the 1940's boom in low wattage mercury fluorescent lamps still remains today.  Despite mercury lamps having now shrunk down to much more compact dimensions, today's 125W lamps are still offered with the now grossly oversized E40s large screw cap.  Remarkably there is still a sufficiently large installed base of old 1940's luminaires to justify making new lamps with this cap type for replacement purposes!  It is surprising that the MBF lamps were still not widely accepted outside the UK though, perhaps owing to the colour quality not being marvellous and worth the considerably higher price those lamps demanded.  Developments in the earliest MBF lamps therefore took place most actively within the British mercury lamp manufacturers of the time - The British Thomson-Houston Co., Osram-GEC, Metropolitan Vickers, The Edison-Swan Electric Co., Siemens Bros., Philips UK and Messrs. Crompton-Parkinson.