Tubular fluorescent lamps are a very common source of lighting. Their chief advantages a long life and good reliability. They appear in such varied applications as LCD-backlights, street lights, office lighting, light-emitting signs, tanning beds and wastewater treatment.

In this write up, we will first take a brief tour of the history of the tubular fluorescent lamp. Then, we will take a look at the various parts of the lamp, the ballast, the electrodes, the plasma and the coating, and how they operate to generate the light we see. Then, the disadvantages of tubular fluorescent lamps will be discussed. Finally, a brief overview of the future of tubular fluorescent lamps will be given.


In 1855, Heinrich Geissler invented the Geissler tube. This is a glass tube, that contains gas under a very low pressure. By passing an electric current through this gas, the gas starts to emit a faint glow. Such a gas is an example of a plasma.

The use of a mercury plasma for lighting, as it is used in a tubular fluorescent lamp, was patented in 1901 by Peter Cooper Hewitt. The lamp gave of a lot of light, unfortunately, it did so in a ghastly mix of green and blue. This had the interesting property of making healthy humans look like walking cadavers. Hewitt realized people would not want this kind of light for their houses, and thus he marketed them for industrial applications.

Hewitt's lamp was found out to emit large amounts of rather short-waved ultraviolet (UV) radiation. This radiation luckily cannot pass through a normal glass. However, several people got the idea to put a fluorescent coating on the inside of the glass that converts the ultraviolet to visible light. General Electric obtained the key patents that describe a tubular fluorescent lamp that is fundamentally similar to the lamps we use today. The first lamps were sold in 1938.


A fluorescent lamp consists of four chief parts:

  • A mercury/noble gas plasma that converts electricity to visible light and UV radiation.
  • A glass tube coated with a fluorescent powder, to contain the gases and to convert the UV radiation to visible light.
  • Electrodes that supply current to the plasma. This is the part of the lamp that is most prone to failure, so knowing how to treat your lamp's electrodes properly might save you money.
  • An electronic device, commonly known as a ballast, that regulates the current you supply to the lamp, and generates the voltage peak you use for starting it.

The plasma

The lamp contains a small amount of mercury. Because mercury is a liquid at room temperature, it will evaporate. The vapor pressure of the mercury is determined by the temperature of the coldest spot of the lamp surface. Typically, a lamp contains several milligrams of mercury. This is considerably more than is needed for the operation of the lamp. The surplus is because the mercury tends to react with the other components of the lamp, such as the electrodes and the coating. By adding more than is initially necessary, there will be sufficient mercury during the operational life of the lamp-a few tens of thousands of hours.

The voltage over the lamp accelerates free electrons, which then collide with the mercury atoms. Sometimes, they excite the mercury to a higher energy state. When the excited mercury falls back, either to the ground state or to another excited level, it may emit light. The light that is emitted in this way is in wavelengths which are typical for the gas. For mercury, the ultraviolet lines at 183 and 254 nm2,3 are very intense, and are the lines that are used by the fluorescent coating to generate visible light. The blue line at 436 nm and the green line at 546 nm2,3 are also rather intense, and are chiefly responsible for the visual emissions of low-pressure mercury lamps without a fluorescent coating.

The plasma is highly efficient, converting over 60% of the emitted electrical power you put in out as visible light

An electron colliding with a mercury atom might also ionize it, creating an extra electron and a mercury ion. This is essential for keeping the plasma operational, as the electrons tend to stick to the tube, and by virtue of the electric field this creates, drag the ions to the tube as well. The tube then acts as a catalyst in recombining the electron and the ion to neutral mercury.

This ionization and recombination is a significant and undesired loss of energy for the electrons. By adding a buffer gas, we can lessen the loss of ions to the wall. When the ions move to the wall, the buffer gas atoms are in the way, and thus it will take longer for the ions to reach the wall, increasing the ion and electron density and reducing the energy loss.

The gases most often used for this are the noble gases neon, argon and krypton. In fact, the buffer gas atoms are more abundant than the mercury atoms by a factor of roughly 100. However, because of the inertness of the noble gas atoms, they play only a minor role in the reactions. As an added advantage, the buffer gas protects the electrons from damage by the plasma, and facilitate the starting of the lamp. Furthermore, using different buffer gases allows one to somewhat tune the voltage at which the lamp operates.

The tube

The tube is made of a thin glass. This glass is extraordinarily sharp when it is broken. Be very, very careful when handling these tubes, I've seen absolutely dreadful cuts from broken lamps. This glass also blocks all of the UV radiation the lamp produces. This is desirable, as the UV produced by the plasma is very harmful to living things. As a side note, this is actually used in wastewater treatment and sterilization of medical equipment. By using a special glass that does allow the UV to pass and not using any fluorescent coating, a strong emitter of hard UV radiation is obtained, which can kill bacteria- and cause burns, cornea damage and skin cancer in humans.

The fluorescent coating absorbs the UV photon that falls on it, and re-emits a new photon with a longer wavelength. By choosing different coating materials, you can get light of practically any color you want out of the lamp-without having to make any changes to the plasma. The exact operation of these coatings and their compositions are patented or secret. Most contain rare earths, such as Europium, Yttrium, and all sorts of other weird metal oxides you probably didn't even know existed.

Because one UV photon gets emitted to one visible photon, which carries less energy, a lot of energy is lost in the conversion process. The energy loss is less when the energy of the emitted photon is close to that an UV photon. This means the lamp is most efficient emitting blue and violet light, and that is why most lamps have a rather cool blue color. Generally, the conversion process is about 50% efficient.

The electrodes

The electrodes of a lamp consist of a very, very thin coil of tungsten wire, not unlike that of an incandescent lamp. In the case of the fluorescent lamp, a mixture of metal oxides is coated on the coil. These metal oxides are chosen so that the electrode more readily emits electrons when it acts as a cathode. The exact composition is either patented or secret.

As mentioned earlier, electrode failure is the chief cause of lamp failure. In particular starting the lamp can cause serious damage. When the lamp operates, the electrodes are hot, sometimes even red-hot. The oxide coating then "sweats" electrons, which supply the current to the lamp. When the lamp is cold, the coating does not supply these electrons. Plasma processes that are beyond the scope of this write-up causes a large voltage drop to establish in front of the cathode, so the electrons get yanked out by the high electric field. This field also accelerates plasma ions, which bombard the electrode surface, destroying the electrode. Now you know why you should not flick your lamp on and off all the time.

The ballast

A typical, garden-variety 4-foot lamp works at roughly 105 V. Why can't you just plug it in the mains when you are an American, or put two in series when you are an European and plug them in your 230-V mains?

First of all, the lamp won't start. You need a voltage pulse to ignite your lamp. Secondly, a lamp that operates has a tendency of increasing its current while lowering its voltage. This means, that the lamp will start to conduct more and more current, all while dropping the voltage, causing even more current to flow. This usually ends in the explosion of something, either the fuse or the lamp.

In places where the mains voltage is about 230 V, people put a large solenoid in series with the lamp. About half the voltage drop will be over the lamp, and half will be over the solenoid. The tendency of the lamp to increase its current while slightly decreasing its voltage is tamed by the almost linear voltage/current relation of the solenoid, causing the lamp-solenoid system to be stable. In places with a 110 V mains, an additional solenoid is used as a step-up transformer. This solenoid is called a ballast, because it stabilizes the lamp, acting as a counterweight of sort to current changes.

By putting a switch in series with the lamp, one can start it using the solenoid. First, the switch is closed. This causes a current to flow through the switch. Now, the switch is opened again. A solenoid cannot easily change the current that goes through it, so it will generate a voltage peak to ram the current through the lamp. An actual lamp uses a special starter that first closes and then opens. This causes the typical flickering when starting a lamp.

While effective, the solenoid-switch system has several disadvantages. The 50/60 Hz alternating current (AC) causes a nasty hum in the solenoid. Furthermore, the 50/60 Hz AC causes the electrodes of the lamp to switch from cathode to anode and back with that same frequency. This is not desirable for stable light output. It is also detrimental to lamp efficiency.

Modern ballasts increase the frequency to several kilohertz. This means the solenoid can be much smaller. The plasma cannot change so rapidly, and is therefore roughly stable in time, causing a more stable light output and increasing the efficiency. Furthermore, an electronic circuit is added that heats the electrodes prior to starting, reducing the damage the lamp electrodes receive when the lamp starts.

In fact, these ballast, at a price in efficiency, can be made so small they can be fitted in the bottom of a mini-fluorescent lamp. These lamps are so small that they can replace an incandescent light. While not quite as efficient as a full-sized fluorescent lamp, which can reach efficiencies of up to 30%, or 100 lumen per watt, they are easily several times more efficient than incandescent lights. One warning: you cannot operate these lamps in a dimmer.


There are several disadvantages to the use of tubular fluorescent lights. Firstly, they are big, unwieldy things, although mini-fluorescent lamps exist that are somewhat more convenient. Secondly, they don't have the pretty reddish warm glow that incandescent lights and candles have. In principle, one can buy lamps that have this pretty color, at the price of some efficiency. Furthermore, they contain mercury, which is not good for the environment. Considering the minute amount of mercury per lamp, and the amount of mercury that is liberated in the burning of coal to produce electricity, the net effect of a tubular fluorescent lamp on the amount of mercury in the environment might be a positive rather than a negative one.

The Future

The major threat for the tubular fluorescent lamp is the LED (Light-Emitting Diode), which is much more compact for the same light output. Cost considerations still confine LEDs to specialty applications for the moment. Another alternative to tubular fluorescent lamps are high-pressure lamps. These lamps produce far more light in a small volume, and do so with efficiencies comparable to that of tubular fluorescent lamps. In this case, it is also chiefly costs that stops the adoption, in combination with the large amount of light that high-pressure lamps emit, which is impractical for many applications. In short, the future for tubular fluorescent lamps looks bright, especially for the thinner tubes, operated with high-frequency ballast, which are becoming more commonplace.


Tubular fluorescent lamps are a mature technology, that has been with us for nearly 70 years already. It is one of the cheapest and most efficient ways of generating light in amounts that are practical for everyday use. Incremental improvements, mainly to the electrodes, fluorescent coating, and to the ballasts, have made smaller and more efficient lamps possible, which are gradually replacing the larger, older tubular fluorescent lamps and incandescent lamps as well.


  3. C.J. Sansonetti, M.L. Salit and J.Reader, Applied Optics, volume 35, number 1, page 74, 1996.