Photocathode technology

1. Photomultiplier tube (PMT) basic principle

A photomultiplier tube is a vacuum tube consisting of a light input window, a photoelectron emissive surface (photocathode), an electron multiplier and an anode, assembled into a hermetically sealed container as shown in Figure 1 (a). When light enters the window, photoelectrons are emitted from the photocathode and are then accelerated and focused so as to strike on the first electrode (dynode) where electron multiplication takes place by secondary emission. This secondary emission is repeated at each of the subsequent dynodes resulting in clusters of electrons multiplied to 106 to 107 times or more being collected by the anode.
Photomultiplier tubes are usually grouped by the direction of light incident on the photocathode into a side-on type and a head-on type. The direction of photoelectron emission relative to the photocathode also differs between these two groups as shown in Figure 1 (b) and (c). The side-on type photomultiplier tubes are usually compact while the head-on type photomultiplier tubes offer more design freedom in selecting the photocathode size compared to the side-on type.

 

Figure 1 (a): Cross-section of photomultiplier tube (head-on type)

Figure 1 (a): Cross-section of photomultiplier tube (head-on type)

 

Figure 1 (b): Side-on type photomultiplier tube

Figure 1 (b): Side-on type photomultiplier tube

Figure 1 (b): Side-on type photomultiplier tube

The side-on type photomultiplier tubes usually have relatively high gain and are widely used in spectrophotometers and general photometric systems.

 

Figure 1 (c): Head-on type photomultiplier tube

Figure 1 (c): Head-on type photomultiplier tube

Figure 1 (c): Head-on type photomultiplier tube

The head-on type photomultiplier tubes have a photocathode formed directly on the inner surface of the light input window and are frequently used for radiation measurements since a scintillator can be easily coupled to the light input window.

 

2. Photocathode

2-1. Alkali photocathode

Figure 3: Dr. Alfred H. Sommer (left) visited Hamamatsu Photonics on Oct. 25, 1984
Right is Teruo Hiruma, Chairman of Hamamatsu Photonics

Figure 3: Dr. Alfred H. Sommer (left) visited Hamamatsu Photonics on Oct. 25, 1984
Right is Teruo Hiruma, Chairman of Hamamatsu Photonics

Compared to other types of photodetectors, the photomultiplier tubes deliver excellent performance characteristics that definitely come from use of a low-noise electron multiplier. To make use of this feature and achieve even higher sensitivity, the quantum efficiency of the photoelectron emissive surface (photocathode) must be further improved. Figure 2 shows quantum efficiency (QE) versus wavelength of typical photocathodes currently used.

A multialkali photocathode is produced by making a layer of Sb (antimony) react with Na (sodium), K (potassium) and Cs (cesium) in a process discovered by Sommer (USA) in 1951 (left in Figure 3). This photocathode has sensitivity over a wide spectral range from ultraviolet light to 850 nm and is used in spectrophotometers and fluorescence measurement in biology and gene-related fields.

A bialkali photocathode is produced by making a layer of Sb (antimony) react with K (potassium) and Cs (cesium), and has high sensitivity in the region near 400 nm. Photomultiplier tubes using this bialkali photocathode are widely used in radiation measurements using scintillation counting because the spectral response characteristics of this photocathode are a good match with the emission wavelength of NaI(Tl) scintillators. Incidentally, this bialkali photocathode was also invented by Sommer in 1963.

Since this discovery by Sommer, many improvements have been made in these two types of photocathodes, making them the most widely used photomultiplier tube photocathodes today. Photocathode operation is described using band models. This work has led to the development of NEA semiconductor crystal photocathodes and high-sensitivity bialkali photocathodes, opening the way to enhanced photocathode sensitivity and extended the spectral response range.

 

Figure 2: Quantum efficiency versus wavelength of various photocathodes

Figure 2: Quantum efficiency versus wavelength of various photocathodes
Quantum efficiency (abbreviated “QE”) is the number of photoelectrons emitted from the photocathode
divided by the number of incident photons, and is usually expressed as a percent.

 

2-2. Photocathode band model

Figure 4: Band model for alkali photocathode

Figure 4: Band model for alkali photocathode

ince an alkali photocathode is a semiconductor, its operation can be described using the energy band theory expressed in terms of the energy gap (energy gap), electron affinity (electron affinity), Fermi level (Fermi level), work function (work function), and so on. Figure 4 shows an alkali photocathode band model. When a photon strikes a photocathode, electrons in the valence band absorb photon energy (photon energy) and become excited to the conduction band, diffusing toward the photocathode surface. If the energy of these electrons overcomes the vacuum barrier, then they are emitted into the vacuum. This electron emission process was expressed as follows by W. E. Spicer.

 

Reflection coefficient:Reflection coefficient
Full optical absorption coefficient of photons:Full optical absorption coefficient of photons
 Absorption coefficient when electrons are excited to a level greater than the vacuum level:Absorption coefficient when electrons are excited to a level greater than the vacuum level
 Diffusion length of electrons:Diffusion length of electrons
 Probability that electrons can escape into the vacuum:Probability that electrons can escape into the vacuum
 Frequency of light: Frequency of light

 

This is called Spicer's three step model which explains the photoelectron emission process using three steps: optical absorption process, electron diffusion process, and escape process. This expression is utilized to enhance the quantum efficiency by extending the diffusion length L to improve the crystalline properties of the photocathodes and also reduce the electron affinity by increasing the Ps.

 

 

2-3. High-sensitivity alkali photocathode

Figure 5: Typical QE curves for UBA, SBA and standard bialkali photocathodes.

Figure 5: Typical QE curves for UBA, SBA and standard bialkali photocathodes.

In 2007, Hamamatsu Photonics succeeded in significantly enhancing the quantum efficiency of an alkali photocathode by improving photocathode activation process. In this photocathode, an average QE of 43% at 350nm was achieved at peak wavelength and is named the “ultra bialkali” or UBA for short. Besides this “ultra bialkali”, we also developed another photocathode with moderately high sensitivity called the “super bialkali” or SBA for short and which delivers an average quantum efficiency of 35% at 350 nm. Figure 5 shows typical spectral response characteristics of these UBA and SBA along with those of an ordinary bialkali photocathode.

 

2-4. Development of crystalline photocathode

n parallel with improvement of alkali antimonide photocathodes, researchers have been very active in developing photocathodes using semiconductor crystals such as GaAs (gallium arsenide). This work revealed that forming an electrical double layer of cesium oxide (Cs-O) on the surface of a semiconductor crystal processed by Cs-O activation causes the energy band in the surface to curve downward so that the electron affinity has a negative value. These photocathodes are called NEA photocathodes (NEA: Negative Electron Affinity). Figure 6 shows a band model for single crystal GaAs activated with Cs-O. Since NEA allows electrons at the bottom of the conduction band to escape, its sensitivity extends to 900 nm which corresponds to the energy gap (Eg). As predicted from the band model, semiconductors with a higher energy gap exhibit a larger NEA. So development of GaAsP (gallium arsenide phosphor) photocathodes commenced following GaAs photocathode development. Figure 7 shows a band model of GaAsP. Currently, GaAsP photocathodes with a peak quantum efficiency exceeding 50% can be found in actual use in related fields. (See Figure 2)

 

Figure 6: Band model for GaAs photocathode

Figure 6: Band model for GaAs photocathode

Figure 7: Band model for GaAsP photocathode

Figure 7: Band model for GaAsP photocathode

 

2-5. Near-infrared photocathode

To obtain adequate sensitivity at wavelengths longer than 1.1 μm, an InP/InGaAs transfer- electron photocathode was developed. This photocathode is fabricated by depositing a metallic thin film (approx. 50 angstrom thick) such as silver on the surface of a semiconductor crystal in order to form a Schottky junction. Applying a bias voltage between the Schottky electrode and the backside of the semiconductor crystal forms an electric field in the photocathode. This greatly lowers the surface barrier and accelerates the photoelectrons, emitting them into the vacuum. Figure 8 (a) and (b) show an energy band model for photoelectron emission of a hetero-junction field-associated photocathode. When no bias voltage is applied, the photoelectrons excited in the InGaAs light-absorbing layer are unable to reach the emitting surface due to the conduction band barrier ΔEc to the InP electron-emitting layer as shown in Figure 8 (a). However, when a specified bias voltage is applied as shown in Figure 8 (b), a depletion layer is formed from the silver Schottky electrode toward the inside the photocathode, eventually reaching the interface between the InGaAs light-absorbing layer and the InP electron-emitting layer, so that the photoelectrons excited in the InGaAs light-absorbing layer can overcome the barrier to the InP electron-emitting layer. Moreover, the photoelectrons are accelerated within the InP electron-emitting layer and make the transition from the bottom of the conduction band Γ valley to a higher energy band L, and are then emitted into the vacuum from the emitting surface while maintaining a high energy level.
This photocathode covers a broad spectral range spanning from a wavelength of 0.3 mm (300 nm) in the ultraviolet region through to a wavelength of 1.6 mm (1600 nm) in the near-infrared region which corresponds to the InGaAs band gap. A nearly flat quantum efficiency of 2% is obtained over the entire spectral response range. (See Figure 2.)

 

Figure 8 (a): Band model for InP/InGaAs photocathode

Figure 8 (a): Band model for InP/InGaAs photocathode

Figure 8 (b): Band model for InP/InGaAs photocathode with bias voltage applied

Figure 8 (b): Band model for InP/InGaAs photocathode with bias voltage applied

 

2-6. Photocathode for cryogenic temperature operation

“Dark Matter” has been a hot topic in astrophysics research lately. It has been proposed to detect “Dark Matter” by using photomultiplier tubes to catch faint ultraviolet photons, which are emitted by a rare collision of incident Dark Matter and scintillator atoms. Liquid xenon (-108 ℃) or liquid argon (-186 ℃) is used as the scintillator. At such low temperatures, the sheet resistance of the photocathode becomes significant, causing the photocathode current to be limited. The possible distortion of linearity can be very critical for some experiments.

Hamamatsu Photonics has developed a new photocathode, which can be operated even at such a low temperature .

Conventional photocathodes designed for cryogenic operation have an aluminum underlay beneath the photocathode. Figure 9 shows typical spectral response characteristics of a conventional photocathode with an aluminum underlay along with this new photocathode for a low-temperature application. The new photocathode exhibits a quantum efficiency of approximately 28% at 420 nm, which is slightly lower than the SBA photocathode but is 1.5 times higher than the conventional photocathode. Figure 10 compares the linearity of a photocathode having an aluminum underlay with the new photocathode designed for a low-temperature application. When operated at -100°C, the output linearity of the conventional photocathode begins to drop severely at approximately 0.5 nA, while the new photocathode or a low-temperature application maintains linearity up to 1 μA. Here, the linearity is defined as the current at which the output deviates by -5% from the initial level.

 

Figure 9: Spectral response characteristics of the new type bialkali photocathode.

Figure 9: Spectral response characteristics of the new type bialkali photocathode.

Figure 10: Linearity comparison during operation at -100°C

Figure 10: Linearity comparison during operation at -100°C

 

2-7. Photocathode for high temperature operation

During oil well logging which is a technique for locating oil and natural gas deposits, a detector probe is lowered into boreholes that may be as deep as 2000 meters (70°C) to 3000 meters (105°C) underground. This application requires developing detectors that resist even higher temperatures not only because replacing the detector is difficult at the drill site but also because there is an increasing trend to drill deeper and deeper to find new oil stratum. Photocathodes for photomultiplier tubes gradually dissolve during operation at high temperatures such as oil well logging. However, using a combination of Sb-K-Na allows fabricating a photocathode that can resist these high temperatures. Recently, Hamamatsu developed a new photocathode capable of operating for more than 1000 hours even at high temperatures around 200°C. This new photocathode also has very low dark current at room temperature making it ideal for low-light-level detection and other applications requiring low noise.
Figure 11 compares output fluctuations in a conventional high-temperature photocathode with those in newly developed high-temperature photocathode. As can be seen, the operating life of the new photocathode at high temperatures is about 8 times longer than the conventional photocathode.

 

 Figure 11: Output fluctuations in high temperature environments of 200°C kk

Figure 11: Output fluctuations in high temperature environments of 200°C

 

3. Conclusion

Photomultiplier tubes are becoming ever more sophisticated and demand has been stimulated by development of innovative photocathode technologies and characteristics while handing over some of their jobs to semiconductor detectors. Photomultiplier tubes are still widely used, for example, in low light-level measurements such as in high energy physics experiments, medical equipment, biotechnology-related equipment, oil well logging devices, and astronomical observation equipment, etc. Requirements for these applications include even higher quantum efficiency, wider spectral response extending to the infrared region, and also higher sensitivity in the ultraviolet region. To meet these special needs, Hamamatsu Photonics will continue to develop new photocathodes that cover an even broader spectral range and with higher sensitivity (QE=100%).

 

Go to top