Photomultiplier

from Wikipedia, the free encyclopedia
Schematic sketch of a photomultiplier
Photomultiplier, length approx. 8 cm; on the right the entrance window with the photocathode, in the middle the dynodes attached to insulating bodies
Photomultiplier, length approx. 17 cm; on the left the entrance window with the photocathode, in the middle the dynodes attached to insulating bodies
View through the entrance window (with photocathode) onto the first dynode stage

A photomultiplier or photo electron multiplier (short photomultiplier , engl. Photomultiplier tube , PMT ) is a special electron tube with the purpose of weak light signals (down to individual to photons to be detected by generating and amplifying an electrical signal). A photomultiplier typically consists of a photocathode and a downstream secondary electron multiplier in an evacuated glass bulb (10 −6 ... 10 −5  Pa ).

functionality

The photons hit the photocathode and, through the external photoelectric effect, release electrons from its surface, like a photocell . The released photoelectrons are accelerated in an electric field and hit further electrodes (so-called dynodes ), from the surface of which each impacting electron knocks out several secondary electrons ( δ = 3 ... 10; δ is the secondary emission ratio ). Thus, the number of electrons increases exponentially from dynode to dynode. In order for this to work, the dynodes must be on increasingly positive potential (from left to right in the diagram). This is usually achieved by dividing the original high voltage using a voltage divider chain. Finally, the electrons hit an anode and flow off to the ground. They generate a voltage drop across a resistor (R a in the drawing ). This voltage is the output signal.

The gain factor increases exponentially with the number of dynodes. Typical multipliers have approximately n = 10 dynodes. If 4 electrons are knocked out of each dynode per incident electron, the result is an increase in the number of electrons (i.e. the current) by a factor δ n  = 4 10  ≈ 10 6 . The number of secondary electrons generated is proportional to the number of radiated photons as long as a saturation threshold is not exceeded, which is around 10% of the so-called cross current (the current flowing through the voltage divider chain). This means that the level of the voltage output in this linear working range is proportional to the number of irradiated photons, i.e. the intensity of the light (analog operating mode).

Due to their high sensitivity, most photomultipliers must be protected from daylight during operation, because the incidence of too many photons generates too high a current and the ability to coat the dynodes (e.g. alkali - antimonides , BeO , MgO and especially sensitive semiconductor layers such as GaP or GaAs P) for secondary emission can irreversibly weaken ( “going blind” ) and it is even possible for the photomultiplier to burn through.

Single photon detection

With very low light intensities in the so-called single photon mode ( also known as "photon counting mode" ), photomultipliers can detect single photons with a time resolution of less than 200 ps. The dynamic range extends from maximum counting rates of a few million photons per second to the lower limit of less than 10 photons per second, which is only superimposed by a (largely thermally caused) dark current . At room temperature, the typical dark count rate, depending on the photocathode material, is approx. 10–5000 1 / s ( cps ).

Pulse behavior

Due to the construction of a photomultiplier, special characteristic impulse responses to short light pulses result in single photon counting , which falsify the actual measurement signals and can lead to misinterpretations: photons or electrons play a role that cannot be assigned to the photocathode. They generate additional output pulses. These are registered as false photon events, which are temporally correlated with the actual light pulse and lead to so-called pre-, late- and post-pulsing (afterpulsing).

Prepulse
Photons that are not absorbed at the photocathode can generate photoelectrons at the first dynodes with a low probability, which leads to weak pre-pulses, since these photons arrive earlier at the dynodes than electrons generated at the photocathode.
Late pulses
Secondary electrons, which are elastically or inelastically scattered back from the first dynode and accelerated again towards this dynode, generate output pulses that are delayed by a few nanoseconds depending on the acceleration voltage present. These typically overlap with the actual pulse in the impulse response and lead to a slight shoulder in the falling flank. (Forward-scattered electrons are also possible, which are noticeable as broadening in the rising flank, as these overtake the corresponding secondary electrons of the scattering dynode.)
Post pulses
The so-called after - pulsing extends from several nanoseconds to several microseconds and is due to several effects and is heavily dependent on the size and geometry of the photomultiplier. The causes include, on the one hand, residual gas atoms in the PMT, which are ionized by the secondary electrons and thus accelerated towards the photocathode. There they in turn generate new secondary electrons, which lead to significantly delayed output pulses, since the ions can be accelerated much more slowly due to their size. Other causes are the possible phosphorescence of the photocathode or the glass window, as well as luminescence of the last dynode stages or the anode caused by strong electron bombardment ( see also: cathodoluminescence ).

application

Large photomultiplier (approx. 50 centimeters in diameter) made by Hamamatsu Photonics , which is used to detect neutrinos in the Super-Kamiokande physical experiment . The neutrinos solve upon penetration into a water tank electron and muon , which in turn is a characteristic Cherenkov radiation produced which is registered by 11200 of such photomultiplier tubes.

In connection with scintillators they are used as detectors for elementary particles . They are often used in large detectors ( Antarctic Muon And Neutrino Detector Array ( AMANDA ) , IceCube experiment, Double-Chooz experiment , Super-Kamiokande) to detect neutrinos in large numbers. The photomultiplier register the photons that are generated by secondary particles, which arise from the extremely rare interaction of neutrinos with matter. Photomultipliers are also used in Cherenkov telescopes to detect the weak flashes of light that are created by high-energy cosmic rays in the upper atmosphere.

In scintillation counters they are also used to detect gamma radiation (e.g. gamma spectrometer or gamma camera ) and in medical technology also in PET systems to detect annihilation radiation , which is generated when positrons interact with electrons (pair annihilation ).

Furthermore, in optical spectrometry and light microscopy, photomultipliers are often used as receivers to detect light in the wavelength range from 100 nm (UV) to approx. 1000 nm (IR) (with special photocathodes up to 1700 nm). In light microscopy, photomultipliers are used as detectors in laser scanning microscopes , for example in confocal laser scanning microscopes and in multiphoton microscopes . In time-resolved fluorescence spectrometers and microscopes, they are used to determine the fluorescence lifetime in the digital operating mode, whereby the method of time- correlated single photon counting is often used.

In the scanning electron microscope , photomultipliers are part of the Everhart-Thornley detector . The secondary or backscattered electrons (SE - secondary electrons or BSE - back scattered electrons ) generated by the electron beam on the sample are converted into photons in the scintillator , which are fed to the photomultiplier via a light guide and converted into electrical signals.

Photomultipliers have also been used in drum scanners . These are no longer manufactured because today's high-performance flatbed and film scanners are cheaper and achieve a relatively good quality. However, when very high resolution or density acquisition is required, drum scanners are still unmatched for high quality scans with great detail.

Other designs

Schematic representation of how a microchannel plate works

A special form of photomultipliers are so-called microchannel plate photomultiplier (Engl. Microchannel plate photomultiplier , MCP-PMT or short MCP). In the microchannel plate, secondary electrons are released from the inner wall of microscopically thin channels along which an accelerating electrical field prevails. They thus represent a homogeneous combination of dynodes and voltage divider chain, with which a time resolution of less than 30 ps is achieved. You will u. a. Used in image intensifiers and preferably in time-resolved fluorescence spectrometers for high temporal resolution (but are many times more expensive than conventional photomultipliers).

A specially coated input side of the microchannel plate can, with some limitations replace the photocathode, and there are also secondary electron multiplier with a single larger channel made, so-called electron multiplier ( English channel electron multiplier ).

Another alternative design is the so-called hybrid photomultiplier (HPMT or H (A) PD for hybrid (avalanche) photodiode ). Here , the classic dynodes are replaced by an avalanche photodiode, which takes on the task of the secondary electron multiplier. Similar to the MCP-PMT, the large differences in the transit time of the electrons across the various dynode stages are avoided and a time resolution of approx. 100 ps is achieved.

Alternatives

Close-up of a silicon photomultiplier (SiPM) consisting of an array of APDs

The semiconductor equivalent of the photomultiplier are avalanche photodiodes (APD) and derived from this the silicon photomultiplier (SiPM), which use the avalanche effect that occurs in semiconductor crystals at high field strengths to increase charge carriers. Individual APDs generate an output voltage that is proportional to the radiation power, but in contrast to the photomultiplier only achieve a gain of <10 3 , SiPM achieve similarly high gains as photomultipliers in the range of 10 6 . Avalanche photodiodes are preferably used for the detection of low light intensities of medium or high frequency , such as. B. in laser rangefinders .

For single photon detection, special avalanche photodiodes, so-called single photon avalanche diodes (SPAD), can be used, where individual photons briefly generate up to a few million charge carriers and can thus be easily registered as electrical impulses .

Very sensitive photodetectors are photo resistors . However, they cannot be used to detect single photons, they have a lot of noise and are very sluggish (seconds).

literature

  • Hanno Krieger: Radiation measurement and dosimetry. Vieweg + Teubner, Wiesbaden 2011, ISBN 978-3-8348-1546-0 .
  • William R. Leo: Techniques for Nuclear and Particle Physics Experiments: A How-to Approach. Springer, New York 1994, ISBN 0-387-57280-5 .

Web links

Commons : Photomultiplier  - album with pictures, videos and audio files

swell

  1. ↑ a b Hamamatsu Photonics KK : Photomultiplier Tubes - Basics and Applications. 3. Edition. 2006, pp. 17–18 - 2.3 Electron Multiplier (Dynode Section)
  2. ↑ Hamamatsu Photonics KK: Photomultiplier Tubes - Basics and Applications. 3. Edition. 2006, p. 126 - 6.1 Analog and Digital (Photon Counting) Modes
  3. ↑ O. Ju. Smirnov, P. Lombardi, G. Ranucci: Precision Measurements of Time Characteristics of ETL9351 Photomultipliers. In: Instruments and Experimental Techniques. vol. 47, number 1, 2004, pp. 69-80.
  4. ↑ HR Krall: Extraneous Light emission from the photomultiplier. In: IEEE Transactions on Nuclear Science. vol. 14, issue 1, 1967, pp. 455-459.
  5. ↑ Hamamatsu Photonics KK: Photomultiplier Tubes - Basics and Applications. 3. Edition. 2006, pp. 30-35 - 4.1 Basic Characteristics of Photocathodes
  6. ^ W. Becker, B. Su, O. Holub, K. Weisshart: FLIM and FCS detection in laser-scanning microscopes: Increased efficiency by GaAsP hybrid detectors. In: Microscopy Research and Technique . 2010. doi: 10.1002 / jemt.20959