PHOTON COUNTINGUsing Photomultiplier Tubes

TECHNICAL INFORMATION

INTRODUCTION

Recently, non-destructive and non-invasive measurementusing light is becoming more and more popular in diversefields including biological, chemical, medical, materialanalysis, industrial instruments and home appliances. Tech-nologies for detecting low level light are receiving particu-lar attention since they are effective in allowing high preci-sion and high sensitivity measurements without changingthe properties of the objects for low valume sample.Biological and biochemical scrutinies use low-light-levelmeasurement by detecting fluorescence emitted from cellslabeled with a fluorescent dye. In clinical testing and medi-cal diagnosis, various advanced techniques are beingused, for example, invitro assay for blood analysis andblood cell counting, immunoassay for hormone testing anddiagnosis of cancers and various infectious diseases, anddiagnostic analysis of genes and single base polymorphismby using “gene chips”. These techniques involve low-light-level measurement such as colorimetry, absorption spec-troscopy, fluorescence photometry, and detection of lightscattering or Iuminescence measurement. In molecularbiology research, optical sensors occupy an importantposition as the fluorescence detectors in fluorescencecorrelation spectroscopy (FCS) - an innovative techniquefor analyzing dynamic interaction between molecules bydetecting fluorescent molecules at single molecule levels.Furthermore, in RIA (radioimmunoassay) which has beenused in immunological examinations using radioisotopes,radiation emitted from a sample is converted into low levellight which must be measured with high sensitivity. In en-vironmental measurement, test methods that make use offluorescence and luminescence measurements are beingused for hygiene monitoring such as for bacteria contami-nation in food processing and for water pollution. Thesetest methods allow rapid measurement, yet with high sen-sitivity.Photomultiplier tubes (PMT), photodiodes and CCD im-age sensors are widely used as “eyes” for detecting lowlevel light. These detectors convert light into analog elec-trical signals (current or voltage) in most applications. How-ever, when the light level becomes weak so that the inci-dent photons are detected as separate pulses, the singlephoton counting method using a photomultiplier tube isvery effective if the average time intervals between signalpulses are sufficiently wider than the time resolution of thephotomultiplier tube. This photon counting method is su-perior to analog signal measurement in terms of stability,detection efficiency and signal-to-noise ratio.This technical manual explains how to use photomultipliertubes in photon counting to perform low-light-level mea-surement with high sensitivity and high accuracy. Thismanual also describes the principle of photon counting,its key points and operating circuit configuration, as wellas characteristics of photomultiplier tubes and their selec-tion guide.

1

TABLE OF CONTENTS

1. What is Photon Counting ? ................................................ 21-1 Analog Mode and Digital Mode (Photon Counting Mode)1-2 The Principle of Photon Counting

2. Operation and Characteristics of Photon Counting ........ 52-1 Photon Counter and Multichannel Pulse Height Analyzer2-2 Basic Characteristics in Photon Counting

(1) Pulse Height Distribution and Plateau Characteristics(2) Output Instability vs. Variations in Photomultiplier Tube

Gain (current amplification)(3) Linearity of Count Rate

3. Characteristics of Photomultiplier Tubes ......................... 93-1 Spectral Response (Quantum Efficiency)3-2 Collection Efficiency3-3 Supply Voltage and Gain3-4 Noise3-5 Magnetic Shield3-6 Stability and Dark Storage3-7 Uniformity3-8 Signal-to-Noise Ratio

4. Measurement Systems ....................................................... 164-1 Synchronous Photon Counting Using Chopper4-2 Time-Resolved Photon Counting by Repetitive Sampling4-3 Time-Resolved Photon Counting by Multiple Gates4-4 Time-Correlated Photon Counting4-5 Calculating the Autocorrelation Function using a Digital Correlator

5. Selection Guide ................................................................... 195-1 Selecting the Photomultiplier Tube5-2 Photomultiplier Tubes for Photon Counting5-3 Peripheral Devices

2

Figure 1 : Output Pulses from Photomultiplier Tube at Different Light Levels

1-2 The Principle of PhotonCounting

One important factor in photon counting is the quantumefficiency (QE). It is the production probability of photo-electrons being emitted when one photon strikes the pho-tocathode. In the single photoelectron state, the numberof emitted photoelectrons (primary electrons) per photonis only 1 or 0. Therefore QE refers to the ratio of the aver-age number of emitted electrons from the photocathodeper unit time to the average number of photons incidenton the photocathode.

Figure 2 : Photomultiplier Tube Operationin Single Photoelectron State

1. What is Photon Counting ?

1-1 Analog Mode and Digital Mode(Photon Counting Mode)

A photomultiplier tube (PMT) consists of a photocathode,an electron multiplier (composed of several dynodes) andan anode. (See Figure 2 for schematic construction.) Whenlight enters the photocathode of a photomultiplier tube,photoelectrons are emitted from the photocathode. Thesephotoelectrons are multiplied by secondary electron emis-sion through the dynodes and then collected by the an-ode as output pulses. In usual applications, these outputpulses are not handled as individual pulses but dealt withas an analog current created by a multitude of pulses (so-called analog mode). In this case, a number of photonsare incident on the photomultiplier tube per unit time as in1 of Figure 1 and the resulting photoelectrons are emit-ted from the photocathode as in 2. The photoelectronsmultiplied by the dynodes are then derived from the an-ode as output pulses as in 3. At this point, when the pulse-to-pulse interval is narrower than each pulse width or thesignal processing circuit is not fast enough, the actual out-put pulses overlap each other and eventually can be re-garded as electric current with shot noise fluctuations asshown in 4.In contrast, when the light intensity becomes so low thatthe incident photons are separated as shown in 5, theoutput pulses obtained from the anode are also discreteas shown in 7. This condition is called a single photoelec-tron state. The number of output pulses is in direct propor-tion to the amount of incident light and this pulse countingmethod has advantages in signal-to-noise ratio and sta-bility over the analog mode in which an average of all thepulses is made.This pulse counting technique is known as the photoncounting method. Since the detected pulses undergo bi-nary processing for digital counting, the photon countingmethod is also referred to as the digital mode.

ARRIVAL OF PHOTONS

LOWER LIGHT LEVEL (Single Photoelectron State)

HIGHER LIGHT LEVEL (Multiple Photoelectron State)

PHOTOELECTRON EMISSION

SIGNAL OUTPUT (PULSES)

SIGNAL OUTPUT (PULSE OVER LAPPED)

ARRIVAL OF PHOTONS

PHOTOELECTRON EMISSION

SIGNAL OUTPUT(DISCRETE PULSES)

TIME

TPHOC0027EB

TIME

1

2

3

4

5

6

7

PHOTOCATHODE

Dy1 Dy2 Dyn-1 Dyn

P

1ST DYNODE

SINGLEPHOTONS

ELECTRONGROUPS

ANODE

TPMOC0048EB

PULSEHEIGHT

3

Photoelectrons emitted from the photocathode are accel-erated and focused onto the first dynode (Dy1) to producesecondary electrons. However, some of these electronsdo not strike the Dy1 or deviate from their normal trajecto-ries, so they are not multiplied properly. This efficiency ofcollecting photoelectrons is referred to as the collectionefficiency (CE). In addition, the ratio of the count value(the number of output pulses) to the number of incidentphotons is called the detection efficiency or counting effi-ciency, and is expressed by the following equation :

Detection efficiency =Nd / Np = η .α

where Nd is the count value, Np is the number of incidentphotons, η is the photocathode QE and α is the CE. Al-though it will be discussed later, detection efficiency alsodepends on the threshold level that brings the output pulsesinto a binary signal. Since the number of secondary elec-trons emitted from the Dy1 varies from several to about20 in response to one primary electron from the photo-cathode, they can be treated by Poisson distribution ingeneral, and the average number of secondary electronsbecomes the secondary electoron emission ratio δ. Thisholds true for multiplication processes in the subsequentdynodes. Accordingly, for a photomultiplier tube having nstages of dynodes, a single photoelectron from the photo-cathode is multiplied by δ n to create a group of electronsand is derived from the anode as an output pulse. In thisprocess, the height of each output pulse obtained at theanode depends on fluctuations in the secondary electronmultiplication ratio stated above, so that it differs from pulseto pulse. (Figure 3)Other reasons why the output pulse height becomes un-equal are that gain varies with the position on each dyn-ode and some deviated electrons do not contribute to thenormal multiplication process. Figure 3 shows a histogramof the anode pulse heights. This graph is known as thepulse height distribution (PHD).As illustrated in Figure 3, the photomultiplier tube outputexhibits fluctuations in the pulse height and the PHD isobtained by time-integrating these output pulses at differ-ent pulse heights. The abscissa of this graph indicates thepulse height that represents the charge (number of elec-trons contained in one electron group) or the pulse volt-age (current) produced by that electron group. It is gener-ally expressed in the number of channels used for the ab-scissa of a multichannel analyzer.

Figure 3 : Photomultiplier Tube Output and PHD

TIM

EC

OU

NT

S A

T E

AC

HP

ULS

E H

EIG

HT

PULSE HEIGHTS (CHARGES)TPMOC0049EB

LOW

CO

UN

TF

RE

QU

EN

CY

HIG

H C

OU

NT

FR

EQ

UE

NC

Y

LOW

CO

UN

TF

RE

QU

EN

CY

Photomultiplier Tube Output in SinglePhotoelectron State

The output signal from a photomultiplier tube in the pho-ton counting mode can be calculated as follows:In the photon counting mode, a single photoelectron e-

(electron charge:1.6 × 10-19 C) is emitted from the pho-tocathode. If the photomultiplier tube gain µ is5 × 106, then the anode output charge is given by

e × µ = 1.6 × 10-19.5 × 106 C= 8 × 10-13 C

Here, if the pulse width t (FWHM) of the anode outputsignal is 10 ns, then the output pulse peak current Ip is

Ip= e .µ .1/t A= (8 × 10-13) C /(10 × 10-9) s= 80 µ A

This means that the anode output pulse width is nar-rower, we can obtain much higher output peak current.If the load resistance or input impedance of the suc-ceeding amplifier is 50 Ω, the output pulse peak volt-age Vo becomes

Vo = Ip × 50 Ω= 4 mV

The amplitude of the photomultiplier tube output pulsein the photon counting mode is extremely small. Thisrequires a photomultiplier tube having a high gain andmay require an amplifier with sufficiently low noise rela-tive to the photomultiplier tube output noise. As a gen-eral guide, photomultiplier tubes should have a gain ofapproximately 1 × 106 or more.

4

Figure 4 (a) shows a PHD when the incident light level isincreased under single photoelectron conditions, and (b)shows a PHD when the supply voltage is changed.

The ordinate is the frequency of the output pulses thatproduce a certain height within a given time. Therefore,the distribution varies with the measurement time or thenumber of incident photons in the upper direction of theordinate as shown in Figure 4 (a).As explained above, the abscissa of the PHD representsthe pulse height and is proportional to the gain of the pho-tomultiplier tube and becomes a function of the supplyvoltage of the photomultiplier tube. This means that as thesupply voltage changes, the PHD also shifts along theordinate, but the total number of counts is almost con-stant.

Figure 4 : PHD Characteristics

CO

UN

TS

(a) When the incident light level is increased

SIGNAL+NOISE

PULSE HEIGHT

INCREASE IN LIGHT INTENSITY

CO

UN

TS

PULSE HEIGHT (b) When the supply voltage is changed TPHOB0032EB

SUPPLY VOLTAGE TOPHOTOMULTIPLIER TUBE

INCREASE IN

5

2. Operation and Characteristics of Photon Counting

This section describes circuit configurations for use inphoton counting and the basic characteristics of photoncounting measurements.

2-1 Photon Counter andMultichannel Pulse HeightAnalyzer

There are two methods of signal processing in photoncounting: one uses a photon counter and the other a mul-tichannel pulse height analyzer (MCA). Figure 5 showsthe circuit configuration of each method and the pulseshapes obtained from each circuit system.In the photon counter system of Figure 5 (a), the outputpulses from the photomultiplier tube are amplified by thepreamplifier. These amplified pulses are then directed intothe discriminator. The discriminator compares the inputpulses with the preset reference voltage to divide theminto two groups: one group is lower and the other is higherthan the reference voltage. The lower pulses are elimi-nated by the lower level discriminator (LLD) and in some

Figure 5 : Typical Photon Counting Systems

cases, the higher pulses are eliminated by the upper leveldiscriminator (ULD). The output of the comparator takesplace at a constant level (usually a TTL logic level or CMOSlevel ). The pulse shaper forms rectangular pulses allow-ing counters to count the discriminated pulses.In contrast, in the MCA system shown in Figure 5 (b), theoutput pulses from the photomultiplier tube are generallyintegrated through a charge-sensitive amplifier, amplifiedand shaped with the linear amplifier. These pulses are dis-criminated according to their heights by the discriminatorand are then converted from analog to digital. They arefinally accumulated in the memory and displayed on thescreen. This system is able to output pulse height infor-mation and frequency (the number of counts) simulta-neously, as shown in the figure.The photon counter system is used to measure the num-ber of output pulses from the photomultiplier tube corre-sponding to incident photons, while the MCA system isused to measure the height of each output pulse and thenumber of output pulses simultaneously. The former sys-tem is superior in counting speed and therefore used forgeneral-purpose applications. The MCA system has thedisadvantage in not being able to measure high counts, itis used for applications where pulse height analysis is re-quired.

ULD

TTL LEVELLLD

ULDLLD

ULDLLD

(a) Photon counter system

(b) MCA system

PHOTONS

PHOTONS

PREAMPLIFIER COUNTER

MEMORYDISPLAY

A/D CONVERTERLINEAR AMPLIFIERPULSE SHAPER

DISCRIMINATOR

CHARGESENSITIVE AMP.

DISCRIMINATOR

PHOTOMULTIPLIERTUBE PULSE SHAPER

ACCUMULATION

=50 s

ULD

LLD

TPHOC0028EB

PHOTOMULTIPLIERTUBE PREAMPLIFIER

CO

UN

TS

TIM

E

PULSE HEIGHTµτ

6

2-2 Basic Characteristicsin Photon Counting

(1) Pulse Height Distribution andPlateau Characteristics

Figure 6 shows pulse height distributions (PHD) of a pho-tomultiplier tube, obtained with an MCA. The curve (a) isthe output when signal light is incident on the photomulti-plier tube, while the curve (b) represents the noise whensignal light is removed. The major noise component re-sults from thermionic emission from the photocathode anddynodes. The PHD of such noise usually appears on thelower pulse height side. These PHD are the so-called dif-ferential curves and the lower level discrimination (LLD) isusually set at the valley of the curve (a). To increase de-tection efficiency, it is advantageous to set the LLD at alower position, but this is also accompanied by a noiseincrease.In contrast to the curve (a) in Figure 6, the curve (c) showsan integration curve in which the total number of pulseshigher than a certain discrimination level have been plot-ted while changing the discrimination level. Since this in-tegration curve (c) has an interrelation with the differentialcurve (a), the proper discrimination levels can be set inthe photon counter system without using an MCA, by ob-taining the integration curve instead.

The LLD stated above corresponds to a pulse height on agentle slope portion in the integration curve. But this is notso distinct from other portions, making it difficult to deter-mine the LLD. Another method using plateau characteris-tics is more commonly used. By counting the number ofpulses with the LLD fixed while varying the supply voltageto the photomultiplier tube, a curve similar to the"SIGNAL+NOISE" curve shown in Figure 7 can be plot-ted. Although, analog gain increases expotentially withsupply voltage, since the photomultiplier tube counts onlythe number of pulses, the slope of the curves is relativelyflat, which makes the supply voltage setting easier. Thesecurves are known as the plateau characteristics. The sup-ply voltage for the photomultiplier tube should be set withinthis plateau region. It is also clear that plotting the signal-to-noise ratio shows a plateau range in the same supplyvoltage range. (See Section 3-8 (2).)

Figure 7 : Plateau Characteristics

Figure 6 : Differential and Integral Displays of PHD

PULSE HEIGHT (ch)

NOISE/DIFFERENTIAL (b)VALLEY

SIGNAL/DIFFERENTIAL (a)

PEAK

SIGNAL/INTEGRAL (c)

CO

UN

TS

(IN

TE

GR

AL)

CO

UN

TS

/CH

AN

NE

L (D

IFF

ER

EN

TIA

L)

1

0 200 400 600 800 1000

2

2

0

4

6

8

103(×106) (×103)

TPHOB0033EC

LLD

800 900 1000 1100 1200 1300 1400 15007000

1000

2000

4000

3000

6000

5000

7000

CO

UN

T R

AT

E (

s-1)

S/N

RA

TIO

PHOTOMULTIPLIER TUBE SUPPLY VOLTAGE (V)

NOISE

SIGNAL+NOISE

PLATEAU RANGE

S/N RATIO

TPHOB0034EB

HIGH VOLTAGE SETTING

7

0.5

1.0

1.5

0600 700 800 900 1000 1100 1200

CO

UN

T R

AT

E (

×106

s-1)

SUPPLY VOLATGE (V)

A

B

VC

VC'

VHVL

TPHOB0038EA

PLATEAU RANGE

(2) Output Instability vs. Variationsin Photomultiplier Tube Gain

If the photomultiplier tube gain varies for some reason (forexample, a change in supply voltage or fluctuations in theambient temperature, etc.), the output current of the pho-tomultiplier tube is also affected and exhibits variations. Inthe analog mode the output current (or gain) of the photo-multiplier tube changes with variations in the supply volt-age as shown in Figure 8 (a). In the photon counting mode,the output count changes, but this is significantly smallerthan in the analog mode.By setting the supply voltage in the plateau region as shownin Figure 7, the photon counting mode can minimizechanges in the count rate with respect to variations in thesupply voltage without sacrificing the signal-to-noiseratio. This means that the photon counting mode ensureshigh stability even when the gain of the photomultipliertube varies as the gain is a function of the supply voltage.For the above reasons, the photon counting mode offers

several times higher stability than the analog mode againstvariations in the operating conditions.

Figuer 8 : Output Variation vs. Supply Voltage

RELATIVE SUPPLY VOLTAGE

0.96 0.98 1.00

1.0

1.2

1.4

1.6

1.8

2.0

2.2

1.02 1.04 1.06 1.08 1.10

CH

AN

GE

RA

TE O

F C

OU

NTS

(PH

OTO

N C

OU

NTI

NG

MO

DE

)C

HA

NG

E R

ATE

OF

GA

IN (A

NA

LOG

MO

DE

)

ANALOG MODE (a)

PHOTON COUNTING MODE (b)

TPHOB0035EA

Speaking it in a broad sence, the integration curve ex-plained in Section 2-2 (1) has plateau characteristics.Here we describe a more practical method for obtainingthe plateau characteristics by varying the photomultipliertube supply voltage.1. Set up the photomultiplier tube, photon counter, high-

voltage power supply, dark box, light source, etc, re-quired to perform photon counting. Preferably, the pho-tomultiplier tube should be stored in the dark box forabout one hour after the setup has been completed.

2. Set the discrimination level (LLD) according to the in-struction manual for the photon counter being used.Then allow a very small amount of light to strike thephotomultiplier tube.

3. Gradually increase the photomultiplier tube supplyvoltage starting from about 500 V. When the photoncounter begins to count any signal, halt there and makea plot of the supply voltage (VL) and the counted valueat that point, on a graph with the abscissa showingthe supply voltage and the ordinate representing thecount rate.

4. Increase the supply voltage with a 50 V step until itreaches about 90 % (VH) of the maximum supply volt-age while making plots on the graph. This will createa curve like "A" shown in Figure 9.

How to Obtain the Plateau5. With the photomultiplier tube operated at a voltage

(Vc) in the middle of the flat portion of curve "A", ad-just the light source intensity so that the counted valueis set to 10 % to 30 % of the maximum count rate ofthe photon counter.

6. Readjust the photomultiplier tube supply voltage toset at VL, then make fine plots while changing thephotomultiplier tube supply voltage in 10 V or 20 Vsteps. This will make a curve like "B" shown in Fig-ure 9.

7. In the flat range (plateau range) on curve "B", thevoltage Vc' at a point where the differential coeffi-cient is smallest (minimum slope) will be the opti-mum PMT operating voltage.

Figure 9 : Plotting Plateau characteristics

8

(3) Linearity of Count Rate

The photon counting mode is generally used in very low-light-level regions where the count rate is low, and exhib-its good linearity. However, when the amount of incidentlight becomes large, it is necessary to take the linearity ofthe count rate into account. The upper limit of the band-width of photomultiplier tubes ranges from 30 MHz to300 MHz for periodical signal. However, the maximumcount rate in the photon counting mode where randomphotos enter the photomultiplier tube is determined by thetime response of photomultiplier tube and the time resolu-tion of the signal processing circuit connected to the pho-tomultiplier tube. The time resolution referred to here isdefined as the minimum time interval between successivepulses that can be counted as separate pulses.Figure 10 shows a typical linearity of the count rate ob-tained from a Hamamatsu H7360 Photon Counting Head.As can be seen from the figure, the dynamic range reachesas high as 107 s-1. In this case, the linearity of the countrate is limited by the time resolution (pulse pair resolution)of the built-in circuit (18 ns).If we let the measured count rate be M (s-1) and the timeresolution be t (s), the real count rate N (s-1) can be ap-proximated as follows:

Figure 11 shows the actual data measured with the H7360Photon Counting Head, along with the corrected data ob-tained with the above equation. This proves that after mak-ing correction, the photon counting mode provides an ex-cellent linearity with a counting error of less than 1 %, evenat a high count rate of 107 s-1.However, the linearity deviates from this corrected datacurve in a region where measurement errors become largedue to base line shift or overlapping of photomultiplier tubeoutput pulses themselves.

M1–Mt

N =

102101

102

103

105

104

106

107

108

104103 105 107106 108 109

CO

UN

T R

AT

E(s

-1)

INCIDENT NUMBER OF PHOTONS (s-1)

TPHOB0036EB

103

+10 %

0 %

-10 %

-20 %104 105 106 107

DE

VIA

TIO

N

COUNTS (s-1)

CORRECTED

MEASURED

TPHOB0037EA

Figure 11 : Correction of Count Rate Linearity

Figure 10 : Linearity of Count Rate

9

3. Characteristics of Photomultiplier Tubes

3-1 Spectral Response(Quantum Efficiency )

When N number of photons enter the photocathode of aphotomultiplier tube, the N × quantum efficiency (QE) ofphotoelectrons on average are emitted from the photo-cathode. The QE depends on the incident light wavelengthand so exhibits spectral response.Figures 12 (a) and (b) show typical spectral response char-acteristics for various photocathodes and window materi-als. The spectral response at wavelengths shorter than350 nm is determined by the window material used, asshown in Figure 13.In general, spectral response characteristics are expressedin terms of cathode radiant sensitivity or QE. The QE and

Figure 12 : Typical Spectral ResponseCharacteristics

radiant sensitivity have the following relation at a givenwavelength.

where S is the cathode radiant sensitivity in amperes perwatt (A/W) at the given wavelength λ in nanometers.

3-2 Collection Efficiency

The collection efficiency (CE) is the probability in percent,that single photoelectrons emitted from the photocathodecan be finally collected at the anode as the output pulsesthrough the multiplication process in the dynodes. In par-ticular, the CE is greatly affected by the probability thatthe photoelectrons from the photocathode can enter thefirst dynode and multiplied. Generally, the CE is from 70 %to 90 % for head-on photomultiplier tubes and 50 % to 70% for side-on photomultiplier tubes with full cathode illu-mination.The CE is very important in photon counting measure-ment. The higher the value of the CE, the smaller the sig-nal loss, thus resulting in more efficient and accurate mea-surements. The CE is determined by the photocathodeshape, dynode structure and voltage distribution for eachdynode.As stated earlier, the ratio of the number of signal pulsesobtained at the anode to the number of photons incidenton the photocathode is referred to as the detection effi-ciency or counting efficiency.

MgF2

100 120 160 200 240 300 400 500

100

10

1

TR

AN

SM

ITT

AN

CE

(%

)

WAVELENGTH (nm)

BOROSILICATEGLASSUV

TRANSMITTINGGLASS

SYNTHETICSILICA

TPMOB0053EA

Figure 13 : Typical Transmittance of Window Materials

Figure 12(a): Transmission Mode Photocathodes

1.0

0.1

0.01

10

100

100 200 300 400 500 600700 800 1000 1200 1500 2000

QU

AN

TU

M E

FF

ICIE

NC

Y (

%)

WAVELENGTH (nm)

BOROSILICATE GLASSUV GLASS

MULTIALKALI

InGaAs/InP

InGaAsP/InPBIALKALI

Ag-O-Cs

EXTENDED-REDMULTIALKALI

GaAsP

Figure 12(b): Reflection Mode Photocathodes

100 200 300 400 500 600700 800 1000 1200 1500 2000

QU

AN

TU

M E

FF

ICIE

NC

Y (

%)

1.0

0.1

0.01

10

100

WAVELENGTH (nm)

BOROSILICATE GLASSUV GLASSSYNTHETIC SILICA

InGaAs/InP

InGaAsP/InP

MULTIALKALIGaAsP

BIALKALI

InGaAs

TPMOB0084EA

QE= S ×1240

TPMOB0083EB

10

Figure 14 shows the relation between the CE and the pho-tocathode to first dynode voltage of 28 mm (1-1/8 ") diam-eter photomultiplier tubes, measured in the photon count-ing mode with the discrimination level kept constant andwith small area illumination. As can be seen, the CE sharplyvaries at voltages lower than 100 V, but becomes satu-rated and shows little change when the voltage exceedsthis. This means that a sufficient voltage should be ap-plied across the photocathode and the first dynode to ob-tain stable CE.

Figure 14 : CE vs. Photocathode to First Dynode Voltage

3-3 Supply Voltage and Gain

The output pulse height of a photomultiplier tube varieswith the supply voltage change, even when the light levelis kept constant. This means that the gain of the photo-multiplier tube is a function of the supply voltage. The sec-ondary emission ratio δ is the function of voltage E be-tween dynodes and is given by

δ = A • Eα

where A is the constant and α is determined by the struc-ture and material of the electrodes, which usually takes avalue of 0.7 to 0.8.Here, if we let n denote the number of dynodes and as-sume that the δ of each dynode is constant, then thechange in the gain µ relative to the supply voltage V isexpressed as follows:

(K is a constant.)Since typical photomultiplier tubes have 9 to 12 dynodestages, the output pulse height is proportional to the 6th to10th power of the supply voltage. The curve previouslyshown in Figure 8 for the analog mode represents thischaracteristic.

Figure 15 : Gain vs. Supply Voltage

Figure 16 shows the relation between the secondary elec-tron emission ratio (δ 1) and the photocathode to first dyn-ode voltage. The incident light is passed through a slit of 3mm × 15 mm for side-on photomultiplier tubes, or is fo-cused on a spot of 10 mm diameter for head-on photo-multiplier tubes. It is clear that the secondary electron emis-sion ratio depends on the material of the secondary elec-tron emission surface. Generally, the larger the second-ary electron emission ratio, the better the PHD will be.

Figure 16 : Secondary Electron Emission Yield vs.Supply Voltage

0

5

10

15

100 200

SIDE-ON:BIALKALI (SECONDARY PHOTOEMISSIVE SURFACE)

SIDE-ON:MULTIALKALI (SECONDARY PHOTOEMISSIVE SURFACE)

HEAD-ON:BIALKALI (LINE DYNODE) (SECONDARY PHOTOEMISSIVE SURFACE)

SIDE-ON:SLIT (3 mm × 15 mm)INCIDENT LIGHT

HEAD-ON:SPOT ( 10 mm )

SE

CO

ND

AR

Y E

LEC

TR

ON

EM

ISS

ION

YIE

LD :

1

PHOTOCATHODE TO FIRST DYNODE VOLTAGETPMOB0085EA

0 50 100 150 2000

20

40

60

80

100

CE

(%)

PHOTOCATHODE TO FIRST DYNODE VOLTAGE (V)

TPMOB0057EB

φ

φφ

28 mm SIDE-ON TYPE(3 mm × 15 mm LIGHT SLIT)

28 mm HEAD-ON TYPE( 10 mm LIGHT SPOT)

GA

IN

SUPPLY VOLTAGE (V)

TPMOB0082EA

102

200 300 500 700 1000 1500

103

105

104

106

107

108

= = (A E ) = A ( ) n n

n

n+1

(n+1)

nα

nαnα nα

α

= V = K V

Vµ δ

A

11

3-4 Noise

Some noise exists in a photomultiplier tube even when itis kept in complete darkness. The noise adversely affectthe counting accuracy, especially in cases where the countrate is low. The following precautions must be taken tominimize the noise effects.

(1) Thermionic Emission of ElectronsMaterials used for photocathodes and dynodes have lowwork functions (energy required to release electrons intovacuum), so they emit thermal electrons even at room tem-peratures. Most of the noise is caused by these thermalelectrons mainly being emitted from the photocathode andamplified by the dynodes. Therefore, cooling the photo-cathode is the most effective technique for reducing noisein applications where low noise is essential. In addition,since thermal electrons increase in proportion to photo-cathode size, it is important to select the photocathodesize as needed.Figure 17 shows temperature characteristics of dark countsmeasured with various types of photocathodes. These aretypical examples that vary considerably with photocath-ode type and sensitivity (especially red sensitivity).The head-on type Ag-O-Cs, multialkali, and GaAs photo-cathodes have high sensitivity in the near infrared, butthese photocathodes tend to emit large amounts of ther-mal electrons even at room temperatures, so usually cool-ing is necessary.

Figure 17 : Temperature Characteristics of Dark Counts

(2) Glass ScintillationWhen electrons deviating from their normal trajectoriesstrike the glass bulb of a photomultiplier tube, glass scin-tillation may occur and result in noise. Figure 18 showstypical dark current (RMS noise) versus the distance be-tween the photomultiplier tube and the metal housing caseat ground potential. This implies that glass scintillation noiseis caused by stray electrons which are attracted to theglass bulb at a higher potential. This is particularly truewhen the tube is operated with a voltage divider circuitwith the anode grounded. To minimize this problem, it isnecessary to reduce the supply voltage for the photomul-tiplier tube, use a voltage divider circuit with the cathodegrounded or make longer the distance between the pho-tomultiplier tube and the housing. Another effective mea-sure is to coat the outer surface of the glass bulb withconductive paint which is maintained at the photocathodepotential in order to prevent stray electrons from beingattracted to the glass bulb. In this case, however, the pho-tomultiplier tube must be covered with an insulating mate-rial since a high voltage is applied to the glass bulb. Wecall this technique "HA coating". Although Figure 18 is anexample of a side-on photomultiplier tube, similar charac-teristics will be observed for a head-on photomultiplier tube.

Figure 18 : Dark Current vs. DistanceBetween Photomultiplier Tubeand Housing Case at Ground Potential

10 7

20 400−20−40−60TEMPERATURE (°C)

DA

RK

CO

UN

TS

(s-1

)

HEAD-ON TYPEAg-O-Cs

HEAD-ON TYPEMULTIALKALI

HEAD-ON TYPEBIALKALI

GaAs

HEAD-ON TYPELOW-NOISE BIALKALI

SIDE-ON TYPELOW-NOISE BIALKALI

SIDE-ON TYPEMULTIALKALI

10 8

10 6

10 5

10 4

10 2

10 3

10 1

10 0

10−1

TPMOB0066EB

DISTANCEBETWEEN METAL CASEAND GLASS BULB

METALCASE

GLASSBULB

ANODE OUTPUT

MICROAM-METER

RMSVOLT-METER

1 MΩ 3 pF

−1000 V

DARK CURRENT

RMS NOISE

10 10-8

10-9

10-10

1

0.1

RM

S N

OIS

E (

mV

)

DA

RK

CU

RR

EN

T (A

)

0 2 4 6 8 10 12DISTANCE BETWEEN METAL CASE AND GLASS BULB (mm)

TPMOC0014EC

PH

OTO

MU

LTIP

LIE

RTU

BE

12

However, in most cases, the input window of the photo-multiplier tube is exposed even with the HA coating. There-fore, in the anode grounded scheme, use of good insulat-ing material such as fluorocarbon polymers or polycarbon-ate is necessary around the input window in negative HVoperation. Otherwise, a large potential difference may becreated at the input window, and could result in irregularand high dark counts.To avoid this problem, adopting a cathode groundedscheme is strongly recommended.

(3) Leakage CurrentLeakage current may be another source of noise. It mayincrease due to imperfect insulation of photomultiplier tubebase or socket pins, and also due to contamination on thecircuit board. It is therefore necessary to clean these partswith appropriate solvent like ethyl alcohol.In addition, when a photomultiplier tube is used with acooler and if high humidity is present, the photomultipliertube leads and socket are subject to frost or condensa-tion. This also results in leakage current and thereforespecial attention should be paid.

(4) Field Emission NoiseThis is voltage-dependent noise. When a photomultipliertube is operated at a high voltage near the maximum rat-ing, a strong local electric field may induce a small amountof discharge causing dark pulses. It is therefore recom-mended that the photomultiplier tube be operated at avoltage sufficiently lower than the maximum rating.

(5) External NoiseBesides the noise from the photomultiplier tube itself, thereare external noises that affect photomultiplier tube opera-tion such as inductive noise. Use of an electromagneticshield case is advisable.Vibration may also result in noisemixing into the signal.

(6) RingingIf impedance mismatching occurs in the signal output linefrom a photomultiplier tube, ringing may result, causingcount error. This problem becomes greater in circuits han-dling higher speeds. The photomultiplier tube and thepreamplifier should be connected in as short a distanceas possible, and proper impedance matching should beprovided at the input of the preamplifier.

3-5 Magnetic Shield

Most photomultiplier tubes are very sensitive to magneticfields and the output varies significantly even with terres-trial magnetism (approx. 0.04 mT ). Figure 19 shows typi-cal examples of how photomultiplier tubes are affected bythe presence of a magnetic field. Although photomultiplier

tubes in the photon counting mode are less sensitive tothe magnetic field than in the analog mode, photomulti-plier tubes should not be operated near any device pro-ducing a magnetic field (motor, metallic tools which aremagnetized, etc.). When a photomultiplier tube has to beoperated in a magnetic field, it is necessary to cover thephotomultiplier tube with a magnetic shield case.

Figure 19 : Typical Effects by MagneticFields Perpendicular to Tube Axis

3-6 Stability and Dark Storage

In either the photon counting mode or analog mode, thedark current and dark count of a photomultiplier tube usu-ally increase just after strong light is irradiated on the pho-tocathode. To operate a photomultiplier tube with good

Figure 20 : Effect of Dark Storage in Noise Reduction

101

0 50 100 150

102

104

103

DA

RK

CO

UN

TS

(s-1

)

TIME(min)

PHOTOMULTIPLIER TUBE LEFT IN DARKNESS

PHOTOMULTIPLIER TUBENOT LEFT IN DARKNESS

TPHOB0039EB

120

110

100

90

80

70

60

50

40

30

20

10

0-0.5 -0.4 -0.3 -0.2 -0.1 0.1 0.30.2 0.40 0.5

TPMOB0017EB

RE

LAT

IVE

OU

TP

UT (

%)

28 mm dia.

SIDE - ON TYPE

13 mm dia.HEAD - ON TYPELINEAR - FOCUSEDTYPE DYNODES( )

38 mm dia.HEAD - ON TYPECIRCULAR CAGETYPE DYNODES( )

MAGNETIC FLUX DENSITY (mT)

13

stability, it is necessary to leave the photomultiplier tube indark state without allowing the incident light to enter thephotocathode for about one or more hours. (This is called"dark storage" or "dark adaptation".) As Figure 20 shows,dark storage is effective in reducing the dark count rapidlyin the actual measurement.

3-7 Uniformity

Uniformity is the variation in photomultiplier tube outputwith respect to the photocathode position at which lightenters. As stated in 3-2 "Collection efficiency (CE)", evenif uniform light enters the entire photocathode of a photo-multiplier tube, some electrons emitted from a certain po-sition of the photocathode are not efficiently collected bythe first dynode (Dy1). This phenomenon causes varia-tions in uniformity as shown in Figure 21. If photons entera position of poor uniformity, not all the photoelectronsemitted from there are detected, thus lowering the detec-tion efficiency. In general, head-on photomultiplier tubesprovide better spatial uniformity than side-on photomulti-plier tubes. For either type, good uniformity is obtainedwhen light enters around the center of a photocathode.

Figure 21 : Typical Uniformity

3-8 Signal-to-Noise Ratio

This section describes theoretical analysis of the signal-to-noise consideration in both photon counting and ana-log modes. The noise being discussed here is mainly shotnoise superimposed on the signal.(1) Analog ModeWhen signal light enters the photocathode of a photomul-tiplier tube, photoelectrons are emitted. This process oc-curs accompanied by statistical fluctuations. The photo-cathode signal current or average photocurrent Iph there-fore includes an AC component which is equal to the shotnoise iph expressed below.

where e is the electron charge and B is the frequency band-width of the measurement system.The shot noise which is superimposed on the signal canbe categorized by origin as follows :

a) Shot Noise Resulting from Signal LightSince the secondary electron emission in a photomul-tiplier tube occurs with statistical probability, the result-ing output also has statistical fluctuation. Thus the noisecurrent at anode, isa, is given by

where µ is the gain of the photomultiplier tube, F is thenoise figure of the photomultiplier tube. If we let thesecondary electron emission ratio per dynode stagebe δn, the noise figure for the photomultiplier tube hav-ing n dynode stages can be expressed as follows:

Supposing that δ 1=5, δ 2= δ 3= ……… = δ n=3, the noisefigure takes a value of approximately 1.3 .At this point, the photocurrent Iph is given by

where P i is the average light level entering the photo-multiplier tube, η ( λ ) is the photocathode QE at thewavelength λ , α is the photoelectron CE and hν is theenergy per photon.

b) Shot Noise Resulting from BackgroundAs with the shot noise caused by signal light, the shotnoise at the anode resulting from the background Pb

can be expressed as follows:

where Ib is the equivalent average cathode current pro-duced by the background light.

c) Shot Noise Resulting from Dark CurrentDark current may be categorized by cause as follows:1 Thermionic emission from the photocathode and

dynodes.2 Fluctuation by leakage current between electrodes.3 Field emission current and ionization current from

residual gases inside the tube.Among these, a major cause of the dark current is ther-mionic emission from the photocathode.

isa = 2eIphFB µ

F=1

1 1(1+ + + + )1 2

1 1 nδδ δ δ δ

• • • • • • • • • • •• • • • • ••

Iph=P i ( ) e

hη λ α

iba = 2eIbFB

Ib =P b ( ) e

h

µη λ α

iph= 2eIphB

(1) Head-on Type (2) Side-on Type

100

50

0AN

OD

E S

EN

SIT

IVIT

Y (%

)

PHOTOCATHODEVIEWEDFROM TOP

SE

NS

ITIV

ITY

(%)

AN

OD

E

50SENSITIVITY(%)

ANODE

PHOTOCATHODE

GUIDE KEY

500 1000

100

TPMHC0085EB TPMSC0030EB

14

In a low bandwidth region up to several kHz, the NEPmainly depends on the shot noise caused by dark cur-rent (the latter component in the above equation). In ahigh bandwidth region, the noise component (the formercomponent in the above equation) resulting from thecathode radiant sensitivity (Sp/µ in the equation) pre-dominates the NEP.The noise can also be defined as equivalent noise in-put (ENI). The ENI is basically the same parameter asthe NEP, and is expressed in lumens (Sp is measuredin units of amperes per lumen in this case) or watts.

(2) Photon Counting ModeIn the analog mode, all pulse height fluctuations occurringduring the multiplication process appear on the output.However, the photon counting mode can reduce such fluc-tuations by setting a discrimination level on the output pulseheight, allowing a significant improvement in the signal-to-noise ratio.In the photon counting mode in which randomly gener-ated photons are detected, the number of signal pulsescounted for a certain period of time exhibits a temporalfluctuation that can be expressed as a Poisson distribu-tion. If we let the average number of signal pulses be N, itincludes fluctuation (mean deviation) which is expressedin the shot noise n = N. The amplifier noise can be ig-nored in the photon counting mode by setting the photo-multiplier tube gain at a sufficiently high level, so that thediscrimination level can be easily set higher than amplifiernoise level.As with the analog mode, dark current may be grouped bycause as follows:(a)Shot noise resulting from signal light

nph = Nph

(Nph is the number of counts by signal light)

(b) Shot noise resulting from background lightnb = Nb

(Nb is the number of counts by background light)

(c) Shot noise resulting from dark countsnd = Nd

(Nd is the number of dark counts)

In actual measurement, it is not possible to detect Nph sepa-rately. Therefore, the total number of counts (Nph+Nb+Nd)is first obtained and then the background and dark counts(Nb+Nd) are measured over the same period of time by

Therefore the shot noise resulting from anode dark cur-rent ida can be expressed as shown below:

where Id is the equivalent average dark current fromthe photocathode.

d) Noise from Succeeding AmplifierWhen an amplifier with noise figure Fa is connected tothe photomultiplier tube load, the noise converted intothe input of the amplifier is given by

where Req is the equivalent resistance used to connectthe photomultiplier tube with the amplifier, T is the ab-solute temperature and k is the Boltzmann constant.

e) Signal-to-Noise RatioTaking into account the background noise (Ib+Id), thesignal-to-noise ratio (S/N) of the photomultiplier tubeoutput becomes

... (1)

Among the above equations, the amplifier noise canbe generally ignored because the gain µ of the photo-multiplier tube is sufficiently large, so the signal-to-noiseratio can be expressed as follows:

................. (1)'

f) Noise Equivalent PowerIn addition, the noise can also be expressed in termsof noise equivalent power (NEP). The NEP is the lightpower required to obtain a signal-to-noise ratio of 1,that is, the light level to produce a signal current equiva-lent to the noise current. The NEP indicates the lowerlimit of light detection and is usually expressed in watts.From equation (1)' above, the NEP at a given wave-length can be calculated by using Ib = 0 and S/N=1, asfollows:

S/N = =2eF(Ipha+2Ida) B 2eF(SpP i+2Ida) B

I pha Sp P i

where Ipha is anode signal current, Ida is anode dark current,and Ipha=Iph =Sp P iSp: anode radiant sensitivity, P i : input power

Then calculate the variable P i

S pP i = 2eF (SpPi+2Ida) B

(SpP i )2−2eF (S pP i+2I da) B = 0

µ

µ

µ

µ

µ

Therefore, NEP is given by P i as follows:

P i = +eF B (eF B )2+4eFIda B

S p S p

µ µ µida = 2eIdFB µ

4FakTBReq

iamp =

Iph

2eFB Iph+2(Ib+Id) +(4FakTB/Req)/ 2S/N =

µ

Iph

2eFB Iph+2(Ib+Id)S/N

15

removing the input light. Then Nph is calculated by sub-tracting (Nb+Nd) from (Nph+Nb+Nd). From this, each noisecomponent can be regarded as an independent factor, sothe total noise component can be analyzed as follows:Here, substituting nph = Nph, nb = Nb and nd = Nd

Thus the signal-to-noise ratio (S/N) becomes

.............. (3)

The number of counts per second for N'ph, N'b and N'd iseasily obtained as shown below, respectively. t is the mea-surement time in seconds.

N'ph=Nph/t, N'b=Nb/t, and N'd=Nd/tAccordingly, the equation (3) can be expressed as follows:

............ (3)'

This means that the signal-to-noise ratio can be improvedas the measurement time is made longer.

By replacing the measurement time and the number ofcounts per second with the corresponding frequency andcurrent, with t=1/(2B) and N'x=Ix/e (e= 1.6 × 10-19 C) re-spectively, it becomes clear that equation (1)' is equiva-lent to equation (3)' except for the noise figure term.

In photon counting mode, if we define the detection limitas the light level where the signal-to-noise ratio equals to1, the number of signal count per second N'ph at the de-tection limit can be approximated below, from equation(3)' under the condition that the measurement time is onesecond and the background light can be disregarded.

............. (4)

At this point, if the dark count N 'd is more than severalcounts per second, the detection limit can be approximatedwith an error of less than around 30 %.

If we let the QE at a wavelength λ (nm) be η ( λ ), the

N'ph t

N'ph+2(N' b+N' d)S/N =

100

101

102

103

104

105

106

107

108

109

1010

10-18

10-19

10-17

10-16

10-15

10-14

10-13

10-12

10-11

10-10

10-9

10-8

PHOTON NUMBER ( S-1 mm-2 )

3.6 × LIGHT POWER ( W mm-2 ) at 550 nm

incident light power Po at the detection limit can be ap-proximated as follows:

.................(5)

For your reference, let us calculate the power (P) of a pho-ton per second as follows:

As an example, the illustration in the bottom of this pageshows the relation between the light power and the num-ber of photons at a wavelength of 550 nm.

In contrast to the signal pulse height distribution (PHD)similar to a Poisson distribution, the dark current pulsesare distributed on the lower pulse height side. This is be-cause the dark current includes thermal electrons not onlyfrom the photocathode but also from dynodes. Therefore,some dark current component can be effectively eliminatedby setting a proper discrimination level without reducingmuch of the signal component. Furthermore, by placingan upper discrimination level, the photon counting modecan also eliminate the influence of environmental radia-tion which produces higher noise pulses and often causesignificant problems in the analog mode. It is now obviousthat the photon counting mode allows the measurementwith a higher signal-to-noise ratio than in the analog mode,which is even greater contribution than that obtained fromthe noise figure F.

Constant

Electron Charge

Speed of Light in Vacuum

Planck's Constant

Boltzmann's Constant

1 eV Energy

Wavelength in VacuumCorresponding to 1 eV

ReferencePhysical Constants

Symbol

e

c

h

k

eV

–

Value

1.602 × 10-19

2.998 × 108

6.626 × 10-34

1.381 × 10-23

1.602 × 10-19

1240

Unit

C

m/s

J.s

J/K

J

nm

Photon Number and Light Power

NphNph

n tot Nph+2(Nb+Nd)S/N= =

Nph+2(Nb+Nd)n tot =

N'ph s-12N' d

W-162.8 × 10

PoN'd

η λλ

W-16

2 × 10

hcP=λ

λ

16

4. Measurement SystemsPhoton counting can be performed with several measure-ment systems, including simple sequential measurementsand sampling measurements depending on the informa-tion to obtain or the light level and signal timing condi-tions.

4-1 Synchronous Photon CountingUsing Chopper

Using a mechanical chopper to interrupt incident light, thismethod makes light measurements in synchronization withthe chopper operation. More specifically, signal pulses andnoise pulses are both counted during the time that lightenters a photomultiplier tube, while noise pulses arecounted during light interruption for subtracting them fromsignal pulses and noise pulses. This method is effectivewhen a number of noise pulses are present or when ex-tremely low level light is measured. However, since thismethod uses a mechanical chopper, it is not suitable forthe measurement of high-speed phenomena. This methodis also known as the digital lock-in mode. Figure 22 showsa block diagram for this measurement system, along withthe timing chart.Figure 22 : Synchronous Photon Counting

Using Chopper

4-2 Time-Resolved PhotonCounting by Repetitive Sampling

This method uses a pulsed light source to measure tem-poral changes of repetitive events.Each event is measured by sampling at a gate timing

INCIDENTLIGHT

LED

CHOPPER

PHOTO TRANSISTOR

PMT AMP

DISCRIMINATOR

GATECIRCUIT COUNTER

SYNCHRONOUSSIGNAL

CHOPPER WIDTH

CHOPPEROPERATION

GATE CIRCUITOPERATION

SETTLING TIME

GATE FOR SIGNAL

GATE FOR NOISE

a

b

c

d

TPHOC0030EA

SAMPLING TIME

slightly delayed from the repetitive trigger signals. The sig-nal measured at each gate is accumulated to reproducethe signal waveform. This method is sometimes called thedigital boxcar mode and is useful for the measurement ofhigh-speed repetitive events. Figure 23 shows the timechart of this method.Figure 23 : Time-Resolved Photon Counting

by Repetitive Sampling

4-3 Time-Resolved Photon Countingby Multiple Gates

This method sequentially opens multiple gates and mea-sures the light level in a very short duration of the opengate, allowing a wide range of measurement from slowevents to fast events. This method can also measure singleevents and random events by continuously storing datainto memory. The time chart for this method is shown inFigure 24.Figure 24 : Time-Resolved Photon Counting

by Multiple Gates

TRIGGERSIGNAL

CLOCKSIGNAL

SIGNAL LIGHT

ADDER CIRCUITGATE SIGNAL

SUBTRACTER CIRCUITGATE SIGNAL

ADDITION GATE

SUBTRACTIONGATE

a

b

c

d

e

f

g

TIME

A1

B2C3

ZnYn

A1 An

B1C1

BnCn

YnZnZ2Z1

T1 T2 Tn

A2

B2C2

TPHOC0031EB

CO

UN

TS

INCIDENTLIGHT

GATE 1

ON

OFF

GATE 2

ON

OFF

COUNT DATA

TPHOC0032EBTIME

17

4-4 Time-Correlated PhotonCounting

Time-correlated photon counting (TCPC) is used in con-junction with a high-speed photomultiplier tube for fluo-rescence lifetime measurement (in tens of picoseconds tomicroseconds). By making the count rate sufficiently smallrelative to repetitive excitation light from a pulsed lightsource, this method measures time differences with re-spect to individual trigger pulses synchronized with exci-tation signals in the single photoelectron state.Actual fluorescence and emission can be reproduced withgood correlation by integrating the signals. Since thismethod only measures the time difference, it provides abetter time resolution than the pulse width obtainable froma photomultiplier tube. A typical system for the TCPC con-sists of a high-speed preamplifier, a discriminator with less

Figure 25 : TCPC System

time jitter called a constant fraction discriminator (CFD), atime-to-amplitude converter (TAC), a multichannel pulseheight analyzer (MCA) and a memory or computer. Figure25 shows the block diagram, time chart for this measure-ment system and one example data for fluorescence de-cay time.Besides TCPC, time-resolved measurement also includesphase difference detection using the modulation method.This method is sometimes selected due to advantagessuch as a compact light source and simpler operating cir-cuits.

Reference: ¡"TECHNICAL INFORMATION Applicationof MCP-PMTs to time-correlated single photon countingand related procedures" (available from Hamamatsu).¡"TECHNICAL INFORMATION Modulated Photomulti-plier Tube module H6573" (available from Hamamatsu)

PULSE LIGHT SOURCE FILTERSAMPLE

PINPHOTODIODE

PMT AMP.

(b)

(c)

(a)

(d)

DISCRIMINATOR

DISCRIMINATOR

CFD

TAC MCA

TRIGGER

COMPUTER

DELAY CIRCUIT

(A) Measurement block diagram

(B) Time chart

(C) Example of TCPC Measurement (Sample : Cryptocyanine in ethanol)

TRRIGER

PMT OUTPUT

CFD OUTPUT

TAC OUTPUT

t 1

v1

t 3

v3

t 2

v2

TPHOC0033ED

FILTER

TIME (ns)0.20 0.4 0.6 0.8 1.0

CO

UN

TS

(s-1

)

104

103

102

101

100

FLUORESCENCE DECAY

(a)

(b)

(c)

(d)

18

4-5 Calculating theAutocorrelation FunctionUsing a Digital Correlator

Autocorrelation is a technique capable of finding repeatedpatterns in a signal or periodic signals in apparently ir-regular phenomena. Autocorrelation is a valuable techniquefor dynamic light scattering (DLS) and fluorescence cor-relation spectroscopy (FCS).When measuring a signal output from a photodetector whileviewing the signal intensity over time, and the signal inten-sities at a certain time t and another time delayed by τ arelarge, and this behavior repeatedly occurs, then that corre-lation is said to be high at time τ. The extent of this correla-tion can be expressed by the autocorrelation function.When finding the autocorrelation function from the signaloutput by a photodetector, the autocorrelation functionversus the time delay τ, G(τ), is expressed by equation(1). In the photon counting technique, G(τ) can also becalculated by equation (2) for time-series data containingthe number of photons ni detected per unit time (∆t: gatetime) as the element. In each equation the values in < >are average values. M is the number of time-series data.

.............(2)

G (τ )= ⟨F ( t )F ( t+τ ) ⟩

G (τ )=G (k∆ t )= nini+k m=M-k∴1m

m

Σi=1

.........................................(1)

Here, if the average of the signal intensity is <F>, the sig-nal intensities at times t and τ can be respectively ex-pressed by using the fluctuations δF(t) and δF(t+τ) as inF(t) = δF(t) + <F> and F(t+τ) = δF(t+τ) + <F>. Equation (1)can therefore be rewritten as follows:

∴

G(τ)=⟨(δF(t)+⟨F⟩)(δF(t+τ)+⟨F⟩)⟩

=⟨δF(t)δF(t+τ)⟩+⟨F⟩2

⟨δF(t)⟩=⟨δF(t+τ)⟩=0

..................(3)

From equation (3) we can see that the autocorrelation func-tion is expressed by the intensity fluctuation term,<δF(t)δF(t+τ)>, and the square term of the average inten-sity, <F>2.

Here, the signal intensity fluctuation at time t, δF(t), canbe given by equation (4). If evaluating only the signal in-tensity fluctuation, the autocorrelation function can be ob-tained by calculating it by normalizing with the averagesignal intensity as in equation (5).

δF(t)=F(t)-⟨F⟩

g(τ)=

=

=

⟨δF(t)δF(t+τ)⟩⟨F⟩2

⟨(F(t)-⟨F⟩)(F(t+τ)-⟨F⟩)⟩⟨F⟩2

⟨F(t)F(t+τ)⟩-⟨F⟩2

⟨F⟩2

∴⟨F(t)⟩=⟨F(t+τ)⟩=⟨F⟩

.........................................(4)

.................................(5)

In the photon counting technique, equation (5) is convertedto:

⟨F(t)F(t+τ)⟩= nini+k

1m

m

Σi=1

⟨F⟩2 = ni

1m2

m

Σi=1

ni+k

m

Σi=1

So the autocorrelation function can be obtained by equa-tion (6) as follows:

m nini+k -

g(k∆t)=

m

Σi=1

ni

m

Σi=1

ni+k

m

Σi=1

ni

m

Σi=1

ni+k

m

Σi=1

Because calculating the autocorrelation function requiresa great deal of time, the calculations were usually madeby dedicated hardware correlators. However, recent im-provements in computer processing speed now allow thesecalculations to be made by software. Our digital correlator(M9003+U8451) uses software to calculate theautocorrelation.

.....................(6)

1

1.1

1.2

1.3

1.4

1.5

1.6

1.7

0 0.01 0.1 1 10 100 1000

TIME t (s)

Fluctuation in fluorescence intensity change over time Autocorrelation function vs. time delay

FLU

OR

ES

CE

NC

E IN

TEN

SIT

Y F

(t) C

OU

NT

(×1

000

s-1 )

TIME τ (ms)

AU

TO

CO

RR

ELA

TIO

N F

UN

CT

IONSingle fluctuation at time t

δF(t)=F(t)-⟨F⟩

⟨F⟩Averagesingleintensity

g(τ)=⟨δF(t)δF(t+τ)⟩

⟨F⟩2

21

4

6

8

10

12

14

16

0 1 2 3

18

20

Fluctuation in fluorescence intensity change over time and its autocorrelation function

19

(5) Dark Count → Lower Detection LimitThe dark count is an important factor for determining thelower detection limit of a photomultiplier tube. Selectingthe photomultiplier tube with minimum dark counts is pre-ferred. In general, for the same electrode structure, thedark count tends to increase with a larger photocathodeand higher sensitivity in a long wavelength range. When aphotomultiplier tube is used in a long wavelength rangeabove 700 nm, cooling the photomultiplier tube is recom-mended.

(6) Response Time →Maximum Count Rate,Time Resolution

The maximum count rate (s-1) in the photon counting modeis determined by the time response of the photomultipliertube and the frequency characteristics of the signal pro-cessing circuit, or by the pulse width.Most photomultiplier tubes have no problem with the timeresponse up to a maximum count rate (random pulse) of3×106 s-1. However, in applications where the maximumcount rate higher than this level is expected, a photomul-tiplier tube with the rise time (t r) shorter than 5 ns must beselected. In time-correlated photon counting (TCPC), theelectron transit time spread (TTS) is more important thanthe rise time itself.

5-2 Photomultiplier Tubes forPhoton Counting

Table 6 (on pages 24 and 25) is a listing of typicalHamamatsu photomultiplier tubes designed or selectedfor photon counting, showing general specifications for thespectral response and configurations. For more informa-tion, refer to the Hamamatsu photomultiplier tubes cata-log. In addition the to photomultiplier tubes listed in thetable, you may choose from other Hamamatsu photomul-tiplier tubes for photon counting. Please consult our salesoffice.

5. Selection GuideThis section describes how to select the optimum photo-multiplier tube for photon counting, along with their briefspecifications and related products. The following notesshould be taken into account in selecting the photomulti-plier tube that matches your needs.

5-1 Selecting The PhotomultiplierTube

(1) Photomultiplier Tube StructurePhotomultiplier tubes are roughly grouped into side-on andhead-on types, so it is important to select the photomulti-plier tube structure type according to the optical measure-ment conditions.Side-on photomultiplier tubes have a rectangular photo-sensitive area and are therefore suitable for use in spec-trophotometers where the output light is in the form of aslit, or for the detection of condensed light or collimatedlight.Head-on photomultiplier tubes offer a wide choice of pho-tosensitive areas from 5 mm to 120 mm in diameter andcan be used under various optical conditions. Note, how-ever, that selecting the photomultiplier tube with a photo-sensitive area larger than necessary may increase darkcounts, thus deteriorating the signal-to-noise ratio.

(2) Photocathode Quantum Efficiency (QE)The QE has direct effects on the detection efficiency. It isimportant to select a photomultiplier tube that provides highQE in the desired wavelength range.

(3) GainGain required of the photomultiplier tube differs depend-ing on the gain or equivalent noise input of the amplifierconnected. As a general guide, it is advisable to select aphotomultiplier tube having a gain higher than 1 × 106,although it depends on the pulse width of the photomulti-plier tube.

(4) Single Photoelectron Pulse Height Distribution(PHD)

Although not listed in the catalog, the PHD is an importantfactor since it relates to photomultiplier tube detection ef-ficiency and stability. Hamamatsu designs photomultipliertubes for photon counting, while taking the PHD into ac-count. As a guide for estimating whether PHD is good ornot, the peak to valley count ratio is sometimes used. (SeeFigure 6.)

20

5-3 Peripheral devices

There are many peripheral units and devices for photoncounting. This section introduces photon counting units,photon counting heads, counting boards, high-voltage powersupplies, coolers and amplifiers available from Hamamatsu,giving a brief description and specifications. For detailedinformation please refer to our product catalogs.

Photon counting headsPhoton counting heads integrate a photomultiplier tubespecially selected for photon counting, a high-voltagepower supply and a photon counting circuit. Photon count-ing can be easily performed by simply counting the outputsignal pulses with a counting board or commercially avail-able pulse counter. The supply voltage for the photomulti-plier tube and the discrimination level are optimized sothat no adjustments are required before use.

PMT modules- Photon counting type -PMT modules include a photon counting photomultipliertube and a high-voltage power supply. Using a PMT mod-ule with a photon counting unit or data acquisition unit al-lows you to easily perform photon counting.

Counting boards, photon counting unitsThese devices are used to count the number of photo-electron pulses converted to logic (TTL) signals. Both PCIbus add-in board and USB interface types are provided.Our product lineup also includes high-speed types havinga digital correlator function or capable of calculating theautocorrelation function. (See section 4-5.)

Photon counting unitsPhoton counting units convert single photoelectron pulsesdetected by a photomultiplier tube or PMT module into apositive logic signal by using an internal photon countingcircuit.

Data acquisition unitsData acquisition units convert PMT module photoelectronpulses into a digital signal with a built-in photon countingcircuit and send the count data to a PC.

D-type socket assembliesHamamatsu provides a full line of D-type socket assem-blies having a voltage divider circuit optimized for photoncounting. Select a socket assembly that best matches thephotomultiplier tube type to be used. D-type socket as-semblies specifically designed for use with coolers are alsoprovided.The lineup also includes DP-type socket assemblies thatincorporate a voltage divider circuit and a high-voltagepower supply.

High-voltage power suppliesThe output from the photomultiplier tube fluctuates withchanges in the supply voltage. To maintain stable photo-multiplier tube operation, it is essential to select a high-voltage power supply that has minimal drift, ripple, andtemperature dependence, as well as good input voltageand load regulation.Hamamatsu high-voltage power supplies are designed tomeet these requirements for high stability.Power supplies come in various configurations includingon-board types for mounting on PC boards, bench-toptypes for general use, and multi-output types for supplyinghigh and low voltages.

CoolersCoolers are effective in reducing photomultiplier tube noise(dark count). We currently provide 4 cooler models eachdesigned for use with various types of photomultiplier tubes.

Amplifier unitsAmplifier units are useful for compensating for photomulti-plier tube gain. In photon counting, amplifier units shouldhave a gain of at least 10 times (20 dB) and a frequencybandwidth of several dozen megahertz. Select an ampli-fier unit having the gain and frequency bandwidth that meetyour application.We also provide high-speed amplifier units that are idealfor fluorescence lifetime measurement.

Power supply for PMT modulesIntegrated into a single unit, these power supplies providethe drive voltage (±15 V) and control voltage (0 to 1.2 V)needed to operate a PMT module.

Magnetic shield casesMagnetic shield cases reduce adverse effects of magneticfields all the way down to 1000th or even a 10 000th oftheir original strength to ensure photomultiplier tube sta-bility. We have a complete lineup of magnetic shield casesfor various sizes of photomultiplier tubes.

21

Photon CountingHead

Low VoltagePower Supply

(+5 V)

+12 V

Data AcquisitionUnit C8907

RS-232C

RS-232C

PCI

Photomultiplier TubeModule

(Photon Counting Type)

+5 V

CommercialCounter

Low VoltagePower Supply

(+5 V)

Low VoltagePower Supply

(+5 V)

Low VoltagePower Supply

PC

PC

PC

PC

PC

Photon Counting Headwith Internal CPU+Interface

Photon Counting Head Counting Board

PCI

Photomultiplier TubeModule

(Photon Counting Type)

Photon CountingUnit

C3866, C6465

Counting Board

USB

Photon Counting Head Counting UnitC8855

Counting Board

Power Supplyfor

PMT ModuleC7169

Power Supplyfor

PMT ModuleC7169

AC Adapter

Low VoltagePower Supply

PCI

PCPhoton Counting

Head

Correlator Board M9003

Fluorescence Correlation Spectroscopy

SoftwareU9451

Counting Board

PCI

PC

PMT

Cooler or Magnetic Shield Case

D-Type SocketAssembly

Amplifier Unit Photon CountingUnit

C3866, C6465

CommercialCounter

Low VoltagePower Supply

Low VoltagePower SupplyHigh Voltage

Power Supply

Related Products for Photon Counting<Configuration Example>

TPHOC0161EA

TPHOC0047EA

22

WavelengthRegion

Type No.Spectral

Response (nm)Effective

Area (mm) Typ.Note

Internal prescaler of division by 4High QE (40 % at 580 nm), cooled type

Cable outputCable outputInternal microprocessor and RS-232C interface

Cable output

Cable outputInternal microprocessor and RS-232C interface

UV to Visible

Visible

Visible toNear-Infrared

UV toNear-Infrared

185 to 680300 to 650300 to 650300 to 720300 to 650300 to 650300 to 650300 to 650300 to 650380 to 890300 to 850300 to 850300 to 850300 to 850185 to 850185 to 900

Input VoltageDC (V)

+5+5+5+5+5+5+5+5+5+5+5+5+5+5+5+5

Dark Count (s-1)

30100100100200156010015012520005000600

10 00080400

Max.80500500300500803005003003753500

15 0001000

15 000200800

4 × 20φ 8φ 8φ 5

φ 15φ 22φ 22φ 8

φ 22φ 5

φ 15φ 22φ 8

φ 224 × 204 × 20

Count Linearity(s-1)

2.5 × 106

1.5 × 106

1 × 107

1.5 × 106

1.5 × 106

6.0 × 106

6.0 × 106

1.5 × 106

2 × 107

1.5 × 106

1.5 × 106

6.0 × 106

1.5 × 106

2 × 107

2.5 × 106

2.5 × 106

H8259H7155H7155-20H7421-40H7828H7360-01H7360-02H7467H9319H7421-50H7828-01H7360-03H7467-01H9319-02H8259-01H8259-02

Table 1: Photon Counting Head

Internal microprocessor, RS-232C interface and prescaler of division by 4

Internal microprocessor, RS-232C interface and prescaler of division by 4

WavelengthRegion

Type No.Spectral

Response (nm)Effective

Area (mm) Typ.Note

Pin outputCable outputCooled type

Cooled type

UV to Visible

Visible

Visible toNear-InfraredUV to Near-Infrared

185 to 680300 to 650300 to 650300 to 720300 to 650380 to 890300 to 850185 to 850

Input VoltageDC (V)

+15+15+15+15+15+15+15+15

Dark Count (s-1)

308080100200125200080

Max.804004003005003753500200

4 × 20φ 8φ 8φ 5

φ 15φ 5

φ 154 × 20

Rise Time(ns)2.20.780.781.01.51.01.52.0

H7732P-01H5773PH5783PH7422P-40H7826PH7422P-50H7826P-01H7732P-11

Table 2: Photomultiplier Tube Module (Photon Counting Type)

Type No.Signal Input

LevelSupply VoltageCompatible OS

5 V/1 A(Supplied from PCI bus)5 V/0.5 A(Supplied from accessory AC adapter)5 V/1 A(Supplied from PCI bus)

M8784

C8855

M9003

Internal CounterGate Time

CounterGate Model

10 µs to 10 s(100 ns or more)*

50 µs to 10 s

50 ns to 12.8 µs

TTL Positive logic

TTL Positive logic

TTL Positive logic

Internal / External

Internal only

Reciprocal mode/ Gate mode

Interface

PCI (half size)

USB (Ver. 1.1)

PCI (half size)

Windows98/98SE/Me/2000Windows98/98SE/Me/2000Windows2000/XP Pro

* Counter gate mode: external

Table 3: Counting Board (Sample Software Supplied)

C3866

C6465

Discrimination Level(Converted into input)

(mV) (s-1)

InputImpedanceType No.

(Ω)

GainRequiredin PMT

-0.5 to -16

-2.2 to -31

50

50

3 × 106

5 × 106

Prescaler

1/11/10

—

MaximumCount Rate

4 × 106

1 × 107

1 × 106

(ns)

Pulse PairResolution

2510

60

(ns)

Output PulseWidth

10Depending on count rate

30

Supply Voltage

+5.2 ± 0.2 V, 150 mA/-5.2 ± 0.2 V, 300 mA

+5 V, 60 mA/-5 V, 120 mA

OutputPulse

C-MOS

TTLPositive Logic

Table 4: Counting Unit

C6438C9663C5594-44c

aAt 50 Ω load resistance.bValue after current-to-voltage conversion by input impedance. cContact our sales office for other connectors for C5594.

Frequency Bandwidth(-3 db) (db) (Ω) (mA)(V)(V)

DC to 50 MHzDC to 150 MHz

50 kHz to 1.5 GHz

Voltage GainType No.

20 ± 3 a (approx. ×10)38 ± 3 a (approx. ×80)36 ± 3 a (approx. ×63)

Current-to-voltageConversion Factor

0.5 mV/µA b

4 mV/µA b

3.15 mV/µA b

AmplifierInput

(Output)±Voltage (non-inverted)±Voltage (non-inverted)±Voltage (non-inverted)

InputImpedance

InputVoltage

505050

±5±5

+12 to +16

Max. OutputSignal Voltage

±1 a

±1.4 a

-2.5 a

Input Current(Max.)

±55±80+95

Table 5: Amplifier Unit

23

MEMO

24

Table 6: Typical Photomultiplier Tubes for Photon Counting

Wavelength Region

VUV to UV

UV to Visible

Visible

Visible to Near-Infrared

UV to Near-Infrared

Configuration

28mm side-on28mm head-on52mm head-on52mm head-on

MCP-PMT13mm side-on13mm side-on28mm side-on28mm side-on28mm side-on28mm side-on10mm head-on13mm head-on13mm head-on16mm head-on19mm head-on19mm head-on25mm head-on25mm head-on28mm head-on28mm head-on52mm head-on52mm head-on52mm head-on

19mm head-on

28mm head-on

28mm head-on

52mm head-on

52mm head-on

52mm head-on

13mm head-on

28mm head-on

52mm head-on

MCP-PMT

13mm side-on

28mm side-on

28mm side-on

28mm side-on

28mm side-on

28mm side-on

Spectral Response(nm)

160 to 320160 to 650160 to 650160 to 650160 to 680185 to 650185 to 680185 to 680185 to 710185 to 710185 to 730300 to 650300 to 650300 to 650300 to 650300 to 650300 to 650300 to 650300 to 650300 to 650300 to 650300 to 650300 to 650300 to 650

300 to 850

300 to 850

300 to 900

300 to 850

300 to 900

300 to 1010

185 to 850

185 to 850

160 to 930

160 to 850

185 to 830

185 to 850

185 to 900

185 to 900

185 to 930

185 to 1010

Effective Area(mm)8 × 24φ 105 × 8φ 10φ 11

4 × 134 × 138 × 248 × 24

10 × 248 × 24

φ 8φ 10φ 10φ 8φ 4

φ 15φ 21φ 21φ 10φ 255 × 8φ 46φ 10

φ 4

φ 10

φ 25

5 × 8

φ 46

10 × 10

φ 10

φ 25

10 × 10

φ 11

4 × 13

8 × 24

8 × 24

8 × 6

3 × 12

3 × 12

QE Peak (%)352222221820231923233026262021222026202228222622200.3200.461

200.38

0.213

0.25220.3230.32212187

24132611252.7252.7248

140.13

Wavelength (nm)220420390390420270330300320320300390390375390390375390375420390390390390340800420800600800340800600800340

1000270800270800300800420600240600270600260800260800330800330

1000

MaximumSupply Voltage (V)

Photocathode QE

125015001500250034001250125012501250125012501500125015001000125012501250125015001000150027002500

1250

1500

1500

1500

2500

2200

1250

1500

2200

3400

1250

1250

1250

1250

1500

1500

Type No.

R7154PR7207-01R585R3235-01R3809U-52R6350PR6353PR1527PR4220PR5983PR7518PR1635PR647PR2557PR7400PR2295R5610PR1924PR3550PR7205-01R6095PR464R329PR3234-01

R1878

R7206-01

R2228P

R649

R2257P

R3310-02

R1463P

R1104P

R943-02

R3809U-50

R6358P

R4632

R928P

R2949

R636P

R2658P

A Characteristics are measured at a supply voltage giving a gain of 2 × 105.B Characteristics are measured at 1500 V regardless of gain.

Supply Voltage Typ. (V)(at 1 × 106 gain)

7007708001500

3000A750800780730720730120090010009009008501000800770900100015001500

1100

770

1100

800

1700

1700

1000

750

1700

3000A

830

830

720

720

1300

1500B

Rise Time(ns)2.21.7

13.01.3

0.151.41.42.22.22.22.20.82.52.2

0.782.51.52.02.01.74.0

13.02.61.3

2.5

1.7

15.0

13.0

2.6

3.0

2.5

15.0

3.0

0.15

1.4

2.2

2.2

2.2

2.0

2.0

TTS(ns)1.21.2—

0.450.0250.750.751.21.21.21.20.71.6—

0.231.0———1.2——1.0

0.45

1.0

1.2

—

—

1.0

—

—

—

—

0.025

0.75

1.2

1.2

1.2

1.2

1.2

Typ. (s-1)510550—1051010101010080108021510020101005

20050

100

300

150

200

600

30

900

4500

20

—

20

50

500

3003

15

50

Max. (s-1)2030151505030105050505040040030400545300603025015600150

250

1000

500

350

2000

150

1000

7000

50

2000

50

100

1000

500—

50

300

Temp. (°C)252525252525252525252525252525252525252525252525

25

25

-20

25

-20

-20

25

25

-20

25

25

25

25

25-20

-20

-20

Dark CountsType No.Remarks

R7154PR7207-01R585R3235-01R3809U-52R6350PR6353PR1527PR4220PR5983PR7518PR1635PR647PR2557PR7400PR2295R5610PR1924PR3550PR7205-01R6095PR464R329PR3234-01

R1878

R7206-01

R2228P

R649

R2257P

R3310-02

R1463P

R1104P

R943-02

R3809U-50

R6358P

R4632

R928P

R2949

R636P

R2658P

Low Dark Counts, Low AfterpulseLow Dark Count Type (Max. 3 s-1) availableFast Time ResponseFast Time ResponseCompact, Low Dark CountsCompact, High Sensitivity, Low Dark CountsLow Dark CountsHigh Sensitivity, Low Dark CountsWide Photocathode AreaHigh Sensitivity, Low Dark CountsCompactCompactLow Noise Bialkali PhotocathodeMetal Package, Compact, Low ProfileLow AfterpulseCompact, RuggedizedRuggedized, Low ProfileLow Profile, Low Dark CountsLow Dark Counts, Low Afterpulse

R464S (Max. 3 s-1) available

Fast Time Response, Low Afterpulse

Low Dark Counts, Low Afterpulse

Low Dark Counts, Low Afterpulse

Extended-Red Multialkali Photocathode

R649S (Max.100 s-1) available

Extended-Red Multialkali Photocathode

High Quantum Efficiency (at 1 µm), For Raman Spectroscopy

Compact

High Gain

High Sensitivity

Fast Time Response

Compact , Low Dark Counts

Low Dark Counts

High Sensitivity

Low dark counts

High Sensitivity

For UV to NIR range

25

WEB SITE www.hamamatsu.com

HAMAMATSU PHOTONICS K.K., Electron Tube Division314-5, Shimokanzo, Iwata City, Shizuoka Pref., 438-0193, JapanTelephone: (81)539/62-5248, Fax: (81)539/62-2205

Main ProductsElectron TubesPhotomultiplier TubesLight SourcesMicrofocus X-ray SourcesImage IntensifiersX-ray Image IntensifiersMicrochannel PlatesFiber Optic Plates

Opto-semiconductorsSi PhotodiodesPhoto ICPSDInGaAs PIN PhotodiodesCompound Semiconductor PhotosensorsImage SensorsLight Emitting DiodesApplication Products and ModulesOptical Communication DevicesHigh Energy Particle/X-ray Detectors

Imaging and Processing SystemsVideo Cameras for MeasurementImage Processing SystemsStreak CamerasOptical Measurement SystemsImaging and Analysis Systems

Sales OfficesASIA:HAMAMATSU PHOTONICS K.K.325-6, Sunayama-cho,Hamamatsu City, 430-8587, JapanTelephone: (81)53-452-2141, Fax: (81)53-456-7889

U.S.A.:HAMAMATSU CORPORATIONMain Office360 Foothill Road, P.O. BOX 6910,Bridgewater, N.J. 08807-0910, U.S.A.Telephone: (1)908-231-0960, Fax: (1)908-231-1218E-mail: usa@hamamatsu.com

Western U.S.A. Office:Suite 110, 2875 Moorpark Avenue San Jose, CA 95128, U.S.A.Telephone: (1)408-261-2022, Fax: (1)408-261-2522E-mail: usa@hamamatsu.com

United Kingdom:HAMAMATSU PHOTONICS UK LIMITEDMain Office2 Howard Court, 10 Tewin Road Welwyn Garden CityHertfordshire AL7 1BW, United KingdomTelephone: 44-(0)1707-294888, Fax: 44-(0)1707-325777E-mail: info@hamamatsu.co.uk

South Africa Office:PO Box 1112, Buccleuch 2066, Johannesburg, South AfricaTelephone/Fax: (27)11-802-5505

France, Portugal, Belgiun, Switzerland, Spain:HAMAMATSU PHOTONICS FRANCE S.A.R.L.8, Rue du Saule Trapu, Parc du Moulin de Massy, 91882 Massy Cedex, FranceTelephone: (33)1 69 53 71 00 Fax: (33)1 69 53 71 10E-mail: infos@hamamatsu.fr

Swiss Office:Richtersmattweg 6aCH-3054 Schüpfen, SwitzerlandTelephone: (41)31/879 70 70, Fax: (41)31/879 18 74E-mail: swiss@hamamatsu.ch

Belgian Office:7, Rue du BosquetB-1348 Louvain-La-Neuve, BelgiumTelephone: (32)10 45 63 34Fax: (32)10 45 63 67E-mail: epirson@hamamatsu.com

Spanish Office:Centro de Empresas de Nuevas TecnologiesParque Tecnologico del Valles 08290 CERDANYOLA, (Barcelona) SpainTelephone: (34)93 582 44 30Fax: (34)93 582 44 31E-mail: spain@hamamatsu.com

Germany, Denmark, Netherland, Poland:HAMAMATSU PHOTONICS DEUTSCHLAND GmbHArzbergerstr. 10,D-82211 Herrsching am Ammersee, GermanyTelephone: (49)8152-375-0, Fax: (49)8152-2658E-mail: info@hamamatsu.de

Danish Office:Skovbrynet 3, DK-5610 Assens, DenmarkTelephone: (45)4346/6333, Fax: (45)4346/6350E-mail: lkoldbaek@hamamatsu.de

Netherlands Office:PO Box 50.075, 1305 AB Almere The NetherlandTelephone: (31)36-5382123, Fax: (31)36-5382124E-mail: info@hamamatsu.nl

Poland Office:02-525 Warsaw, 8 St. A. Boboli Str., PolandTelephone: (48)22-660-8340, Fax: (48)22-660-8352E-mail: jbaszak@hamamatsu.de

North Europe and CIS:HAMAMATSU PHOTONICS NORDEN ABSmidesvägen 12SE-171 41 Solna, SwedenTelephone: (46)8-509-031-00, Fax: (46)8-509-031-01E-mail: info@hamamatsu.se

Russian Office:Riverside TowersKosmodamianskaya nab. 52/1, 14th floorRU-113054 Moscow, RussiaTelephone/Fax: (7)095 411 51 54E-mail: info@hamamatsu.ru

Italy:HAMAMATSU PHOTONICS ITALIA S.R.L.Strada della Moia, 1/E20020 Arese, (Milano), ItalyTelephone: (39)02-935 81 733, Fax: (39)02-935 81 741E-mail: info@hamamatsu.it

Rome Office:Viale Cesare Pavese, 435, 00144 Roma, ItalyTelephone: (39)06-50513454, Fax: (39)06-50513460E-mail: inforoma@hamamatsu.it

Information in this catalog isbelieved to be reliable. However,no responsibility is assumed forpossible inaccuracies or omission.Specifications are subject tochange without notice. No patentrights are granted to any of thecircuits described herein.© 2005 Hamamatsu Photonics K.K.

Quality, technology, and service are part of every product.

TPHO9001E04JUL. 2005 IPPrinted in Japan (1500)