Liquid Scintillation Counting

Liquid Scintillation Counting As unstable nuclides spontaneously change into more stable species, they emit charged particles and/or electromagnetic radiation (Alpha particles, Beta particles (electrons), Gamma rays or X-rays). Gamma rays and X-rays are identical - the gamma ray originating from the nucleus and the X-ray from the orbital electrons. Nuclides that emit gamma or X-ray are best counted with solid crystal scintillator such as sodium iodide.

Beta particles are best counted by liquid scintillation (LS) methods in which the nuclide is intimately mixed with a scintillator. Nuclides used in radioimmunoassays of clinical interest that emit gamma or X-rays include 125I, 131I, 51Cr, 57Co. Some radionuclides that emit gamma or X- ray also emit beta particles or electrons during their decay (131I, 60Co, 53Fe).

The total number of atoms of the radionuclide present which spontaneously decay in a unit time is the absolute activity of that sample, usually expressed as disintegration per minute DPM.

Because of some factors in the environment, every disintegration is not detected by the photo-multiplier tube. Since the counts per minute (CPM) are always a fraction of the actual disintegration per minute, the efficiency of the counting system is expressed as -

% Efficiency = CPM / DPM x 100

Procedure for the Use of Liquid Scintillation Counters:

• The radionuclide emits a beta particle which upon collision, transfers its energy to an aromatic organic solvent molecule;

• The π -electron cloud of these solvent molecules accept this energy and becomes excited to a higher energy level;

• Next, the energy from the activated solvent molecule is transferred to another organic molecule serving as the scintillator (Solute);

• The scintillator, upon return to ground state, emits light, which is detected using appropriate PMT.

Processes in Liquid Scintillation Counting Excited states Solvent Ionization Solvent Diffusion Re-combination & Migration

Internal Conversions Solute π σ Photon Emission Radioactive Quenching Sample β Emitter

2 LSC Scintillator Guide Fluorescence Scintillator Acronym emission function maximum, nm BBOT Prim &Sec. 425-435 Butyl-PBD Primary 360-365 PBD Primary 360-370 PPO Primary 360-365 p-TP Primary Bis-MSB Secondary 420-430 dimethyl Secondary 425-430 POPOP

Solvent PPO* POPOP* Touline

PMT Sensitivity

Light Emission

Wavelength 450 nm The more energetic the beta, more collisions occur, thus, yielding the greater number, N, of excited solvent molecules. Since the emitted ultraviolet radiation is not suitable for detection in the 200-260 nm wavelength range, fluorescent organic materials are added to the solvent which are capable of absorbing the UV radiation. These fluors re-emit the energy at longer wavelengths. The fluor absorbs the short wavelength energy and re-emits it at a longer wavelength, preferably in the visible region.

3 Detection Liquid Scintillation counter: A comparison of detection limits of some other methods of analysis with those detection limits obtained by using radioisotopes are shown below. Common methods of analysis

METHOD OF ANALYSIS LIMIT OF DETECTABILITY REMARKS IR Spectroscopy 1015 molecules Non-destructive UV Spectroscopy 1015 molecules Non-destructive Atomic Absorption 1013 atoms Destructive Flame Emission 1013 atoms Destructive Gas Chromatography 1013 atoms Destructive Radio-Isotopes Non-destructive 14C 1011 atoms (half life 5730 yr.) 3H 109 atoms (half life 12.26 yr.) 32P 6 x 106 atoms (half life 14.29 d)

Characterization of Beta Energy Spectrum- Distribution of beta particles emitted from a single type of nucleus

Ep dN H# 1. Maximum Particle Energy (Emax) dE E max 2. Pulse Height at Peak Energy (Ep)

3. Inflection Point (H# number)

Energy (keV) 4. Average Pulse Height – Spectral Index of the Sample (SIS) n(x)

u Σ x=0 x * n(x) SIS = K Σu x x=0 n(x) Energy (keV)

4 Commonly used beta emitting isotopes

ISOTOPE EMAX (MEV) 3H 0.018 14C 0.156 35S 0.168 45Ca 0.250 32P 1.710 131I 0.610 (F/87% BETA ACTIVITY) Tritium beta particle detection requires – • A more sensitive detector • High efficiency energy converter If a beta emitting material is interspersed with an aromatic solvent, such as benzene, toluene, or P-xylene, the emitted particle produces excitation and ionization in the solvent. Electrons in the conjugated carbon-carbon double bonds of these aromatic solvent molecules are excited to a high energy state by absorbing some of the beta particles energy and upon returning to the lower energy ground state, release their energy in the form of light. This conversion of kinetic energy to light is referred to as the scintillation process

Beta Collision Process

Excited Excited

Solute Solvent

Emitted

Photons

Ground Ground Quenching Level Level Solvent Solute

Sample

5 Solvents Solvents and fluors used in liquid scintillation counting Solvent Fluors Dioxane PPO Toluene dimethyl - popopbutyl PBO p-Xylene PBBO. Some key properties are: a) Purity b) High solubility for sample and the fluors, c) High transfer efficiency for conversion of particles energy into excited molecules d) Chemical stability. Since small amounts of certain substances can produce a dramatic loss of scintillation efficiency, highly purified solvents are necessary.

Toluene P PPO Emission h 1. PPO Absorption o Emission t o n

I n t e n s i t y 0 480 280 Wavelength (nm)

6 The Excitation Process A. Photon Emission 1. The primary excitation in liquid scintillation counting is produced by a charged particle (beta) passing through or stopped by the scintillation solution. This transfer of energy results in the production of a large number of excited molecules and ion pairs. The ions recombine with electrons to produce additional excited molecules. 2. About 60% of the fluorescing molecules are generated by this mechanism. 3. The remaining 40% are the result of direct excitation of the solvent molecule. 4. The amount of light produced by the solution (scintillation yield) will depend on the concentration of the scintillator (fluor). Excimers are the dimers formed by a pair of excited molecule and an unexcited molecule of the same kind. In this equilibrium high concentration and low temperatures favor the excimer formation. The excimer has lower energy than the monomer so that it’s fluorescence light has the longer wavelength.

B. Quenching Effect Quench - anything that interferes with the scintillation process in any of the above four steps prevents light from reaching the PMT, results in a loss in the number of recorded counts and in the apparent energy. The Causes of Quench are- 1) Impurities in the cocktail material, 2) Insufficient amounts of solute or solvent 3) Too much sample volume. Color quenching occurs when colored substances, present in the sample, absorb the light emitted by the scintillator (fluors) before it can escape the vial and reach PMT. The net effect of Color quenching is the reduction in CPM registered by the counter.

7 Scintillation yield The scintillation yield is defined as the total energy emitted by the solution (as light or photons) divided by the energy of the particle that excites the solution to fluoresce. Scintillation efficiency Sx is defined as

Sx = Nph . Eph/ Eex

Where Nph = average number of photons Eph = average energy of photons Eex = energy of the exciting particle

The solvent determines the overall scintillation efficiency. Efficiency has different value for

• each solvent

• different isotopes in the same solvent From the scintillation efficiency - it is possible to calculate the average number of photons (Nph ) produced by a particle of energy Eex.

Nph = Sx Eex / Eph

The average energy of the photons Eph can be calculated from the average wavelength of fluorescence. For the scintillator, PPO, the average wavelength of fluorescence is 3900 oA.

Eph = hν = h C/λ 5 10 5 Eph = (4.143 x 10 eV) (2.99 x 10 )/(3.8 x 10- )

At about 7 gm/liter of PPO in toluene, the transfer efficiency is about 100% and the quantum efficiency is 1.0

3 Nph = (0.04) (10 eV/KeV) (Eex)/(3.2 eV) Nph = 12.5 KeV (Eex)

The number of photoelectrons produced at the photo-cathode is now given by: No. of photoelectrons = Nph x photo-cathode efficiency The photo cathode efficiency is 28% for the bralkalu type phototube currently in use.

8 Quench Indicating Parameters

1. Based on Sample Spectrum: A parameter is derived from the sample spectrum. Quench correction curve is generated by measuring this parameter on a set of standard sources with different quenching effects. The unknown sample measurement is corrected for the quenching using this curve. Different parameters are -

• Spectral Index Sample (SIS): Average Energy of sample spectrum.

• Sample channel Ratio (SCR) Ratio of counts in two channels selected in such a way that the ratio of counts indicates the quenching effect. Channel A

CPM Channel B

Energy

• Spectral Quench Parameter of Isotope (SQP): The spectrum is transformed by log E operation. The Sample Index Spectrum of this spectrum defines the SQP and is the mean log energy of the spectrum.

SQP – Average Counts / log energy Minute

Log (Energy)

B. Quenching Effects 1. Self-quenching: Increasing the concentration beyond the optimal level will usually result in an appreciable reduction in scintillation yield. 2. Solute (Scintillator)-quenching: The reduction in efficiency with increasing scintillator concentration appears to depend on the ability of the scintillator to achieve a co-planer configuration. This may result in either an increase or decrease in measured scintillation efficiency depending on the form (monomer or dimer) that is more favorable.

Inherent interference in Liquid Scintillation Spectrum Analyses 1. Quenching – Two main types are - Chemical: causes energy loss in transfer from Solvent to solute. Colour : Causes Attenuation of photons. UnUnquenchedquenched CPMU & SISU of quenched spectrum of a sample are used to Quenched obtain the DPMU by reference to a n(x) graph between the Efficiency and SIS.

DPMU= CPMU / (% EU) *100

Channel No. Efficiency (E) is obtained from 100 counts (CPM) of a standard sample with known (DPM).

E Different amounts of quenching are % E U obtained by varying the amount of SISU quenching material in the sample but keeping the activity at a 0 constant value. SIS

Quench Indicating Parameters Internal Standards: This process involves counting the same in a normal way, followed by the addition of a small volume of the same isotope as that being counted. The sample is recounted and the increase in count rate is determined. It is assumed that the aliquot that was added to the sample was quenched in the same manner as the original sample isotope.

Sample dpm = (dpm added) * (R1 - R0)/(R2 - R1) R1 = counting rate of sample cpm. R2 = cpm of sample plus added standard R0 = background Some of the disadvantages of the internal standardization technique include: 1) Volume of the standard must be small enough to assure minimal dilution of the scintillation solution 2) the liquid must be measured accurately, 3) increased probability of manipulative errors, and 4) laborious technique when a large number of solutions are to be measured.

11 Quench Indicating Parameters

• Sample channel Ratio (SCR): Upper part of the spectrum constitutes one channel (B) that indicates the quenching. The second channel (A) is extension of this channel to lower part of the spectrum and yields total counts. The ratio of counts in these channels defines the parameter for quench correction.

Ch A 60

Ch B Unknown Counts / 1 % Sample Minute 2 3 E B/A=0.69 4 %E=48

5 0 0 1. Energy (MeV) B/A

Five standards of H3 with various amounts of quench agent (Nitromethane). ______S. Scintillator Quench Agent H3 CPM CPM %E Ch. Ratio No. ml (Nitromethane) dpm Ch. A Ch. B A/dpm B / A ______1 15 0.005 115,000 57,614 40,873 50% 0.71 2 15 0.010 115,000 46,452 28,756 40% 0.62 3 15 0.015 115,000 30,226 14,480 26% 0.48 4 15 0.025 115,000 23,051 9,061 20% 0.40 5 15 0.030 115,000 14,887 4,514 13% 0.30 ______

Unknown Sample: Counts A = 29,158 cpm, Counts B = 20,160 cpm, B/A = 0.69 %E from graph=48. Sample dpm= 29,158 / 48 * 100 = 60,745 dpm

12 Quench Indicating Parameters 2. Based on External Sample: External Standard Source conventionally a gamma emitter - produces the Compton Spectrum in the liquid scintillation vial. A set of samples with different quench factors are exposed to the external source and spectrum data are collected. A parameter derived from this spectrum is used to generate quench correction curve. Different parameters are -

• External Standard Count: The counts in the upper part of the Compton Spectrum. This method dependent on the volume (Compton effect is more with large volumes).

• External Standard Ratio: Two integral channels – one with nearly all the spectrum and the second with upper part are defined. Ratio of the counts defines quench indicating parameter.

• Inflection point of Compton Spectrum: Compton edge of the external standard spectrum. Has the disadvantage of – 1) questionable reproducibility, 2) require high activity source and 3) has poor precision with color quenching.

• Spectral Index External Standard (SIE): The average energy of the Compton Spectrum. Lower energy end for SIE computation is set to non-zero value to remove the wall effect.

• Sample Quench Parameter of External Standard: End point of external standard spectrum estimated after Log transformation. It is not a sensitive parameter for quench changes. In addition it has color and wall effect dependency.

• Transformed Spectral Index External Standard (tSIE): Transformation on the Compton spectrum of the external source is carried out to eliminate the wall effect, volume effects and colour quenching.

13 Quench Indicating Parameters

Efficiency Tracing: Instrument must have a multi channel analyzer for this purpose. The unquenched standard spectrum is measured and stored. Six region of this spectrum are analyzed and efficiency is determined. Percent efficiency in each of six regions is calculated and graph between the efficiency and actual counts in six regions is plotted.

For unknown sample the spectrum is analyzed in the same six regions and the counts in these regions (x coordinate values) are plotted. A curve is fitted on least square and extrapolated to 100 % efficiency yields sample DPM. Each isotope requires storage of reference spectrum and the efficiency tracing values. This technique is used for pure beta and gamma emitters. Tritium can have high errors for highly quenched samples. It can not be used for radionuclides with isomeric transitions and electron capture, as there is emission of x- rays or auger electrons. It is difficult by this method to estimate the True activity of these isotopes.

14 Gamma Ray Spectrum in Liquid Scintillation Counter: Composition and Characterization E < 30 keV: Conversion 30 < E < 2 MeV: Continuous electrons (57Co, Compton 51Cr) and Auger Spectrum Electrons 125I. from 0 to E*. E*= 2 E2γ / ( 2 Eγ +0.510) Where Eγ gamma ray energy in MeV.

E > 2 MeV: Pair Production. dn Constitutes the Compton part of dE liquid Scintillation spectrum.

Energy ( keV)

Characterization of Compton Spectrum: The conventional parameters used to characterize beta spectrum are used to characterize gamma ray Compton spectrum. Only SIE index is defined accurately and is used for this purpose. The parameters are –

1. Emax : Maximum end point energy of the spectrum – Not well defined. 2. Ep : Pulse Height at Peak Energy – not a well-defined point. 3. H# : Inflection Point – it is usually at the Compton Edge energy. 4. SIE : The Spectrum Index of External sample – can be used to identify the Compton Spectra. It is the average energy of the spectrum. Summation is started from non-zero energy level (L) to ignore the lower end of the spectrum that is subjected to counting interference.

u Σ x=L x * n(x) SIE = K u Σ x=L n(x)

15 External Standard Channels ratio methods, Quench Correction An external Standard channel Ratio (ESCR) is a method to detect and measure the occurrence of quenching. Quenching reduces both the number of counts detected (lower efficiency) and shifts the apparent energy of the counts to lower region.

Unquenched Sample Spectrum Quenched Sample n(x) Spectrum n(x)

Log Energy Log Energy

137Cs source is used to bombard the cocktail with gamma rays. Compton electrons produced by gamma rays inside the scintillation vial, give a constant and repeatable spectrum for a given type of sample.

Compton Spectrum Compton Spectrum of Unquenched of Quenched Sample Sample n(x) n(x)

Log Energy Log Energy

Method for correction: 1) Count of the sample activity, e.g. 3H in two energy ranges called windows. 2) 137Cs source is placed beside the scintillation vial, and activity (3H + 137Cs) is counted again in the two windows. 3) Subtract sample activity from the total counts. 4) Ratio of counts in the two windows or channels in calculated. 5) % Efficiency for this ratio is obtained from the graph between the channel ratio and Efficiency. 6) DPM for sample are obtained from 3H CPM and efficiency.

17 Quench Correction with External Standard Channel Ratio

Channel Channel B Channel Channel B

3 H 137 137 Cs Cs n(x) n(x)

Log Energy Log Energy Unquenched Sample Quenched Sample

ESCR number = 60 3 137 3 ( H + Cs)B - ( H )B 114332 - 11 Unknown Sample = = 0.362 B/A=0.362 %E=20% 3 137 3 ( H + Cs)A - ( H )A 320428 - 4511

0 0 1. ESCR B/A The ESCR number is also used to correct for quench. By setting up a quench series and plotting the counting efficiency Vs ESCR number of each member of a series of samples one obtains a quench curve. DPM is obtained as - DPM = 4511 / 20 * 100 = 22555

18 Sample Vial

PMT -1 PMT -2 A D ADD ON D CARD 2 O N

Coincidence C

A R D Summing Circuit (Analog Signal) 1

Gate Earlier Vintage Hardware PHA Counting System

ADC, FIFO, & Amplifier Amplifier Amplifier SPECTRUM A B C FORMATION H V ( MCA) PHA PHA PHA A B C PC based Spectrum SCALAR SCALAR SCALAR Analyses & A B C Counting

Contemporary PC based system has two ADDON cards (1. HV and 2. Spectrum Acquisition). The Vintage system has hardware and only a three-channel analyzer.

All LSCs use two PMTs for better performance- 1. To allow coincidence techniques to be used in the reduction of noise 2. Pulse summation ensures that the output is based on all photons and balances out the effect of positional effects of different events.

With Summation

n(x) Individual PMT Signal, i.e. no Summation

Energy Pulse Height

20 MULTILABEL COUNTING It is possible to count multi-label samples so long as the maximum energies are sufficiently different, e.g. 3H and 14C. 1. Based on Spectrum Index of Sample (SIS)

Spectrum of a Mixture of Isotopes, Estimation of Individual isotope activity

Total Spectrum (SIST, CPMT)

3 Tritium H (LE) Spectrum (SISL , CPML)

n(x) 14 Carbon C (HE) Spectrum (SISH , CPMH)

18.6 156 Energy (keV)

SISH - SIST CPML = CPMT SISH - SISL

SIST - SISL CPML = CPMT SISH - SISL

21 MULTILABEL COUNTING

2. Based on Channel Ratio: For a mixture of two radionuclides, two regions A and B are selected. Region A has sole contribution from 14C while region B that is appropriate to 3H will receive a contribution from 14C. Quench correction curves are constructed for each nuclide independently in regions A and B using the external standard.

14 CB 14 CA

3 HB 3 HA

SIS with External Standard

14 Let NA = total counts in region A , NB = total counts in region B , CA = C efficiency 14 3 3 in region A, CB = C efficiency in region B, hA = H efficiency in region A , hB = H efficiency in region B. For a mixture of isotopes with carbon activity aC tritium activity aH, counts in region A and Region B can be written as

NA = hA aH + CA aC

NB = hB aH + CB aC Solving these equations -

14 NB - NA(hB/hA) Activity of C, aC = CB - CA(hB/hA)

3 Activity of H, aH = NA - NB(CA/CB) hA - hB(CA/CB)

Chemical and Colour Quenching Chemical quenching - is the interference with energy transfer between the solvent molecules and is independent of position in the sample. Colour quenching - is a function of the distance (x) traveled in the sample by photons:

-µ x [I = I0 e ]

Where µ = absorption coefficient. Hence, colour quenching is dependent on the position of the β-particle emission in the sample. Thus quench calibration curves are different. The two curves are produced using two sets of standards. One set with an agent that produces chemical quenching. The second set has the agent that produces colour quenching.

Set of Standard Samples with Chemical Quenching 100

% E

l1 Set of Standard Wrong I2 Samples with Colour Value quenching

Point P SIS of unknown Correct sample Value

SISs For an unknown sample, a value of SISs is computed from the spectral data. Normally only one quench correction curve is generated. If the value of SISs is low, a large error may occur if the sample quenching is different from the standard curve. If point P represents the actual, but unknown, solution, then in order to define P, the distance from the quench correction curve is required, i.e. the proportion of colour / chemical quenching, i.e. l1 and l2.

23 Position Effects on Colour Quenching and Correction to Efficiency Value.

β-emission at X1, has the same path length for photons entering either PMT. For emission at X2 more photon attenuation is for PMT1 than for PMT2. Difference between pulse heights in PMT1 and PMT2 at X2 to some extent is proportional to the amount of colour quenching. The average energy (SISd) of the difference in the signal in PMT1 and PMT2 plotted against the average energy of the sum (SISs) produces a divergence between chemical and colour quenching.

I3 SISd From a Set of Colour quenched Samples

P P X X2 I M 1 M 4 T Unknown T sample From a Set of 1 SISd & 2 SISs Chemical quench samples

SISs The equivalent of point, P is located by measuring SISd and SISs. The distances l3 and l4 are measured.

It is found that the ratio F1 = I1/ I2 & ratio F2 = I3 / I4 are related. Distance I for an unknown Unknown 1 Sample F sample can be determined from 1 2 the graph between F2 and F1. Then the correction to the F2 F1 value by efficiency for colour quenching interpolation I = F * I can be made for low SISs in the 1 1 2 graph between Efficiency & SISs. 0 0 1 F1

α /β DISCRIMINATION IN LSC

Many naturally occurring sources contain a mixture of α - and β -emitters. For example, the 238U and 234U and α -emissions from 238U and 234U, and β -emissions from 234Th and 234Pa.

238U α  > 234Th β  > 234Pa β  > 234U α  >

The first five disintegration commencing with 222Rn involve 3 α -emissions and 2 β - emissions.

222Rn α  > 218Po α  > 214Pb β  > 214Bi β  > 214Po α  >

Consequently, the monitoring of either of these very important schemes must take account of the mixtures. Alpha Particle Spectrum Alpha emitters produce high-energy particles (energy from 4 MeV to 6 MeV) that have high LET values. Energy is lost in short distances but interactions with liquid scintillator have low photon yield. The light yield is approximately 10 times lower than the beta particle of similar energy. The mono-energetic alpha particles yield a broad peak (peak width approximately 1 MeV). The separation of spectra for different isotope is difficult.

Improved resolution is obtained by single PMT counting and decreasing the sample to scintillator cross-sectional area (narrow diameter tubes).

Observed Observed C C O O 233U U U 233U N N 241A T T 241A S S

0 10 0 10 Energy (MeV) Energy (MeV) 15 ml Sample 6x50mm Sample

α /β DISCRIMINATION IN LSC - 2 The ability to discriminate between α - and β -particles lies in the small difference in pulse shapes. 1. The initial interaction of the α -particle is much stronger than the β - particle due to a) the double charge, b) the low velocity. The α -range is much less than β -particle so that the ionization produces a very high-density track. 2 Compared with the β -particle, the α -particle produces fewer excitations (∼ 0.4%). Pulse height depends on the number of excitations. Pulse height produced by α is approximately 10% of pulse height produced by β of similar energy. 3 Due to the high density of ions, the probability of recombination of an ion and electron is greater for α than β. Recombination may produce a ground state molecule or an excited molecule. The excited molecule may be in a singlet state or triplet state. Number of excited states for a singlet is 1 whereas that for a triplet is 3. e- + X+ → 1X* (excited singlet) or e- + X+ → 3X* (excited triplet) 4 The excited singlet state return to ground state by emitting a photon in a very short time (i.e. no delayed emissions). 5 The excited triplets have a much longer lifetime due to the low probability of changing spin from 1 to 0. There is a probability of two 3X* molecules colliding leading to triplet annihilation and production of phonons. 3X* + 3X* → 1X* + 1X* + phonons The 1X* decays rapidly but has been delayed by the lifetime of the 3X* molecules, i.e. produces "delayed fluorescence". 6 The enhanced delayed fluorescence contribution for α produces a longer tail to the output pulse compared with β.

26 Alpha / Beta Discrimination in LSC

1. Based on Zero-Crossing: Raw shape of the signal is difficult to discriminate due to variable size of the pulses with the continuous β -spectrum. Using two CR circuits, the double differential of the pulse profile exhibits different zero-crossing points.

V Alpha

0 Time (ns) Beta

A time comparator is employed to determine the zero-crossing point τ relative to a reference. All the α pulses cross over at a time greater than τ while β pulses cross over at a time less than τ. Two spectra - ‘α ‘ pulses longer than τ and ‘β ‘ pulses shorter than τ are generated. Due to the statistical fluctuations in pulse shape, there is some spillover. The optimum value of τ has to be determined experimentally.

Alpha Beta Discriminating Method – 2: Filtering and Summation of PMT Signal

Alpha Beta

TIME (n sec) 400 From PMT 0 Anode Output 1.Alpha (> 0) RC W Filter Sum

RC W 2. Beta (< 0) Filter

From PMT Dynode

27 Alpha Beta Discriminating Method –3: Gated Integration of Signal.

Interval for Total Signal

Interval for Discriminating Signal Alpha

Beta

From PMT 0 TIME (n sec) 400 Anode

Gated Gated Integrator (B) Integrator (A)

Output Normalization 1. Alpha: Logic 1 (Divider A/B)

& TTL 2. Beta: Logic 0 Conversion

28 Cerenkov Counting in Liquid Scintillation Counter

1. Cerenkov radiation is produced when a charged particle travels through a transparent medium, at a velocity greater than the speed of light in the same medium. 2. Energetic beta particles in an aqueous Solution produce a faint, blue- white light that is amplified by the liquid scintillation counter's PMT to produce pulses in the usual manner. 3. A beta emitter must have energy greater than 263 keV to be detected in water by Cerenkov counting. 4. Common nuclide measured by the Cerenkov counting technique and their characteristics are summarized below

Isotope Emission β Spectrum Efficiency

> 263 keV Emax keV in water Counting

36Cl 46% 714 7% 32P 86% 1710 53% 90Sr / 90Y 61% 545 25%

5. Chemical quenching (not significant), Colour quenching, Volume effects and wave-length shifter effects are corrected by the methods applicable to beta spectrum analyses i.e. SIS, SIE, external standard, and channel ratios. 6. Cerenkov light is highly directional. As a result, light photons generated may be detected by only one PMT n(x) and thus rejected as a count by the coincidence network. Hence counting efficiencies is low. Efficiency can be increased by the addition of a wavelength shifter to the solution 0 and/or by deactivating the Energy (keV) coincidence circuitry. 25

29 Inherent interference in Liquid Scintillation Spectrum Analyses. 2 Chemiluminescence: Production of light – single photons - as a result of chemical reaction between the components of sample – typically the alkaline tissue solubilizers and emulsifier type scintillator. The spectrum overlaps the Tritium Spectrum. The coincidence circuits eliminate these at low count rates. These counts Lc are calculated by the following equation –

Lc = C1 * C2 * 2t

Where Ci is count rate in PMT i (i =1,2) and t is the coincidence window. Approximately 6.7 counts per minute are observed for count rate of 105 and 20ns coincidence window. At ten times this value approximately 667 CPM are observed.

PMT-1 1-t Delay

Coincidence Coincidence Circuit Circuit

PMT-2

Total Background / Events Luminescence Events

Method to detect these events is based on coincidence of delayed (by one t interval) PMT-1 signal and PMT-2 signal. The genuine beta events will be ignored. With the assumption that these events are at least two-t interval wide, this coincidence circuit will detect these. These events are then subtracted from the total counts. Spectrum of these events is also be generated and used for correction of observed spectrum.

Inherent interference in Liquid Scintillation Spectrum Analyses

3. Static Electricity: Static charge is produced due to friction or pressure between the two materials. Discharge leads to emission of Photon imitating the electronic pulses.

Coincidence counting and threshold removes the PMT noise. Surface charge ions are removed by creating an atmosphere of counter charged ions by • Ionization radiation: exposure to alpha emitter before counting the sample. • Electricity: Electronic static controller that removed the static while the sample is being lowered in to the counting position.

Inherent interference in Liquid Scintillation Spectrum Analyses

4. Wall Effect: It is significant for plastic vials. Diffusion of the liquid scintillation cocktail in to the plastic vial walls contributes to the Compton spectrum. It affects the efficiency and quench-correction methods based on external standard.

The energy spectrum is distorted by addition of this effect in the low energy region (up to 40 keV). The correction methods based on the SIE though not significantly affected, can be compensated by calculating the SIE as

Σ u x * n(x) x=L SIE = K u Σ x=L n(x) where Wall Effect on the L is the wall effect cutoff level. Compton Spectrum of

External Source

n(x) Wall Effect Level – (< 40 keV)

Least Square fitted Energy (keV) 1000 curve to six points Counts (regions of spectrum) k CPM

0 80 100 % Efficiency

Inherent interference in Liquid Scintillation Spectrum Analyses.

5. Scintillation Volume Variation: Maximum value of the response of the photo-multiplier tube is at the center of photo- cathode and the response decreases with the distance form the center. Thus the detection efficiency varies with sample

volume and is low at low volumes. The SIS is affected in the same way as for quenching and the correction for volume can be made with reference to a graph between the Efficiency and the sample volume.

PMT Cathode 100%

70% 90%

60% 80%

50%

100

Sample Volume % E Effects on Efficiency

5 50 VOLUME OF SAMPLE (ml)

Low volume will distort the External Standard Compton

Spectrum – low energy gamma rays will contribute more as compared with the high-energy gamma rays (less volume implies less absorption of high energy gamma rays). The inflexion point H# is not affected but the average energy i.e. Spectral Index of External (SIE) source is affected. Alteration is not significant as the effect is only in the low energy region,

however these can be corrected by using transformed spectral index (tSIE).

Inherent interference in Liquid Scintillation Spectrum Analyses 6. Heterogeneous Sample: It is assumed that the radioactive

sample is uniformly distributed in liquid scintillation cocktail while applying the corrections for quenching. In those cases where this assumption is not valid i.e. counting of paper strips source, insoluble form of the source, the efficiency of counting is affected due to absorption in the support material. Counting efficiency for the dissolved substance is more than the

efficiency for the part that is bound to support. Thus the heterogeneous sample counting should be avoided in liquid scintillation counter. The correction is however made by relation between the external standard quench parameter (tSIE) and the spectrum end point (SEP).

Inherent interference in Liquid Scintillation Spectrum Analyses 7. Random Noise: Sporadic background pulses are encountered

in generation of the spectrum. The sources of these pulses are – noise in electronic circuit, line interference (high voltage transients and switching noise) and radio frequency noise (generated by switches, motors and fluorescence tubes). The spectrum has extra peak in low energy region. Digital filtering technique is used to smooth the spectrum and remove this

noise.

Noise + Sample Spectrum

Random Noise Spectrum n(E)

Smoothed Spectrum

0 Energy keV 18.6

Inherent interference in Liquid Scintillation Spectrum Analyses

8. Background: Source of background pulses are

a. Instrument background: due to dark current noise and after pulse noise of PMT. Contribution in well-designed instrument is approximately 10%. b. Cross talk: Spontaneous release of photoelectrons from

cathode of one PMT and seen by the other PMT. It is limited by the coincidence interval of the gate. Can be further reduced by PMT masking and cross talk correction. It contributes 22% of the total background. c. Vial glass and PMT face: These events are produced by cosmic and environment radiation. Presence of 40K in the glass vial and PMT face contributes 37% to the background. These events have uniform distribution over a broad energy range. d. Scintillator Background pulse is again caused by the environment radiation and contributes 31% to the background counts. This background spectrum is stored in the spectrum analyzer and the observed spectrum is corrected.

Background from sources a (10%) & dN b (22%) of total Background

dE Background from sources c (37%) & d (31%) of total Background.

0 Energy 2000 keV

36 The background in LSC Background sources comprise 1) γ-rays from radioactive rocks/materials near detector; 2) High energy electrons/muons in cosmic rays; 3) α -emitters as impurities in sample/vial. From 1), the γ-rays interact by photo-electric effect and Compton scattering to produce electrons. Hence 1) and 2) effectively produce interactions of high-energy electrons (β - particles) with the scintillator. Alternatively, it may be that Cerenkov pulses are produced. These are shorter than the normal β -pulses. Obviously, these sources add to the β background. The only α background is from 3), but compared with non-discriminated samples, the background is extremely small and so the minimum detectable level following α /β discrimination is much less for α than β.

37 Background Event Detection based on Three Dimensional (3-D) Spectrum Analysis & Innovative Method of Gated Integration

Fast Component (2ns)

Volts Delayed Component (up to 900 ns)

900 ns Time (ns) Peaks following the fast Fast Component (2ns) component are seen in the delayed part (up to 5µs) of the background pulses. The Volts number varies with the type of background. Pulse index specifies the pulse number following the fast pulse.

Time (ns) 5µs

3-D Spectrum – Counts vs. Gated Integration Energy for different delay time Interval (Pulse Index). Integration Interval Pulse index is used to (t1 to t2) discriminate Background pulses. PMT Signal Sum

Gated Integrator

Comparison

Output 0: Beta Event Threshold 1: Background 38 Quality Control and Quality Assurance Periodic checks on the Liquid Scintillation Counter are required for proper functioning of the instrument. 1. Tests are carried out at the time of installation and are called “Acceptance Tests.” 2. Tests with standard sources are carried out at regular intervals and are termed as Quality Assurance tests.

Tests required being performed daily (D), once a week (W) and verifying that the results are within the statistical limits set by the manufacturer.

Test Standard Isotope ------14C 3H a. Total Counts in Window D D b. Counting Efficiency D D c. Background Counts D D d. Quench Calibration W W

39 Common Problems and Cause in Liquid Scintillation Counter

Problem / Symptom Possible Cause a. High Background 1. Light leakage in sample chamber- shutter malfunction. 2. High ripple or unstable High Voltage Supply. 3. Counting chamber contamination. 4. Loose connection in cables or printed circuit boards, 5. Coincidence gate malfunction. 6. Improper position of external source while using for calibration. 7. Failure of single channel analyzer to provide differential (window) counting. b. Large Variations in replicate counting 1. All causes mentioned above (in section a. High background) and 2. Improper positioning of the counting vial in sample chamber. 3. Faulty scalar / timer. 4. Faulty trigger logic or malfunctioning oscillator that generates timing clock pulse.

40 Common Problems and Cause in Liquid Scintillation Counter

Problem / symptom Possible Cause c. Poor Efficiency 1. Improper positioning of the vial in sample chamber 2. Drop in high voltage power supply 3. Poor light collection due to bad condition of the reflectors or PMT window. 4. Poor quality of vials. 5. Change in the discriminator threshold in coincidence circuit. 6. Faulty coincidence gate generation. 7. Misalignment of window measurement (slow) channels and associated summing amplifier. e. Erratic movement of vial 1. Vials of improper size 2. Fault in motor control circuit. 3. Malfunctioning of position sensing switch. 4. Improper tension in conveyer belt. 5. Misalignment of sprocket wheel with respect to slot of the rack. 6. Faulty microprocessor interface signals or malfunctioning of interface buffer.

41 Common Problems and Cause in Liquid Scintillation Counter

Problem / symptom Possible Cause f. System not responding to input commands 1. Defective user interface 2. Loose connections in cables, printed circuits boards, etc. g. Unacceptable results of data processing or faulty printouts 1. Faulty central processing unit or memory malfunction. 2. Improper anchoring or connection of signal cables. 3. High electrical noise in the computer system 4. Poor quality mains supply.

42 Chi-square test

Test is performed at the time of installation or after major repair of the Liquid Scintillation Counter. A radioactive sample is counted n times accumulating at least 10,000 counts (for 1% statistical error) in each observation. If Xi i= 1, 2, 3, .. n represent the i'th observation. M the mean of these count values, the chi-square with (n-1) degrees of freedom is given as

λ2 Σ n 2 = i=1 (Xi - M) / M

The significance level is found from the chi-square distribution table given in the standard statistics textbook. The acceptance limits are from 0.1 to 0.9.

Fluor Characteristics Most aromatic solvents are not good fluors and so solutes which are, have to be added. 1. High absorption of energy from solvent molecules and emission of light matching spectral sensitivity of photo cathode;

2. Solubility in solvent; 3. Adequate chemical stability; 4. Short fluorescence time.

The energy transferred in the form of photons is a function of fluor concentration.

45 Quenching Quenching refers to a reduction in the amount of light incident in the photocathodes. This is by chemical colour and optical quench. Chemical Quenching Caused by the presence of materials in cocktail which interfere with transfer of energy from solvent to fluor molecules. Mechanisms include:- 1. Acid quenching resulting from interaction of H+ with primary or secondary fluor; 2. Excessive concentration of one component; 3. Dilution quenching - dilution increases average distance between solvent molecules; 4. Dipole-dipole quenching resulting in non-radioactive loss of energy or increase in vibrational energy; 5. Electron capture preventing energy transfer from e- to solvent molecules. Colour Quenching A coloured material in cocktail absorbs light photons emitted by fluor. In practice, all materials have an absorption spectrum. Reduced by bleaching or decolourizing, e.g. blood samples have been bleached with peroxide, but this increases the chemical quenching. Optical Quenching Absorption of light by condensation, finger prints or residue on counting vial. Quenching reduces the number of light photon striking photocathodes and consequently decreases the size of each voltage pulse. For samples labelled with nuclides other than 3H, this reduction in pulse height causes a shift of the pulse height spectrum downwards.

Influence of chemical quench agent CCl4 on the pulse height spectrum for 14C Because of the small number of light photons released, quenching in a cocktail containing 3H reduces the number of interactions which liberate photoelectrons simultaneously in both photocathodes. Hence, more events are rejected by the coincidence circuit and the height of the pulse height spectrum is suppressed, often without a particularly noticeable shift of the spectrum.

47 The effect of quenching produces: 1. A decrease in the number of pulses produced, i.e. a reduction in counting efficiency; 2. A downward shift in the pulse height spectrum and a reduction in the value of the mean pulse height, H.

QUENCH CORRECTION

Quench correction is a method for the determination of the counting efficiency of a sample and hence, in conjunction with the count rate, the activity

Counting efficiency = count rate/activity when the disintegrations are 100% β -emission. All quench correction techniques use a set of standards, i.e. known activities (dpm) to which different quantities of quenching are introduced artificially to produce a range of quenching to cover that for the samples.