9‘.»
7 5 .
:..!¢r
. 21B.»
5... It.)

.c...
C
a. It
flint... .A...‘ v.
f. .u fir.
_.

,r.

r . ,r. V. 5“
1 it..-

: “.551...
z 529;.

. a . .
I .p 2.! .A
. .15... glut
,5: if. 7,
r7 It£~l|..

1‘
.25....
1‘ It»; . I:

 

27.15.:
hcerax
, \AJ&....5»!P
titxu ..
.. r-

 

. ..

 

>Y’I‘A [flamin-
- .20.. . ('5

E. £25 I... S
. :2? 1...; 1.-
.3 :hiitvvzt t

"fJD’ ‘7

’l"ll‘..x
. a
of ..
-.aFI.¢o .
Jud-Hp“ ‘
t‘ . I. u
. v1 1")

\
:1 I-..A

\.;J

.5.
7.1...: I}. ,

c
4.))..49344.“ .A . .nurwd. . Jan 1.4,

.Ané:

Mufiwfl... a 5.9.5.. L“.
k n . . x .

, $5.: i. f .otl I". , A
9| 4 .vfl ..¢.1; 41s..

 

THESlS

5305019

ll l IllIlllll’lllllllllllllllllllllfllllllllllll

31293 00627 9214
"HWY
”CHM Shh

1|an

 

 

 

 

 

 

 

 

 

 

This is to certify that the

thesis entitled
Ultrasensitive Detector System for a High

Performance Hybrid Mass Spectkometer

presented by
Luis A. Romero

has been accepted towards fulfillment
of the requirements for

Master..— degree in W

453:??? (\/Q>

Major\pj ofessor

Date Nov. ], 1989

 

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

PLACE IN RETURN BOX to remove thlo checkout from your record.
TO AVOID FINES return on or boron dole due.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

= fil J

MSU Is An Affirmative AotioNEqual Opportunity Inuitution

._.._ .7 ..__._._ _ _ _ _ . _______ ,—

 

 

 

 

DLTRLSEISITIVE DETECTOR SYSTEI FOR 1

HIGH PERFORHAICE HYBRID IRSS SPECTROIETER

BY

Luis A. Romero

l THESIS

Submitted to
nichigan State University
in partial fulfillment of the requirements
of the degree of

EASTER OF SCIENCE

Department of Chemistry

1989

c{

i"

U

(0 05 4

ABSTRACT
by

Luis A. Romero

An ion beam (double focusing) mass spectrometer coupled
to a collision cell and followed by a quadrupole mass
filter, is being constructed in the MSU Chemistry
Department. It will be a powerful instrument that will
provide experimental information concerning bond energies
and threshold energies of reactions, among others (between
organic molecules and atoms/ions/complexes). However, in
order to perform such experiments, it is necessary to
implement a detection system capable of measuring low
fluxes of ions. The subject of this thesis is the design
and construction of the detection system for the ion beam

instrument.

The photon counting detection was chosen as the optimum
method for this application. The different criteria
involved in its selection are discussed. Comparison of
different types of detection schemes are also included.
In addition, models of different types of photon counting
systems are examined using a simulation program (SIMION
4.0) which allows the relative efficiency to be determined.
Design criteria and operational procedures for the

detection system are described. Discussion of the

different problems encountered during the design and
construction process is included, as are the different
approaches taken to solve such problems. The current

status of the working detector is documented and further
work is recommended in order to complete the

characterization process of the overall system.

To my family and my wife

ACKNOWLEDGMENTS

To those who have contributed in my personal growth and have been

there in every time when their moral support was most needed.

I would also like to offer my must sincerely thanks to the following
people, all of whom have made direct contributions to the development of

the detector system (the subject of this thesis).

George Leroi and John Allison for their guidance, support,
encouragement, advice and patience (among others) during all this time of
my graduate student life. Also Paul O'Connors by all the help and

information that he shared with me during all this work.

Marty Rabb, MSU Chemistry Electronic Designer for designing the

pulsating circuit used as well as for all his advice in some of the

electronic problems encountered during this work.

Russ Geyer, Dick Henke and Dick Vatters, the staff of the MSU
Chemistry Department Machine shop for their carefully machining of the

vacuum chamber, enhancer dynode and other parts of the detection system.

Reg Renninger and Jehn Yurkon from the Cyclotron Facility which
provide us with the scintillator material and coated with the aluminum
oxide and also provide us with enough technical information to start

iii

developing the detector.

Finally, I would like to acknowledge the financial support provided

by Michigan State University during all these years as a graduate student.

iv

TABLE OF CONTENTS

List of Tables .........................................
List of Figures ........................................
CHAPTER 1. INTRODUCTION...... .........................
CHAPTER 2. DETECTION SYSTEM DESIGN..... ........... ....
2.1. Detection System Selection Criteria ........
2.2. Electron Multiplier ........................
2.3. Scintillator Detector System ...............

2.4.

2.5.

CHAPTER 3.

3.1.

3.2.

3.3

l. Scintillator Material Characteristics and
Selection Criteria ......... . ............
2. Photomultiplier Selection Criteria ......
Design.... .................................
1. Model Systems ...........................
2. Constructed System ......................
Pulse Counter ..... . ........................

1. Design Considerations ...................

EXPERIMENTAL APPARATUS AND OPERATIONAL

PROCEDURE .................................
Detector Chamber .............. . ............
Mass Spectrometer and Ion Optics ......... ..
Detection System ..... ... ...................

1. Photonultiplier Electronic Circuit......
2. Detector-Amplifier Interface ............

V

page
vii

viii

10
ll
11
12
20
21

23

24
24
26
28
32

37

3. Amplifier ...............................

4. Pulse Height Discriminator and Counter..

CHAPTER 4. RESULTS AND DISCUSSION ......................
4.1. Characterization of Photomultiplier

Electronic Circuit .........................

1. Anode Response Time .....................

2. Signal-to-Noise Ratio .............. ' .....

4.2. Amplifier-Discriminator Signal Coupling....

4.3. Mass Spectrometer Test System ..............

CHAPTER 5. SUMMARY AND CURRENT STATUS OF THE

DETECTION SYSTEM ..........................
5.1. Summary ....................................
5.2. Current Status of Detector System ..........

REFERENCES .............................................

vi

39
4O

42

42

42

47

50

52

60

60

61

63

LIST OF TABLES

table page

1.1. Ion Detector Characteristics .............. 4
4.1 Characterization of Photomultiplier

Signal Count Rate ......................... 49
4.2 Count Rate as a Function of

Discriminator Voltage ..................... 51
4.3 Photomultiplier Count Rates Using

the Mass Spectrometer ..................... 58

vii

LIST OF FIGURES

page

figure

1.1. Proposed HSU Ion Beam Instrument ............. 2
2.1. Schematic Diagram of Scintillator

detector with energy diagram .................. 9
2.3. Schematic Diagram of the Strausser

Scintillator-type Ion Detector ................ 14
2.4. Simion Representation of Strausser

Detection System .............................. 15
2.5. Schematic Representation of McCormick

Daly and Powell Scintillator Detector ......... 16
2.6. Schematic Representation of Daly

Scintillator Detector ....................... .. 17
2.7. Simion Representation of Designed

Detection System .............................. 19
3.1. Schematic Representation of Detection

System Chamber.. .............................. 25
3.2. Schematic Representation of: ................. 27

a) Lens Controller Voltage Divider

b) 30 xv Power Supply Regulator Control System
3.3. Schematic Representation of the

Detector System.......... ..................... 29
3.4. Schematic Representation of Photomultiplier

Tube Housing with Scintillator .......... . ..... 31

3.5. Schematic Representation of Photomultiplier

viii

3.6.

4.1.

4.2.

4.3.

4.4.

4.5.

4.6.

4.7.

Electronic Circuit ............................ 34
Schematic Diagram of: .......... . ........ ...... 38
a) LED Pulsed Circuit

b) Test System arrangement

Anode Response to LED Pulsating Circuit ....... 43

Effect of the Length of the

Transmission Cable ............................ 45
Effect of the Value of the Terminator

Resistor on the Anode Output Signal ........... 46
Beam Visualization of Test Background

Signal in the Vacuum Chamber .................. 53
Mass Spectrum of the Background

Gases in the Vacuum Chamber ................... 54

Frequency Dependence Signal from PMT

Electronic Circuit without Low Pass Filter.... 55
Anode Output Signal of Background Gas ......... 57
a) without Discriminator

b) with Discriminator

ix

CHAPTER I

INTRODUCTION
A new ion beam tandem mass spectrometer is being developed
through collaboration between the Allison and Leroi research groups in
the MSU Chemistry Department. The instrument will consist of: (1)
single beam Varian MAT CH-S DE double focussing mass spectrometer, (2)
deceleration lens system, (3) radio frequency-only octopole ion beam
guide/ collision cell, (4) quadrupole mass filter and (5) ion flux
detection system. Figure 1.1 shows a schematic diagram of the

instrument.

The goal in the development of this instrument is to provide
insights into the nature of reactions between organic molecules and
metal atoms\ions\complexes. It will also allow experimental
determinations of bond energies and energetic reaction barriers and
provide information about ionic structures. The apparatus is designed
to allow accurate measurements of ion energies in an experimental
laboratory energy range from 0.05 to 500 eV. Therefore, it requires
a detection system, the topic of this thesis, capable of measuring low

ion fluxes at these threshold energies.

In general, mass spectrometers require sensitive, fast-response ion
detectors to measure very small currents or to record very rapidly

scanned mass spectra. This can be accomplished by the use of a direct

   

Magnetic

 

 

 

 

 

 

 

 

 

 

 

 

Sector
Electric
Sector
- Deceleration
r Lens
‘L—(‘ Octopole
Ion Beam Guide
Source
Gas [:
Cell
Quadrupole II 'I
Detector

Figure 1.1. Proposed ISU Ion Beam Instrument

or an indirect method of detection‘1v2). Unfortunately when the

direct method of detection, (e.g., Faraday cup) is employed to record
relatively rapidly scanned mass spectra (less than 0.1 seconds for a
mass range of 50 to 500), its sensitivity is usually low (10 '1‘ A).‘1’
Vith the indirect method (e.g., scintillation, detection), higher
sensitivity can be achieved (10 “2°A),‘2’ even for relative rapidly

scanned mass spectra (less than 0.1 s for a mass range of 50 to 500).

In order for the detection system to be useful for the chemical
experiments of interest in this project, it must meet certain
requirements. These requirements include: (1) high sensitivity (10'
2° A), in order to measure ion intensities at threshold energies of
reactions, (2) fast response time (nanosecond range), (3) good signal-
to-noise ratio, (4) short and long term linearity, for quantitative
work, (5) long and short term stability, (6) long serviceable life
(little deterioration over long periods of use), and (7) simplicity and
cost-effectiveness. The detection method that can best meet these

requirements is the indirect method of detection.

The first task undertaken in this research project was the design
and construction of the detection system. The design process included
a comparison between different indirect methods of detection such as
electron multipliers (discrete and continuous dynode versions), and

scintillator detectors. Table 1.1 presents a summary of their main

Table 1.1

Ion Detector Characteristics

 

 

Type Maximum Nain Use Advantage Disadvantages
Sensitivity
(A)
Faraday Isotopic Reliable Poor sensitivity
cage abundance,
10-17(0)
Quantitative
analysis,
Research
Electron
Multiplier Research, High sensitivity, Short life,
10-20
Isotopic Ion counting Cost
abundance
Scintillator
Detectors Research, High sensitivity,
10-20
Isotopic Ion counting, High voltage
abundance Long life

 

(a) Uith vibrating reed electrometer; less with other types of anneters

characteristics.

Computerized modeling of several scintillator detection systems has
been done with an ion trajectory simulation program, SINION 4.0.‘3’
This is an interactive program developed for microcomputers (e.g., IBM
AT/XT and Apple NACINTOSH) which allows the simulation of ion

trajectories through ion optical components.

The selection as well as design considerations of the detector
system will be the subject of chapter 2. Chapter 3 deals with
operational procedures for the detector. Experimental results and
discussion of them will be presented in chapter 4. A description of

the current status of the system is given in chapter 5.

CHAPTER 2
Detection System Design

2.1 Detection System Selection Criteria

The ion beam tandem mass spectrometer under construction is
designed to allow accurate measurements at threshold energies of
reaction. Under such conditions the number of ions/sec that will be
detected are in the range of 103 to 101 or fewer. This corresponds to
ion currents of 10'15 to 10“8 amperes. Therefore, the most important
criterion for the detection system is its sensitivity over that range.
In addition, the detector should provide: (a) S/N ratios that allow the
measurement of the above currents to a precision of at least 1 10%, (b)
fast response time (nanosecond range), and (c) short and long term
stability. The two most common methods of detection for this range are

ones which utilize electron multipliers or scintillators.

2.2 Electron Nultipliers

Electron multipliers use the principle of secondary-electron
emission amplification.(1) They are classified as either discrete or
continuous dynode electron multipliers. The difference between them is
the way in which the secondary-electron emission surface is constructed

(discrete or continuous).

Electron multipliers can amplify ion currents by factors up to 10° .

depending on which type of multiplier is used and on its condition.‘4>

The amplification factors obtainable using electron multipliers give
high sensitivity to the current-measuring system. ‘These detectors are
also characterized by fast response time (nanosecond range for
continuous and some discrete dynodes)<5’ and good S/N ratio (1 10% for
ion currents of 10'19A).(“ These properties allow the use of electron

multipliers in analog as well as in pulse detection modes.

Unfortunately, electron multipliers can offer only short term
stability. Degradation in performance is observed as a function of
time of use. The initial gain decreases rapidly (from values of 106
or higher to 105) and then more slowly, progressively to lower
values.‘4’ This can be attributed to: (1) ionization of residual gases
which are normally present between dynodes and (2) the catalytic action
of secondary electrons which polymerizes hydrocarbons on a dynode
surface, thus lowering’ secondary electron jyields. The rate of
deterioration, therefore, depends strongly on the space current of
electrons flowing through an electron multiplier and on residual gas

pressure.“’

2.3 Scintillator Detector Systems

Scintillator detector devices, as do electron multipliers, use the
principle of secondary-electron emission amplification. In these
devices a form of ionizing radiation (i.e., a-particle, S-particle,
electron, etc.) interacts with scintillator material which produces an

excited electron and a low energy photon. The initially excited

electron dissipates its kinetic energy by exciting other electrons from
the valence band in the scintillator material into the conduction band.
When these excited electrons return to the valence hand, some of them
generate scintillation photons, which are subsequently detected by a
photomultiplier tube (PNT).(7’ Figure 2.1 shows a schematic diagram

of the process.

Scintillator detectors can provide current amplification in the
range of 108 to 109, depending on the scintillator material,
photomultiplier and electronic circuitry involved. The response time
of a scintillator system is in the nanosecond time range (depending on
scintillator material and photomultiplier used). Signal-to-noise
ratios improves by more than 10% for ion currents of 10’19 to 10'2°A

can be obtained.(°’

Photomultipliers in scintillator detector systems suffer long term
instability due to the same factors that affect electron multipliers:
(1) ionization of residual gas present between the dynodes, and (2)
polymerization of hydrocarbons on the dynode surfaces (present in all
vacuum systems). Fortunately, in scintillator systems these factors
can be minimized by placing the photomultiplier outside of the vacuum
system.‘°’ Therefore, scintillator detector systems best meet the

basic requirements for low particle flux measurements.

 

 

 

(:
B
{A
cownucrrom saws
Pnomocnrnons ”
(
(
0
TIPLIBR
PHOTOHUL F ' { ,+___E

 

 

 

 

 

 

 

 

VTULETNIB
BAA“) . c)

Figure 2.1. Schematic Diagram of a Scintillator
detector with energy diagram (adapted

from reference 7.) .

1: optical coupling B: reflector (11:0: powder)
C: protective cover (Al) D: 30 RV electrons

E: excited electron from scintillator

F: decay of scintillator electron with emission of

light 6: holes

10

2.3.1 Scintillator Eaterial Characteristics and Selection Criteria

In general, a scintillator material can be defined as a gas, a
liquid or a solid in which flashes of light can occur. The two most
commonly used media are liquids and solids. Liquid scintillators are
used in systems in which the ionizing radiation to be detected is very
low in energy and the sample to be ionized can be dissolved in the
scintillator medium (e.g. radioactive materials).‘1°) The solid type
is used with samples that cannot be dissolved in the liquid
scintillator material and can be used with high as well as low ionizing
radiation energy systems. In either case the number of photons
produced is essentially proportional to the energy of the incident
radiation. The maximum energy that can be transfered to the photo-
electron is given by:

E max = E/[l 4' 1/(20)] (2.1)

where E = incident energy, a = E/(DOCZ), mo = mass of the electron at

rest and c = speed of light.‘1°)

In order for a scintillator material be useful it must meet certain
requirements ‘7'“): (1) optimized scintillation efficiency
(measurement of its ability to convert absorbed radiation to detectable
light), (2) wide emission spectrum with a maximum.near the maximum of
the spectral response of photomultiplier, (3) fast decay time, and (4)
high collection efficiency. High collection efficiencies can be

achieved by: (1) appropriate selection of the shape and size of the

11

scintillator to match the photocathode of the photomultiplier, (2)
coating the surfaces of the scintillator (except the one facing the
PET) with a reflective material for the wavelength of the photons
emitted, (3) appropriate match in refractive indices between
scintillator, PNT and the medium between them to avoid internal
reflections, and (4) appropriate thickness and diameter selection of

the scintillator.

2.3.2 Photomultiplier Selection Criteria

The photomultiplier tube used in a scintillator detection system
must have certain characteristics: (1) photocathode diameter which
matches scintillator material diameter, (2) emission from the
photocathode at a given wavelength (no variation in the quantum
efficiency as a function of the position of the photocathode) . (3) high
photocathode'sensitivity to blue and near-ultraviolet wavelengths, (4)
good stability (no or little shift in gain with long term use), (5) low

dark current (1 to 2 nAmp), and (6) fast response time (1 to 3 nsec).

(11.12)

2.4 Design

The design of the scintillator system can be divided into two
sections: (1) simulation of the ion optics transmission system by the
implementation of SINION 4.0 (3’ and (2) physical space constraint

considerations.

12

2.4.1 Model Systems

Modeling of various scintillator systems has been done by the use
of the microcomputer program SINION 4.0. This program simulates ion
trajectories through an electrostatic field using an interactive over-
relaxation method. The over-relaxation method is used to calculate the
electrical potential at up to 16,000 (grid) points in the electrostatic
field. This interactive program allows the user to design the
geometric arrangement of electrodes, as well as to introduce: (1)
voltages for the electrodes, (2) initial ion masses and velocity
directions and (3) relative positions of the ions in the field.

The scintillation-type detector systems simulated with SIMION 4.0
are the ones presented by: (1) Strausser (1970),"3’ (2) Daly,
McCormick and Powell (1968),‘2’ and (3) Daly (1960).<9’ These systems
are designed to respond only to ions whose energies lie within a pre-
selected range. The basic principle involved is that the scintillator,
which is coated with a thin film of metal, responds to high energy
electrons that can penetrate the coating. A single ion striking the
target, an enhancer conversion dynode, causes the release of several
secondary electrons which in turn will penetrate the metal coating of
the scintillator and produce photons which are measured by a
photomultiplier tube situated outside of the vacuum system. The number
of electrons emitted per ion will depend not only on the voltage
applied to the dynode but also on the type of material used in its

construction. For example, if an ion strikes the surface of an enhancer

13

dynode made of aluminum which is at -30 keV, the number of electrons
emitted by the dynode will be approximately 7 electrons per ion,
whereas if the dynode is made of stainless steel only 5 electrons per

ion are emitted“".

The first modeled system (Strausser type) included an evacuable
envelope containing a secondary emitter target electrode structure
having a bore of conical shape (enhancer dynode). The inside surface
of the bore was bombarded by the ions to be detected; the secondary
electrons emitted by the dynode were focused onto a scintillator facing
the exit end of the bore to give the optical photon output which was
transmitted through the window to a photomultiplier.‘13’ Figure 2.2
shows a schematic diagram of the system and Figure 2.3 presents a

SIMION ion trajectory calculation of it.

The main advantages of the in-line arrangement of the Strausser
system are the simplicity of the design as well as the low range of
voltages applied to the scintillator (0.1 to 10 keV). Unfortunately,
the simulation program (SIMION) has shown that over 95% of the ion
trajectories will miss the enhancer dynode (see Figure 2.3). Several
modifications were made to the system during the modeling process in
order to improve the transmission( of ions, (i.e., focusing and
deflection lenses were added, the scintillator and PET were floated),

but less than a 10% improvement was accomplished.

14

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 2.2. Schematic Diagram of the Strausser
Scintillator-type Ion Detector (from

reference 13.)

A: tubular body

B: high voltage lead and insulator
C: secondary emitter electrode

D: scintillator phosphor layer
E:transparent optical window

F: photomultiplier tube

15

    
  

IIIIIIIIIIIIIIIIIIIIIIIHIUII]

 

 

 

 

 

 

 

    

IllIlIIIIIIIIIIIIIIIIII

 

 

I
)A

o -1..."
(3

Figure 2.3. Simion Representation of Strausser Detection System.
Ion Energy is 50 eV and its lass is 100 amu.
A: quadrupole = 3000V B: top deflection lens = 2000V
C: bottom deflection lens 8 0 V
D: housing - 0 V E: enhancer dynode - -30 kV
F: photomultiplier 0: ion trajectories

B: countour lines

16

 

EV-r- '--C3

(__________> E
T _— A- -
c.

{-11

 

 

 

 

 

Figure 2.4. Schematic Representation of lcCormick,
Daly and Power Scintillator detector

(from reference 2.)

A: secondary electron emitter (enhancer dynode)

B: aluminum coating C: scintillator
D: glass window E: photomultiplier
F: ions trajectory G: secondary'electrons

I: light detected by photomultiplier

17

 

Figure 2.5.

 

 

 

 

 

 

 

 

Q-m

 

 

 

 

P1

Schematic Representation of Daly

Scintillation detector ( from

reference 9.)

pumping holes

ion beam entrance slit

anti corona ball

glass metal Rover seal

secondary electron beam

plastic scintillator

B: positive ion beam

D: stainless steel rod

F: -40 ReV connection

I: stainless steel with
A1 coating

J: Al coating: reflects

light,¢:uHMct ahmnxuns

L: photomltiplier tube

18

The McCormick-Power-Daly system has a similar arrangement to the
Strausser, except that the enhancer dynode is a flat surface and has
a scintillator electrode (see Figure 2.4). The function of the
scintillator electrode is to repel the ions into» the enhancer,
increasing the number of ions reaching the enhancer. One of the
limitations of the system is that if an ion has a greater energy than
the scintillator electrode it will hit the scintillator and not the
enhancer dynode, thus decreasing the detection efficiency of the

system.

The last system under consideration is the Daly scintillator
system. In this system the enhancer dynode as well as the scintillator
and PET are at 90° to the ion trajectory (Figure 2.5).The enhancer
dynode consists of aluminum, vacuum-deposited as a thin layer on a
stainless steel base ground optically flat and highly polished. This
increases the secondary electron yield”) and also decreases the
probability of field emission. The voltage applied to the enhancer
dynode (-40 keV) assures that all of the ions entering the detector
region strike the enhancer dynode, thus increasing the probability of
secondary electron emission and decreasing the probability of ions
reaching the scintillator. The only drawback of this system is the
number of safety precautions that must be used due to the high voltage
present. (For further discussion the reader is referred to references

2,6,9 and 13.)

19

mo

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

111111111111111

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘ o
O
”—f
K

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. vvvwv va v
III.ICIOIIICIOICIIIIIIOIOICU...ICIIIII-......I...-III-III....-II.....II-IIIIIICIIDICIUCI

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i

~

~

m

H

H

N

" lH-H

N Ii

P‘ III-I

H .0.

"‘ ml-

: ~-

: :C
m

H ..C

H nu

"' mu

" mu

" mu

" un-
l

H I:
mH
IN
IH
IN
IN

I um

um

 

 

 

O-uu
m-
m

Figure 2.6. Simion Representation of Designed Detection System.

Ions Energy is 50 eV and its lass is 100 amu.

A: quadrupole - 3000 V B: extractor lens = -2000V
C: top deflection lens 8 2000 V

D: bottom deflection lens

I: housing 8 0 V F: enhancer dynode I -30 kV
6: photomultiplier B: ion trajectories

I: countour lines

20

2.4.2 Constructed System

The scintillator ion-type detector constructed was based on the Daly
model since this system will provide the highest secondary electron
emission. Several modifications were made to the initial model in
order to improve transfer and transmission of the ions. Figure 2.6
presents a simulation of the trajectories of ions in this system
obtained using SIMION 4.0. The system consists of: (1) transfer optics
section, (2) enhancer conversion dynode, (3) scintillator and (4)
photomultiplier tube.

The transfer optics section consists of: (a) an extraction lens
which will be held at an electrical potential between -2000 to 100 eV
to ensure the complete extraction of the ions from the quadrupole
(Figure 1.1), and (2) a deflection lens to focus the ion beam to a
specific area of the enhancer dynode at angle between 30° and 60°
(thereby increasing secondary emission (1"). The enhancer dynode
consists of an aluminum base, ground Optically flat and highly polished
on which aluminum is vacuum-deposited as a thin layer (to increase
secondary electron emission), coupled to a -30 keV power supply (GAMA
RC 10-30?) which has an output current of 0.33 mA via a 30 kV feed-

thru (EFT 281 2093).

The scintillator material used (BC 408) has a decay time of 2.1
nanoseconds, and its wavelength of maximum emission is 425 nm. The
light output of this material is 65% with respect to anthracene, the

conventional actinometric standard. For an incident electron with an

21

energy of 29 keV the scintillator will emit 8 photons per incident
electron.”5’ The scintillator is covered with a reflective thin layer
of vacuum-deposited aluminum (5000 A), which helps increase the
collection efficiency of photons by the photomultiplier, which reduces
damage to the scintillator from air moisture, and which provides a
grounding contact surface. The scintillator) is coupled to the

photomultiplier tube by a thin coating film of clear epoxy (Lucite).

The photomultiplier used (R329-02 Ramamatsu) has 12 stages, a gain
of 1.1 x 106, and an anode dark current of 0.7 namp at 1500 V. It has
a total spread time of z 7nsec: thus, its upper frequency response
limit for resolved pulse operation is z 140 MHz. The average charge
packet for this PNT, which arrives at the anode in about a 7 nsec time

interval, has an average current of 2.5 x 10-5 A.

2.5 Pulse Counter

The inherent advantages of photon counting such as: (1) direct
digital processing, (2) discrimination against dark current not
originated at the photocathode, (3) elimination of reading error and
(4) system stability against drift, have made the technique very useful
in many quantitative applications.‘1"”’ This technique is very
attractive for systems in which the radiation level is so low in
intensity (where the signal-to-noise ratio approaches unity ‘1“) that
it is difficult to obtain reliable measurements by conventional

techniques (e.g., dc current, lock-in amplifier and noise

22

voltage).‘19) The limit of detection for this technique depends on
thersystem under consideration as well.as on the required precision and

accuracy.(2°’

In photon counting systems the current pulses from the photo-
multiplier tube are usually passed across a load resistor, and the
resulting voltage pulses areeamplified by a fast pulse amplifier before
entering a discriminator and counter. In some systems in which a high
gain PET is used the amplifier can be eliminated, and the load resistor
can be connected directly to the discriminator. The final readout of
the system will be affected by the counting circuitry components (RC
load, amplifier and counter). The width of the pulses from the PMT or
the voltage pulses in the counting circuitry results in a finite
probability of pulse overlap. ‘21'32) For photomultiplier tubes with
upper frequency responses between 70 and 200 MHz the probability of
pulse overlap is assumed to be negligible.(‘9'23’ In cases where the
counting circuitry’limits the frequency'response (time interval between
pulses or dead time, I) this may be attributed to several factors: (1)
slow RC decay time with respect to its rise time, which increases the
width of the peak seen by the discriminator, (2) improper discriminator
reference voltage level, (3) improper selection of the counter
speed.(¢1’ Therefore, the selection of these components is critical in

the performance of a photon counter system.

23

2.5.1 Design Considerations

In order to properly implement the photon counting technique several
requirements must be met by the system. These requirements are: (1)
high sensitivity, (2) variable sample rate control, (3) high counting
stability, and (4) high count resolution. All these requirements are

basically interrelated as will be seen below.

Photon counting circuits should provide a high sensitivity in order
tt» measure signal-to-noise ratios less than unity. This can be
accomplished by the appropriate selection of the: RC circuit,(21'22’
discriminator level‘2" and counter sensitivity (chosen to miss only
1 to 2% at an average count rate below 2 NHz).‘23-2‘) The counter
should also have a variable sample rate controller. Since the arrival
rate of electrons is a random process (some of them having arrival time
less than 10 nsec), by optimally adjusting the sampling rate of the
counter a greater number of pulses can be detected.‘“'“) In
addition, the photon counting should have a high count stability in
order to provide precise long time integration (greater than 10 sec),
thereby reducing the noise bandwidth and allowing differentiation of

individual pulses.

24

CHAPTER 3
EXPERIHENTAL APPARATUS AND OPERATIONAL PROCEDURE
The objective of this chapter is threefold: to present a
description of the detection system constructed for the MSU beam
instrument, to document the experimental conditions under which the
detection system is operated, and to discuses the experimental problems
encountered during this work, many of which were due to the author's
lack of experience. The following discussion will help future users

solve some of the problems that could occur again.

3.1 Detector Chamber

The detection vacuum chamber, shown schematically in Figure 3.1,
consists of an aluminum box with outside dimensions 21.5 inches long
x 11.5 inches wide x 9 inches high and 0.75 inch thick walls. The
vacuum system coupled to the chamber consists of: (l) a mechanical
pump, Helch Scientific Co. model R-1376, (2) a Varian 2 inch diffusion
pump model H52, which has a pumping speed of 285 l/sec, and (3) a
Varian cryotrap model NRC 325, filled with methanol as refrigerant
liquid. The function of the cryotrap is to minimize the amount of oil
reaching the chamber from the diffusion pump. The speed of the
diffusion pump is reduced to 125 l/sec due to the presence of the
cryotrap. An extension flange 7 inches long x 5 inches diameter is
coupled to the chamber in order to fit the quadrupole mass filter

contained in the main chamber with an ion source and a leak valve

25

.uemmumu Iceman mowuoeuem «o mowaeumomemmom owuumeaoa .«.n ousuwh

ll .22
223:8

..l. .3528

 

_ am: 3
T g

T 325:5 mg

 

 

 

 

5:28 .533

 

 

 

a
as $50

26

located at the entrance of the ion source.

The pressure inside the chamber is monitored with an ion gauge tube
series 260 and an ion gauge controller model 274005 from Granville-
Phillips: the transducer is located in an extension tube at the

opposite end of the chamber (with respect to the ion source).

3.2 Hass Spectrometer and Ion Optics

The reactant gas is introduced directly into the ion source through
a leak valve. The ion source is powered by a Harrison Lab. Inc., power
supply (model 868A). The ions formed are guided from the ion source
and focused into the quadrupole mass filter with four electrostatic
aperture lenses. These lenses were supplied with the Extranuclear
quadrupole mass filter (model 324-9). The dimensions of the mass
filter rods are 1.9 cm diameter x 22 cm long. The lens system is
powered by an Acopian t 120 V power supply model 120J05D, while the
mass filter is powered by Extranuclear power supply model 311, equipped
with a high-Q head model E (mass range 0-200 amu). The ion source,
lens system and quadrupole are mounted on an optical rail and are
easily moved several inches in a direction parallel to the quadrupole
axis. This allows the user to make any modifications in the system
more easily (e.g., to move system components or to isolate different

components).

27

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

VOLI- 1 R— ‘t R — (.1 4 (—
) p , .—
m - 1 -1-Z_/( I'3 AN“ ”('5 /('5 21201»:
1 (
sum e-fi’ms on new r, Li, Ll,
R1: 129 KO
(a)
' 39 K9
POUCH SUPPLV
1
§ *i--O’5 2 o—w :9
mo 13" 3°- SK
3" f 4 °_ 10 KWU-W
(ION-91

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(b0

Figure 3.2. Schematic Representation of: (a) Lens
Controller Voltage Divider, (b) 30 AV Power

Supply Regulator Control System.

28

The voltages applied to the lens system are controlled by a home-
made voltage divider (Figure 3.2a). The m/z values of the transmitted
ions from the quadrupole is controlled by a dial on the front panel
of the quadrupole power supply or by a frequency applied to a connector
in the back of the quadrupole power supply. The detailed procedures

are listed in the Extranuclear quadrupole mass filter manual.

3.3 Detection System

The detection system consists of: (1) two electrostatic aperture
lenses (extractor and deflection lenses), (2) a. housing, (3) an
enhancer (or conversion) dynode, (4) a scintillator, (5) a
photomultiplier, and (6) photon counting electronic circuitry' (Figure

3.3).

The extraction and focusing lens is 2 15/16 inches in diameter x
0.017 inches thick and has an aperture of 1 inch. The objective of
this lens is to extract the ions from the quadrupole and to focus them
into the center of the deflection lens. The deflection lens is 2
15/16 inches in diameter x 0.035 inches thick: it has an aperture of
l in., and an extension tube 1 cm long which couples the lens to the
detector housing entrance. This lens is divided into two sections in
order that a different voltage can be applied to each section and
thereby provide more control for the deflection of the ion beam. The

separation between the sections is 1/8 of an inch. In order to test

29

82.8 :2

b

 

mots—Eu mg... .

 

 

 

 

.IJ-

 

LE

1

 

J

552.. mg“:
Sc ”85%

.meuemm .3330: 05 «o consume-mums: umuemeaom

 

 

 

 

 

 

 

  

 

secs gas
ma°_su=zes smaism
5,2 masses
2 L.
A 29 spasms
J) 3:28
E353... 5.58
30523 as: -2?
.2: 8 W-
Ea). 4.. “wheels
ages .2

4
m5.— 225359

:2: a“: 2. I.

x8 ”:53 I

 

30

the system, one section is kept grounded while different voltages are
applied to the other section via a high voltage feedthru coupled to a
flange located on the right side of the chamber.

The deflection lens ends at the entrance of the detector system
housing. The entrance aperture for the detector system is 1 1/8 inches
in diameter. A housing was designed to serve two purposes: (1) to
constrain the high voltage applied to the conversion dynode into a
specific area, and (2) to minimize the amount of light not originating
from the plastic scintillator that can reach the photocathode. Located
inside the housing is the conversion dynode, the scintillator and the
photomultiplier tube. The housing is made of a brass tube with 3 1/4
inch O.D.and 1/16 inch wall thickness. It is placed 3/16 inch from the

chamber wall.

The conversion dynode is made of aluminum and is coated with 10‘
A of aluminum oxide. This material was selected based on its weight
(a light metal is needed in order to be supported firmly by the high
voltage feethru connector) and its secondary electron emission
properties at the voltages of interest (Al has a ratio of secondary to
primary current of approximately 7).(1" The size of the conversion
dynode is 1 cm diameter x 1 cm long and is shaped like a door knob.
It is connected to an external GAMMA -30 Rev power supply via a high
voltage feethru. The high voltage power supply is regulated by a home-

made controller system (Figure 3.2b). The control system allows the

31

.uouezwuewou A»? 2:32. 04:... 3333332:— uo sown-unease.— swan-emu»

mam—5: 5325:2035

 

 

 

 

 

 

 

ewes euca¢o¢
“34553381.. 32:. .o
v A s 4
so m
3.33
n .5553 ......
a we: fizz—.558: 1.)) 35::sz
2. .
.H .II

 

 

 

 

 

 

 

 

 

 

 

_
2.2% .55.. 32533.. $25.... 258: Ea

32

operator to monitor, via a Fluke multimeter, the voltage, applied to
the conversion dynode (the reading of the meter should be of 1 V for
every 10 kV applied to the conversion dynode). The organic plastic
scintillator and the photomultiplier tube are located across from the

enhancer dynode inside of a stainless steel housing.

The photomultiplier tube is mechanically coupled to the organic
plastic scintillator by the pressure that a set of springs in the PET
base exerts on the photomultiplier (Figure 3.4). The use of this
configuration offers certain advantages: (1) it allows the system to
be tested without the need of a coupling medium (epoxy or silicon oil
used to avoid the reflection of light due to differences in refractive
index .between the photomultiplier and the scintillator); (2) it
provides isolation from external rf fields; (3) the photomultiplier
tube will have a longer life (since it is not under vacuum): (4) facile
changes of photomultiplier tubes without perturbing the chamber

experimental conditions can be made.

3.3.1 Photomultiplier Electronic Circuit

There are two ways in which the photomultiplier tube can be
arranged for pulse counting: (l) the photocathode at a high negative
voltage and the anode at ground, or (2) the photocathode at ground and
the anode at high positive voltage. Both arrangements have their

advantages and limitations. The advantage of having the photocathode

33

at a high negative voltage and the anode at ground is that it is
easier to process the signal, since no coupling capacitor is needed.
The limitation of this arrangement is the presence of high levels of
dark noise that can affect low signal currents. The second circuit
arrangement (anode high potential at +HV) offers the advantages of low
dark:current levels from the PMT and.low steady-state DC levels induced
by ambient light levels.(25’ It is also more easily coupled to a
plastic scintillator that is at ground potential. Its limitation is
that the signal is difficult to process since a coupling capacitor is
required in order to separate the signal from the applied voltage.‘12’

The coupling capacitor should have good stability.

Based on the advantages previously mentioned, the photomultiplier
configuration selected for the detector system was that with the anode
at a high potential, +HV. The specifications of the various components
(resistors, capacitors etc.) were selected based on the ratios
suggested by Hamamatsu.‘“’ The value of the gain resistor, R6
(between the high voltage source and voltage divider) was chosen based
on two factors: (1) the stability of the photomultiplier tube (for good
phototube stability, the voltage divider current should be equal to or
greater than 50 times the average value of the phototube current), and
(2) the fact that the value of Ra should be less than or equal to one

half the voltage divider resistance (Rn).

-(

34

(ERG R
(I mMJIIm is. in in L 51.4 L

 

 

#7

Figure 3.5.

 

 

 

 

 

 

Schematic Diagram of the Photomultiplier
Electronic Circuit.

I:=2.2 ID, la=la=Rs=°°'=Rs=470 KO.

Realie' 750 ID, Rnitlis= 920 an, R1s=l.5 ID,
Rt=2 10, 36-470 10, R136 ED, C1= 0.02 nF,
C28 0.047 pr, Cs= 0.1 hr, C48 0.22 pF,

Cs= 0.47 pF, Cc= 0.001 pF, Cc: 0.005 pF.

35

For the ratios for the resistors suggested by Hamanatsu‘37’ the values
for the individual components in the voltage divider were determined
(see) Figure 3.5). The value obtained for the voltage divider
resistance was 8.49 x 106 ohms and the maximum value for the gain
resistor R; was 4.3 ND. The exact value of the gain resistance can be
determined based on the desired voltage drop aeross it. For

most phototubes a voltage drop of 100 V across Rc changes the
gainapproximately by a factor of 2; this change in gain is sufficient
to maintain a good linear operating voltage range for the phototube.
Since the gain of the photomultiplier is one of the criteria used in
the selection of the photon counting systems, the value of Ra was
chosen to keep the change in gain with the change in the PMT voltage
as small as possible. Therefore, the value of the gain resistor was
selected such that the change in gain obtained was less than 2 (Rc=470

k0 gives a voltage drop of 79.0 volts.)
The value of the phototube resistance is given by:
anr= E(Applied)/I(dark current) (3.1)
For the R-329 phototube the dark current is 0.8 x 10-9 amp, which will
yield a value of Rpur = 1.88 x 1012 D at an applied voltage of 1500V.

The value of RPMT >) Rn, which is one of the criteria required for

good stability of the electronic circuit. The stability gain is

36

defined as:

S.G.= ID/IP (3.2)

where In is the current of the voltage divider and Ip is the current

of the photomultiplier. The photomultiplier current is defined‘“’ as:

Ip = [(4 counts/s)(# p.e/keV)(Vc.n.)(GPMr)]/(1 C-s-A‘1/Qe-) (3.3)

Assuming 10,000 counts/s, 8 p.e.lkeV (p.e.= photoelectrons) emitted by
the scintillator, the voltage applied to the conversion dynode: 30
keV/count, gain of PHT=1.1 x 10‘, and Qe-= 1.6 x 10'19C/e' gives a
value of Ip = 4.22 x10’7 amp. The value of In = 1500V/(8.49 x 106 +
Re) = 1.67 x 10'“I amp. Therefore, the stability gain is 396.6, which

is a good stability value.(29)

The system has a low pass filter, consisting of a 0.005 uF
capacitor and a 2 k0 resistor, at the input of the +HV, in order to
eliminate high frequency components in the signal. This filter will
allow only frequencies lower than 15 kHz to pass. The coupling
capacitor used in the system (selected based on the criteria of
stability, voltage rating and physical size) was 0.001 uF t 2% (3000

V).

37

The time constant (I) and rise time (tr) can be determined using

the following equations:

tr =O.35/f =(O.35)(2nRC) (3.4)

RC (3.5)

{-4
II

where tr is the rise time, R is the load or terminator resistance and

C is the capacitance.(2°’ In the designed system the time constant

as well as the rise time will be determined by the values of stray
capacitance of the transmission line (30 pF/foot for RG-58/U coaxial
cable), the load resistor (6 k0) and the terminator resistance (50 0).
For two feet of RG-58 cable the stray capacitance will be 60 pF and
a load of 6 k0 (without a terminator resistor) the value of the time
constant and rise time will be 360 and 792 nsec respectively. If a
the terminator resistor is added with a value 125 or 50 D the time
constant will decrease to 7.5 or 3 nsec and the rise time to 16.5 or
6.6 nsec, respectively. The selection criteria involved for the
appropriate time constant and rise time will be discussed in the next

section.

In order to test the photomultiplier as well as the linearity of

the pulse electronic circuit, an LED pulsating circuit was designed

   

38

 

  
    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5))
5))
9
220 Q
20 9F
musr __, 1m torn.) 9T 3" J'L
mamzss 0an m "in “cc 13+
5 n a 3 m
LEDJ'L 7412) I Q
_ A 398
58 nsec. 0 BRN 2 |
7 ,
(a)
1:51 cugmm
PHOTOI'IUUIPLIER HOUSING
1
F " H0
up ELECTRONIC :=O PHOTOHULTIPLIER TUBE SIGNAL
cmwn LED . (3::
new ;

 

 

 

 

 

 

 

 

 

 

 

 

’ |
PM HOUSING "OLDER

 

 

Figure 3.6. Schematic Diagrams of: a) LED Pulsed Circuit,

(b)

and b) Test System arrangement.

 

 

39

(Figure 3.6 a).‘3°’ This circuit generates a light pulse of 50 nsec

and has an operating repetition rate of 1 kHz to 5 MHz.

3.3.2. Detector-Amplifier Interface

The output current pulses from the PET are converted to voltages
through the RC load. The value of the capacitance as well as the value
of the coupling resistor must be kept as small as possible to maintain
the high frequency response of the photomultiplier tube. The value of
the total capacitance (PHT' output capacitance +- amplifier input
capacitance + stray capacitance of the signal cable) should be about
30 to 60 pF. This can be accomplished by having a short signal cable
(RG-58/U see section 3.2.1). The value of the resistor selected must
be a compromise between that which provides high response frequency of
the photomultiplier and amplifier and a value that maximizes the gain
of the amplifier. If a 50 D resistor is selected in order to keep the
PET response frequency high, the amplifier cannot respond to such fast
pulses and the signal will be distorted. Also, if the gain of the
amplifier is not sufficiently high to amplify'such a small signal, even
when it has a fast frequency response, then the advantage of using a
small resistor is lost since the signal cannot be measured by the

subsequent circuitry.

40

In the experimental setup used, an input resistance of the
amplifier of 125 O was utilized. This resistor value develops
sufficient signal voltage pulses and provides a good frequency response

from the photomultiplier.

3.3.3. Amplifier

In order to have good amplification of the pulsed signal voltages
the amplifier must meet certain criteria: (1) minimum input capacitance
so that the degradation of the input pulse be a minimized, (2) low
noise in order to avoid inductituxof noise in the subsequent electronic
components, (3) stable gain, and (4) low output impedance so that the

load can be driven at high frequency.

The amplifier used in this work is an Ortec linear amplifier model
410. This amplifier has a constant input impedance of 125 0 and an
output impedance of 1 0 that can be shunted on the input BNC connector
with terminator resistance to achieve 93 or 50 D. It also provides

a total stable gain factor in an RC shaping mode between 0.35 to 480.

It has a rise time of 80 nsec and an RC pulse shaping with time
constants from 0.1 to 10 nsec t 2%. The maximum amplifier band pass
is within 3 db from 700 Hz to 4.3 MHz. The amplifier equivalent noise

is less than 7 uV(rms) at the maximum gain, and its nonlinearity is

41

less than 0.1% from 200 mV to 10V.‘3°)

3.3.4. Pulse Height Discriminator and Counter.

Pulse signals with intensities of less than 50 mV must be
amplified in order to be shaped by the discriminator and registered by
the counter. In order to increase the intensity of the signal an Ortec
linear amplifier model 410 was used (described in a previous section).
The discriminator used in this system is an Ortec constant fraction
discriminator model 463. This model allows the user to select between
various types of scintillator detectors from its front panel and has
a minimum discriminator setting voltage of 50mV (3” ( any signal below
that voltage will not be seen). The selection of the discriminator
level is done via a timing single channel analyzer, E&G&G Ortec model
420,(32) which is coupled to the linear amplifier output and the
counter input (see chapter 4 for further discussion). By selecting the
appropriate discriminator level the' number' of background signals

reaching the counter can be minimized.

The final stage of the detector system is the counter. The counter
used during these experiments was a Hewlett-Packard model 5216A. This

counter makes frequency, period, ratios and time interval measurements.

Its sensitivity can be set between 0.1 and 10 V. It can operate over

a frequency range from 3 Hz to 12.5 MHz and to 1 MHz in the Period

42

setting. The range interval of time measurements is from 10 nsec to
10 sec. For further details on the counter the reader is referred to

the counter manual.‘33)

43

CHAPTER 4

RESULTS AND DISCUSSION

4.1. Characterization of the Photomultiplier Electronic Circuit

In order to test the photomultiplier electronic circuit, a
pulsating LED circuit was built (see Figure 3.6a), which generates a
50 nsec pulse of light. The test was conducted inside a black wooden
box. The first circuit utilized had a gain resistor of 470 k0 and a
decoupling capacitor of 0.004 'uf (see figure 3.5). The pulse
frequencies applied to the LED circuit ranged from 1 kHz to 5 MHz. The
length of the transmission cable used to couple the signal from the

output of the photomultiplier to the oscilloscope was 3 ft.

4.1.1. Anode Response Time

The response of the PET electronic circuit is presented in Figure
4.1. The rise and decay time of the anode under the given conditions
were 0.035 nsec and 0.04 nsec respectively. The signal full-width at
half-maximum was 0.05 nsec. In Figure 4.1 the presence of a small
shoulder on the sides of the signal rise and decay can be observed.
No explanation could be given at that time due to the lack of a better
resolution oscilloscope (a 100 MHz scope was used in these
experiments). Later with the aid of a 300 MHz oscilloscope and by
turning on only the LED pulsating system, it was determined that the

shoulders originated from the LED electronic circuit and were coupled

 

44

 

 

5 psec

Figure 4.1. Anode Response to LED Pulsating Circuit.
Applied frequency = 5 kHz; PET applied voltage=
1500 V; Oscilloscope set at 5 nsec/div and
5 mV/div. Intensity of the peak 8 19 mV,

FVHH =0.9 nsec.

45

to the transmission cable. Several approaches were taken in order to
try to solve the problem. Among them were: (a) increasing the
separation between the LED and the photomultiplier, (b) adding an extra
external ground at the different input and output BNC cables and
connectors, (c) enclosing the LED electronic circuit inside of a
grounded aluminum foil cover, (d) changing the frequency generator, and
(d) rebuilding the LED electronic circuit with new components after
testing each one. Unfortunately none of these steps eliminated the

problem.

In order to determine the effect of the low pass filter on the
anode rise time this component was eliminated. It was observed that
the anode rise time had some improvement, in the microsecond range, but
the resolution of the scope did not allow the quantification of such
improvement. This improvement was attributed to the disappearance of
the shoulder observed previously. No explanation could be found for
the fact that in the presence of the low pass filter noise originating
in the LED circuit could be coupled to the system. Since an
improvement of the signal was observed this circuit was used to test
the lens system that was employed to focus the ions onto the enhancer
dynode and to test also the enhancer dynode and the plastic
scintillator. This electronic circuit was later modified for reasons

that would be discussed later.

5 mV 1: Ilflfllihhifly‘ ’ ‘Eflflh

5 nsec

(a)

5 mV ‘1 lvh-~.‘w‘.

5 nsec

(b)

Figure 4.2. Effect of the Length of Transmission Cable
on Anode Output Signal: (a) with a 3 ft
cable and (b) with a 1 ft cable.
Oscilloscope set at 5 nsec/div and

5 mV/div.

47

5 Irv ]: S~IIHIVV~N' ;;flflfil

 

 

5 nsec

(a)

. ..1

 

(b)

Figure 4.3. Effect of the Value of the Terminator Resistor
on the Anode Output Signal: (a) 5.8 in and
(b) 50 D. Oscilloscope set at 5 nsec/div and

5 mV/div.

48

The newly modifying PHT electronic circuit again utilized the low
pass filter and included a new decoupling capacitor ( 0.001 uf). This
capacitor was used in order to determine the effect of the decoupling
capacitor on the anode rise time. No improvement in the anode rise
time was observed. Therefore the value of the decoupling capacitance
does not limit the anode rise time. Improvement in the anode rise
time was found when the length of the transmission cable was changed
from 3 ft to 1 ft and terminated in a 50 D resistor. Figure 4.2 (a)
and (b) shows the effect of the length of the transmission cable on the
output signal of the anode while Figure 4.3 (a) and (b) shows the
effect of the terminator resistor on the anode output signal. From
figure 4.2 it can be observed that the output signal broadens and
decreases as the length of the transmission cable is increased. In the
other situation where the‘value of the terminator resistor is increased
an increase in the signal intensity is observed but also a broadening

of the signal (Figure 4.3).

4.1.2. Signal-to-loise Ratio

The signal-to-noise ratio will determine if the detection system
under consideration will be useful for the ion beam instrument, for
which the signal intensities from the PMT are expected to be in the mV
range. Therefore, the higher the value of the signal-to-noise ratio,
(1000 or more), the greater the probability of detecting the

scintillation signal.

49

The value of the signal-to-noise ratio was determined as follows:
(1) the photomultiplier tube was powered at an specific voltage for a
three hour period in the dark, (2) the anode signal was coupled to a
counter, (3) the counter was set to its maximum sensitivity (0.01 V)
and the maximum counter time (10 s), and (4) after the counter was
left to stabilize for a time period of 20 minutes the number of dark
counts was recorded. The same procedure was used to determine the
number of counts when the LED was operated at 1.5 kHz. Table 4.1
presents the average values for the number of counts registered under

the stated conditions.

The error in the number counted is given by<3" :

Es 3 (Rs + Rb)“ (4.1)

where Es is the error in the signal count, R. is the signal count rate,
Rn is the background count rate and if T is the observation time for
both the background and signal counting the signal-to-noise ratio can
be expressed as:

S/N = (Rs“T*)/(1 + 2Ru/Rs)* (4.2)
If the background count rate is small with respect to the signal count
rate this equation reduces to:

S/N= RsxT" (4.3)

As shown in Table 4.1, the background count rate is much smaller than

50

Table 4.1

Characterization of Photomultiplier Signal Count Rate

 

 

 

PET voltage Dark Counts Signal Counts Error SIN
(V) (counts/sec) (counts/sec) (Eh)
x 10" x 10" x 10'Is*
1000 0.0000 1.4396 3.7942 119.98
1100 0.0000 1.4389 3.7933 119.95
1200 0.0000 1.4377 3.7917 119.90
1300 0.0004 1.4393 3.7938 119.97
1400 0.0003 1.4366 3.7907 119.86
1500 0.0003 1.4395 3.7944 119.98

 

LED frequency = 1.5 kHz,

seconds.

counter sensitivity

0.01 V, gate time= 10

51

the signal count rate; therefore the expression to be used is equation
(4.3). Table 4.1 also shows that the signal-to-noise ratio is 100 for
a total count rate of 103 counts/sec. For the experimental conditions
at 1500V PHT voltage the average photocurrent is 1.8 x 10'10 amperes.
The average dark current for a rate of 0.3 counts/sec is 5.3 x 10‘1‘
amperes. This represents a reduction in the value of the dark current
by a factor of 1000 compared with the anode dark current reported by

Hamamatsu Co., (0.8 namp).

4.2. Amplifier-Discriminator Signal Coupling

The output signal from the photomultiplier tube was directly
coupled to the 410 linear amplifier with the amplifier output in turn
coupled to the constant fraction discriminator. Amplification of the
signal was necessary because the intensity of the signal from the PET
was too small for the minimum value required by the discriminator (50
mV). Two problems were encountered when the anode output signal was
coupled to the amplifier and discriminator: (1) the minimum time
constant of the output signal from the amplifier was 0.01 usec, and
(2) the discriminator setting control was very sensitive to any small
change. The first problem (limitation on the time constant) did not
allow a true reproduction of the anode output, since it broadens the
signal to a value of 0.01 nsec and causes pulse overlaps. The second
did not permit the setting for the correct discrimination of the number

of dark count with respect to the signal. Table 4.2 presents the

52

Table 4.2.

Count Rate as a Function of Discriminator Voltage

 

Discriminator Voltage Dark Counts Signal Counts

 

(mV) (counts/sec) (counts/sec)
110'“3 810"3
50.040 9.2153 18.6704
50.060 9.3532 16.6708
50.080 1.5688 16.6617
50.100 0.3203 16.6516
50.120 ------ 16.6770
50.140 ------ 16.6531
50.160 ------ 16.6711
50.170 -------------

 

Amplifier conditions :

constant = 0.1 nsec.

gain=

3.

fine

PET voltage = 900 V.

integration time

LED frequency = 1.5 kHz.

53

counts values after the signal was passed at different discriminator
values. It can be observed that the number of signal counts did not
change even when the discriminator voltage was set at a value which
should discriminate against the dark counts. This could be attributed
to some pulse overlap in the signal or to noise developed in the

amplifier.

In order to try to overcome the problem a Colinear preamplifier was
placed at the anode output and coupled to a single timing channel
analyzer. Unfortunately, the preamplifier used had a 50 0 input
resistor and output that decreased the signal-to-noise level. The
single timing channel analyzer required a minimum signal level of 100

mV that could not be met the by preamplifier signal level. The same

configuration was used with the constant fraction discriminator but the
same signal level obtained from the amplifier could not met the
limiting input voltage of the discriminator. Since no improvement
could be made in the characterization of the photomultiplier
electronics and signal processing using this configuration a mass
spectrometer was employed in order to try to simulate the real system

conditions.

4.3. lass Spectrometer Test System
The description of the system was given in the previous chapter.

In order to use the ion beam from the mass spectrometer to generate the

54

 

(b)

Figure 4.4. Beam Visualization of Test Background Signal in the Vacuum
Chamber. Conversion dynode voltage = -30 RV. (a)
Deflection lens voltage = 0 V, (b) deflection lens apply

voltage = 70 V.

55

.uoamaau mssoe> on» :« noneu casouuxoem «o msuuoemm nae: .m.¢ emsowm

mass. amt!

 

 

 

 

 

 

 

HIBMZmHa>

56

Mn .4.AL¢:1.\M‘

unr- spasm“

 

Figure 4.6. Frequency Dependent Signal from PET Electronic Circuit
without Low Pass Filter. PET voltage = 1500 V,

Conversion Dynode voltage = -30 EV.

57

anode output signal it was necessary to make the ions reach the

conversion dynode and to observe an output signal (electrons emitted)
via a Faraday plate at, the position where the scintillator and
photomultiplier tube was going to be placed. This experimental set-
up has a threefold advantage: (1) it allows the optimization of the
ions exiting from the source by adjustment of the voltages of the
lenses between.the source and the quadrupole; (2) one can validate and
optimize voltages of extractor and deflection lenses between the
quadrupole and the conversion dynode and (3) the utility of the Daly

conversion dynode can be determined in the experiments.

The signal output obtained after optimizing the previously
mentioned experimental arrangement was 2.0 x 10 ‘9 amp. at an applied
conversion dynode voltage of -30 kV. It was observed that decreasing
the conversion dynode voltage resulted in a decrease in the signal
current was observed. This agrees with the theoretical modeling done
with SIEION 4.0. In addition, a beam visualization was. done using the
scintillator as a visualizing screen. Figure 4.4 shows the signal

image obtained during the experiments.

Figure 4.5 presents the mass spectrum of the background gases
present in the chamber. After the background mass spectra were
obtained the photomultiplier tube for pulse counting was placed in the

chamber. The electronic circuit used initially did not contain a low

58

(a) 5 “‘°C

 

Figure 4.7. a) Anode Output Signal of Background Gas in
the Chamber: (a) without Discriminator,
and (b) with after discriminator with amplifier.
Eass = 18 amu, discriminator set at 50.04 mV and

amplifier at a gain of 1.

59

Table 4.3.

Photomultiplier Count Rates Using the Eass Spectrometer

 

 

PETvontaoo Dark Counts Signal Counts Error S/H
(V) (counts/sec) (counts/sec) Eh
x 10-3 x 10"
1500(" 0.0003 6.6002 77.476 244.98
1500‘”) 0.0402 608.098 780.31 2467.3
1500(°’ 0.0003 6.4170 80.106 253.32
1500“” 0.0402 644.566 802.85 2538.51

 

(a) monitoring H20, m/z = 18, without amplification.

(b) amplifier conditions: gain = 1, fine gain = 1.5, integration time
constant = 0.1 nsec.

(c) monitoring Na, m/z = 28, without amplification.

Quadrupole controller setting: high cal e 1.2, low cal = 0.75,

low mass =0.3, resolution = 5, delta E = 0.

60

pass filter. Vith this electronic circuit in the system presented

one problem was encountered; when the mass spectrometer was scanned and

approached a region where an intense peak was present a frequency
dependent signal was observed, Figure 4.6 shows the frequency
component observed. This problem was attributed to the presence of rf
noise on the signal output.‘3°’ It was corrected by placing a low
pass filter in the photomultiplier tube electronic circuit . Figure
4.7a shows the anode signal output and Figure 4.7b shows the output of
the anode signal after passing through the constant fraction
discriminator amplification. It can be seen from Figure 4.7 that the
anode signal output has a greater intensity than the one generated by

the LED test circuit. Table 4.3 presents the count rate of the system

under consideration. The procedure used to obtain these data was the

same as the one used with the LED circuit (Section 4.1.2).

For the experimental conditions employed the average charge content
of the pulses changes with the applied photomultiplier voltage, thus
causing a photocurrent change. The average charge content at 1500 V for
the signal is 1.2 x 10-9 amperes without amplification and with
amplification is 1.1 x 10" amperes. The equivalent dark current was

5.3 x 10-1‘ amperes without amplifier and 7.1 x 10-12 amperes with it.

61

These values represent a reduction in the dark current of the
photomultiplier tube, and demonstrate the inherent advantage of the

photon counting technique in systems were the signal intensity is very

low.

 

62

CHAPTER 5

SUEEARY AND CURRENT STATUS OF THE DETECTION SYSTEN

5.1. Summary

The data presented in the previous chapter for two different
arrangements of the photomultiplier circuitry for pulse counting
indicates a fast response time (in the nanosecond range.) The exact
characterization of the anode response time was not possible at the
time of this work due to various factors. These include: (1) lack of
a pulsed emission light source that has a pulse width of 5 nanoseconds
or less, and (2) the presence of a frequency component in the anode
output signal that did not allow the exact determination of the
response time of the signal. Since the LED pulse circuit could not be
operated at such small pulse widths and because the inclusion of a low
pass filter caused an interference signal at the same frequency as the
anode output signal, only an estimate of the value of the anode

response signal could be made, and this is 50 nanoseconds 1 10%.

In addition to the characterization of the PET electronic circuit,
an attempt to determine the linearity of the counting system was made.
Knowledge of the pulse height distribution would give information on
where the discriminator can be set for the best stability and
discrimination against dark counts. Unfortunately the discriminator

used did not allow the proper setting for the discrimination of the

63

used did not allow the proper setting for the discrimination of the
dark count from the anode signal response ( see Table 4.2) Although
this test could not be performed properly the signal-to-noise ratio was

obtained by determining the rate of dark counts and the rate of signal
counts (see Table 4.1.) The dark current of the photomultiplier was
observed to be lower than the value reported by Hamamatsu Co., for the

PET operating at 1500V.

These series of experiments could not provide the complete
characterization of the system, nonetheless, since the PET pulse
circuitry was operating in the nanosecond range it was possible to test
the rest of the detector system (extractor and deflection lens,
enhancer dynode and scintillator). It was found that the SIEION
modeling correctly predicted the behavior of the system. It was also
observed that the intensity of the signal decreased as the voltage
applied to the enhancer dynode was decreased, and that signals obtained
with an operating voltage (applied to the enhancer dynode) over -15 keV
did not require any type of amplification. In addition, a decrease in
the dark current and a good value for the signal-to-noise ratio was

observed (see Table 4.3.)

5.2 Current Status of the Detector System
The detector system for photon counting is at the present moment

in its final stage of characterization. In order complete its

64
characterization a new discriminator and an amplifier or preamplifier
(with a smaller time constant) will be used. This will permit the

determination of the linearity of the system.

After its final characterization the system will be coupled to the
ion beam instrument. At that stage experiments similar to those done
with the quadrupole mass spectrometer will be performed with the entire
system (determination of: dark counts, signal counts, shape of the
output pulses, linearity and stability measurements). Once those
experiments are finished it will be necessary to automate the system.
This will allow the user to program voltages, discriminator settings
and counting times via a computer system and also will facilitate the

handling of the data obtained during experiments.

 

65

REFERENCES

1. Eelton, C.E.: Principles 9; Eass Spectrometry and Eggative Ions,
edited by Marcel Dekker, New York, 1970, p.141.

2. Daly, N.R.,EcCormick, A., Powell, R.E.; Rey. Sci. Instr., 39,
1163, (1968).

3. Idaho National Engineering Laboratory, EG&G Idaho Inc., Idaho
Falls, ID.

4. Harris, E.E., Trott, G.V., Eorgan, T.G., Brenton, A.G., Kingston,
E.E., Beynon, J.H.; Eass Spectrom. Rev., 3, 209, (1984).

5. Hamamatsu Corporation, Technical Eanual RES-0795, Bridgewater,
N.J.

6. Dietz, L.A., Hanrahan, L.R.; Rev. Sci. Instrum., 49(9), 1250,
(1978).

 

7. Hurlbut, C.R.; Plastic Scintillators a Survey, (1985), Bicron Co.,
Newbury, OH.

8. Bellian, J.G., Dayton, R.R.: EaI m) Sciptillation Detectors, Bicron
Co., Newbury, OH.

9. Daly, N.R.; Rev. Sci. Instrum., 31(3), 264, (1960).

 

10. Burle Industries Inc.; RCA Photomultipliers Handbook, Lancaster,
PA., (1980).

11. Ealmstadt, H.V., Barnes, R.E., Rodriguez, P.A.; LChem. Educ., 41,
263, (1964).

 

12. Hamamatsu Corporation, Application Eanual RES-0790, Bridewater,
N.J. .

13. Strausser, Y.E.; U.S. Patents Applications No. 3,538,328, (1970).

14. Bourne, H.C.Jr., Cloud, R.V., Trump, J.G.:g. Appl. Phys., 18, 596,
(1955).

15. Vasko, B.A.; Personal Communication, Bicron Technical Sales Dept.
16. Viterbini, E., Adriani, A.; Rev. Sci. Instrum., 60(2), 280, (1989).

17. Hinefordner, J.D., Vickers, T.J.: Anal. Chem., 42, 206R, (1970).

66

18. Nakamura, J.K., Schwartz, S.E.; Appl. Qp_., 7, 1073, (1968).

19. Franklin, E.L., Horlick, G., Ealmstadt.; Anal. Chem., 41, 2,
(1969)

20. Rosenstock, H.E.; Int. g; Eass Spectrom. 19g Phys., 20, 139,
(1976).

 

21. Ingle, J.D. Jr, Crouch, S.R.: Anal. Chem., 44, 777, (1972).
22. Ingle, J.D.Jr., Crouch, S.R.; Anal. Chem., 44, 785, (1972).

23. Darling, E.J.; 2230. Thesis, Eichigan State University, 1978.

 

24. Dietz, L.A.; Rev. Sci. Instr., 36, 1763, (1965).

 

25. EcHose, R.C.; RCA Fast Photggultiplier Tube Technique AN-4797, PA,
(1985).

26. EcHose, R.E.; RCA Photodetector Ag-4884, PA, (1972).

 

27. Hamamatsu Corporation; Photomultiplier Tubes For Scintillation
Counting g High Energy Physic , pp. 26, N.J., (1985).

28. Earti Rabb; Personal Communicatiog, Electronic Designer, E.S.U.
Department of Chemistry.

29. Bicron Corporation; Application Notes SC-110, (1985), Bicron Co.,
Newbury, OH.

30. E&G&G Ortec Co; Instruction Eanual Eodel 410, (1966), E&G Ortec
Co., Oak Ridge, TN.

31. E&G&G Ortec Co; Instructiop Eanual Eodel 463, (1966), E&G&G Ortec
Co., Oak Ridge, TN.

32. E&G&G Ortec Co: Instruction Eanual EOdel 520, (1966), E&G&G Ortec
Co., Oak Ridge, TN.

33. Hewlett Packard: Instruction Eanual Electronic Counter 5216A,
(1969), Hewlett Packard Co., Santa Clara, CA.

34. Norton, G.A.: Appl. th., 7, 1, (1968).

35. Harsh, E., Prat, L., Gilton, T., Cowin, J.P.; Rev. Sci. Instrum.,
60(9), 3070, 1989.

"ITrwfmmmmmES