The Production and Use of Proton-Induced Ultrasoft X-rays · 1 The Production and Use of...
Transcript of The Production and Use of Proton-Induced Ultrasoft X-rays · 1 The Production and Use of...
The Production and Use of Proton-Induced Ultrasoft X-raysJones, Elizabeth Anne
The copyright of this thesis rests with the author and no quotation from it or information
derived from it may be published without the prior written consent of the author
For additional information about this publication click this link.
http://qmro.qmul.ac.uk/jspui/handle/123456789/1487
Information about this research object was correct at the time of download; we occasionally
make corrections to records, please therefore check the published record when citing. For
more information contact [email protected]
1
The Production and Use of Proton-Induced
Ultrasoft X-rays
by
Elizabeth Anne Jones, B. Sc., A. R. C. S.
A thesis submitted for the degree of Doctor of Philosophy
of the University of London'.
Dept. of Medical Electronics and Physics The Medical College of St. Bartholomew's Hospital Charterhouse Square London EC1 June 1988
2
ABSTRACT
A 700 keV Van de Graaff accelerator was used to accelerate protons
onto solid targets of different light elements to produce ultrasoft,
characteristic X-rays (< 5 keV). The proton energies were calibrated
using the (p, y) resonances at 633 keV in Aluminium and at 340 and
483 keV in Fluorine.
The X-ray emission characteristics of Aluminium, Carbon, Gold,
Silicon/Carbon, Silicon/Nitrogen and Titanium/Boron were studied as
a function of incident proton energy, angle of inclination of the
target (30° - 60° to the proton beam) and angle of detection of the
X-rays (40° - 130° to the beam). Detection of the X-rays was achieved
using a gas-flow proportional counter directly coupled to a low-noise
pre-amplifier. The resulting spectra, recorded on a multichannel
analyser, were well fitted by linear combinations of single Gaussian
curves to give peak position (X-ray energy), width and area (X-ray
intensity).
Carbon contamination of the target surface was studied in detail
for the Aluminium target. A number of low beam currents onto the
target (10 - 70 nA) were used for total irradiation times of up to
17 hours in order to establish the degree of overall X-ray energy
mixing.
The information gained from the study of both the Carbon
contamination and the X-ray emission characteristics was used to
propose practical optimum conditions for the production of ultrasoft
X-rays by proton bombardment in their application to biological and
biochemical irradiations.
A computer code, capable of following the electron track histories
resulting from ultrasoft X-ray interactions has been used to compare
the details of such energy deposition with the results of mammalian
cell irradiations made at the M. R. C. Radiobiology Unit for a number,
of different ultrasoft X-ray energies. Such a-comparison has been used
to try to identify the mechanisms of radiation action. Included in this
work is the application of the computer code to a variety of. characteris-
tic X-ray photon energies, thus extending the available, calculated data.
3
ACKNOWLEDGEMENTS
I am deeply indebted to many people for their help towards the
completion of this work. Foremost among these is my supervisor, Dr. F. A. Smith, who has patiently steered me throughout the duration
of this study and has been a constant source of inspiration,
encouragement and advice. Sincere thanks are also due to Dr. D. T. Goodhead at the M. R. C. Radiobiology Unit (Chilton, Didcot) who was always available for discussion and guidance on all aspects of the
work.
My thanks are also due to Prof. N. F. Kember for making the facilities at the Physics Department of the Medical College of St. Bartholomew's Hospital available, and I am grateful for the
comradeship afforded to me by him and all the members of the department during my three years there. Special thanks must go to Len Eastgate for constructing various parts of the equipment and James Oriel for his
constant advice and help with the electronic equipment (and breakdowns! ).
I am also grateful to Peter Browne and his colleagues for their assis- tance with the computing side of the work.
During the duration of this study I made several visits to the M. R. C. Radiobiology Unit and I am grateful to all those who made these
visits possible. I am especially grateful for being allowed access to
the computing facilities at the Unit and appreciate the help given by
members of the Computer Services Division. I also appreciate the friend-
ship and advice given by members of the Cell and Molecular Biology
Division. In particular, I would like to mention Dave Charlton (now at Concordia University, Montreal, Canada) and Hooshang Nikjoo - who was
an invaluable source of advice and information concerning the use of the
Monte-Carlo track structure codes. They, along with Dr. Goodhead, also
provided me with unpublished data, for which I am very grateful.
I am indebted to the Science and Engineering Research Council for
the award of a research studentship and to the Medical Research Council
for covering the cost of the visits to the Radiobiology Unit. I am also
grateful to Mrs. D. Boyle for deciphering my writing and typing the
thesis.
4
Finally, I would like to thank my family. My brother, Hefin
has always been ready with his support and helpful advice, and
submitted himself willingly (? ) to reading the final versions of
this work. My parents have given continual encouragement, support
and devotion over the years and this was reiterated by their
willingness to help check the data presented in Appendix 2. It is
as a token of my appreciation that I dedicate this work to my parents
and brother.
Diolch o galon i'bob un ohonoch!
5
TABLE OF CONTENTS Page
Abstract 2
Acknowledgeme nts 3
Table of Contents 5
List of Figures 7
List of Tables 10
Chapter 1 Introduction 11
Chapter 2 Expe rimental Apparatus 22
2.1 The Van de Graaff accelerator 22
2.1.1 The Generator 22
2.1.2 The Accelerator 24
2.1.3 The Pressurised Tank 25
2.2 The Beam Line 26
2.2.1 Accelerator tube extension 26
2.2.2 Collimators 29 2.2.3 The Scattering Chamber 29
2.3 The Target System 30
2.4 The Detection System 34
2.4.1 The Proportional Counter 34
2.4.2 Associated Electronics 43
Chapter 3 Calib ration of the Van de Graaff machine 45
Chapter 4 Data Analysis 55 4.1 Subroutine SIMPLX 56
4.2 Subroutine FCN 58
Chapter 5 Expe rimental Section 61 5.1 Experimental Procedures 61
5.1.1 Surface contamination study 61
5.1.2 Angular distribution study 62
5.1.3 Proton energy dependence of 62 X-ray emission
5.2 Results 62 5.2.1 Surface contamination study 79
5.2.2 Angular distribution study 85
5.2.3 Proton energy dependence of 92 X-ray emission
6
Pace
Chapter 6 Theoretical Treatment of Results 96
6.1 Compensation for target surface 96 contamination
6.1.1 Example of surface contamina- 101 tion calculations
6.2 Calculation of theoretical angular 103 distribution
6.2.1 Example of the theoretical 108 angular distribution calculation
Chapter 7 Application of Monte-Carlo Track Structure 115 Codes
7.1 Introduction 115
7.2 Local Energy Deposition of Radiation 116 Tracks
7.3 Calculations using a Monte-Carlo structure 118 code'
7.3.1 Track Structure and Scoring 119 Programme
7.3.2 Results 120
Chapter 8 Discussion 141
8.1 Comparison of the observed angular 141 dependence of proton-induced ultrasoft X-rays with theoretical calculation
8.2 A practical beam line for proton- 143 induced ultrasoft X-rays
8.3 Monte-Carlo electron track structure 150 data
Bibliography 154
Appendices:
Appendix 1 Subroutine FCN 165
Appendix 2 The absolute frequency of energy depositions 169 greater than a given amount in each elemental cylinder per Gray of absorbed dose to the
medium
7
LIST OF FIGURES Figure Page
1.1 Frequencies of dicentric exchange aberrations 12 in human lymphocytes after irradiation with high and low L. E. T. radiations.
1.2 Dominant mode of absorption of Aluminium K X-rays 14 by an Oxygen atom.
1.3 Energy deposited locally by electrons from low 15 L. E. T. radiations.
1.4 Typical X-ray energy spectrum from a hot-filament 18 X-ray tube.
2.1 Simplified drawing of the AN700 Van de Graaff 23 accelerator.
2.2 Schematic diagram of the beam line. 27
2.3 Schematic drawing of the target system. 31
2.4 The effect of applying a positive bias voltage 32 to the electron suppression shield.
2.5 The angular dependence of the target charge 33
collection time.
2.6 The Manson Gas-flow Proportional Counter. 35
2.7 The transmission of ultrasoft X-rays through the 36 VYNS window.
2.8 The effect of adding Polypropylene filters above 38
the detector window.
2.9 Magnetic deflection of scattered protons. 39
2.10 Calculation of the magnetic field required to 40 deflect the scattered protons away from the hole in the steel shield.
2.11 The correction to the required magnetic field due 42
to the limited extent of the available field.
2.12 Schematic drawing of the ultrasoft X-ray detec- 44
tion system with the associated electronics.
3.1 Proton capture. 46
3.2 Schematic diagram of the beam line as used during 48 the calibration of the Van de Graaff.
3.3 An example of the energy spectrum from the CaF2 50 target.
3.4 Thick target y-ray yield curve for 19F (p, y)
20Ne. 51
3.5 Thick target y-ray yield curve for 27A1 (p, y)
28Si. 52
8
Figure Page
3.6 Calibration graph for the proton energy as 54 a function of the generating voltmeter reading.
4.1 A two dimensional simplex. 57
5.1 Examples of the spectra obtained for each of 65 the targets when bombarded with 500keV protons.
5.2 Derivation of the BK X-ray intensity. 73
5.3 The transmission characteristics of 6 pm 74 Polypropylene at low energies.
5.4 Energy calibration graph for the proportional 76 counter and MCA system.
5.5 The variation of the CK X-ray yield with 82 collected charge on the target.
5.6 The variation of the A1K X-ray yield with 83 collected charge on the target.
5.7 The A1K : CK X-ray intensity ratio as a function 84 of the charge collected on the target.
5.8 The build-up of Carbon contamination X-ray 86 intensity with the total amount of charge collected on an Al target.
5.9 The rate of Carbon contamination build-up as a 87 function of the proton beam current.
5.10 The relative X-ray intensity as a function of 89 the detector angle.
5.11 The mass attenuation coefficient of low energy 93 X-rays in Carbon, Silicon and Gold targets.
5.12 The measured X-ray intensities of CK and A1K X-rays. 94
5.13 The ratio of the measured Carbon-to-Aluminium 95 K-shell X-ray intensity as a function of proton energy.
6.1 The mass attenuation coefficient of low-energy 97 photons in Carbon.
6.2 Calculation of the depth of Carbon contaminant 100 on the target surface.
6. ý Simplifying assumptions used for the calculation 104 of the theoretical angular dependence of the emitted X-rays from a thick target.
6.4 The variation of proton energy with depth into 105 the target.
6.5 The K-shell X-ray production cross-section for 106
proton bombardment of various low Z targets.
9
Figure
7.1 Diagram illustrating the virtual sphere enclosing the simulated electron track(s) and crossed by the scoring cylinders arranged at random.
7.2 The frequency of energy depositions obtained when considering the interaction of CK X-rays with soft tissue.
7.3 The frequency of energy depositions obtained when considering the interaction of A1K X-rays with soft tissue.
7.4 The frequency of energy depositions in a 2 nm x2 nm cylinder by simulated electron tracks from various low Z characteristic X-rays.
7.5 The frequency of energy depositions in a 10 nm x5 nm cylinder by simulated electron tracks from various low Z characteristic X-rays.
7.6 The frequency of energy depositions in a 25 nm x 25 nm cylinder by simulated electron tracks from various low Z characteristic X-rays.
7.7 An inter-elemental comparison for the frequency of energy deposition in a2 nm x2 nm cylinder.
7.8 An inter-elemental comparison for the frequency of energy deposition in a 10 nm x5 nm cylinder.
7.9 An inter-elemental comparison for the frequency of energy deposition in a 25 nm x 25 nm cylinder.
8.1 The sin 6 dependence of the difference in the distance travelled within the target at either end of each pm increment.
8.2 A schematic diagram summarising some of the properties of a practical beam line for biological irradiation by proton-induced ultrasoft X-rays.
Pace
122
124
128
133
134
135
137
138
139
142
149
10
LIST OF TABLES
Table Page
5.1 The energies of the characteristic X-rays studied. 64
5.2 The average parameter values for the various 71 targets.
5.3 The observed yield of X-rays. 78
5.4 Comparison between the observed and documented 80 yields of proton-induced ultrasoft X-rays.
6.1 The linear attenuation coefficient of ultra- 98 soft X-rays in Carbon.
6.2 The A1K intensity - correcting for surface contami- 102 nation.
6.3 The energy loss experienced by a 500 keV 109
proton in an Al target.
6.4 The AlK production cross-section. 110
6.5 The distance travelled by the X-ray within the 111 target.
6.6 The product of the A1K X-ray production cross- 112 section and the attenuation of the X-ray in the target.
6.7 The contribution of the A1K X-rays produced in 114
the first pm of the target.
7.1 Input parameters for the electron generating 121
code, MOCA7.
8.1 The mass attenuation coefficients of low energy 152 X-rays in Oxygen.
11
CHAPTER 1
INTRODUCTION
In 1938 K. Sax published the results of a study of chromosome
aberrations induced in cells by ionising radiations. These results,
together with further studies (for example : Sax, 1940,1957; Evans,
1962; Revell, 1974; Savage, 1975), showed that ionising radiations were
capable of inducing the exchange of chromosome material and it was
suggested that an aberration arose as a consequence of two radiation
damaged chromosomes undergoing an exchange. Radiations of different
ionisation density or linear energy transfer, L. E. T., (L. E. T. is a
measure of the energy imparted to a medium along the path of a charged
particle [ICRU, 1970]) were observed to have varying degrees of effect
on the irradiated cells (Figure 1.1). Combining this information with
the size of the chromosome structures involved led to the suggestion
that the required distances between pairs of damaged chromosomes for
aberration induction (termed the interaction distances) were of the
order of 0.1 to 1 pm (Lea and Catcheside, 1942; Lea, 1955; Wolff, 1959;
Heddle and Wolff, 1966; Neary, Preston and Savage, 1967). In comparison,
the diameter of cell nuclei is typically of the'order of 10 Pm and the
diameter of a DNA double helix is approximately 2 nm.
During the same time period, developments in experimental physics,
particularly with the introduction in the late 1950s of the low-
pressure (Rossi) proportional counter (Rossi and Rosenzweig, 1955),
enableddirect measurements to be made of the energy deposited by
radiation within a small simulated tissue volume. Such measurements,
and their analyses, became known as microdosimetry - the study of the
properties of radiation in small tissue volumes usually smaller than
the nuclei of mammalian cells.
12
m 0 16.
d CL U) C 0
IM I- 0
vý "L
C
0 0
HIGH LET LOW LET
Fission neutrons Y- rays
Dose (Gy)
Figure 1.1 Frequencies of dicentric exchange aberrations in human lymphocytes after irradiation with radiations of high L. E. T. (Lloyd et al., 1976) and low L. E. T (Lloyd et al., 1975). Above each curve is a diagrammatical representation of the track of ionisation produced by a typical secondary charged particle from each radiation (from Goodhead, 1982 with permission, copyright by Academic Press, Inc. ).
02468
13
By using microdosimetric techniques attempts could be made to
measure some physical properties of ionising radiations that corres-
ponded to the properties of radiation resulting in chromosome
aberrations and other biological effects in cells. Such studies
might then elucidate the mechanisms of radiation action, enabling
the radiological risks to human beings to be better understood and
quantified.
As mentioned earlier, L. E. T. is often varied within studies of
radiation mechanisms. The radiations used frequently deposit energy
along tracks which are long in comparison with sub-cellular dimensions,
thus making it difficult to pinpoint the cause of any specific damage.
Radiations which produce tracks of short, defined lengths have been
available, but practical limitations such as their attenuation co-
efficient have limited their use.
However, ultrasoft X-rays of less than ti 5 keV, although having
very large attenuation coefficients and a limited availability, have
been used intermittently since the late 1920s (Goodhead and Thacker,
1977). The interaction of ultrasoft X-rays with matter takes place
almost entirely by photoelectric absorption (Figure 1.2) leading to
the production of photo- and Auger- electrons of low, well defined
energies, capable of producing only very short tracks (< 0.1 pm).
Recent studies on mammalian cells have shown that ultrasoft X-rays are
highly effective in inducing biological damage (Cox et aZ., 1977;
Goodhead et al., 1979,1981a; Virisk et al., 1980; Thacker et al., 1983;
Brenner et al., 1987; Raju et al., 1987) and have been found, in general,
to be more effective per unit dose than hard X-rays and gamma (y) rays.
Although other ionising radiations, such as hard X-rays and y-rays,
produce a large number of low energy electrons (Figure 1.3), the effect
14
> > > C) Cf C) Y Y Y
LO co C') U) r 1ý O) to O O r
.r. "Y
ý C C " p p " 1. 6.
i d C)
" 1
4.0 0
" O üi L 3
a Q
.
E 0 co c a) rn x 0
be CL Igt
>% Co M
qt CO
xoz 41 n0Nn
co 'o o cam, sa "c -o
ý w o Cd
w o Ln
vý O iýs
cC , ýG Oh UN
U 'Z3
u cd O
to i "-4 one
-P4 4j Cd w
ri r4
r-+ 00 d "ý 4ö
WUU S4 O ,. C a 4-,
4J v r. N O OD
", 4 O C) ß. "3H 0
n -+ o rn o ,0 M U) w
w 0 0.1
4-a 0PU H O +-i 41
cd uO
'U 0 r-+ D, co O (1) a) . -a CO Q to 1N "%--%
xN i- .4 CO Cd OU llý r4 r. <4 ä
" El b o0o "v-1
A .o co i- 3
N
r- 0) L 7
LL
15
1.0
0.8 m
0
0 u. 6
v- O
G 0
0.4
I. w
ey 0.2
0 L- 0.1
/º
o,
}V
JC I,
ý
ti0/ `tai ry/
. ý:
0 10
, eA
"/ . t"ýý, /ýMe ego
1
Figure 1.3 Energy deposited locally by electrons from low L. E. T. radiations (D. T. Goodhead and H. Nikjoo, (pers. cormn. ) based on data by Burch (1957) and R. J. Munson (unpublished data) using the method of Burch). 35 - 45 % of the energy is deposited by electrons of 5 keV energy or less.
1 10 100 1000 Electron Energy
, keV
16
of the low energy electrons cannot be studied in isolation from the
effect of the higher energy electrons (Raju et al., 1987). The effec-
tiveness of ultrasoft X-rays was also observed to increase with
decreasing photon energy.
Numerous models and concepts attempt to explain radiation-induced
biological damage (Goodhead, 1987a, 1987b) and, consequently, studies
involving short track length radiations have become increasingly
appropriate. The observation that small, highly localised energy
depositions are effective in inducing biological damage is not only
important to the understanding of the mechanisms of radiation action
but also sets tight constraints on any proposed model of radiation
action (Goodhead, 1977,1982; Goodhead, Thacker and Cox, 1978; Goodhead
et al., 1981b; Thacker, Goodhead and Wilkinson, 1983).
The full potential of ultrasoft X-rays as a fine probe of the
energy and the spatial requirements of mechanisms of radiation damage
in biological systems can only be realised with a wide, versatile
range of sources. A variety of sources have, and are being used to
produce ultrasoft X-rays including cold-cathode discharge tubes, hot-
filament electron tubes, hard X-ray induced fluorescent characteristic
ultrasoft X-rays from secondary targets, laser induced X-rays and
synchrotron radiation sources (First International Ultrasoft X-Ray"
Workshop, 1987). All these techniques have their advantages and
disadvantages (Goodhead and Bance, 1984).
A further method of producing ultrasoft, characteristic X-rays is
by the use of heavy ion bombardment of solid targets. When a target
is bombarded by a heavy particle (proton, deuteron, alpha particle),
electrons are ejected from the electron shells surrounding the target
nuclei. If an impact removes an electron from an inner shell, the
17
subsequent filling of the vacancy by an electron from an outer shell
leads to the emission of X-rays characteristic of the element bombarded.
The production of characteristic X-rays by electron impact is a
familiar phenomenon and gives rise to characteristic peaks super-
imposed on the continuous bremsstrahlung background (Figure 1.4).
The ionisation of an atom by the impact of a heavy particle on the
other hand, and the subsequent emission of an X-ray, has received only
sporadic attention since Chadwick and others first observed and
identified the characteristic X-rays of several elements (ranging from
Aluminium to Uranium) which were bombarded by alpha (a) particles
(Chadwick, 1912,1913; Chadwick and Russell, 1913,1914; Thomson, 1914;
Slater, 1921). A more comprehensive investigation of inner-shell
ionisation in various elements by a-particles was published in 1928
by Bothe and Franz who used a Polonium source for the production of
a-particles. They studied both the K-shell ionisation for targets
of atomic number, Z, greater than 12 and L-shell ionisation for targets
with Z> 34, as well as the M-shell ionisation of Bismuth (Z = 83).
In 1941, Cork used deuterons of energies up to 10 MeV and
examined the blackening of photographic plates by X-rays from 38
chemical elements. Cork found that for K-shell ionisation by deuterons,
the yield of the emitted X-rays maximised in the region of Z= 28 whilst
for L-shell ionisation, the yield maximised at Z= 64. Although Barton
(1930) had unsuccessfully searched for X-rays from a Copper target by
low energy (15-25 keV) proton bombardment, it was Gerthsen and Reusse
(1933) who performed the first successful experiments with protons as
the incident particles. Protons in the energy range 30 - 150 keV were
used and the K-shell radiation emitted from Aluminium and Magnesium as
well as the L-shell radiation from Selenium were. observed. Further
work was carried out by Livingston, Genevese and Konopinski (1937)
18
,ý1 .0 N C 0)
C
0.5
4'
Z c0
Figure 1.4 Typical X-ray energy spectrum from a hot-filament X-ray tube with a Tungsten (Z = 74) target.
/
01 50 100 150 200 Photon Energy , keV
19
using protons up to 1.72 MeV. They measured the intensity of the
emitted X-rays as a function of the atomic number and estimated the
order of magnitude of the cross-section for proton-induced X-ray
production.
Numerous experiments have since been carried out with protons of
energies less than a few MeV. These have been primarily in the field
of particle-induced X-ray emission (P. I. X. E. ) for analytical purposes
(Rutledge and Watson, 1973; Johansson and Johansson, 1976). Protons, as
opposed to a-particles and other heavy ions, have been favoured as the
bombarding particle mainly as a result of the availability of small
proton accelerators in many laboratories. Furthermore protons have
been shown to give the best sensitivity in most practical applications.
There is also a greater problem with target heating and deterioration
when a-particles are used and because of their large energy loss it
is necessary to limit the beam intensity.
Measurements presented by Goodhead and Bance (1984) indicated that
proton bombardment is also a favoured method of producing characteristic
ultrasoft X-rays with greater, and more widely variable, intensities
than are available from sources such as the cold-cathode discharge tube.
The main advantage of proton-induced ultrasoft X-rays, especially when
compared to hot-filament electron-induced sources, may however be con-
sidered to be the relatively low level of background contamination
(which is dominated by secondary electron bremsstrahlung emission).
Goodhead (1987c), for example, has found that for a proton energy of
600 keV, the bremsstrahlung contamination contribution in air, in a
practical irradiation configuration, in the detected photon beam is
less than 0.2% when using a Carbon target and approximately an order
of magnitude less using an Aluminium target. Comparable values for
electron-induced sources are in the region of 1 to 5% (Goodhead and
Bance, 1984; Raju et al., 1987).
20
To date virtually all proton-induced ultrasoft X-ray radio-
biological information has been obtained with Carbon or Aluminium
targets (First International Ultrasoft X-Ray Workshop, 1987) producing
the corresponding K-shell characteristic X-rays (CK X-rays at 0.28 keV
and A1K X-rays at 1.49 keV). However, the application of ultrasoft
X-rays in radiobiology and radiation chemistry would benefit greatly
from the availability of other X-ray energies thereby producing
different electron energies and varying ratios of photo- and Auger-
electrons.
Using a 700 keV Van de Graaff accelerator at St. Bartholomew's
Hospital Medical College the study reported here set out to investigate
the possibility of using proton bombardment of other solid target
materials and to look in greater detail at the angular distribution
of the emitted photons. The study was limited to the characterisation
of the ultrasoft X-rays and no biological irradiations were undertaken.
From the information gained, certain practical optimum conditions for
the production of proton-induced ultrasoft X-rays for biological and
biochemical irradiations are proposed.
As previously mentioned, radiobiological studies involving ultra-
soft X-rays are important in furthering our understanding of radiation
damage mechanisms. Their effectiveness in producing complete biological
effects, even though their energy deposition is small and highly local-
ised, increases the need for further understanding of the physical
properties of radiation interactions, particularly those which may be
correlated with the biological effect of practical radiations.
The biological features of ionising radiations should be
contained within the details of their track structure. Hence, the
critical characteristics of radiation action may possibly be identi-
fied by using computed Monte-Carlo track structure codes to simulate
21
the radiation tracks. By recording all the spatial and energy transfer
information (Paretzke, 1980) these can then be compared with the observed
relative biological effectiveness (R. B. E. ) of the different radiations.
The M. R. C. Radiobiology Unit, Chilton, Didcot possess one of the codes
which is capable of full simulation of electron track histories, inter-
action by interaction. In addition to the experimental work on the
beam line characteristics, the work reported here also applied this
code to a variety of characteristic X-ray energies, thus extending
the available, calculated data.
22
CHAPTER 2
EXPERIMENTAL APPARATUS
2.1 The Van de Graaff accelerator
The proton beam was produced using a 700 keV Van de Graaff positive
ion accelerator (Model AN700, High Voltage Ltd. ). The Van de Graaff
machine produces high voltages by driving charges along a conveyor belt
into a conducting terminal, and consists of three main parts: the
generator (drive motor, power supply, moving belt, high-voltage electrode),
the accelerator proper (the ion source and the accelerator tube) and the
pressurised tank.
2.1.1 The Generator
A simplified drawing of the AN700 Van de Graaff accelerator is
shown in Figure 2.1. The charging belt transports electrons electro-
statically from the high voltage terminal to ground potential by
revolving around a system of two cylindrical pulleys, the drive pulley
situated at the base of the accelerator at ground potential and a
pulley alternator at the high voltage terminal end. The belt charge
is provided by a fine-mesh, stainless steel screen which is insulated
from the generator base plate and positioned such that the corona points
face the outer surface of the belt at the drive pulley. The drive
pulley, on the inner surface of the belt, is grounded through a Carbon-
brush contact, and the charging screen is maintained at a positive
potential by a power supply located on the generator base. A similar
(collector) screen is connected directly onto the high voltage terminal
plate, with the fine stainless steel mesh trimmed to present a series
of sharp corona points to the positively charged belt. Electrons from
the terminal are attracted to the belt through the collector screen,
electrostatically held by the belt and transported to the base plate
aý rn
23
to H
p C o m m
C C 0 V1
Ö E Q 0
m m I- O ++ 10 C
O t aº
C fl
E H rn ýc
N N eý pM > 41 N >
H ý 0 ä ,=E °' 0
a N 0 z --
m
a a,
O 0 0 4-1
c3 y U
0 v a. U
" w w
ö C) U b C
`" m 0 0 ^ M C
rn C) co 4
w O °D C r4 m ". 4 m
U N b
O 'b
14-+
O '4
7 a `"
a a m 3 ME
24
where they are removed by the positive potential at the charging screen,
leaving behind them a positively charged belt.
The high voltage terminal is mounted at the top of the accelerator
column and houses the oscillator which supplies the radio frequency
(r. f. ) power for ionising the source gas. Also situated on the high
voltage terminal plate are the power supply units for providing voltages
to the terminal plate, collector screen and the probe of the ion source
bottle. The high voltage terminal is insulated from the generator base
by a column made up from a number of circular equipotential planes
separated from one another by porcelain insulators. Appropriately
shaped holes in the equipotential planes allow access for the accelerator
tube, the charging belt and the various control rods.
The equipotential planes are electrically connected through series
resistors. These form a voltage divider network and maintain a constant
voltage gradient between the terminal and ground. Each plane is provided
with an Aluminium gradient bar placed such that it faces the upcharge
side of the belt, thus decreasing the possiblity of discharge between the
belt and the plane.
2.1.2 The Accelerator
Hydrogen gas is introduced into the ion source bottle through a
Palladium (Pd) leak assembly. This consists of a Pd thimble surrounded
by a quartz sleeve which serves to insulate the thimble from the spiral
heating element wound closely around it. Power is supplied to the heater
by a5 to 10 Volt transformer secondary winding which warms the Pd, thus
making it porous to Hydrogen and its isotopes. The gas passes through
the Pd and flows through a gas line connected to the thimble into the
ion source.
25
Within the ion source bottle the gas is ionised by r. f. energy
with the resulting plasma being magnetically concentrated at the exit
canal of the source bottle. Positive ions are initially expelled
through the exit canal into the acceleration path by a potential
applied to the probe of the source bottle. Further acceleration is
provided by the voltage gradient developed along the column. Focusing
of the beam can be achieved by applying a negative potential to the
focus electrode situated at the ion source end of the accelerator tube.
The accelerator tube is evacuated to 10-6torr (1.33 x 16-4 Pa) and
provides a long mean-free path for the accelerated particles. The tube
is mounted on the generator base and extends up through the column of
equipotential planes to the high voltage terminal. It consists of a
number of ring-shaped insulators cemented between dished Aluminium
electrodes, the number of electrodes in the tube being equal to the
number of equipotential planes in the column, each tube electrode being
electrically connected to a column plane by a spring connector.
2.1.3 The pressurised tank
The generator is housed within a cylindrical tank containing a
high pressure gas mixture in order to prevent the continuous discharge
from the high voltage terminal to ground. As the pressure of a gas
increases the distance between the molecules becomes correspondingly
shorter as does the mean-free path of an ion with the effect that any
gaseous discharge will appear at higher electrical fields and thus
enable stable operation up to a higher voltage limit.
The AN700 Van de Graaff accelerator was operated at gas pressures
> 115 lb. in-2 (7.9 x 105 Pa) with a mixture of Freon (CC12F2).
Carbon dioxide (C02) and Nitrogen (N2). Initially these gases were used
in the proportion 15% Freon, 35% C02 and 50% N2. However, a reduction
26
in the percentage of C02 proved to give a more stable mode of operation.
As a result the accelerator was operated in the later stages of the
work with 30% Freon and 70% N2 in the tank, the Freon being introduced
first to bring the evacuated tank up to about 30 lb. in-2 (-2.1 x 105
Pa). Whilst the tank was filled to a total pressure of 115 lb. in-2
(7.9 x 105 Pa) the subsequent warming up of both the Freon and the
Nitrogen meant that the accelerator was invariably operated at gas
pressures of 120 - 130 lb. in-2 (8.2 x 105 - 9.0 x 105 Pa).
2.2 The Beam Line (Figure 2.2)
2.2.1 Accelerator tube extensions
On leaving the accelerator, the proton beam travelled through an
assembly of different tube extensions that provided an evacuated
channel between the accelerator and the scattering chamber. Included
in the tube extension assembly were:
(a) Penning ionisation gauge. The vacuum in the accelerator tube
system was monitored by a Penning ionisation gauge (C. V. C. Products
Inc. ) capable of measuring vacuum pressures in the range 1x 10-6
to 1x 10-3 torr (1.33 x 10-4 - 1.33 x 10-1 Pa). The vacuum pressure,
as measured by the gauge, was monitored on the accelerator control
panel.
(b) Main gate valve. The main gate valve was used to isolate the
accelerator tube from the tube extensions.
(c) The Associated Vacuum System. The accelerator-tube system could
only be operated under a high vacuum (low pressure) to minimise
collisions between the accelerated particles and gas molecules in the
tube. This was achieved using an Edwards Speedivac Oil Vapour Diffusion
Pump (Model 203B). The pump was water cooled, had a pumping capacity
of 250 - 300 m3/hour and contained 60 cm3 of Diffusion Pump Fluid 704
(Dow Corning). A mechanical backing-pump (Edwards E2M2 Double-stage
27
a)
0 Co ai
a)
w 0
ä 00 Co
b U u CO Ei a 4 U
u2
N
N
a, L
ü) LL
a 14 CC o º+ u U
CC 0 .ý 00
0 14 o 1. + 0
v .., Cv .d s-I 0 41 $. 4 d Z U
.ý y )
"rI
41 a
. 14 r-4
:1
ü0 r4
E
"4
0 41 lC ".
o0o Cl1
v
l0 ý, ' 4-1 r-4 CC 10 Ei C) P-4 OD
N
ü
", 4 0) 41 u Ü
cn
"O N N
ä
c 14 4j c)
w
0
tu P-4 a
4-+
4+ a to
12 .u
.C H
0 0
. -4 ä
0 4
-
P4
a 3
ýo
0 U
Ü
1 0 41
-0
a
"-
c: 0 .1 14 0
0 14
0 F-4 44 1 Co R S cQ
5
-
"0 C) Cl)
,0
>
O (D M +
. -4 0
1+ C) aj Q)
"cC 0
9
r.
> r-4 td
4i
Cu oo
14 p-, V-4 H X
d
E
p.
. 0i to :1
w 44 º- ca
cl
r-4 co >
tad ba
c A "0
Z
0
41
vii "e4
"oo
r. ., 4 r.
44 44 tu
1.4 O
d b
D
Ch
O
14
U
14 o0 C)
a) to F+
n co c'. O
28
Cl)
rýN r
O
>1 O
co
1,
N 1
1 lf)
I co
ý]
i - --=-J
L
29
Rotary Vacuum Pump) was used as a secondary system to remove the exhaust
from the main oil-diffusion pump to the atmosphere. Situated between
the oil-diffusion pump and the proton beam line was a double walled,
liquid-Nitrogen cooled cold trap to serve as a hydrocarbon condenser.
(d) Auxiliary gate valve. This gate valve was used to isolate the
scattering chamber from the tube extensions. Work could therefore be
carried out on the chamber without interrupting the evacuated state
of the accelerator tube.
2.2.2 Collimators
Prior to entering the scattering chamber, the proton beam encoun-
tered an Aluminium (Al) collimator with a 7mm diameter central hole.
Once within the scattering chamber the beam passed through another Al
collimator (1 mm diameter hole). Previous work (Khan and Potter, 1964)
has established that electrons produced by collisions of imperfectly
focused protons on the tube walls pass down the accelerating tube
with the protons. Secondary electrons are also produced at the edge
of the collimating discs so the collimator within the scattering
chamber was held at 300 V positive potential (supplied from a Harwell
2000 Series E. H. T. Unit, Type 2147 - 2) to reduce backstreaming-of the
electrons from the collimator to the accelerator.
2.2.3 The Scattering Chamber
The scattering chamber consisted of a 43 cm diameter, 23 cm wide
cylinder, orientated with the cylinder axis horizontal, with both circular
ends forming removable, vertical lids. The target and detection systems
entered the chamber through opposite lids through '0' ring seals and
the required electrical connections were made using B. N. C. pressurised
bulkhead adaptors (including one high voltage version)-and insulated
glass-metal seal 'feed-throughs' on the chamber walls. Separate pumping
systems were used for the scattering chamber and the accelerator, the
30
scattering chamber being evacuated with an Edwards Speedivac Oil Vapour
Diffusion Pump (Model F203) with a Welch Duo-Seal (Sargent-Welch
Scientific Company) backing pump. A double walled, liquid-Nitrogen
cooled cold trap was mounted onto the scattering chamber. The vacuum
pressure in the chamber was monitored by an Edwards Penning Gauge Head
(Model 6) positioned on the cylinder wall, and was generally at
' 10-5 torr (1.33 x 10-3 Pa) at the start of an experiment.
2.3 The Target System (Figure 2.3)
Thick (6 mm) solid targets of Aluminium (Al), Carbon (C), Gold (Au),
Silicon/Carbon (SiC), Silicon/Nitrogen (Si/N) and Titanium/Boron
(TiB2). were clamped onto an insulated, semi-cylindrical brass holder
which entered the scattering chamber through one of the removable lids
(Section 2.2.3). The TiB2 (Borax Consolidated) SiC and Si/N
(Morganite Special Carbons Ltd. ) targets were hot compressed, the
Si/N target believed to be predominantly Si3N4 (ti 90%) and the remainder
consisting of an unknown ratio of Si to N atoms. The target holder
could accommodate all six targets together, each target being suffi-
ciently wide (> 1 cm) to allow the proton beam to strike at least
two positions. Movement from one target position to another was
possible without breaking the vacuum in the scattering chamber as was
varying the target orientation with respect to the beam. This latter
facility permitted a complete 360° rotation about the target holder's
axis although in practice, the need for electrical connections res-
tricted it to < 180°. The angle which the target made to the beam
could be monitored from outside the scattering chamber.
The target holder was surrounded by an electron suppressor shield
which was insulated from the target holder and the scattering chamber.
Using a similar design to that used by Sterk, Marks and Saylor (1967),
31
O vº 3C
Z'
a 4- rn ý
vi C
_ý .. m r- .r tV NN
Nr =a or -
L0
CCP to
w
"C
VN N
U)
C .` 1
0
io
w cm a) 0 ou-
mN N
. 92 Co Co
mm
O
2
öC to 0u
0 a
01 C
N ýp t0 ++ h.
9 Q) u Co U)
v 00
4J
0 bO
c3 b u
v u
M
C,
lJ..
32
N
0 r
U
N CV
V ,4
0 0
O
d
.xg
m E
d rn cv m
Figure 2.4 The effect of applying a negative bias voltage to the electron suppression shield with the target at 45° to the beam. The points shown are the average values (± s. d. ),
over five readings, for the time taken to collect a charge of 25 pC on the target.
0 10 20 30 40 50 60 70 80 90 Target bias voltage (V
33
`n 5 N Q
v
U
U, c, 1 4-
04 a, ö 0
0
c YI c9
. 6. -
(D E
4-
d Of
(9 0 b. i d
Figure 2.5 The angular dependence of the target charge collection time for an unbiased electron suppression shield (") and with a -90 V bias voltage on the shield (0). The points shown are the average values (± s. d. ) taken over five readings.
Target-to-beam angle
34
the target was biased at +90 V relative to the electron shield using
a power supply operated with its output floating (but referenced to the
shield). This was found to be sufficient to prevent secondary electrons
from escaping (Figure 2.4) thereby causing a false indication of beam
current - an effect which was observed to be target-angle dependent
(Figure 2.5).
The current onto the target was monitored using a charge integrator
(Figure 2.2) constructed in the Physics Department of St. Bartholomew's
Hospital Medical College. The current onto the target was transferred,
through a coaxial cable, to the integrator where a2 NF capacitor was
charged. The voltage across the capacitor was measured and monitored
on a 25 V full scale deflection (f. s. d. ) meter, thus giving a charge
measurement of 50 pC f. s. d. The integrator had a number of different
capacitors (1 nF to 2 pF) in its circuitry such that a suitable value
could be chosen for the particular current being measured.
2.4 The Detection System
2.4.1 The Proportional Counter (Figure-2.6)
The emitted X-rays were detected using a gas-flow proportional
counter (Model 04, J. E. Manson Co., Inc. ) directly coupled to a low
noise pre-amplifier (Manson Model PAL-01). The counter had a Nickel
plated brass body with cylindrical internal symmetry of 1.95 cm diameter
and length 7.3 cm. The 3 mm x 10 mm slotted window consisted of a VYNS
(a copolymer of 90% Vinyl chloride and 10% Vinyl acetate: C22H2302C19)
window of 28pg. cm 2 nominal thickness ('11 0.1 Nm) on a 60% open 400
lines per inch (ti 160 lines per cm) Nickel mesh. The transmission
characteristics of the window at low photon energies (Manson literature)
are shown in Figure 2.7. The X-rays transmitted through the window
were absorbed in the cylindrical cavity (maximum diametric gas path
35
A. Position of anode
wire
3mm x 10mm slotted V YN S window
EXX
Anode high voltage "
Output PAL-01 Pre-amplifier E
±15 and Gnd
B.
Window <--- Gas supply
-AAA4Aý[ Anode wire v ray
r-4 ý-
Pre -amplifier
Figure 2.6 The Manson Gas-flow Proportional Counter : A. Showing the 3 mm x 10 mm slotted VYNS
window. B. Cross-section through the centre of the
counter, showing the window, anode wire and gas path.
36
100
80
60 0 I::
Figure 2.7 The transmission of ultrasoft X-rays through the VYNS window (28 pg. cm 2 film on a 60 % open mesh). The position of some light element K-lines are indicated.
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Photon Energy, keV
37
length of 2.05 cm) by P10 (90% Argon and 10% Methane) gas at a flow
rate of 12.5 cm3. min-l. A bias voltage of +1700 V was used on the
central 50 um diameter stainless steel anode for all the X-rays except
Titanium K-shell (TiK) X-rays, for which, because of their higher energy
(4.5 keV), a bias voltage of +1600 V was used to produce less amplifi-
cation. The collecting area was defined by a1 mm diameter hole in a
Cadmium disc placed in front of the counter window. This also reduced
the X-ray intensity and hence the counting rate, avoiding pile-up.
Directly above the counter window was placed 6 pm of Polypropylene
(CH2 = CHCH3) which, when combined with the counter window, was found
to be sufficient to prevent backscattered protons entering the counter.
The thickness of Polypropylene used was determined experimentally by
observing the X-ray emission spectra from a Carbon target, when
bombarded with 500keV protons, for varying thicknesses of 'filter'
above the detector window. The optimum condition was taken to be the
point at which additional layers of filter served only to attenuate
the X-rays being produced with no effect on the background (Figure 2.8).
An alternative, magnetic method was used in the earlier stages of the
work (Figure 2.9). Using the expression for the magnetic field, B,
required to deflect charged particles of rest energy Wo and kinetic
energy T, through an arc of a circle with radius r (Rosenblatt, 19'68):
1
IT (T +2 Wo)] 2 2.1
300 r
where T and ö are in MeV, it was possible to estimate the field
strength required to deflect any scattered protons away from the
window. A simplified drawing is shown in Figure 2.10 (A). With the
magnet pole pieces 7 mm apart (Figure 2.9 (A)) and the steel shielding
(present to prevent the magnetic field affecting the operation of the
pre-amplifier) having a3 mm aperture, the maximum value of 'x' was
38
1
1
1
U)
O V
0 ö z
1
of Polypropylene
Figure 2.8 The effect of adding Polypropylene filters above the detector window. The spectrum from a Carbon target bombarded with 500 keV protons is shown with filters added in 2 pm increments from 0 to 10 pm of Polypropylene.
0 100 200 300 400 Channel no.
39
A.
ý7mm
Spac
B. Target
tt tt tI
; scattered , protons
Nt TS
-ff
r
out of page
X-rays
Proton be-am
Al collimator
UMNII-Steel shielding -Cd pinhole
Proportional counter
Figure 2.9 Magnetic deflection of scattered protons : A. Schematic diagram of the magnet (with
a magnetic field of 0.3 T between the pole pieces).
B. The experimental arrangement used.
Pole piece
40
A. r
---ý x-
For y»»x, y=I and a. tan'' xl Cyl
r . 180 ,1 Fa
B. 7mm E-r
Magnet pole Epiece p
--ý 5tnm(-
`ý Shield
3mm
2.2
Figure 2.10 Calculation of the magnetic field required to deflect the scattered protons away from the hole in the steel shield.
A. The simplifying assumptions used for the calculation of the radius of the arc of the circle through which the protons are deflected.
B. The maximum distance over which the protons are deflected (x).
41
taken as 5 mm, where a proton at the edge of the field just avoids the
edge of the aperture (Figure 2.10(B)). The distance between the magnet
and the shielding was 6 cm (_y). Using the approximations that the arc
length, Zwy, and ax tan-1 (x/y) in Equation 2.2 gave a value of r
0.7 m. When used in Equation 2.1 for 500 keV protons this then gave
a required magnetic field of 0.15 T.
Since the magnetic field used was 0.3 T, it should have been more
than sufficient to deflect even the most energetic backscattered protons.
This was not, however, found to be the case. The presence of the magnet
between the target and the detector proved to be largely ineffective in
reducing the amount of backscattered projectiles entering the counter.
There are three possible-reasons for this:
(i) It is possible that the protons reached the magnet pole pieces at
an angle other than parallel to the axis, and that therefore some protons
which were originally well off-course from the 3 mm aperture were curved
back to it.
(ii) In the preceding calculation it was assumed that the proton was
subjected to the field B throughout the distance y. In practice how-
ever, the field only extends for the 7 mm within the pole pieces, after
which the proton will continue on a tangential path to the curve it was
following. As a result the field required would be higher if the same
eventual deflection was to be achieved (Figure 2.11).
(iii) Buck, Wheatley and Feldman (1973) showed that for proton energies
ti 100 keV, a large fraction of the backscattered projectiles were
neutralised. Consequently any attempt to remove the recoil particles
from the X-ray flux by electrostatic (or magnetic) deflection techniques
would be expected to be only partially successful.
It was therefore concluded that an absorbing filter was necessary to
prevent back-scattered protons and Hydrogen atoms from entering the
proportional counter.
42
r i
I.
4 - X-ý
Path of a proton subjected to a magnetic field over a distance: (a)--t y
(b)--f--- Z
Figure 2.11 The correction to the magnitude of the required magnetic field due to the limited extent of the available field (z =7 um).
x' x tan a=-=z z ry
For xz =5 mm (i. e. xz =x)
z. xz MI= = 0.6mm WU
V
Assuming Zzz (Figure 5.10(A)) and using Equation 2.2, r=8.5 cm which, when substituted into Equation 2.1, gives B=1.2 T.
*- XZ -4
43
2.4.2 Associated Electronics (Figure 2.12)
The bias voltage to the counter anode was provided using a
Harwell 2000 Series E. H. T. Unit, Type 2147-2 with a variable output
from 0 to 3 kV. The Manson PAL-O1 pre-amplifier required a+ 15 V and
a- 15 V input which was obtained from a Harwell 2000 Power Supply
Unit, Type 2015-4.
The input stage of the pre-amplifier was a junction field-effect
transistor followed by quasi-Gaussian double differentiating and
integrating filters with time constants of 0.75 ps. The output pulse
was bipolar with the positive lobe leading. The output from the pre-
amplifier was fed into the input of a Norland IT-5300 Multichannel
Analyser (MCA) (Norland Instruments/Ino-Tech. Inc. ). Using the pulse
height analysis mode of the MCA the energy distribution of the output
from the pre-amplifier was obtained and recorded onto the GOULD PN6000
mini-computer.
44
r. 0
. r., 41 u v a)
w 0 Co t+
a
U) U
Ö
c) N
a) b N
U O y y
N
a (1) 41 U)
0 co
cd i+
"d
U " rl 1-i td E
C) U)
N I
N
a,
Li
45
CHAPTER 3
CALIBRATION OF THE VAN DE GRAAFF MACHINE
The potential at the high-voltage terminal of the Van de Graaff
accelerator was monitored using an in-built generating voltmeter. The
voltmeter assembly consisted of a motor-driven rotor, with four equally
spaced, 45° sectors cut out of it and a stator plate divided into
eight insulated 45° sectors. As the rotor rotates, it alternately
exposes the high voltage terminal to the stator and shields from it
each sector. Triangular wave alternating current (a. c. ) voltages are
thus electrostatically induced between adjacent sectors of the stator,
the voltages being directly proportional to the terminal voltage.
Four alternate stator sectors of the voltmeter are connected in parallel
to the a. c. terminal of a bridge-rectifier circuit, located on the
voltmeter housing, which is connected to the generator voltage meter
on the control console of the accelerator.
The generating voltmeter was calibrated using the (p, y) resonances
method (Hunt and Jones, 1953). A low-energy proton may be captured by
a nucleus which is thereby transformed into a residual nucleus in a
highly excited state (Figure 3.1). Instead of expelling the incident
particle again (scattering), the residual nucleus may decay to the
ground state and rid itself of the binding energy of the particle by
the emission of one or more gamma (y) rays. The capture of a proton
depends upon the energy being of the correct amount needed to enable
the residual nucleus to be formed in one of its excited states, and is
thus resonant in nature.
For the calibration of the Van de Graaff accelerator, resonant
proton energies were looked for in Fluorine and Aluminium. Strong
46
M( A+lY *)
on energy
M(A+1Y)
0
Figure 3.1 Proton capture : the target nucleus (AX)
captures a proton (rest-mass, M(p), kinetic energy, T(p)) to form a residual nucleus in an excited state, A+ ly*.
47
resonances have been observed at 340.4 keV and 483.1 keV in Fluorine
(Hornyak et at., 1950) and at 633 keV in Aluminium (Meyer, Venter and
Reitmann, 1975). For these cases the expected reactions are as given
below, with the available excitation energy, E. being calculated from:
EM (AX) +M (p) +T (p) _M (A+1 Y) 3.1
where M, the rest mass, was obtained from the Table of Isotopes, 1978,
and T (p) was the kinetic energy of the proton.
For Fluorine, the resonance reaction is given by
19F (P, Y) 20Ne 3.2
with M (19F) = 17697.032 MeV, M (20Ne) = 18622.977 MeV and M (p) =
938.280 MeV. Thus for a proton of energy T (p) = 340.4 keV, the
excitation energy was calculated to be 12.675 MeV, and for T (p) =
483.1 keV, E= 12.818 MeV. For Aluminium, the reaction is given by
27A1 (p, Y) 28Si 3.3
with M (27 Al) = 25133.333 MeV and M (28Si) = 26060.537 MeV giving an
excitation energy of 11.709 MeV for a 633 keV proton.
For both Fluorine and Aluminium the calculated excitation energies
are such that the energy levels of the residual nucleus are very close
together. For example, for the 20Ne nucleus there are over 125 energy
levels in the region of E= 11 to 23.5 MeV. As a result, an increase
of only 0.14 MeV in the energy of the protons interacting with the
Fluorine nuclei is sufficient to give rise to a second resonance
position.
The emission of y -rays'from thick (2,6 mm) targets of Aluminium
and Fluorine (in the form of a Calcium fluoride, CaF2 target) was
studied using a 3" diameter x 3" length (75 mm x 75 mm) cylindrical
Thallium activated Sodium iodide (NaI(Tl)) crystal detector in
conjunction with a Norland IT-5300 MCA. As is shown in Figure 3.2 the
48
ti
CD
N
9 0 w E
m C7 m
oý ýv oý
a
L() a
oº c
aý d Aý vA Nt
OV H
r. 0 M 41 Co Ei 1+ u r1 N(0
H G4 >lb cu Co 0 C) U
b0
.C Cl O Hrl
I+ q 41 q R1 N O
'O I oO O d 1-i u
.t P td C3 aý a E-4
Co O Rf .a
cSt a
Ki 0 e-i H
W
cd 'L3 1.1 C1 Gl 0
a) u > +-
b Co 4-+w > O 4-+ .. 41
cu oo > aý cu H r-I 41
cd }a _1 '-' C C O ýH p o0
ý Co N Cl z 0) ý,
ý 41 w b x: Co Co 0 NM GO '-I u td rl
.rA> b DC F+' 'r 4. + W -rl x
H. ä
Cn 0 -4 r1 ºr1 r-
N
M
V
ü_1 M
49
target was positioned at 45° to the proton beam with the detector at
900. In order to eliminate most of the background counts the NaI(Tl)
detector was encased in 5 cm Lead shielding. The accelerator
terminal voltage was varied in 10 kV increments around the expected
resonance positions (340.4 ± 0.5 and 483.1 ± 0.5 keV for19F, 633 ±1
keV for 27A1). The energy spectrum from the NaI(Tl) detector was
collected and, using the integrating facility on the MCA, the number
of emitted Y-ray counts measured (Figure 3.3) as a function of the
accelerating voltage. As shown in Figure 3.3, for the Fluorine
resonances it was possible to integrate a specific Y-ray peak which,
using the two 1-rays from a60Co source for calibration, was found at a
photon energy E1 ti 7 MeV. However, for the Aluminium target, no such
peak was observed - the 1-ray energy believed to be > 10 MeV (High
Voltage Ltd. literature) thus extending beyond the range of the 1024
channels of the MCA. However, for a 75 mm x 75 mm cylindrical NaI(Tl)
crystal a large proportion of the photon interactions would be
expected to be of the Compton effect kind, where only a portion of the
energy of the photon is deposited in the crystal thus giving rise to a
range of reduced height pulses which would appear on the MCA scale.
The resonance position for the Al target was therefore deduced by
integrating the total energy spectrum above channel 100 (thus eliminating
most of the low energy, accelerator voltage independent noise).
Having normalised all the measurements to the proton beam current,
thick target y-ray yield curves were obtained (Figures 3.4 and 3.5).
The resonance positions were identified at the point of maximum gradient
of the thick target yield curves versus accelerating voltage (Khan and
Potter, 1964):
19F(p, Y)
20Ne - 340 ± 10 kV
480 ± 10 kV
27Al (p, y) is Si - 640 ± 10 kV " 14,
?! äff ý``
50
kV independent noise
N 2.0 0 r
x u) 1.5 c 0
1.0 0 Position of Y-rays
from60Co source
-130 5 - . E z
ON I 0 200 400 600
Channel number 800 1000
Figure 3.3 An example of the energy spectrum obtained from the Ca12 target with an accelerating voltage of 350 kV. The y-ray counts were obtained by integrating the peak between the 'Start' and 'Stop' positions. An energy calibration based on the y-rays from a 60Co
source (at 1.17 and 1.33 MeV) gave the energy at channel 800 as z7 MeV (assuming a linear energy scale through the origin).
kV dependent peak
Stop Start
51
Jfl)$/SIUflO3) oI
`aonino i96VJe; )10! 43 J0 3ua! peiE)
0 LD ICJ -0
Ln a) G)
+' Fi co 0 U)
cv u O ý3 O +' Ln 3p
Co 10 Z O
O > aW
C o r1 44 4. J
O d Nä O u al
-v-4 Ö
E . ý. r N
O 0 >
4 Nv LU C! OO. L]
4J OO d0+- N ý+ D N4) rl
a) 0 x10 as 0 uvn M ý O H
b 0+ ý
cr Om N N
m U.
N0 Co Co 'e c4 O (zOi. x) e `anano 1e6aei )1oiy1'swunoo I(ei-A,
52
(n)Iis}unoo, pl. x) . `aAJn3 ; a6Je; 43141 10 ; uaipei! E) " 0)
O
°D ä3
O .. ý 40 >
N 41 `{ iý
0 ago 0 4- .v R4
C N 00 "e
C N i Ai O 41 Q) ä
"v r. >
aý b °0 c 4.4 rd rn 41 >' E cd p
ý-º 00 W O OUO 'ty
HO
> cu 4-3 j2 "0 41 0
P-4 N 4J . ýG ": O -rl UUN
rl i--I H bO W v HU
0 L
0 1n 0 to 0
ivoLx) " `anano ia6ie; 40141'slunoo Aea-A.
53
In order to calibrate the proton beam energy against the accelerating
voltage, as monitored by the generating voltmeter, the observed
resonance positions were compared with the known resonance energies as
given by Hornyak et al., 1950 and Meyer, Venter and Reitmann, 1975.
This information was then used to produce a calibration graph for the
actual proton beam energy in terms of the generating voltmeter reading
(Figure 3.6). The vertical error bars were smaller in magnitude than
the size of the plotted points.
From the three resonance positions observed it was seen that the
terminal voltage as measured by the generating voltmeter was within
t5 kV of the proton beam energy, thus lying within the experimental
accuracy of t 10 kV under which the major part of this work was
undertaken (Chapter 5).
54
701
060,
>I 50 rn
c 40
30 0
20 a
10
Figure 3.6 Calibration graph for the proton energy as a function of the generating voltmeter reading. The regression line is shown (y = O. 99x + 1.59) with a correlation coefficient, r=1.0.
.ý
0 100 200 300 400 500 600 700 Voltmeter reading, kV
55
CHAPTER 4
DATA ANALYSIS
After recording the energy distribution spectra of the emitted
ultrasoft X-rays on the GOULD mini-computer, the data were transferred
to the Amdahl 470 V/8 computer at the University of London Computer
Centre (U. L. C. C. ). Using the Amdahl, multiparameter least-squares
curve fitting was attempted using the CERN Program Library Routine
called "MINUIT" (James and Roos, 1975,1985).
A large class of problems in many different fields can be reduced
to the problem of finding the smallest value taken by a function of
one or more variable parameters by minimising the difference (chi-
square* X2) between the theory and the experimental data. This
difference is represented by the function F (X n) where n are the
unknown parameters. The function F (X n) need not be known analytically,
but may be specified by giving its value at any point X in the space of
the parameters.
MINUIT is a package of programmes that minimises a function of n
variables, computes the covarience matrix and finds the true errors
(James, 1978). This system of programmes incorporates three different
minimisation methods,: 'each of which may be used individually, or in
combination with the others, depending on the behaviour of the function
and the requirements of the user. The three minimisation subroutines
available are: -
(i) 'SEEK' -a Monte-Carlo searching subroutine.
(ii) 'SIMPLX' -a subroutine employing the simplex method of
Neider and Mead (1965).
(iii) 'MIGRAD' -a minimisation subroutine based on a variable
metric method (Fletcher, 1970).
56
MINUIT is generally used to minimise a X2 function or a negative log
likelihood function. It is assumed that the point where the function F (X)
takes on its lowest value, FMIN, determines the most likely or best-fit
parameter values, and that the region over which the function takes on
values smaller than (FMIN + UP) corresponds to a confidence interval for
which the confidence level is determined by the value of the positive
constant UP. For example, if F (Xn) is a X2 function of one variable
parameter (n = 1), a value of UP = 1.0 determines 'one-standard-deviation'
errors corresponding to a confidence level of 68%.
For the analysis of the X-ray energy spectra, only the subroutine
SIMPLX was used. This is a relatively fast method even when far from the
minimum but will also converge to the exact minimum. It does not compute
the covarictnce matrix but gives order-of-magnitude estimates of its dia-
gonal elements - the parameter errors (James, 1972).
4.1 Subroutine SIMPLX
In the minimisation method of Neider and Mead (1965) the information
about the function F (X1, ----, n) consists of its values at n+1 points,
forming a simplex. A simplex is the smallest n- dimensional geometrical
figure with n+1 vertices :a triangle for n=2, a tetrahedron for
n=3, etc. How the method works can be shown by considering a two dimen-
sional case as in Figure 4.1.
The three starting simplex points (P1, P2 and P3) are supplied by the
user and the function F (X n)
is evaluated at each point. Let the point
PH be that at which the function value is highest (worst) and PL that at
which it is lowest. Let P be the centre-of-mass of all points in the
simplex except PH , that is
n i=n+1 {E pz , pH } 4.1
i=1
57
X2
Pý*
Pz
L
Xl
Figure 4.1 A two dimensional simplex formed from P19 P2 and P3 (refer to text for detail).
P, = PH
58
From the original simplex, a new simplex is formed by' replacing Pff with
a better point if possible. The first attempt to find a better point
is made by reflecting PH with respect to P, producing P* =P+ (P - PH).
If F (P*) <F (PL), a new point is tried at P** =P+2 (P - PH), but
if F (P*) >F (PH) then a new point is tried at P** =P-2 (P - PH).
The best of the new points then replaces PH in the simplex for the
next step, unless neither of them is better than PH in which case a
whole new simplex is formed around PL, with dimensions reduced by a
factor of 0.5.
A convenient convergence criterion for the simplex method is
based on the difference F (PH) -F (PL). The iterations are stopped
when this difference is less than a pre-set value. As a final step,
the function is evaluated at P, which is often slightly better than F (PL
MINUIT is designed to handle any function, F (X n
), and hence the
definition of the function to be minimised must be supplied separately.
This is done through the subroutine "FCN" in which F (n) is calculated
and which is therefore supplied by the user.
4.2 Subroutine FCN (Appendix 1)
The energy distributions of the characteristic X-rays, as detected
by the proportional counter, were found to be well fitted by a linear
combination of Gaussian curves super-imposed on a constant background -
one Gaussian curve describing the energy distribution for each target
characteristic X-ray energy.
If the area under a Gaussian curve is 'A. ' then the distribution
function is defined as:
59
2
Q/2, Tr . -XP {-2 (x Q u) } 4.2
(Bevington, 1969)
where x is the value of a random observation,
u is the mean value of the parent distribution
and a is the standard deviation of the distribution.
The standard deviation, a, can be defined in terms of the full-width at
half-maximum (F. W. H. M. ), r, of the curve (Bevington, 1969):
r=2.354 a 4.3
and substituting this into Equation 4.2 gives
PG (x, , r, A) =
0. ý4. A . exp {-2.77 (x T u)2 } 4.4
The subroutine FCN was written to include a possible three
Gaussian distributions with separate u, r and A combined with a
constant background, C. Thus the complete distribution, 1theor
as described in FCN is given by:
Y theor = 0.94. A1
.e{-2.77 x- u1 2
{}} r1 p1
+ 0.94. A2
. exp {-2.77 2
{x- u2
}} r2 r2
+ 0.94. A3
. exp {-2.77 (x ru2
3+c4.5
33
where the variable parameters were c, AZ, u2 and r2 (i = 1,2,3)
with x, the MCA channel number, being directly proportional to the
photon energy.
Having calculated the theoretical spectrum obtained above
('theor)' FCN proceeded to calculate the X2 using the definition of
James (1972):
2K Yk - Tk(X) 2 X (X) =Ei6)4.6
k=1 k
where yk and Qk are measured values and errors, and Tk (X) are the
values predicted by the model, depending on some parameter(s), X.
60
In the case of the X-ray energy spectra considered in this work,
Equation 4.6 became:
x_ NPTS
( yobs (x)
- ytheor (X) 2 4.7
x=1 yobs (x)
where NPTS represented the number of channels considered on the
MCA and yobs fix) was the observed X-ray energy spectrum as collected
on the MCA. The error in the observed values was taken to be given by
i the square root of the value, (yobs (x))2.
The function F (X n)
required by MINUIT for minimisation was
defined in the subroutine FCN to be the X2 as given in Equation 4.7 with
ytheor (x) being defined by Equation 4.5. Before starting the iterative
process of minimising F (n), MINUIT requires an initial estimate of the
parameter values (c, Ai, u2 and r2, i=1,2,3) to facilitate the loca-
tion of the minimum. The choice of initial parameter values were based
on estimates taken from the MCA as the spectra were collected.
Once MINUIT had located the minimum value of the function Fý( n
and hence had calculated the best fitted curve to the experimental data,
the calculated values for each parameter were returned to the user. Of
particular interest were the mean energy of the photon - given by the
mean value (u) of the distribution, the width (r) of the distribution
and the intensity (A) produced for a given proton charge onto the
target.
61
CHAPTER 5
EXPERIMENTAL SECTION
5.1 Experimental Procedures .
The characterisation of proton-induced ultrasoft X-rays in this
study comprised three. experimental investigations:
(i) a study of target surface contamination and the way it
effects the intensity of emitted X-rays,
(ii) the angular distribution of the X-rays resulting
from target absorption and attenuation effects, and
(iii) the dependence of X-ray yield on proton energy.
5.1.1 Surface contamination study
With the target and proportional counter at 45° and 90° to the
beam respectively, the proton beam was run continuously onto a target
with an X-ray energy distribution spectrum being recorded for each
25 pC increment of charge collected on the target as measured by the
charge integrator. This was done for both the Aluminium and Carbon
targets, with the proton beam energy being kept constant at 500 t 10 keV
and the current kept, at 35 t1 nA. Each spectrum took - 12 min to
collect and the beam was run onto the target for a total time of 8 hours
in the case of Carbon and 17 hours for Aluminium. In the latter case
there was an interval of about 12 hours after the first 9 hours, during
which the beam was off.
For the Aluminium target, the dependence of the rate of contami-
nation build-up on beam current was also considered. Keeping the proton
beam energy at 500 ± 10 keV, spectra were collected for beam currents of
10,40 and 70 nA. Spectra were also recorded for a beam current of
50 nA but with a reduced proton energy of 300 ± 10 keV. All the spectra
were collected for a constant charge of 25 pC on the target.
62
5.1.2 Angular distribution study
To study the angular distribution of the emitted X-rays the proton
beam was run onto the target for a total collected charge of 25 pC
(taking approximately 500s). The proton beam energy was kept at
500 t 10 keV with a beam current of - 50 nA. For each target angle (6)
X-ray spectra were recorded over a full range of detector angles
(40° a5 130°), in 10° steps. Four complete sets of data were
taken for each target element, with 0= 30°, 45° and 60°, and the data
combined to give average yields. A clean target surface was used for
each complete set, cleaning being achieved using a combination of
abrasion with fine emery cloth, then soaking the target in one molar
solution of Hydrochloric acid and finally immersing it in an ultra-
sound bath for about three hours.
The starting angle with which a given set of data was taken was
varied between experiments in order to eliminate any possible effect
of the order in which data were collected. In addition to the full
scan of detector angles for each target angle, data were also obtained
for a= 900 at the start and end of each target angle scan so as to
provide information on any surface contamination build-up (Section 6.1).
5.1.3 Proton energy dependence of X-ray emission
Using the Aluminium and Carbon targets, the effect of varying the
proton beam energy on the yield of X-rays was studied. With the target
and detector at 45° and 90° to the beam respectively, spectra were
collected for 25 pC of collected charge on the target. The proton
beam energy was varied between 225 t 10 keV and 700 t 10 keV, the beam
current being maintained at 35 ±1 nA throughout.
5.2 Results
Within the text of this, and following chapters, reference is made
to the characteristic X-ray energies. The values given are taken from
63
the X-Ray Data Booklet (1986) and, where the energies of the lines differ,
the energy of the K line is used. However, the compilation of Storm and a1
Israel (1970) show that the weighted average of the K-shell X-ray energies
varies less than 1% from the energy of the K line, even for the highest al
atomic number element considered for this work (Titanium, Z= 22). The
characteristic X-rays are denoted as 'As' where A represents the chemical
symbol of the target element and s denotes the electron shell (K, L, M).
For example, AlK represents the K-shell characteristic X-ray of Aluminium.
The characteristic energies studied in this work are given in Table 5.1
(with the Aur1 X-ray energy taken from 'Table of X-ray Emission Energies').
Having collected the X-ray energy distribution spectra on the
MCA the data were recorded on computer and transferred to the Amdahl
470V/8 computer at U. L. C. C. where MINUIT (Chapter 4) was used to fit
Gaussian curves to the data. The output from MINUIT came in the form of
a plotted spectrum with both experimental data and theoretical fit shown.
Also given were the theoretical positions, F. W. H. M. and area under the
peaks (with an estimate of the uncertainties), the background level and
the average value of X2 (Equation 4.7) per channel. This latter value
gave a measure of the goodness of the theoretical fit and was found to be
2 per channel in all cases. The error on each of the variable para-
meters was less than 1%.
Shown in Figure 5.1 ((i) - (vii)) are examples of the spectra
obtained (both experimental and theoretical) for each target. Figure 5.1
((i) - (vi)) were obtained with 1700V on the proportional counter anode
(Section 2.4.1) whereas Figure 5.1 (vii) was obtained with an anode
voltage of 1600V. Also shown is the variation between different
qualities of Aluminium (Figure 5.1 (iii) A and B), where the 'pure'
Aluminium spectrum gives a far better fit between theory and experi-
mental data in the region of channels 200 - 400 than the 'Dural'
Aluminium alloy. Over these 200 channels the average X2 value for the
64
Characteristic Energy X-ray (keV)
BK 0.183
CK 0.278
NK 0.392
A1K 1.487
SiK 1.740
AuM 2.120
TiK 4.511
Table 5.1 The energies of the characteristic
X-rays studied.
65'
Figure 5.1 Examples of the spectra obtained from each of the targets when bombarded with 500 keV protons:
(i. ) TiB2 target, 1700 V on the counter anode wire (Peak at channel 75)
(ii. ) C target, 1700 V on the counter anode wire (Peak at channel 91)
(iii. A) Dural Al target, 1700 V on the counter anode wire (Peaks at channels 90 and 585)
(iii. B) Pure Al target, 1700 V on the counter anode wire (Peaks at channels 90 and 580)
(iv. ) SiC target, 1700 V on the counter anode wire (Peaks at channels 90 and 690)
(v. ) Si/N target, 1700 V on the counter anode wire (Peaks at channels 90 and 690)
(vi. ) Au target, 1700 V on the counter anode wire (Peaks at channels 90 and 850)
(vii. ) TiB2 target, 1600 V on the counter anode wire (Peak at channel 690).
4
i. 11
N1 C
O 0
%6- 0
ö1 Z
ii. 1
1
N rr C
O
0 0
z
1
66
Channel no.
100 200 300 Channel no.
67
iii.
1
U)
O V
y_1 0
0 z
1
0 200 400 600 800 1000 Channel no.
Iil. B,
1
N rr
7 O V
v- O
0 z 10
10 200 400 600 800 1000 Channel no.
.. " "" ."
68
iv.
N
Cý 7 O 0
O
O z
V. 1
N
c1 0
'I. - 0 ö z
1
0 200 400 600 UUO iuuu Channel no.
0 200 400 600 800 1000 Channel no.
69
vi.
1
Cl) ++
c1 0
O
O Z
1
vii. 1
1
Cl) C
O V
ö1 ö z
1
0 200 400 600 800 1000 Channel no.
'0 200 400 600 800 1000 Channel no.
70
Dural spectrum is 10 per channel compared with a X2 of 2 per channel for
the pure Aluminium spectrum. Dural Aluminium comprises 95% Al, 4% Cu
and 1% Mg (Goodfellow Metals Ltd. ) with both impurities having
characteristic X-rays slightly lower in energy than the A1K X-ray at
1.487 keV (MgK X-ray at 1.254 keV and CUL X-ray at 0.93 keV). It is
probable that the X-rays from the impurities in the Dural are therefore
responsible for the increased X2 in the region below the A1K X-ray peak.
The values obtained from MINUIT for the peak position and F. W. H. M.
for each target are summarised in Table 5.2. Whilst the error on each
individual parameter as given by MINUIT is -1%, the errors shown in
Table 5.2 are the standard deviation (s. d. ) of the averaged parameter
values taken over all the collected spectra for each target (about 140
in total). The deviations from the average values arose primarily as
a result of slight fluctuations in the gain of the proportional counter,
particularly when spectra were taken on different days with possible
small deviations around the optimal anode voltage of 1700V. The widths
of the peaks, as given in Table 5.2, compare favourably with the speci-
fied performance of the proportional counter (Manson literature).
For example, the F. W. H. M. is expected to be 115% of the peak energy
for the BK peak and 36% of the peak energy for a MgK peak (1.254 keV).
There is also agreement between the width of the Aluminium peak as
given in Table 5.2 with that measured for a similar proportional counter
by Hoshi et at. (1985) where the F. W. H. M. of the AlK peak was about 30%
of the peak energy. Hoshi et at. (1985) showed that the spectral resolu-
tion of the proportional counter is limited almost entirely by the
statistics of the small numbers of ionisations within the counter.
For the spectra taken with an anode voltage of 1700V it can be
seen that a low energy peak (at about channel 90) was present in all
cases. This has been shown (Section 5.2.1) to be the result of Carbon
contamination of the target surface, resulting in CK X-rays being
71
G 0
". -I
3 a. wx
a
N
Ox, Z. 7'.
Y3 W
ýý. 0.
ON x ". 4 g2 Co 41 () ", 4 a Cn
O R+'-
91. 0 41
" 0'. - 3
V-
0 z
W CL
P4 t)
O 0) x"r4 cd 4J
ß4 y .L 0 ,2 924
1- 0) 4-j "a 0)
v rn L f0
F-
1 CV ON C'V %M p M t'7 cl c'1 N cn
O tý co n co r-
+1 +1 +1 +1 +1 +1 -T ^ Oh N O ºr1
- C14 C14 c N N N N
r" N 0 00 uy ý- N "-- (V C14
1I +1 +1 +1 +1 +I +1
-cr 0 cV 0 00 N QO 00 O. ON -I7 ON In Il .O .o 00 "o
OO00- 00 0 N ON O\ ON O\ 00 ON N
-I7 N (V cV N N Výp +1 +1 +1 +1 +I +1 -H +1 O N CV 0 cV ON O 0 ON 00 00 Co co r- co M
[- u1 In In -7 In .o c+1
+1 +1 +i +1 +1 +i +1 +1 In "- r- M 0 0 Co In tý ON OO Co O'. 01 00 cV
00000000 00000000
r4 r-A
N PQ H 0 v 0)
w ä
u ., 4
v)
z :i pl
H
a
w 0 U) Ei s"+ a)
Ha
O
W
"d
rl cd
O
+I
4J '
i. ý
Co
cli
0
rr
O
00
fa 41
ON .ýN 1J
Ei 4-4 O O
.+N
Cl1 U
N
In
t0
F-
72
emitted along with the characteristic X-rays of the target. The AlK,
SiK and Aum X-rays were well resolved from the contaminating C. X-rays
but this was not the case for the; Boron- and Nitrogen-containing
targets.
For the TiB2 target, with a 1700V detector anode voltage, the
low energy X-ray peak was observed at a lower position than in any of
the other examples at 1700V (channel 75 t7 as opposed to channel
90 t 5). Since the BK X-ray energy is 183 eV and the CK X-ray energy
is 278 eV it was concluded that the peak observed from the TO2
target was a combination of Boron from the target and Carbon from the
surface contamination. This conclusion is supported by the increased
width of the combined peak (90 ±4 channels rather than 81 ±2 channels).
Since the rate of Carbon build-up was found to be constant for a given
current in the surface contamination study (Figure 5.8) an estimation
of the BK X-ray yield could be found by subtracting the CK X-ray yield
from the Aluminium target data (assuming that the Carbon contamination
rate does not depend on target material) from the combined BK and CK
X-ray peak of the TiB2 target (Figure 5.2).
Whilst a similar effect might have been expected for the Si/N
target (with a low energy peak slightly higher in position than channel
90, consisting of a combination of CK and NK ( 392 eV) X-rays), this was
not found to be the case. As is shown in Table 5.2 there was no observ-
able difference between the low energy peaks of the Si/N target and
the SiC, Al or Au targets. Since there was no reason to suspect that
the NK X-rays were not being produced from the target it was concluded
that the lack of observable NK X-rays was the result of their increased
attenuation in the Polypropylene filter (Section 2.4.1) since there was
no significant difference in the attenuation of CK and NK X-rays in the
VYNS window (Figure 2.7). Using the mass-attenuation coefficients of
Hubbell (1977), the transmission characteristics of 6 pm of Polypropylene
73
1
1 N 4.0 C
O v
O
0 z
Figure 5.2 Derivation of the BK X-ray intesity:
Spectrum from TiB2 target (as fitted by MINUIT).
------- CK X-ray peak derived from the Al target.
""""""" BK X-ray peak calculated by subtracting the CK X-ray peak away from the TiB2 target spectrum.
0 100 200 300 C hannel no.
74
CO I-
d' r-
N I-
o> T C)
Y
CO W
C W
(O 6
it 6
N. 6
CO
'1"4
G)
0
a1
aý v
o cd
o r4 P4 -+ F1 I. -
U) I
1.0 x wz 0
vi uca 1J 0 N
v-1 Q)
d 1j 41 uW co
Cli 4O u Vr4
J. 1
O U) . r-1 0 Un a U)
W V. Cd 60 I r. 41 , -1
(U O 44 H ul
M
LL
0pp0 O Co pl
UOISSIWSUUJ; %
75
were calculated (Figure 5.3). It is seen that whilst Polypropylene will
transmit 35% of the CK X-rays, less than 1% of the NK X-rays will be
transmitted. It is not surprising therefore that the NK X-rays were
not seen by the proportional counter.
With an anode voltage of 1600V the TiK X-rays from the TiB2
target were observed (Figure 5.1 (vii)). In this case the BK X-rays
from the target and the contaminant CK X-rays appeared as a narrow peak,
high in counts, at the low energy edge of the spectrum. Whilst the
overall MINUIT fit gives X2 = 1.7 per channel, there is a region
(channels 100 - 300) where the theoretical curve underestimates the
experimental data. This may be a result of the lower energy peak being
calculated to be narrower than it actually is, or a result of small
amounts of other X-ray energies being present, such as the L-shell
Titanium X-rays (- 0.5 keV).
All the spectra shown in Figure 5.1 ((1) - (vii)) have a constant
background level ('c' of Equation 4.5) calculated to be 10. The main
contribution to this background arose from electronic noise within the
detection system itself - collecting a spectrum for 500s with no proton
beam but with an anode voltage of 1700V on the detector gave an average
of 8t3 counts per channel over the 1024 channels of the MCA.
From the data summarised in Table 5.2 (columns 3,6) it was possible
to draw an energy calibration graph for the proportional counter and
MCA (Figure 5.4). All the data (at 1700V counter anode voltage) were
found to lie on a straight line (with a linear correlation coefficient,
r=1.0), including the deduced BK X-ray position.
Using the values obtained from MINUIT for the areas under the
Gaussian peaks to represent the number of X-rays detected by the
proportional counter, an estimate of the absolute yields of X-rays could
be derived. The number of X-rays produced per steradian per proton onto
the target, NX, was calculated using (modified from Hoshi et at., 1985):
76
to
N
0
N
to
r
d . 5t
rn
w
N
o
o . r_ -H 41 as P44 00 Pl. o 4o a Y+ aj a ,4 4N
.No oa
44 a 41
aao a0 $4 -ri bo 3
0 r. 4 O
. ri {ý w
ý+ a `i- to
Rf N u
ýU a bo ge i
W co N
U)
`% a%
C) lot
a 'ý co W ot C%4
IOU IGUU843
77
Nx _ 1.6 x 10-13. cx 5.1 Q. T. e. (1 - T)
The number of X-rays detected per pC of charge on the target, CX was
obtained from the area under the photon peak as calculated by MINUIT,
correcting for surface contamination effects (Section 5.2.1). The
solid angle, Q, subtended by the area of the pinhole aperture was
given by irrt 5.2 R2
where r, the pinhole radius, = 0.5 mm and R, the target-to-pinhole
distance, = 15 cm. T represents the proportion of the emitted photons
transmitted through the counter window and the Polypropylene filter at
the different X-ray energies and was calculated from the product of the
window (Manson literature) and the filter (Hubbell, 1977) transmission
coefficients, the low energy region of which are shown in Figures 2.7
and 5.3 respectively. The detection efficiency, c, of the counter was
calculated from X-ray absorption coefficients (p/p) for P10 gas (Handbook
of Chemistry and Physics, 1987-88) using a gas density, p, of 1.677 x
10-3g. cm -3 and the maximum gas path length, x=2.05 cm (Section 2.4.1).
A combined pressure and temperature correction of 1.09 was applied to
account for ambient pressure (9.87 x 104 Pa) and temperature (18°C), thus
giving u1 e= 1- exp-l ppx1.09
5.3
The dead time, T, was kept at < 3% of the counting time throughout.
In order to compare the calculated yields with those available in
the literature, the number of X-rays detected at 900 to the beam when
the target was at 45° to the beam was considered. Using Equation 5.1,
the number of X-rays produced was calculated for each target element
(Table 5.3).
However these calculations represent only the number of X-rays
in the direction of the counter for one particular target angle since
no corrections are included for target self-absorption effects. The
78
Target Photon Energy No. of photons Nx (± 15%) per 25 pC on (photons per
(keV) the target proton per steradian)
B 0.183 45404 4.5 x 10-4
C 0.278 700491 1.2 x 10-3
Al 1.487 155669 7.6 x 10 5
Si 1.740 75584 3.5 x 10-5
Au 2.120 34619 2.0 x 10-5
Ti 4.511 6933 9.2 x 10-6
Table 5.3 The observed yield of X-rays at 90° to the
proton beam for a target angle of 45°.
(The uncertainty in the value of Nx is the
result of the combination of the experimental
uncertainties in the parameters involved. )
79
absolute X-ray yield per proton would require integration over the
complete sphere taking into account the variation in absorption length
within the target at all angles. However an estimate may be obtained
by assuming a constant yield at each angle such that the total X-ray
yield is given by 4iNX, which may then be compared with the yields
given in the literature (Table 5.4). For the cases where direct com-
parison can be drawn (that is, where the proton energies are the same)
it is seen that the yields calculated in this study are approximately
a factor of two less than the documented values, but there is insuf-
ficient information on these calculations to enable the identification
of a possible systematic reason for this.
In calculating NX (Equation 5.1) the anode wire of the proportional
counter was assumed to be infinitely thin. In practice the 50 um dia-
meter central wire would be expected to reduce the detector efficiency.
However, as this effect is more noticeable for small pinholes it was
assumed to be negligible for the 1 mm diameter pinhole used in this
(study (Hoshi et at., 1985 for example, calculated that the required
correction when detecting AlK X-rays was 0.89 for a 30 pm pinhole and
0.92 for a 200 pm pinhole).
In the following sections (5.2.1 - 5.2.3 and Chapter 6) reference
is frequently made to the 'intensity' of the X-rays. Whilst intensity is
normally used to define the rate of energy transfer per unit area normal
to the direction of propagation, in the context of this work it is used
to describe the integrated number of counts under the peak.
5.2.1 Surface contamination study
It has already been shown (Figure 5.1) that continuous bombardment
of a target by a proton beam results in the production of not only the
target's characteristic X-rays but also a low energy photon similar in
energy to the CK X-ray. This was assumed to be caused primarily by
80
C) U C C)
U
w
-. -ý o
v 94 G) o
C) to
o o 0
ýsz,
"ß 1 4.1
0\0 Ci , 94
> f1' LX G)Z
o t= 44 --r
0 x aý
OD n ý'' w Wa
u1 u1 ýlo N
CYN
w w
r""1 ý7 r-i 4-4
ý n 3 3 c
10 . 10 10
41 , p4 %0 41 Co 4. ) CYN 41 Z
v r- o
w xw o
ce Q) cu P.
20 ti
N MMM C ýO ýOýOýO
DC >C DC DC DC M 1ý. %D N
OO', O
NN. --
N ý7 - 1 O
1 O
1 O
I- X
In In In
rn
.ý i. % Co
i-1
U ni
Rf . C. N Uy
' C) Ol UU
C7 0
1b s~ a) O +º 41 C O +-1
U)
a U) wm ow U bo aý ý, x vo Q) o
aý n ý Wa 0 qci
to
b 0) C)
0 o Wo 0 P.
+- H
U) N 3ý
cl
.a1 x O +) v) W
"ý 0 iI 0 cd cd
1J o r+ 0
00 rý 0) . 4.4 L. rý OD O 01? -
_C - º11
C ý- N 0.. WO ý- -7 F-
81
a deposition of Carbon containing contaminant on the target surface due
to the decomposition of organic vapours from the diffusion pump oil
(Khan, Potter and Worley, 1965). However, work performed by Gaines
(1981) has shown that hydrocarbon build-up occurs on target surfaces
even at the very low pressures (< 10-6 Pa) achieved by turbo-molecular
pumps, thereby suggesting that all the contamination may not have been
due to the deposition of the pump oil vapours.
Since the X-ray line observed from the contaminant was indis-
tinguishable in energy from CK X-rays, it would indicate that the surface
contamination is Carbon being deposited on the target's surface. This
assumption was borne out by the observed effect of continuous proton
bombardment onto the Aluminium and Carbon targets. In the case of the
Carbon target, the intensity of the CK X-rays remained constant (t 1%)
throughout 8 hours of bombardment at 35 nA (t 1 nA) beam current
(Figure 5.5). However, the AlK X-rays showed a reduction in intensity
with time (Figure 5.6) accompanied by a simultaneous increase in the
intensity of the low energy X-ray.
The observed effect of continuous proton bombardment on the Alumi-
nium and Carbon targets can be explained if the contaminant was indeed
Carbon. A reduction in the intensity of AlK X-rays would then be expected
as a result of absorption in the Carbon layer. However, no effect would
be expected for the Carbon target since any surface deposition would
only serve to increase the target thickness somewhat - the absorption of
the CK X-rays in the contamination layer would be equivalent to the
self-absorption in the target itself.
The decrease in the A1. K X-ray intensity was found to be exponential
with time, having a characteristic exponent of = 37 hrs-1 at a beam
current of 35 nA. The ratio of the intensity of A1K X-rays to that of
the surface CK X-rays was found to fall from 15 to 0.3 over a total
proton bombardment time of 17 hours (Figure 5.7).
82
8
6
N 4+ C 7 O V
4 0 r-
`w +1 ýN
C
2 C
Y U
Figure 5.5 The variation of the CK X-ray yield with collected charge on the target for 500 keV protons.
0 500 1000 Collected charge on
target, ýC
83
w a) 4-+ 4) 4) 4+
,. + 00 ci a i
'-4 O OH u 4' O 0
N Cl) 41 ril
3 41 Cd 'd o3 r
o 0) cd :3 Po
In O4 O P'4 ONN u1v "--ý Di
ý.. ý4 is co tt Cu -O
1 V-4 G) 4-1
.0O C °1J41
0 C O ,w CO , r1 p 1 04 , S. N
3 L G aý
G cfl O44.1 -+
" H a-)
v O cd rq r4 0 9: 4
"ý º4 . 1-1 4J
N > 6600 ä
co
O V CD Cd (1) ö 1
. . 4 .1 Ln HU 41 4-4
C ci
tD
LL s; unoo Sotx 'ABisuejui IiIV
84
10
0 Co
..
U, t d
C
Y U
0.1 L 0
Figure 5.7 The A1K : CK X-ray intensity ratio as a function
of the charge collected on the target for 500 keV
protons bombarding an Al target. I denotes the point at which the accelerator was switched off for a time interval of z 12 hours.
500 1000 1500 2000 Collected charge on target, uC
85
The effect of varying the proton current on the rate of Carbon
build-up was also studied. In Figure 5.8 the intantty of the CK X-rays
emitted from an initially 'clean' Aluminium target (Section 5.1.2) is
presented as a function of the collected charge on the target at
varying currents. It is clear that the gradients for the three 500 keV
proton beam currents (10,40 and 70 nA) are different, implying that
the surface deposition was current related. It would also appear that
the highest beam currents produced the lowest rate of Carbon build-up
and that varying the proton energy had very little effect on the rate
of Carbon build-up (Figure 5.9). It is difficult to extrapolate this
effect to large currents to find conditions under which Carbon contami-
nation is reduced to an acceptable level. Clearly however, there is
always the possibility that Carbon build-up takes place even at
currents of the order of mA.
Having established that Carbon was deposited on the target surface
it was possible to correct for the absorption of the target X-rays in
the contaminant layer (Section 6.1). The intensities referred to in
the following sections have been corrected for absorption in the
Carbon layer.
5.2.2 Angular distribution study
For each target, four sets of data were taken - each set
consisting of a full detector angle scan (40° .5aä 130°) at target
angles, 9, of 30°, 45° and 60°. In the latter case (0 = 60°), the
detector angle was limited to aä 110° since at a= 120° the detector
was in line with the target edge resulting in an intesity reduction owing
to the restricted view of the target by the detector. After compensa-
ting for absorption in the surface Carbon layer the four sets of data
were combined to give average values (t s. d. ) of the X-ray intensity at
each target/detector angle combination. In order to compare the angular
86
10
8
tos C
O V
r x
.. 4
C O r. ý C
02
Figure 5.8 The build-up of Carbon contamination X-ray intensity with the total amount of charge collected-on an Al target.
o: 500 keV protons at 10 nA O: 500 keV protons at 40 nA ": 500 keV protons at 70 nA A: 300 keV protons at 50 nA
OR
0 100 200 300 Collected charge on target, »C
87
401
C. )
N 301
0
a
'. 320 .o C O
c0 U
0 10 0 t0
cr.
Figure 5.9 The rate of Carbon contamination build-up as a function of the proton beam current.
" 500 keV protons 300 keV protons
ýO 10 20 30 40 50 60 70 80
Proton beam current, nA
88
distribution of intensities with theoretical expectations (Section 6.2)
the intensities were normalised to the intensity at e= 300, a= 60° -
this being the maximum theoretical intensity.
The resulting distributions are shown in Figure 5.10 ((i) - (vi))
where the experimental data are given along with the theoretical cal-
culations (details of which are given in Chapter 6). Also shown in
Figure 5.10 are the calculated'angular distributions assuming that the
X-rays are produced only in the first 1 pm of the proton's path in the
target (for example, Table 6.7). The errors on the theoretical values
are encompassed within the data symbols used on the figures. In the case
of Titanium (Figure 5.10 (vi)) there was insufficient difference between
the calculations based on the assumption that X-rays are produced over
the whole of the proton range and those based on production in the first
1 pm only, to warrant their separate plotting.
From the results shown in Figure 5.10 it was deduced that for K-shell
X-rays the degree of angular variation showed an approximate inverse
proportionality to the X-ray photon energy, with the X-ray intensity
being constant over a far wider range of target/detector angles for the
TiK X-rays (4.5 keV) than the CK (278 eV) or BK (183 eV) X-rays. It was
also observed that the degree of agreement between the experimental data
and the theoretical calculation improved with increasing photon energy.
However, this would appear not to be the case when considering the AuM
X-rays (2.12 keV) where it might have been expected that the agreement
between experimental and theoretical data for the AuM X-rays would be
similar to that of the SiK X-rays (1.7 keV). In addition, the degree
of angular dependence observed for AuM X-rays (Figure 5.10 (v)) is
larger than that for K-shell X-rays at approximately the same energy
(Figure 5.10 (iii) and (iv)). This may be due to the self-attenuation
coefficient of Gold for its M-shell X-rays being nearer in magnitude
to the self-attenuation coefficient of low energy X-rays, such as the
89
Figure 5.10 The relative X-ray intensity as a function of the detector angle, a, for target angles 0= 30°, 45° and 60°.
O Experimental data
f Theoretical calculation (X-rays produced in the total proton range)
" Theoretical calculation (X-rays produced in the first 1 pm of the proton's range only)
(A calculated point is not shown where it lies
within the experimental error bars)
(i. ) Boron K-shell X-rays (183 eV) (ii. ) Carbon K-shell X-rays (278 eV)
(iii. ) Aluminium K-shell X-rays (1.487 keV) (iv. ) Silicon K-shell X-rays (1.740 keV)
(v. ) Gold M-shell X-rays (2.120 keV) (vi. ) Titanium K-shell X-rays (4.509 keV)
90
gcW 45' I.
c'S
Y £, G1 ".
N
ä; °C 0
40 80 120 40 80 12
Detector angle, a
0:
Ii 30" 45"
.
c "
"5 "
Y "ý
ci
d
0 40 80 120 40 80 120 Detector angle, a
ý0 III
" 30" 45
. 1 Qoooo Z
10 ý"
" I ý
4 q
1 i1
C y 1
r C
"5 Q "
0 40 80 120 40 80 120
60
""
.¢
" .
60*
" A
.
1
40 80 1
60
ituý
i
Detector angle, cc
91
p 30'
iv.
N C d
C
ac "5
m
m 0:
0 40 80 120
p 30*
V.
y"1
a:
d
0 40 80 120
0 30* VI.
4. .N c 0
c "E "5 Y
a>
45.
1
A,
to 80 120 Detector angle, «
45'
"
40 80 120 Detector angle, c
45
ý¢ýQ4-". ýýý¢
60.
o°'Pk9ýý
40 80
60'
ýýýý :. ¢ý .. .ý
.ý ýý
60
01 aÖ 8Ö 120 40 80 120 40 80 12 Detector angle , cr
92
CK X-rays in a Carbon target, than for higher energies such as SiK X-rays
in a Silicon target (Figure 5.11). A further anisotropy in the emission
of M-shell X-rays might also be expected because of the angular momenta
change which occur in the system upon the emission of M X-rays.
5.2.3 Proton energy dependence of X-ray emission
The variation in X-ray intensity with proton energy (at constant
current and for a fixed amount of charge onto the target) was measured
for the Aluminium and Carbon targets. Three sets of data were taken
for each target, the average values being shown in Figure 5.12\ with the
uncertainties (± s. d. ) being less than the size of the plotted points.
Whilst the X-ray production cross-section for 500 keV protons onto
a Carbon target is 15 times that for an Aluminium target (Figure 6.5)
only a factor 4 difference was observed in practice. This may however
be explained by the decreased transmission of the CK X-rays through the
VYNS window and the Polypropylene filter compared to A1K X-rays -a
factor of 2 in each case (Figures 2.7 and 5.3). It was also seen that
whilst the intensity of the A1K X-rays continued to be increasing
rapidly as the proton energy approached 700 keV, the intensity of the
CK X-rays appeared to be levelling off. As a result the measured ratio
of the Carbon-to-Aluminium intensity decreased with energy (Figure 5.13).
93
1
i
% 1=f
: icr C !
r "
ý ý f " "
w Qt
f t
"i ý" i"ý
0 it "iý,
C i" fr
m 1O "" i""i
N
ai
i
Ck S! kAum%%
"1 1 10 X- ray energy, keV
Figure 5.11 The mass attenuation coefficient of low energy X-rays in Carbon (" ), Silicon (A and Gold ( ) targets with the positions of the CK, SiK and AuM X-rays indicated.
94
1
N C
O C. )
T
N C d C
I-
X
1
CK
ýýK
Figure 5.12 The measured X-ray intensities of CK from a Carbon target ( ) and A1K from an Aluminium target (A) as a function of proton energy.
U 2UU 400 600 Proton Energy, keV
95
12
10
oý E cý
.NE C d r+ C
Q
0
C
Figure 5.13 The ratio of the measured Carbon-to-Aluminium K-shell X-ray intensity (as given in Figure 5.12) as a function of proton energy.
Proton Energy, keV
96
CHAPTER 6
THEORETICAL TREATMENT OF RESULTS
6.1 Compensation for target surface contamination
The results'of the surface contamination study (Section 5.2.1)
showed that most, if not all, of the observed target surface contamina-
tion was Carbon. Whilst this would appear to be of no relevance for
a CK X-ray beam (Figure 5.5) it most certainly does affect the intensity
of other photon energies. (Figure 5.6) because of their absorption in
the Carbon layer. In order to ensure that any intensity variations
occurring in the angular distribution study (Section 5.2.2) was a result
of changing the target and/or detector angle, and not due to the build-
up of Carbon on the target surface, corrections were applied to the
angular distribution data to take into account the depth of Carbon the
proton beam had to traverse on leaving the target.
Using the data taken with the detector at an angle of 90° to the
beam at the start and end of each target angle scan, it was possible to
deduce the effective depth of Carbon on the target surface. If the ini-
tial intensity of the target's characteristic X-rays was II and this falls
to an intensity 12 by the end of a scan then, using the expression for
the attenuation of a homogeneous X-ray beam in traversing a thickness
'x' of matter (Harnwell and Livingood, 1933):
12=1II. exp (- ux) 6.1 where u is a constant of proportionality known as the linear attenuation
coefficient.
The u-values used were derived from the mass attenuation co-
efficients (u/p) as given by Hubbell '(1977) (Figure 6.1), taking the
density, p, of Carbon as 2300 kg. m-3 (Science Data Book, 1978). The
linear attenuation coefficients used for the surface contamination
97"
ýCD N
E u1
w
C
v
d
81
c 0 co c
cv U) Cl) cti
Figure - 6.1 The mass attenuation coefficient of low- energy photons in Carbon.
Energy, keV
98
Characteristic Energy W X-ray (keV) (cm 1)
BK 0.183 1.38 x 104
CK 0.278 4.95 x 103
NK 0.392 5.75 x 104
A1K 1.487 1.56 x 103
SiK 1.740 989
AuM 2.120 575
TiK 4.511 57.5
Table 6.1 The linear attenuation coefficient, p,
of ultrasoft X-rays in Carbon (derived
from Hubbell, 1977).
99
correction are given in table 6.1. These values were then used in
Equation 6.1 to give the depth, x, of Carbon laid down on the target
surface between the measurements of I1 and I2. In doing this, it was
assumed that the incoming proton energy loss in the Carbon layer was
negligible. Calculations (Section 6.1.1) showed this assumption to
hold within 5%.
Knowledge of the amount of charge which had been collected on the
target between the measurements of 11 and 12 enabled the Carbon contami-
nation to be expressed in terms of 'thickness per charge' (nm/pC).
It was therefore possible to deduce the depth of Carbon on the target
surface at any stage of the angular scan. However, as this calculation
gave the thicknesslat 90° to the beam (in line with the detector position),
a geometrical correction factor had to be included when the detector
angle was varied before the depth of Carbon through which the X-ray
photon had to traverse on leaving the target could be calculated
(Figure 6.2).
Having calculated the amount of Carbon on the target surface at
any particular stage of the angular scan it was possible to modify
the intensities obtained for the characteristic X-ray yield to com-
pensate for any attenuation occurring in the Carbon layer. If IM was
the measured intensity, as given by MINUIT (Chapter 4), at target
angle, 0, and detector angle, a, then the real intensity, before
absorption in the Carbon layer, IX, was given by
IM (8, a) = Ix (6, a) . exp (-u. ye, a) 6.3
where m is the linear attenuation coefficient of Carbon for the X-ray
energy and Yea is the calculated Carbon thickness on the target surface
as given by Equation 6.2 (Figure 6.2). The actual X-ray intensity was
therefore taken as I
IX (e, a) -- M (B, a)
6.4 exp (-p. ye, a)
100
T----. --.
:e
P
Detector
Figure 6.2 Calculation of the depth of Carbon contaminant on the target surface.
ß= 900 -a, y= 900 -0
Using the Sine theorem :
x sin (180 - (ß+y)) sin y
X. cos 9 6.2 sin (a+A) yA,
a
101
6.1.1 Example of surface contamination calculation
An example of a set of data for the Aluminium target is shown in
Table 6.2, with the measured AlK X-ray intensities given in the second
column. In order to calculate the thickness, x, of Carbon laid down
on the target surface between the first (I1 - 164269 counts) and last
(12 a 159694 counts) measurements, Equation 6.1 was rearranged to give Zn (I2/Il)
. x-6.5 u
Substituting the values of I1 and I2 and using p=1.56 x 103 cm 1
(Table 6.1) gave
x=1.81 x 10-5 cm (181 nm).
Since all the measurements were normalised to the charge collected on
the target, each spectrum being collected for 25 pC, the total charge
collected between the measurements of 11 and 12 was 300 pC. This gave
a Carbon build-up depth of 0.6 nm. pC 1
or, alternatively, 15 nm per
25 pC of charge collected on the target.
Taking the geometrical correction factor given in Equation 6.2 into
consideration, the thickness, y, of Carbon through which the AlK X-rays
had to traverse on leaving the target was calculated for each detector
angle (Table 6.2, column 6). It was then possible to use Equation 6.4
to calculate the real intensity of the AlK X-rays having compensated for
the absorption in the surface Carbon layer (Table 6.2, column 7).
Finally, the intensities were normalised to the theoretical maximum
flux at 6= 30° and a= 60° (Section 6.2) - Table 6.2, column 8.
As can be seen from Table 6.2, the total Carbon depth at the end
of the scan was 195 nm at 90° to the proton beam. This corresponds to
a depth of 0.3 pm in the direction of the beam (a = 0°) which would
result in the proton beam losing - 20 keV in the Carbon layer (Figure 6.4),
with-a 5% reduction in the X-ray production cross-section (Figure 6.5).
102
0
t! º N . ti C ß (x 0
O O O*,
CO C%
O O
0 O
O O
O O
o OA
o O%
in O,
in 00
-7 O O
C to- 0 IIM
.- O O . - r- '- ^ ö .
O O .
O .
O -- C *' m
+ý+ In L -t L/1
00 -e
O C%
r_ M
ul 1-
Ul O
rý CV
-e 1"
%D .O
M 00
-Y' '-
00 1ý
r N
±ä 'c X O ýC -7
ýD
C% 1- LM
%D . - C
u1 -
0% M
-t ýt
%D -7
in M
%M N
in LM
ýt ON
O, %m
%D t
2 + C_ W "- "-
' "--
% r-
%D %O . -
ýO "-
%D "-
% "-
in r-
M . -
O %D
Or 4- O
0 ,Q
92 O %M N O 00 (D %D M LM V) N O 4'' in N. c N lD N u1 N Oa 1z 00 in 0
++ _
fa + - N M in % 00 O
"- M 00 in O-. C\
Q + . - . - N - 00 . -
Z3 F O
O N Oa
00 00
fl, 00
00 00
N C%
O O
M . -
in M
M r
C' vl
Q ON
O O
ÜC '" O O O O O r- . - N . to
U 0 +
0 a) s E5 ,n ä c, w`
ö ý- in "-
O M in - O .O in N. 0 a in O O
cv in M O ºn in
.o O 00 in v% n, W c'
r- r "- r r- ._ . -
LZ tu el
coaom -% Z: L V8 tu
tj : 3.
in N
0 LO
vl N.
0 O
in N
O vi
in r-
0 O
in cV
O L
in O in
-70 O "' . - cV N
n N
r- N
O M
CV (n
VO 0 H
"ý 3
Q% ,m -t M O
0 N -
0 tu N
r18 -t
cV I.
'O O
N M N N
0 - 0
- 00
N
C'4 to C%
.W %M in %0 'G % % %0 N -t m 0' in
I-
++ C 0% - ý
%M N. O O O O . CD
Co 0% p
103
6.2 Calculation of theoretical angular distribution
K-shell characteristic X-rays are emitted isotropically as a
result of the initial state having an angular momentum of 1/2 (Folkmann,
Cramon and Hertel, 1984). The only angular dependence therefore expected
is a geometrical effect due to the self-absorption of the X-rays in
the target.
The complete description of, the effect of self-absorption in the
target would require detailed information on the-energy and X-ray produc-
tion cross-section of the proton beam as a function of the distance
travelled within the target. However, for the purpose of comparison
with the experimental data a more simplified approach, outlined below,
was used (Figure 6.3).
Using the 'Stopping Power' graphs of Anderson and Ziegler (1977)
the energy of the proton beam, Ep, at 1 pm increments within the target
was calculated, with EP = 500 keV at the target surface (Figure 6.4).
This does assume a constant stopping power within each 1 pm interval
and will therefore cumulatively underestimate the energy loss in each
interval. However, the range of the proton as calculated in this manner
was within 3% of the range as given by Anderson and Ziegler (1977)
and therefore the underestimation in the energy loss was deemed negli-
gible. The average energy of the proton beam in each um, EZ, was taken
to be that of a proton at the centre of the increment, that is for
xi = x1 + (i - 1) Ax 6.6
with x1 = 0.5 pm, Ax =1 pm and i=1 to 6. Using the analytical cross-
section formula of Paul (1984), the K-shell X-ray production cross-section
at each position, ai(E. ), were calculated (Figure 6.5): 71
SC X
10' Q 6.7
Z2.2
where f and S, are functions of proton energy, E, and target atomic
number, Z.
104
0
Xi E-
ýýa `" .ý .ý
d i
P
Figure 6.3 Simplifying assumptions used for the calculation of the theoretical angular dependence of the emitted X-rays from a thick target.
d2 xi . sin 0
= sin (A + a)
x Xý X . ýcL
0= 30 °, 45 ° and 60 °; 400 4a< 1300
105
500 "�
400- 6.. Error bar for "'"ý '", ý each point
300 '° "' ''ý
. d0
o. rn
w 200 '° '"o ö
'ý '" 0
100 '". q
00 23456 Depth Into target, Nm
Figure 6.4 The variation of proton energy with depth into the target.
Gold target 0 Titanium target
A Carbon target (a similar curve is obtained for Boron)
E3 Aluminium target " Silicon target
106
1
1
E1 u
N
70 r
K
w
C 0
a) N
N N 01 L.
C 0
4.1 V
01 1- a
cv
X 1
1
Proton Energy, MeV
Figure 6.5 The K-shell X-ray production cross-section for proton bombardment of various low Z targets (using the method of Paul (1984))
0 Boron Carbon
A Aluminium Q Silicon
" Titanium
107
The proton flux was assumed to be constant over the whole
of the proton range, which is generally the case provided EP » Ik,
the electron ionisation energy (-1 keV). No account was taken of the
divergence of the proton beam, the assumption being that this will only
be of major importance towards the end of the proton range and that
even if divergence is occurring it may still be assumed that on average
the protons follow a central path.
The number of K-shell photons emitted from a thick target into a
solid angle 0 per N incident protons of energy E and range R0 is given by
(Lewis, Simmons and Merzbacher, 1953):
Ro - ud(x)
I (Ro) _2. Nnp fe. a {E (x) } dx 6.8 4 ir 0
where n represents the number of target atoms per gram, p represents the
density of the target, ji is the absorption coefficient of the target for
its own characteristic X-rays (Hubbell and Veigele, 1976; Hubbell, 1977)
and d is the distance which the X-ray travels in the target. In the
simplified calculation considered here, the photon flux in the direction
of the detector was taken to be proportional to:
6- ud. 0 0, a=
E Q. (E. ) .e6.9 i=1
where a (Z) was calculated using Equation 6.7 and dZ (Figure 6.3) was
calculated from
x" sin A d. Z 6.10
Z sin (A + a)
00, a
was calculated for each target/detector angle combination and
normalised to the maximum flux (found at 0= 30° and a= 60°).
Also calculated were the normalised flux when considering only the
first interval of the proton's interaction within the target, that is
for the first 1 pm only. In this case the photon flux was taken to be
108
represented by
-ud1 ýeý a Q1 (E1) .e6.11
and all values normalised to 1030,60
In the case of the Gold M-shell X-rays a similar treatment was
considered but with the production cross-section, a (Ederived from
the total M-shell ionisation cross-section, a M, (Johnson, Basbas and
McDaniel, 1979) combined with the M-shell average fluorescence yield,
wM, (Bambynek et at., 1972):
I vZ = am .w6.12
6.2.1 Example of the theoretical angular distribution calculation
For 500 keV protons penetrating an Aluminium target, the energy
of the protons along their range (- 5.5 pm) was calculated using the
stopping power data of Anderson and Ziegler i(1977), (Table 6.3).
These values were plotted (Figure 6.4) and used to find the energy
of the proton beam at the centre of each pm increment (Table 6.4,
column 2). The X-ray production cross-section at each energy was
then derived from Figure 6.5 (Table 6.4, column 3).
Using Equation 6.10, the distance travelled by the X-ray through
the target (d2) was calculated for a given target and detector angle
combination for each of the distances travelled by the proton
(xi, i=1 to 6) (Table 6.5). Taking the Aluminium self-absorption
coefficient as 1.08 x 103 cm- l (Hubbell and Veigele, 1976) the
product, OZ, of the production cross-section (a2) and the attenuation
in the target (e-sdi) was calculated (Table 6.6).
For each target/detector angle combination, 0 e, a
(Equation 6.9)
was calculated from 6
Sýeý a ZEI
of 6.13
109
Depth into Proton energy Stopping Power Proton energy target of Aluminium after 1 pm
(um) (keV) (keV/um) (keV)
0 500.0 69.3 430.7
1 430.7 74.1 356.6
2 356.6 81.3 275.3
3 275.3 90.3 185.0
4 185.0 105.4 79.6
5 79.6 125.3 -
Table 6.3 The energy loss experienced by a 500 keV proton
on entering an Aluminium target, assuming a
constant stopping power within each 1 um interval.
110
Depth into Proton X-ray production target Energy cross-section
xi E. ci (awn) (keV) tx10'24 cm2)
0.5 466 190
1.5 392 130
2.5 316 75
3.5 230 32
4.5 132 6
5.5 14 < 10-5
Table 6.4 The AlK production cross-section for the
proton as it penetrates the target.
111
Detector di (pn) Angle i123456
a° xi 0.5 1.5 2.5 3.5 4.5 5.5
40 0.266 0.798 1.330 1.862 2.394 2.926
50 0.254 0.762 1.269 1.777 2.285 . 2.792
60 0.250 0.750 1.250 1.750 2.250 2.750
70 0.254 0.762 1.269 1.777 2.285 2.792
80 0.266 0.798 1.330 1.862 2.394 2.926
90 0.289 0.866 1.443 2.021 2.598 3.175
100 0.326 0.979 1.632 2.284 2.937 3.590
110 0.389 1.167 1.945 2.723 3.500 4.278
120 0.500 1.500 2.500 3.500 4.500 5.500
130 0.731 2.193 3.655 5.117 6.579 8.040
140 1.440 4.319 7.198 10.078 12.957 15.837
Table 6.5 The distance travelled by the X-ray, d., within the
target for a target angle, 0= 30°.
112
Detector
Angle
a° i1
X. It. : 0.5
2 1.5
3 2.5
d1
4 3.5
5 4.5
6 5.5
46 30, a
(ERZ) 46 30. a
30,60
40 185 119 65 26 5 N 400 1.00
50 185 120 65 26 5 E 401 1.00
60 185 120 66 26 5 G 402 1.00
70 185 120 65 26 5 L 401 1.00
80 185 119 65 26 5 I 400 1.00
90 184 118 64 26 5 G 397 0.99
100 183 117 63 25 4 I 392 0.98
110 182 115 61 24 4 B 386 0.96
120 180 111 57 22 4 L 374 0.93
130 176 103 51 18 3 E 351 0.87
140 163 82 34 11 2 (< 1) 292 0.73
Table 6.6 The product, of the AlK X-ray production
cross-section, aand the attenuation of the X-ray
in the target (exp [-µd ]) for a target angle,
0= 300.
113
and the intensities normalised to the maximum flux occurring at 0- 30°
and a= 600. Finally 06, a
(_ 01) was considered (Table 6.6, column 2)
and normalised to 0 30,60
(Table 6.7).
114
Detector 30, a Angle 1
a° 30, aß 30,60
40 185 1.00
50 185 1.00
60 185 1.00
70 185 1.00
80 185 1.00
90 184 0.99
100 183 0.99
110 182 0.98
120 180 0.97
130 176 0.95
140 163 0.88
Table 6.7 The contribution of the AlK X-rays
produced in the first pm of the
target normalised to those produced
at a target angle, 0= 30° and
detector angle, a= 60°.
115
CHAPTER 7
APPLICATION OF MONTE-CARLO TRACK STRUCTURE CODES
7.1 Introduction
This chapter sets out to consider the microtopography of the
radiation tracks produced by the interactions of ultrasoft X-rays
of varying energy in soft tissue. Significant differences in patterns
of energy deposition may result in differences in biological effective-
ness which are open to experimental investigation.
Whenever photons interact with matter they create secondary elec-
trons by means of three competing interactions :
(a) the photoelectric effect, in which most of the incident photon's
energyAs expended in the ejection of an orbital electron;
(b) the Compton effect, in which part of the primary photon's energy
is transferred to a single atomic electron and a 'Compton scattered'
photon is emitted with energy and direction determined from relativis-
tic momentum-energy conservation;
and
(c) pair production, requiring a minimum of 1.02 MeV, in which an
electron-positron pair is produced. The relative importance of each
interaction type shifts with increasing energy such that for photon
energy, lhu < 30 keV the photoelectric effect is dominant (in low
atomic number material) whilst for hu > 10 MeV it is pair production
which is most important. Thus the energy deposition by X- or y- ray
fields on a microscopic scale is determined almost exclusively by the
energy loss of their secondary electrons. The importance of the energy
deposition of electrons is in fact common to all radiation fields,
including heavy ions and neutrons (Paretzke, 1980). A detailed theore-
tical understanding of the microtopography of events occurring within
the track of a charged particle is consequently important. The
116
availability of the necessary cross-section data (for example Opal,
Beaty and Peterson, 1972) and the development of sufficiently powerful
computers has contributed towards an increase in our knowledge of the
energy deposition by ionising radiation by permitting the calculation
of the local concentrations of activations on a nanometer scale using
Monte-Carlo track structure simulation codes (Berger, 1974; Heaps and
Green, 1974; Hamm et at., 1976; Paretzke, 1978).
Ionising radiations can induce in cells a variety of biological
changes. The cell nucleus appears to be the main region of radio-
sensitivity. However, a large proportion of the ionisations, excitations,
and their products, produced by the radiation appear to have no permanent
effect on the cell (Goodhead and Brenner, 1983a; Goodhead et aZ., 1985).
Permanent effects such as cell death (loss of proliferation capacity),
mutations and chromosome aberrations appear to be caused by some'rare,
critical damage induced by the radiation. Differing types ('qualities')
of ionising radiations have differing efficiencies in producing biological
effects thus underlying the need for information on the physical proper-
ties of radiation interaction which may be correlated with the biological
effect of practical radiations and assist in the understanding of the
mechanisms of radiation action.
7.2 Local Energy Deposition of Radiation Tracks
Previous experiments with ultrasoft X-rays have shown that complete
biological effects can be efficiently produced by small amounts of highly
localised energy depositions (Goodhead, Thacker and Cox, 1979;
Goodhead et at., 1980; Thacker, Goodhead and Wilkinson, 1983). For
example, the photo-electron produced as a result of the absorption of
a CK X-ray photon in soft tissue deposits < 280 eV in <7 nm. These
studies indicate that the damage produced by ultrasoft X-rays is similar
in nature to that produced by conventional low L. E. T. radiations such as
hard X-rays and y-rays (Goodhead, Cox and Thacker, 1981).
117
The small dimensions (down to < 10 nm) involved in the energy
deposition of low-energy electrons, including those produced by ultra-
soft X-rays are, unfortunately, below the region of most experimental
measurement techniques ( -1 - 10 pm). However, a possible way of
identifying the critical characteristics of radiation action is by the
use of Monte-Carlo track structure codes. These simulate the radiation
tracks in the form of complete atomic, interaction-by-interaction
histories, recording all spatial and energy transfer information
(Paretzke, 1980). This information can then be compared with the
observed relative biological effectiveness (R. B. E. ) of the different
radiations.
During the late 1950s analyses of the relative effectiveness
of heavy ions of different L. E. T. indicated that it might be possible
to describe their mechanism of action in terms of a threshold energy
deposition in small sites (- hundreds of eV in- nm) (Howard-Flanders, 1958).
The availability of Monte-Carlo track structure calculations (Paretzke
et al., 1974; Hamm et al., 1978; Terrisol, Patau and Eudaldo, 1978;
Wilson and Paretzke, 1980) supported by low-pressure cloud chamber
measurements (Budd and Marshall, 1980; Kwok, Budd and Marshall, \, 1981;
Brenner, 1982) has enabled the threshold energy concept to be reassessed.
Previously reported experiments (Cox, Thacker and Goodhead, 1977;
Goodhead et at., 1980; Thacker, Cox and Goodhead, 1980;: Virsik et at., 1980;
Goodhead, Thacker and Cox, 1981a) have shown that very low energy photons
have considerably greater effectiveness per unit dose than hard X-rays
or y-rays, the lowest energy X-rays having the greater biological
effect. Since the X-rays deposit energy throughout the irradiated cells
only in highly localised, small quantities, complete biological lesions
may be produced by these well defined energy concentrations. Goodhead
and Brenner (1983b) set out to investigate the threshold energy concept
in terms of a distance and an energy such that the probability of this
118
amount of energy (or greater) being deposited by the X-rays within a
volume correlated with the observed R. B. E. of the radiation. By com-
paring data obtained from experiments (Goodhead and Brenner, 1983b and
references therein) with those calculated using the Monte-Carlo track
structure codes (Brenner, 1982; Brenner and Zaider, 1982; Zaider, Brenner
and Wilson, 1983) it was concluded that, for low L. E. T. radiations at
least, there was correlation for threshold energies of about 100 eV in
a spherical volume of diameter -3 nm. This correlation did not hold
for high L. E. T. radiations.
Whilst spherical volumes are mathematically convenient, intra-
cellular structures at the nanometer level of organisation (such as
the D. N. A. double helix, nucleosome, chromatin fibre) are better
modelled by cylindrical volumes. Since 1983 a method for scoring
energy depositions in randomly orientated cylinders has been developed
(Charlton et al., 1985a, 1985b; Goodhead et al., 1985) and using a
Monte-Carlo track structure code, based on that of Wilson and Paretzke
(1980), an extensive calculation and tabulation of absolute frequencies
of energy deposition by mono-energetic protons and a-particles has
already been produced (Charlton et al., 1985a). The process is being
extended to include mono-energetic electrons and photons (H. Nikjoo,
pers. comm. ).
7.3 Calculations using a Monte-Carlo structure code
The energy deposition in small cylindrical targets by simulated
electron tracks from the absorption of Carbon, Aluminium and Titanium
K-shell X-rays has already been calculated (Nikjoo, Goodhead and
Charlton, 1987). The Monte-Carlo technique employed has the capability
of dealing with extensive input-tables on the relative photo-absorption
and atomic de-excitation probabilities for a given irradiated medium,
such as soft tissue. However, the construction of complete input
119
tables to include all modes of absorption of an ultrasoft characteristic
X-ray by atoms of all elements in soft tissue, weighted to take account
of the photoionisation cross-sections and the percentage by weight of
each element in soft tissue, is very time consuming. It would therefore
seem practical to try to reduce the time taken by considering only
the most significant contribution(s). In this way a wider range of
photon energies may be investigated and any discontinuities in local
energy deposition as'a function of photon energy observed. ' These may
have biological consequences and therefore be open to experimental
investigation with appropriate proton-induced ultrasoft X-ray beams.
Goodhead and Thacker (1977) have shown that Carbon, Nitrogen and
Oxygen atoms are responsible for - 99% of the absorbed dose when
mammalian soft tissue is'irradiated by AlK X-rays. The most signifi-
cant contribution (- 90%) is the result of absorption in the Oxygen
atom's K-shell - Oxygen accounting for - 75% by mass of soft tissue
of the same atomic composition as spleen of 'reference man' (I. C. R. P.,
1975). It might therefore be expected that an input table which
showed the production of electrons resulting from the absorption of
characteristic X-rays in the Oxygen atoms only would approximately
simulate the effects taking place in soft tissue having its full atomic
composition.
7.3.1 Track Structure and Scoring Programme
Using a Norsk Data mini-computer at the M. R. C. Radiobiology Unit
and the electron track simulation code 'MOCA7' contained within the
charged-particle track structure code 'MOCA14' (Wilson and Paretzke,
1981) along with the scoring programme 'ESCORE' (D. E. Charlton and
H. Nikjoo, pers. comm. ) it was possible to carry out scoring for
energy deposition in small cylindrical volumes for the K-shell
characteristic X-rays of selected elements within the atomic number
range 45Zä 20.
120
A number of photons were considered for each X-ray energy, taking
only absorption by the Oxygen atom into account. This resulted in
either one or two electrons, depending on whether the photon energy was
sufficiently large to cause the release of an Auger electron as well as
the initial photo-electron (Coghlan and Clausing, 1973) (Figure 1.2).
A minimum amount of energy required to free an electron ('residual
potential energy') of 32 eV was assumed for Oxygen in soft tissue
(H. Nikjoo, pers. comm. ). The resulting input parameters for all the
elements considered are given in Table 7.1.
The code MOCA7 was then used to generate the tracks of electrons
in water vapour adjusted to a density of 1 g. cm 3 (103 kg m73), the
electron histories being followed until their energies had fallen to
10 eV. Although liquid water is of greater biological relevance than
water vapour (the cell consisting of - 80% water), there are fewer
experimentally measured cross-sections on which to base a liquid
water transport code. Although it is known that there are differences
between macroscopic cross-sections in the liquid and vapour phases
(Goodhead, 1987d) a comparison of spatial patterns of energy deposi-
tion, -for electrons, between liquid water codes (Hamm et al., 1976;
Terrisol, Patau and Eudaldo, 1978) and those obtained from a water
vapour code (Turner et aZ., 1982,1983a, 1983b) indicate significant
differences only in energy deposition over small (< 1 nm) distances
(Paretzke et al., 1986).
7.3.2. Results
Scoring of the energy deposition of the electrons was carried out
within a gross volume defined by a virtual sphere whose radius was at
least 1.1 times that of the length of the longest electron track.
Chords cutting the virtual sphere were generated using 'p-randomness'
(Kellerer, 1971) and these served as the axes of the cylindrical
121
Element Atomic No.
z
Photon Energy
(eV)
Residual potential energy
(e V)
Photo- electron energy
(eV)
Auger electron energy
(eV)
Be 4 109 32 77 -
B 5 183 32 151 -
C 6 278 32 246 -
N 7 392 32 360 -
0 8 525 32 493 -
F 9 677 32 140 505
Na 11 1041 32 504 505
Mg 12 1254 32 717 505
Al 13 1487 32 950 505
Si 14 1740 32 1203 505
C1 17 2622 32 2085 505
Ca 20 3692 32 3155 505
Table 7.1 Input parameters for the electron generating code,
MOCA7, considering the interaction of characteristic
K-shell X-rays with Oxygen.
IlIAoujl a iesiduai polenl; al e ieºtgq of 32eV
was used to all Ae aLove eases a al; 9yýý JoweI
value would be oioýc aPPýcP{ýaýe ýaý
e sinj j io n; seal cases (Be - O)
122
Figure 7.1 Diagram illustrating the virtual sphere enclosing the simulated electron track(s) and crossed by the scoring cylinders arranged at random.
123
volumes in which the energy deposition was scored (Figure 7.1).
Each cylinder was then divided into small sections, beginning at a
random point along its length, and the total energy deposited in each
section recorded. The lengths of the sections were chosen as multiples
of the radius of the cylinder. The frequency of deposition of given
amounts of energy in each cylindrical section was recorded and normalised
to the frequency which would have occurred had the macroscopic region
containing the cylinder been subjected to an electron flux delivering
one Gray (Gy). As a check on the sampling accuracy, the ratio of the
volume of the virtual sphere to the volume sampled was compared with the
ratio of the energy deposited in the virtual sphere to the energy
deposited in the target cylinders. These two ratios were generally
within 5% of each other. The diameter and length of the target cylinders
could be varied and data were collected for dimensions similar to those
of biological structures:
Biological Structure Cylinder diameter (nm)
Element of DNA 2 Nucleosome 10
Element of chromatin fibre 25
In each case, energy deposition was calculated for four cylinder
lengths, 1, given by
1=nxd7.1
where d represents the cylinder diameter and n=0.5,1,2 and 4.
Scoring was initially carried out for Carbon and Aluminium
(Z =6 and 13 respectively) to enable a comparison to be drawn between
the data obtained from considering absorption by the Oxygen atoms only
and those-already obtained by H. Nikjoo (pers. comm. ) from considering
the full elemental composition of soft tissue (as given for spleen of
'reference man' (I. C. R. P., 1975)). The results obtained are shown in
Figures 7.2 and 7.3. The error bars shown on these and following
figures were calculated from the assumption that the statistics of the
124
Figure 7.2 The frequency of energy depositions greater than a given amount, E in cylinders of various sizes, per Gray of dose to the surrounding medium. Shown are the frequencies obtained when cosidering the interaction of CK X-rays with the full elemental composition of soft tissue (o) (H. Nikjoo (pers. corrnn. ))
and, where they differ by more than the size of the plotted points, considering the interaction of CK X-rays
with Oxygen atoms only (A). The error bars were cal- culated using Equations 7.2 and 7.3 in the text.
(i. ) 2 nm diameter cylinder -A: 8 nm length B: 4 nm length C: 2 nm length D: 1 nm length
(ii. ) 10 nm diameter cylinder -A: 40 nm length B: 20 nm length C: 10 nm length D: 5 nm length
(iii. ) 25 nm diameter cylinder A: 100 nm length B: 50 nm length C: 25 nm length D: 12.5 nm length
125
i 4
i 0
> O r`
W
7 01 ýd
C W
0 -0 N r.
0 t N G) I-
O '_ O ý'
r
0 Co
0 to
0 1a
O N
0
'' ý93< uo,; lsodep ABioue jo Aouenbei: j
126
d ýw. 'w
oº
,c w
v >0 Jt
y I-
i s~
i 1
v
i
l, Ag' 3< uoillsodep Aßjeue 10 A3uenboi j
127
0 Co c4
D !O N
0 v N
D N N
D D N
co r
0
co. w
o OI rc
w V
0ö Nt ry
C!
0 Co
0 KO
0 e
0 N
D
ý. AD `3< uoillsodep ABieue jo Aouenbajj
128
Figure 7.3 The frequency of energy depositions greater than a given amount, E in cylinders of variousisizes, per Gray of dose to the surrounding medium. Shown are the frequencies obtained when considering the interaction of A1K X-rays with the full elemental composition of soft tissue (Q) (H. Nikjoo (pers. corrnn. ))
and, where they differ by more than the size of the plotted points, considering the interaction of A1K X-rays with Oxygen atoms only (o ). The error bars
were calculated using Equations 7.2 and 7.3 in the text.
(i. ) 2 nm diameter cylinder -A: 8 nm length B: 4 nm length C: 2 nm length D: 1 nm length
(ii. ) 10 nm diameter cylinder -A: 40 nm length B: 20 nm length C: 10 nm length D: 5 nm length
(iii. ) 25 nm diameter cylinder -A: 100 nm length B: 50 nm length C: 25 nm length D: 12.5 nm length
129
N
> d
U) W
I-. oý I- 0) C W
v 0 t
ry .y
t
N O
O
ýAE)13 < uoillsodap i5Jeua 3o Aauenbei3
130
Q CD 0
b+ p4-4 ºý º-4--4 Q 07+ 47 º-d7
OO -4-p b-4 a QQ ý-4Q
Qp -40 º-4-40 OO '-47 º-'4-O
Qp º40 F-4-a C2 p º-4Q º-4-ý p
Cl Q N. 40 º-4412 OD '40 '-4-4 O DQm º-4-4 p
4] ' 4: ] 413 º44 O 4] O 40 4O QO 47 413
QO 413 40 QQ 40 40
pQ 4O 40 pO 47 O
QOmm pO 47 QO
pQ 47 O QO 4] O
Q 40 4) 4] Q 4] 40 40
QI Q] 4) Q] 40 43 ßl Q]
4] mmm 4 4] 47 47 O 4] 4] 40
OOQm QQOO
pQDO pOQO
pQQO pQQp
QpQQ OOQO
QQQO pOOQ
OQQO OQ7O
DOOO pQOD OQOQ
QpQD QOpQ pQOQ
DQDQ QpoQ
QQpQ pQQp
pQOp DOOO
Q. pOQ QDQO D Cil Cl O
Q 131 131 D O [il Ol O
p [41 O1 Q O [ml (ii O
O [i! QQ OO Q1 Q
D til O Cil Q Cii fý p [91 O Cif O ßA O Oi
LM to pi Q1 0
us Ql Ci1 "I1
o0 0 r. "- "-
t-(D `3 < uolllsodep Aßioue jo AouenbeJd
ti
T
> m Y
IT . LU
214 rn a) c w v
co ö N G) I.
N
r-
-o .Q r-
131
I-
Qm C) G QQ Q
44 4 4 4 Q4 --d
44 4 4 ®Q Q4 4) 44 4 4
44 4 4 Q1 0 QJ Q 444 4
444 4 131 Q3 Q Q 444 4
444 4 QQQ Q
4444
4444 QQQm
4444
4444 QQQD
4444
4444 D ßl QD 4444
8a 4444 DQDQ 4444
4444 El 0DD 4444
r1
4444 C31 m CSl ß 4444
ý4 ä
4444
XE) 13< uolllsodap A6iaue jo Aouenbeid
0
Co
hY
W
ýý .ý
c w
iý O
r N d
ar H
M
O O
.
132
data were represented by the number of hits within the target cylinders.
If 'H' represents the total number of hits within the target cylinders
and 'f>E' denotes the frequency of energy deposition greater than E per Gy,
then the percentage error on each frequency was calculated using:
nE = f>E
xH7.2 f>0
% error =� nE x 100 = (rtE) 2x 100 7.3
nE
If the entire track of a CK X-ray absorption in soft tissue is
approximated as a highly localised ('point') event in a large scoring
cylinder, then as the cylinder length increases the frequency of energy
deposition greater than a given amount, E, would be expected to increase
proportionally. From Figure 7.2, (ii) and (iii), it is seen that the
frequency of energy deposition does double with cylinder length for the
larger diameter cylinders, but for the smallest cylinder diameter (Figure
7.2, (i)) as the cylinder length is doubled the frequency increases by
more than a factor 2 at high threshold energies (E > 50 eV) and less
than 2 at lower energies. This suggests that as the target cylinder
length increases it is more likely to contain large energy depositions
(large hits) and less likely to have just a small amount of energy
deposited in it. Figure 7.3 shows that AlK X-rays follow the same
tendency but, because of the larger dimensions of the tracks, a larger
sized target (> 25 nm diameter) is required before the approximation
holds that doubling the target size doubles the chance of hitting it.
The results of scoring the CK and AlK X-rays showed sufficient
agreement between the data considering absorption in Oxygen atoms only
and the data taking the complete elemental composition of soft tissue
into account to warrant the continuation of scoring using absorption by
Oxygen only. Using the data of Table 7.1 as input, scoring was there-
fore carried out for the remaining elemental characteristic energies using
the code ESCORE. A limit of Z <_ 20 was set because efficient scoring
133
1
1I
C, U; Al
O
O a
rn
c d
. 1.1 O
0 c
v d I- LL
Figure 7.4 The frequency of energy depositions greater than a given amount, E in a2 nm diameter, 2 nm long
cylinder by simulated electron tracks from various low Z characteristic X-rays.
O BeK X-rays
A CK X-rays
A1K X-rays
" CaK X-rays
Threshold Energy 'E'. key
134
I- C
W
A c c
1
Figure 7.5 The frequency of energy depositions greater than a given amount, E in a 10 nm diameter, 5 nm long cylinder by simulated electron tracks from various low Z characteristic X-rays.
o BeK X-rays
A CK X-rays
A1K X-rays
" CaK X-rays
0 .' .2 ,3 .4 .5 Threshold Energy 'E ý. keV
135
a t9
w A C 0
U, 0 a m v
rn m C m w 0
0 c m
c 0 I- U.
Figure 7.6 The frequency of energy depositions greater than a given amount, E in a 25 nm diameter, 25 nm long cylinder by simulated electron tracks from various low Z characteristic X-rays.
O BeK X-rays
A CK X-rays
A 0K X-rays
A1K X-rays
" CaK X-rays
0 "1 "2 "3 "4 "5 -6 Threshold Energy 'E '. keV
136
can only be carried out for low-energy electrons, where the entire
electron track is confined to a relatively small volume. For high-
energy electrons, multi-stage sampling procedures have proved necessary
to enhance the scoring efficiency (H. Nikjoo, pers. comm. ). The com-
plete energy deposition data are given in Appendix 2, with selected
samples, including the lowest and highest (Z =4 and 20) photon energies,
plotted in Figures 7.4,7.5 and 7.6. The target cylinder sizes were
chosen to represent:
(i) element of DNA
(ii) nucleosome
(iii) element of chromatin fibre
2nm diameter x 2nm length
: 10nm diameter x 5nm length
25nm diameter x 25nm length
The cylinder lengths chosen for the elements of DNA and chromatin
fibre are arbitrary.
Having generated the basic data it was possible to draw an inter-
elemental comparison by selecting a particular size of target cylinder
and plotting the frequency of events depositing more than a given energy,
E, in the target as a function of characteristic X-ray energy. Such
data are presented in Figures 7.7,7.8 and 7.9. Whilst comparison with
existing experimental data (Goodhead and Brenner, 1983b; D. T. Goodhead,
pers. comm. ) would suggest that threshold energies of the order of 100 eV
within targets of - 2nm diameter is the important region in terms of low
L. E. T. radiations such as X-rays, larger sized targets are of particular
interest when considering high L. E. T. radiations such as a-particles
(Goodhead et al,, 1985).
Figure 7.7 shows that low energy X-ray photons (< 1 keV) are the
most efficient in depositing small, localised amounts of energy within
a target. Within this region, the data show varying degrees of struc-
ture, the more pronounced being at the higher energy of deposition
(E - 200eV). The decrease observed at the lowest X-ray energies
(BeK, BK and CK) is probably due to the limit set by the X-ray energy
137
b
W> 0
" " "-t,
" i i
i . i
+ i i"
i i .ý tip
O 0 N $4 tO W (1)
co a) Cd
0 144
41
b 44 0
0) N lý w
L4 Gl W -W d
4 cd N 10
O rl W
m F. ' N
N Y
w M ro (if `r
N m a >, w 60 ö U
i- -i Im
I- I
co :3 1J O 'L7 C: F1 r-4 as cd O
º *-1 Z- V 11Q! N i
bO i. J
Y a) aý ctS O
a 41 Ö
º-. r-+ O-
, -4;. "
V- m_
O
ii[) 1 3. < uopflsodep AB., eue jo Rouenbojd
138
I
W m
i
"
4
i
i
i i
o 0 14 c) N M a Cl
41 'J
ýQ b0 N
t! ) O 4-4 Ö
rl N 41 O IH W U) O Na a as aýb ý ý
M r l
b0 ý NU
G 00 UÖ
4. -
44 r4 0 Ü
N V
v' N 4l N N 1ý
44 aJ
U cd 41
N Ö R
cn -H
- Q ry ON UwN
` ri qN
ö)- 41 0 "0 0
ai cu _ ý 'ý d a iqc n
i i i Z- m P-1 >(L) Y N GO ý+
U. + cd aJ
d41ö
0 ;0 41 in
m- 3 o_ m lL
0 .- .- r- V.
g_At) '3< uollisodep ABJeue lo AouenbeJ, 1
139
A
'o I-
0 Wm in In o C) t p
N
tM
M
Ln
" N
Y 1 _
CI
C!
ii ii Z- to
ii 6 o- - ý
i U-
0 10 'o o 'o ýI j` 3<u ollisodepýAß eua lo ýIauenbäjd
.N
Glr-l
a, HH DO
0 v1 W A O
-rl 0 1. ) W "4 is H O0 av
"0 _H
0o u W oO Q) O W O U v1
41 w
o' H 41 d N ýº
W OJ
N Co "c ri 4) -0
H E3 Oý
W
ýN O
Hý
Ei r4 OH Vwy
Co :i 4»+ O
Ei (L) cu O
p `4
H OO 4º a) 41 C Q) r
M g2 4.1 ctf
41 0
N
ül L1
140
itself; as E approaches the X-ray energy the frequency of energy
deposition greater than E will obviously fall. Assuming the uncertain-
ties are represented by Equations 7.2 and 7.3, the fluctuations in the
region of the NK, OK and FK X-ray energies are more than can be accounted
for by scoring statistics alone. Referring to Table 7.1 it is seen that
for Fluorine, an Auger electron is emitted as well as the photo-electron.
Looking in particular at a threshold energy, E- 200eV, the photo-
electron (at 140eV), if acting independently, should be incapable
of depositing energy > 200eV within the target and hence the probability
of energy deposition should be dominated by the Auger electron (505eV).
The Auger electron is similar in energy to the photo-electron from the
interaction of OK X-rays with the Oxygen atoms in soft tissue, yet the
probability of energy deposition > 200eV per Gy for the FK X-rays is
double that for the OK X-rays. This would suggest that there might be
a substantial combined effect resulting from the overlapping of the
electrons as opposed to each electron acting independently.
Figures 7.8 and 7.9 again show that the low energy photons are the
most efficient at depositing small amounts of energy within a target.
With a target size of 25nm x 25nm (Figure 7.9) the electrons produced
by the low energy photons are usually entirely enclosed within the
targets (for example, the CK X-ray (278eV) gives rise to an electron
with a range of < 7nm (I. C. R. U., 1970)). As the photon energy increases,
whilst the electrons produced are still enclosed by the targets there
will be fewer of them per unit dose, and hence the efficiency of deposit-
ing a threshold energy falls (from BK to Na, X-rays). Further increase
in the photon energy (> 1 keV) leads to electrons starting to 'escape'
from, or enter into, a target and so targets are more likely to be hit
(from Na K to higher energy X-rays). As the threshold energy increases
(from 150eV to 600eV) the benefit of additional hits is lost because
these higher energy (lower L. E. T. ) electrons have a low probability of
depositing large energy in a target even if they do pass through it.
141
CHAPTER 8
DISCUSSION
8.1 Comparison of the observed angular dependence of proton-
induced ultrasoft X-rays with theoretical calculation
Both the observed and calculated angular dependence of proton-
induced ultrasoft X-rays are shown in Figure 5.10 ((i) - (vi)).
For K-shell X-rays the agreement between the experimental data and
the theoretical estimate, based upon the consideration of X-ray
production over the total proton range in the target, is best when the
target is inclined at an angle of 30° to the proton beam. As the target-
to-beam angle decreases the experimental data becomes more in agreement
with that calculated from consideration of the X-rays produced in the
first 1pm of the proton range only.
A certain amount of disagreement between experimental data and
theoretical estimates might be expected because of the limited (fpm)
resolution steps of the calculations and the simplifying assumption
that the proton energy remained constant over each fpm increment
(Section 6.2). In fact, the proton energy (and therefore the X-ray
production cross-section) would be greatest at the target edge side
of each increment where self-absorption effects would be less. The
effects would be expected to be more pronounced at larger target angles,
0, due to the sin 0 dependence (Figure 8.1). Further reasons for dis-
agreement might also be expected as the result of an irregular target
surface profile. Studies by Gaines (1981) have shown that anisotropy
additional to that produced due to self-absorption in a smooth target
can be expected for surface roughness with average peak-to-valley
distances greater than -0.1Nm. Methods of quantifying the effect of
target roughness were considered, the most promising being the intro-
duction of regular rulings onto a smooth target having an average
142
e=so'
Y- x= sin B 1
Figure 8.1 The sin 0 dependence of the difference in the distance travelled within the target at either end of each 1 pm increment.
143
peak-to-valley distance of, < 0.05pm. However, the cost of producing
such a high specification target meant that such a study extended
beyond the scope of this work.
8.2' A practical beam line for proton-induced ultrasoft X-rays
From the results of the angular distribution study (Section 5.2.2)
it is seen that for the maximum intensity of characteristic X-rays a
small target-to-proton beam angle is desirable. Whilst practical limita-
tions in this work, resulting from the need for electrical connections,
have meant that the narrowest target angle studied was 30°, geometrical
considerations (Equation 6.10) show that narrower angles would maximise
the X-ray yield still further. For any particular target angle, the
optimum detector angle, that is the irradiation position, would appear
to be perpendicular to the target surface with the X-ray yield decreasing
as the irradiation position moves further away from the perpendicular.
With particular reference to the target angles and detector angle varia-
tions used in this work, the maximum X-ray yield occurred for a target
angle of 30° and a detection angle of 60° to the beam. These effects
become less important as the X-ray energy increases -a direct conse-
quence of the reduction in the target self-absorption of the X-rays.
Whilst increased proton energy does, in general, lead to an
increased X-ray production cross-section (although for the lowest Z
elements such as Boron and Carbon, the cross-section does reach a
plateau at proton energies of a few 100 keV (Figure 6.5) before falling
as the energy increases further towards 1 MeV and greater), the
advantage thus gained must be weighed against the necessity of elimi-
nating all scattered protons from the photon beam. Since this would
appear to require some form of filtration (Section 2.4.1), the higher
the proton energies the greater will be the filter thickness required
to stop the scattered protons. This will result in a corresponding
144
decrease in the transmission of the photon beam thereby offsetting the
increased production at high proton energies.
Previous work (Sterk, 1965; Khan, Potter and Worley, 1965; Gaines,
1981) has shown that Carbon contamination of the target surface is a
common problem when studying proton-induced ultrasoft X-rays. The
indications are, however, (Section 5.2.1) that this may at least be
reduced by making use of high (> NA) proton beam currents. Beam currents
of the order of 500 pA have been used at the M. R. C. Radiobiology Unit
and whilst Carbon deposition was still observed on the target surface,
the rate of build-up of Carbon and the subsequent reduction in charac-
teristic X-ray yield (from an Aluminium target) is far less pronounced
(D. T. Goodhead, pers. comm. ) than with beam currents of - 35 nA as used
in this work. The beneficial effects of high target currents may be
due to target heating.
When considering the irradiation of biological systems, and in
particular, mammalian cells, additional factors have to be taken into
account. Although all the measurements in this study were made within
the evacuated scattering chamber (Section 2.2.3) the detection/
irradiation position for mammalian cell irradiations would almost
certainly need to be at atmospheric pressure in order for the cells
to be able to survive. It is necessary to irradiate an attached mono-
layer of cells such that their position is well-defined and there is
minimum attenuation across the cell layer. The subsequent removal,
counting and biological assay of the cells necessitates the irradia-
tion of a large number ( -. 104 - 105) of cells and hence an irradiation
dish base of the order of 3cm in diameter is required. A well-defined
entrance surface is required for dosimetry purposes and therefore
irradiation cannot be from the air side (as opposed to the dish base
side) of the cells since, even if the growing medium was poured away
there would remain an unmeasurable, variable layer of liquid, and if
145
the dishes were dried the cells would die. As a result of these
considerations the preferred system in use is to make an irradiation
dish with a thin plastic base on which cells are grown, and irradiate
through the dish base (First International Ultrasoft X-Ray Workshop,
1987). A typical experiment (for example, to obtain a dose-response
curve) would require the irradiation of many such dishes.
Whilst cells may be grown on a variety of materials, the best
combination of small, well-defined attenuation across the irradiation
dish base and cell growth compatibility appears to be provided by
1.5Nm Polyethylene terephthalate, C10H8 0 4,
('Mylar' (Du Pont))
(D. T. Goodhead, pers. comm. ). In order to achieve uniformity of dose
to a large area monolayer of cells, very little distortion of the dish
base can be tolerated and so to avoid a pressure differential across
the dish base, atmospheric pressure is required on both sides of the
irradiation dish. This leads to the requirement of a 'vacuum window'
to bring the photon beam out of the vacuum chamber, in which the target
is positioned, into atmospheric pressure. There will, therefore, be a
'flight-path' between the vacuum window and the dish base.
Assuming an irradiation dish base size of - 3cm diameter, a
required uniformity of 14
beam across the base means that the vacuum
window has to be of comparable diameter. However, the vacuum window
also needs to be able to tolerate the pressure differential of
10"3 - 104 Pa on the vacuum side and atmospheric pressure (105 Pa)
on the other, and an unsupported window thick enough to tolerate such
a, pressure differential over a circular area - 3cm in diameter would
greatly reduce the X-ray flux. A thin window (for example, 2.5pm
'Hostaphan', '(Hoechst)) on a supporting grid would therefore appear to
be more 'appropriate. There is then the additional problem resulting
from the 'shadowing' of the transmitted beam by the grid wires.
146
The irradiation dish must not therefore be too near the vacuum window
in order that the natural divergence of the photon beam from the finite-
sized source will cancel out any shadowing effects.
some filtration of the scattered protons, additional filtration will be
required in order to ensure that no protons enter the irradiation dish.
The vacuum window-to-dish base separation may be used as a flight tube
with low atomic number gas (Hydrogen, H2, or Helium, He) flowing at
atmospheric pressure. Whilst Helium would be preferable from the safety
point of view, the increased attenuation in Helium at energies <1 keV
would make Hydrogen a better choice at the lowest energies. Both H2 and
He are efficient at stopping protons (500 keV protons having a range of
< 10cm in both gases) and have lower attenuation coefficients, p, for
ultrasoft X-rays than air. For example (Hubbell and Veigele, 1976):
for CK X-rays
Whilst the vacuum window and the irradiation dish base will offer
'AIR 6.2 cm-1
u-H - 3x10-2cm1 2
PHe 0.6 cm- 1
and for A1K X-rays
For the X-ray beam to achieve a uniformity within 5- 10% over the
' 'AIR 1.3 cm 1
PH - 1.6x10cm1 2
pHe 3x 10-3 cm 1
base of the irradiation dish, the photons need to be produced over a
large target area. The nearer the vacuum window is to the target
position, the larger the required production area since there will be
less distance over which the photon beam may diverge. In terms of
maximising the dose rate to the cells however, the nearer the vacuum
window and the dish base are to the target, the better. One possible
alternative to using a large diameter proton beam bombarding the
target would be to scan the proton beam across the target surface such
that the overall production surface area is larger.
147
Finally, consideration must be given to the Carbon contamination
X-rays, the elimination of which will require additional filtration to
that offered by the window, dish base and flight tube gas. Whilst it
was possible in this study to deduce the BK X-ray yield by subtracting
the CK X-ray yield (taken from the Aluminium target data) from the
combined peak, for biological irradiations deduction of effectiveness
by subtraction of observed effectiveness of mixed beams would introduce
very large errors and would usually not be practical. The necessity of
growing cells on some material, usually polymers, and the need for a
vacuum window (both of which will in general have transmission charac-
teristics dominated by the CK edge) means that it would also be very
difficult to filter out the CK X-rays from a combined beam of Carbon
and some other low Z elemental X-ray without removing the X-ray energy
actually required. X-rays from targets of atomic number lower than
Carbon (Z = 6) have the added disadvantage of having higher attenuation
coefficients in the flight tube gases than Carbon. Hence, even if it
were possible to filter out the CK X-rays produced along with, for
example, the BeK X-rays from a Beryllium target using a Be filter
(X-ray Data Booklet, 1986) it is unlikely that a sufficiently intense
beam of BeK X-rays could be obtained at a practical irradiation position.
For elements with Z near, but greater than 6, a similar problem
exists. - Whilst it might be possible to use a Nitride or Oxide filter
to transmit the NK or OK X-rays respectively at the expense of the CK
X-rays, these elemental X-rays would have had to have been produced
from compound targets such as Si3N4 or A1203, SiO2 and MgO, 2and
hence the'photon beam would also consist of the higher energy charac-
teristic X-rays which could not be filtered out efficiently whilst
still leaving the lower energy X-rays. However, once the characteris-
tic photon energies are >1 keV it should be possible to use a filter,
ideally of the same Z as the target, such that the CK X-rays are removed
148
without over-reducing the yield of the required X-rays. Additional
filters would also reduce the gas tube length required to ensure the
removal of all scattered protons from the photon beam.
Because of the difficulty in achieving a total removal of the
contaminant CK X-rays from other characteristic low energy (< 1 keV)
X-rays, it would appear that, other than for Carbon targets, there is a
very limited biological use for proton-induced low target Z charac-
teristic X-rays. It may however, in a well characterised system where
the Carbon build-up is kept to a minimum and in which a detailed study
is made on the rate-of build-up under specific operating conditions,
be possible to look at the effect of a combined energy photon beam on
biological systems and then, by comparison with the effectiveness of
CK X-rays alone, "deduce the additional effects due to a second X-ray
energy.
A schematic diagram summarising some of these practical considera-
tions is shown in Figure 8.2. Specific distances (such as target-to-
window, window-to-dish base) will vary according to particular systems.
Proton beam energy, available beam diameter, supporting grid wire size
and separation, and practical limitations on target and irradiation
angles all have to be considered.
The beam line shown in Figure 8.2 is by no means a unique solution
to the practical problems of biological irradiation with proton-induced
ultrasoft X-rays. Different systems can be, and are being, used (First
International Ultrasoft X-Ray Workshop, 1987). Attenuation problems
are greatly reduced by actually growing the cells on the vacuum window,
thus eliminating the need for a flight tube. However, in this case it
would be necessary to incorporate filters into the vacuum system to
ensure that scattered protons do not reach the cells and, other than
for Carbon targets, to absorb the CK X-rays. Also, assuming that an
149
jEPi H2 01
3
.1
.
. '_Kra
ý �
1
z
r
Figure 8.2 A schematic diagram summarising some of the properties of a practical beam line for biological irradiation by proton-induced ultrasoft X-rays.
1. Target 2. Window (2.5 pm Hostaphan 3. Flight Tube ( =; 10 cm in length), with H2
or He gas flowing to prevent scattered protons from reaching the irradiation position
4. Variable filter - depending on target - to absorb CK X-rays
5. Irradiation dish (cells grown on 1.5 pm 'Mylar)
\i
150
irradiating beam area of - 3cm diameter is required, the vacuum window
would still need to be supported by a grid so that the cells would have
to be grown carefully in between the grid wires if they were to be
uniformly irradiated. Growing the cells on the vacuum window would
also mean that the area immediately inside the window would have to be
let up to atmospheric pressure prior to the irradiation of each sample
for the window to be changed.
The pressure'differential across the vacuum window could be reduced
by irradiating the cells under reduced pressure. Instead of bringing
the photon beam out into atmospheric pressure the cells could be enclosed
in a 'wet irradiation chamber' where saturated water vapour, at a pressure
of. - 10 4
Pa (0.1 atmospheric pressure), was admitted into an evacuated
chamber. Another alternative would be to bring the photon beam out
through a much smaller vacuum window, such as a narrow slit, and then
scan the irradiation dish across the slit. The reduced area of the
vacuum window might also eliminate the need for a support grid.
8.3 Monte-Carlo electron track structure data
Previous comparisons (Goodhead and Brenner, 1983b; D. T. Goodhead,
pers. comm. ) between experimental data and the calculated frequencies
of energy deposition using Monte-Carlo simulated tracks of randomly
emitted photo- and Auger-electrons generated by ultrasoft X-rays seem
to suggest that energy deposition of > 100 eV in small volumes
( -2nm diameter) may be a critical property when considering the
biological effectiveness of low ionisation density radiation such as
X-rays. The effectiveness of various K-shell X-rays at depositing
threshold energies of - 100 eV within small cylindrical (2nm diameter
x'2nm'length) targets is shown in Figure 7.7, where a certain degree
of structure is observed at the low (< 1 keV) photon energies. This
suggests that biological irradiations with these characteristic X-rays
151
would be an useful indication of the validity of the threshold energy
concept.
However, biological irradiations with proton-induced ultrasoft
X-rays in the region of interest (CK to FK X-rays) are subject to a
number of experimental problems. Irradiations with proton-induced
CK X-rays (278 eV) have been undertaken at, for example, the M. R. C.
Radiobiology Unit. The main problem at this energy is one of attenua-
tion within the cells, the absorbed dose being reduced by 50% over 1.2Nm
depth of the cell (Goodhead, Thacker and Cox, 1979). This compares with
an average 'flattened' cell thickness of - zum (Goodhead and Thacker,
1977). Using the data of Henke et al. (1982) for the mass attenuation
coefficients (p/p) for the various X-ray energies in Oxygen (Table 8.1)
to represent the attenuation in soft tissue, the attenuation coeffi-
cient for FK X-rays is found to be twice that for CK X-rays. FK X-rays
would therefore be a very unsuitable choice for practical irradiations.
From Table 8.1, it is seen that in terms of attenuation, NK and
0K X-rays would be a better choice of low energy photons, but there are
further practical difficulties concerning their production. If proton
bombardment is to be used as a means of producing the characteristic,
ultrasoft X-rays then, as has been already discussed in Section 8.2,
there are problems with scattered proton filtration and the elimination
of contaminant CK X-rays as well as the higher energy characteristic
X-rays, such as SiK, A1K or MgK X-rays, which will inevitably be
produced from the compound target.
It would therefore appear necessary to use an alternative method
to proton bombardment of solid targets for the production of low
energy photons (other than CK X-rays) if these particular biological
irradiations are to be undertaken. A strong contender would be the
use of synchrotron radiation sources which are capable of producing
152
X-ray Photon u line energy p
(e V) (cm2. g-1)
BK 183 1.65 x 104
CK 278 6.04 x 103
NK 392 2.53 x 103
0K 525 1.20 x 103
FK 677 1.24 x 104
Table 8.1 The mass attenuation coefficients (u/p)
of low energy X-rays in Oxygen (from
Henke et aZ., 1982).
153
high intensities of monochromatic ultrasoft X-rays. These are not
confined to producing characteristic energies and would eliminate the
need for filtration of scattered protons and contaminating X-rays.
There are, however, other practical difficulties such as a small
beam size and the requirement for vacuum windows capable of with-
standing ultra high vacuum on one side and atmospheric pressure on
the other, but these problems are not insurmountable. It might, there-
fore, be possible, 'using'synchrotron produced ultrasoft X-ray energies,
to carry out biological irradiations at low photon energies such as
NK (392 eV) and'OK (525 eV) X-rays and compare their biological effec-
tiveness with the frequency of threshold energy depositions within
small volumes as calculated by Monte-Carlo techniques. Such informa-
tion would aid the further elucidation of the critical properties of
ionising radiation action.
154
BIBLIOGRAPHY
(Abbreviations are given according to the
World List of Scientific Periodicals. )
ANDERSON, H. H. and ZIEGLER, J. F. (1977) Hydrogen: Stopping Powers and Ranges in All Elements. In: The Stopping and Ranges of Ions in Matter. (Vol. 3) Pergamon Press, New York and London
BAMBYNEK, W., CRASEMANN, B., FINK, R. W., FREUND, H. U., MARK, H., SWIFT, C. D., PRICE, R. E. and RAO, P. V. (1972)
X-ray Fluorescence Yields, Auger, and Coster-Kronig transition probabilities. Rev. Mod. Phys. 44 716 - 817
BARTON, H. A. (1930) Comparison of protons and electrons in excitation of X-rays by impact. J. Franklin Inst. 209 1- 19
BERGER, M. J. (1974) Some new transport calculations of the deposition
of energy in biological material by low energy electrons. In: Proc. IV. Symp. Microdosimetry (Verbania-Pallanza, Italy, 24 - 28 Sept. 1973 [EUR5122] (eds. J. Booz, H. G. Ebbrt, R. Eickel and A. Waker) 695 - 711 Commission of the European Communities, Luxembourg
BEVINGTON, P. R. (1969) Distributions. In: Data Reduction and Error Analysis for the Physical Sciences. 27 - 53 Mc. Graw- Hill Book Company
BOTHE, W. and FRANZ, H. (1928) Röntgen Radiations Excited by a-particles. Z. Phys. 52 466 - 484
BRENNER, D. J. (1982) Calculation of ionization distributions in a tissue equivalent cloud chamber gas mixture. Radiat. Res. 89 194 - 202
BRENNER, D. J., BIRD, R. P., ZAIDER, M., GOLDHAGEN, P., KLIAUGA, P. J. and ROSSIE, H. H. (1987) Inactivation of Synchronized
Mammalian Cells with Low-Energy X-rays - Results and Significance. Radiat. Res. 110 413 - 427
BRENNER, D. J. and ZAIDER, M. (1982) Proceedings of the Workshop in the interference between Radiation Chemistry and Radiation Physics. In: Argonne National Laboratory Report. (Argonne, U. S. A. )
BUCK, T. M., WHEATLEY, G. H. and FELDMAN, L. C. (1973) Charge states of 25 - 150 keV H and 4He Backscattered from solid surfaces. Surf. Sci. 35 345 - 361
155
BUDD, T. -and MARSHALL, M. (1980) Analysis of Ionization Distributions in a Low-Pressure Cloud Chamber. Radiat. Res. 82 225 - 234
BURCH, P. R. J. (1957) Some physical aspects of relative biological efficiency. Br. J. Radiol. 30 524 - 529
C HA DWIC K, J. (1912) y-rays excited by the a-rays of Radium. Lond. Edinb. DubZ. PhiZ. Mag. 24 594 - 600
CHADWICK, J. (1913) Excitation of Y-rays by a-rays. Lond. Edinb. DubZ. PhiZ. Mag. 25 193 - 197
CHADWICK, J. and RUSSELL, A. S. (1913) Excitation of Y-rays by the a-rays of Ionium and Radio-thorium. Proc. R. Soc. A88 217 - 229
CHADWICK, J. and-RUSSELL, A. S. (1914) Y-rays of Polonium, Radium, and Radio-actinium. Lond. Edinb. Dubl. Phil. Mag. 27 112 - 125
CHARLTON, D. E., GOODHEAD, D. T., WILSON, W. E. and PARETZKE, H. G. (1985a) Energy Deposition in Cylindrical Volumes :
(a) Protons - Energy 0.3 keV - 4.0 MeV (b) a-particles - Energy 1.2 MeV - 20.0 MeV. M. R. C. Radiobiology Unit, Monograph 85/1 (Chilton, Didcot, Oxon. )
CHARLTON, D. E., GOODHEAD, D. T., WILSON, W. E. and PARETZKE, H. G. (1985b) The deposition of energy in small cylindrical
targets by high LET radiations. Radiat. Prot. Dosim. 13 123 - 125
COGHLAN, W. A. and CLAUSING, R. E. (1973) Auger Catalog. Calculated Transition Energies Listed by Energy and Element. At. Data 5 317 - 469
CORK, J. M. (1941) Production of characteristic X-rays by deuteron bombardment. Phys. Rev. 59 957 - 959
COX, R. ', ''THACKER, J. and GOODHEAD, D. T. (1977) Inactivation and mutation of cultured mammalian cells by aluminium characteristic ultrasoft X-rays.; II. Dose-responses
of Chinese hamster cells and human diploid cells to
aluminium X-rays and radiations of different LET. Int. J. Radiat. Biol. 31 561 - 576
EVAN S, H. J. (1962) Chromosome aberrations induced by ionizing
radiations. Int. Rev. Cytol. 13 221 - 321
156
FIRST INTERNATIONAL ULTRASOFT X-RAY WORKSHOP (1987) (Milton Hill House, Abingdon, Oxon., 17 - 18 July 1987) Report in Press.
FLETCHER, R. (1970) A new approach to variable metric algorithms. Conrput. J. 13 317 - 322
FOLKMANN, F., CRAMON, K. M. and HERTEL, N. (1984) Angular distribution of particle-induced X-ray emission. Nucl. Instrwn. Meth. B3 11 - 15
GAIN ES, J. L. (1981) Low Energy X-ray Calibration Sources at the Lawerence Livermore National Laboratory. In: A. I. P. Conference Proceedings No. 75 : Low Energy X-ray Diagnostics (eds. D. T. Attwood and B. L. Henke) 246 - 251 A. I. P.
GERTHSEN, C. and REUSSE, W. (1933) Stimulation of X-radiation by Canal-ray impacts. Phys. Z. 34 478 - 482
GOODHEAD, D. T. (1977) Inactivation and mutation of cultured mammalian cells by aluminium characteristic ultra- soft X-rays. ' III. Implications for theory of dual radiation action. Int. J. Radiat. BioZ. 32 43 - 70
GOODHEAD, D. T. (1982) An Assessment of the Role of Microdosimetry in Radiobiology. Radiat. Res. 91 45 - 76
GOODHEAD, D. T. (1987a) Physical Basis for Biological Effect. In: Nuclear and Atomic Data for Radiotherapy and Related Radiobiology : Proc. I. A. E. A. Advisory Group Meeting (Rijswijk, 16 - 20 Sept. 1985) 37 - 53 I. A. E. A. Vienna
GOODHEAD, D. T. (1987b) Biophysical Models of Radiation Action : Introductory Review. In: Radiation Research : Proc. VIII. Int. Congress of Radiation Research (Edinburgh, Scotland, 19 - 24 July 1987) (VoZ. 2) (eds. E. M. Fielden, J. F. Fowler, J. H. Hendry and D. Scott) 306 - 311 Taylor and Francis
GOODHEAD, D. T. (1987c) Development of radiation sources and dosimetric techniques for radiobiological studies at low and high dose-rate. In: Radiation Protection : Progress Report'[EUR10953] 93 - 96 Commission of the European Communities, Luxembourg
GOODHEAD, D. T. (1987d) Relationship of Microdosimetric Techniques to Applications in Biological Systems. In: The Dosimetry of Ionizing Radiations (Vol. 2) (eds. K. R. Kase, B. E. Bjärngard and F. H. Attix) 1- 89 Academic Press
1 57
GOODHEAD, D. T. and BANCE, D. A. (1984) Proton-induced characteristic X-rays :a versatile source of ultrasoft X-rays for biological and biochemical investigations. Physics. Med. Biol. 29 535 - 541
GOODHEAD, D. T., and BRENNER, D. J. (1983a) The mechanism of radiation action and the physical nature of biological lesions. In: Radiation Protection, VIII. Symp. Micro- dosimetry (Jülich, F. R. G., 27 Sept. -1 Oct. 1982) [EUR83951 (eds. J. Booz and H. G. Ebert) 597 - 609 Commission of the European Communities, Luxembourg
GOODHEAD, D. T. and BRENNER, D. J. (1983b) Estimation of a single property of low LET radiations which correlates with biological effectiveness. Physics. Med. BioZ. 28 485 - 492
GOODHEAD, D. T., CHARLTON, D. E., WILSON, W. E. and PARETZKE, H. G. (1985) Current Biophysical approaches to the under-
standing of biological effects of radiation in terms of local energy deposition. In: Radiation Protection, V. Symp. Neutron Dosimetry (Munich, Neuherberg, 17 - 21 Sept. 1984) [EUR97621 (eds. H. Schraube, G. Burger and J. Booz) 57 - 67 Commission of the European Communities, Luxembourg
GOODHEAD, D. T., COX, R. and THACKER, J. (1981) Do ultrasoft X-rays produce lesions characteristic of high LET or low LET, or neither? In: Proc. VII. Symp. Microdosimetry (Oxford, U. K., 8- 12 Sept. 1980) [EUR71471 (eds. J-. Booz, H. G. Ebert and H. D. lHartfiel) 929 - 939 Harwood Academic Publishers for the C. E. C.
GOODHEAD, D. T., MUNSON, R. J., THACKER, J. and COX, R. (1980) Mutation and inactivation of cultured mammalian cells exposed to beams of accelerated heavy ions. IV. Biophysical interpretation. Int. J. Radiat. Biol. 37 135 - 167
GOODHEAD, D. T. and THACKER, J. (1977) Inactivation and mutation of cultured mammalian cells by aluminium characteristic ultrasoft X-rays. I. Properties of aluminium X-rays and preliminary experiments with Chinese hamster cells. Int. J. Radiat. Biol. 31 541 - 559
GOODHEAD, D. T., ' THACKER, J. and COX, R. (1978) The conflict between the biological effects of ultrasoft X-rays and microdosimetric measurements and application. In: ! Proc. VI. Symp. Microdosimetry (Brussels, Belgium, 22 - 26 May 1978 [EUR60641 (eds. J. Booz and H. G. Ebert) 829 - 843 Harwood Academic Publishers for the C. E. C.
158
GOODHEAD, D. T., THACKER, J. and COX, R. (1979) Effectiveness of 0.3 keV Carbon ultrasoft X-rays for the inactivation and mutation of cultured mammalian cells. Int. J. Radiat. Biol. 36 101 - 114
GOODHEAD, D. T,., THACKER, J. and COX, R. (1981a) Is selective absorption of ultrasoft X-rays biologically important in mammalian cells? Physics. Med. Biol. 26 1115 - 1127
GOODHEAD, D. T., VIRSIK, R. P., HARDER, D., THACKER, J., COX, R. and BLOHM, R. (1981b) Ultrasoft X-rays as a tool to investigate
radiation-induced dicentric chromosome aberrations. In e Proc. VII. Symp. Microdosimetry (Oxford, U. K., 8- 12 Sept. 1980) [EUR7147] (eds. J. Booz, H. G. Ebert and H. D. Hartfiel) 1275 - 1285 Harwood Academic Publishers for the C. E. C.
HAMM, R. N., WRIGHT, H. A., RITCHIE, R. H., TURNER, J. E. and TURNER, T. P. (1976) Monte Carlo Calculation of Transport of Electrons
through liquid water. In: Proc. V. Symp. Microdosirnetry (Verbanis PaZZanza, Italy, 22 - 26 Sept. 1975) [EUR5452) (eds. J. Booz, H. G. Ebert and B. G. R. Smith) 1037 - 1050 Commission of the European Communities, Luxembourg
HAMM, R. N., WRIGHT, H. A., TURNER, J. E. and- RITCHIE, R. H. (1978) Spatial Correlation of Energy Deposition Events in Irradiated Liquid Water. In: 'Proc. VI. Symp. Microdosi-
metrzy (Brussels, Belgium, 22 - 26 May 1978) [EUR6064] (eds. J. Booz and H. G. Ebert) 179 - 186 Harwood Academic Publishers for the C. E. C.
HANDBOOK OF CHEMISTRY AND PHYSICS (1987 - 88) Mass attenuation coefficients. In: Handbook of Chemistry and Physics (1987 - 88) (eds. R. C. Weast, M. J. Astle and W. H. Beyer) E136 - E138 C. R. C. Press
HARNWELL, G. P. and LIVINGOOD, J. J. (1933) Experimental Atomic Physics. ' Mc. Graw-Hill took Company
HEAPS, M. G. and GREEN, A. E. S. (1974) Monte Carlo approach to the spatial deposition of energy by electrons in molecular hydrogen. J. appi. Phys. 45 3183 - 3188
HEDDLE, J. A. and WOLFF, S. (1966) Estimation of the rejoining distance for chromosome exchanges induced in Drosophila sperm by
combined doses of X-rays and neutrons. Int. J. Radiat. Biol. 10 207 - 210
159
HENKE, B. L., LEE, P., TANAKA, T. J., SHIMABUKURO, R. L. and FUJ I KAWA, B. K. (1982) Low-energy X-ray interaction coefficients
Photoabsorption, Scattering, and Reflection. At. Data & Nucl. Data Tables 27 1- 144
HORNYAK, W. F., LAURITSEN, T., MORRISON, P. and FOWLER, W. A. (1950) Energy levels of light nuclei\III. Rev. Mod. Phys.
22 291 - 372
HOSHI, M., GOODHEAD, D. T., BRENNER, D. J., BANCE, D. A., CHMIELEWSKI, J. J., PACIOTTI, M. A. and BRADBURY, J. N. (1985)
Dosimetry comparison and characterisation of an Al K
ultrasoft X-ray beam from an MRC cold-cathode source. Physics. Med. Biol. 30 1029 - 1041
HOWARD-FLANDERS, P. (1958) Physical and Chemical mechanisms in the injury of cells by ionizing radiations. Advances in Biological and Medical Physics 6 553 - 603
HUBBELL, J. H. (1977) Photon Mass Attenuation and Mass Energy Absorption coefficients for H, C, N, 0, Ar and seven mixtures from 0.1 keV to 20 MeV. Radiat. Res. 70 58 - 81
HUBBELL, J. H. and VEIGELE, Wm. J. (1976) Comparison of Theoretical and Experimental Photoeffect Data 0.1 keV to 1.5 keV. In: N. B. S. Technical Note 901. N. B. S. Dept. Commerce, Washington, D. C.
HUNT, S. E. and JONES, W. M. (1953) The Absolute Determination of Resonant Energies for the Radiative Capture of Protons by Boron, Carbon, Fluorine, Magnesium and Aluminium in the Energy Range below 500 keV. Phys. Rev. 89 1283 - 1287
IC RP (1975) Report of the Task Group on Reference Man. ICRP Report No-23 Pergamon Press, Oxford
IC RU (1970) Linear Energy Transfer. ICRU Report No. 16 ICRU, Bethesda, Maryland
IC RU (1983) Microdosimetry. ICRU Report No. 36 ICRU, Bethesda, Maryland
,..
JAMES, F. (1972) Function Minimization. In: Proc. 1972 CERN Computing and Data Processing School. (Pertisau, Austria, 10 - 24 Sept. 1972) [CERN 72 - 211 CERN, Geneva, Switzerland
160
JAMES, F. (1978) Interpretation of the Errors on Parameters as given by MINUIT. Supplement to Routine D506. CERN Computing Centre Program Library, Geneva, Switzerland
JAMES, F. and ROOS, M. (1975) MINUIT -a system for function minimization and analysis of the parameter errors and correlations. Comput. Phys. Conanun. 10 343 - 367
JAMES, F. and ROOS, M. (1985) Function Minimization and Error Analysis. CERN Library Routine D506. CERN Computer Centre Program Library, Geneva, Switzerland
JOHANSSON, S. A. E. and JOHANSSON, T. B. (1976) Analytical Application of Particle Induced X-ray Emission. NucZ. Instrum. Meth. 137 473 - 516
JOHNSON, D. E., BASBAS, G. and Mc. DANIEL, F. D. (1979) Non-relativistic plane-wave Born-approximation calculations of direct Coulomb M-subshell ionization by charged particles. At. Data &ýNucl. Data Tables 24 1- 11
JOPSON, R. C., MARK, H. and SWIFT, C. D. (1962) Production of Characteristic X-Rays by Low-Energy Protons. Phys. Rev. 127 1612 - 1618
KELLERER, A. M. (1971) Considerations on the random traversals of convex bodies and solution for general cylinders. Radiat. Res. 47 359 - 376
KHAN, J. M. and POTTER, D. L. (1964) Characteristic K-shell X-ray Production in Magnesium, Aluminium, and Copper by 60- to 500- keV Protons. Phys. Rev. 133 A890 - A894
KHAN, J. M., POTTER, D. L. and WORLEY, R. D. (1965) Studies in X-ray Production by Proton Bombardment of C, Mg, Al, Nd, Sm, Gd, Tb, Dy, and Ho. Phys. Rev. 139 A1735 - A1746
KWOK, C. S., BUDD, T. and MARSHALL, M. (1981) Track studies in tissue equivalent gas and water vapour with a low
pressure cloud chamber. In:, Proe. VII. Symp. Micro- dosimetry (Oxford, U. K., 8- 12 Sept. 1980) [EUR7147] (eds. J. Booz, H. G. Ebert and H. D. Hartfiel) 347 - 357 Harwood Academic Publishers for the C. E. C.
LEA, D. E. (1955) Actions of Radiations on Living Cells. Cambridge University Press, London
161
LEA, D. E. and CATCHESIDE, D. G. (1942) The mechanism of induction by radiation of chromosome aberrations in Tradescantia. J. Genet. 44 216 - 245
LEWIS, H. W., SIMMONS, B. E. and MERZBACHER, E. (1953) Production of Characteristic X-rays by Protons of 1.7- to 3- MeV Energy. Phys. Rev. 91 943 - 946
LIVINGSTON, M. S., GENEVESE, F. and KONOPINSKI, E. J. (1937) Excitation of characteristic X-rays by protons. Phys. Rev. 51 835 - 839
LLOYD, D. C., PURROTT, R. J., DOLPHIN, G. W., BOLTON, D., EDWARDS, A. A. and CORP, M. J. (1975) The relationship between
chromosome aberrations and low LET radiation dose to human lymphocytes. Int. J. Radiat. Biol. 28 75 - 90
LLOYD, D. C., PURROTT, R. J., DOLPHIN, G. W. and EDWARDS, A. A. (1976) Chromosome aberrations induced in human lymphocytes
by neutron irradiation. Int. J. Radiat. Biol. 29 169 - 182
MEYER, M. A., VENTER, I. and REITMANN, D. (1975) Energy levels of 28Si. Nucl. Phys. A250 235 - 256
NEARY, G. J., PRESTON, R. J. and SAVAGE, J. R. K. (1967) Chromosome aberrations and the theory of RBE. 'III. Evidence from experiments with soft X-rays, and a consideration of the effects of hard X-rays. Int. J. Radiat. Biol. 12 317 - 345
NELDER, J. A. and MEAD, R. (1965) A simplex method for function minimization. Comput. J. 7 308 - 313
NIKJOO, H., GOODHEAD, D. T. and CHARLTON, D. E. (1987) Energy deposition in small size targets by high and low LET radiations. In: Radiation Research : Proe. VIII. Int. Congress of Radiation Research (Edinburgh, Scotland, 19 - 24 July 1987) (Vol. 1) (eds. E. M. Fielden, J. F. Fowler, J. H. Hendry and D. Scott) 87 Taylor and Francis
OPAL, C. B., BEATY, E. C. and PETERSON, W. K. (1972) Tables of secondary-electron-production cross-sections. At. Data 4 209 - 253
PARETZKE, H. G. (1978) On limitations of classical microdosimetry and advantages of track structure analysis for radiation biology. In: Proe. VI. Symp. Microdosimetry (Brussels, Belgium, 22 - 26 May 1978) [EUR6064] (eds. J. Booz and H. G. Ebert) 925 - 935 Harwood Academic Publishers for the C. E. C.
162
PARETZKE, H. G. (1980) Advances in Energy Deposition Theory. In: Advances in Radiation Protection and Dosimetry in Medicine. (eds. R. H. Thomas and V. Perez-Mendez) 51 - 73 Plenum Publishing Corporation
PARETZKE, H. G., LEUTHOLD, G., BURGER, G. and JACOB, W. (1974) Approaches to Physical Track Structure Calculations. In: ' Proc. IV. Symp. Microdosime try (Verbania-Pallanza, Italy, 24 - 28 Sept. 1973) [EUR51221 (eds. J. Booz,
" H. G. Ebert, R. Eickel and A. Waker) 123 - 138 Commission of the European Communities, Luxembourg
PARETZKE, H. G., TURNER, J. E., HAMM, R. N., WRIGHT, H. A. and RITCHIE, R. H. (1986) Calculated yields and fluctuations for electron
degradation in liquid water and water vapour. J. chem. Phys. 84 3182 - 3188
PAUL, H. (1984) An analytical cross-section formula for K X-ray production by protons. Nucl. Instrwn. Meth. B3 5- 10
4....
RAJU, M. R., CARPENTER, S. G., CHMIELEWSKI, J. J., SCHILLACI, M. E., WILDER, M. E., FREYER, J. P., JOHNSON, N. F., SCHOR, P. L., SEBRING, R. J. (1987) Radiobiology of Ultrasoft X-rays. I. Cultured and QOODHEA) P-T Hamster Cells (V79). Radiat. Res. 110 396 - 412
R EV E LL; S. H. (1974) The Breakage-and-Reunion Theory and the Exchange Theory for Chromosomal Aberrations Induced by Ionizing Radiations :A Short History. In: Advances in Radiation Biology (VoZ. 4) (eds. J. T. Lett, H. Alder and M. Zelle) 367 - 416 New York Press
ROSENBLATT, J. (1968) Particle Acceleration. Methuen and Co. Ltd.
ROSSI, H. H. and ROSENZWEIG, W. (1955) A device for the measurement of dose as a function of specific ionizations. Radiology 64 404 - 411
RUTLEDGE, C. H. and WATSON, R. L. (1973) Cross-sections for K-shell ionization by 1H, 2H, 3He and 4He ion impact. At. Data & Nucl. Data Tables 12 195 - 216
SAVAGE, J. R. K. (1975) Radiation-induced chromosomal aberrations in the plant Tradescantia : Dose-response curves. I. Preli-
minary considerations. Radiat. Bot. 15 87 - 140
SAX, K. (1938) Chromosome aberrations induced by X-rays. Genetics 23 494 - 516
163
SAX, K. (1940) An analysis of X-ray induced chromosomal aberrations in Tradeseantia. Genetics 25 41 - 68
SAX, K. (1957) The effect of ionizing radiation on chromosomes. Q. Rev. Biol. 32 15 - 26
SCIENCE DATA BOOK (1978) (edt. R. M. Tennent) Oliver and Boyd, Edinburgh
SLATER, F. P. (1921) The excitation of Y-radiation by a-particles from Radium Emanation. Lond. Edinb. DubZ. PhiZ. Mag. 42 904 - 923
STERK, A. A. (1965) X-ray Generation by Proton Bombardment. In: Advances in X-ray Analysis (Vol. 8) 189 - 197 Plenum Press
STERK, A. A., MARKS, C. L. and SAYLOR, W. P. (1967) Production efficiencies of X-ray emission spectra by proton bombardment. In: Advances in X-ray Analysis (Vol. 10) 399 - 408 Plenum Press
STORM, E. and ISRAEL, H. I. (1970) Photon cross-sections from 1 keV to 100 MeV for Elements Z=1 to Z=100. NucZ. Data TabZ. A7 565 - 681
TABLE OF ISOTOPES (1978) [7th Edition] (eds. C. M. Lederer and V. S. Shirley) John Wiley andjSons, inc. '
TABLE OF X-RAY EMISSION ENERGIES Princeton Gamma-Tech, Europa.
TERRISOL, M., PATAU, J. P. and EUDALDO, T. (1978) Application a la microdosimetric et a la radiobiologie de la
simulation due transport des dlectrons de basse dnergie dans l'eau et 1'etat liquide. In:; Proc. VI. SUmp. Microdosimetry (Brussels, 'Belgium, 22 - 26 May 1978) [EUR60641 (eds. J. Booz and H. G. Ebert) 169 - 178 Harwood Academic Publishers for the C. E. C.
THACKER, J., COX, R. and GOODHEAD, D. T. (1980) Do Carbon ultrasoft X-rays induce exchange aberrations in cultured mammalian cells? Int. J. Radiat. Biol. 38 469 - 472
THACKER, J., GOODHEAD, D. T. and WILKINSON, R. E. (1983) The role of localised single-track events in the formation of chromosome aberrations in cultured mammalian cells. In: Radiation Protection,, VIII. Symp. Microdosimetry (Jülich, F. R. G., 27 Sept. - 1 Oct. 1982) [EUR8395J (eds. J. Booz and H. G. Ebert) 587 - 595 Commission of the European Communities, Luxembourg
164
THOMSON, J. J. (1914) Soft RSntgen Radiation produced by impact of positive and slow kathode rays. Lond. Edinb. DubZ. PhiZ. Mag. 28 620 - 625
TURNER, J. E., PARETZKE, H. G., HAMM, R. N., WRIGHT, H. A. and RITCHIE, R. H. (1982) Comparative study of electron energy deposition
and yields in water in the liquid and vapour phases. Radiat. Res. 92 47 - 60
TURNER, I. E., PARETZKE, H. G., HAMM, R. N., WRIGHT, H. A. and RITCHIE, R. H. (1983a) Comparison of electron transport calculations
for water in the liquid and vapour phases. In: Radiation Protection, VIII. Symp. Microdosimetry (Jülich, F. R. G., 27 Sept. -1 Oct. 1982) [EUR8395) (eds. J. Booz
and H. G. Ebert) 175 - 185 Commission of the European Communities, Luxembourg
TURNER, J. E., PARETZKE, H. G., HAMM, R. N., WRIGHT, H. A. and RITCHIE, R. H. (1983b) Calculated yields and fluctuations of electron
energy losses in water in the liquid and vapour phases. In: Proc. Int. Congr. Radiat. Res. 'VII., Session A (Amsterdam) Al - A53 Nijhoff, The Haque
VIRSIK, R. P., GOODHEAD, D. T., COX, R., THACKER, J., SCHÄFER, CH. and HARDER, D. (1980) Chromosome aberrations induced in human
lymphocytes by ultrasoft Alk and Ck X-rays. Int. J. Radiat. Biol. 38 545 - 557
WILSON, W. E. and PARETZKE, H. G. (1980) Calculation of ionization frequency in small sites. Radiat. Res. 81 326 - 335
WILSON, W. E. and PARETZKE, H. G. (1981) Calculation of distributions of energy imparted and ionisations by fast protons in
nanometer sites. Radiat. Res. 87 521 - 537
WOLFF, S. (1959) Interpretation *of induced chromosome breakage and rejoining. Radiat. Res. SuppZ. 1 453 - 462
X-RAY . DATA BOOKLET (1986) (edt. D. Vaughan) 2.13 - 2.17 Lawerence
Berkeley Laboratory, Univ. California, Berkeley, California.
ZAIDER, M., BRENNER, -D. J. and WILSON, W. E. (1983) The Applications of Track Calculations to Radiobiology. '\I. Monte-Carlo Simulation of Proton Tracks. Radiat. Res. 95 231 - 247
165
APPENDIX 1
SMBRO TINE FCN(NPAR, G, F, X, IFIAG) C C GAUSSIAN FITTING: T1 GAUSSIANS ON A CONSTANT BAC[E(JND C
DIMENSION DESCP(40), TITLE(40) DIMENSION DRAW(90) INTEGER SPEC (1024), DRAW, BlANK, DOT, PLUS, LINE, STAR, RUB DOUBLE PRECISION T EO1(1024), TED2(1024), THEO3(1024), R, R2, RR DOUBLE PRECISION AVX1, AVX2, AVX3, Sl, S2, S3, BCK DOUBLE PRECISION X(15)1FWHM1, FWHN2, FWHM3, AR1, AR2, AR3 DOUBLE PRECISION SUM(1024), CEII(1024), EPR(1024) DOUBLE PRECISION XMAX, XNIIN, P, Q, T, RM, RN, U, V DATA BLANK, DOT, PIUS, IXNE, STAR/'
C C PARAMEII; R DEFINITIONS C
BCX X(1) AVX1=X(2) F. ' M=X(3) AR1 X(4) AVX2=X(5) FWHM2=X (6 ) AR2=X(7) AVX3=X(8) FWHM3=X(9) AR3=X(10)
C C READ IN RAW DATA C
IF (IFLAG. GT. 1) GO TO 25 READ(5110) (TITLE(I), I=1,40) BEAD(5,10) (DESCP(I), I=1,40) READ (5,15) NPI'S K=(NPTS/5)+1 DO 30 J=1, K NUJ*5-4 N=J*5 IEAD(5,20) RUB, ZERO, (SPEC(I), I=M, N)
10 FORMAT (40A1) 15 FORMAT (14) 20 FORMAT(/, A1, I4,1X, I6,1X, I6,1X, I6,1X, I6,1X, I6) 30 CONTINUE 25 CONTINUE
DO 35 I=26, NPTS R=DFLD1T(SPEC (I) ) ERR(I)=DSQRT(DABS(R)) IF (DABS(ERR(I)). LT. 1.0D-03) ERR(I)=1. OD+00
35 CONTINUE C C COMFTTE ! IHE RETICAL FUNCTION C
166
C C FIRST PE1K C
DO 40 I=26, NPTS 1N=I-1 IF(AVXI. EQ. O. OD+00) CO To 50 FAC1=AVX1-DFLOAT (M) IF (PS HM. LT. 1. OD-03) FWHM 1. OD+00 F1=FAC1/F HMJ. ARG1=-2.7726D+00*F1*F1 IF(ARG1. LT. -10) GO TO 50 S1=0.9394/FWHM1 THEOl(I) AR1*Sl*EXP (ARGl) GOT040
50 THEM (I) =0. OD+-00 40 CONTINUE
C C SECOND PEAK C
DO 42 I=26, NPIS N=I-1 IF(AVX2. EQ. O. OD+00) GO 'P0 52 FAC2 AVX2-DFLOAT(M) IF(FWHM2. LT. 1. OD-03) FVDV=1. OD+00 F2=FAC2/PS 7HM2 ARG2=-2.7726D+00*F2*F2 IF (AI 32 . LT. -10) GO TO 52 S2=0.9394/FWHM2 THE02(I) AR2*S2*EXP(AI32) GO TO 42
52 THEO2(I)=O. OD+00 42 CONTINUE
c C THIRD PEAK c
DO 44 I=26, NPTS ICI-1 IF(AVX3. EQ. O. OD+00) GO TO 54 FAC3 AVX3-DFLO T(M) IF(FWHM3. LT. 1. OD-03) FWHM3=1.0D+00 F3=FAC3/FWHM3 ARG3=-2.7726D+00*F3*F3 IF (ARG3 . LT. -10) GO TO 54 S3=0.9394/MUM 'HE03(I) AR3*S3*EXP(ARG3) GO TO 44
54 THEO3(I)=O. OD+00 44 CONTINUE
C C CALQJLATE TOTAL =RETICAL VAIÜES C
DO 60 I=26, NPrS SUM (I) =TH]. (I) +'I%ä702 (I) +'IHID03 (I) +BCi(
60 OONTINUE
167
C C miSQUARE OVER ALL CHMNEIS C
F=0.0D+00 ISI RI=26 ISTOFWTPI'S DO 70 I=ISTART, ISTOP R2=DFLO T(SPEC(I)) SPECPI`=(R2-SUM(I))/ERR(I) CHI(I)=SPECPT*SPECPT F=F+C I(I)
70 CONTINUE IF(IF? AG. NE. 3) GO TO 80 FES =FIOAT(ISTOP-ISTA C) FIDIM4=FEXP-SQRT (2*FEXP) FtJPLºII@FIPFSQ T(2*F XP) CHIIPC H=F/FF. }P
C C OC rair RESULTS C
WRITE(6,100) (TITLE(I), I=1,40) WRITE(6,100) (DESCP(I), I=1,40)
100 FORMZ T (/, 4OA1) WRI'T'E (6,110) F, FEXP, FLOLIM, FUP'LIM
110 FORW(////, 20X, 'CEIISQUARE IS: ', D10.3, //, 1X, 'EXPECI'ED VALZJE IS: ', 1F7.1,1X, 'WITH LIMITS: ', F7.1, lX, 'AND: ', lX, F7.1) WRITE(61115) CHIPCH
115 FOPMAT(//, 1X, 'CHISQUARE PER allUU L IS: D10.1) WRITE (6112 0)
120 FORMAT(///, 5X, 'PEAK POSITION', 5X, 'F. W. H. M. ', lOX, 'AREA') WRITE(6,130) X(2) , X(3) , X(4) , X(5) , X(6) , X(7) , X(8) , X(9) , X(10)
130 FOFMi T(6X, F7. O, 11X, F7.0,4X, F10.1) WRITE(6,140) X(1)
140 FORM7T(//5X, 'BACK I UND IS: ', F7.0) C C GRAPHICAL OUTPUT C
XMAX=O X IIN=1E08 DO 1000 I=ISTART, ISTOP IF(SUM(I). LT. 1) SUM(I)=1 IF (SPEC(I). LT. 1) SPEC(I)=1 IF(SUM(I). GT. XMAX) XMAX=SUM(I) IF(SUM (I) . IT. XMIN) X[IT=SUM (I)
IF(SPEC(I). L2. XMIN) XrIr=SPEC(I) IF(SPEC(I). Gr. XMZx) XM X=SPEC(I)
1000 CONTINUE WRETE(6,1020) XcIN, X X
1020 FORMAT (/8X, ' YM]21 ', F8.1,17X, ' IAGAi2II IC SCALE' , 23X, ' YN X=' , F8.1) DO 1100 I=1,90
1100 MAW (I) =ILINE WRITE(6,1030) DRAW
1030 FORMAT(/4X, 'Cii'12X, 90A1,5X, 'CAL', 5X, 'OBS'18X, 'CUI', /)
168
P=DLDG10(XMAX) Qý=DIDG10 (XMIN) T=89 ./ (P-Q) K= (NPIS/5) +1 DO 1400 J=6, K I=J*5-4 IF(I. EQ. 1) I=2 DO 1200 L=1,90
1200 t 1W(L)=BLANK PR=DFIOAT(SPEC (I) ) I DIOG10(RR) RN=DICG10 (SUM (I) ) ur* (PM-Q) V=T* (P-Q) DRAW(U)=DOT DR1 W (V) =PLUS IF (U. EQ. V) rRAW (U) STAR M=I-1 WRITE(6,1040) M, DRAW, SUM(I), SPEC(I), C: HII(I)
1040 FORMAT(2X, I4,2X, 90A1,1X, F7.1,4X, I5,4X, D9.2) 1400 CONTINUE
WRITE(6,1060) 1060 FOFIMAT (/, 40X, ' SYM3DL: . ABS. , 4-CAL., *= BS. AND CAL. ')
80 CONTINUE RETURN END
169
APPENDIX 2
The absolute frequency of energy depositions greater than a given
amount in each elemental cylinder per Gray of absorbed dose to the
medium
The tables are headed by the photon energy for the particular
scoring run, and the number of interactions considered. The diameter
of the target cylinder is given and also the number of infinitely long
cylinders that were hit (a measure of the statistics of the run). The
body of the table shows the frequency of hits of energy greater than
E, and data for four different lengths of cylinders are given. The
frequencies apply to a single target cylinder placed at random in water
irradiated uniformly with an absorbed dose of 1 Gy. In some cases, the
last values in each column are subject to very large uncertainties
(Equation 7.3) resulting in the repetition of the same value, and may
therefore be ignored.
The frequency averaged mean specific energy in the targets, z, is
given at the base of each column. This was calculated from the full
frequency distribution of hit sizes and independently of the absolute
normalisation of the scoring frequencies in the Table. Let f(E) be the
frequency of energy depositions ('hits') of energy between E and E+ dE.
Then the frequency mean hit size, E, is given by
Co E_0E. f(E). dE
A2.1
ö f(E). dE
irrespective of the normalisation of the frequency distribution.
Hence, the frequency mean specific energy is
z=E A2.2 m
where m is the mass of a target (I. C. R. U., 1983; -Goodhead, 1987d).
170
(In I. C. R. U. notation this frequency mean specific energy is written
as ZF (I. C. R. U., 1983)). The probability of a hit of any size is given
by z-1 and should, with ideal scoring, equal the frequency of a hit
greater than 0 eV (f>o). The difference between these values is a
measure of the statistical uncertainty of the data. The 'scoring
efficiency', S, is defined by:
f>o - z-1 I S (%) _ -1
x 100 A2.3 z
Finally, the number of hits from which the statistics are dervied is
given at the bottom of each column.
171
dý m o00 . ýo o oinn . -1 1(1 NH
41 3i
a
o öö
zHº,
0
dý om
mo .v OOON rl 1[1 rl /'1
wo
0ý0
d0 ö0
"b
u
ýwa"1
o ill
PIAAAMMPI r1r, r) r, O
44444444444 4 N
4 ° wlNNN00Nawwww w °
° 0% No O S
040 nM4 D 0% 4% in
00440 1
. "i0O%9 00n %040181/1 PIO
Co
rl NN
P1r NNNNNNNNNO rl
AhAhAM11'1 N1 AP1 f'1 O
444444444444 N
4 w
Ö ° N In Y l ind
V 1'1 P1F)NNr-I
N
rl "rl rl rl r* p4,4 Nº1 rlO
N0) rl
nÖN Uli i-1O 1 N
ar 1fl N
0%U) 0)N so d(p (O In EOO 1(1PIN01 O n ßßr) N. 01n
Ul N
N fNV
eýýrýeriýeuiö e .i
X44 44 4444 a wýalýw1fýwwwwwwww r" c) 0 1 w ° o,
[1 V te W bb r N 0 0 0 N NN "i W 1 r I '1 -10 )
ri P1 P1 ei ri r) mm r) F) NO N
w OOOOOOOOOOOOO
ii N fl d in 4O r co Of c). 4
fV 1
i ý k
r N
_1 y
" ddddddddddO PI
44 HUHN www P L Co oo oý 1NI ) Yl O1
fiý0Smfi Co : °Ný Co
:Q N
44 m NNNN rl 04 . 4.4.1.4.40 d
d . 4. ein inekn to Lnint1O
44444444444° P1
Co O wwwww ý}
ý1f.. ý]
wwww Co 0 1
wyý d
O 0) Ö
º1 bVV
O) NNÖ .i N F) N ri e"1001 d00 )O)C1O NO Pf
AAA0; %U%0ooNNN%0 ö Co
r-1 ý'
in IAU1 in in Kn in IAU)in inO
44444440444° d 0
° o {1 .. ý( . -]i
wwwwwww w w 0 1
w (0 d NO ý r 11 fV
(%. 1.40 dNN OOi ON 111 b
N ri c%NNO)r dN 0%0%Ulr10 4O
If1 If1 Y1 dVd 'V P1 r) P1 r) O
ri
I[) to tn U1 in U1Ifl Il in tn V1 O d C" 4444444444 z 4 ° o rý. wwwwWawww ýo Cl A 01 w 'o 0%
1f1 : 10rn
U) 0)
0 dn ý""ý O Ifl 0 N
fV P i
m9IloinnN. +mmnin° M NNNNNNNrl rl IIlI 0 m
w 0000000000000
A N/'f din '0N CO 0A 2e-IN ý-1 '"1 lýI v
tll Jý
N Wý
40%0%0 %D%O%Orrrrr O
4444 4 444 4444 U)
° ý] [aý (aý w1 14D (yCl fw
d .4
NN N w ý, °b
COO r l 1 l
d '0 'O nOMd %D ri r1 P1 O N ri 40 CO d 1(1 . ""1 40 N PI O C) U) rO If1
Pf P1N N. i. -IO)n dP1r10 N
ýoýoýcýonnrnNNNO 444444444 4
tn 4 o www1aww14 414114 1) o 00 [ý] r o CO
d dd (
1(ý1 0) Cl O1
mn Co 00.44 ÜV Nb ö b di % A %O
rI rI r1 ei C0 ýO dAN r1 0D 0 Ifl
ri 0nn00000 3 r, Co Co 0 444444444444 tn
°+ o ! wwIawNawwwww O O w e 05 0 y cl
Nd N
O) N/0) O1 od wn)
0 Onrn01vo
O O' dd0 01 N lO .1 01 WO 0° p1
ph ';
0.1 CO bMm MNrlr"1el MO 0%
NN NN NN 000%CO CO0
4444444444 44 4o
4 o wý. lwwwwwww www m 0NOO °. ooodO 1 w rn 0 r-1 0 NO ý N O
NO 0) NCO O F) O IO nodrmýnloNO n N r
in If1 MN rl . 04 l0 dmHO e1
w 0000000000000
r1NMI d U10%N W 01 r0-IýN .1 v 1ý
N (Aý
172
1o MMMMMMMMMMMMM ei MMMMM0
44444444444444444444 IO aw lwww
N
4 N r1 W ww 2 CO o0 %l0 to mW 1[1 BOO M W 1 ý] { n
W O
w c2 W ON OO Oh WNnMn OýýO N0ýý0 dO P. T in o om Co Wn %o %0 %0 U, in in r .s .rm MO
'0 W m
N ry
'Ty N"-l rl ri ri rl r-1 rl rl rl "-l rl rf HMmm r4, -l0 ' W FCý7
MO° O I( Oý
.. ý, NN ý" MdVV sº d? V a} a} V .º ýº t dr dO
4044444444444444 N
4 "' "" »"""" 1 o ö wwwýarý. wwWeiriNrNW 0)www
MW OO, m Mm 01 0 V "f M OW ri M m 1 O1O 0
+1 to N WnV Co MO1V eO M Np 7O Ln Omr Oh Co ýO v Mn Of m%01f1M 44 0
M fý
N N r
ri 010101 Co Co to Co Co Co NNNNN hNIOI0O 01
UI M
-1 . VVVVVV ýf VVVVVVeeeo
44444444444444444 M
o wwwwwwýawwý. lrýýawwwwýaýaww MMIll l()O 01 0 OI NM Oý aý0 1 ýo rm oý dM N
iý O CIN
M'P 0NMPý00% CO f . 1ofN WW U) oO 001 P ý0 .! V MNý00ý01 W PLO b If1MN0 V
pý NN
HH I)V O'V V' .1wV MM MMMMMMM0 I. 1
Me 49) a
f+ý
V Vif V"! V V VeTV . fV VVV VO 444444444444
M
° y. ý C
w wuý. ýaww wrý. wwwwww
wwwww
ýO 01 WO OIA M'0 Wo NO. t m in C) 1
w In
m ü
ö+l 44
N rl
ON 1A . -1 .! O w(y W MOý10 NWV. U1m NO Yl d1'1MNNr-I r-10001 p1C W CO mhW (00
M Co
NN
O Ft O `ý
NNNNNNNNNNei ri "-I rl ri º1 '1 A
. 10
rj
E Aý W
OOOOOOOOOOOOOOOOOOOOO r-INMYI[l bfr W 01ýýNtI H1(lWNw ci0
rl rl .i r1º1º1 1N tIý
1 Q r N
vy y
(7 d dd -q- -e dd dd tnm U1 NN to Y1 U1 1A O 044 4ý°4444444444444 m
4 o O 0 tn WwýaWww O ý-1 1[1 N
0% N, ObON
1[1 O 1/1 On Oý1'1 Co OD /'1 O1 Pln rl d Nd2M/'1114Ö w N oo d
`0 Wd ýO 0 1"1 N (4 -I 14 Co M Co Pl op vn -I d0 W N 1 1
J 7ý rl r1.1. -1. -11-1 .1 rl H. i at o% co co n n%o%o mO Ln
000 W00 Co "+ ýº+r+
C" to to to in to to ýn to in in to to ýn nm in to o 00 0 o14
14001 11 4444444 ww v 0 .... .. -
+ io ö
wwWwwlwwwwwrw4ww www
rnU)WO mrýr' NO ui n noomwo 1 ?On il w lI1º11f1 U1n 1'1 %ON%O NOXO %D O
it m
n Ov
W NN
F. 1 N N NOS dr-1 OX lO dr1 O1 XOd r10, n P10 NO
Co Co n%0b %0 Ln in in Ul Vdddm mMmN0 . -1
r-1
o öIA
ö to in IA Öö
1(1 in I(1 in ö
wp
Ln U1U1H1NO 4 444 4444 1, HNZ www w wwwwwwww +
3ý d O%0ý mO %0 r, 0N ONý NfF'1v VO1 (. ý
b Co
Co O rl
NO F)n mOX ul NN O%%OP1 . -1 co 'O U Co m SO O r10 ýOd NOOK c2 r, U) d P'NO OX 9 X0 U)F)0
V 1'I P1 P1M NNNNNNNN vI I l l l
U) P7
Nm
r r v r 0 N
0
O
U1 to U1e ui in IA IA U1 ui vi U1 to N%o b%o %D %oO
WWWWWWWWwWI WWWWWWWf+IW w
F) ('yw n
ýY
O If1
ý n 010%ýO%0 NI Nd1! )d «4 r) Co 0 00 O P1 fit in Nd Co 1-1 ýO Oý 11 ýO 01 1ýf t0 U) Co O1 . -I O CO n 4O U) dd F) NN .400 4D
ýNdd0 tn 1-4 4O
m N F)
00 a w
v
wS
N 44 14 rl A rI vl ri. -l ýI ri rl e. 4 vl 0% of CO r- %00
OOOOOOOOOOOOOOOOOOOOO 14 NF)dl[)%O n CO 0% oß-1 1'ld wn m0n o
d
N M
U'' ýo 10 40 4
j a
0 '0 10 NNNNN fN NN Co CO CO
4 c0 m4O
04444444444444444 ow wwww4
U n
4 ý ýw (±1 wwwwwwww In 0% 01 0N NC) Nul CO pldryNMN100 vi Co w
Co A r. Co 0% r, 4 C) 10 M Ö 11 Co ri AM M MOI V CI If O1 01f110 M V NN91 r10 MMNH rl H 01 P %0 Ve MNH rl co MMAOO
0%
N
d
M8°. I
°' NF-NPnrnrn Co Co Co Co Co 4 rn o
444 444444444444444 in
4 .+ :=
o wwww 4kýww14 WW 4l ww ww ww ww bM N U) kl M0M 000' m0 ,
w Q
M Co Co
'. 47
ý1 0% 1ý 040 in Nin O r1r Ow OU)fý r. Co in 0 ýO I(1 CIN M01NNO .M wýO rI I()n ý0oo
ý
N CO
A 0%
ý-1 ý-I r-I O P Y1 V 1"I MN rl r1 01 ý0 VN r1 ýO N0 in
CI NNNNNNNNNNc CO IDID ornrn oo . 444 444 4444444ýý+ o
%o ryýýý 10
to ww ýa ww ýl {ý w w'
0% Co 0% N CO COO r10 V) 0 ýy MOM Of )O VO
(ý7 v rn 0 . 4N N 0 C ANN 0ý1ý01 MfýM --t fýM V OMaO . -1O
Ö
.4 O%COIA V MNNr-1r-1 ri N1flMNei OIU1r1InO r-.
UvY ý
. NNNN1-Nc0t0aD[000 W 0%0%01010%OrlO
40444444444444444474 b
owww 4ý ww w+ Www Co ID
wwww 4ý ww 01040 oO1M CI Nw mCo ro m m r w .ý M
W 1
%4 OH
lNC OO a0N O fl -IN COC of In In Co m NoO 1(1 CO M ýO NNVI ýO oMY ýp 10 40 fV o o
n rj
14 O0 in V MN rI rI 0% %OVMN .4 CO VNr4,1O .4
W OO 000000000,00 OOOOOOO S
ý-INM YY1 ýO fý (D 010 r1ry M Vuý 40F-tOmO ý-1 r-1 ý-1 r-1 r-1 r1 rl r-1 rl rl N v
yý
N -3
173
Co 09,4 rn No to N N IA NN1
41
. -1
M ryl W FG
[ý O (((aaa Ö
1ý Ö
X
yAý
äi E
rGN o0" m
N oo%0 N 1f1 r1 V
o ill 0
dý Co ogm
0NMN I1
öº} ö
z Nah N
1'
°; W-? o 4°4Q4444444 o wwN wwwww wý wwww wwNwwý w w co m . 1w "ON Go Nmao ONNý Mo Om oll wN
O 0 Ch ý11'Ih
mm NNNpý 1ýyýýy OOOO NO
w Iomry 0)ow w r, aº ONN 1'1 V PI PI PINNNr-Ir-I rd rl ýi00000t0 X011 r-101 fý V r10ý1(10 n 14 It
1-1 rl vl rl rl. i ri rl rl rl ri rl rl ei rl ri rl º1 OýOýOý Oh cow 10 t0PP o to
OON
V MW U
11{1(. Jý
0{ryaý 0 v({. ýjý f1Maýin ONN Ip 0 1ýNO of 1 Y1 O NOON Vy mv mIflNrl O r-1 Oývl ý-1 r-1O ýO fýfý Vpý aDr OvNM 1! 1 V4 cO'I 01 aD PLO hY M'!.
. C1 CO
. U1 ýpMN0011ý OM
Ntoo 1040 1010 In to in to IA in in in in in VrVVVVVV MMO vl
o wwý"q' w wýwww ýwýwww wwwww w. a mM Nv Nve Mono rn loom NuiMý 11 Ch N cri N0% NOýýy inrnrir- .f In 0 In0InOOV M Co ryV m coO Co NM N lO In VM NrN i ºi 00% 0% CO mnn Ui 0 ID %0 In In o Ul MN Nrl r-IO 0%ID0
MM MM M MMMNN N NN NN NN NNNNN 94 44 fV M rlp fy
ýg4ý in wýýwww ý g ýwwwwwwýwwýwýw wm v P1 ww
Ohm m OhN mOm Pl ON 00 d1O11 Oý N 14 ch N to P1 01 %0 N OS I0 fN U1 Pb 0N Ul N0NvN Of
-9901, e9
%O NN PI r-/ mNNý0 ý0 Ui Ui Ui tVvV P1M nNINNNNr1 ý-I rI ON
14 rl .i . -1 ý-1 rl rl ý-1 rl rl rl ri rl rl .i rl rl rl rl rl e-1 rl rl ei rl rl Oý Oý O 44
0000000000000000000000000000000 WSNNNYnbnmgo 14 : 4: 3 V 1fý bnma0NN Mw in NNbNwa0 r414 r-1 r-1 rl ý-l eýl rI NNNNNNN NNN /'1 Vv yJ
N Vl
4kn in 4°°Y4°4tn °&n in in öLn in tn in in in 0 ö O O
A W WWýWý+ikWWý1WWýIW WW WA WdWWWW p WW W M N d r Il Nri ö 11 1 « 11 (+ý M o%
Wd M
0% % 40 oO11 O . ry (p oOZ U) N0
M d10n 0N 0 I(1 Nn 0 PN Il 1 fV Co N rl lD U1 N 01 b Mon d1 r! 01 lO Nod o
bb
dd rl d . . A14H 14: f iof
of Co Co r ri ez bbb
ui ui AAvvv ri ri mm 44 ö
[+1
!! {y U1 gn eN to to in in in in U11(1 in to to in to I v) in U1 in in U1 w% in U1O 40444444 4444444444 44444 44444° ý© 1 o
ö ý www wwwww W 14 +ýlwwwwww ww W ww4lww W w2
Co N OýNCO fw d MMI(1 M 19N IAIl d ý6 MNO . 11() 0) ri U1 OII N NN Co MN MO W'0 dIA c00 P(001llOb N(0 dm 1N M N. -1 M0
,. wM 01 in %0 N 10d OD'. ONOOD M Mri Olt bd N. 101 CO b in m NO 011O'. 0U1N O
Ob-'n-U1d VV dMMMMMMNN NNNN r4 mH rl mH0
Co d d
r1
o
in Ln in 1fl Ifl 1f1 U1 I! 1 kn in U1 e in U1 in U1 e to in Ifl N1(1bbbýDbO OO 0000 OOOOOO0400440
. 44.4441.... 4.. 44 4. wwcý. ww lw wr . ww lw
,
° O
1
mý ý ý wýa ( wwww
11 1l1 Md NN d QO M MN OWV 01 ýNOI O Nd 1/1Ny NO0) W NN OZON d1/1 O Mb(. -1dN d ONM O N0)10101[1 dN 1001 c o M
ö , "1
ýI r r 0 DN ýD UId dMN ri r10(0 r1 'tu MMNNNNNNNýI, 4 of rl M rlrl v1 r1 r1 r-ýri ý-1 pýpýap .b
N
N l 0
zz ýJ O
1A 1f1 U1 LA ßf1 kfl in in U11f11f1 bbbbbb %o bbbbbb b'. O'. p bo 4444404444444444444444444444° 4{w l4www +iw Wwww WWwW wwwwwww w b
j L IA
ý wwl14www `I,; N UIN 0( OM[O NMh M 00 MIA NN 00 MIA O) d sr d0 11 bP 0 01 r-I Y101 NIl1 O MOr11t101 N
MI +1 [O rl
kýaif
d Nrl Nd'. O M01 b1t, U)o Co 1ý 1f1 ýf MMNrl r1OO1(1 O 1/10 dOtO N a0 dO Md. -1dO AAAAAAAA 44 x10101 Co Co 1. hb %0 U1 1A Uld e MMMNO
d ý-1 In d
j t
W OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
/-IN Md If1101. m01 014 M d14 0I l4
0 rI NMbr, Co C% O (4 ý"
ý. -4.1 '4 r
01 lNNNNN fo NN ri N f4 NNM v
N f/1
u' 10 10 10 10 10 to nnnnnrnrn to Co Co to to to to Om o. o) °° 4444444444 444 000 000000.. 0 4 4i1I411
An
0 o 111,, + wwwýIwwwwwwwwwwwcýwwýarý. rý. wwwwww m to 0 NN Oo mm NN 10)0 Mm ryO +r (. ý e ,, m vWM m'00 1111 NOWto1c1n V. -1 MAO r-Im Onn ýO Ofrrnmvr11/1.10 ý-1
W - NON Co V rl V'00)rrmNWN V r-4 to V Mo01(1 lDNMO MOQ O
MMNrI 04A Oh rA V MNN r1 0%rvM44. -4 'le MN 44, e 0 en
o
WWW 4 rnnnnn
4 nnnnCo Co to Co Co Co Ch im aCh at moo o 000000000 00 00.0 441444441111111114114111+ WWW ý. IWWWW(. ýW WWWWWWWWWWWWWWW
n ý
in 0 +M W ýo ýn V v CO ul N o Co rym vvO1ý mm .om .om nv Co w nQ 1111 r) r ms NOr1n m NMMn 01v10 vr 11( 1o ý01f1 1 Y1N m O 0 ý N° Co
IIfl q. C! r r 1MrM m01 Mri rlw ný11f11f1 VO N P O1
H ''1 AA Ot r in eMNN rl AA rZ 111 VN x-1 0% %0 V 44 HHN0 U1
r+ ý.
°
NNNrnnrrnaommmmmmaaarnwoooo. 0 000 oOoOoON. 4 o . 1.4444.44.1.. 4444.44.1... + ww "lwwww
ID 4e
N
ý öNÖýýrn m ý, vin MMMaT
(ýýW, Fnmuwiö 1111 dNNW U)IAOW IA (7)NMOýrl IA dln0ýy NIA to nm(y v0 Cl x0011010 NWO NW NMm m 01M 1010 ri MM Ul r f. ) '0O
1 %0
O4 Oh 10%0 VMNN rf ri at r to r) 44 -4 A 0%%0 eN10, OM 44n0 /-1
o iý4ýmcomQqýýý°o°°. ý. 1. ý° 0000, .
1-10 wwwwwl c . wwrw 0° O
.i ý ý tý" wwwww w wwww OMmý, °p,
r, ý{dMeoý`Oqý výýýýýnNNO 111111 vuimCM Nmaoýo. ýNO. ý. ýp r4 (14 OM in nN 01 1 1 to n
ýo v rý ° x ch M Co Cl 0)10 NCO rl vV ýO W toö in 4n MNý1 rl m ýOYNri r-1r V rvr; Arm
rlm U1N NO
N n
el
W. o° o00000o000000000000000 0° o0000 . 1NC)I? in 0 nmtO. 4 MvinZO nwtO. 1NM,? IAWNmrno
Cl NNNNNNNNN Cl MV
NN
174
0 ., m
9 w N
ö n +
°V Co in
w In .+ e
w $ 'm r4 O. 1'1
ri
W 1ý
W
V
Ch
N t/1 »V{ .T
dE o .+
n ý°n
a
.a
o diö
dE NOON O. O" 11 IfI0O
44
.ý
° in in Y°°4in °tn 4°4Ytn 44444Y4444444 444W4WWW4I4 o ýwwwwwwwwwwwwwWwwwwwwwwgwwýwwwwwwýwýwwwww
1(ý W-r0mrnnn
4" r-i mvovrvýýý
cn rvWWWOe m ýneo
«4 2d ri
ýOOn lO ll!
.W
v' iN
Co bd . Wl
112 .. NO
Co m
in r1 Aäh v1 Pý ý1 01 N In VNOO WW O
ºi rl r-1 rl 01 W co co hhh %O %O %p in 1f1 in vi ./drrv PI M 1'1 /'1 NNNNN ýi -f ýt ýi ýt, 4 c% hO
v
In U11(1 In in vi kn in In in In In In Ifl in In in In In U1 in IA IA In in in in U1 in In Ifl in %o%O %o ID %o %o H %0O
°44°4`? SSY N°ýY°ý44ý44ý4444`ý ? 4ý4°YO 44ýY4444 o wýawwwwwwwwrywwwwwwwwwtý. cýwwwwwwwwýawýawwwww o Co rn mo
IDIn IDNONn. - y r-100
ýv N °W uri
ý1 PN'1mN1 rIm 0ý10N1 W
rbu1 CSU ulv(v 0olmnno
NrO NO ýO V NO01 tý 1f1v Nr-101 W fý 1! 1 v P'1NOf`ýý01[1 v 1'1Nrl O I(11ý O1 O AP111101
In in kn Y1 V -1-T `1 mm PIP t) MNNNNNNNNN r4 H 4M rl ºýýI riý01 W t-P%0 bn AO
1l1 in I(1 U1 If) in t(1 in If1 I[1 U1 U1 in U1 U1 in in U1 If1 If1 to U1 in to b %o b %o %o %o %o %o %o %o'o %0 %0 %D'o o
44444444440444444444444444444444444444° o wwwwww4 wwwww4 rywwwwwwwwwwýawwwrýwwwrWwwwwwwwwww r, WOm w'? v kn
ÖvÜ bw mrIN U1 MAIn m-in
Ö-owmw %0 -41n N1ý h
NCO b MONO 0 wco a
ri 01 aD l01n 1'1N OOýcDnP lo«, v r'N r4 r100 '0'Co P11: NCO . -1 IO NtO ' 'n -1 [00
NN 44 N 44 94 N ri A r4 mA vl vt A rl of Ar fA rI rI rt ob Co Co h h%0%0 to AVV rim PiN 44 vf0
In in in in in in In In in 'o i0 %o %0%0 %0eb %0%0 '0'0 %0 %0 b %0 %0 %0 %0 %0 %D %0 b %0 I0 %0 %0N NNN 0 4°444044444444444444444444444444444444444
O wwwwwwwrýwwwwwwWWwwWWWwwýwwwwwwwwwWawwwrýwww V 1(1 Y1 ý0 N1ýO 01101n0 NCO r1 Or Oý 10n cl In0 --
pý ýO CON 0010 r1 m r4 0
NN IA v ANº. 4,4ON o0, a0 P1 O, V0bN OI10 YI Ob1'1 .4
0U)11OW ý°O vN 0ri
,O lý100
rim- I. 4rl rl 'i ri '-I Cl Of Co Co NN%DWO%D 9) 191 VVY ýl ei mmmmNN 44 N, i 'i rl . 4.40% N In V0
0000000000000ýy0000000000000000000000000000 mW r-1 NV tow ýaD 01
rH-I. 1.1 .. -t 4.. 4r. -IýIýýNNNNNNN NN NI'1 /'/fN'1 (ý1 fý'11ý'1 P1 fn'1 PEA V
°%0 °4ý4°° www w4444°4Y°404YYý4ý4ý1111 o wAwAwwwwIwwwwwwwwwýwwwwwýý wN d cO O ý0 ýD OI(l nO"0 fNOO1ý N Or'1 N dri N Co Co NýONnd0OW ONaDd IA fC% f0 V NO Iflei A %O 0 00d Ou) A0N
NON Co drfNd a0 ýO %0 Co in O. U1 W U. IAOr1 U1 OI , . ON O PI PI N . 4rl. 401 NN in e 49 N 44 . 4.4 eim 10 V PI N rl. i. 401 %0 wv ei AA 411
f- r- ý rý rý r, r- Co Co Co Co 0 Co
o wwwwýwwwwwwwwwwwwýfl wwwNww w Ul O1N Oý fýNN V'm o ýO rIN V ..?
'
ý/ V 001/lOON U1 1ý10ýO NN o I(1 %0 %0V o
Co Co «4 e ONr00 Owp
ri rl rl OIPM mNNH Ar4 h in V PI N 44 ei rl PAV mNNýi r4 Af in
rý r- rý f- r- r- r- r, r- to Co Co Co Co Co Co Co Ch 0% Ch ý Ch ý0000m0 %D -
O r1 wwwwwwwwwwwwwwwwwwwwwwN rn O c0 ul
NO Pl OV u)
e ry N00m O( Oý
O O% h 0%
O fýl m 44 y Vý ?N -1 NN0O0OV 10 O 000 Np 00 1"1 O 0% ri
rI N %O N ß-1'O N 01 %0 01 %e WO O1 lfl rl ýO v YI BO O7 .1NvvO .
99 ON O1 Co %D Om 44 N r4 H Co %0 V PI NH r1 111ý In pt N 44 mr-101%O V NCOO re
NNNNNN Co Co Co Co Co wm rn m rn rn rn o0000Hob
owwwwwwwwwwwwwwWWwwwwAwww (ý NN OWN v OHO N1ý vyý ýO V I1N.
oo A NN Ill O111 ý6
.io. + n %0
o oo Co 0W
rlN co %D vN . oo 0000 o0 O 0000 Co 0 01 OýNV hýONVtD Co N1ll OD 00 hpý(ý1
oV fD oap1p
Gp00000 lO N
in in N1 44 H r4 tDM V 94 rf x400111 P1 NH ri Co UI PI Nm Co V NNO H
000000000000000000000000000000 WS rI Nm vn Wf'. m 010 N1'1 Ow fh X001 o rI N P1 Vtow t00ý S
r4 ri ýW
IUIr1NNNNN NNNNNM Z ;P 44
;
175
44ý4VY44ýQ44? 4444ý44444ýýýýý ýw ci o ýawlý. wwýawwwwwkýwwwuýwwwrý. wwýIwwwýIwrýww
NOO O em ONry I(1 Ifl ýO NN Vaý m f'1 V w 1'/
a oý N
NW N Co º1 0% %0 Co W(ý WinW VrV Yl1ý f00% m00i1-I P1 Co S 42 1 ý ý V Nd (. j N N "ýý WV r101 /'IMN V W ý001(l NW 1'10 výON l'f 01 V m .ý
A /7 N r1 "'1 Co %O If1 r 1"I N . 'I rI r1 Wn in VAN rI r1 0b %O .1A 1'1 NP r1 N
N
in
o. W NN.
-1 %0%0%0 rrnnrnnnr Co Co Co to Co Co Co Co ci 0%0100000
4444444444ýý4ý47. -74 44444444 N
4 i, « ""« ýuýawýawýawwwý. IwwwwwwwH" N+Iý1a ww w ýo wn OO e
H NH w "e W W^ iýrPl
HnV ul OWV Oýry dY1 ý0h 0% nr4m 0% N U% %n z, 8
b hOýrri A/'1 WW Wci NO Xo A01f101 ºi r-1 U r Nd
. -1 +F1.
rl rl"-IOb%0 Mr MNAAA m%0 V PI NAAA 01%0 Y/ r) NA 01 Y1 AA0 H)
ý rnnrrnnr Co Co Co Co
4m Co 0%Oi 0% O 000 00
ý' 4ý444444444444444ýýýý77'. ý4 U1
Y , w o wcý. ý7wwwkýýawý. Iwtý" ýl wwwww{ý. (ýj wwýrý. ww
N Ifl d c0 H
Pl O 01 OO 2 -OIIII
m l ýO %
'n w m
WN n b
{ r 0% I ci ob in ý001 NmN ONr V
-1 PI r' 0 01 pýp fV m
Oin r
NoýYl Y/ '11'11! P12 P1 OýON II. Oý V OOirTr 01 lici N mV r-10 N r
N rl
º1 ýF
i{ 01 Ob %0 I PIN d4 e-1 n 1f1 ri Nl rl 01 %0 V PI N rl al Ifl PI N vl O Gi
M4 01OY!
W
0
rrrrr4n Co to Co mm o% Oi 0i O0Ho Ch 47'. 4 4444444444444
% 4 U w o wwwwwwwýlýaýwwýlý"lw wwN"Iww
oýnmoýn Nv oeýoo mo 1111111111 w v Wm ý uim
rl ei A Oh n 0% rl Co On CO rb u1 .i cm mom to Co 0 Oh fV O ö O nein V Yl"-I UlO rN dN mýD ýO UI ON r-1 O! O in N 02 p
%o in ei rt 4 AA> tnri N Ac%in INHo Hn PIN0 º1
wOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO H 44 ri /IAWN Co 010 of v1 WO x00 01 OHNei V UI%ON Co CO rl riýýrl rl rl rl rl rl N NN N NN NNN N PI
ryý
N Vl . T.
N 41
dÖ
9 :° n .; N Co
01
9 W V
% o
Ö
r rl N1
ä
a
y
N 0%
ö
o a PI
.in
N
.. N
N
N y
dE or ell as
ÖNr
m Ul N"
I ý'
Pl.
oÖ H
Ö
O
ii
I
o wwwwýwwwwwýwwwwýwwýwwwwwwwýwwwwwww ö -öööömev°nmmmnnýöäntoen. 4 nýöroýý ýmrr°vö r4 AAr4 ýHHM m ýýýmmmco mrowriririrriririeeeWeerirrrhvivwö
4444 wMwMwr1wMwwwwwwwwwV 4 X444 oýw wwwww wwwVýww wwwwýwwwwwwwwwwýý ww4fý7 co c4 0, fo O a1,
vM Öä
co Pb Y1v Vm Nui r%
4 oboin co
icoocor- thinVmraA ÖÖ
a%ODD n'O 60 r1IN
an N0
M Yl In in IA In VvVVV rr Or vVVm Pf PI PI PI VI PI PI PI PI PI PI Pl f% NNNNN M44 N0
C14 Ö
fV 1(1 1m 0
In 16 co Wä
rl 0 O
I[1 ý WN V
ifl i0 ä
aW0 1ý ýNONÖý
1ý OWi . -1 fr ei iA
NINNI
w ONffi
U1 b1 NmMO ýOrycp v r-11ý V OP PI 0, 'o N'O %101 V ýh Nco P1 co co n0 N co a0 a0 n1O YVN In lfl 1'f in NN rI rlO 000101 OýoD a01ý1ýfr ýD %O W! hlflddAAN Nrl I ri O
N 44 NNNNN 44 N 94 NNNNN 04 NN rl r4 v4 el rl rl I rl rl rl M r4 r4 A X1.4H N""1 "4 A0
ýýýg444ýVý4Y4ýýý44444ý44444444ý4444V4Y4Y0a wý ýwww C% Ul%0 wwwwwww wwwwwwý ýwww www NbvoOONNNaNbO a} nwar ,OV Co in Ot %0 V t03 vry O Co %0 NO CO 0%30 1(ý
% %O 01O rl O co O, ZO Ifl P1U1%0 UlcOO, Q
r-1 vd1'I A NNN N rI rl rl ei00ý00aD ýD If1 PI rl per YlN ri 01 NO VN0 tO lO Ani 00i r! -
. ý. i. 4AHH. i. iHm. f. i11r1mH. f14. fof0obob ma0aoaömnrrrreeeeinuiuiö
(a1 0ýN('fVhý1ýOD
0 rß-1
9A rn rri ßn-1 ºi
mNNNNNNNNW M. 0 $N'IAA (A ( NW C?. 0
176
44QYý4 44 44gZ 4444 44Y44Yw 2 o wýawrý. w44
cn wwcý" ýrwww www - b Co OO Yl 9B i1 ý ý N O
pý ýa o pr-1
W 00 2 M1 f1 1ý . -1 1ý ON CO N 1f1 in -
p to MmY1N m 10 N01b V2 ýO to U0M 0 to N010 U) M . 10110 llf 1
Yl N N O1
y T S
0 c% 01 co co co P 1- Pb %o b %o in 1! 1 in 1f1 .; r -º .pMMMMNNNO Oh
U) O CO
0 ý j
C" N
NNN 4 4ý ° 44 44? ? ? 4Y 4.4?? 44?? 42 2 .. .. »«
iý t ,1 t, 1W 4]ý ] 41 W Ov rv .1W. ýo fý
Ö wmw b : -1 . -1
w Kl
WWWW W(aý WWWWW .mm 01MNbN Vm ýOOO ONb b0 M
cO OO
. -1 ' 7 IC
UNOM m uly O . 4001toN 10
fý NNp h Oý v Nr-I OýNb V MNOOý1ý 101! 1 dNý001mP ýO V MNO 01 NN 01 Fý ý ý ý
'+'7 'f VVMMMM MM MN NN NN NNNrl rl rl rl rl rl rl0 i-1
44? "444, s44VQ. Y4? 4g4444tn 444 o www(ý] wwwwwwww(ý{aý wwwwww w(ý wwww 1ý hMOM fV O IA Co 01 OOV Pl
w xn O% Mp Co Ol ýO Of n . -1 N Co O bý1Ný U)to toOdN OM AO 01NN OMm NWWOý11101M dO N dM NtMOOIt toI IO b in VMMNN. 400 N . 1100 M 1000
{ý oI o . -1 1 Co .4N Ob
NNN44m14 HAA mm 11.4 rl r4 rf rf. f rI rl ob Co Co NbbO M
VV dd V 'UI 10In U1U1in In In U)in U)U1U1 V1in U)in Ul U1U110O 4444444444444444444444444444
M
W a wwwwwwýawwwwwwwwwwwwwwwwwwýaww Nr rfm V OýMmMm Ifl Clm OOO1fl 01 Oý 01N MO 0 -I ýN 4 ra o 94
M OM m . M .4NM fý Oý dNONN 1() bwo
ri MN r-1 0 PNm VºihMOf Y1 Nm VrIm Y1N 01 Y1NaD l[l0 MNN
O Ij O r1 'i ºl rl r-lºt ch c» co co co r- t- b%o %o bn mina VV MMMNNO f-
qqqßýýý
M OOOOOOOOOOOOOOOOOOOOOOOOOOOO
W NYbmO d%OCo ON Vw Co ON V %OCo ONV %D Co O Nd ýýN. "ýýNN NNNM }I
Q MMMM VVVV ß/1[l lnh ýv
N Vj
b to bb ID bb M
H .i O WWWWkýý. lW +N1
IC e %0 CO r, 1 (Y% n
F W r Niirnaý1U)nö c;
1D in M/'/ Co
bbbbbbbö ww 4
o wwwý
i wwN
. ýi f4 ri -e c - m W
O
r4 il e r OO Co fN 01kn OW N mO U)
WW N bN0/ V v1NU1 A rl
MM NNN r1 r1 . -1
ä$ ý' °444??? 2 O wýyw4ýw W ww wNN
r 1 N
1 Co m
. "4
VN
r1
ý W! le U)
O 0N1
N M
g
01 N to in M
o o
wwwwwwý wNv 0
0 to n
CO CO
öý rUi n'm
UI IO N Co in f4 OV 9 NNN. -4.4.4.4 A
0000000 )b N Co 0l ryý W MV'?
...... -
1ý -
N (A
dE om
ui o" . -4 N ui o r, in N rl 1'1
.ý 9
14 ölö
vvnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn'01010 44444444444444044444444
w 44444444444444444444 o ww wýwww pwwwwwmpwwwwwwwwwwwwwxwwwwýwwww OV 0% co Ohry ý'I cn nNn
0, vw co n wm co n rI ms n vw voD 0: 3 v+ m rI NPI NOmN O r1 0 ONOWNr
O, I in Nm r4 NN Nt'1ý-1wry NNOe-1 PI NOn oo oý
OOOýry ý d r100 bOnN n NOS /ý1 Oý ýD Pf OPýO PI . -I Of PnNO aD ýO ý} 01 0m wn VN -W t- n
AAAm 01 w co hP %o %a %o in in in nVVVVmmmmmNNNNNN rl rl ºi rl rl rl r-1 rl 01 mN
r
.i N
444444444?? Y444°44444444 44Sj j'1444 V44Y?? ýwwwwww wwwwwwwwww wwýwwý NwwýwwN o wOpwwwwwww co MM O Of Md dý001 O, p1 N ý0 Oa0 Oý ddd dN0ý CON Ob
Mhý-1 co bfr tý MMO 01 H 1(1 co Nw O r1 %0 co
:3 to rI Vt Ow %0 %0
1 t0 1[l H r-I MddP 1ý O OH tp rl rý p N Yl dO 00 1(1 N ý'1 01 (ý ýO dM ý-1 Oý W ýO 1f1 MM .1O 01 t0 fý ýO hdMN rl OON ýO hOMPN Ir M CO
h1f1h V dddMMMMMMNNNNNNNNý-I rl rl rl rl r-I rI rl rl rl r101 CO /qP Pip yýhddM
N In In UI M In hM in Vi to in in IA U1 in in U IA U1 in %o b %o %o %o %o bb .O tO .b .0 .0 .0 .0 10 10 10 10 10 10 4444444444444444444444444444444444444444444
o E, ýwwwwwwwNawwýawwww m 0ý c" ww N. w cýIw wwww 1(1 O 10 tO M MOCh M NN NI(10% V Ch 01 VN Co m ri MO "( tn A Oo m (001 O ab
dM cý " cnn
fo p, p' 1 i1 NV00 1(ý 00 N 01 fý 01 r0 fA Mý (' N UI M fD 0M If1 Co N 1f1 rý O. eM If1 N ri 0 01 Co gý
. t.
.! M. cý. O 01 NN CO 1f1 NOPV r-1 pý fý Y1 O %O rI I[1 tO M O.
. 191
N NNNNNri ti AH riA AAAAAAAA/ ch 0% Co lZ n%0%0 Y1 in -e v"l MM MANN Nrl riA
.,
in U) U) in IA Ill U1 U) N U) WO WO W to b t0 to to %o 10 %0 %0 b tD %D %0 '0 %0 bb %0 ID %0 b %0 NNNNNNNN
4444444444444444444444444444444444444444444 O 'wwwýawwwwtý" rý" wwwwwýawwýawrýwwwýwwcýwwwcýwwwwwcýý"+ww 1`1N 0ý m1'1 ýOv mOýN[Ov va0 N If10ý0 011 OWN ýO OýOýdO0 Nd h OI OP OWN1(ý cO Nr-I fý ýO Y Oý W hý-I Y11(1 d0ý v00. -I ri0 1'1 VN EO N1ý M t01f1pý OOH ON
aO" Ul U1PIN Nr-100 X001 P'11ýrl ýO r-Ih P1 Oýhý-I CO hr100 Y1N OW ýO V Plr-10ý oD 010/'11fl ýý0
ý-Ir-I rl ri r-I rl ri ºI rl ý-/OýOD CO h1ý10 b1f11f1 VVV 1'1A1'1NNNNrl rl rl i-I ri ß-101h 10 ýOhV VPI
W 0.0-INN
1(10IýC0D 0ýý-I
ri ri rl. 4. . 4.1.. 1.4 NNNNNNNNNN f'l /Ný1M Pei ei ei r9 te'i'l VVV
177
m
n n ,o
d
N ýo
U 0 14
O
000
N r 1o
E NOO fV w
r-1
N O
Ö Ij
Ö
01
E
ooh NN
O rl
e-1
6ý ýHI
%0 QQQ4444444YQV444444Y4444 o wýýwwýwwwwwýwýwwwwwwwýw wa a vv oo cv a
NcWwoono oll v0 O NO CO UI V PlU 0M Co 14 V N1; 01 0V X00 QPl O V OW N 1! 1 01 V 0% in 0 in ºI rO1 1f1 PI r1 0e P in ei O! fý 10 0NO I'1 OA ýI
Co 90 r- %0 '06n in in 'f V l/ ri ri mNNNN nI vl 14.4.40'. f00
II 44p44444444440 4%04Y4ýý4Y4Y
ýýwwwwwww wwwwýýýwwýww wnN0 o www rn v omrr co Nmm om vo 11v0 Om P'1 mn ýOVm
vNo mN'O 0 N Gov n.. non .o mu, nrro %D A ýº Nvq, r 11.!, r IV1 99m ý0 1[l vA ry Ov CV ý X0
. 9i9 O
"VM Pf MNNNNNNHM r4 4 r1 ý lH M 011 WMVN0N
44ýý44V4444ý4ý444rý4444 o ýýýýýýwwwwýýw wýwwýwwýww ö dameýööýnrnrvrvvr-nývmýörl onöýý
Co m
-v ri OD 1[1 1'1 N rl O 1! 1 d 1ý Ifl dO ýO O 1- ý1 ,O cV CO %e rv 1-. rl P19 Co Am rf rI rf rs e4 c» OD PP %o %o in in v4 PI PI N fV N-m %o Om
.ý w ýI,
CO Co oCo to Co Co Co to to 0 44444444ýQV44444444444444° o wwwwwww (. ýWwww(ý wwwww w(. ](. ý www wN rn Vfy wO V 00ý NtýaDOmily Ntýltl rfýfr Co OHO 11 rt n h Or-1 VO N%0 Oh%O Ul 0 NO U) O ri WO IA O. Y1 ID rrNV0 1D r% U)
vmv. +r: r! 01 uºsrrl. +omvmnorvvo. ro .i V 1'I P/ Pi 44 44 r4M AHAA
AAA LO m%0 U1 Vmm P1 Nri 0 %0
0000000000000000000000000000 WS -vvnenmmouirvn vv -w vvvo to n to in in eF, co moýN mvin ýn co mnv jJ ýio
n in in nýeWýýýv ý%o e N fA rTi
tn in tn to kn in in in tn in in to in in in w) InU, inu, uiinin in viviin kn to into in to viin Into in in in in in e 4444444444444444444444444444444444444444444 ow wwwwww www wwwwwww w www ww ww ww öö rn in öcomwö nv nm cc vrpow A, in co AZo öýeAZo n vi ko.
"%m
v nýwrneýýmevýýrnnnýýmýýýmnvýýmnrvnmnemvýn. om 01 mmnP. nnw %o www Ul in in to in VVOvmmmAmN fV WNNNN rl rl A º1 .1 vl A-4,4 Ot
.ý
tn in O tn O in NN in in in in in in N v) tn N in NNN tn tn in in tn in NNN in mbbbbbbobb
4444444Y44Y44444444°444Y44 w ý4ý4444444444444 o ww w wwwwwwý wwww 4w?
ýwwwwwýwwwýNr)U1 Yl O (ý1 Vw W. W. 9w a0 ýOh 1'1 f- P1 O
Am vv 1'1 N Co O 1" c0 N 1ý1
O ýO1 1/1 ý0 . -1 [O ýO V Ul-W N Co I, r, Co 1, W ID 0% r1
1Or %0 m Ln%D Pfl Oý r1 CO W Ov02 Co P1 N 0A V Co rl N ýD If1 rIN aD X01[1 V P1Nr1001 a01ý ýO In V N00ýGD fý ýD I(1 V 1'IN Nr-IOO A1"-I lr /'1P NODN 01
VV 0( Pf mm ri mrr PI 44 NNNNNNNN 'I 444,4 vl (4
,4 "t, 4 1 01 ftl Co PP %0 N UI HI -I
N In in in M in in in in in in in in in to in ui in in m in %0 %0 tO in in in in in tO in iO W in in in iO in in in in in in 4444444444444444444444444444444444444444444
o wwwýawrawwwCawnýr wwwwwwwýýýawwrýwwwwwwrý. w www wwWa Co
dd 1'1 I'1 901 N Co Nm ri P1 U. in 00 cc,
Ul N O. O1 Co m d00 inp 14 00 O mNO1
OON0
N Mr) 1'
in 101 P1 ýo 94 N %ý001l11f1 pt P11(1d r-10 d1111
N9 dri Co N 0ý0101m IA
.? Nl(1 e1 dd r-1 O C! Co lý P LO ýO ll! U) dd /'1 AN r1 OOmN lfl rl b rl l0 Nm Ifl '! ýO 01 01 10 dN 01 PdNO
NN N NNm 14 H 1'4 rfA 14 r m't rf rf rl rl ºI vl Of Oh Co Co r- 1-%0 %0 Ul in in "}. i/ A/'11'11'IN NN N 84
-,
d
4444444°44 441? C> C>44444°44°4°44ý4444 w ýý3Z
o wwýý wwwNwwwww. 1wwwýwwwwwww wý wý www Co m wN
rvm M Co Mno mnM vmrvm m oe m rv mo vmm Co m in %0 . 1vN Am mM
mmONn eN V Ul 0H d m-rl OMmýD dý0
N In Q er O ýT
InN YºY1M Nx100 "OW i-. .VO
bNe Nmd 010 ýf rIm 1[1Mr101n vl VN OIfl Ifl ný
r-1 rl rl 1.4 rl x10) 0) 01 Co Co nnn %D %O In Ifld dVM MMN 44 N Nß'1
M Mi
rl r1 rl rl. 4
x o% mit %o in in
0000000000000fy00 O1cýO OO 000000 OOOOOOOOOOO0000000 [aý F
rl N l'1 V In ý0 NCO O'. 04
ýei r4-4I r1 .wINi rO1mNN
NN NNNNN MO ANf M'I vow Amrr V
g4%o q44rý WWWwwwwwww4wwwwwwwww ww wwww4 Ww o wýýwwýýwwwwwýwwwwýwwýwwwwwwýwwwwwww wAn V el
ý(º P10 co VN0 00 ýO co Ov OýýDOO
mW 0pý0pý
in Vp0 m %o ON1.4hdr1ANr11Nrco
0 14 m 14 %O i1NNNmN Omi 1000n.
4IONNÖahWNOO V NNýh ý. 4 .4
PI NN rI ri r10%h%J U1VMNNr-I r10% h%a to mNN ri r4 0n CO Y1V MMM 04 rl rl r1 A
I OOONNNNNNNNNnmCo mmmoooa 0% 0
o IN 4g4
w 444444ýwwwMww4wWW
w in wwwwwww WWAAAAZAii wE ou w? ýwwwýwwwwý'wwwA le 0 CO %0 0
Co .4 CO
0M CO Öý b YN1
(4 nn ýI in Ob11(1
Nr VV ODD Ir ýf a0
ÖÖ D
ý1 CýOrý
ý1 rl rl 0% n in V P) PI N rl º1 rl OI n In /'/ NN º1 º/ Co %0 V N1 N rl %0 PI M Pl PI YI M P1 PI %0
NNNNNNnnnnmmmmmmmrnmrna 0000 ýrýýý+ 40 44444MUM1zUmsý rwiýýv -e %nM
ri fV U ("1 NOS nM UI 00N P1 0 01030% V0 %o ýO o%n ýO mNnl'ýo n Oý V r-I Pl ýO rlnr N rf 00 N190
Of min V 1'1NNýI ri rl n 111 V P1N rl rlm bV Nv1 .i PI Nr-10 N
.ý v ýIý
nnnnnnwmcOtOWCOWaaamogoooýo to 4444044444444444444... 7° 4
o wwwwýwwwwýýwwýwwwwwwwwww w NNOIrý uot NW v0OUiomýoýouiNO 111111111111d .i oin. i inuiooaouir mmnnaoov
o monvin'onv'oo "+ oNm n NO1 V U! ri PI V o'i n O. in r1 "! It U) CDO
ý01f1 PI N rl r1 co wVMN rl r-1 4O V NA r4 ß-1%a P1 A Ao .4
000000000000000000000000000000000000 wSNnvbn rn o ry nvNbnoNvNnm rn oNnvN ý ri rl r-1 r/H. -1 . -1 ri . -1. -1 tV NNNN NNMNNAAAA A/`l `ý N V1
178
d
N n ýo
9
III
d n F- '. 0
a 00 0" Oý In in r-1 NNN
..
0 0 öy
ü
lC w
w 0
V IV -W -W -r -0 ýýo ýH-V IV
ww wýp + OV CCU
N 1[l ul f')
bnV 0OD V
in b Cl NNb. -4
Cl ÜI Omi to Cl0
w0 .4 1'1 r 0 1.1.. 4 9Oýl0 N ID in v++ n PI N Nvl. 4O O! C! aDNN Y1 ON
API AANN NNNNNN NN NNN NN vl v$A M '1 N0 rl
N
W
44Z4m4 ? Mzzz° YgQV4444(HRO ww o wýwwwýwwwwwwwwUZ w" N1 OOO Vv NO
%OP1r. 4wc0u) V
wtn ir/p4No O 1D Vw W r1f PdN rn
O In VVI Cl N Cl Cl r1 1119 4C- VN (10 O% In
ý1 1'f N in w 1O W! Ul IPI 991 ., Ml O09 v
r4 rlrl_4_Mr4 rlr4 . rl r4 r1 rl. 1 M4 rl r4rl 0) 010% CO Co NO N
nnnnnnnnnonnnnnnnnnnnnnnno n 4444444444444444444444444° 4
O wwwIýw., wwwwwwwwwwwwýt. wwwww pl VV P/ rn nnOO V Oi'O 1I d n ul IA N4 n rlw y hr10N N to WO lb 1 In bAN ID 01 W f N N Plr-101 '01'I O COb 1'I o01WN r10ýtD ýO nri Pf N OO W fýt0It N9C! CD
aöomrrrlýýýýýýV; M; V; V1V; vwV c- *IV -e. i ,rd V
.ý
v
In LA IA In IA IlL IA IlL IA UI IA IA IA IA IA IA LA IA IA Vi IA IA IA IA IlL 0 4444444444444444444444444° 4
nw ýwýWwwwNwwwwwwýwwwwwwNw wnn . -1 Q wbn
V r-I0
mNIrbL In
d-P1c' . 4O
ä00 1wfýf-no' ,
n1 MN
P7I Pf Pf 1'1 Pf n PIN NN NNNNNNNNN . 4.4.4 r4 . 40 co
00000o0000000000000000000o00 Mvnw r- co rnoInrvinvininnmmcl ýfnnv%tonmrno FF
nain in wwain ow %a %a e10 eeeeen N f/1
4444.4444°. 4 o Own O% ýo Nw CO
wd r-wwwWWWwwwwwwWwwwwww1wMwwwwwWwwwwwwwWw
NN ý ICI OD N Co M
dd O0 NOd AON NN. vr/
CO -e
0Ndd X00 d ý3, Or-1 r-1 m %O CO Co O. -1 A ýD UD OWb0 00 Co %ý0 %O ýO P f- a0d 1f1 %D ýO %0 %O Ob i-1r-1O 0001
0 O% Co in d ft of O 01 c0 a0 1ý ýD Ifl d 1'1 NN rl 01 00 PbdA r-1 O 01 mrb Ifl d 1'1 N ý-1 O p1 GD fD Pbd 1'1 PhPPPPbbbbbbbýÖbbv
ui A in in in Iii bn in -p e
de es výir vmAmA
ri mm
I o wwwwwýwwwwýwýýýýwwýýwwýwwwýwwwý M Mut
mr c) P1
w) a ZO aON 21 P1NO^Do
co m co e m(ý VOPm0
cn C, IOA Om1o!
n I:
9
ýO Irfl Ifl VVOM PIN NHm
.40 O1 Cmnb Ifl hYVMM
fm4 94 ý-1 ÖÖN
0ý1 001m EOD rb
P1 P1 P'1 en M PI PI PI fl PI M PI PI PI PI PI m PI NNNNNNNNNNNNNNNNNN r$ .. 4 .4
rl ri . -i
YY44444gY 4 44444v44ý"ýýývýfýýýýý4ý444g44Y o {"ý (. Jwýý-J1 wwwww (lýpwwwwww(: ý-1 ww yý1ýwwipww wwy Nw www(ýý-1 ww ýn 1(0%
to 1N Nn
in NÖ NP/N O0i n"ý
l'101 W
OION 0ý1ý r1 Nb
dN Co rl Öým
0% N 0! 0! co co co týfý týlr ýo bý0 ý01f11n
0 cn V'o"l V P' mN NNNrI rI rI r1011ý
." It
".. ". """"".. ". ". "" It
.. "".. """". . 991
rl rl"-1 r4 rf H rl rl rl r4 rlº1 H r4 rl MH rl rl HI -1rl-"i r4 rl ri . -1'-1 rl ri H rf H 0101 OI 01 co cow
.4
4 °°44444m 444ý4444e 44°4QYY444444WW$4WW14 gn in in
n wwýwwýwwwýýwwýwwwwýw iDNIMwww MWw wwu Oý N 1'1 dNN ýO 1 1ý 1(1 Oý dWdW ýO f ý1 Wd Oý d P1 OV
fy 1ýOO W t11(1 SON WOV NN W dýý00P Oý WVW Abpý dýýONN WU r0 iNvlot Of rl OOO 01 r: 1l1 d r! 1: lD Ul PI 1: u! d rl l1 P 1f. dNO 01 W ýO h 1'1 O 01 ý. in dN .1.
.i A- 4cnrno%o±oiotohmmmCoCoCoaönnnnr- roeee%cvinwlW)u)V)4nin, *, *.::: +r+
S 0ýimm V owc, w4hoýl=N V
nyý= cOO OÖ imm-wow h wo
ÖANM vw f-moo-i-IN
H14 ri ý
. 1,4.1.1.1 rIN NN NN NN NN Nr iiP/ AAA 1'f hl'1 AOýI "
179
d 0
0
äi
0 .a
U 0 14
P
Q
0
I O Nf
O" "T 00M Nri V
40 414
0 00 rO o "n AN O7 O
.1
I HI
ii
444tn 444%o 444g4%o Y44Y4'o 44444q444V4 wwwýwýýýý wýwýw wwýw1fý ed o 0(Aul wwwcl el m
V 1'ý1NNýÖÖ11r 41 1" V
0! VýNvN WVQ
Ccc bm
rläbeNý0lNN'O NO+
r1 e- AAA Aat co to PP 1010 w H1 U1 Mrrvmmmm r4 NNN NrI P4 rl M º4
1' Y44YQV4444YY%0 %D 444444444444444Z
Co "
o W{ýw (aý (. ýw www{týVJ wwww w{lDýJ ww wwww w. ý
ww win %0 O.
Nm, ,H,: l,,
N/'f ü10Nlo ýN Ö
r0-Iýh mw Co -e r1frprl Oou1
woN0 "I O NN 1'IP dr/l'fý OýýO rN Ow 1ý dN %0 O e9 %O h. 11'IN
ýD%0 Ln A bn V/I PI PI P1 Pf cl NNNNNm4 rl vl rl rl rl fi x101 Co Co Co N
4ý4444Y44%0 444%0 4444444ýý4ýýý4qý44 " o wwýýwwýwwýýwwýwwwýýwwwwýwýýw vv
ö ýmamvthXAý v$nn öärnýsWonmmýöW l0 WUI nr d wvA ..
C! ............... .9 ....... 94 NN 44 Hrl rl rf r-I ri H. -I rI r1 Oh 010D 0 nw %0 In in 'r rr AAI'I AN 44 M 'r
11
d I
ý4444ý4444ý444444g444444q4 ýw o wwwwwýwýwwwwwwýwwwwwýwwwwww co
to 01 to mm tp %0 oM nýw + oonpp oorn e Of.
4 In
0 %0 1l1N co
C, Ö
a. o VII co 0 r-
0W Yý1 ýI
Pmt rn-IÖÖr(('I hÖA
In 10.4N Nn
In 0
%0%0 In In -P VMM P1 NNNNN rl 14 44 . -1 rl rl 1401 CO P P%D'D Ul U1+V M %D
00000o000000000o00000000000000000 I
Pl Yl co 0% 0,4 N P1 V Yl Wr co On ON P1 VN CO % 1.0 O 0,4 mm V I(1 r-I W or v. rv. evvawininýinanininto %ýovv%oww%o %oýovrrrrrr ay NN
in H1 U1 Uf in IA in U1 to U1 IA in U1 IA U1 IA IA IA in in IA in in IA U1 U1 U1 IA U1 U1 in IA IA IA in IA 0 IA IA IA U1 IA IA 4444444444444444w4444444444444444444444444Y
o wwwwNNwrý. ýawwwwwwýawwwwwwwwwýawwwwwroaýaw ýawwr.. pýaw m in v co voýy 1oaon ýomýuiconýový1 v1nppý1 ao uiv v ui O ýO IRIN NN 01 ca x100 Nl ri "'S 011'fb a0 v01 --CO Y1. -Im1(ýn ýO r10ýop IA Npý 0 coO NrI "- N V
'o Oart cc in V No aD %o Y NO OD h inm 001 O010 1f1 V Nri 0010D t01f1 V /'1 N00ý co a0 ýO 1f1
Co Co hP%a W%a%0%a kn In in in to VVVVV .W P1 PIM MMMPI ANNNNNNNNA
rl r-1, -1-ri rl
I' U1 UI U1 U1 U UI UI in U1 U1 in UI UI N1 to UI %n WI U1 U1 N UI M In U1 %fl UI UI U1 UI in U1 UI vi in In %o %o b %0 %D '0 b
4444444444444444444444444444444444444444444 o wwwwwwwwwwwwuýrýwwwwwwýrwwwwwwýýwwýawýawww wwrý" www oooomari ddn"'n H ýondo dn ?, nn Od 1'1 N1NN rl %0
1ý1 in r- 01 NOn . d-1
dhO fz 1"1 s CO
ui Od
co
NN .
1!,
Co Co O
ao d
,p ONO 03 a0 "-1 N x1 r. WI v P1 NHHO 0% r- in dP'1N r-I x1001 01cO 111 n. e dd P/Nm 0 1001d 0 001(1 r1
dd MM M PI P1 P1 PIN NNNN Mm NNNNr1 p414 H rl A 11 rl 1-4 rf 14 A rl MN ro 01 0% 0% CD CO PP
in In 1n in In U1 in Ifl in 1(1 In U1 U1 Ul If1 UI U1 In U) UI M %o %o %o b ID bb %o b %o bb %0 bb %0 %0 %0 %0 %0 IO %0 4°°444444°44444ý44444Y44°4ý444g44444ýq44444
o . A. wwwwwýarýwwwwwwr-e-e wN wwwc%-e 0% o Co wwýawwwwýaý Oý 0 MMV' X [O rLO
m , r4
V %A
1fý M0,0, C r,
UI Co GO CO f0pý M[O o aTnM öN [O CO O rI OD D-1 1ll O1 MN ýO r-/ ý0 ý-I O r110 ý-I 01 01 ED M Co 1f1
I. roVON O1
Na 01 in A
'D . -ý MN oo mwn ýnýn n ýo vvMMNN . -ý rl oonN fA Voh ýT r1 rMo ýO M 01Ifl N0 m1N n Yl No om
NN NN rl HA rl ri .1 .$ -I ýi rl . 4.4.4.4.. 4.40% chCo Co Co NNn
%0 %0 %01111A VYVMM MM MN
.,
,ý v
(f1 to to U1 U1 in U tf1 *0 %o %o %o *0 *0 *0 10 *0 %o *O b *0 *0 %0 %C b %0 *0 *O b%b *0 %0 %0 %0 %0 %0 *0 %0 %0 PPP 444444444444444444
MN
4444Y4444444444444444444 ow v wwýawwwcýwwwýaýw w ýawwwwýawwwaýwwýacý. wwwýawwwwwww OM Om 0% OO%N 1l1 %t) vFf r) CO 0% CO NN w
%D CO %D in On ao CO LO h P1 ZD OWN 01c)a r4rl0%1'1 v Pl N in 0%1Z%0 nw 0h IZ in H in OiO ýT O ýO dOl On O ýTry f'/ Ill Y1 P1 NN'! Pell P101V Ole c0 V Vre 00 Y11'1 0 t0 ýO VNO Oý fý ýO V C1 ri 000 ý-I Hl r4HH 14HHAm 010%mmel P t. %0%0Nn %n in VVVPI P) m P1 N 44 NNNHAMH4MAA 0% COP
0000000000000000yý000000000000000000000000000 pl 0
rIN P1 ýf 111 ý01ýCD 0114- 44 rlýý, -I rý-1N
W O. NNNNNNNNNN 0.4 NI IMf /V1 /O11O
NCO O. 0 VN
%D %0 %0 %0 C% t% C% y r% r. co co co co cc at Ch 0% wl
o ýwýwwwwýwwwwwwwýwýýýwwmwwwýwww n to o morv v0. Ur v rvoo nad m rool w .iwhw rl Vv 01 h 1(1 rl(ý NIfl O 01N ýO rl O1! 101 hI(1 ý1 PIN NNr-1OO NO Oý OÖ NCN r40 v 0hN ri 1f10: 1fl r-11'ß in co YNU 0ýD V ý-I ýO rI NO 19 U 00 O r-1 P1 NAAAx of Pin V /'1 PI V4 .4 rI r4 0% rý MV YIN Nei -I ri QOM V M04 It rl%00 Pf
I
I
mmenr, nnnNNNrnmCo mm mm as mrnoooo tn
ýe 44444444444444444 w44ýý$ýý44ý44ýý4 w o wwww w www www wwwwwww w tg bi www wrm
ýoo m mrvrv mmmý ýmrnNaýmN mý etnmA0S a a 1-1n Oa rl '. O O0 rI N '. OO'. O'. NN CO In 10 NO a (NN. -1 '. O 0 UI Ul U) P1 N00^0h
ý1 rl r11ýý? Ulm > PI O Pf v) %0 a%00a M01 m rO bm YlOOOIfl O 1ý O
rl rl r101 P1l1V ANNri rl r-ImbVNIN rI r1 a %01fl VP1 rte' rl rl rlh ýO r10 in
4444ýQY440 444Y44444ý? 0 000ý o wwwwww w wwww wwww www www n '0OrnmoNW mWmuýnmýCo 0ONmCo
rivkn o11111111w0 NW mÖn Y11:
C! W!
.DO ma
ND ý(mý1 r4 -1.
O-I MYINÖNÖN 01fj0 OO
O- m%o V P1 NN HeI co %o in PI N Nº101%o V mAAmb%n mr- O . "ý
rrrrrrCo Co Co Co Co Co Ch (N m aOOOOOO0.40 444qý4ý44ý444ýý %D
'l g2 ob o ww w wwwwwwwwwww wwwwwwww wm '0 r, rvrvmvrnuiLnmvm ýommvooom0 111,1 11111r .io. rvrv r Co In r o%0 w m. + m C) m%0 mmnö NO
. -1 /'1 ri V %O 1.1 o U) mO o1 A 1(1 1! 1 If1 ý-1 rv 10 1.4 mrm . -1 b
%0 in m 94 r1 11 Co in PI 44 º1 ./ Co in P1 44 ri Co Om rl .1 . -1 rl m0 -4
000000000000000000000000000000000000 W AN in V in bP co B%oANA V: q to f, ýqpý 0 r1N P1 d LO %O NW0 º1Nh V 1(1 . 4.4 -I. 4 ri º4 rI rI rI NNNNNNNNNN PI P1 P1 Yf m P1 vv
NN
180
ýW rz
�-
w ý^ ýý
ýö .. ý
N
0[711
Y
H O
QO
d
r-1 v O r1
dE A00 ö OU4
.iN (V /V
51 0
11. N
tn in in tn in tn to Ln to to in to ee ri mmg444ýY44 it 0e o 0U) NNNNwwwww 0 UO N)NNNNNNU)ONNO O PI a1 . r1 O" 0U)NN00)N WVN N0ý1A N 01f1 N01ý1I11'/001 Y1 1O
14 V MPI M 1'1 N 94 N rl v4ºi rl rýPf 11
C
I U1 in Ul U1 U) Ifl U1U1%ObWO%o%ob P1
44444444444444 4 0 o w tlwNwwwww O wILd. Na ' ýO ý w P1 O
t 1 N Pl ELI ýD OV IP10N01ý Or-1 ýy P W " hN
ýl N ei rI N ri 01 P LO 1f1 N ý-1 O a0 a0 ýD Pf Oý fý 1'1
44 rlº1 r-1 rl rl rl vl Co %0 H)Plrl Pi
in 90 (0 b '0 (0 %0 *0 %D (0 %0 %0 WO N Pf
ý4( . ýw 44www4wwwww4{ w
0 n 1( ,ý N C Ui w
If1 N
1 WU)Ui N 0Nb v bOirf
P V O O*O rl N In OD N In 0ON r4 0
ii N
PI N
A Co N'0 In in VV ei 44 NaD 'D
w
%0 bbb %0 %0 b %0 %D %0 %O P- PP
.n ýwg((faaýýýwýý1tl]wwwwww44ý
(V PrO dN IDÜ
0 bdfNN /'1
d
N . 1Oi Ö
19 N
dd vPI 49 NNN r1.4 r1Ch in N e1
00000000000000 w rn oý CIS mv In ýn 00 rn Wýnm
ooco Go w0%onrnmat mma 0% (n ý" v
±ay} N to , T.
44Pý4444444Y-W 11 4444444444444444444 o w[rý-1(f. ýý w(1ýý ww(1ýp wwwww(l. 6ý w11wýr, ]1 wwwwwwýs]ý. ]ý] ww{ý w(. ý w wwww(fýý wi(ý-i ww
co 0% co co co C-1 O ODD 01 c, o q II- Ö
in tv V ÖýO.
-I vim ÖtO
A ono w 'co ui ri ý6 Npý-wm
onin ZryuuU, C3 jN
VmV ui. q uIb
0 P1 N H1 ri0010ýcO CO n1- 1O 1O W! lfl dVn 1'1 NN1! 9C, cc N 0, rl ýO v1ýfI Cý1O fV m 1! 1 14 mNNNNNN r4 rl r4 HH r-4 .4H ý4 r1 ., rl AA rl A rl A rl rl rl O, mw co co PP %a w in U1 in -º 0
U 1'
44Q44444in 44444444Y4444444444gY44q44444444 o wýwwwwýwwwww1qwwwwwwýwwwwwwwwww wwwwýww12 ww
`r Co q, 0 In 0' r4 -4 r 94 rl)
IA NºýI"Ai 000 Co ul N r-11-In ri OIL Yý1 mr10ý V NO1ý dN2O1b V
NÖa(4 ýI
O Pf n
rý101 OO ýO ÖN
rlrl r. 4 r4 eil r4 A 0% 0% 0h 0% Co Co Co Co IýPPhP %0 Vw %D in in In flH1-ºd VVV PI PI PI PINNNNft
1[1 in Y1 In tf1 U1 WI to in M In in U1 In tf1 N In n in in In in In In in M If1 UI In in LA UI in U1 WI U1 WI U1 In in to In UI
4444444444444444444444444444444444444444444 o Nrl (If awwwwwwwcýw wwwwwrý. wwwwwwwrý. wwwwwwwwýwww Yl OtO 01 NV0 mrw 1ý "D I) wtclvmm 0101.10 PI Pf Ul WI(9 O V. T O1 N (9 II) VON tD 1f1 UI19 N0 ýTOOW01N OW O1 O1'ONIn 01 O1 r- WN r1P"DYl P101 NV 1D )OIANL001 OO1 N. 1NO N 0011 10 V Pl I 01NWO'0(9Nr100 C! N'o1O cl
ý 0! CII Y1 V P)N. -IOTmO IOd a1'IN r/
%a to in in in in in w vv vvv vVmm 17 mm en mMMNNNN Cl NNNNN r4 rl rl r4 rl rlH H rl
., ai v
UI In in IA UI IA IA IA In UI IA In UI UI IA IA UI In Ui IA UI IA UI Ul IA UI UI Ui UI UI Ui 1010 %0 %0 ee%e%0 ea 4444444444444444444444444444444444444444444
n wwwwýýwwwýw wwwwwwwwwwwww1zýýzuwwwwwwwwwýw NMV
rv m rv VoaNNMOnbonM Co V O) ONO 0) Onm
Co : Ibn %D 0 nvN1N rN-1Ön1 ONl n0
Wh b10
Ul i fl /N'/N N, ß-1
rß-10 Ohl v001r oPl1
(ý. Oi
Ifl ºnl ri NNO01
API P1 mNNNN 44 NNNNNNN ,4H4HHmH rf of Hrf v4 r1 14 r4 A ch 0% 0mmt. n 10 %0 nn
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 57 -O P1 d Ul -O rm 0e 0 . -I NMW 9191
9O rm-0 . -I N %%O
r Co r- r-
fV P1 V rc r0 fl
Co 0% Q 14 r, m 03 0 ri le mw vvvv+ý*aýnkninawininnaýo%ovýo%o %v%o%o%o w ýorrnrrrriýrrm
N u1 mmm
g4444qý44°4°4°414.44Y4Q4ýgý444? ýý4444 V444Y4 o wow wwýwý'wwwwwwwooP
"w wwwýww ww ýwwwwww 01 tD co 'w NW O1 O'. Yl U1 VW ýO c0 eO O t0 M 0% in O in Nd to
O %N
N'P OOn co ri Pf Ifl N i-1 r1 1(1 NW O% P1 in b Pf OA UI O% PI WNV CD in IA V %O O'. O CO .4 in bbIN rl O 01 a. co P 1O W! 9 11 N º1 OO Ol Oý co fý ýO ýO I(1 dV PIA N r-1 r-1 OCý 01 C0 h 1- Ul Ul
U1 ;o
in in to OVVVV of VVY -W wVVAAAAAA PI mm en AAAANNNNNNN
I dd dd ýd did dd ýdýd ad
o wwwwwý Diww01wOO UNION
m wwwwwwýwýwwwwwww OO P'1 V' V ýO m Co Oý mOVmO ýD v ry ONm Od Oý N YI O ýO 1'1 ObNm ýO r-1 01 d 1ý NV rl C% bNm 91 m in f4 co 1f1 1'1 Oý in N IA mmý0 ý01l1 Y1 `10 d1'11'1NNNr-1 000ý0n 01mmtOPPP O0 b1Il U11(1 dv P1 P11'IN NN
N NNNN NN 94 94 44 44 44 44 NNNN 44 r4A 4H -4 Al iM ro M ýI ro 4rf r' M rf rf º1 rI BfA of iA
ý4Y4? ???? 44 4°4444444Y4444444Q444°Z4 ow ww wwwwwwwwwwwwww www wwwwwwwwwýwwAwgw
Z
m oo oonmomovmnMOrnry m rnvmoNmerul mr cl Co 1I1 u1 N m0 vN 00%%01(1 P10 Tr u1 N r-1'P1ýýNýONNn . -I Co Pl dm rl lfl Nv rOf mri ý0 Nvd P/NI NNN NNýýrl rl r! O OOOO Im. OYNv102 fý bd 00 V 1'f , -10ý1ý b dAºý
. i. i. i. i. i. fMr 4141-4 oh oh o% ch oh et CO CO CO CO oCo r rr-r %0 %0 e%0 e%0
14 61
in U1 N Ui Ui In IA N In in - U1 U1 . Ul U1 U1 in V1 Vi IA in U1 in UI in in in U1 U1 U1 in M1 mM in IA U1 IA in Vi U1 in
44444444444wwwwww444444wwwwww44444444444444 n wwwwbA2 Awwwwww rý. wwwwwuýwwwwwwwwýawwrýwFZwwww omor mo "'mrvnrv O In ui Comm nýnvmovom non N %0 X700 i. 4 ýo%omr. fm-r3, ýo. +uir+nre, Gov, nrnm. ýnýnýorvvrvrvvooýo. ýrmrvvýoý. ýoýo
. ýoýný. ioýruieriNOmrýoýnnN. ýmmr121 i-! ommrloer+. iommrlonsn ýoICo CCNNNNNN Nýeýeeý%0IIwin nýiuiui%nevvv-i-e vvmrimmr mm
0000000000000ry000000000000000000000000000000 14 rI N1"1 V IA SON CO 01041.4
M, 4HH.. 4,4I N NN N NN NNNNMM M1'11'11'I MM Mn wVV
181
U)
O
0
L
O
000
v
rý
I
I o . -r o0 N 00
r-1
Id
E 00 ,o o "v Am
9
w 0 0 0v
44°%0 44g4444to Y4444%D 444444444Y%0 4444 w o wwý wwýwwwýwwwwý Co wýwýwwwwwýwt" we '7O eO fV mP Nm ' O» O, Naý
4O Pr/m et" Oe
V ýn-I rOl 901.
W! . '1 Pr9r
V ce
b fm`1
br! c2b
rc Ö9r: IV
ýÖ01ýle V Co i cn -e
Yf N PI "
rl ýI of On Ot co r- P%o%o in kn m e-I O P/rb Pf P1mmN W fV Nrf A rf /-1 el rl r1 rl
ý.
.ý
QY44qý44ýg4q444444444Y44ý22wy 4W Co Owýw Inw www. 1U)wwwwýýwýwwwwýwwýýw
O I[1 PI ýO m0
w0 O
11100 V bfr
ýO cD ry ýb
r'. OND fib Ný 1'1 ý/ N Na N P1 ON r) ONVO IN 1V Pl
. "! 99%e In V P/ N ri O U) 9Na W'0 r) r) N
YlA in ,. r V -f Pi PI AmNNNNNNAAA rf A Arlrl Oh Co 00 P%0%0 w#n wl N
%o %a e%0 ee%0 eko nr, nnnnnr- ., ý 11 nnnnýnnýý 444ý4444444V4444 44444444444ý44ý to ow ýýýw'wwýwýwwwýwwwwmwwwwwwwwýýwý h MvOVM toN V cDUIMM m0 m0 %O O 1ý 0 N0. V in 1%01[1 M in fN IAN OrI dN bOfV 0O. 0ý NN W WUl in V co 'O YI -I r1 ri 0a NVi V MNow ri W Ch M%0 r4 M fýN Oh t- ri %0 Nr1 a0 aD ýO V MNOM Oý
N fV MM ri M vl rl r10%CWW PP%D%0 Y/ V-I MMMNNN ri rl M r4 r4 M r-I M
nnnnrrnrrrnnrnoCo Co Co mCo Co Co Co Co Co Co oCo CO Co mv
0 4(ly6ýwg{1ýpwwwwwý14Wýwý(b.
ýwwwwwwý(rý{wwwQ(lýý]wwwwwwwý w
(V If) rß-1 OD 1N0 fr'1 Nm r- to e, º1 Ö
00 P1 O1 r1 b
r-1 mOm
CND 1! 1 1[I 1! 1 1ý'1 ONw 1'/ 1N0
-p ei P111 NNNN. rlrli-4ºt rl Co r in VV PA PI NN rI .4 .4 rl. 1.4.1 In mN vi
000000000000000o00000000000 00000 Ws rvalmnmrnoinrvn Nmnmmotirvna"enmrnöArvrvam avaaaaannanýýýýýýeeeýeýýýmmnnnnnn
N f7
4444444444444444444444444444444444444444444 o ýawýwwýaýarý. cýýawýawwýawww('1 N ww; awrýwwwwwwwrý. ý wcýwcý min"' O of co 0% c0 Ul Oý co m in s 5% CD N O, U) co NN co W O at
C4 Ln aD O r1 O 1(1 dv If/ 7 1, %O in Ifl 1/ Ch
"" a0 %o Nvý to NO ýD 1ý1 d 1fl v ul vo ul ýO b
1'1N 10 Pf ri c0 ý. 1'1 x101 Ir Nl (ý t: W! P1 9C0 .V
PfN YOa. r: O ul r! Nrl C! C1 C0 Fý%O Yl V 1'f N co co PP P%D-D%0%0 toto in kn in OVVOVVAA in m PI Am r4 NNNNNN N'-11.1 4 rl , rl rl , -1
In to In in If1 N N1 to In in in in in in in In In in UI In U1 In In in In in In N In N in in in in in '0 %a %o b %o 10 10 b
4444444444444444444444444444444444444w4444 o wýaw41 ; Nwwwwwwwwwwwwwwwwwrý. wrýuýwwýaw ýawwwwwwwcýrý !V Ofý al .rwO. a 01v m ýr O 1(1 V ""Q
vi' 0 Oý VW co ýO Nm 0 C% O tNNa)N
Ww-w NAoO 451N 0000ýy
u1 OW
NUl On NI'f om rl Ulfs BOO r11f1 co V IA r4 to
O PIO
Ifl O1(1 O ýO rl N NNC01ý ýO V ANrlOO1 oU 1ý ýO In V P1 (R. r: w., OW! Ir! "! 11! 1991O
wV PI mm r) m r) MmNNNNNNNNNN rl m -I 4r#rlH H V4 M r-I H rIHx m0%CO co P P%0 ýo
in in in 1f11fl in in 1f1 in Y1 in %n N Min to vi in U1 If1 %o bb %D %o bo %o %o b %o %o %o b %o %o %o to %0 %o %o b'O 4444444444444440440444444444444444444444444
o wwwwwwtý"wwwwýawwwwwwwwwwwwwwýawwyýaýwwwwwýýaýwcý. wýa o öim Co Co
°ö'H vbm vr
n oý°
rn rv X00
vm výl v° n%0 om1 1mrv 4
1.4 -wriHQornmn%o%oiný. rýrvrvcý. roomv`ýýonmrvmýn. ýnrvmýorv. rmarvomýorv 44 NN NN -I . 4.4.. 4 tr$ r . 4.4.4.4.4.4.4 O, as as Co Co N N%0 %D 501! 1 Sfl VVV PI PI M PI N NN
In NO In in in NNbbw b'0'0 to b 10 to ID bw to to 10 10 bbbbb ID b ID b% to NNNNNN 4444444444444444440444444444444444444444444 o www wwwww[i! ý1 w
(ýww(1apý ww(n. ý www(r.
-]ý w(i-ý1 ww(:.
-ý1 ww{rýi (. ý(. ý{aý www (aý(. ý(ýý.
} w(. ý
111 d0 P1 d co
Im -m -w a9M Nrýtn in co o
Oaf ý[O
in f-P1 CDN 01 r/Nd to
riv -tý4 ý4
mry d0 w IO'O d Ml N. 1.4001 Of Yf 01d OId O In N U'. U1 N 0110 P1 r1 W In M00'. N In dNrl O. -1010 NI0
ß4r-,. -1 ri . -1.1.4.4 a% ct wwN -O%a w to in dddmmr in N Cl Cl Cl rl rl r4 . 4. l rl rl 0l CO %a w U'. d
W ýrl
N tý'1-w to %o FZ a001 ý-i r1 MH ri rlNNNN NNNNN0
0. - P1 fN'11'`fl
U)'0Nlml fß'1
6A V
o wwwýwwwýwwwwýwwwwýwwýwwwwwN Wwww wnr CO 0% cn O1 R Pl r4 V Ch f'1 in 0% CO CO 1ý ýO ýO ýD VMNN7V ri N Co ryN O0 1ý Co NVnV0 HO r, 0 U) N NIP PNN U) U) ) IA OO in IA 0000 O1 O PI
r1 . 4ON%0 ew! rZ SON in rV ýO CO rI N P1 O NNNOO NNUIOOO w .4
MMmAAA 01 P to V MNre AA 0lP in eM Nrt AA 4 r-V V mN"1 mrZ in to in 44
I'
ch Co Co 0 ]???
44 (ý]00(ý]0(ý]0(ý] ]4444ý]
444(ý]4( ]°? ( ]? ( ]??
? ?? (r
4( ]4(
t ý y . ý ý ý
MNý N*OMMOmr%
öN Oý A f'1
ON If1NOOOInÜ
ÜOIII ý' . in u)1 1 0. -4W! 4 Na V
-p d
0 V 11
Ai-1A m Pin e PI N NAA 0%r. m mNN rlrlri %D %n P1 t, Nr4.14 4 r4 A0 in
0444fl 4r- e 44444qý44ýý44ýýýýýý 2'ZZ o ýwwýýýwwwwwwwwwwwwwwww wev rn vm MVrnoCo o. vmNmm on ýomýo 1110 ýo 1 cl N 0N0%
N V'. DÖ p
4r, "INe4Ö NYNIÖ
ClN WN
Off d ÖN
NN NN «4A kn
0 c) ui U)
O
r1 0h 10 V PA NN rI r1 Co %0 VMN rI H> 1"I N_I 16 ýI%0%D P1 1-1%0%D%0%0100 01
rý r- CO Co Co 00 owwwwwwwwwwwwwwwwwwwwWN
wCO rl výn mOr Co vv - ýnao(oo 11111i111hy Co
Ill mm
fr'I 1N A YI ? U)WIn wO dW Of)O W
a0 N/'1 NN. ý-4
ß-4O hO
%01f1 P1 N rI rI P in P1 fV ri 01%o e NNAa; Y1 NX o%%o%o ß'1P10 rl
00000091wmo 000000000000000000000000 V in wb WNNo r4 (ý1ý vb ni moNMvNbnmNoNh H r4 r-I rl e-1 H r r-I rINNNNNNN NN N/'11'1 /'f API /'1 J
N N'. 47
182
d in ca
,3 v
0 ýr.
A m
v in rl
i In
aýi
>O 0 VU
0C" ui uNi In 00 i . -4 r1
ýr
0 OO N4
0
H1 U1 in U1 in U1 in U1 N IAU1U1u U1N P1 444444444444444 4 o wwwwwwwwwwwwuýww In w0V 1') W N01 O. w rlN O c w 0 m in O
. o Na) OýP1 fý t- Wrl W NPIOV1ýp %a
W in
0 W r! 0! fý V NOO1f-Ul P/NO f-iO 1n
14 01 01 co co co co co NNNNNW 10 10
444°4Q44°44444° " NNrl 0m ID. 1.10ý N in e .l
OnýO VP1N
. ;i
0OlWNb
.. "w
. fß'1 PNl
""".. ""
tp I[1 rl
. YVVV e-e V Pf PI P1 PI PI PI M in N
tnviinkn tnin vi invi tnIninIninin ei
o wwwýawwwwwwwIaIDww uiwm o m INnom nv wý m 44 N
yý Co NW N Co mN . 4NNN
I0 h %0 1'1 N. OT 01 c. 9 f- Ir w ý0 % -! Y1 "m v in "i
. NNNNN rl rl rl vlrlrlrl. 4. t 11 ID
.4 it
ýD'O ýO tO tO tD tO tO tO tO tO tO tO tO b
444444444444444 .n wwwcý. wwwwwwýwwww
2 01 V tl P/NO1(1 aD NO
N I(1 Nm If10D NO NOILI rI Vm aONO _4 01 h 1'/ r1 01 tO I- to IA NO aD W IA .. r
rt
4 w ýn m '-4 In pý
0 v Pl N
%0 %0 %0 %0 nnnnnnIn : d.: -e at no 0000000000000
WS enmrno. INukvU)eromo w co coco ca mosmosasosmma aso C) . a1 }
N to . Tý
o www , wwwwýýriwwwwwwwwwýOIDwýýwýýw0NNw 0 U) M0O NI wO NN at pp0p%. ý ON QN 1[10(10 ONV 01 ww If) N co WT �p0% 1'ýOt Mw ýO V Ob 1'11- NOt to O, ýO N00 VA in CO /'1 01 45 N 0% 1f1 at OfýNr-1
0 Oý aD fý Pb 1f1 vV 1'1 PI NNN "'I 911 O! aD aD 1ý tr 1ý ýO ýO b 1(1 vV P1 1'1 1'1 NNN. O)
1 rl r1 OOO NNNNNNNNNNNNNN 94 N C-4H 4f 4 rl vl rl. -Ir. I. -4 rlrj. -l. 4.1 r4 IM. 4.4.4 r4 rl rl 14. -1
II e g4-e 4r r 44e 4°qe 44kn 44444444444in Y444444Q°4444
o ww V wý ýn ýwwwwwýwwwwwwwwwwwwwýwwwwwwwwwwwwwý oOVoobnNo Co mVN m oVo00mbrr Co vi O %0 N 01111N 0r- pl cn 0ffy c2 Co 1(1 ý-In%0 ml, MNN N If11fl 1'I rfZ Nrl in lflýy 0mmN
1ý ýO lfl IA VV P1 P1 P9 P1 N rl rl r-I rl O00 v NOS ,Od NN rl N Olýdº101 m ýO V r11 fýýD IIl l'1N O
ririri. .4 , i. 0 .4 . i.. 4H . iiii. iio mý0 %oboto; cö
m Co Co Co Co Co rnereeIowwIo in uiInIAIIn i
Ifl U) IA In U1 In In In in tn In in in in In In In In in in Ifl In In In in in in in In Ifl Ifl U) In Ifl in U1 It) In In in in in U1
4444444444444444444404444444444444444444444 o wwwrlww. l10awwwwwwwwwwwwwwwwwwrýýaww wwwwwýaýawwýw tD0%" . ON NCONO CONIi w O4 VO Co 0% M
NI(10%ch NO UI NU101NTNO VMO AO (ý o0 1ý . -1 OM rl ýO vaM s} bM . -1 . -1 ýO N ýO fý NVvM V- 0% N0 (1 MNNVMP rý CO V %D co r4 N
N Mr-1011ý 101f) O MN01 W 101(1 V MNO m1ý lO 1f1d MNr-1 O01P bdMM N-IO01 OD th %O lfl Ifl Ph %o %o b w%o %o %o w) in in in in in in än -e VVdV«*, e VV MMMMMMMM MNNNNNNNN
.ý
w
in tn U1 U1 m tn 1[1 1l1 N in tn U1 1(1 Lf1 U1 Ill in 1f1 tn N U1 I11 tn tn U1 N tn tn IA vi tn 1f1 tn in U1 Y1 b %0 %0 %D %D %D '0 4444444440444444444444444444444444444444444
uº wýawwwrýwýawýýwwýawwwwwwwýwcýwwýwwwýawwwwwwwwrý. wýaww r rnvo an nmmv 0 rv rv rn rv
rn Co ýn omo intnmrýrvrvtý rv ro N in av rnmo. ý rv r) ým rv vmn CNrn li o . iýoov. ýmvooomomrvonn rl U)4 1
°ý Co o Ch mmn %0 in ri . -ýo mm mn%0 w v/'1/'I n ev . +. +. ioov(vm v. +u). ý
MM r) PI M 94 44 94 NNNN 44 44 N 44 IV HHAHAA vt iH of r4 MMAAH vl Ar chc%CO CO 90 PP
000o0o000000000o000000000o000000in%0 000000 evr
U) tn Co q
Oe O rl fV mv Ul WO n Co On ONV kn %0 n Co %%O
H rV nVM
rO mo rl N
Co V v
W v+ -* v 'ýº n amýºninnaºýcºn%n %ýnv %ýo%cýoýo%o%o %ýorrrrrrr9 nnO2 mmmmmm
qý44444g44°4w g, 4 44ý44ý444 o wwwwwwIwwwggzý wwwwwwww
44444 44qß 4ý4 ýýýwwýw wwwwwwwýwww 00 01 N0 0ý OD Vý Co MN0,4f, f4 O U) O O% Co c0 O 1lf a vo m Co ri U1
O N1ý N7ý0 r-IN V Oýy CD C% %00h v to Co 0 %0 rý 0% Co 0w %00%N Yl O ri c% f4 V On d t, r1 e cD 00 91P le W! VVr! r! Nß-1 0 0! lII loh D eU1 v PIN Nr-I r100e [O fA 1ý1: lO M1[1 V r, 4, r44 Z
ýD %0
%O wi in w) in to MmM in in in v44YVVVVvVV -w, * 44 P1 ei mm m PlA P1 m Nt ei PI mP1 PI N
:1 -F -W ON
o ýwwýw$4 ZWwwýýww
ýwýwwwýww wwwwwwwwwwJww co V
nömmmr: r, nrbbbadVVÄNNNömämämmnnneeW! W!
v
ANNNN 14 N 44 N 94 NNNNNNNNNNNNNNNNN r1 .4 .4 rl .4 .4 .4 .4 rl rl .4
.4 .4.. 4 .4
in 44 ýý4ý4444ý444 44444111?????? 44 o ýwwwI14ww
4 wwwwrýwýw
4ww wwwýýw w wýwwwýw wwwww mim No mmmn to my d rNV v rn in to co N nmrin rn
N in In vVV PI I'1 P/PI P1m m440
rNl r0 /o o in oc,
o bV
1N/r1Ö amp iou)r'INo
cQiß AH ýi 1.4 11 '. H1.4 . 4"0 11.4.4.4.414 . 4_4 . 4AA. 4.4 mm0% ob 0. Co Co Co Co 90 Co r nn
4444444 4°44Yý44ýWwwwwwwwI4WW4ýýY4444444g4 wwwwwwwwwwýwwýwwwwwwwww wwwwwwwýwýwwwww 1ý
Oa0
A
Co (2 01N cl 0% tn
wOOddIO 01 V
oU1 P1 ýO dýO dO O0NNIý 09
ry ro d 1110 4dm rIr X010 ri ooYl 111 dý-I r1 X010y 0d dN01ý1'1M NUl 11ON N CO
Co . -1 U1 P1 c1 t- 1lld Nr10ý00 h ýO dMN 011ýý01(1 dN00' P ýO d 1'I x-1001 c01ý ýO Y1d N100ýfý Ö1 Ö1
Co Co Co Co Co m
fý fý NPhhr (ý %0 %0 %0 %0 %0 %0 %0 in Ul Ul in H1 Ill Ifl MVVd -e
VVV 'V
V P1 P'
1
[ý 000000000000000000Q000000000000000000000000 W iý 0 94 00 0% 0 Co 001m -W: 2 %nOU. 0.4 NM0 0% r Co 000 94 Mr in %r Co O0AN
1--4 4A 4 NNN fV NNNNNNMMMMMMMMMM ý1 V
183
iar
r- Co
w r co v
0
41 v
9 0
w
r m v
R oco
o00 In
41
F
0 ill
E 00 .i rlNý
ri
I IN
ti
U1 UI ko w%D%0 w bb%a %0 w%o b%ob ko%o b 40 w%0-O w wo ww bPP
4444444444444 44 4 44444444444444444 o o wwwwwwrý. wuwwwýawwwýawwwwwwwwwwwýwwww motumoko o ýnm%o o%o % mm o. to c fn n In N wom C v me nNO o C; win0%Dnmr+cl OmNmNOmNri rmmo'omrvrn. ýýouioýon n in
.ýr 0 v oým.! vovmvo1oIo! muiorn1oI19onýnvNN. ýooým o ºiý01mmNN %D hY1 VV 1'f in A1'I NNN NNNr1 r-4 ýl rl r-4 rl m x101 40 m
I
v
%D %D %0 eee%0 %0 W%D %D e%0 %0 %0wg0g nnnnnnnnnnnnn o 444444444444444444444444444444444 Y Co o wwwýwwwwýwwwwwýýwwwwwýwýwwýwwww o Co (D OO NCO V ONNN O Oý Wno 110ý Nor-b tn Co w
rI01 ri O r1 Ch 101'1 P1 %001 r1pý P1n NP/r1000D 00 mV Co
r- Co 010 Co
OO nV 1'/ N fi N rn r!
. dN 0! ý. dr10 00 n 10
.N x-10 V en P1 Y1 W co AO ODNOOý p1
If1 YV PI mmmNNNN .1 rim vl vl vl rl O% Co nn %0 Ifl Vr ýI P9 /'1 PI PI N rl
mmýmmemNNNNNNNNNNNNNNNNbNrrrnmwo 4444444444444444
wwY444444 aU 444YQQ444ý o o wwwwwwwwwwwýwwwwwwýýwwýwwýý woe UNO
wo moa mrvm mýdmmmNý ma o O
CO CO W Ifl IfI OO CON a r0W 0a
a 1[1m NN m Or :i r-1 -m N
1f1n NQ1
Or ý/ rl Oh m ý0 Y11'/Nr101N N W! ýD C! r: r! Ie lD N ch2 a N'iON Oý NIl1 h
r-I r11-I rl r-Ir-I r10% 0% 00 r'0%DSO VV PI in PI 44 NNN 11º1. -1"0 re v101rr%0 m
444g4ý444444w44Vwem CO Co 4co4co 44444444 o wwwNwwwNwwNýwbi 1ýc"cNCowwwwwwwwwwwww qo Vm
1Ný 0, b
ry vm Co m9ANW NO ý
O. .
le
. -I r. 0! N0u% %0 NN 0101
b 1ý1
M On1
äb b0 W r1 P1
Co V Yý1 c" c3 Co Co Vm PI mNNH ei H H. 4 rt co co CD%o%o v) VV mNNNN N rl Ar oºf A r4 rl %n
00000o000000000o000000000o0000000 v-elenCo 0% 0 Inrvnvin%0 rLomo11 rvnvhenCo rnor-rvrvn W0
vvvvvvvin nnin nin ain Lo neo %o %o ee%o eeeennnnrn ýx
e1 N
to in U1 In If1 WI M WI WI in in in U) in in in in in in in IA in WI to U1 U1 to IA to in N IA OO UI U1 IA in U1 U1 in in to 44444Y44444Y44444444Y4Y444Y44YYY4YYY444Y4Y4
o ýawwwýaýawýawwwwrý"wýwwwýawwýawwwýwuýýawwwwwwýawwwwww OO I O OD OO
tO coa r- NNa in V 00 Oi co co 7 OD V U1 fD 1(l für l(1 Pf b ry d 01 W'oP O NOS Pfýy V wh ON GD c0 4 w'O r-/rl ýD t0 MIfl 1'1 tnNpý %O qr1
m NPl kjN AV V 'i 1[1 N 1ý V O1: V NN1f1 f'1 C! r: Ifln 'Im co ul T Arl 0! rý1O ul tANOOý OD 1: 10 "! vN1 NNr-1
0% Ch wwr. P 1-b%D%a U1 In to In VVVVVm PI P1 MMMNNNNNNNN e4 r4 M ri MM rl M MM
U1 In UI to In In In in in in In In U1 In In in In in In in In in In In In In in In in Ui In In in b %o %o b %o 10 10 %o 10 %o
4444444444444°44444444444444444444444444444 o wuýrý. wýaýarý. wwwwuýwwwwwcý. wwwwýawwwwwwýýcýwwwwwýawwwwcý. vN in r mi m On Av c4 NO tO Mnvomm nNONaV oo No000 O oo Ulm Co OO Co Co N Z0 0m'D
W In N ON
% Inm m NNN1ý O P'1N ONN 10 IW N mit V NOS nlD V M. -IO Co lfldMN'f0 mitýO YI V MI'INNrI tI Ul r-1 V 011(1 rlm V
VdVOMMMMMMMNNNNNNN NNý-l rl ri ý-l º4 4.. $. 4. lri r-I ri r10101m tO N *0 bblfl 111
Ml N in NNNNNN Ml NO in Ml in in NNNN to bb b'0 bb b% b b% b0bbwbbwebb 4444444444444444444444444444444444444444444
o wWWWNwwwwwwwwwwwwwwil www ww www w wwww www vOOo '0 0Omm rv NNNNO rv r4 10'n In`0 m rv rv Mdennw rv Oe 0% )N 0% w ýO rý 1Z c% r1 0 rI V 0% m !Z' In 00 Ln veoNZ
CO Ze
ý0 1f1 N Y1 '0 MNN rI PI ,
ýO aD a Oý ý-1 P ýOd Mr-10 p1 O01ýý0 ý01f1 V P1N Nr-1001[1 [O OV 01ý , -I co OD in If1 N 01 ý0 YI rý 1(11"1 NO
NNNN NN'1 4ri r-Ir-Ir-1. -l rl rl r-1 r1 r1 r1 .ImW co co N r. %o %o in Ul in -e -r+ 17 MMMNNN NN
In N In NN In in N N%0 b%0 % b% bbbbbbb b%0 b %D %0 b b%0 b b%0 b b%0 nNNNNNN 4444444444404444444444444444444444444444444
o wwýýwwwwwwwwNWwwwwwwwwwwwwNw wwwwwwNwwwwNw MN In WW
P1 Co WONUO
UI Co CO ýy uin
tno %0 l-O
YI 'in c" rO
Co NOPdA0 00 M WNin C)N.
Co 0dcOd
01 Dl ? O1nN
dO,. 1 ei U) rN
0)N N0) 1f1 NN "4 04 In cm
N 0N 00 W 0 a0 9 Ifl d1'1 Nr100dtýr-Id 01d 0ý0 NcO dr1 a01(1N O. P dN 0 U) 01 tý Uld 1'1N0 It) NON If10ýin lO
rl rl . -1 rlrl 1.1rl AH 01 Co CO 1N ID %0 %0 U) I)ddd P1 m P1 NNNN Nr-1 rl v1ri vl ei ºi ob CO P ID in U)d
00000000cDmoH9 000000000000o0o000000000o0000 W rlN/'1VYIýOl-ý00ý H H-I
r-1ý 4rri oNrW W O1NNNNN NNN 0. -4ei nffn
in n co nW n eO Oe
ooeýýýnrnnrnrrnmmCo Co Co oCo omrnrnmrnmrn oo In °444°ý44Yý444°Y°4°°ý4ý°4°°444Yýý4ýý7 Co rd
Wmo cO N Co PI V Co P1 M CO P1 .f dr ON rIONN iO V On V
o%f0 Co 00 10
Op co OV
wO%
cOr NfO CO N1 0n ro OHO ll)e00n 1l1NrbN7%O NCO U10ýt0 Wr vN np PVr! t0 P1 0V rý ONd lý le r! al n l0 N 1(1 01 P1 9 1l1 1f1 ON Ill l1 c2 lO le ViN
P1 M ra AAH co %o in V mNrim Hcnr- %o V P1«4A ri co r- %o in V mri A AA'4 co 4
I o oOOOOOOO O OOOOo Co OO0OOOQ4ýý Q
o wýýOOwýýýww, w Aý . Aw 2
ý, w2ýýwwwwAwIfl ul Aw Cl Z V Of W Pf O r'/ mN "T aý
Od1! 1 Ifl O Ifl %tOp
ClN N
ri Cl P r1HN V
Co r- 10
WNä 0 Co bN Co
Ill , 'm
14 0% ! dH
V 1O'1
(. 1H
NH ÖÖ ry
O rl
r-1 A rl 01'0 IA V PI 44 rl HH Co ID in m r, 44 r1 H 01 IO . 94 NHA rl Ch N -e f'1 r1 4-1 rl rl 1f1
4rý4444r-4c% CID mco4w 4Ymatat 4400gISItIII 1-1 ++o o wwuýwwpwwwwwwwwwwwwwwwNwwwwwwwwww w co in Oma! NN r'1t000NýO a mm"ww W no GDto rl IlO r)U) I11(1 O1111! 1 0 N rl hhON 7%o co i co to 1- (n1(ý mg co ui r4 m o. ý ko to v m%o Aw ui rno co 0 .1
r-101 V c0 Y1(1 aD MO V 1n 01 'iW! Nc0 /'11ýOIfl r/CD 1ýýO 1-1 r11(1O NNO ON
rl at %o V P/ N vl ri rl 1N in mNN rl rl CO in ONN rl H%O VmmN rl 4 to MO 0)
v I rrrnfý nmmCIO mmmrnrnrnmmoooooý0 to 44444044444444444ý7754774
o Wwwwwwwwww47 wwwwwwwwwwww o r-1 rn NvmNmww. m .ivmOvNm rI v rn 1O111111111111% ONr t` ObVONNNnrO00O r 03 VM UI 0N 0% r1 N0'0 X0 01 r1 VNWWMMNO IA N
10 If) PI 94 r4 rl 1- V PI N . 401'O VNN rl 0 10M N. 1 M0 .4
00000000o000000000o00000000000000000 WS1.4 NnaW) erCo mo rl rýýy ýv 1ý
rl er Co mo r4 Nnv c4 en c4 mo rt Nnvr NN rl 'ON
NN NNNWa . -4 PI P1 NN
184
d N
O
r- w
. -1
9 N
43
0
a
d
r- co
dý r o. I co 0"W
14 0i n2
ä ý,
aý
.i0 +i O ýýýjjJ ÖÖ
0
w
ýd
w 0
C], 4Y4 ý 444 41 w wwwýwýwW ww o wwwwý w N N in Co Co -r Nmm
NN V Co r) ö
VO Nn N
0NWn Co O O m O OmY1N 01 ,0 vNOOým ý01f1 v .i .T
rý %0%0%0 Ul wf In in Un h. p VVVV rf
444444444444444 4 o WWPWWWWWwWWryWWW ry Co rVý r) N O
Lo NN Nfýr-I Co NN ß-11'f in r1 O Co
0% Co 0
ine in In Yl V Nri G1 OI9 r: l01fly P1NN a)
V Oý
r) m In INNNNNNNNNNN fN
In in in U1 in in in in in in iO in in in iO 444 44 44444444 4
Ff
OWWWWW WW W . ý. 0$N UI O VW PI
W (Yl %0 N0O
r4 N0NV
r1 I00 N 0 N O N r; Co
0 0 V rl N . -I 0 N i-I N N 'Vi rl00 01ý1-1 `r VN 1ý Yl N Ifl
I
v
to %D io %D %0 %0 W %%0 to to %0 v%0 oWý www ww ww wwwwww
fý rNnO .1VVoo to
rl ONNONNN
In N %D OO a) O -1 %0 bVO Of O0401[1v V! r! e-1 00M
r o +Nb ý
NVN N
m P1 m" NNNNNNNN 14 N 14 co
000000000000000 bnmmo at NvMbn Ch rn o W OD CIO co co mam0% 0% M mrnmao
ýt y
o wwwwwwwwwwwwww wwwww wwwwww wwwwww ýwwww vmomýrvvým°e ýý rvýným mv v nom 9N4 Inm mavm ö MrvrnrvrnMmovrnýýnoýmvmNrv v nmýý oomonnmýrvmmýen ommmnneýývvMNNýoorn rn mnn Si
hvvnNNýýoomrvmýnmýn ý'1 NNNNNNNNNNNNNNNNH r-1 r-1 rl ý-1 rl rl ri ri H r1 14 r1 11 1414 r1 14 QMm0 cp rNn
I dd dd d ddddddddd dd in If1 In In in If1 in in in in in in in in in in in in in in in in4 44044444444444w444444444444444444444U)ininU1 4 o wwwwwwwwwýawwwwwýj w wwwwwwwwwwcowwwwwwwCaw
cocor, ý. 00Md O
'0 r, r4 Co Wd
0N 0 IA ri M CO Nd MV CD m oo
ULn to
)Nm1ý M01M
Nd Q1 ýO d rl OA O1ýd O InM O, dN" N pý co 01 0n N ri co N1ý co ul N a%9 0 %nr4 M Nfý
d r4 ^ ýD IO fl in ddddMMMNNN1 1-1 rl OO 01 ýO rl co 1(1 MO ýO dO fý in 01 CO I(1 M r1 Of ýO / 01 .
r4 r-1 rl r4 r4 rlH r4 ri ri rl H ri H r1 HH1 11 11 Co. cow tZPh%a 10winvIYl Ul VId dd d dMMM
I[1 in 1fl ui U(1 in Yl U1 v1 u) in in U1 in UI U1 in 1! 1 in Vl in U1 U) U1 in U1 in in in in U1 It) U1 in U) ui in in N V1 V1 V1 1A
nm ui ý°`ýYý°444444°ý`ý44ýý444444444444444ý`ýQ4g4
o wwwwwwwwwwwwwwwww° ww wwrý. wwwwwwwwwwýawwýwwww on ao 'o u 'omN coo'n o 'o . oN nN rn eon, o vo .r NCO v LO inmýýon`. ýuivýoýrnoov o aýon ýnonnin wom. rrnmvinýoýoýaor+o Nb P1N 00 00 ýO d C! C01i oP1ýpl
ppn Yl PI Na n I(1 V Nr-10ý W ýD 1/1 V P1 r4 14 Ow wn w lfl n nnnbýG ýO ýD býO Ill 1[1 If1 In hfl .f ar ýf "y . fd r' r1 n t' Nm 14 NNNNNNNN HMS 4M 4r4
.4
ui In If In in in II) In In to In Ifl IA In IA V1 U1 IA IA Hl in U1 IA In U1 U1 U1 %o b %o %o %o 10 %o ID ID tD IO %o %o %o %o %o °°44°°ýY4s4ý4Y4`? °44444ý44444444ýY4Y4444Y444 ýn wwwwwwwcýwuýwwýwýawýawwwwwwwwwýawýawwwýaýwwwýawwcý" ry V OfývNCO ýD aD 1L10ý Oý NmýO V ýT Of 10 1'1 Nrl
Oaf 1Aw%0
brß-1r)v Co rý ul I(11/1 Oý1ý1%0 Co c0 n1'/rl v 1ýýO ýf !D Ul
/'1 Oý V
Co vPl fý0 ýO Ifl 1'1 Nr-1O Cým ýO Yl vf'1 r4,4 O OOf Co raw in v ri N1, Or -4%0Nb r ri mO
P)PI/'11'1 N1NN NNNNN NNN1 r-1 i rl r-1 r4 , -1 rl rl rl rl OI mCo %0I0 in U1 VvVVV Pf
0000000000000000000000000000000000000000000 "-TOW nmmoHanwowrwm O"" m P" Wn wo OHM nvown WOO HaMeN .Vvvvvv0m4 in ul inn 0 00 bbwwbwbbwWnnnnnnrnC. nwmww00
dddddddd It d It ddddddddddddddadddddddddddddVdd
°44444°Y4°°44°4444°444°ýV°44°°44°44444°ý o wwwwwwwwwwwwýýwwwwwwwwwwwwwwýwwwwwwwwwýww 1(1Ný00 dtOOd dONPI op ýO tp ýp cprv mO Nlýf It CO Op1N NO0 pý LA CO NCO aDN dN O P7 C0ry OýOýO NI ýOn NI(1N nr11-I YIn N
d 11 Ylýinnmvroooýo coo l'Id k0 t01'1n N1(11f10% O rl OýCO oO nýO hfl l(1 dý'1N Nr-1001 O" 1O n'Od"-001 ODn %I.. r P1 NCO
ebeiiriýtiuiinuýýciýtiuiýciuºvivvvvvvvvvvsvvviririoirirºvimmmm ri
N
I vvvavvvvv'I. vvvv vvIF
www wwý vvv swv va 4°°444? 404°4 ° ?? -?? zzz zg 4a4 44 4 o wwwýwwww www wwwý wwwww wv ýwwwwwww
to Oh 01 N OD 1ý1 O ONO MN PlýO1 wOýd ýO P'1 N OYlh V vOý ýO 1"1 ýO 1A V O N01 X001(1 O hm p1 %a Nw 14w NOS w Nmw Nw V"w O1f1 PI01U10 ýO NCD a}. i Op ýO NIA V 1[1 N'i 00 01 01 00 [0 P I, ý ýO bW to u) It Vw 1'1 /'1 1'/ NNNI r1 000,01 0! OD CO fý hVb ýO h Yl
OI m PI m 14 NNNNNNNNNNNNNNNN 04 NNN Cl 04 N 94 N rf MM 4rl j MM rf M rl ý$
ý4 °°°°4ýg4ý44444g g4°444444 ?????? 4? 44? 4? 4 o (ý]wwwwwwww(ý] wwwwýaj(aý(aý ww[aý wwwwww(ý] www wwww(ý w{ý www ri Yl ul 0% (A 0% co M r1 CO IV co M iý 1[1 ý0 0 1ý rl dNNNWd Iý 0% 1-w OpN %0 01 N t6
min r-1 1f1 fý N
N NNr dN O a01, cn W in M ri 01-%Od NO co %o dfy 01101(1 MO d Nf- %o d r4 r- in in co N
N 1ýý 10 ý01f11f1 Y1h d dddMM MMMM NN NN Nri r/. 1 r-100000 W bM 1101P If1It N0ý r/rf f1H'4 rl HP4riH rlH rlH ri rl 11H rl H e4 rl riH H r-1 rl v4 HM ri r401 at M Vi Co W Co CO CO P
VVV tn tn 1f1 in 1(1 1(1 1f1 N U1 Ifl tn U1 tn N 1t1 in tn in in m in in tn U1 in U1 m 1f1 in tn in Ln in to to U1 in Y1 In in
4444444444444444444444444444444444444 u w4wwr!. www4 wwwNwww4 w4 www0 wwwwwww4 w(ý] wOý. -1 wý] wwwwwwwwwwwww O ýO v Oý Co Co NN 1'1 01 0% CO fl N Co 1'1 W 0% OA 01 v I(1 Oý rl O ýO 1ý aD 1`l vv ýO ýD N O1 fý N
0% I, NNM r) Ul %0 O, vr f- 2VahhNP
rI rý O . -1 LA N'n -f ri rl ri ON t0 N ýO N Oý Pl 01 .1N [O Ifl º1 999 1l11°ý. -I a'. w O týI NO W 1O V 1'1 ri o wN In rf NO C. N Ul vN .'4
. «e9 wMmN 0001 fN'. 0
, 4,4A at to of 0101 co co co co co m^ r. n^ r. n %o%o%o%o %o %o wiin bn in in in VVevVVVY PiNf
00000000o0000fy000000000000000o00000000000$00 W 9. r1N r1 V 111 N01ý Co Oýrii 1.4
.4 . 4.4.4.4.4 NNN tNV N".., - Co m0 14 r4 r) -e k91 %0 > CO
NN ei NNr P1 m r) mm ri r) mnV rr
185
O 0 r
0
0 .4
43 v
9 0 0 1
17
1 93
d 0
E om o ". + oow
. -4 r+ o
di
E o .1 o ", o
h N1'1 rl
vý
0 ö1ö
,
44%0 4w www ý4444444ý44Q44ý4444444ý 1 o wwwwwtowwwwwwwwwww. 1wwwN w wwwNw f4 f4 Co "d r N'NNo 4 ö moriommrnný Homnemööön°ön'4Nri%n0m'4v'4 v
vmnýanýmvonýondNmn9nom9. ...
. slirv .....
ýoeoýn :3v ................... mmmnneernn-e vvýM ri MNaNN. -t. 4.4.4.4.4.4.4 Ch Ch Co n. m
I'
w
I
4%0°4444ýý444°P444444ý444444444 m o wl wwwwwwwwwwwwwwwwýwwwwwwýww w 't Co OmN wl mO %0 r4 01 N0%Wo fV 01 'O [0001 V PIM Co 11 O0 Owv ý0 O 1'1 O WO P1 0d NOW P1 . -I OWVN in 1ý1 r1 01 %0 O Co P1 01 1fý rl Ot O N P1 rI lD 10 AºI aD ýO V N000 NU) dl'1 NOO NN ýO ýD vN011f1N Co V, 4 CoV
YNI Nf ri MIN 44 NN Nrl rl rl rl rl x-11-1 ºI C1 Co %0 in 66 in Vv vA PIPINei rl
mmmmemrrrrrrrnnrnrrnnrnnr4rmmmmmm 444444444444444444444444444444Y
o wwwwwwwýýýýýwwýwwýýwwýw wwwwww .4o mw orvvrvNmo orv m mit Z5 Irnövvv öN O NN N Om mmOtorr 1'1 r-1r1 VO ýr-irlrlr
N 01 CO 1O hd Ar10 H10ý0 Nm OV OP10ýNm A0! f11°IN r10 rn OIf1 OQqy
r4 rl r-I rl 14 '40%w mr %a %a U1 to VVm PIN NA AA AA r1 ýO b Y1rA 1'1 A 1'1
wn uzzý4ý44444ýý444w444444ýý 4m
o wwwwwwwýwwýwwýwwwwwwwwwwwwwww w00w 0 Nco at r r) CO rvM onnn mmr avveý. ý If1 v r1 O r/ O in co P4 n If)-f A Im co rNN %o v
Or rr OO fV fV IA in n P1 rf PI nr rv rl WMN aD rd rl .
99 co vONN b'-1 Oý 01 pý vv º1 r/ r Vý d ýO n0
m P1 P/ fV NN ri º1 r4 rl r4 01 r %o In vVmm r4 4.4.4 14 m r4 to 00 co in to in in V
0000000000000000000000000o0000000 f"1 dný0 NW Of Or1N 1'f d NEON [p NPI dn%01- Co c. 0m N I'I do wi dddddddnnnnnnnnnnývýoýoýoýoýoýoýovýonnrnnn
e z7 q
ý`1 N
dw in in Ui in Ui to to to MMiin Mto in in to MiMiin in InMiMiMiMiUiMiMiin tn Miin Miin MiMiMito in Ln Ui 444444444444444444444444444444444444444444 o ýwwwww,. wwww, ýwpwýýwwwwww wwwww www wwwwýw. Od1[)elý NOVV' VM Ob 70 co mw 0%0 11m ui nvrý CO O
NO1 O I'1 1' NI If1 NOO 'w rI %CO hOb- i0 to C0 0 1(1 at 1- NwV1 01 P %O V7 fn NV %D h pý N
VOO 01 1'1 hM 01 in rl 1ý Pf O Ir V a0 ýO V ei 0n 1 fý Icl 1'1 NO co fý ýO 'P 1'1 NO 01 mh 1O Y1 V Pl N rl OO
"-4 r10f 0100 aD Ptah bbb IA IA in VVV 1'f /'f r) MM /'f NNNNNNN. 4.4.4.1.4.1 4 . 1.1.4 1
In in In WI In 1A I[1 in If1 In In in In In in In UI in in In in UI IA IA to In U1 in in UI UI UI W %O w %o %o to %o %D wb %0
NOCIO 4444444444444444444444444444444444444 o wwwwwwýawwwwNawtý" wwwcý" wýawuýwwwtýwwwwwr.. cH. wýwc4wt. uýýý OT vv ýy Iý 1[l OVOO
On cm' OP 'D O ýT 0 01 N f'1 01 u1 N 1ý MOVO tr1 ýO r, nr If1'o ý fý 0 t,
N f-w A c-4
wVNo in
w 01 in -r
1ý 01 14
in O1 ýN410V r-1 V000 N 1O Ill IW 1l1 NWVN00:
. 11
.... in co W 1.. 0
. Y1 V
. 9c!
. -1 00 Y1 O1O! "!
...... in IA to VVVV r) In mmcl m NNNN NN rl lrlrl. v-4 .4 .1 .4 lr1x10101 WP 1rW %a U1 U1M-f
tn In in in to in (AIAin in in IA in in IA IA IA IA IA ID ID '0 %a IDIDe10IDID%a IDIDIDe'0%1010%to 10%0 10 00ýoo444?? O??????? O???? O? OQ????????? o44 44 o wwwwwwwwwww Qwwwwwwýwwwwwwwwwwwwwwwwýwwwýwýw
,, o !I Lcn o MmOdMNOýO hI(1 c0 f'1 MM iO SON ulO Mý0 ý0 ýD ddNUld mlllO aDM OdO 01 C4 -W in
10.1 c, j 0 IOdMN O010Dh%o in d AN'i rr11o
co Nw r11k1N bN co v 0r n in 2,
i ä. ý.
ým. ýww
Cl P7 4N4N4NA
rl rl rl A. 4 rl rl rl AA c4 o% W fA {-C-SO SO Y1 Ul 111 VV4A Cl A4N4NN .4 .4
NN in UNNN in N%o b ID b ID to bbb ID b to ID b to ID to wbbobobbNNNNNNNNN
o ýwwýw4wwwwwwýýwwwwýw4wwwwwwýwww4wwwwwwwýww
kn 0 :Q
%D "r "
%00 0 I-F 0% ri O t0 IO IA AN .i O1 rl PI NqM
%a co C% co 01 Yl O ýO N Oý Yl N Oý 1ý d NOW cO U) Pf NO Pl U) IO WO 1O O IO rl
NN. 4 "1 `4 r" 11 -1 "1 01 01 CO 10 In 111 I[1 VV PI P1 P1 NNNN 04 rl rl ri rl rl rl - 00 Pw to IflV VV
000000000000000000000000000000000000000000 oA g: 3 A cl to rý co a% wä0AN 1'I d in %O fý cc O% N4ý
ý1 H Ný
UNNNNNNNNNN PI rl
M P1 P/ N1
'0 r) W
M P1 MVYN
w
w 0 ýi
ýoýoýoýoýoýonnnnnnnnnCo Co CO CO wCO CO CO rn mm mmrno%$o in 00 00 000
wwww 0
ww1 044444ý`? 41?
WW .. Y1ý11 0 ? 1+ o o wwwwwwýawwwwwrýýawwwwý. Iwwwýawwwwwwwwww w .+o KOMM NNO 11vb MAD Oý000 ryM OýNm mmýDm OIf10 A CO N001 NCO UIMrin 0ý0 V ýOn Y1Nm VvV m0ý'-1 . -In NvmmO ri OO
H1mMO94 ob q Co Valnm V dNM 0ýM
, 0Nr NOO CM
ýO r-1
MMN A141i m%0 VM MN rI 14 ri m%01f1 VMN rI r10% n in in ý7 M r1 ri m ri 0N
v I
%0 to WNNNNN NNrco mooco owoammma OOOOOOO tn 4444°44444444 w 44444444444wwwwý
m 4ýww 7ww7ýý+
o wwýý6 wwwwwwwwwwwwwwýww w
ý. o no rn1nvrrnrnN0r 1[ý0N meUvrl"ooon oll to d OI'11'I d t11'I NN NNOý a0 bV If1N NN 01U1 N'O N, O Vr 4r4000 0 OfDNNýON00/'I CO M Or-I rl ýO lfl lý Olfl N 011'101(1 d0d rI dV 10100 to
N riH 01%D MV M C4 P4 rl r4 Co %0 V MNN rl r4 Co 10 -W ri NN r40% U1 In in M 91 O
WWN omo ow omrnrnrnrn ooooo4.4o oooON0NOn 0
nOoooOOoOOOOO447ýH 71ý N o www ww ww ww wwwwwWwwwwwwww w-o aomoinýoooýnonýmmýotý-ýýorýýýn 0111111t to
to r4 0 to
................... 19
..... co! "i r1 %0 0 PI 04 ri r1 01 1N U) -f NN . -4 ri N VI MN .1 0% 10 IHH 0) 0) 0 ca
rrrNNNCO CO oommrnrnmmoooo. 00q ý° wwýo wWWWAN
galz444 o
oQ www w .+ wwýwwwwwwýa(.. ý wwww O Oý N PI 00 NN a0 Yl Co fý O% a
NVN Co O1111111411111O 1.4 Cori. nin. Imr4tnrlniýor2A In Ul O rlu)oo
0öe ýp tp V P) . fl ON Oý d t'ý 1f1 rN Co N ý-1 NON 1'1 1f1 1! 1 O 1ý1
%o en mN vl rl rV /'i N . 40110 P1 t4 . 4e- V N. 5 VVo º1
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Fi S rlN/'Idlllbfý0010r1rytý v1! ý Torcomo, -ýrvnvuiýnnmm o. +rvnvýn ýy
v ý-I ri r-I r-I , -I rI rI rI NNNNNNNNNN I`7 /'1 /'1 1'1 /'f 1'1 vv
N ýA : J. J
186
i In
0 Nn
n in n
a
0 v r
d oo no vo "m onn oH
44
o ötýö aý
444444444444444 O WWWWWWWWWWWWWWW WN co oNOWWovmNOU)or, n0 oý
to 90NoN U) r- 94 Co vrnC. to 0N 42 vC, 0, r, evaommný n eeininALO vdri ri mri A
I in tn tn tn kn in tn tn tn tn tn Ln U1 in U r 444444444044444
o wwWwwww4wwwwwWw Ul IL)O ON P NO NN N .1o O 4
. Ol OVn4 r, -e N 1f)Of0 I(1 c0 r-N n 0
.4 ON
I[1 cl VNN CON ý1 v1001GD00nn n
1'1 NNNNNNNNN .4 rl rl ri .4 N
In U) U) IO %D %0 10 0 %D 10 10 0 %0 %D %0 Pf 444444444444444 4 o wr. ýýaw w
o% %D ýo ww wýa
vo 0o on ý, rl N yý ýÖw N N wvÖ bV Öb
rv CYN
N ON
ºI frI O i o! l 0 1
.i 14 "4ob0ic (' Nre%U)UI( e
.,
w
'0 %D 0 %D %o %a %D '0 %0 %0 kO %o %D %o 'D Pf
u NWI4WWWwwwwwwwww rý ., N V Noo V tO NN Oh Nmm 1(1 N(AN N01 o1. Plr co 0 (ýI
r-1 OAP1DNO tOWNtO VV NN Nr1 N tV NNNN x411 1.4 .4 rl 1-4 rl rq H rl
00o000000000o00 Wnmrno4 arN vin WNWrno W0 Co Co Co Co 0% 0% m0% 0% m(rb C)% Q% mQ
NN
4WWNW °°ý4ý°444°Y4 444444444°°QýY4°Y4444g4 o wwwwwýwýwýwwwwwwwwwwwwwwwýwwwwwwwwwww ýw4w monvo 0 rnoýmrnrnmmeýrnmýmý mrnmvvmovm mrn m rn O OH OONW ýO mvm of Oý vOO r1 0 ýf Oý O PI 1ý ON r1 1ýI co m 1ý1 O Ofl oNr
r4 % p. ri IN
rl r-1 O01 mýO in V aT P1 N"ý. -1001 m1ý1ýýD Ifl Ylvd h NNrl riOOO mm N On 1-Pf O tý "...... ".. "..... ".. ".. """"... Cl MCl NNNNNNNNNNNNN rl rl ri r4 "-1º4 HH Hl qMH -1 rl r4 HHH 0% 0% co co fr r. PP iO
VV iT VVVVVVVVV dd eV In in UI in In in UI to Ul UI H1 U) U1 in IA U1 UI in M Ul UI UI V1 Ul Y1 IA in
4404444444444444444444444444444444444444444 o wwwwwwwuýwwwwwwwwwwwwýwwýawýwwwwwwwwglwwwwwwcý 1'1 "! O vN01(1 00ý V 11 ýT 0ýý0 V'1(11[1 c0 P1 ýO V Pf MOWNMM OO M-0 NEON O 0Lnry Ori ri V tD r101 V 0ýý01'l0r1)al
rl Co Co r-N cOmpp OLn%o v0rr in"ulMri A0 in ýO III I(lv V PI P/NN Nýri 0000 ýO d drl lý Vrm 0ýý01'1 O CD ýO /'1 r1 Oý ýD 1111'10 Co (ý in VN
. i. i. i. i. i. i. iH iA. i. i. -. - AAcio+o%co co mrrr- %o %o %o %o vintnw)vv-ee emrioimri `j aý
444444440444444444444444444444444444444444 o wwwwwwwwwwwit wwww WW ww wwwýwwwýw wwýw wwwww
Co ulwo in Q in Co (4 %0 ri rý -e :2 Co -e ri N Oý "D V r1 a0 ý0 V r1 0ý W ýO VNO aD 1ý in vA r1 0h r2 ýO d P1 ri 0% t0 ý0 1(1 V r! NOO 0% r2 I[1 ul
., VIN
rrh (-%o%o%D %o Yl in in Ul to in eYVdemmm ei mPl NNNNNNN N fNA rf A Am AAA
v
1! 1 In in to In In In In In In In to to in In In In In In In In in in In in %o %a %o %o %o %o ID %o %o tD %0%0 to 1O %0 %a ID %0
In xz'L ýwýwwwwýwwwýww4wwgwwwwwwwwwýwýwwwýwwwýw
ýy v01ý0°"an,
46wmminnä°výýö maýOOöýO°vövmmM°° CA Nmö nrnn n ri V
Hom, %01(1 P1N". W hý01f11(1 d1'IN OO%O CD V %O PI 1O N rý 6 19 0! 19 dr-19
In In r) In NNNNNN NN. 1.4. lIt . 4... 1.4 1.4.1.41101 co co r- 1. ID%0 M Ul In VV IT In In In m In
0000000000000000000000000000000000000000000 (m e kn'Oh Co c%O 4N m eu1%01.03 0%0 1 N1 V tn%0 rý 00010't 44 PI Y in W 000%014 N ri vIn Wwvvvsvv
vin ui in in win in in v) in %o %o %o %o %0 Zo %0 %0 %0 %o nnnnnrrF: nrm0 Co Co 40 Co
wýwwwýýwýýwýwwwwwwwýwwwmowwýwwýýwýwwwýwwww rn ýnm om nmmrnemý oemm wN n dn 0% O ON O0(, NO r1 O f- co OriN On m wnm mrl d1f1 t" co cn Nd UlO %D O N. -101 ONd dm en m OmO Cl m 1O in d en NOm 5% mh 1(1 d rl 1119 O Oý m fý ýO 1[l 1f1 d P1 P1 rl rl O Oý n fý ýO 1! 1 d 1'1 P1
!ýh lZ%a bb wb bb b Lain to Ln Ln ui in wi m vi vi dddddddd ddd d enm Mm en mmmm
1' o NMýwNwwwwwwwwww(yýw
wýwýwýwwwwwwwwwwwww O awMr cj 010 ýry in rIbNmrm-1'o'n rl 4 -w MHhMpý
b01b NbÖ to
N co
A'l Ö r- -W (n 1ý ýO 1Ild dM NN rl rl 000101m co 1ý 'O Lf11f1 d MMNN. -1ý-1000Oý O1m mr
MMMMMMMMMMMMNNNNNNNN Cl NNNNNNNNNNNNN It ºi 14 .4H . 4.4 rl
ý44ýý°Q°ýý44°44Pvý444? 4444ý444444ýp OO0
Oý O 00", wwwýwwwwwwwwý(aj w wýwwýwwwrywýwwwwwwwww ýO CON w NCO NAT N
P1 co at %O O 00 ý N
BOO rlNin M10100 fN'IýIIIO of
N fý VONN Oi ýO r1 01n in N co to c4 OvN 01,1f1N O 1ý 1/1tV 1f1 N 00ý 01 co co co 1, C, ii) ýO ý01l11(1 Yl VYVV h1'/Pf N1 NNNNý-1 rl rI r1000Oý, O dNOrib tNý1 . -1
N ri r-1 r-1 rl vl rl ,4 .iH rl rl rl . -1 rl vl v-4 r4 ,i r4 .4 rl r-1 rl ,1. H rl .4 rl ý-4 .4 rl rl 01 01 0% a% co fö wm 00
v
4°v 4ooöoin o4in in tn 4°44°4°4tn 4tn 444in in °in in Ytn 44444Y in wýNwwwwwNNwwwwwwwwwwww wwww www wwwwwww ww OnNWOON ry nO rv Orw in V co oNbr, 0N
C'tf'IrlP . iw in to
OQN P1 0,
N OUN1 a1 fm '1trvý1 r-I
ý (ýfýfýcD1 01 r1 r-IN co I OýN. ww wcory Minn UO
NNr,! Q
1ý101f1 . -1 r-1 0 O001ý1(1 N0 1 1(1 NO co ý0 V nCh
h to m0 co %a V mA cO O dN rl co . -1rl----. -1--. -1 O- On On a%t0 to co a0 1Z PPPw %o ýo ýo 10 w in to in in in ui VVvvV....
0000000000000ry000000000000000000000000000000 14 ý
rIN 1'I V 1! 1 ýO Iý aD Oß . -1 .
ý-I ri ,
N-1ý rýiýr^/rý-/rý-INNN NN
NN NNNI"1 /'11N'I
n f11ýI ('1MAL dVd
187
d N N to N
d N N to N
0 .ýx
ro
0 c; in 0 Co
1
ýö w
4-2 o öýö
E 00 NN
I UN
O Ö4Ö
N yý N
a
bbbbbbbbbbbbbbbbbrrnn nn W r- rrrrnnn 4W444°4444444°4W44444444WWWWW4
. NN o wwwwwww ýI wwW w%w0wo Ww rw wwwwwwwNwww wW iý v in 0 m. 1NNCWr0c0umrnmmch 0, rnmmm o .10AA s ovmnrv0U)4obýonýowmnmmnnýýýNObrýý ........................
n W N I .
o el e m
....... in 1f1 vvV 1'1 M1 n P1 NNN r-I r4 r1 ºI r-1 01 aD 1D If1 1fl 1f1 OVd 1'1 1'I NN rl ý4 r-1 b
1' %0 4444w Y444WWW44WWWW444444444ý44°4
o www w wwwwwww w wwwww ww co mWN0 0U)ýýýanrvnýroýýv
ýý'o co, - o ö rvwnoaoýeoorvrrvmnrvvmv mmmvvvmmvrmrnm Nnn
n nvrvoco rin AAAvMrvi in io rrv-örrvrvrvnýHomvmm rv 14 NNN ri M rl rl MM 0% OD %0 NO In V PI rf ri N rl rl MM rlºi r-l rl ri wV NN .4
nnnnnnnnnMUN nmmmmmmmm mI CO mm Y4w r-r-44444gCo4
o mN Nww w wwwwýwww6www pý VVO1 Yl Z1 NONNV et Vd... ö vooonmm%ov. +m. irnvmnNnmmrnmvvvesmmrnmmm ri "! n in mgnO., nN 1D Ir: 1ý 1! 11N n O: CEO., GINN NN NCI VdVvV PI
v o wm tO V0
co vº ýn N
OI co it %0 V1 10 V PA PI Y/ NN .4 .4 r1 Co Co %* 1f1 NNNNNNNN .1 .4 .1nnn N
w I
ýrnmCo Co Co mmmCo Co Co Co rnrno 4444444444444444° 40
o wýwwýwwwwwýýwww w W f4 mvVVNO000I111111I1II1I111 ri 0 O
%0 m r- by gn '044 a
NO U) Co Co %0
. ýi mU vvv
N-e 0
A14 rI Of Co m1-in V vmAA r11-m0 PI
00o00o000000000o0000000000000000o V to %D rC 0l o fV md w0 r co to o rIN V N%OrM at O r-I N P1 V to rvvvvvv Wmr
LO M in nN0" wl LO to to w %o %0 %ýo %o -o %o %o %o wrnnrrn ;N
VVdd of N1 U1 in U) in U1 N Ui U1 I[1 in U) in in U1 U1 U) U) U) U) in in U) U) U1 U) H1 U) Ul U) V1 in 'O W0 b %o %o "D
4444444444444440444444444444444444444444444 o ýwwwwwwcýwwwwwwwwwwwwwwwýaýawcýwwwwwwww c3 Fir
"0 r, , INb0
Vf Ucc) IN w'O dUl CO 1(l -4-4
m. ý-ImN 0 ZU OýNýýDN OW10ý01
'ONrIac0 'r
dv NA 0N r, 2
wi5 W YI N vAco in týl rl co N 1(11'1 e-1O 01N ýO Y) V 1ý1N r1O COON01 NYl .... ".. ""........ "... ". ""..... ".. HAAAA 0% Co Co rý rý %0 %0 M M1 vav r) rim rm NNN 94 NNH of of mHHHm rI rl 0% O% Co 1- h %0
I In If1 in In to In In In In In In In U1 to In to In to In IA In In U1 U1 in in In W %o (O %o to %o to to (0 to %D (0 %D %D (O %o
YYY44444444Y4Y4YYYYY4444YY44444YYY444YY4Y4Y O WWWý1WWklWWWWWWWWWWWWWWWWWWWWWFýWWOýWWWWWWWWWWWW
OÖNN oNnrý^Ör 1i1 rr C4
Ö01n bY. V Prl
-W IN r-I
ÖÖIdýI tO ýÖ
1ý Nn
N C4A
0 C4 R at in F4
O It ,4 co NM co
fD Xi to
N1O It W! . It
N IN(O In In dddmfnri(A NNNN Nr 4 r4 rl r4 r'I r1.4.4.4ß4 at Co WNN %D %D In In lA ddm P1 PI Pf
in to IA IA %n in in U1 u1 IA IA in IA in In IA in b %o %o % ID %0 %0 %0 b %0 b %0 %0 ID %0 %D %0 %0 %0 ID ID b %0 %0 '0 %0 4444444444444444444440444444444444444444444
o uýwwwwwwwwwwwwwwwwwwwwwýywwwNaIwwcýrý. wwwwwwwww! ww OA
b1l1 ý0
Ö ýMýOý W 01 ri rýiN
n Ar-1
v Nfý1f1nÖP ýO
Üld Omi
NN ýr r Oý
ýn
. 'i N. /ul NO1 ýO P1 rl 01 c0 ýD 1l1 N. -1 OOr! v! 9N lO rl lO Na0 Yl rl9 NO9 W! NO a0 d PlNO VV /'/PINN NNH AH AH AH HA mw rr rß%0%0 M UI VVV mri mm NN NN rg A rIrl AA, 1
., ,ý
0 h
w
in kn in in tn in 4in 44Yt04Y44444WW°44444WWYY4444414 44444 wwwwwwww wwwww wýýwwwwwwwwwww wwýwwýwwwww 011(1 N NlD
Nd co Nin N0 %0 OI(1*0
NONV Co d0 00 01 d
CO d ý0
N(001 bÜ (0*0
VN 1ON1'11 ONO
N 1'f hOýd NCO NLD c% CO
O d01 dd dr-1 d 1'fN *00
fV N
NCO0 led -! rnýodN. roýrnoNýoomOdo
NN 44 HH r/ rI ri H 0% wP %D %0 in in ddm P1 ri NN 44 .4 rf ýf ºf m -1 'f 0% Co 1- rý %0 U1 In mm r) 94 44
0000000000000N01ý10000000oaccoooooooooooooo0000 ri N 1'1 d 111 %O N OD 01 ý. 4
:4 r-1 r-1 .
r-1 rr-1rß-1 rml NNN4NM
mo., " '/ 1'1 I'1 fß'1 1ý1 A /'1 tn'1 fwl 1m'1 Vd
oýo"
eeýýýNNNNNNnnnmmmmmmmmmmrnma0000000 u-t OO4Od °° o00opo oO °Op1.40 0 111ººººº411111º111ºººººº++
owwwwwwwwwwwwwwwwwwwwwwwwwNN emN ýouivrn rvmv mnvm rnuiýo rnrv rvrvn ot1mN
Co . 40 rl ý-1 o 0% d nb Uf CN o Co nN Ul V on 0N0%1'In n Co Co mm 0 '0 0 01
.e dN 4rºOýn
ul V N1 N r1 .im it In Y% NN r4 A vI CO %O PI N . 40% 0% wVVV0N
CO 00 43% Ch Ch (n eeerööönöw 00C? 0% ö
OWWWWWWWWWWWW fý WWW tý. WWWWWWWWWNN co m to Olnm 1'1 rnN UIO mmv at r11ý mam r1111111m In r mZO mýrr%O V 1(1nlfl l(1 X m-1N V %D 1(1 al r-1n mOýO% 01o nö 01 NONE-11l1n In ID a. V. 1NO VN Nm soNN0Cl P1 Of 1/1 ý-4 Cl P1n0 11 N IV x4 01 WY Cl N . -4,4 .4 Co l0 V Cl C4, -# W 50 U1 V PI N, 4 DIN NNN0 . i/
4Y4444t, r-44400 4ý4ý44ö0ri0 Hýr4Hý owwwwwwwwwwwwwwwwwwwN -W
om rv m 0% '0 rv min v%oNMno14, NNo IIIIIIII1o N Ul rl co O O'. to N %0 V 1N 0 hfl r% NM %0 Pl Pl. 40% co Co (it OA OA pý 0 ri OO
Nr1 ýO %O rIN %O . -$ -I tO IN W NN NN 9r! _4 010'. Iflr . $r19 O4
ri r-1 %0 VrNýf .4 CO Ill VN ri rl CO If) M Cl rl . 110 IA IN r4 º1 0 tO
NNrnnmCo ooCo Ch UN Ch mmooooooo14 oe
w0w00 fýIH
0 4wwww wwwwwwwwýwwWýww PrvaO
' uino+minO %0 AIAN^no 1111111111110N .1 In v. r%oin NInuiCh ri nrvnr. incommm9 0%Co0 m
r! r! W! lN Vl r'! 9 ! l0 lD r! 1ý N PI 01 rl f: lý r: .
r101 O li Pbm 44 r10. %0 Y1 «4A Oh U11 mA Co V. iA H rI 4 U10 'i
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO WS rl N Pl V UI w1ý omO . 1ý-I(y Pl v "1o 1, c0 0ONNNNNNNN0OmN P) v Ul v jJ ý r-Ir/ rl rl rl r-1 .l rl r-INNNNNNNNNNAA/'I PI M11'1
-l NN". 47
188
ä in
n v a
H O
A
v N N CO N
i In N
,ý
w 0
ii
d N N
N
äý NO P/ NO. r wo In in .1 N [V N
v
ä ý,
g v
ö 144 44
ü
O C;
in U) in in U) Ul in U) in ui to to in U) to
WWWWWWWr4wNwWWw ýnomrvmmn voono'o N to ri too' nr Oi U! ) In %On rico of Oi
N1oruivr OOQiWniOUne
N
W 0 In
n
W NN
rim P1 NNNNNN r4.4 4 ýI 1 . -1 Oi
o ö
n
Ul in in in IA %0 %D ID ID %a %o %OW %o IO 0 C, wwwww1wwwwwwW4w
rn rl Io Un 111 fl u) CO omrn n 0 10 -4 r ow EO '1 V CO n r. -IN Pl rvN oo Ins-Inn Nn o0 w ri
rf
w n O1 CO
in m r4 lV N
. 4. -l i4 i 4oiwNNtZWIIuiri .i
1'
bbbb t0'. 0 b'. O b ýD'. O b tO b ýO Pf ,
oW Ni co CO O'. 0 rn Nm Co Co vN r0Co O in U) N O'. U) tN Pl ri P1 O Oa 4 Nv 10 NN
44 %%D N to tD C P1 r4 0 U)ri ri r1 .$bbVN V
1'1 1'1 M1 P1 /'1 1'1 NNNNN rl rl rl rl 1'1
.. v v
inn"N r- t% Nnnn r- nnn r u Iwwýwwwwwww M-3m44Q44 4 4w +rN
to w N %0'0 N CC 0NNrl r) r %O W %0'0 dN e1
00N Cl N co co m co CC 0' "' 0' o
rl NN '00 U) 0101 C) PI PI Pf r4 r-1 . -4 .4V
In In VVINNNNNNA '41-011 In
oooooaooooo0000 GON a0. rvC-1 vatatratrno WWWwmmm 41% rn at at w at 0V..
{ NNT.
0oö4ýýY444°° °e, 4-e? °°ßq44°44°in in in 4in in in mtn44Ln 44 o wýwýwwwwwwýwwwwýýýwwwwwwýwwwwýwýwýw0(4Ul rNoNMNTmrvVNOoONoVoNNm ö oýýrnýrýnmmrnmonommýmmvomoný omomvryýmýmýýto ul 0 in VN c4 00ý mmrý b'7 '. ºeiN Nr10001l11ý1011(lm V ri n vl rý DN O1 hN N9
. ll
.".... ".................. ri NNN NN NN rI rI HAH rI 0 rI of H A- frl rlx m 0%m 00nnr%0%D%0 Ul v; -/ VdV ýf V Nein
I -f aT VVVVV In 1! 1 in lfl In in In in In m In I In In IA IA Ifl to InIn In In In In In in In In Ifl In to In In Ifl In In
4444044444444444444444444444444444444444444 o ýawwwwrý. wýawwrywwwwwwwwwwwwwwwwwrýwwwýýawwuýuýwwwwww kn N O CO Z Off bra., (n V ri [ON V. ý V. i r1 ý0
O InN
0. Ný0 Ln Co Co (9 Nn1ýýOnÖN. ' 000(0 V In
Ill fN 900 t, dN a0 e"10! 1'! 01 h x'101 t-1"10 fO O ýTNOx010111 A'i 919«! V NO W cO lr ý01l1 O
rivlm of ri rl r-10% 0% T Co 0P 1-%0%0%0 in kn e in vV 44 V rb P1 P1 NmmNNNNNA ArI A 44
to ÖNÖÖ In In In ÖInN in in in in Ö in in vi ÖÖ in in in in Ö in Ö in in in in to%o%o %o b b-0 *0 b*0 ö YýYýýýYYýýYYýýýaYýYýýýYYY4YYYY4YYYYYý
o wwwww Yw Yw YwwNwwwwwwýawwwwwrý.
wwwwwwwwwýarýwwwý1wwwwww N rl M-qNr/Öý
v r! 0: 1b
vý 001 orDb Ul N 19
. OO I:
b" W!
vfý1 V! .4
-W ww.. ... r! Q!
VV. W! "...... """""
It .. *O Ifl IA IA IA indd VdInPIIn Pf NNNNNN NN r1rlrl. 4.4. -1rl. 4.4.4rl01 N r. %D IA in in in 44
v
N In kn NN In In In in NN in NLn bbbb b% ID 10 ID ID b to b %0 %0 b%0 w ID to to to NNNNNN 444444444444444444044 4 4w 4w44 4 44w 4w
wNwwwww44 44 44 4
wwwwww4ww 4
Nwwwwwwww u)wNwwýwwwww NommbrmNoombm rv NmN rv vnmmNmNornmdNo N fý I(1 Oý NdV r1 NmNmdmNobrN ul oNNwWm
Co U1 dw rl OmlN: lO Oý 0 1)
... 4 0% m-NN
01 l FO O)
oNN rl d 1'l Oý l/l d lel -9W Ul lý Wm P1 l0 t: W! NnN mm ýD O r- dN
NNNN r-1 rl ri ri rl rl º1 rl "i rl 01 01 Co 1' t. %O %O I[1 Ifl Ifl d4 Pf A 1'1 1'1 NNNN r-1 rl , -1 01 Co PP 1D In
00000000000000000000000000000000000o0000000 Ww MInr4nýr-t-nnnmmcocomm
Mvvvvvvvvvvvvvvvavvvavvvvv vvv vvvv vvvvva Y°4Y4Y°4YY44Y444Y44444Y444Y4ýYY4Y4444Y o wwwwwýwwwwwwwwýýýwwww4 wPI wwwwwýwwwwwwwwww On If10N nV OIl1nnNO nNýD 1'/O Nn P1O
N ý1N
P1 000
-I 01
Yw
01 NNON0 O ON0IMýTONOaýr-1nýDa0ý011)b V nýOýcp. -1pýýprlýOnU11l1. wn
cý .
NO. . .N.
a0 w! P1H Oln 1! 1 NO CD wO af NO01n %0 ui
D -401 con bm
4P1N . -ý N"-1 OOn
02 .
n rO 16 -
"1010101 Co '0 CO Co nnnn ný0'010'0101[1 M Wb Ul In VVdVVVV P1 P1 in ri ri mm ei mm t4 NN
1' o wwwwwwwýwwwwýwýwwwwwwIwN wwwýwwwwww NVNOv rv m (n rmm rn m ommmm co 'D W) co at v CO ö vorrl .0 in 0hccn rýýrvomtmoýwmýeaöný0
COr ýWoý ..
mnm .N
WYo.. Wn. ..
bN.... VMNo.... nbv. v.. NN 19 o. omC. .^n.
N ....
VN ... . Ifl lfl wVVddVVM NI in mmmmmmNNNNNNNNNNNNNH r4'1 4 , -4 vl rl - '. 0 ,44.. 4
ee44e2g2ee22ýe222222ý-e4in4Q4tn444QS4in o ýýwýwwwwwwwwwwwwýwwwwwwwwwýwwwwwwwwwww MO N 1! 1 e a0 v vNMON «h 'O MM MO NM co M maONN c001l1 (O u1 N011O1 I)a0 0%NM %0O01
.1 inc%M %0 O M%0 O» M(O r-IM IfI I 01 NNV r1 1llY 0% NVM 1(1 d Nfý
N [O fý'O 1( .. 1
.v.. """. " MN rl
.O 01 OE
.. CD W P%0
., In
.. ".. " MM NN
"""n. 10
".. 00 ro1/1N O" ý lO V. 1 [ IfIN O. ý. bM
"... ". "... IV NNNNNNNN fV NmH rl HHAH ri HA rl H rf H rf of .i rf A 0% T 0% Co Co 0mr. PP %0 %0 %0
ý+
dddddddddddd tn in in NN to tn N in in in N tn in NNN vi -N in tn NNNNNI
4 44ý°444°4°°g4ýýýý4444ýP44444Y44Y4ý444°44 wwwwýýwwwwwwwwwwwwwwwwwwwwwwwww wwwww ww MNoOOd rv mrNMNomNNmooNonmdNoNN fý O%NN u10ýr1 v11ý N ADO Out Oo odm001 at NUl dN at Ary dd P1d 01. -1 U1NOU1 dN N 'D "i NN ýO Ill 1! 1 d1'//'1 NN . -1OOýOO h. -1m V Ofý I 1O drlm ýDdNOOýP ýO d Nrl Oý m1 h
ý4 -4 1.4 H ý4 4 14 ß"l /1 H4HH r4 0f 01 mmnn rý %0 %0 %a wo to In 0 1* -W ddVA 1'1 AM P1 PINN NN
57 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO (Ij
A . -4 N1'1 V 40 %O t, x001 ýOf 'I
N14 ri
N . Oi 14 ý
". i (V f4 NN
((4 fV NNN N P1 PI Pf POI Pý1 r, PI/r9 mm V
d1-4 d
189
>O N O0 ýO
>0 N 01 ýO A
0 .4
r
O" N O f1 1! / 0 Co
. -1 Co
..
M
CYyG.. ý
-rl N +i O 44 Ö
0
o. 4 NN ýO
r-1
00 iiö
%0 e%0 %D ee%0 %0 %D %0 %D %0 %0 %D eew en rNNNNNNNNNNNNNN im 4440444444444444444444444444 +a o ww4ww
4ww 44wwwwwwwýwwýýwwwwýwwwwwwww wNr ýnNnýý rýoým g No dumdd m O d0 0 1n'f dri Olw OP1 Of00h "-t O (4 %0 r-IN r10101m1°/rvo. -
ýy1 Onrn
nNn aD bP1NN (IN V Co
d rl wdO mýO dP10ý coNbbl YýNN r10U)r Odrrr ý1P, O a ow [O G1 d 1'1 1'1 NI NNNN rl N r1 rl rl rl ri ý'i r-1 Oý a0 aD r ýO ýO 1f1 hh 1f1 ddd1. //ý1 d
II %0 %D %0 %D e%0 rn 4 rrrnnrrnnnrnnnnnnnnnn nn n 44444444444404444444444444444ý44 4
o wwwwwwwwwwwwwwryMwww w wWwwWwwwýWW wfe 40 Co O Of
b113 b' Co
ÖUffN001 b`r r4 0 aD
b%0 v mle 14 rl
00 Uii ei) CO Co to ./ä N 0%r Inm NºIOOýýOnNr NW na0 rl U)In r1WN oA0001 lD aD If1M P1 P9 'O
ri A vI rf H rI A 0b Co rn %0 %D to in rV PI Pf P1 NNNN 44 44 'f "-I rf r4 of rf re 01
NNNNNNNNNNNNNNNNNNr- Co mCo mCo mCo Co Co Co oCo Co Co °Y444444V44ýý4444444444ýV4ý44444 gn o www wýwýýwýwwwNwww ýwwwwwww ýýýý 1! 1 co Co ý'1 1'1 lD ry O fý N 01 01 OD OONN
1'1 OO ýO r 04N
'OO 1(1 VONV mNp1 b Oýý0001ýfý V CO
NCO Co CO NNrO rl 01N if)r-11ýY0%0 V N0 DI011ýbv
'n 'n i Nrl rlh YOB 0ý01AAaO m nlýlýfý 1ý f-1
w %0 in ve API PI 44 NN Nrl rf A rf AAA 010 rh %0%0%0%0 in in in evve 11
.,
w
t- mco mco mco co mmmmmco co mCO mmco mrnaamao 444444444444444444444444444
o wýwwwwwwwwýýýwwwwwwwwwwww wö -W
ýýový ýnuiýnvvvrmmmmm 0111111W ýn moonvnvm mrvuº rnmrn'D vU%U ,o ýLn 'a nrnrvmýcºomn. +mýoýý. inwýnomrnmmmnö
Hm r-%0 %D In In U1 OI MMNNNNNMM 14 rl P 11 fýlýfýYlO ry
n 00 V 000000000000000000000000Q00000 In%D r- Co Ch Q 14 N rf V W) ýo t, CO MO1.4 N P1 N ,f %0 rýCo 0%O of NmY to W
ar vvvvsvva n in Iq w a0a
in
in n 1010101 10 %010%01010 e, nnnrn ý'
NM
dddddd in in U1 U1 1A tn V1 in Ul in tn tn U1 IA u)Ul m U1 U1 U1 U1 U1 tn in U1 U1%o%o%o b NO%0"O %0 bb 4444444444444°44444444444444444444444444444
o wýawwwwwwwwwNWID Nawwwwwwwuýýawrý. wwcýwwwwwwwwwwwww ao ao Omr to 4r rotýýo d NN"O N mc", u%
O 0% w
OOn o1'10N ao Oý Copý oO d%0 Ln Co U)%0 w OHO ýO 1'1 O OýO d O, O Oh O N011ý1 in Oý 1[ý er dr-1 Pl Oý. -1O a0 Oa0"O". -1 d OX IA PI r1or P9 1(l lýl99 4n rlw ll1m oN ýO dri0 Co "O in dNN ei Ofr O P)oN d o, u rr d
"-1 rf H rl ri H ob CO P%01D to Yl dd PI PI P1M NNNNNrl - Hrf - rf r1x c%cn co co P%o in in in Vd
..
..
.. ý w
In in In U1 in to In in in IA U1 N1 in In In in In Ul U1 IA IA in U1 '. 0 %o wWb %o 10 %o %o to WO '. O WO %o b 10 %o %o bb
4ww4444444444444444444444444444444444444444 o wwwwwwNcý. rýwyýcý" wwwwýawwwwýwýawwrýwcý" wwwwwýawwwwwrr. wwww Ulk %0 0N m1IA ÖýO
N Na0
VO NU01NOmNW V
ri NAON Co 1
PlNrN-1 W %0 -e Pl ^WY`tl Co 0 Q1f..
r'1O1Nb iAr, O
01 01 co %o %o bn VV ri rb mNNNNH Al f rf r4 H rl H c% co co P %o %o m'A wVVmmm f4 NNN 44 N
Y4in4kn4tn44444tn4°°4444to 4444444444to °°4444ýQY o wwwwo wýwwwwwýwýwýwwwwwwwwwwwwwýwwýwwýwww öronrmSN b"'
m° in vo cio nö
in "0 mv ö'D 'o b 0' mb
rn co co nömmmÄä rl 1"I rI PI NCO W, 1ýv1 Y1 OONb OWN W 1'1 CD A01ý Pf 01ýbm r1 mco ýD Yl mC400f NNh
In in VmmNN Nei rl rl rl ri 40%01 CO r- 1D 10 M In V 10 MPI MNNNN. 1.11.44.1.1 0101 mP
nnnnnnnnen%0 %0 en%0 e%0 nn%0 ee%0 %0 %0 n%0 n0 N -NNNNNNnnNnn YY44YYYYY4Y44Y4YY4444Y44YYY4Y4Y4Y4444444Y44 o wwwwwwwwwýwýw w wwwwwýwwwwwwýwwwýýwwwwwwww ry mmmmvrn n nAOýnrn w o ný vCo ýM1ý rnmm o mmmvýow 1n
N pp V
NfýN 0 On ýO ri r$nn Nrl n O40O
N v1
NIlO. $UO .
Nn V oD PI NN CO . NN NOYOPdN l0 V "le a.
. . OHO P.
. ý0 d r1 0%
.AN r1 0N 01 OD Nn Oý A ýO OHO NO ýO A
P1 mNN r1 .$AA OI Co IN %0 nnVVm PI N 44 94 N ri r1 ºI t. 4 .1 .4 0) 1N %0 ID nVV PI M 44 44 N r4 r4
WS0 r1 NM ýf II1 b
fý t0 01 0 0r-ý 000 M1fý WH
N oD 0Ö . '"i
NMA "No bN
WD Ö
101 ((V () d
lU) bN 00
WM04N W rl rl rl rlý-I WH
rl ý r-I NNNN NNNNN N I'1 MMMMPIMM MM vv -
%0 %0 e%0 %0 NNNNNNNNW W0mmCo aarn rnrnrna00000 an 44444 4 .. 144414 4 ° °4444°44°44°444° Z, 423 ýý o wýýwwwýwwwwwN wwwwwwwwwww ww wnn O Co I(1 O 1'/ Oý N P1 1f1 1f1 v Oý nNO 1I1 Oý 01 Oý In In OO 0% O% Oh ch O11N Co NVONV r1In Ul s If1 VInN 01V OýryN NN .
NO 0
YI o O tO vl d 10 rI N OON VII InI le VýO', ON V ýcln e'ýrl V Oý1"1 rl rlnNNNN NOO ý-1 V 94 rf m Co %0 V mNHH mCo U1 V PIN AH rt Co Y1 Pf 44 44MA H-I rVr0N
I ,4144 -4 -7
co o ýwwNOww Npý1 wN Nwýýwwýwwýw wwwwwýýwý wN0 lDd N, 04 V vv vV ma. Ul O11N d wN dOOý P1N'0 N01 NUItN N1y toto N1ON NrI OýDNN IývvvN ryN ýy O %a Yl Iß-1 NN 1O O O. 1! 1O rl O 1ý 01 N. -1 VNO, r1 a00% PI NO I1 P1 NNN vlº1, -1 r1 O 0% 94 NH CO In V NNH r10%%PV MNHH Co %D MNN V4 r1%0%0 V V-0 NNN NO
.ý
v ý
ý0000000? 000900 444444g4qCo 44 ý
e20 4 o ýýwwwwwwwwwýwýwwwwýýýwýwýw nö aD OWN Nmrl
N Ch Co
I1n c00 O.
NN ul NO11 hAa01'11ý01'/f'I N1'1 aD V 1/1 ýO ýO tO0 Y1 W 0h N CO[Oryry fyN fy ýOb O 01 in N rl r'/ r/ fý O2 ý'1 V lO PNN Yl ýO N fý I(1 MVO,, . -1 rý r rº OýO OOOO PI A r1 %0 V 44 H x1 01 10 V ri f4 .i 0% %D Vm CY H of Co %0 mNNN 44 44 N rI 1.4 ºf ºf 0P
n NNnmmooomrnmm 000000000000 ýo e, 444444444444 1ý4 oC wWNwýWWWW.. wwýw wwwwwwwwww(ý. ý wr ry O ýO 1W ýO N
CNN NNNNNNN fV fV OIIII$
0% in N
r1 Co rl r-1N ýfN1"10010 ýO Oß1 V 01 «, P1 rl NPlN Co CX) OýýO ýO e ýO 0w ýO ýO r-1 riO
44 N/'11'IONPºI r1 ONIA V 1'1 V Irr-11f100000000 P1 P10 N rý V PI NriP Mm NH hin N1 AAh 4m-l r4 r'l r0 - . 4r4 .4 U1 Y1 0A
0000000000000000000000000o0000000000 WSNNVbno ff (y Nvbno rv N It hno rv NV HH ri ri , -1 H rl ri rl NN N NN NNNN Nr flm frm
ý' NN
190
d N On
1'1
9 N
N
PI
ä In N
0
ä N
rO P1
E go
o "n In in C% rý
41
III
O O 0
0 in
0Ln to 0
in in in to 0in
in in 00
Ul U1OIll OOOOOOOOOOOOOO WWWWWWW W1
IW WWWWWW OO VvaT VC aC DOýWON. t 010. NNN N1fl U) . N1n N(O N 01 (OU)ýnvvý. a. ýrnmmwt-. u
..............
N O {ý
(O .r
.
1fý in
"O VV
NN N04 N NN N. -1 . -I rl rl r-I r-1 rl h
If Ul In in %D %0 %0b %0%0%D %0%D %OWOw r OOO O4O O4O4OOO4O
.... ....... 0
+ 0 wwwwwwwwwwwwwww w In ö 'n C. Q r, 'o v, nn tn n rt rmn ui NN. -frn%orlnrl00r1nv
r to vr ui Oo0nnsýoýo. ýmm
-i. IH0%0%oiCo Co Co nnnNIOtO . -i
'0 to tO tO V to to %D %D %D tD to %0 to to n oo°Oo0044o4Ooo4 ° nýawom ö
L L NO + v v k:
Yl n o t) ri ao N CO Nmrr CO r m(0 rnn 0l tO r . -1
IA V '? N n N00 0t tO -I0000%%0 IA
..... ".. "" " IA IA
. . . Pf PI PI mPf 44 NNN NN Nrf H rf N
.ý ä .ý ,ý v
nrnnnnnrrnrrrnn 44°°ý44°°4444°4 wwwwwwýwwwwýýýý %a %a wka e%0 %D %D w%0 N4 co U) in U) U) V NOl0)O)(D (0 NN
.+ nwoooo'orvýnr)0, mrir
ri ý w in ui nV
rn w n in
............... CD nnnrn%0%oto &o tnvvmen . m
000000000000000 brwmoNo as nmmoN Ww OD fD f0 01 C'. G1O 01 01 01 01 C1 G1 C1 OM }1
NN
Vr V .fVV V' dVVV. V. VVV in u in in in In Ul in in in in in in in to in in in in in in in in in in in in in ýýýý ýýýý 4404444444444444444444444444444
o wwwww4. wrWý. 'Wwwwwwwwwwwwýauýwwwwwrý. wwýawwwwwwwwwww o in
f'1[ý v mNN in
O UI ýO WOO 01 m1 0O cr1 co21'n r'P1O UI, no
O, IO !1N 1n u1 m ul OO in pf N 'O .-
I41
in O ýD wn U1 P1 "'D 'O 'O I" wOOVt, co co "T m rl 4 0,0 C0 Fý1ýý0 k0 in Ov P1N AH 00ýO OIýN10 m in 1'100! 1(1 . -I 0%h d t4 Ohm 1ýýO ýf V r/O m1ý1ý
. -i .4 rl .i# 4- f. 4 ,4 , 4i 4HH ri v4 m at mmn l- %o %a bw Vº in in -* svv rf mmm r% mm r'1
NNN
in tnin ui to tn to in to to to Ill in U) in to in to tn in ui in tn U) U1 in to tn to in tnN to in in tn to to U1 U1 IA in U1
0000000004000 °°°°°° °4°°4444°44°444444444 ............ 44...... 4. II o wwwwwwnýwwwwwwwwwwrýwwwwwwwwwcýwwwwwwwwuýwýww nm mwW, %o
mo nmr,
w0"0m 01 mN CO .Tm lfl N ýO ýO f'1 K
111 co o nO
PIa OnnV V' V In
v ý0 ry oV'n
ý11f11(1WNW V Oý1'1Ylo t1 UlmoNoUln N. -1nNvW vrl v .. i
w nn. ..
v+m. voninNWn. It .. vmN19ll ml ýnrn.
. mr: 1.11. U!
.. . V!
........................ O1 m co mmnnnýw %0 w1 91 hdVYV /'f 1'1 P1 1'11'1 NNN 44 NN rl . -1 rl ri rl r-1 rl . -1 r-1 rl r-1 rl 'i
v F
t[1 Ifl in in in ul tn W tn in V1 In in in In in tn U1 tn tn in In 1! 1 in In V1b %o to %o ID %o b %o %o ID b ID%o%o%o to %o OOOOOOOOOO 00000z00000
.. Aý 4ýýýZ4ýý4ýýý4ý, w
`? w ww
ZQYZZ43? ZZ?? 4 0 wwwwrýwwwwwwwwwwwwwýwwwwwwwýawwwwwww Nvrn N CoNNcn(h%000%01L)NU1OONIn +vi00v nuirnin IoýIno N rvömnLo rmýrnwnvnc°výmmýeu^inýN.
Nýöön ý : °meýnrnmo e 03u m YVm4.1 mmm PIN NN 44 NNN -4 ý4 H vl of of ii rf Ai r4 of Co Co nn %0 in Ul tn in VVVm P1 P1 f
In Ifl WI V1 u1 N U1 in tn 1/1%010%0%01010%0%0%0%0 %0 %0 %0 %0 %D %0 %0 %0 %0 b %0 %0 %0 tD b I0 %0 I0 b IO nnn 00000000OOOOOOOOOOOOOOOO
wwwwwwwwwwwwwt2 wwwuA 'AAwrýwwwwwwwwwwwwu4wwwwwwc.. N r`'nW
. C-4 44 . IoomOIW `n.
100 ri ný ono ýnWNI0Wýmmýrnrnv Oc0NONNN
ýI -e ri
I0 ID IA VV 1'1 rn-1 . -100! 9PrI ONID Nnvn
1 P1 r1 lý tDIn P1 NÖO. ' ID V r! r, '1 . -I
0 Il 0C., i
rl rl .4 rl rl AAA iº1 c% co nNn %o in In V Omr rI rI NNNNN e4 4 .4Am. i. 4. -l . 4.4.101 c%01
0ooooo000000000000000000000000000000000000o wvmmrmmoulNnvmmnmmoarvnvmmrmrnoNrvr-vt-mnmmooorvrvm Ww vvvvevvmmnmmnýmwammmmemmmmmrnnrnnrrrrmmmmmm
MMMMMN0vvvvvvvvvvvvvvVVvOvvVvVvVvvvVvVvvv
00000000000 0000000 o wwwwwwwwwwww{a] wwwwwwww(ý wwwwwwwwwwwwwwWw(. ý wwww
r'1Nm tO 1- U10011- "-W C4
P1c00ý0: -d a0 NP1 tD NN OOw0 to 0m wtýfý. -I U1NN(ý0ý Ö m"Nrrgwm mov0ý11v0 MmA., q ow o
11. " no w m", w,.., nvo
lä eM'1 0% -I . r1Ö
rl r-I ri r4 r-4 r4 Hvl O. at co co PPn%o%aw w into in VVVVMMMMM in NNNNNNNNN NN
I evvvvv. VIP 'TV-0 IF vv vv "v It v It V IF vvVvavvsevvN
44444444444444444gqq444444444444444444 4 o wwwwwwwwWý-1(i. ýpwwwwwww(ý(ý(ý ww1Vwwwwý[a1ý(f(aý1 wwfWýwwwýiaO] wwW1ý(iý-1w(ýwww O
ýNt01ýONmw 410,4 to fýN
Ifl [O a0 19 co r-
.O co n c, Oa 14 PI oO
co r. N
UI co
N ri O, co 1n tO ,, Yl v P1 (V N ". -1 .4OON
%o%o%o%o in in to in vVVV PI PIm mm en c-4 c4 NNN fV c% fV rl r1 e4 Hmm rl rl r4 x Hºi r4 r4'lM Oh
"v -q °44U00ýý4ý""o" QYqýý44Q44444Q44444444 o Nwwwwwwwwýr4wwwwwr4wwwWWWwwwwNwwwwwwýwwwwwýwW Go vrnnmrvvn nenn vow mtýoos nNrvmmm oýnorn rvn
N ný1(1 C/OcOO r1)ýNrýiO
OMiý bh
m eß'1 N cri ÖÖ
V wmw-rm/w0n0W Wp
440WD1O V
ý4 v w
NI e7 (4 #4 PI PI N44NNNN4 A-1 ri rl vl rl I-1 ºi .4 .4 ri 9
.4 01 WWPNN 'O 'O W 1[1 Ifl U1 If) U1 -I 'V
44-4,4öýv 00ý-v 44in in in 44444°in in in 44Yin tn 44°4°4444444 wýwwwwwwwwwwwwwwwwwwwwwýwwwýwwwwwýwww N Ol Ný'I P1001 dOddOýýO
w pi(1
w NI
wwOw PIU1
wd N1'1" 01' 00 OHO UI 1'1ýNOM NN W
ýy O ýO 1001 ýO 1(1 PI rl OO riry d ýO'-101 Pl %D Nrl 00 Nd 01 co If)ýO r1 Ol 01 N co 1! 11'1 NaD in IA ' r/$NdN
u1 ri d f'ýri c0 Nw Ifl d1ý1 Nr-1O N Plm Pl cp d O1ýnOý w m0 w in NI N OOýýD ýO d P1 OOý W 1ý
N NNr-1. -I rl rl 'i rI H rl H rl 0j O, tD% P %D %D %G In llld d ddMM MM MNNNNNNNNrI ý-4 rl
0o00000000o00000000000000000000000000000000 WS ý4 NP1 V Ifl W 91W Oh o. -1ry qvm w(Z w m6A Pl V 6w r woo 0v1Nn 0ýOr W m0"N
14H Cý4,4 rl, 4 NNNNNNNNN NP1 Pf PI fýI P/ Mi mC. ) (1 VVV