130 Pages
Gain access to the library to view online
Learn more

High efficiency purification of liquid scintillators for the solar neutrino experiment Borexino [Elektronische Ressource] / Ludwig Stefan Niedermeier

Gain access to the library to view online
Learn more
130 Pages


Published by
Published 01 January 2005
Reads 13
Language English
Document size 5 MB


Lehrstuhl E 15
Institut für Astro-Teilchenphysik
Fakultät für Physik
Technische Universität München
Prof. Dr. Franz von Feilitzsch

High Efficiency Purification of Liquid Scintillators
for the Solar Neutrino Experiment Borexino

Ludwig Stefan Niedermeier

Vollständiger Abdruck der von der Fakultät für Physik der Technischen Universität
München zur Erlangung des akademischen Grades eines

Doktors der Naturwissenschaften

genehmigten Dissertation.

Vorsitzender: Univ.-Prof. Dr. W. Weise

Prüfer der Dissertation: 1. Univ.-Prof. Dr. F. von Feilitzsch
2. Univ.-Prof. Dr. K. Schreckenback

Die Dissertation wurde am 22.12.2004 bei der Technischen Universität München
eingereicht und durch die Fakultät für Physik am 31.01.2005 angenommen.

High Efficiency Purification of Liquid Scintillators
for the Solar Neutrino Experiment Borexino


Ludwig Stefan Niedermeier

Introduction 1

1. Solar Neutrinos 2
1.1. The Standard Solar Model 2
1.2. Neutrino Oscillations 5
1.3. Solar Experiments 9
1.3.1. Radiochemical e 10
1.3.2. Water Cherenkov Detectors 11
1.3.3. Scintillation 13
1.4. Answers and Open Questions 15
1.5. Future Experiments 16

2. The Experiment Borexino and its Prototype CTF 17
2.1. Motivation and Future Possibilities 17
2.2. Detector Equipment 22
2.2.1. Liquid Handling
2.2.2. The Borexino Detector 23
2.3. Background 24
2.3.1. Radioactive Metal Isotopes and Radon 25
2.3.2. Further 27
2.3.3. Muon Induced Radioactivity
2.4. The Counting Test Facility 28
2.4.1. CTF Detector 29
2.4.2. Test Campaign CTF1
2.4.3. Test Campaign CTF2 30
2.4.4. CampCTF3
2.4.5. Current Status 32

3. The Module–0 34
3.1. System Description 34
3.1.1. Components 36
3.1.2. Safety Issues 39
3.1.3. Surroundings 41
3.1.4. Construction and Development 42
3.2. Cleaning of the System 44
3.3. Radon Emanation Measurements 45
3.3.1. Motivation 45
3.3.2. Methods
3.3.3. Results 47
3.3.4. Conclusions 49
2203.3.5. Rn Emanation 49
3.4. Particulate Counting 50
3.4.1. Motivation 50
3.4.2. Method 50
3.4.3. Results 51

4. Silica Gel Chromatography 54
4.1. Theoretical Considerations 54
4.1.1. Static Model 54
4.1.2. Dynamic 56 4.1.3. Loop Purification Mode 57
4.1.4. Batch 58
4.1.5. Comparison 59
4.2. Laboratory Experiments 61
4.2.1. Liquid Scintillator Loading 61
4.2.2. Activity Detection and Analysis Methods 62
4.2.3. Free Enthalpy Determination 71
4.2.4. Laboratory Batch Purification Test 73
4.2.5. Purification Efficiency Limitation 76
4.3. Experiments with CTF3 79
4.3.1. CTF3 Loop Purification Test 79
4.3.2. Batch 86
4.4. Conclusion 88

5. Adsorption on Surfaces 90
5.1. Motivation 90
5.2. Experiments
5.2.1. Adsorption on Teflon and Glass (combined) 90
5.2.2. Adsorption on Stainless Steel 91
5.2.3. Glass 93

6. Water Extraction 94
6.1. Realisation Possibilities 94
6.2. The Borexino Water Extraction Plant 94
6.3. Experiments with CTF3 96
6.3.1. Test Set-up 96
6.3.2. First Test Phase 97
6.3.3. Second 99

Conclusion 102

Appendices i
A. Piping & Instrumentation Drawing of Module −0 and the CTF Inner Vessel i
B. Scintillator Solvents iii
C. Operation Procedures of Module −0 iv
C.I. Operation of Module −0 Safety Valves iv
C.II. Transferring of Liquids iv
C.III. Mixing of Scintillator v
C.IV. Volumetric Loading of the CTF Inner Vessel vi
C.V. Volumetric Unloading of the CTF Inner Vessel vii
C.VI. Silica Gel Chromatography viii
C.VII. Nitrogen Extraction ix
C.VIII. Water Extraction xi
C.IX. Liquid Supply to the Skid on the CTF Tank xi
D. Recommended Alarm Set Points in Module −0 xii

Tables & Diagrams



7The liquid scintillation experiment Borexino aims at a determination of the solar Be neutrino
rate as well as at the observation of anti-neutrinos from supernovae, solar and geological
sources. A low background rate is crucial for the success of such low counting rate
The Counting Test Facility, a prototype of this experiment, has been constructed for
background examinations, mainly concerning the radiopurity of the liquid scintillator. The
examination of scintillator purification methods constitutes the main goal of this thesis and
aims at reaching the required activity level.
In this context, a liquid handling and purification system for both the CTF and the
Borexino detector had to be installed at the experimental site, the Laboratori Nazionali del
Gran Sasso in Italy. The functionality and radiopurity of this system had to be tested in order
to further carry out various scintillator purification tests with the CTF detector. These tests,
concerning a silica gel chromatography column and a water extraction plant, are illustrated in
this thesis, together with interpretations of the observed effects. Based on the results of this
so-called CTF3 testing campaign, well determined after a data taking period of two years, the
status of the Borexino experiment can be evaluated.
The potential of the purification systems installed on-site – this thesis focuses on the
silica gel chromatography plant – can be estimated by the development of a theoretic model,
aiming at applicable purification methods. Small-scale laboratory tests have helped to
experimentally confirm the assumptions and to determine important parameters of the model.
On this basis, a simulation of the silica gel chromatography process supposed to be applied on
the Borexino scintillator can clarify its potential with respect to necessary extensions of the

11. Solar Neutrinos

Looking at all neutrinos that are destined to cross the earth on their way through the infinity of
space, the sun is by far responsible for the biggest part of them. Moreover, the solar neutrino
flux remains reliably constant and its direction is well known; so solar neutrinos offer us an
excellent possibility to study the characteristics and behaviour of the neutrino itself. As
neutrinos interact with matter at ultra low
rates, of course it is not easy to detect them.
But once done that, neutrino spectroscopy
can serve as a unique method to look into
regions, where no other kind of radiation is
able to escape from in an undisturbed way.
Like this, we hope to get further
information about supernovae, far galaxies,
but also on the composition of the earth.
Diagram 1.1 shows the neutrino flux of
many different sources and their energy
ranges. The production of atmospheric
neutrinos and reactor neutrinos is – like the
solar neutrino flux – quite well understood
and additionally helps to understand the
nature of the neutrino.

1.1. The Standard Solar Model

The study of neutrino properties on the sun
as neutrino source requires a sufficient
knowledge of the solar structure itself.
Fortunately, this is the case. The sun
consists of concentrically distributed gas,
mostly hydrogen, held together by
gravitation. On the other hand, the
thermodynamic gas pressure prevents it
from collapsing. Now, we know that the
sun loses energy by radiation, which
normally would cause the gas to cool down
and to contract more and more in order to
maintain the pressure. In this case, the sun
would never have been able to assume a
stable state over the time period it has been
existing for. But as with further contraction
the gas temperature inside the sun
increases, from a certain point on the
thermal energy of the hydrogen nuclei is
able to override the Coulomb barrier
between two protons, so that nuclear
Diagram 1.1: Neutrino flux on the earth reactions become possible. They are able to
from different sources [Kos92] provide a lot of heat, to compensate the
2radiation losses of a star over a very long time period and thus to stabilise the sun in its
current state. There, hydrogen actually gets fused to helium. The solar neutrinos arise from
these fusion reactions, illustrated in Diagram 1.2.

Diagram 1.2: The pp-cycle reactions of the helium fusion in the sun, together with the
1 + - 4 2+emitted neutrinos. The sum reaction is 4 H + 2e → He + 2ν + 26,73 MeV. e

In particular, a quantitative neutrino flux determination requires the knowledge of the sun
7 temperature, up to ~1.5 10 K in the core, its density profile and the cross sections of the
nuclear reactions. A detailed description of the standard solar model can be found in [Bah89]
and [Bah01], where also the Table 1.3 has been taken from. The rate of the main reaction, the
pp-fusion, is very well settled by the value of the photonic, solar energy radiation. These
5photons travel 10 years to leave the dense matter of the sun, whereas neutrinos take about 10
minutes only; but normally the sun, as a light star, is assumed not to change significantly
5within 10 years, so that both fluxes refer to equal solar states when arriving on earth. Not
mentioned in the above diagram is the so-called CNO fusion cycle, fusing helium out of
hydrogen via a chain of nuclear reactions involving C, N and O atoms. The rate of this cycle –
normally suppressed – gets especially enhanced at higher temperatures; and indeed, the
easiest way to induce changes on the solar model consists in a variation of the sun
temperature. As the probabilities of the different fusion reactions – generating neutrinos of
different energy – depend diversely on the temperature, their rates get diversely modified in
the case of a temperature change. That would indeed question the prerequisites of the solar
neutrino rates in a relevant way. A comparison of the standard solar model, in particular of its
density profile, to seismological measurements shows an excellent agreement with a
discrepancy smaller than 1% [Bah01].

3neutrino branch neutrino flux average energy maximal energy
-2 -1 Φ in cm s E in MeV E in MeV ν ν ν, max
10 pp (5.95±0.06)10 0.2668 0.423
7 9 Be (4.77±0.48)10 0.863 (0.386) 0.863 (0.386)
8 +1.01 6 B (5.05 )10 6.735±0.036 ≈15 -0.81
8 pep (1.40±0.02)10 1.445 1.445
3 hep 9.3·10 9.628 18.778
13 +1.15 8 N (5.48 )10 0.7063 1.1982±0.0003 -0.93
15 +1.20 8 O (4.80 )10 0.9964 1.7317±0.0005 -0.91
17 6 F (5.63±1.41)10 0.9977 1.7364±0.0003

Table 1.3: Energy and flux of solar neutrinos according to [Bah01]. The first five lines
belong to the pp-cycle, shown in Diagram 1.2. The last three lines arise from the CNO cycle,
where C, N and O atoms catalyse the fusion from H to He. The lines with equal average and
7maximal energies are mono-energetic, Be causes two of such lines.

Diagram 1.4: Solar neutrino spectrum and according uncertainties, compared to the
thresholds of some neutrino experiments [Bah04]

The reaction cross sections generally arise from the strong interaction theory and have
partially been confirmed by experiments, like the LUNA experiment at the Gran Sasso
3 3 4laboratory measures the cross section of the reaction He( He,2p) He in the relevant energy
Of course, some uncertainties remain also here; but the standard solar model and the
nuclear reaction cross sections, sufficiently known to be considered as presumptions, have
generally been used to examine the behaviour of the neutrino. The according solar neutrino
experiments are described in the following paragraphs.