41CA measurement with low energy accelerator mass spectrometry (leams) at the Centro Nacional de Aceleradores.

UNIVERSIDAD DE SEVILLA
DEPARTAMENTO DE F´
ISICA APLICADA I
41Ca measu emen wi h Low Ene gy
Accele a o Mass Spec ome y
(LEAMS) a he Cen o Nacional de
Acele ado es
A hesis submi ed o he deg ee o
DOCTOR IN PHYSICAL SCIENCES AND TECHOLOGIES
PRESENTED BY
CARLOS VIVO VILCHES
SUPERVISORS
D . Jos´
e Ma
´
ıa L´
opez Gu i´
e ez
D . Ch is o Vockenhube
2018
Jos´
e Ma ´ıa L´
opez Gu i´
e ez, P o eso Ti ula del Depa amen o de F
´
ısica Aplicada I de la
Uni e sidad de Se illa;
y
Ch is o Vockenhube , Cien
´
ıfico Ti ula del Labo a o io de F
´
ısica de Haces de Iones de la
Escuela Poli ´
ecnica Fede al de Z´
u ich
CERTIFICAN: Que la p esen e memo ia ”41Ca measu emen wi h Low Ene gy Accele a o
Mass Spec ome y (LEAMS) a he Cen o Nacional de Acele ado es”ha sido ealiza bajo su
di ecci´
on conjun a en el Cen o Nacional de Acele ado es po Ca los Vi o Vilches pa a op a
al g ado de Doc o en Ciencias y Tecnolog
´
ıas F
´
ısicas.
Pa a que as
´
ı cons e, y en cumplimien o con la legislaci´
on igen e, fi man el p esen e docu-
men o.
Se illa, a X de junio de 2018
Jos´
e Ma
´
ıa L´
opez Gu i´
e ez Ch is o Vockenhube
A mi amilia, con odo mi amo

In his age o specializa ion men who ho oughly know one field a e o en
incompe en o discuss ano he . The g ea p oblems o he ela ions be ween
one and ano he aspec o human ac i i y ha e o his eason been discussed
less and less in public. When we look a he pas g ea deba es on hese subjec s
we eel jealous o hose imes, o we should ha e liked he exci emen o such
a gumen . The old p oblems, such as he ela ion o science and eligion, a e s ill
wi h us, and I belie e p esen as difficul dilemmas as e e , bu hey a e no o en
publicly discussed because o he limi a ions o specializa ion.
Richa d Feynman
Ag adecimien os
Danksagung
Ring aziamen i
¿Qu´
e queda po deci , adem´
as de alea iac a es ? Pues al a ag adece a oda la gen e que
me ha acompa˜
nado du an e es e doc o ado, que no es poca. Voy a in en a i dejando los
dis in os ag adecimien os en el o den m´
as es ´
anda , ya que en es o de los ag adecimien os
de esis ya hay cie a ju isp udencia. ¨
Ube diese Teile dass ich au Deu sch sch eibe, muss
ich sache en schuldigung, weil ich mein Deu sch e gessen habe, und ielleich e was is nich
¨
ollig ich ig. Aunque en el
´
ı ulo ambi´
en a˜
nado la palab a ing aziamen i, no me a e e ´
e
a esc ibi m´
as cosas en i aliano que alg´
un g azie suel o. Suficien e i aliano in en ado han
sopo ado los des ina a ios du an e es os a˜
nos.
Y como sigo un o den es ´
anda , empeza ´
e es os ag adicimien os po mis di ec o es. Ag ade-
ce a Jos´
e Ma
´
ıa su cons an e supe isi´
on, su disponibilidad y su buen ollo. Si alguien se
me ece el econocimien o p incipal po la di ecci´
on de es a esis y deci el d
´
ıa de ma˜
nana
”a ese chico le di ig
´
ı yo la esis”, es sin duda ´
el. Ich muss also mein co-Be eue , Ch is o ,
danken. Obwohl e noch nich mein Be eue w¨
a end meines 4-Mona e Besuch nach ETH
Zu ich wa , ha e wie mein Be eue mi mi benommen. Seine E ah ung in 41Ca AMS und
in de be ich igen de Kaliumin e e enz wa en no wendig ¨
u die Messunge in Se ille. Y debo
ambi´
en da las g acias a Manolo, quien me dio es a opo unidad y ha es ado siemp e a en o
a odos los p oblemas bu oc ´
a icos a pesa de es a siemp e has a a iba de abajo. Que
alguien con an o jaleo es ´
e siemp e que puede disponible es muy de ag adece , y es una
pena que al final no cons e como mi di ec o .
Tambi´
en engo que ag adece al es o del g upo de AMS del CNA, po que c eo que es di
´
ıcil
encon a un g upo con el mismo buen ollo, e imposible con mejo : a Ja i, po sus a des
de juegos y su sen ido del humo ; a Elena, po que su inigualable conocimien o de SARA ha
sido una ayuda cons an e, amable y a menudo eque ida du an e es os cua o a˜
nos; a Juan
I
CONTENTS
B.2 The expe imen a SARA ................................ 112
C FORTRAN p og ams o 41Ca measu emen analysis 115
C.1 Inpu file ........................................ 115
C.2 Ou pu files ....................................... 117
C.2.1 ou -K-co ec ion ................................ 117
C.2.2 ou - uns .................................... 118
C.2.3 ou -blank .................................... 118
C.2.4 ou -s ds ..................................... 118
C.2.5 ou -final- a ios ................................. 119
Bibliog aphy 121
Cu iculum Vi ae 137
VIII

Lis o Figu es
1.1 Basic scheme o he s ipping p ocess. ....................... 5
1.2 Basic scheme o an AMS sys em. ........................... 5
1.3 Layou o se e al AMS acili ies. ........................... 8
2.1 Pic u e and schema ic iew o he 1 MV AMS sys em a CNA Se ille. ....... 14
2.2 Pic u e and schema ic iew o he 600 kV AMS sys em a ETH. .......... 15
2.3 Scheme o a high-cu en spu e ing ion sou ce. .................. 16
2.4 Simula ion o ions passing h ough an Einzel lens on SIMION®. ......... 17
2.5 Pis on aking a a ge ou o he SARA sample ca ousel............... 18
2.6 SIMION® simula ion o he SO-110B ion sou ce. ................. 18
2.7 Ta ge wheel used by he MC-SNICS ion sou ce. .................. 19
2.8 SIMION® simula ion o he ion sou ce om TANDY. ............... 19
2.9 Basic ope a ion p inciple o he sys ems used by he 2 di e en ypes o pa icle
accele a o o ge he high e minal ol age. .................... 22
2.10 Mean cha ge s a e o u anium ions as a unc ion o he ion ene gy o di e en
s ippe ma e ials. ................................... 24
2.12 SIMION® iew o an elec os a ic quad upole. ................... 26
2.13 Pelle chain used o anspo he cha ge in he pelle on om TANDY. ..... 27
2.14 Focusing in he X axis o magne ic dipoles depending on he inclina ion o he
bo de s. ......................................... 29
IX
LIST OF FIGURES
2.15 Focusing in he Y axis o magne ic dipoles depending on he inclina ion o he
bo de s. ......................................... 30
2.16 Scheme o an ESA wi h sphe ical geome y. .................... 31
2.17 Beha iou o di e en ions wi h he same mass bu di e en ene gy in an
ach oma ic sys em. .................................. 32
2.18 GICOSY simula ion o 41Ca2+ ions in he HE side o SARA. ............. 33
2.19 Fa aday cups used in AMS measu emen s wi h SARA. ............... 34
2.20 GICOSY simula ion o 41Ca2+ ions in he HE side o TANDY. ............ 34
2.21 Scheme o he ope a ion p inciple o a GIC. .................... 35
2.22 S opping powe o 41Ca and 41K ions wi h an ene gy o 26.5 MeV in bu ane
acco ding o SRIM. ................................... 36
2.23 De ec o signals in a 41Ca measu emen wi h he 5 MV AMS sys em a SUERC,
Glasgow. ........................................ 37
2.24 Scheme o he compac GIC a SARA. ........................ 38
2.25 Compa ison be ween he old HVE GIC, and he new compac GIC. ....... 38
2.26 Pic u e and scheme o he e ac able GIC used in he 41Ca measu emen s wi h
TANDY. ......................................... 39
3.1 SRIM simula ion o he ajec o ies o se e al 41Ca ions a wo di e en ene -
gies in a GIC de ec o . ................................. 42
3.2 Spec a simula ion wi h SRIM o an equal numbe o coun s o 41Ca and 41K a
di e en ene gies. ................................... 42
3.3 E olu ion o 39K/40Ca a ios om blank and s anda d samples in a ge holde s
om di e en ma e ials. ............................... 47
3.4 Time e olu ion o he 40Ca2+ cu en , he 41M2+ and 39K2+ a es, and he 39K/40Ca,
41M/40Ca and 41M/39K a ios o a blank sample (blank 1) du ing a measu e-
men a SARA. ..................................... 49
3.5 Time e olu ion o he 40Ca2+ cu en , he 41M2+ and 39K2+ a es, and he 39K/40Ca,
41M/40Ca and 41M/39K a ios o ano he blank sample (blank 2) du ing a mea-
su emen a SARA. ................................... 50
X
LIST OF FIGURES
3.6 Typical scheme o a 41Ca measu emen wi h a compac AMS sys em imple-
men ing he K-co ec ion. ............................... 53
4.1 (40CaF3)-cu en ou pu om he SO-110B ion sou ce a SARA o di e en
blank a ge s mixed wi h sil e o niobium a di e en weigh a ios. ...... 63
4.2 (40CaF3)-cu en om 2 di e en blank samples o e a long ime ange. . . . . 64
4.3 40Ca2+ ansmission and 41M/40Ca backg ound as unc ions o he He s ippe
p essu e a SARA. ................................... 65
4.4 40Ca3+ ansmission as a unc ion o he He s ippe p essu e a SARA. ..... 67
4.5 Cha ge s a e dis ibu ion o he (40CaF3)-→40Caqs ipping p ocess in He a
225 keV and 400 keV. ................................. 68
4.6 Rela i e posi ion o he FC3 depending on he iso ope whose cu en is mea-
su ed and he adionuclide which is uned o c oss o he ESA. ......... 69
4.7 Expe imen al a simula ed spec a o 41Ca2+ ions eaching he compac GIC a
SARA. .......................................... 70
4.8 Expe imen al spec a o 41Ca3+ and 39K3+ ions eaching he compac GIC a
SARA. .......................................... 71
4.9 Rela ionship be ween he 41M a e eaching he compac GIC and he de ec o
efficiency. ........................................ 72
4.10 Linea co ela ion be ween he 41M/40Ca and 39K/40Ca a ios in blank samples
a SARA. ......................................... 73
4.11 Scheme o he calcium isola ion me hod o u ine samples. ........... 75
4.12 Compa ison o he 41Ca/40Ca a ios om he biomedical samples a TANDY and
SARA. .......................................... 81
4.13 Tempo al e olu ion o he u ina y 41Ca/40Ca a io in subjec 4, including mea-
su emen s a TANDY and SARA. ........................... 82
4.14 Tempo al e olu ion o he u ina y 41Ca/40Ca a io in subjec 5, including mea-
su emen s a TANDY and SARA. ........................... 83
4.15 Tempo al e olu ion o he u ina y 41Ca/40Ca a io in subjec 6, including mea-
su emen s a TANDY and SARA. ........................... 84
XI
LIST OF FIGURES
4.16 Tempo al e olu ion o he u ina y 41Ca/40Ca a io in subjec 7, including mea-
su emen s a TANDY and SARA. ........................... 85
4.17 Tempo al e olu ion o he u ina y 41Ca/40Ca a io in subjec 8, including mea-
su emen s a TANDY and SARA. ........................... 86
4.18 Tempo al e olu ion o he u ina y 41Ca/40Ca a io in subjec 9, including mea-
su emen s a TANDY and SARA. ........................... 87
4.19 Tempo al e olu ion o he u ina y 41Ca/40Ca a io in subjec 10, including mea-
su emen s a TANDY and SARA. ........................... 88
5.1 40Ca neu on cap u e c oss sec ion and he mal neu on ene gy dis ibu ion. . 90
5.2 Scheme o he calcium isola ion p ocess o conc e e samples. ......... 93
5.3 E ec o he acid diges ion o conc e e sample depending on he p esence o
agg ega es. ....................................... 94
5.4 Scheme o he segmen a ion o he bioshield o Jos´
e Cab e a and 3D pic u e o
he shape om each block. .............................. 96
5.5 Schema ic iew o he posi ion o he d ill co es. .................. 97
5.6 Posi ion o he samples aken by sc apping he la e al su ace o he sub-blocks. 98
5.7 41Ca/40Ca esul s om he d ill co e samples and co espondence wi h he mal
neu on fluence. .................................... 99
5.8 C oss sec ion o he Jos´
e Cab e a eac o essel and bioshield when he plan
was ope a i e a he heigh o maximum neu on fluence. ............ 100
5.9 41Ca/40Ca esul s om he la e al samples. ..................... 102
XII
Lis o Tables
1.1 Common adionuclides measu ed by AMS. ..................... 6
1.2 41Ca pe o mance pa ame e s om di e en AMS acili ies. ........... 7
3.1 Ene gy loss s aggling and a e age ene gy di e ence wi h 41K ions o 3.4 MeV
41Ca ions a e a silicon ni ide window as a unc ion o he oil hickness. . . . 43
3.2 Ra es om di e en ions in an aluminum dummy and a blank sample du ing
he expe imen wi h he 600 kV AMS sys em a ETH. ............... 46
4.1 Vol ages a he accele a o e minal and he ESA in o de o selec di e en
ions wi h he HE spec ome e a SARA. ....................... 73
4.2 Blank co ec ed 41Ca/40Ca a ios o he s anda d samples used in one o he
41Ca measu emen s a SARA using he 2+ s a e. .................. 77
4.3 Compa ison o he 41Ca pe o mance pa ame e s a SARA and TANDY. ..... 78
4.4 Measu emen o he ETH in-house 41Ca s anda ds B10 and B8 a SARA. . . . . 79
XIII

LIST OF TABLES
XIV
Ac onyms
AMS Accele a o Mass Spec ome y
CNA Cen o Nacional de Acele ado es
ESA Elec os a ic Analyze
ETH LIP ETH Labo a o y o Ion Beam Physics
FC Fa aday Cup
GIC Gas Ioniza ion Chambe
HE High Ene gy
HEM High Ene gy Side Magne
HVE High Vol age Enginee ing Eu opa B.V.
LE Low Ene gy
LEM Low Ene gy Side Magne
LSC Liquid Scin illa ion Coun ing
NEC Na ional Elec os a ics Co p.
NPP Nuclea Powe Plan
NUS Na ional Uni e si y o Singapo e
PWR P essu ized Wa e Reac o
XV
Abs ac
The accele a o mass spec ome y (AMS) echnique, de eloped 40 yea s ago o 14C da ing,
was soon used o measu e o he long-li ed adionuclides. One o he adionuclides mea-
su ed by AMS since hese ea ly yea s is 41Ca and o a a ie y o applica ions. Some o hese
applica ions a e: he calcula ion o he e es ial age o me eo i es, he s udy o nuclea
eac ions o as ophysical in e es , he unde s anding o he calcium me abolism, and he
cha ac e iza ion o low-le el nuclea was e.
The main challenge in 41Ca AMS is dealing wi h he in e e ence caused by i s s able isoba ,
41K. This in e e ence is educed by using calcium fluo ide (CaF2) samples and he ex ac ion
o he (CaF3)-ion. A la ge AMS acili ies, besides, 41K ions can be disc imina ed om 41Ca
ions using di e en de ec ion echniques based on he ene gy loss dependence on he a omic
numbe . The 41Ca measu emen wi h low ene gy AMS sys ems, like he 1 MV sys em a
he Cen o Nacional de Acele ado es (CNA), is qui e challenging, since his disc imina ion is
no possible. Ne e heless, he 41K con ibu ion can be es ima ed and, he e o e, co ec ed,
hanks o he sequen ial de ec ion o he o he s able iso ope o po assium, 39K (K-co ec ion).
Al hough he sensi i i y achie ed in 41Ca AMS a low ene gies is 3-4 o de s o magni ude
lowe han hose achie ed a la ge acili ies, i allows he compe i i e measu emen s o
biomedical applica ions, and he cha ac e iza ion o conc e e samples om nuclea eac o
bioshields.
Since 41Ca AMS a low ene gies is limi ed o he es ima ion o 41K in e e ence, i is ad isable
o s udy he di e en ways owa d he p oduc ion o his in e e ence. Some ac o s ela ed
o i ha e been s udied in di e en expe imen s pe o med wi h he 1 MV AMS sys em a
CNA (SARA) and he 600 kV AMS sys em a he ETH Labo a o y o Ion Beam Physics in Zu ich
(TANDY). Fo ins ance, we could demons a e ha 41K can be injec ed also as he (41K57Fe)-
molecula ion. As a consequence, 41K in e e ence is dependen on he ma e ials used du ing
he sample p essing. We also p o ed ha , e en when bo h 41K/40Ca and 39K/40Ca a ios
change o e ime, he ela ion be ween bo h, 41K/39K emains cons an . The e o e, he K-
co ec ion is a obus me hod o es ima e he 41K in e e ence.
XVII
CHAPTER 1. INTRODUCTION
Nowadays, hough, he mos common applica ion o 41Ca does no come om i s na u al
cosmogenic o an h opogenic p oduc ion, bu om i s biomedical capabili y as a ace o
calcium me abolism [Elmo e e al.,1990;F eeman e al.,1997]. The use o 41Ca has none o
he disad an ages om o he calcium adioiso opes o i s s able iso opes:
•In he fi s case, he adiological haza d o he adminis a ion o a dose o 45Ca o 47Ca,
whi h hal -li es o 165 days and 4.5 days espec i ely, would be highe , and hey allow
much lowe acing imes because o hei decay, while 41Ca is only limi ed by i s bi-
ological exc e ion. By con as , 41Ca o al doses up o 3700 Bq a e ac ually been used
[Denk e al.,2006]; aking in o accoun ha he dose coefficien o inges ed 41Ca is
2.9 ×10-10 S ·Bq-1 [ICRP,2012], he dose in 50 yea s would be a ound 1 µS . This
dose is o ally neglible, since i is much lowe han, o ins ance, he annual na u al
backg ound exposu e ( ypically highe han 1 mS ).
•In he second case, s able iso opes o calcium such as 42Ca o 44Ca equi e high chemical
doses so ha hei iso opic abundances change significan ly. Since 41Ca is no na u ally
p esen , i he dose has a ela i ely high 41Ca/40Ca a io e y low calcium con en has
o be adminis a ed in o de o be able o measu e 41Ca concen a ions in issues we e
calcium is inco po a ed.
A e y ecen compendium o calcium me abolism s udies pe o med wi h 41Ca acing and
AMS measu emen s is ound in Wea e e al. [2017].
1.2 Accele a o Mass Spec ome y (AMS): finding a needle in a
hays ack
Con en s om any adionuclide wi h a ela i ely long hal -li e, such as 41Ca, a e in insically
ha d o measu e by adiome ic echniques because o he e y low ela ed ac i i ies. I s
di ec de ec ion wi h con en ional mass spec ome y sys ems mee wi h he in e e ence o
molecula ions wi h he same mass ha he adionuclide o in e es . All he mass spec ome-
y sys ems ollow he same basic scheme: ioniza ion o he sample, he use o elec omagn ic
sepa a o s o selec i s mass/cha ge a io, and de ec ion o hese ions. Since he con en s o
long-li ed adionuclides a e ypically o de s o magni ude below he pa pe million (ppm)
le el, e en a ace p oduc ion o isoba molecula ions would make i impossible o measu e
o he adionuclide o in e es .
Accele a o Mass Spec ome y (AMS) sol es his p oblem hanks o he elec on s ipping
4

1.2. ACCELERATOR MASS SPECTROMETRY (AMS): FINDING A NEEDLE IN A HAYSTACK
p ocess wi h a solid o gaseous a ge ha akes place a ene gies a ailable wi h pa icle
accele a o s. In his p ocess, he molecula backg ound is des oyed (see Figu e 1.1).
Figu e 1.1: Basic scheme o he s ipping p ocess.
O he sup ession p ocesses, hough, a e necessa y o sol e he possible in e e ence om
s able isoba s. Figu e 1.2 shows he basic scheme o a common AMS sys em, oge he wi h
he o he isoba supp ession echniques used in AMS:
•Ioniza ion: nega i e ioniza ion p o ides, in some cases, a fi s sup ession o educ ion
o he in e e ence om he s able isoba , because o he ins abili y o nega i e ions
om elemen s o molecules wi h a nega i e elec on affini y (EA). This is he case, o
ins ance, o 14C, since 14N does no o m nega i e ions [Benne e al.,1977]. E en i ha
EA is posi i e, bu e y low, i s in e e ence is educed se e al o de s o magni ude,
since he nega i e ion p oduc ion efficiency inc eases exponen ially wi h he EA [Al on,
1993].
Figu e 1.2: Basic scheme o an AMS sys em. While he possibili y o sup ession o s able isoba s
depends on he adionuclide and expe imen al sys em, he use o he s ipping p ocess o des oy he
molecula isoba s is common o all AMS measu emen s.
5
CHAPTER 1. INTRODUCTION
•De ec ion: echniques based on he di e en ene gy loss can be used because o he
high ene gies o he ions. Since he ene gy loss in a ma e ial depends on he a omic
numbe Zo he ion, he di e ence in he ene gy loss be ween he adionuclide o in-
e es and i s s able isoba can be used o educe he in e e ence en e ing he de ec o
and sepa a ing bo h signals.
The AMS echnique, ini ially de eloped o 14C measu emen s, was soon applied o many
o he long-li ed adionuclides, including 41Ca [Go e e al.,1979;Elmo e and Phillips,1987].
Table 1.1 shows some o he mos impo an ”AMS adionuclides”.
Table 1.1: Common adionuclides measu ed by AMS. Hal -li es om Na ional Nuclea Da a Cen e
[n.d.], excep o 10Be [Ko schinek e al.,2010]. The main echnique used o supp ess/ educe he
in e e ence caused by he s able isoba is also shown: ”Ioniza ion” e e s o hose cases whe e he
nega i e ion o he s able isoba is uns able o me as able; ”De ec ion”, o hose cases whe e he
di e en ene gy loss is used o dis inguish he adionuclide om i s s able isoba .
Radionuclide Hal -li e (My) Nega i e ion S able isoba Supp ession / educ ion
10Be 1.387 (BeO)- 10B De ec ion
14C 0.0057 C- 14N Ioniza ion
26Al 0.717 Al-
(AlO)-
26Mg Ioniza ion
De ec ion
36Cl 0.301 Cl-36A
36S
Ioniza ion
De ec ion
41Ca 0.0994 (CaH3)-
(CaF3)-
41K Ioniza ion + de ec ion
129I 15.7 I- 129Xe Ioniza ion
236U 23.42 (UO)-@
1.2.1 41Ca AMS
In he pa icula case o 41Ca, wo di e en chemical o ms and hei espec i e nega i e ions
can be used in o de o deal wi h i s s able isoba , 41K: calcium hyd ide (CaH2), and ex ac ion
o he (CaH3)-ion; o calcium fluo ide (CaF2), and ex ac ion o he (CaF3)-ion. The choice o
one ma e ial o he o he depends on se e al ac o s, since each one has bo h ad an ages
and disad an ages:
•CaH2: he main ad an age o selec ing he (41CaH3)-ion in he LE side is he ines abili y
o he (KH3)-ion. A 41K/40Ca in e e ence in he le el o 10-13 is ypically ound om he
6
1.2. ACCELERATOR MASS SPECTROMETRY (AMS): FINDING A NEEDLE IN A HAYSTACK
con ibu ion o (41KH2)-ions whe e one o he 2 hyd ogen nuclides is a deu e ium (2H)
[Fink e al.,1990]. The main disad an age is he high hyg oscopy o CaH2, which makes
i s handling p oblema ic, making i necessa y o minimize i s con ac wi h ai [Sha ma
and Middle on,1987]. This choice is sui able o applica ions whe e ew samples a e
measu ed and e y high sensi i i ies a e equi ed, like he measu emen o acc e ion
a e o ex a e es ial 41Ca in An a c ica [G´
omez-Guzm´
an e al.,2015].
•CaF2:in con as , calcium fluo ide chemical p epa a ion is mo e simple and i is a chem-
ically s able sal [Kubik and Elmo e,1989]. Howe e , he (KF3)-ion is s able, e en when
i s s abili y is much lowe han ha om (CaF3)-, leading o a highe 41K/40Ca in e e -
ence in he le el o 10-12-10-11 [Zhao e al.,2010]. Ion sou ce ou pu cu en s a e also
sligh ly lowe [Middle on,1990], bu in his case he di e ence is no so impo an . This
choice is sui able o applica ions whe e la ge numbe s o samples a e in ol ed and
ela i ely lowe sensi i i ies a e equi ed, like he biomedical applica ions de ailed in
sec ion 1.1 [F eeman e al.,1995].
In any case, in con en ional AMS acili ies, which ypically use e minal ol ages highe han
3 MV, he di e en ene gy loss on a ma e ial, as i was al eady s a ed, can be used o sepa a e
he signals o 41Ca and 41K in he de ec o [Fink e al.,1990;Vockenhube e al.,2005;Wallne
e al.,2010;Hosoya e al.,2017]. Be e sepa a ion be ween bo h signals can be achie ed a
highe ene gies, esul ing in a lowe ac ion o 41K coun s ha in e e e wi h he 41Ca coun s.
The pe o mance pa ame e s o 41Ca om se e al AMS acili ies a e p esen ed in Table 1.2.
The layou s om hese sys ems a e shown in Figu e 1.3.
Table 1.2: 41Ca pe o mance pa ame e s om di e en AMS acili ies. Only da a om measu emen s
using calcium fluo ide a e used, e en when Wallne e al. [2010] p esen s also pe o mance pa ame-
e s wi h hyd ide.
Facili y VERA
(Vienna, Aus ia)
DREAMS
(D esden, Ge many)
14UD ANU
(Canbe a, Aus alia)
Re e ences [Wallne e al.,2010][Akhmadalie e al.,2013]
[Rugel e al.,2016][Fifield e al.,2010]
Te minal ol age (MV) 3 6 14
S ippe ma e ial O2gas A gas O2gas + C oil
HE side cha ge s a e 4+ 4+ 7+
Backg ound (×10-15)50 2 1
7
CHAPTER 1. INTRODUCTION
(a) VERA
(b) DREAMS
(c) 14UD ANU
Figu e 1.3: Layou o se e al AMS acili ies: he 3 MV sys em a he Uni e si ¨
a Wien (a), he 6 MV
sys em a he Helmhol z-Zen um D esden-Rossendo (b) and he 14 MV sys em a he Aus alian
Na ional Uni e si y (c).
1.3 AMS a low ene gies
Du ing he fi s decades o AMS i was belie ed ha , in o de o des oy he molecula back-
g ound, cha ge s a es highe o equal o 3+ had o be selec ed (”golden ule”o AMS). Molec-
ula ions wi h hose cha ge s a es become uns able because o he elec os a ic epulsion
[Su e ,1990]. Wi h compac pa icle accele a o s wi h e minal ol ages ≤1 MV, hough, he
mos popula ed cha ge s a es o ca bon ions a e 1+ and 2+ [Jacob e al.,2000] and molecu-
la in e e en s would s ill su i e. Despi e his ac , Lee e al. [1984] and Su e e al. [1997]
showed ha , i he s ippe mass hickness o he s ippe was high enough, his molecula
backg ound was educed o neglible le els.
Based on hese esul s, he fi s low-ene gy AMS sys em o 14C measu emen s was de el-
oped: he 600 kV AMS sys em a ETH Zu ich, TANDY, de eloped by he ETH Labo a o y o
Ion Beams Physics (LIP) in collabo a ion wi h Na ional Elec os a ics Co p. (NEC, USA) [Synal
e al.,2000]. Soon, he capabili ies o his sys em o o he AMS adionuclides we e s udied
8
1.4. MOTIVATION AND OUTLINE
and, since hen, he sys em has also pe o med ou ine measu emen s o 10Be, 26Al, 41Ca,
129I and ac inides (236U, Pu iso opes...) [S ocke e al.,2004,2005]. The fi s 1 MV AMS sys em
designed by High Vol age Enginee ing Eu opa B.V. (HVE, Ne he lands) was ini ially designed o
measu e no only 14C, bu also o he ypical AMS adionuclides, such as 10Be and 26Al [Klein
e al.,2006,2007], and ins alled a he Cen o Nacional de Acele ado es (CNA) in Se ille. The
peculia i ies om hese sys ems and he de elopmen s implemen ed on hem a e p esen ed
in chap e 2.
These small accele a o s, ne e heless, s ill ha e some p ope ies in common wi h la ge
machines, like he need o using some diele ic gas, as sul u hexafluo ide, o elec ically
insula e he e minal om he accele a o ank. This necessi y does no apply any mo e when
e en lowe e minal ol ages a e used, and acuum insula ion is enough o a oid discha ges.
This is he case o he MICADAS sys em, a sys em de eloped by he ETH LIP whose e minal
ol age is p o ided by a 200 kV HV powe supply [Synal e al.,2007]. The MICADAS sys em
is designed only o 14C measu emen s. A new mul i-elemen AMS sys em whose e minal
ol age is p o ided by a 300 kV HV powe supply is now being de eloped also by he ETH LIP
[Maxeine ,2016].
The use o low-ene gy AMS sys ems educe he ope a ing and main ainance cos s o hese
acili ies. The o iginal cos o he sys em is also lowe . Fu he mo e, hei sizes allow hem o
be placed in con en ional labo a o y ooms and no huge and dedica ed halls a e equi ed o
be buil .
1.4 Mo i a ion and ou line
The main goal o his hesis has been he se ing up and op imiza ion o he pe o mance
pa ame e s o he 1 MV AMS sys em a CNA o 41Ca. Since i s ope a ion s a ed in 2006
[Klein e al.,2006], he AMS esea ch g oup a CNA has looked o exploi ing he mul i-
elemen capaci y o he sys em [Klein e al.,2007;Chamizo e al.,2008b]. The same se ing
up and op imiza ion o he pe o mance has been pe o med, o ins ance, o 10Be and 26Al
[Ruiz-G´
omez e al.,2010;Padilla,2015;Scognamiglio,2017], o 129I [G´
omez-Guzm´
an,2010;
G´
omez-Guzm´
an e al.,2012], o 236U [Chamizo e al.,2015a], and o Pu iso opes [Chamizo
e al.,2008a;Chamizo,2009]. Se e al echnical changes and de elopmen s ha e been pe -
o med in collabo a ion wi h bo h, HVE and he ETH LIP [Chamizo e al.,2015c;Scognamiglio
e al.,2016].
Nowadays, he e exis se e al AMS acili ies a ound he wo ld simila o he 1 MV sys em a
CNA. Thus, he s udy o he pe o mance o i s di e en elemen s a e also use ul o o he
9

CHAPTER 1. INTRODUCTION
esea ch g oups, so hey can exploi he capaci ies o hei HVE 1MVAMS sys ems. Some o
hese g oups ac ually included 41Ca in hei accep ance es s [Klein e al.,2013;Heinemeie
e al.,2015], so i can be said ha hey a e in e es ed on pe o ming ou ine 41Ca measu e-
men s in he u u e. The esul s p o ided by his hesis, specially hose also published in
Vi o-Vilches e al. [2017], a e use ul o hese g oups as much as i is o ou s.
This hesis included a s ay o ou mon hs a ETH LIP in o de o pe o m some expe imen s
ela ed o he 41K in e e ence in 41Ca AMS a low ene gies. The s ay also included he collab-
o a ion wi h he ETH LIP in he 41Ca measu emen s o biomedical applica ions pe o med
by hem o he Nu iT ace g oup a he Na ional Uni e si y o Singapo e. The in o ma ion
and expe ience acqui ed du ing his s ay was eally help ul in o de o op imize he 41Ca
measu emen pe o mance o he 1 MV AMS sys em a CNA Se ille.
E en when he s udy o he capabili ies o he sys em o new adionuclides and echnical
de elopmen s hemsel es a e wo o i s main esea ch lines, ou g oup has also ocused on
some impo an applica ion esea ch lines. These esea ch lines include he de elopmen o
chemical p ocedu es o he sample p epa a ion. Nowadays, he mos impo an applica ions
pe o med by ou g oup a e:
•En i onmen al applica ions, specially ela ed o oceanog aphy [Chamizo e al.,2015b,
2016;L´
opez-Lo a e al.,2018;Vi o-Vilches e al.,2018].
•Cha ac e iza ion o Low-Le el Was e (LLW) om ope a ing and decommissioning nu-
clea powe plan s, in collabo a ion wi h he Spanish adioac i e was e managemen
agency, ENRESA [L´
opez-Gu i´
e ez e al.,2013].
Radioca bon applica ions a e no longe pe o med wi h he 1 MV AMS sys em since he g oup
acqui ed a MICADAS sys em in 2012.
The 41Ca AMS measu emen s in he g oup ha e been implemen ed ini ially in he LLW cha -
ac e iza ion p ojec . 41Ca/40Ca a ios ha e been measu ed in conc e e samples om se -
e al loca ions di e en in bo h adial and e ical coo dina es om he p ima y shield o a
decommissioning nuclea powe plan (NPP): he Jos´
e Cab e a NPP. The esul s om hese
measu emen s will be p esen ed in chap e 5.
E en i we did no pe o m measu emen s o biomedical applica ions a CNA Se ille, we did
chemical es s wi h u ine samples. We also showed ha he pe o mance o he 1 MV AMS
sys em is well sui ed o pe o m hose measu emen s o any in e es ed g oup.
This fi s chap e was p o ided in o de o in oduce he eade o he applica ions o 41Ca
measu emen s, he AMS echnique and i s e olu ion o low ene gy sys ems.
10
1.4. MOTIVATION AND OUTLINE
In chap e 2, a desc ip ion o he undamen al p inciples o he mos impo an elemen s in
an AMS sys em is p esen ed. The wo sys ems used in he expe imen s ca ied ou du ing
his PhD hesis a e also concisely desc ibed, om which a mo e de ailed desc ip ion can be
ound in Chamizo [2009]: he 1 MV AMS sys em a CNA Se ille, SARA (Spanish Accele a o o
Radionuclide Analysis); and he 600 kV AMS sy em a ETH Zu ich, TANDY.
The specific challenges ela ed o 41Ca AMS a low ene gies and how he 41K in e e ence
can be es ima ed hanks o he de ec ion o he o he s able iso ope o po assium, 39K (K-
co ec ion) [Vockenhube e al.,2015] a e p esen ed in chap e 3. Some expe imen s pe -
o med on bo h compac AMS sys ems (SARA and TANDY) p o iding in o ma ion abou he
p oduc ion pa hways o he 41K in e e ence a e also p esen ed, as well as he specific com-
plexi y in oduced o he da a analysis by he K-co ec ion. Because o his complexi y, 2
simila p og ams, one o each AMS sys em, we e w i en in FORTRAN code o pe o m he
da a analysis.
The expe imen s in o de o se -up and op imize he 41Ca measu emen s wi h he SARA sys-
em a e p esen ed in chap e 4. These expe imen s include he cha ac e iza ion and op i-
miza ion o he ion sou ce cu en ou pu , he ansmission using He as a s ippe gas, he
de ec ion efficiency, and he pe o mance wi h blank samples (wi h neglible con en o 41Ca)
and s anda d samples (wi h known 41Ca/40Ca a ios). A compa ison o he pe o mance pa a-
men e s wi h hose om he TANDY sys em is shown, oge he wi h he measu emen s o
some o he ETH in-house s anda d samples used in he measu emen s wi h ha sys em and
an in e compa ison expe imen ca ied ou on some o he biomedical samples which we e
measu ed du ing he s ay a ETH LIP.
Resul s om he 41Ca measu emen s wi h he SARA sy em o he conc e e samples om he
p ima y shield o he Jos´
e Cab e a Nuclea Powe Plan a e p esen ed in chap e 5. These
esul s a e p eceded by he exposi ion o he way he 41Ca is p oduced in he conc e e o he
p ima y shield, and how he 41Ca/40Ca can be gi e in o ma ion abou he he mal neu on
fluence he shield was exposed o du ing he ope a ion o he powe plan .
The las chap e summa ize he mos impo an conclusions o his wo k.
11
CHAPTER 1. INTRODUCTION
12
Chap e 2
Compa a i e desc ip ion o he
expe imen al sys ems
2.1 Gene al o e iew o he sys ems
In his chap e , he basic physical p inciples om he ypical elemen s o AMS sys ems a e
p esen ed, oge he wi h he cha ac e is ics o each one o hese elemen s o he sys ems
used du ing his hesis ( he SARA sys em a CNA Se ille and he TANDY sys em a ETH Zu ich).
Bo h sys ems a e desc ibed in de ail in Chamizo [2009]. Addi ional in o ma ion abou he
TANDY sys em can be ound in Jacob [2001]. This chap e , he e o e, ocuses on he changes
implemen ed since hen and he ea u es which a e specially impo an o 41Ca measu e-
men s.
This fi s sec ion p esen s a gene al o e iew o bo h sys ems. The nex sec ions p esen :
•The p oduc ion o nega i e ions by Cs spu e ing (sec ion 2.2).
•The ope a ion and op ical p ope ies o andem accele a o s, oge he wi h he basic
cha ac e is ics o he s ipping p ocess (sec ion 2.3).
•The kinema ic fil e s used in AMS, and he wo di e en sec o s: he LE injec o and he
HE spec ome e (sec ion 2.4).
•The cha ged pa icle de ec ion wi h gas ioniza ion chambe s (sec ion 2.5).
In some o his sec ions, simula ions pe o med wi h GICOSY [Weick,n.d.] a e p esen ed. The
inpu files om hese simula ions a e shown in Appendix A.
13
CHAPTER 2. COMPARATIVE DESCRIPTION OF THE EXPERIMENTAL SYSTEMS
The lack o elec os a ic lens be o e he accele a o suppose a es ic ion in he V0
VT a io so
ha he LE accele a ion ube ocus he ion beam in he s ippe cen e . The eason o his is
explained in he nex sec ion.Jacob [2001] calcula ed ha he op imal a io was V0
VT= 0.102 ≃
0.1. This means ha , o hose adionuclides whe e a e minal ol age o ∼500 kV is used
(10Be, 26Al and 41Ca), he ene gy o he nega i e ions in he LE side should be ∼50 keV, while
o hose adionuclides whe e a e minal ol age o ∼300 kV is used (129I and ac inides), his
ene gy should be 30 keV.
2.3 Tandem accele a o s
In a andem accele a o , he nega i e ions a e a ac ed o i s cen e ( e minal) by a la ge
posi i e po en ial VTV0; he same po en ial accele a es again he posi i e ions a e he
s ipping p ocess which akes place in he e minal, by elec os a ic epulsion. A each side
o he accele a o e minal he e is a se o elec odes connec ed wi h esis o s which a oid a
sudden change in he elec os a ic po en ial (accele a ion ubes). The po en ial in hese ubes
can be supposed o inc ease linea ly i all he esis o s o each ube ha e he same esis ance.
In his case, he elec os a ic o ce which he nega i e ions, wi h a cha ge −e, expe ience in
he fi s accele a ion ube (LE ube) is
FLE− ube =eVT
LLE
,(2.2)
whe e LLE is he leng h o he LE ube. The posi i e ions p oduced a e he s ipping p ocess,
wi h a cha ge qe :q∈Z+, expe ience a o ce in he second accele a ion ube (HE ube) o
FHE− ube =qeVT
LHE
,(2.3)
whe e LHE is he leng h o he HE ube. In he es o he zones (be o e he LE ube, in he
e minal and a e he HE ube) he po en ial is cons an , so he elec os a ic o ce is 0.
I he nega i e ions selec ed in he LE side ha e a mass M0, and he posi i e a omic ions ha e
a mass o M≤M01, he kine ic ene gy Eo he ions is:
•eV0a he en ance o he accele a o .
1M=M0i he nega i e ion was he same a omic species, e.g. 14Cqions om he s ipping o 14C-.M < M0
i his posi i e a omic ion is one o he agmen s om a molecula nega i e ion, e.g. 41Caqions om he s ipping
o (41CaF3)-.
20

2.3. TANDEM ACCELERATORS
•e(V0+VT)a he exi o he LE ube, his is, in he e minal be o e he s ipping p ocess.
•M
M0e(V0+VT)a he en ace o he HE ube, his is, in he e minal a e he s ipping
p ocess. This M
M0 ac o comes om assuming ha he molecula ion b eaks elas ically,
so he eloci y o i s agmen s is he same, dis ibu ing he kine ic ene gy among hose
agmen s p opo ionally o hei masses.
•M
M0e(V0+VT) + qVTa he exi o he HE ube.
In each one o hese 4 loca ions, he e is a change o he elec os a ic o ce ∆Fwhich has a
ocusing e ec whose op ical powe 1
is di ec ly p opo ional o ∆F
E[La son,1974]2. I we
define he e ec i e cha ge s a e o he posi i e ions as q∗≡M0
Mq[Maxeine ,2016], and aking
in o accoun ha VTV0, i can be deduced ha he op ical powe is di ec ly p opo ional
o:
•VT
V0
1
LLE a he en ance o he LE ube.
•−1
LLE a he exi o he LE ube.
•q∗1
LHE a he en ace o he HE ube.
•−q∗
1+q∗
1
LHE a he exi o he HE ube.
The VT
V0 ela ionship becomes he key ac o in he op ical ocusing o he nega i e ions in he
e minal, while he e ec i e cha ge s a e q∗defines he op ical beha iou o he posi i e ions
a e he e minal, in he he HE side o he AMS sys em.
Depending on how he e minal ol age is achie ed, he e a e wo main ypes o pa icle ac-
cele a o s used in AMS (no aking in o accoun he HV powe supplies used by he MICADAS
sys em):
•The ande ons: hese accele a o s use a Cockc o -Wal on ci cui based on a RF powe
supply connec ed o a se ies o diodes and capaci o s which, a e some ime, inc ease
he ou pu ol age o a maximum alue. In he case a o Figu e 2.9 he mos basic
e sion o his kind o ci cui is shown, bu he numbe o diode-capaci o s ages can
be inc eased, being he ou pu ol age di ec ly p opo ional o he numbe o s ages.
The main disad an age o his kind o accele a o is ha only e minal ol ages up o
6 MV ha e been achie ed wi h hem. On he o he hand, hei e minal ol ages a e
e y s able and hey equi e li le main enance.
2E en when his app oxima ion o a hin lens is no sui able o ge he ac ual beha iou o a ce ain accele a o
ube, he p opo ionali y can be used o compa e he beha iou o di e en accele a o ubes o di e en ions.
21
CHAPTER 2. COMPARATIVE DESCRIPTION OF THE EXPERIMENTAL SYSTEMS
•The pelle ons: hese accele a o s use a Van de G aa gene a o consis ing in a chain
o me allic pelle s which a e cha ged posi i ely ou side he e minal, and deposi his
cha ge on i , inc easing he e minal ol age (see he case b o Figu e 2.9). The pelle s
can also be cha ged nega i ely be o e lea ing he e minal, so he posi i e cha ge de-
posi ed by each pelle a e passing h ough doubles. The ol ages ha some o hese
accele a o s can achie e a e e en highe han 20 MV. The main disad an age om he
pelle ons is ha he use o he pelle chains equi es a egula main enance.
In bo h cases, he high elec ic field be ween he e minal and he accele a o ank (a g ound
po en ial) equi es he use o a gaseous diele ic ma e ial in o de o a oid he loss o he ol -
age by elec ic discha ges. The mos common ma e ial o his pu pose is sul u hexafluo ide
(SF6).
(a) Tande on (b) Pelle on
Figu e 2.9: Basic ope a ion p inciple o he sys ems used by he 2 di e en ypes o pa icle accele a-
o o ge he high e minal ol age.
The s ipping p ocess
Se e al la ge AMS sys ems use bo h solid and gaseous ma e ials as s ippe ma e ials. The
angula s aggling caused by solid s ippe s is no so impo an in hese sys ems since his
e ec is in e sely p opo ional o he kine ic ene gy. In con as , compac AMS sys ems, like
he 1 MV AMS sys em a CNA Se ille and he 600 kV AMS sys em a ETH Zu ich, only use
gaseous s ippe s, since he posibili y o easily change he mass hickness o he ma e ial is
c i ical o des oy he molecula backg ound wi hou se e ely inc easing unwan ed angula
s aggling o sca e ing e ec s. Wi h gaseous s ippe s, he desing o he pumping sys em o
he gas in and ou o he s ippe ube is ex emely impo an in o de o minimize he leakage
o he s ippe gas o he accele a ion ubes ( esidual gas). This educes he backg ound
22
2.3. TANDEM ACCELERATORS
p oduced by ions wi h di e en mass/cha ge a io which a e sca e ed by collisions wi h he
esidual gas, leading o a ajec o y which makes hem each he de ec o .
The s ipping p ocess is caused by he cha ge exchange eac ions be ween he s ippe ma-
e ial and he ions. I he s ippe mass hickness is high enough, hese cha ge exchange
eac ions each an equilib ium whe e he e a e no nega i e ions, and he posi i e ions and
he a omic species a e dis ibu ed in se e al cha ge s a es q≥0. The ac ion Φqbe ween he
numbe o ions wi h cha ge s a e qp oduced and he o al numbe o nega i e ions s ipped
is called ”cha ge s a e ac ion”. I is clea , hen, ha
Z
X
q=0
Φq= 1, whe e Zis he a omic num-
be o he species. The ”mean cha ge s a e”is, so, defined as ¯q≡
Z
X
q=0
qΦq, and depends on
he s ipping ma e ial, bu also on:
•The kine ic ene gy which he ions ha e in he e minal Es ip =M
M0e(V0+VT), also called
s ipping ene gy. The highe his ene gy is, he highe he popula ed cha ge s a es
a e. The mos popula ed s a es a low ene gy AMS sys ems like SARA and TANDY o
mos o he adionuclides a e ypically 1+, 2+ and/o 3+, being Φq≃0 : ∀q > 3. This
is he minimum cha ge s a e eached a la ge AMS acili ies using e minal ol ages
highe han 2.5 MV. This dependence ac ually is on he eloci iy; he e o e, o he same
s ipping ene gy, wo iso opes om he same elemen ha e sligh ly di e en ¯q.
•The a omic species o he ions. The highe he a omic numbe is, he highe he popu-
la ed cha ge s a es a e. A SARA, o ins ance, he mos popula ed cha ge s a e in he
He gas s ippe o 10Be is 1+, bu 2+ o 41Ca and 3+ o 236U.
The s ippe ansmission o a cha ge s a e qis defined as he a io be ween he a e o
posi i e ions a his cha ge s a e ha exi he accele a o and he o al a e o nega i e ions
which en e i . The e o e, his ansmission is sligh ly lowe han he cha ge s a e ac ion,
because o he op ical loses due o he angula s aggling in oduced by he s ippe . I is
es ima ed by measu ing he cu en −I0o he nega i e ion om he s able iso ope en e ing
he accele a o , and he cu en Iq om he posi i e ion o cha ge s a e q om s able iso ope
ion. The ansmission, hen, is
Tq=Iq
qI0
.(2.4)
Bo h sys ems desc ibed in his chap e ini ially used a gon as s ippe gas, bu changed o
helium in he ecen yea s [Vockenhube e al.,2013;Scognamiglio e al.,2016].
23
CHAPTER 2. COMPARATIVE DESCRIPTION OF THE EXPERIMENTAL SYSTEMS
The use o He gas as s ippe ma e ial means an imp o emen a low-ene gy AMS because o
he dependence o he mean cha ge s a e wi h he eloci y. An example o his can be seen in
Figu e 2.10, which shows da a om Wi kowe and Be z [1973] using u anium ions (because
o he high mass, eloci ies a e much lowe han o o he ypicial ions as ca bon o simila
ene gies): while o high ene gies he mean cha ge s a es o o he ma e ials a e highe han
o helium, o low ene gies he case is he opposi e. This makes He gas he bes op ion o
AMS a low ene gies and/o o hea y ion AMS [Vockenhube e al.,2013;Winkle e al.,2015].
In addi ion, he angula s aggling is educed because o he low mass o helium.
Figu e 2.10: Mean cha ge s a e o u anium ions as a unc ion o he ion ene gy o di e en s ippe
ma e ials. Da a om Wi kowe and Be z [1973], p esen ed also in Vockenhube e al. [2013].
2.3.1 The 1 MV ande on a SARA
Figu e 2.11 shows he scheme o he elemen s wi hin he 1 MV ande on used by he SARA
sys em. All he esis o s om each o he accele a ion ubes a e equal, so i can be supposed
ha inside hem he e is a cons an elec ic field =VT
L, whe e L=LLE =LHE = 72.5cm. In
bo h cases, he elec ic field is in di ec ion om he e minal o he accele a o ex emes.
A he en ance o he accele a o he e is an elec ode (Q-Snou ), whose po en ial VQ−Snou >0
can be changed. The Q-Snou lens is used o inc ease he ene gy wi h which he ions en e
he LE accele a o ube and, so, educing he ocusing powe om he ape u e o his ube.
24
2.3. TANDEM ACCELERATORS
The con e gen powe o he ube en ance is, in his case, p opo ional o VT
V0+VQ−Snou
.
Chamizo [2009] calcula ed ha , in o de o ocus he beam in he cen e o he s ippe ube,
his ol age has o be
VQ−Snou = 0.082 ·VT−V0.(2.5)
In he s ippe ube, he gas p essu e is measu ed only a i s cen e (p0). In o de o cal-
cula e he s ippe mass hickness, i is assumed ha he gas p essu e ollows a iangula
dis ibu ion, so he a e age p essu e would be p0
2, and he mass hickness,
ρA=MgasL
RT
p0
2,(2.6)
whe e Mgas is he mola mass o he s ippe gas, L he leng h o he s ippe ube (30 cm
o SARA), R= 83145 mba ·cm3·K−1·mol−1, and T he gas empe a u e, which is a a oom
empe a u e o 294.15 K (21 ºC).
Since he ocusing powe o he en ance o he HE accele a o ube is p opo ional o q∗≡
M0
Mq, a quad upole iple (Q-Pole) is placed a he exi o he accele a o o e ocus he beam.
An elec os a ic quad upole is a se o ou elec odes, wo o hem acing in he X di ec ion
and connec ed o a po en ial VQ, and he o he wo acing in he Y di ec ion and connec ed
o a po en ial −VQ(see Figu e 2.12). The elec odes ha e a geome y so ha he po en ial φ
in places close o he cen e is app oxima ely pa abolic, his is, φ∝(x2−y2)(so i ollows he
Laplace equa ion ∆φ= 0).
Figu e 2.11: Scheme o he sys em o he 1 MV accele a o om SARA. Pic u e om Chamizo [2009],
modified by Scognamiglio [2017].
25

CHAPTER 2. COMPARATIVE DESCRIPTION OF THE EXPERIMENTAL SYSTEMS
Since each quad upole ocuses he beam in one axis, bu un ocuses i in he o he , se s o
mul iple quad upoles (mul iple ) a e used, whe e each quad upole uses he opposi e po en-
ials as he p e ious one. In he case o he quad upole iple a SARA, he po en ials a e, o
aVQ>0:
•VQin he X axis o he cen al quad upole, and in he Y axis o he fi s and hi d
quad upoles.
•−VQin he Y axis o he cen al quad upole, and in he X axis o he fi s and hi d
quad upoles.
Figu e 2.12: SIMION® iew o an elec os a ic quad upole. The a ows ep esen he elec os a ic
o ce which a posi i ely cha ged pa icle would expe ienced o a VQ>0.
2.3.2 The 600 kV pelle on a TANDY
TANDY uses a 600 kV pelle on om NEC, which has only one cha ging pelle chain, shown in
Figu e 2.13. All he esis o s o he HE ube a e equal, so i can be assumed ha he e is a
cons an elec ic field HE =VT
L, whe e L=LLE =LHE = 28 cm is he leng h o each one o
he accele a ion ube. The esis o s o he LE, ins ead, ha e a esis ance in he fi s hi d pa
o he accele a ion ube, bu 3
2 imes ha esis ance in he es [Jacob,2001], leading o wo
di e en elec ic fields: a field 1=3
4
VT
Lin he fi s hi d pa , and a field 2=3
21=9
8
VT
Lin
he o he wo hi d pa s, so 1L
3+22L
3=VT.
Thanks o his lowe elec ic field a he en ance o he LE ube, he op imal V0
VT a io o ocus
he beam in he cen e o he s ippe can be achie ed wi h he ene gies a ailable wi h he
ion sou ce e en o e minal ol ages o 500 kV3. The elec ic field di e ence a e his fi s
3I he elec ic field we e =VT
L=4
3ε1, he op imal V0
VTwould be also highe in o de o ha e he same op ical
powe .
26
2.4. ELECTROMAGNETIC KINEMATIC FILTERS
hi d pa o he ube also has a ocusing e ec , bu his is much lowe han he one a he
en ance because o he highe ene gy o he ions in ha place.
In his case, he s ippe gas p essu e is measu ed no only in he cen e , bu also in he
s ippe ape u es. Toge he wi h simula ions, his allows a mo e accu a e es ima ion o he
s ippe mass hickness. Ne e heless, he simula ion on Jacob e al. [2000], o ins ance,
es ima es an A mass hickness o 2 µg·cm-2 o a cen al p essu e o 0.058 mba , whe eas
using Equa ion 2.6, a mass hickness o 2.04 µg·cm-2 is es ima ed, showing he ag eemen o
he iangula app oxima ion wi h mo e accu a e es ima ions.
Figu e 2.13: Pelle chain used o anspo he cha ge in he pelle on om TANDY.
2.4 Elec omagne ic kinema ic fil e s
The kinema ic fil e s mos commonly used by AMS sys ems o selec only he ions wi h a
ce ain mass/cha ge a io a e:
•Magne ic dipoles, which use a magne ic o ce o selec he p oduc o bo h he mass/cha ge
a io and he ene gy/cha ge a io (EM
q2).
•Elec os a ic analyze s, which use an elec os a ic o ce o selec he ene gy/cha ge a io
(E
q).
•Wien fil e s, which con on wo o ces (an elec os a ic one and a magne ic one) o
selec he eloci y o he ions ( ∝qE
M).
27
CHAPTER 2. COMPARATIVE DESCRIPTION OF THE EXPERIMENTAL SYSTEMS
The AMS sys ems desc ibed in his hesis use only magne ic dipoles and elec os a ic analyz-
e s. Magne ic dipoles a e always used in bo h, he LE side and he HE side.
A common cha ac e is ic o hese elemen s is ha ions wi h bo h he same mass/cha ge and
ene gy/mass a io ha e he same beha io . In o de o dis iguish he adionuclide o in e es
om o he ions eaching he de ec o , i has o be able o measu e he kine ic ene gy o
hese.
Magne ic dipoles
A magne ic dipole is an elec omagne which gene a e a magne ic field ~
Bin a specific a ea.
Fo ions wi h a cha ge qe which a e mo ing wi h a linea momen um ~p pe pendicula o ~
B,
he magne ic o ce will make hese ions ollow a cu ed ajec o y, also pe pendicula o ~
B,
wi h a adius
ρ=p
qeB .(2.7)
I he ajec o y o hese non- ela i is ic ions o mass m(p=√2Em) mus ha e a adius ρ=R
in o de o exi he magne and con inue i s ajec o y h ough he sys em, he p oduc o he
mass/cha ge a io and he ene gy/cha ge a io mus be
Em
q2e2=B2R2
2.(2.8)
In an AMS sys em, mass selec ion is needed be o e (LE side) and a e (HE side) he ion ac-
cele a ion in he andem accele a o . Taking in o accoun he ion kine ic ene gy and cha ge
s a e in each o he sides, he mass selec ed by he LE side magne (LEM) will be
M0=eB2
LEMR2
LEM
2V0
,(2.9)
while he mass/cha ge a io selec ed by he HE side magne (HEM) mus sa is y ha
M
qV0+VT
M0
M
q+VT=eB2
HEMR2
HEM
2.(2.10)
I can be demons a ed ha his ela ionship be ween BHEM and M
qis bijec i e so, heo-
e ically, o a cons an BHEM, he use o he HEM would be enough o selec he equi ed
mass/cha ge a io. Howe e , some ene gy dispe sion o cha ge exchange p ocesses wi h he
28
2.4. ELECTROMAGNETIC KINEMATIC FILTERS
esidual gas in he HE accele a o ube could make some ions wi h di e en mass/cha ge
a io ha e he same EM
q2 ela ionship. Tha is he eason ha , o he sensi i i ies equi ed in
AMS measu emen s, an addi ional fil e , like he elec os a ic analyze , mus be used in he
HE side.
Magne ic dipoles also play an op ical ole. This ion ocusing e ec depends on he angle
be ween he beam and he bo de s o he magne [Wollnik,1999] (see Figu e 2.14). I he
en ance and he exi o he magne a e pe pendicula o he beam di ec ion (Z axis), ions
di e ging in he X axis (pe pendicula o he beam bu pa allel o i s cu a u e adius) om
a poin (objec ) be o e he magne will con e ge in a poin (image) ha is lined up wi h bo h
he objec and he cen e o cu a u e o he ion beam. Ions di e ging in he Y axis (pe pen-
dicula o bo h he beam and i s cu a u e adius), ne e heless, would no be ocused (see
Figu e 2.15) since Lo en z o ce, pe pendicula o bo h he magne ic field (in he Y di ec ion)
and he ion di ec ion, would be in he X di ec ion. E en in he bo de s, whe e he magne ic
field has a li le componen pe pendicula o hem, he magne ic o ce would s ill emain in
he X di ec ion.
I he bo de s a e no pe pendicula o he Z axis, his componen pe pendicula o he bo -
de s do has a ocusing e ec in he Y di ec ion, since he magne ic field in hem would now
ha e componen s in he 3 axis [Wollnik,1999]. This con e gen powe om each bo de is,
app oxima ely,
1
y
= an ε
R,(2.11)
whe e εis he angle which he bo de is inclined espec o he plane no mal o he beam,
(a) Bo de s pe pendicula o he beam. (b) Bo de s oblique o he beam.
Figu e 2.14: Focusing in he X axis o magne ic dipoles depending on he inclina ion o he bo de s.
29
CHAPTER 2. COMPARATIVE DESCRIPTION OF THE EXPERIMENTAL SYSTEMS
peak ene gy (27.5 MeV o 41Ca in bu ane, acco ding o SRIM), his s opping powe is ∝Z2
p·Z
E
(Be he-Bloch o mula), whe e Zpand Z a e he a omic numbe s o he ion (p ojec ile) and he
coun ing gas ( a ge ), espec i ely. Fo lowe ene gies, as hose eached a SARA and TANDY
o 41Ca (one o de o magni ude lowe ) he s opping powe becomes ∝Zp·Z
(Z2/3
p+Z2/3
)3/2√E
(LSS o mula).
In bo h cases, he highe he a omic numbe o he p ojec ile, he highe his s opping powe
is4. This can be use ul o dis inguish he adionuclide o in e es om i s s able isoba . Fo
ins ance, in Figu e 2.22 i can be seen he di e en dependence o he s opping powe wi h
he dis ance co e ed by he ion o 41Ca and 41K. Sec ioning he GIC anode in di e en ones,
he e o e, he signals om bo h isoba s can be dis inguished. I he gas ioniza ion chambe
has wo o mo e anodes, he ene gy deposi ed on each one o hem by he s able isoba
ion will be di e en ha he one deposi ed by he adionuclide o in e es . The o al ene gy
deposi ed in bo h anodes will s ill be he same o bo h ions. An example o his sepa a ion
wi h 41Ca and 41K ions wi h an ene gy o 27 MeV can be seen in Figu e 2.23.
Figu e 2.22: S opping powe o 41Ca and 41K ions wi h an ene gy o 26.5 MeV in bu ane acco ding o
SRIM. The bu ane densi y was se o s op he 41Ca ions be o e co e ing a dis ance o 500 mm.
4E en when his is no so i ial o he LSS o mula, i can be easily p o en ha he unc ion (Z) =
Z
(Z2/3+x)3/2 ulfills ha d
dZ >0.
36

2.5. GAS IONIZATION CHAMBERS
A low ene gies, hough, he high angula and ene gy loss s agglings wi hin he coun ing gas
leads o a highe ene gy s aggling o he signals om he di e en anodes, and he dis inc-
ion be ween 41Ca and 41K is no possible (see chap e 3). Fo hea ie ions, he ansmission
o he kine ic ene gy o he whole gas a om/molecule (nuclea s opping powe ) becomes im-
po an a low ene gies.
Figu e 2.23: De ec o signals in a 41Ca measu emen wi h he 5 MV AMS sys em a SUERC, Glasgow.
Pic u e om Schulze-K¨
onig e al. [2010].
37
CHAPTER 2. COMPARATIVE DESCRIPTION OF THE EXPERIMENTAL SYSTEMS
2.5.1 The compac gas ioniza ion chambe a SARA
The compac GIC a SARA, which is p esen ed in Figu e 2.24, was designed a he ETH LIP,
and i is loca ed a e he ESA. I has wo anodes, each one o hem 5 cm long. Compa ing
wi h he old HVE de ec o (see Figu e 2.25), he lowe olume o he compac GIC educe he
elec onic noise and he design minimizes he non-ac i e a ea.
Figu e 2.24: Scheme o he compac GIC a SARA. Pic u e om Scognamiglio e al. [2016].
Figu e 2.25: Compa ison be ween he old HVE GIC (le ), and he new compac GIC ( igh ).
38
2.5. GAS IONIZATION CHAMBERS
2.5.2 The e ac able gas ioniza ion chambe a TANDY
The e ac able GIC a TANDY (see Figu e 2.26) can be inse ed a e he ESA, in he image
o he 3+ cha ge s a e. I has wo anodes, 3 cm long each one. Fo o he adionuclides, a
compac GIC simila o he one a SARA placed a e he HEM 2 is used [M¨
ulle e al.,2012].
Figu e 2.26: Pic u e (le ) and scheme ( igh ) o he e ac able GIC used in he 41Ca measu emen s
wi h TANDY. Pic u e om S ocke e al. [2005]. Scheme adap ed om a figu e p esen ed on S ocke
[2006].
39
CHAPTER 2. COMPARATIVE DESCRIPTION OF THE EXPERIMENTAL SYSTEMS
40
Chap e 3
41Ca AMS a low ene gies
In his chap e , he main challenges ela ed o 41Ca AMS a low ene gies a e de ailed, and
some expe imen s ela ed o his pe o med du ing his hesis wi h bo h sys ems, SARA and
TANDY, a e also p esen ed. The mos impo an esul s we e also published in Vi o-Vilches
e al. [in p ess].
3.1 The 41K in e e ence
Kinema ic fil e s commonly used a AMS only selec ions wi h a ce ain mass/cha ge a io.
Ne e heless, he measu emen o he ene gy o he ions in he de ec o allows he disc imi-
na ion o ions wi h di e en mass bu he same M/q. The e o e, he main challenge o 41Ca
AMS measu emen s is dealing wi h i s s able isoba , 41K. As i is s a ed in chap e 1, la ge AMS
acili ies deal wi h his po en ial in e e ence using he di e en s opping powe in he gas o
he GIC [Hosoya e al.,2017]. A low ene gies, on he o he hand, sca e ing p ocesses wi h
he gas i sel p oduce an impo an angula s aggling. This s aggling, oge he wi h he one
on he ene gy loss i sel , also en ails a lowe ene gy esolu ion, since di e en ajec o ies
lead o di e en ene gy losses in each one o he anodes.
Figu e 3.1, o ins ance, shows a simula ion pe o med wi h SRIM [Ziegle ,n.d.] o he be-
ha iou o 41Ca ions in a GIC wi h 2 anodes: sca e ed ions will co e a longe pa h in he fi s
hal o he de ec o , lea ing mo e ene gy in he fi s anode. This way, a lowe enegies, 41K
ions sca e ed an angle θKcould lea e mo e ene gy in he fi s anode han some 41Ca ions
sca e ed and angle θCa < θK, e en when he s opping powe o he po assium ions is lowe .
The simula ed spec a co esponding o 41Ca and 41K ions a di e en ene gies in a 2-anode
41

CHAPTER 3. 41CA AMS AT LOW ENERGIES
GIC de ec o like he one p esen ed in Figu e 3.1 a e p esen ed in Figu e 3.2. A ela i ely
high ene gies, he spec um shows 2 coun a eas: 41K ions deposi less ene gy han 41Ca
ions in he fi s anode (ΔE) and mo e ene gy in he second one (E es). A lowe ene gies, like
hose eached wi h he 1 MV AMS sys em a CNA, he 41K and 41Ca coun s a e mixed in he
spec um.
The highe ela i e ene gy and angula s agglings a low ene gies also a oids he use o ech-
(a) 26.5 MeV (b) 2.3 MeV
Figu e 3.1: SRIM simula ion o he ajec o ies o se e al 41Ca ions a wo di e en ene gies in a GIC
de ec o wi h a 50 nm silicon ni ide window, filled wi h isobu ane and wi h 2 anodes 2.5 cm long
each. Isobu ane p essu es we e adjus ed on each case o s op he ions jus be o e he GIC end.
(a) 26.5 MeV (b) 2.3 MeV
Figu e 3.2: Spec a simula ion wi h SRIM o an equal numbe o coun s o 41Ca and 41K a di e en
ene gies. The ΔE axis ep esen s he ene gy deposi ed in he fi s anode; he E es axis, he ene gy
deposi ed in he second one.
42
3.1. THE 41K INTERFERENCE
niques based on he di e en ene gy loss on silicon ni ide oils, such as he ΔTOF echnique
[Vockenhube e al.,2005]. As i can be seen in Table 3.1, he ene gy s aggling in oduced
by hese oils in 41K and 41Ca ions wi h an ene gy o 3.4 MeV (41Ca3+ o a e minal ol age o
1 MV) would be, in mos o he cases, highe , han he di e ence on he ene gy loss. E en a
he e y bes , his di e ence is only sligh ly highe han he ene gy loss s aggling. Fo hicke
oils, besides, he e would be an addi ional con ibu ion o he ene gy s aggling because o
he angula one.
The use o CaH2, selec ing he (CaH3)-ion in he LE side, would educe his po assium in e -
e ence because o he ins abili y o he (KH3)-ion. This only lea es he mino con ibu ion
o he 41K in e e ence o he (41KH2)-ions whe e one o he hyd ogen a oms is a deu e ium
one. Howe e , because o he hyg oscopic cha ac e o CaH2, i s chemical p epa a ion and
handling is p oblema ic. This makes his chemical o m no sui able o applica ions which
in ol e la ge numbe o samples, like hose ypically pe o med wi h low ene gy AMS. The
possibili y o ex ac ing he (CaH3)-ion om o he ma ices, like he mix o calcium sal s wi h
i anium hyd ide, ha e been es ed in he pas [Fink e al.,1990;Baba e al.,1997], and e-
cen ly he e a CNA, in he las yea o his hesis. Un o una ely, only uns able cu en s o
ew nA o (40CaH3)-we e measu ed in all hese cases. Mo e in o ma ion abou hese expe i-
men s is p o ided in Appendix B.
The e o e, a SARA and TANDY, CaF2samples, ex ac ing he (CaF3)-ion in he LE side, a e
used, as i is also done in mos o he la ge AMS acili ies. The s abili y o his compound
makes i e y sui able o handling o a la ge numbe o samples. The nega i e poin , is ha
he 41K in e e ence is inc eased by 2 o de s o magni ude [Fink e al.,1990], his is, ypically
in he 10-11 le el. Ne e heless, mos o he samples measu ed wi h hese sys ems p esen
41Ca/40Ca a ios highe han 10-10. Fi s 41Ca measu emen s a TANDY we e pe o med ea -
ing his in e e ence as pa o he backg ound [Schulze-K¨
onig e al.,2010]. Howe e , since
he ela i e con en o po assium could di e om one sample o ano he , an es ima ion o
Table 3.1: Ene gy loss s aggling and a e age ene gy di e ence wi h 41K ions o 3.4 MeV 41Ca ions
a e a silicon ni ide window as a unc ion o he oil hickness. The ene gy s aggling ( ull wid h a
hal maximum o he Gaussian dis ibu ion) has been calcula ed wi h he semi-empi ical Sun’s o mula
[Sun e al.,2007;M¨
ulle ,2009]. A e age ene gies a e he oil we e ob ained om SRIM simula ions.
Foil hickness (nm) 41K-41Ca ene gy di e ence (keV) Ene gy loss s aggling (keV)
50 11 23
100 30 32
250 70 51
43
CHAPTER 3. 41CA AMS AT LOW ENERGIES
he 41K in e e ence would be mo e sui able. The bes way o es ima e his in e e ence is he
measu emen o he o he s able iso ope o po assium, he 39K [Vockenhube e al.,2015].
Po assium has 2 s able iso opes: 39K, wi h a na u al iso opic abundance o 93.2581%; and
41K, wi h a na u al iso opic abundance o 6.7302% [Na ional Nuclea Da a Cen e ,n.d.]. Apa
om sligh a ia ions, hese iso opic abundance a e always he same, independen ly om
he sample, so he 41K/39K a io should be 0.072167 in any ma e ial. The e o e, since he
41K/40Ca in e e ence is ypically in he 10-11-10-10 ange, 39K/40Ca a ios a e in he le el o
10-10-10-9, a ios e y sui able o be measu ed by AMS. Consequen ly, measu ing no only
41M/40Ca1, bu also 39K/40Ca a ios, he ac ual 41Ca/40Ca a io could be calcula ed as
R41,K−co =R41 −αR39,(3.1)
whe e R41 deno es he unco ec ed 41M/40Ca a io; R39, he 39K/40Ca a io; and α, he mea-
su ed 41K/39K a io. Sligh di e ences on he s ippe and op ical ansmission be ween bo h
po assium iso opes would make αno being exac ly he na u al 41K/39K a io. This expe imen-
al 41K/39K a io (K-co ec ion ac o ) can be calcula ed pe o ming a linea eg ession whe e
he alues o he dependen a iable xia e he 39K/40Ca a ios om di e en measu emen s
o he blank samples, and hose o he independen a iable yi, he 41M/40Ca a ios. The ex-
pe imen al 41K/39K a io, α, is he slope o his linea eg ession. Mo e in o ma ion abou he
ela ed da a analysis is p esen ed in sec ion 3.4.
In o de o sequen ially injec 41M and 39K ions in o he de ec o , a oiding changes in he
magne ic field o he HEM, he accele a o e minal and ESA ol ages ha e o be changed
sequen ially, as i is done in ac inide measu emen s [Chamizo e al.,2008a;Fifield,2008].
The e minal ol age o injec he 39Kqions should make i s ene gy 41
39 imes ha one o he
41Mqions when hese a e injec ed. Howe e , his swi ching has o be much slowe han he
one ypically used o injec he nega i e ions in o he accele a o , because o bo h: he need
o coun ing imes o se e al seconds o bo h ions, and he ela i ely slow change o he
e minal ol age.
The mos popula ed s a es a he s ipping ene gies achie ed wi h hese compac AMS sys-
ems a e he 2+ and 3+ s a es. Using he la e , he e is no need o using ela i ely high
s ippe p essu es in o de o des oy he molecula backg ound, since molecula ions wi h
cha ge s a e 3+ a e uns able. Up un il 2018, i was hough ha he K-co ec ion would no
allow he use o his cha ge s a e: he de ec ion o 39K3+ would no be possible because o he
a es o 13C+we e expec ed o be so high ha hey would sa u a e he de ec o . Ou ecen
s udies a SARA, ne e heless, show ha 13C+ a es a e no so high, and he K-co ec ion can
1Whe e 41M deno es all he ions wi h mass 41, his is, 41M≡41K+41Ca
44
3.2. STUDIES OF THE SOURCES OF THE 41K INTERFERENCE
be also used when he 3+ s a e is selec ed (see chap e 4).
Using he K-co ec ion, he 39K/40Ca a io becomes a ele an ac o in ol ed in he final
unce ain y o he esul s. Besides, since 39K and 41M a e no measu ed simul aneously,
bu sequen ially, huge changes in he 39K/40Ca a io should be a oided. I is clea , hen,
ha unde s anding he ac o s ela ed o he p oduc ion o he 41K in e e ence is ex emely
impo an o 41Ca AMS a low ene gies. In he nex sec ions, se e al expe imen s ela ed
o his in e e ence, pe o med du ing his hesis wi h bo h sys ems, SARA and TANDY, a e
p esen ed.
3.2 S udies o he sou ces o he 41K in e e ence
Du ing he fi s es s o 41Ca AMS using calcium fluo ide, i was hough ha he (KF3)-ion
is a o ally uns able ion, and he 41K in e e ence is no caused by he p oduc ion o (41KF3)-
ions, bu om o he molecula ions whe e 41K is p esen and wi h he same mass, 98 u [Fink
e al.,1990]. Zhao e al. [2010] p o ed, hough, ha F+ions we e p oduced om he s ipping
o ions wi h a mass o 96 u, consequen ly iden i ying hose ions as (39KF3)-, confi ming he
exis ence o he ion. The sligh s abili y om he (KF3)-ion, he fluo ine p esen in he CaF2
sample, and he ela i ely high ace amoun o po assium ypically ound in Cs, make possi-
ble he p oduc ion o (41KF3)-ions, e en in he heo e ical case whe e he e is no po assium
ace amoun wi hin he ca hode.
The exis ence o he (KF3)-ion, ne e heless, does no mean ha his ion is he only nega i e
ion leading o 41K in e e ence. The ela i ely high mass o he (41KF3)-ion would allow, also,
he o ma ion o o he se e al di e en ions wi h he same mass,98 u, and whe e po assium
is p esen . Fo ins ance, a common obse a ion du ing expe imen s du ing his hesis is he
de ec ion o ela i ely high a es o bo h 41K and 39K ions when ma e ial wi h no fluo ine
was spu e ed. This has been sys ema ically obse ed wi h bo h sys ems, SARA and TANDY.
Zhao e al. [2010] p oposed he heo y ha ano he impo an nega i e ion ela ed o 41K
in e e ence would be he (41K57Fe)-ion, since i has also, a mass o 98 u. They also checked
ha adding i on o he samples inc eased he 41K a es. This would explain he high 41K a es
when me al is spu e ed.
In o de o p o e ha his heo y is co ec , and ha hei inc ease on he 41K a es was no
due o a highe ace amoun o po assium in hei i on ma e ial, an expe imen was pe -
o med wi h he TANDY sys em du ing hw s ay a he ETH LIP. This expe imen consis ed in
he measu emen o he 41M2+,39K2+,57Fe2+ and 19F+ a es in a blank sample and an alu-
minum dummy. 41M2+ and 39K2+ a es we e measu ed sequen ially using he slow swi ching
45
CHAPTER 3. 41CA AMS AT LOW ENERGIES
41Ca AMS a low ene gies, as i was s a ed in sec ion 3.1, he only way o es ima e he whole
con ibu ion o his in e e ence is du ing he da a analysis. This K-co ec ion becomes one
o he key s eps o his analysis.
The specific ea u es o 41Ca AMS da a analysis a low ene gies depend on he sys em used,
since a ailable in o ma ion sligh ly depends on he elemen s p esen on he sys em; o in-
s ance, he possibili y o measu ing he (40CaF3)-cu en on he LE side a TANDY makes
possible using disca d c i e ia based on he s ippe ansmission. Howe e , he basic s eps
a e he same.
In o de o educe he ime consumed by his da a analysis, 2 p og ams we e w i en in
FORTRAN code du ing his hesis: one o he measu emen s a TANDY, w i en du ing he 4
mon hs s ay a ETH LIP; and ano he one, based on i , o he measu emen s a SARA. The
code changes a e only ela ed o hese sligh di e ences in he in o ma ion p o ided by each
AMS sys em. Mo e in o ma ion abou hese p og ams is p o ided in Appendix C.
In his sec ion, an o e iew o he s eps ollowed by hese p og ams is p esen ed.
3.4.1 AMS measu emen
In o de o unde s and he da a analysis o he measu emen s, i is e y impo an o know
how he measu emen s a e pe o med. Figu e 3.6 shows he ypical measu emen p ocedu e
in 41Ca AMS a low ene gies. Each sample i3is in oduced in he machine a numbe n uns o
uns. Du ing each one o hese (i, j) uns, bo h ions, 41M and 39K, a e sequen ially measu ed
using di e en sequence imes 41 and 39 o each one. Bo h ions a e measu ed he same
numbe o sequences nseqs. Each 41M sequence (i, j, k), excep he fi s one om each un,
(i, j, 1), is p eceded by he 39K sequence (i, j, k −1) and ollowed by he 39K sequence (i, j, k).
Typically, 5 sequences o 41 = 30 s o 41M, and 39 = 10 s o 39K, a e used in he measu e-
men s a SARA and TANDY. This ime can be lowe i a ce ain numbe o ion coun s in he
de ec o is eached.
The measu emen o he 40Ca cu en is pe o med in a di e en way a SARA han a TANDY.
In he fi s case, a 40Ca sequence is added jus a e he 41M one, using he same e minal
and ESA ol age han hose used in 39K sequences; he eason o his will be explained in
chap e 4. In he second case, he 40Ca cu en is measu ed using he as swi ching o he
LEM bounce du ing he 41M sequence.
Theo e ically, a sys em wi h 2 FCs in he HE side and wi h he p ope so wa e could mea-
3This includes no only he samples whose 41Ca/40Ca we wan o know, bu also blank and s anda d samples.
52

3.4. DATA ANALYSIS FOR 41CA AMS AT TANDY AND SARA
su e he 40Ca cu en du ing bo h, he 41M and he 39K sequences, and i would be possible
o di ec ly ge he 41M/40Ca a io Rseq
41 (i, j, k)and he 39K/40Ca a io Rseq
39 (i, j, k) om each
sequence. A SARA and TANDY, he 40Ca cu en is assumed o change linea ly om one
sequence o he nex in o de o calcula e hese a ios.
I he unce ain ies o he 40Ca cu en and he measu ing ime a e dis ega ded, he s a is ical
unce ain y om each o hese Rseq
A(i, j, k) a ios, whe e Ais ei he 41 o 39, is
Useq
A(i, j, k) = Rseq
A(i, j, k)
pNA(i, j, k),(3.2)
whe e NA(i, j, k)is he o al numbe o ions wi h mass Ade ec ed in ha sequence4.
Figu e 3.6: Typical scheme o a 41Ca measu emen wi h a compac AMS sys em implemen ing he
K-co ec ion.
4The unce ain y om NA(i, j, k)is pNA(i, j, k)
53
CHAPTER 3. 41CA AMS AT LOW ENERGIES
3.4.2 Calcula ion o he expe imen al 41K/39K a io (K-co ec ion ac o )
In o de o ge he expe imen al 41K/39K a io α(o K-co ec ion ac o ), he 41M/40Ca and
39K/40Ca a ios om each un o he blank samples a e used in o de o pe o m he linea
eg ession. These a ios a e calcula ed as
R un
A(i, j) =
nseqs
P
k=1
wseq
A(i, j, k)·Rseq
A(i, j, k)
nseqs
P
k=1
wseq
A(i, j, k)
,(3.3)
whe e wseq
A(i, j, k)is a weigh unc ion dependen on he 40Ca cu en in ha sequence IA(i, j, k)
and he sequence ime A(i, j, k), defined as
wA(i, j, k)≡ IA(i, j, k)· A(i, j, k)·Cseq
A(i, j, k).(3.4)
This weigh unc ion makes mo e impo an hose sequences wi h highe s a is ics. The
Cseq
A(i, j, k)is a logical ac o which is 0 i any o he selec ion c i e ia (minimum 40Ca cu -
en , o ins ance) is no passed, and 1 o he wise. The common selec ion c i e ia used a
SARA and TANDY a e p esen ed in subsec ion 3.4.5.
Once all he R un
41 (i, j)and R un
39 (i, j) a ios om he blank samples ha e been calcula ed, a
linea eg ession is pe o med by he me hod o leas squa es, using he R un
41 (i, j) a ios as
independen a iable x, and R un
39 (i, j)as dependen a iable y5. The slope om his linea
eg ession is he K-co ec ion ac o α±U(α).
3.4.3 Applica ion o he K-co ec ion and sample a e age
Since each 41M measu emen sequence is pe o med be ween wo 39K sequences, he 39K/40Ca
a io which should be applied o co ec he Rseq
41 (i, j, k) a io is no exac ly Rseq
39 (i, j, k), nei-
he Rseq
39 (i, j, k −1). Ne e heless, since he expec ed changes in his 39K/40Ca a io a e sligh
in he ime scale be ween wo sequences, du ing he 41M sequence i can be supposed o be
he a e age om he p e ious and he ollowing one. The K-co ec ed 41Ca/40Ca a io om
each sequence, he e o e, is
5[x;y](i, j) = [R un
41 (i, j); R un
39 (i, j)]
54
3.4. DATA ANALYSIS FOR 41CA AMS AT TANDY AND SARA
Rseq
K−co ec ed(i, j, k) =
=Rseq
41 (i, j, k)−α·Rseq
39 (i, j, k −1) + Rseq
39 (i, j, k)
2
:k= [2, nseqs].(3.5)
The unce ain y om his K-co ec ed a io is calcula ed as
Useq
K−co ec ed(i, j, k) =
u
u
u
u
u
[Useq
41 (i, j, k)]2+hα·Rseq
39 (i, j, k −1) + Rseq
39 (i, j, k)
2pN39(i, j, k −1) + N39(i, j, k)i2+
+hU(α)·Rseq
39 (i, j, k −1) + Rseq
39 (i, j, k)
2i2.(3.6)
Using he same weigh unc ions w41(i, j, k)used o calcula e he unco ec ed 41M/40Ca a ios
om each blank sample un, he K-co ec ed 41Ca/40Ca a ios om each un can be calcula ed
o e e y sample as
R un
K−co ec ed(i, j) =
nseqs
P
k=2
wseq
41 (i, j, k)·Rseq
K−co ec ed(i, j, k)
nseqs
P
k=2
wseq
41 (i, j, k)
.(3.7)
and he s anda d de ia ion is
S un
K−co ec ed(i, j) =
=
u
u
u
u
u
u
Ne seqs(i, j)
Ne seqs(i, j)−1·
nseqs
P
k=2
wseq
41 (i, j, k)·[Rseq
K−co ec ed(i, j, k)−R un
K−co ec ed(i, j)]2
nseqs
P
k=2
wseq
41 (i, j, k)
,(3.8)
whe e Ne seqs(i, j)is defined as he numbe o sequences in he un (i, j)whe e Cseq
41 (i, j, k) =
1. The s anda d de ia ion is only calcula ed i Ne seqs(i, j)>1.
Assuming a i ual numbe o 41Ca coun s in each sequence equal o hRseq
K−co ec ed(i,j,k)
Useq
K−co ec ed(i,j,k)i2, he
unce ain y o R un
K−co ec ed(i, j)is
U un
K−co ec ed(i, j) = R un
K−co ec ed(i, j)
snseqs
P
k=2 hCseq
41 (i, j, k)·Rseq
K−co ec ed(i,j,k)
Useq
K−co ec ed(i,j,k)i2
,(3.9)
The weigh unc ion w un(i, j) om each un is defined as
55
CHAPTER 3. 41CA AMS AT LOW ENERGIES
w un(i, j)≡(
0i C un(i, j)=0
hR un
K−co ec ed(i, j)
U un
K−co ec ed(i, j)i2i Ne seqs(i, j)=1
minnhR un
K−co ec ed(i, j)
S un
K−co ec ed(i, j)i2;hR un
K−co ec ed(i, j)
U un
K−co ec ed(i, j)i2oi Ne seqs(i, j)>1
,(3.10)
whe e, e iden ly, a leas one o he c i e ia making C un(i, j) = 0 is ha Ne seqs(i, j) = 0.
This way, hose uns whe e he beha io was mo e s able and, he e o e, wi h lowe ela i e
unce ain ies, ake mo e impo ance. The sample a e age o he K-co ec ed 41Ca/40Ca a io
RK−co ec ed(i)and i s s anda d de ia ion SK−co ec ed(i)a e calcula ed om he esul s o he
uns in an analogous way as hese esul s o he uns a e calcula ed om he esul s o he
sequences. The unce ain y is calcula ed as
UK−co ec ed(i) = RK−co ec ed(i)
sn uns
P
j=1
w un(i, j)
.(3.11)
3.4.4 Blank and s anda d co ec ion
The nex co ec ions a e common o almos all he AMS measu emen s, no ma e he sys em
o he adionuclide. The backg ound Bis calcula ed as he non-weigh ed a e age o he K-
co ec ed a ios om all he blank a ge s. This way, he blank co ec ed a io om each
sample is
RBLK−co ec ed(i) = RK−co ec ed(i)−B.(3.12)
In his case, no weigh unc ion is added since no eason makes some blank samples mo e
scien ifically ele an han o he s. The unce ain y om his a io is
UBLK−co ec ed(i) =
=(p[UK−co ec ed(i)]2+ [S(B)]2i Zs
0
χ2[x;Ne uns(i)−1]dx < 0.95
p[SK−co ec ed(i)]2+ [S(B)]2i Zs
0
χ2[x;Ne uns(i)−1]dx ≥0.95
,(3.13)
whe e S(B)is he s anda d de ia ion o he backg ound, Ne uns(i)is he o al numbe o uns
om he sample iwi h w un(i, j)>0, and s≡[SK−co ec ed(i)
UK−co ec ed(i)]2·[Ne uns(i)−1].
56
3.4. DATA ANALYSIS FOR 41CA AMS AT TANDY AND SARA
The nex s ep is he co ec ion o he s ipping and op ical ac iona ion, o s anda d co ec-
ion. The Rexp(`) om each s anda d ma e ial `, wi h a nominal 41Ca/40Ca a io o Rnom(`), is
calcula ed as he non-weigh ed a e age o he BLK-co ec ed a ios om all he a ge s o ha
s anda d ma e ial. I ha a e age has a s anda d de ia ion S a io(`), he nominal/expe imen al
ac o (`) = Rnom(`)
Rexp(`)will ha e a s anda d de ia ion σ (`) = S a io(`)
Rnom(`)· (`).
Since, ypically, se e al di e en s anda d ma e ials a e used du ing measu emen s, he final
nominal/expe imen al ac o Fapplied o he s anda d co ec ion will be he non-weigh ed
a e age o he di e en (`) ac o s. No weigh unc ion is applied because he same easons
as hose o blank co ec ion: no eason makes some s anda d samples mo e scien ifically
ele an han o he s. I his p oduces a s anda d de ia ion σ(F), he final unce ain y o he
s anda d co ec ion ac o Fis
U(F) = max nσ(F); F
P
`
[ (`)
σ (`)]2o.(3.14)
The final 41Ca/40Ca om each sample is, he e o e,
RSTD−co ec ed(i) = F ·RBLK−co ec ed(i),(3.15)
wi h an unce ain y
USTD−co ec ed(i) = p[U(F)·RBLK−co ec ed(i)]2+ [F ·UBLK−co ec ed(i)]2.(3.16)
3.4.5 Common selec ion c i e ia used a SARA and TANDY
The only selec ion c i e ia used only in 41Ca measu emen s a TANDY, and no a SARA, a e
hose ela ed o he s ippe ansmission, since ha in o ma ion is a ailable in he mea-
su emen s wi h ha sys em. Typically, sequences (i, j, k)whe e he s ippe p essu e has a
di e ence o 5% o highe wi h ei he he a e age anmission h ough he whole measu e-
men , o he a e age ansmission du ing ha un (i, j), a e disca ded. Ne e heless, hese
possible changes in he ansmission a e no expec ed o be ele an a SARA, whe e he
op ical losses a e lowe o 41Ca.
The ewi h, he es o selec ion c i e ia, used also in he measu emen a SARA, a e p esen ed,
and he easons o apply hem a e explained.
57

CHAPTER 3. 41CA AMS AT LOW ENERGIES
Igno ing fi s un
As i can be seen in sec ion 3.3, du ing he fi s spu e ing minu es he ion sou ce cu en
ou pu ypically p esen s a qui e uns able beha iou . I is ad isable no o use hese esul s
in he da a analysis, so a common selec ion c i e ion is ha
C un(i, 1) = 0 ∀i. (3.17)
Minimun 40Ca cu en ou pu
A e y low ion sou ce cu en oupu wi hin a sequence (i, j, k)can be symp oma ic o a em-
po al ailu e, which makes he esul s o he a ios om ha sequence no ep esen a i e o
he eali y. Besides, a low cu en ou pu is a common ea u e o ime in e als whe e ha
same ou pu is mo e uns able; o ins ance, when he a ge is s a ing o un ou o sample
ma e ial. In o de o wo k wi h sequences whe e he beha iou o his ou pu is s able, a
minimum 40Ca cu en Imin can be es ablished. Fo any sequence (i, j, k) : IA(i, j, k)<Imin,
he weigh unc ion will be se as 0. This minimum cu en should be es ablished aking in o
accoun he a e age cu en s obse ed du ing ha measu emen , as well as he ypical cu -
en s obse ed his o ically wi h ha AMS sys em. A SARA o ins ance, we ypically use a
minimum 40Ca2+ cu en o 20 nA.
Maximum ion a e
GIC de ec o s ypically do no wo k p ope ly when he o al ion a es a e oo high. In hese
cases, mainly 2 e ec s educe he de ec ion efficiency:
•The pile-up, o combina ion o he signals om 2 o mo e ions, p oducing a signal
p ope o an ion wi h a highe ene gy.
•The sa u a ion o he elec onic chain ha ans o m he signal p oduced by he ion in o
a digi al one.
I is sui able, he e o e, o es ablish a maximum ion a e, so he sequences whe e ha 41M
o 39K a es a e highe a e disca ded. To es ablish his limi , a s udy o he ela ionship
be ween he ion a e and he GIC de ec o efficiency should be pe o med. Mo e in o ma ion
abou he efficiency o he SARA GIC and he ypical maximum a es in ou measu emen s is
p esen ed in chap e 4.
58
3.4. DATA ANALYSIS FOR 41CA AMS AT TANDY AND SARA
Maximum 39K/40Ca a io
In o de o ake he measu emen s o each sample whe e he 41K in e e ence was he
leas , and o a oid he uns able beha iou o he 39K/40Ca a io when i si highe , a maximal
39K/40Ca a io RMAX
39 can be se , so all he uns (i, j)whe e he a e age 39K/40Ca a io R un
39 (i, j)
is highe han RMAX
39 a e disca ded. This also a oids he ex apola ion o he K-co ec ion ac-
o o R un
39 (i, j)highe han hose ound in blanks samples, p o ided ha his maximum
39K/40Ca a io is lowe o equal han he highes 39K/40Ca a io ound in blank samples. This
maximum 39K/40Ca a io is ypically se in 1-2 ×10-9 ange, so he 41K/40Ca in e e ence is
expec ed o s ay in he 10-11-10-10 ange.
59
CHAPTER 3. 41CA AMS AT LOW ENERGIES
60
Chap e 4
Se up o he 1 MV AMS sys em a CNA
o 41Ca measu emen s
In his chap e , se e al expe imen s pe o med h oughou his hesis ela ed o he pe -
o mance pa ame e s o 41Ca measu emen s a SARA a e de ailed. These pa ame e s a e
ela ed o he final efficiency o he 41Ca ion de ec ion in o de o p o ide a s ong s a is ics,
and he final 41Ca/40Ca backg ound and sensi i i y. The mos impo an esul s p esen ed
he e we e also p esen ed in Vi o-Vilches e al. [2017]. The specific pa ame e s s udied and/o
op imized ha e been:
•The (40CaF3)-cu en ou pu wi h he SO-110B ion sou ce (sec ion 4.1).
•The ansmission in he He gas s ippe o he cha ge s a es 2+ and 3+ (sec ion 4.2).
•The de ec ion efficiency and signal iden ifica ion ΔE-E es spec a o he compac GIC
(sec ion 4.3).
•The linea co ela ion be ween 41K and 39K ions and he final 41Ca backg ound a e he
K-co ec ion (sec ion 4.4).
•The op ical ansmission o 41Ca h ough he HE spec ome e by he compa ison o
nominal and expe imen al a ios in s anda d samples (sec ion 4.5).
A compa ison o hese pe o mance pa ame e s wi h hose a TANDY is also p esen ed in
sec ion 4.6. In o de o comple e his compa ison, measu emen s o he ETH LIP in-house
41Ca s anda ds [Ch is l e al.,2013] and se e al samples o biomedical in e es p e iously
measu ed a TANDY a e also shown.
61
CHAPTER 4. SETUP OF THE 1 MV AMS SYSTEM AT CNA FOR 41CA MEASUREMENTS
highe mean cha ge s a e wi h He as s ipping gas o low ene gies ag ees wi h which has
been obse ed in u anium [Wi kowe and Be z,1973]. On he o he hand, semi-empi ical
o mulas applied o highe ene gies, like he one om Schiwie z and G ande [2001]3, p edic
he opposi e: ha , a la ge acili ies, hese mean cha ge s a es would be highe using A as
s ipping gas han using He. The e o e, i is clea ha Schiwie z-G ande o mula canno be
applied a low ene gies: o 40Ca wi h an ene gy o 400 keV, he mean cha ge s a e acco ding
o his o mula would be 0.9, less han he hal o he one we calcula e. The e o e, he mea-
su emen s o his kind o pa ame e s a e uly impo an in o de o p o ide da a abou he
s ipping p ocess a low ene gies.
Figu e 4.5: Cha ge s a e dis ibu ion o he (40CaF3)-→40Caqs ipping p ocess in He a 225 keV and
400 keV. The cha ge s a e ac ion o he 4+ s a e was supposed o be 0 a TANDY.
4.3 40Ca cu en measu emen wi h he FC3 and 41Ca de ec ion
wi h he compac GIC
A special ea u e o he 41Ca measu emen s a SARA is ha he measu emen o he 40Ca
cu en wi h he FC3 canno be pe o med when he accele a o e minal ol age is he same
3This o mula can only be applied o 40Ca wi h ene gies ≥800 keV.
68

4.3. 40CA CURRENT MEASUREMENT WITH THE FC3 AND 41CA DETECTION WITH THE
COMPACT GIC
which is used when he 41Ca is being de ec ed wi h he GIC. The eason is he geome y o
his FC. As i can be seen in Figu e 4.6, he hole o 129I measu emen s in ol es ha , when
he e minal ol age is uned o de ec 41Ca in he GIC, pa o he FC measu ing 40Ca cu en
would block he way o 41Ca ions o he ESA. Fo una ely, his ea u e is no p oblem, since
he FC can be placed whe e 40Ca is de ia ed when 39K is being de ec ed. This way, 40Ca
cu en is measu ed du ing 39K sequences, ins ead o du ing 41Ca sequences. In o de o ge
he 41Ca/40Ca a ios o each sequence, he 40Ca cu en is supposed o be he a e age om
hose measu ed jus be o e and jus a e ha sequence.
As expec ed, he 41Ca ΔE-E es spec um a he GIC when he 2+ s a e is selec ed in he HE side
is e y simila o he one a TANDY o he same cha ge s a e [Vockenhube e al.,2015] being
82Se4+ he only o he ion eaching he de ec o . Figu e 4.7 shows how, when he isobu ane
p essu e is low enough o le he 41Ca ions each he second anode, he 82Se4+ ions lose
only a pa o i s kine ic ene gy in he gas, ma ching up wi h he spec um p edic ed by SRIM
simula ions. Bo h signals can s ill be dis inguished in he spec um, bu he e exis s a emo e
possibili y o some 82Se ions in e e ing wi h he 41Ca signal.
Ne e heless, since a hese ene gies he signals om 41K and 41Ca canno be dis inguished,
he e is no ad an age in using his ΔE-E es spec um. In o de o ha e a 82Se4+ peak wi h
well-defined ene gy, he gas p essu e inside he GIC can be inc eased so ha bo h ions a e
s opped be o e eaching he second anode. This way, he signal om he fi s anode would be
di ec ly ela ed o he o al ene gy wi h which hese ions each he de ec o . As i is shown in
Figu e 4.7, wi h his configu a ion, he 82Se4+ peak p esen s a well-defined deposi ed ene gy
o 2 imes he ene gy o 41Ca2+, as expec ed. This configu a ion is, he e o e, he one used
(a) 129I (b) 41Ca (c) 39K
Figu e 4.6: Rela i e posi ion o he FC3 depending on he iso ope whose cu en is measu ed and he
adionuclide which is uned o c oss o he ESA.
69
CHAPTER 4. SETUP OF THE 1 MV AMS SYSTEM AT CNA FOR 41CA MEASUREMENTS
du ing ou ine 41Ca measu emen s a SARA.
The 41Ca ΔE-E es spec um when he 3+ s a e is used (see Figu e 4.8) does no p esen many
o he peaks which can be seen a TANDY o ha cha ge s a e, whe e ions wi h di e en M
q
a io each he de ec o because o he wo se op ical anspo h ough he HE sp ec ome e .
A SARA, only 82Se6+ ions each he de ec o wi h 41Ca, as i happens when he s a e 2+ is
used. The ela i e p oduc ion o 82Se6+ ions is much lowe in his case, p esumably because
o a much lowe s ippe ansmission o such a high cha ge s a e.
The in ege M
q a io o he 39K3+ ion implies he de ec ion o se e al o he ion species, being
he mos impo an o hem 13C+,26Mg2+ and, when coppe a ge holde s a e used, 65Cu5+.
13C+ a es when spu e ing a blank sample in coppe a ge s a e in he 1000-2000 s-1 ange;
(a) Expe imen al ΔE-E es (b) Simula ed ΔE-E es
(c) Expe imen al o al E
Figu e 4.7: Expe imen al a simula ed spec a o 41Ca2+ and 82Se4+ ions eaching he compac GIC a
SARA. In he expe imen al ΔE-E es spec um, he dashed lines ep esen s he a e age ene gy deposi ed
by he 41Ca ions, and wice ha ene gy. In he In he SRIM simula ion, each do ep esen s a single
ion.
70
4.3. 40CA CURRENT MEASUREMENT WITH THE FC3 AND 41CA DETECTION WITH THE
COMPACT GIC
his is se e al o de s o magni ude below he 13C+ ound when ying o de ec 26Al2+ [M¨
ulle
e al.,2015]. This is no o ally unexpec ed, since he o ma ion o nega i e ions whe e 13C is
p esen and wi h a mass o 96 u, when injec ing (39KF3)-, should no be as impo an as he
o ma ion o (13C2)-and (12C13CH)-clus e s, when injec ing 26Al-.13C+and 26Mg2+ when Al
a ge holde s we e used we e much highe , leading o impo an pile-up in e e ing wi h 39K
coun ing egion.
The ela ionship be ween he o al 41M a e and he compac GIC efficiency has also been
s udied o cha ge s a e 2+, in o de o p o ide a eliable maximum a e o he da a analysis
(see chap e 3). This was done by connec ing a pulse gene a o o he de ec o , and com-
pa ing he measu ed a e om his pulse wi h i s nominal alue, p e iously measu ed wi h
an osciloscope. Figu e 4.9 shows ha , o 41M a es lowe han 1000 s-1, he efficiency keeps
highe han 90%. Fo highe a es, efficiency becomes in e sely p opo ional o his 41M a e.
I is con enien se ing he maximum 41M a e in 3000-5000 s-1, o ins ance, and using he
pulse gene a o du ing he measu emen s o con ol he de ec o efficiency. This way, he
efficiency will s ay s able and highe han 80%, bu i will be possible o measu e 41Ca/40Ca
a ios up o 10-8 e en when he sou ce ou pu cu en is ela i ely high.
(a) 41Ca (b) 39K
Figu e 4.8: Expe imen al spec a o 41Ca3+ and 39K3+ ions eaching he compac GIC a SARA. Di e en
amplifie gains we e used du ing he adquisi ion o hese spec a.
71
CHAPTER 4. SETUP OF THE 1 MV AMS SYSTEM AT CNA FOR 41CA MEASUREMENTS
Figu e 4.9: Rela ionship be ween he 41M a e eaching he compac GIC and he de ec o efficiency.
The 41M a es p esen ed ha e been al eady co ec ed di iding he measu ed a e by he efficiency.
4.4 K-co ec ion and 41Ca backg ound
As i was s a ed in chap e 3, he sequen ial de ec ion o ions o wo di e en masses, 41
and 39 u, equi es he use o di e en e minal and ESA po en ials. A SARA, he bes way o
do so is uning, fi s , he ol ages which would make 40Ca ions each he GIC, hanks o an
inse able Fa aday cup which can be placed jus a e he ESA (FC5). This also allows a be e
uning o he o he pa ame e s, like he Q-Pole iple ol age.
F om hese ol ages, i is ela i ely easy o calcula e he e minal ol age o make ano he ion
ha e he same EM p oduc so ha i is selec ed by he HEM. Since he ene gy is changed,
so has o be he ESA ol age. Table 4.1 shows ypical ol ages used a SARA o 41M and 39K,
based on he p e ious uning o 40Ca in he FC5.
One o he fi s s udies a SARA o he ela ionship be ween he 41M/40Ca and 39K/40Ca a ios
in blank samples is p esen ed in Figu e 4.10. As i happened in his case, mos o he 41Ca
measu emen s a SARA ha e shown a R2 alue highe han 0.9 in his linea eg ession, wi h a
K-co ec ion ac o ypically be ween 0.065 and 0.09, his is, a ound he na u al 41K/39K a io
(0.0721).
72
4.4. K-CORRECTION AND 41CA BACKGROUND
The 41Ca/40Ca backg ound o he 2+ s a e, a e applying he K-co ec ion, is usually on he
(2-8) ×10-12 ange, and only has been highe han 10-11 when samples wi h 41Ca/40Ca a ios
highe han 10-8 ha e been measu ed. This backg ound le els a e simila o hose ound a
TANDY. Since 41K in e ence is co ec ed, he main cause o his backg ound, s ill much highe
han hose ound in la ge AMS acili ies, would be he one p oduced by sca e ing p ocesses
Table 4.1: Vol ages a he accele a o e minal and he ESA in o de o selec di e en ions wi h he
HE spec ome e a SARA.
Ion 40Ca 41M39K
LE side ion mass (u) 97 98 96
Bounce magne ol age (V) 0 -357 364
Te minal ol age (kV) 940 915 965
2+ ene gy (keV) 2280 2225 2335
2+ ESA ol age (kV) 44.7 43.6 45.9
3+ ene gy (keV) 3220 3140 3300
3+ ESA ol age (kV) 42.1 41.1 43.2
Figu e 4.10: Linea co ela ion be ween he 41M/40Ca and 39K/40Ca a ios in blank samples a SARA. 7
di e en blank samples we e measu ed du ing 3 uns.
73

CHAPTER 4. SETUP OF THE 1 MV AMS SYSTEM AT CNA FOR 41CA MEASUREMENTS
on esidual gas, ape u es, e c. This e ec , which is inc eased because o he lowe ene gies
and he need o using highe s ippe p essu es o des oy he 2+ molecula backg ound
[Su e ,1998], akes also addi ional impo ance in 41Ca measu emen s because o he high
mass di e ence be ween he (41CaF3)-ion and 41Ca. This ac makes easie he p oduc ion o
ions wi h mass 98 u whe e s able iso opes o calcium wi h masses close o 41 u, mainly 40Ca
and 42Ca, a e p esen . Theo e ically, he c oss sec ion o his e ec should be lowe a SARA
han a TANDY, because o he highe ene gy. Howe e , a TANDY, he p esence a he exi
o he s ipping ube o one o he pumps which eci cula es he s ipping gas could educe
he p esence o esidual gas in he HE accele a ion ube, compensa ing he highe sca e ing
c oss sec ion.
The 41Ca/40Ca backg ound o he 3+ s a e has no been s udied ye .
4.4.1 Tes s wi h blank u ine samples: adiochemical me hod
One o he applica ions whe e 41Ca backg ounds ound in low-ene gy AMS sys ems allow
o pe o m measu emen s comple ely compe i i e wi h la ge acili es is he biomedical one.
Fo ins ance, adminis a ing o al doses o 41Ca o 100 nCi (28.23 nmol), minimum u ina y
41Ca/40Ca a ios4a e in he 5-10 ×10-11 ange [Denk e al.,2006,2007]. As i was s a ed in
chap e 1, such a dose can be conside ed adiologically exemp . E en s udies whe e 41Ca
doses a e one o de o magni ude highe (and so would be 41Ca/40Ca a ios) a e s ill consid-
e ed o ha e no adiological haza d [Wea e e al.,2009], since he dose o e 50 yea s would
be 10 µS 5[ICRP,2012].
The capabili ies o he SARA sys em o measu e his kind o samples was es ed in an in e -
compa ison measu emen wi h he TANDY sys em (see sec ion 4.6). These samples we e sen
o he ETH LIP in CaF2 o m by he Nu iT ace g oup a he Na ional Uni e si y o Singapo e.
Ne e heless, in o de o be able o pe o m 41Ca measu emen s o biomedical g oups a
CNA, including he chemical p epa a ion o he samples in ou se ice, chemical p epa a ion
es s wi h blank u ine samples ha e been pe o med.
The chemical p ocedu e was based on hose p esen ed on he bibliog aphy [F eeman e al.,
1995;Mille e al.,2013;Jackson e al.,2013], bu wi h wo impo an changes: he ini ial
p ecipi a ion o calcium ca bona e ins ead o calcium oxala e; and he omission o he use o
ion-exchange esins. The use o calcium ca bona e a oids he p esence o his sal in he final
4U ine samples a e he mos common biomedical samples use o con ol he changes o he 41Ca concen a-
ion in he ex acellula fluids on his kind o s udies.
5The yea ly backg ound dose because o na u al sou ces is ypically highe han 1 mS .
74
4.4. K-CORRECTION AND 41CA BACKGROUND
CaF2p ecipi a e, since he ca bona e ion is des oyed i he sal is dissol ed in an acid6. The
ion-exchange s ep was disca ded because ou es s wi h ca bona e samples did no show a
clea educ ion o he final 39K/40Ca a ios. The ime consump ion and sample loss because
o his p ocess a e, he e o e, no wo hy.
The s eps o he fi s pa o ou chemical p ocedu e, he calcium ca bona e isola ion (see
Figu e 4.11), designed o 24 h samples and es ed wi h 1 l collec ions om olun ee s7, a e:
1. Addi ion, pe 100 ml u ine, o 5 ml o concen a ed HCl and se ling o he solu ion o
1 h.
2. Decan ing o he supe na an in a 2 l (o la ge ) beake and addi ion o concen a ed
NH3un il pH=11.
3. Addi ion, pe 100 ml u ine, o 10 ml Na2CO3solu ion (0.4 g/ml), hea ed un il i almos
boils.
4. Se ling o he beake o e nigh wi h some inclina ion. I is also possible o le he
beake up igh , bu hea ing wi hou boiling o ake ad an age o he indi ec ela ion-
ship be ween empe a u e and he solubili y o calcium ca bona e8.
5. Disca ding o mos o he supe na an and ans e o he es , oge he wi h he p e-
cipi a e, o 50 ml cen i uge ubes.
6. Cen i uga ion and disca ding o he supe na an . T ans e p ecipi a es o an unique
50 ml cen i uge ube pe sample.
Figu e 4.11: Scheme o he calcium isola ion me hod o u ine samples.
6(CO3)2- +2H+−→ CO2+ H2O
7Since hese olun ee s we e no dosed wi h 41Ca hese samples a e conside ed blanks.
8This also a oids he p esence o o he calcium sal s since, unlike calcium ca bona e, mos o he compounds
p esen a di ec ela ionship be ween empe a u e and solubili y.
75
CHAPTER 4. SETUP OF THE 1 MV AMS SYSTEM AT CNA FOR 41CA MEASUREMENTS
7. Washing wi h 20 ml sa u a ed Na2CO3solu ion (0.2 g/ml). Cen i uga ion and disca ding
o he supe na an . This s ep should be pe o med wice.
8. Washing wi h 20 ml Na2CO3solu ion (0.02 g/ml). Cen i uga ion and disca ding o he
supe na an . This s ep should be pe o med wice.
9. Washing wi h 20 ml o MilliQ wa e . Cen i uga ion and disca ding o he supe na an .
10. Dissol ing o he p ecipi a e wi h 5 ml o concen a ed HCl.
11. Addi ion o concen a ed NH3un il pH=9.
12. Cen i uga ion and disca ding o he p ecipi a e (mos ly magnesium hyd oxide).
13. Addi ion o some d ops o concen a ed Na2CO3solu ion (0.2 g/ml).
14. Cen i uga ion and disca ding o he supe na an .
15. Washing wi h ace one.
16. Cen i uga ion and disca ding o he supe na an .
17. D ying a 70 ºC o 1 h.
The CaF2p ecipi a ion pa is a much mo e simple p ocedu e. The ca bona e is dissol ed wi h
1 M HCl, and jus some d ops o concen a ed HF a e necessa y o p ecipi a e he calcium
fluo ide. A e cen i uging and disca ding he supe na an , he p ecipi a e is washed wi h
MilliQ wa e . Then, he cen i uga ion and disca ding o he supe na an is epea ed. The
final p ecipi a e is d ied a 105 ºC o e nigh .
K-co ec ed 41Ca/40Ca a ios om hese blank u ine samples was equi alen o hose ound in
ou ypical blank samples, a ound 5 ×10-12. This ag ees wi h he ac ha , in la ge acili ies
whe e backg ound le els a e in he 10-15-10-14, he 41Ca/40Ca a io om p ocessed u ine
blanks is ypically lowe han 5 ×10-13 [Jackson e al.,2013], one o de o magni ude lowe
han ou backg ound alues.
Du ing his hesis, we ha e been in con ac wi h se e al biomedical g oups om di e en
uni e si ies in Spain in o de o pe o m hese kind o 41Ca measu emen s. Un o una ely,
hese con ac s we e no success ul. Ne e heless, ou sys em is o ally capable o pe o ming
he measu emen s in he case ha any g oup is in e es ed in he u u e.
76
4.5. MEASUREMENT OF THE HE SIDE OPTICAL TRANSMISSION WITH STANDARD SAMPLES
4.5 Measu emen o he HE side op ical ansmission wi h s an-
da d samples
The p esence o a quad upole iple a he exi o he 1 MV ande on allows an op imal
op ical ocusing no ma e he e ec i e cha ge s a e q∗≡M0
Mqo he ions (see chap e 2).
This is one o he mos impo an di e ences o SARA wi h he TANDY sys em a ETH LIP o
41Ca AMS. The non-op imal op ical beha iou o 41Ca h ough he HE side spec ome e a
TANDY implies op ical losses o , a leas , 15% o he cha ge s a e 2+, and 33% o he cha ge
s a e 3+ [Ch is l e al.,2013;Vockenhube e al.,2015].
A SARA, he e o e, he e should exis an imp o emen on his ansmission ela i e o he
one a TANDY, and no such a dependency on he cha ge s a e. An app oxima ion o his
op ical ansmission can be ob ained by checking he measu ed 41Ca/40Ca a ios om he
s anda d samples, p o ided ha he de ec o efficiency is co ec ed. A SARA, his s udy has
only been pe o med, ye , wi h he 2+ s a e.
The AE701 s anda ds om he Ins i u e o Re e ence Ma e ials and Measu emen s (IRMM)
o he Eu opean Commission’s Join Resea ch Cen e [Hennessy e al.,2005] ha e been used
o s anda d co ec ion in 41Ca measu emen s a SARA du ing his hesis. Minimum mea-
su ed/nominal 41Ca/40Ca a ios highe han 80% we e ypically consequence o sligh changes
in he e minal ol ages, while he op imal condi ions lead o measu ed/nominal ela ionship
e en highe han 99%. Taking, o ins ance, all he 41Ca measu emen s pe o med du ing he
yea 2017, he a e age o his measu ed/nominal ac o is (93.8 ±5.5)%, using he s anda d
de ia ion as unce ain y.
Table 4.2 shows an example o he blank co ec ed 41Ca/40Ca a ios om each o he 3 s an-
da ds used du ing one o ou 41Ca measu emen s a SARA; his is, he 41Ca/40Ca a ios calcu-
la ed in o de o pe o m he s anda d co ec ion (see chap e 3) Du ing his measu emen ,
2 ca hodes om each s anda d ma e ial we e measu ed.
The op ical ansmission o he 3+ s a e has no been s udied ye .
Table 4.2: Blank co ec ed 41Ca/40Ca a ios o he s anda d samples used in one o he 41Ca measu e-
men s a SARA using he 2+ s a e.
S anda d Nominal 41Ca/40Ca Measu ed 41Ca/40Ca Measu ed/nominal (%)
AE701/3 (1.0181 ±0.0069) ×10-8 (0.988 ±0.019) ×10-8 97.0 ±2.0
AE701/4 (1.0479 ±0.0071) ×10-9 (1.029 ±0.049) ×10-9 98.2 ±4.8
AE701/5 (1.0520 ±0.0071) ×10-10 (1.000 ±0.013) ×10-10 95.1 ±1.5
77
CHAPTER 4. SETUP OF THE 1 MV AMS SYSTEM AT CNA FOR 41CA MEASUREMENTS
Figu e 4.15: Tempo al e olu ion o he u ina y 41Ca/40Ca a io in subjec 6, including measu emen s
a TANDY and SARA.
84

4.6. COMPARISON MEASUREMENTS WITH TANDY
Figu e 4.16: Tempo al e olu ion o he u ina y 41Ca/40Ca a io in subjec 7, including measu emen s
a TANDY and SARA.
85
CHAPTER 4. SETUP OF THE 1 MV AMS SYSTEM AT CNA FOR 41CA MEASUREMENTS
Figu e 4.17: Tempo al e olu ion o he u ina y 41Ca/40Ca a io in subjec 8, including measu emen s
a TANDY and SARA.
86
4.6. COMPARISON MEASUREMENTS WITH TANDY
Figu e 4.18: Tempo al e olu ion o he u ina y 41Ca/40Ca a io in subjec 9, including measu emen s
a TANDY and SARA.
87
CHAPTER 4. SETUP OF THE 1 MV AMS SYSTEM AT CNA FOR 41CA MEASUREMENTS
Figu e 4.19: Tempo al e olu ion o he u ina y 41Ca/40Ca a io in subjec 10, including measu emen s
a TANDY and SARA.
88
Chap e 5
41Ca measu emen in conc e e
samples om he bioshield o a
nuclea powe plan
P e ious chap e s a e dedica ed o he echnical aspec s o he 41Ca AMS measu emen s.
This las chap e is dedica ed o he fi s applica ion pe o med a CNA Se ille o he 41Ca
measu emen s: he cha ac e iza ion o conc e e samples om he bioshield o a NPP o he
Spanish adioac i e was e managemen agency, ENRESA.
The di ec ela ionship be ween he 41Ca/40Ca a io in he conc e e bioshield and he he mal
neu on fluence is explained in sec ion 5.1. Mos o he in o ma ion om his sec ion is based
on Chap e 3 o Lewis [2008].
The samples had o be chemically ea ed in o de o ha e he CaF2 o be measu ed by AMS.
The chemical me hod de eloped o ha , and applied o all he samples measu ed, is de ailed
in sec ion 5.2.
The final desc ip ion o he samples and he esul s ob ained a e shown in sec ion 5.3.
5.1 P oduc ion o 41Ca in he conc e e bioshield o nuclea eac-
o s
41Ca is p oduced in he conc e e bioshield o nuclea eac o s by he neu on ac i a ion o i s
s able iso ope, 40Ca. The ene gy spec um om hese neu ons eaching he shield p esen s
89

CHAPTER 5. 41CA MEASUREMENT IN CONCRETE SAMPLES FROM THE BIOSHIELD OF A
NUCLEAR POWER PLANT
wo main g oups:
•Fas neu ons (ene gies in he 1 MeV ange) p oduced a e he fission p ocesses which
escape he eac o essel be o e being he malized by he eac o mode a o .
•The mal neu ons (ene gies much lowe han 1 eV) om he mode a ion o he neu ons
p oduced in he fission which escape he eac o essel be o e ei he p oducing a new
fission eac ion, o being abso bed by ano he ma e ial wi hin he essel (con ol ods,
walls, e c.).
Ne e heless, he p oduc ion o 41Ca by as neu on cap u e can be dis ega ded since, as i
can be seen in Figu e 5.1, he c oss sec ion o he 40Ca(n,γ) eac ion is a ound 2 o de s o
magni ude lowe o hese neu ons han o he mal ene gies. In ac , he main con ibu ion
o as neu ons o he p oduc ion o 41Ca would come om i s mode a ion in he conc e e
ma e ial i sel , due o he elas ic sca e ing p ocesses wi h he hyd ogen p esen on i . This
way, i can be supposed ha 41Ca is en i ely p oduced by he he mal neu on cap u e o 40Ca,
being hese ei he o iginal he mal neu ons eaching he shield, o as neu ons which ha e
been he malized by he ma e ial. The elas ic sca e ing no only he malize he neu ons o
oom empe a u e, bu also causes a di usion o he he mal neu ons.
Figu e 5.1: 40Ca neu on cap u e c oss sec ion and he mal neu on ene gy dis ibu ion. C oss sec ion
da a om IAEA [n.d.].
90
5.1. PRODUCTION OF 41CA IN THE CONCRETE BIOSHIELD OF NUCLEAR REACTORS
The 41Ca/40Ca a io in any loca ion o he shield, he e o e, would be
41Ca/40Ca =Z∞
0
σ(E)dΦ(E)
dE dE, (5.1)
whe e σ(E)is he 40Ca(n,γ) c oss sec ion as a unc ion o he ene gy, and dΦ(E)
dE is he en-
e gy dis ibu ion o he neu on fluence. The ene gy dis ibu ion o he mal neu ons can be
conside ed o be Maxwellian, his is,
dΦ(E)
dE = Φ h
E
(kT)2e−E
kT ,(5.2)
being Φ h he o al he mal neu on fluence, k he Bol zmann cons an , and T he empe -
a u e o he medium whe e he neu ons ha e been he malized. In p essu ized wa e e-
ac o s (PWR), o ins ance, he wa e mode a o is ypically a a empe a u e a ound 600 K.
Howe e , hese neu ons a e easily he malized again o oom empe a u e (∼300 K) when
hey each he conc e e shield, mainly because o he hyd ogen con en on i 1. The e o e, in
his a ea i can be conside ed ha kT = 0.0253 eV.
A he mal ene gies, neu on absop ion c oss sec ions a e in e sely p opo ional o he e-
loci y o he neu ons. Typically, he c oss sec ion o neu ons wi h an ene gy E0= 0.0253 eV
is epo ed in he bibliog aphy, which is called ” he mal neu on cap u e c oss sec ion”;
o ins ance, C ans on and Whi e [1971] epo a he mal neu on cap u e c oss sec ion o
400 ±40 mb o 40Ca. Consequen ly, he neu on cap u e c oss sec ion o o he ene gy in
he he mal ange is
σ(E) = σ(E0) E0
E.(5.3)
The e o e, he 41Ca/40Ca a io p oduced by a he mal neu on fluence Φ h would be
41Ca/40Ca = Φ h ·σ(E0)·1
(kT)2Z∞
0 E0
EEe−E
kT dE =
= Φ h ·σ(E0)· E0
kT Z∞
0
√ e− d =√π
2 E0
kT σ(E0)Φ h.
(5.4)
Defining he he mal a e age c oss sec ion as
¯σ(T)≡√π
2 E0
kT σ(E0),(5.5)
1Neu on ene gy loss is mo e impo an when he sca e ing p ocess akes place wi h ligh nuclei.
91
CHAPTER 5. 41CA MEASUREMENT IN CONCRETE SAMPLES FROM THE BIOSHIELD OF A
NUCLEAR POWER PLANT
he o al he mal neu on fluence in a loca ion is hen di ec ly ela ed o he measu ed
41Ca/40Ca he e, being
Φ h =
41Ca/40Ca
¯σ(T).(5.6)
In conc e e, whe e neu ons a e he malized o oom empe a u e kT =E0, he he mal
a e age c oss sec ion would be ¯σ=√π
2σ(E0). Fo he 40Ca(n,γ) eac ion, his c oss sec ion
would be, he e o e, 354 ±35 mb = (3.54 ±0.35) ×10-25 cm2.
No e ha , since p oduc ion o 41Ca is di ec ly p opo ional o 40Ca con en , he 41Ca/40Ca
a io is independen on he calcium concen a ion in he ma e ial. The only o he impo an
ac o ela ed o his a io, besides he he mal neu on fluence, is he empe a u e o which
hese neu ons a e he malized.
The egions o maximum neu on flux in bioshields om comme cial nuclea eac o s can
su e he mal neu on fluences close o 1019 cm-2. This would p oduce 41Ca/40Ca a ios
highe han 3.5 ×10-6.
In o de o calcula e he 41Ca specific ac i i y2 om 41Ca/40Ca measu emen s, i is necessa y
o know he calcium concen a ion in he conc e e, which can s ongly a y om one ype o
conc e e o ano he .
The 41Ca/40Ca a io can be used, also, o calcula e a om a ios om o he adionuclides p o-
duced by neu on cap u e. Fo ins ance, he 36Cl/35Cl a io in he same loca ion would be
36Cl/35Cl =σ[E0,35Cl(n, γ)]
σ[E0,40Ca(n, γ)]
41Ca/40Ca,(5.7)
independen ly o he empe a u e, in his case, since he same ac o would co ec bo h he
he mal neu on cap u e c oss sec ions o 35Cl and 40Ca.
5.2 Radiochemical ea men o conc e e samples o 41Ca AMS
measu emen s
Ou chemical me hod o ea he conc e e is, in some aspec s, based on hose me hods used
o 41Ca LSC o in e compa isons be ween his echnique and AMS [Hampe e al.,2013;Hou,
2005;Wa wick e al.,2009]. Ne e heless, many s eps which a e no necessa y o 41Ca AMS
2The 41Ca specific ac i i y in any ma e ial is defined as he 41Ca ac i i y pe ma e ial mass uni .
92
5.2. RADIOCHEMICAL TREATMENT OF CONCRETE SAMPLES FOR 41CA AMS MEASUREMENTS
ha e been elimina ed, and some s eps which imp o e he po assium supp ession ha e been
added. The me hod, as he one desc ibed o u ine samples in chap e 4, is di ided in wo
pa s: he calcium isola ion as calcium ca bona e, and he calcium fluo ide p ecipi a ion. A
simple scheme o he calcium isola ion p ocess is p esen ed in Figu e 5.2.
Each sample is fi s diges ed in 20 ml o aqua egia du ing 2 hou s. Typically, be ween 100
and 500 mg o each sample is used. The amoun depends on he aspec o he sample:
•When mos o i has a pale colo and is easily g ound, mos o he sample comes om
he binde , which is s ongly dissol ed by acid diges ion, so less amoun o conc e e
sample is needed.
•When he sample includes ha d and da k agg ega es, which a e ypically silica es which
a e no dissol ed by aqua egia, a ela i ely la ge amoun o sample, be ween 300 and
500 mg, is ea ed.
The 41Ca/40Ca a io does no depend on he chemical o m in which he calcium is ound.
Consequen ly, he ac ha he acid diges ion does no dissol e pa o he calcium on he
sample, like ha calcium p esen in silica es, has no huge ele ance. In Figu e 5.3 he di e -
ence be ween how he acid diges ion a acks he sample depending on he agg ega e/binde
ela ionship can be obse ed. In one o he 2 samples, i can be obse ed how an agg ega e is
no a ec ed by he acid diges ion. This s ep elimina es mos o he Si p esen in he sample.
The acid solu ion is cen i uged, disca ding he p ecipi a e. 20 ml o concen a ed NH3is
added o he supe na an in o de o p ecipi a e hyd oxides om o he elemen s p esen in
conc e e, like i on o magnesium. I necessa y, mo e concen a ed NH3is added un il pH is
be ween 9 and 11. The solu ion is cen i uged again, disca ding he p ecipi a e.
Figu e 5.2: Scheme o he calcium isola ion p ocess o conc e e samples.
93
CHAPTER 5. 41CA MEASUREMENT IN CONCRETE SAMPLES FROM THE BIOSHIELD OF A
NUCLEAR POWER PLANT
conc e e s uc u e o e en a highe mode a ion o he as neu ons a some poin s.
The case o hose d ill co es a 30 cm om he cen al ones in he ho izon al di ec ion (CDL) is
special, since hey a e ex emely close o he ape u es owa d he ins umen a ion wells (see
Figu e 5.8). This ape u es we e filled wi h ai , so hey could be conside ed ” anspa en ” o
he neu ons, which eached he side walls o hese ape u e. The di usion o he neu ons
in he conc e e would cause ano he ”neu on on ”, pe pendicula o his wall. Fo deepe
loca ions o he CDL d ill co es, he neu on fluence added by his on , ela i e o he one
coming om he ca i y su ace, would be mo e impo an han o hose loca ions close o
he eac o essel. In any case, his ape u es add complexi y o he beha io o he neu ons
fluence in hese loca ions, so he non ideal beha io o he 41Ca/40Ca dep h p ofiles in hese
d ill co es is no unexpec ed.
The lack o symme y be ween he esul s o bo h blocks, on he o he hand, is qui e unex-
pec ed, bu could be also a consequence o he huge he e ogenei y o he ma e ial.
Due o he 30% Ca con en o he conc e e, he samples closes o he essel, wi h 41Ca/40Ca
a ios a ound 3 ×10-6, would p esen 41Ca specific ac i i ies o 3000 Bq/g. On he o he
hand, he sample wi h he minimum 41Ca/40Ca a io, 3 ×10-10, would p esen 41Ca specific
ac i i ies o 0.3 Bq/g.
Acco ding o he esul s, maximum he mal neu on fluences on he bioshield su ace a e
e y close o 1019 cm-2. This esul has a g ea ele ance, since i is known ha conc e e
Figu e 5.8: C oss sec ion o he Jos´
e Cab e a eac o essel and bioshield when he plan was ope -
a i e a he heigh o maximum neu on fluence. The posi ions whe e d ill co es CDC and CDL we e
aken, o blocks 1, 4 and 6. Fo he es o he blocks, CDL co e was aken om he opposi e side.
100

5.3. MEASUREMENTS OF 41CA IN THE BIOSHIELD OF THE JOS´
E CABRERA NUCLEAR POWER
PLANT
ma e ials s a losing some o he mechanical p ope ies when hey unde go o al neu on
fluences abo e 1019 cm-2 [Field e al.,2015]. Taking in o accoun ha ou esul s only assess
he he mal neu on fluence, i is almos su e ha he o al neu on fluence eached le els
much highe ha 1019 cm-2 a hese poin s. These e ec s on he mechanical p opo ies,
ne e heless, may s ongly depend on he ype o conc e e. Consequen ly, he s udy o he
mechanical p ope ies o his conc e e, which ENRESA will pe o m, oge he wi h ou esul s
o he he he mal neu on fluences and addi ional s udies o he as neu on fluence, could
help o a be e assessmen o he e ec s o neu on adia ion in hese p ope ies.
Ano he capabili y o hese esul s is he es ima ion o he specific ac i i ies om o he a-
dionuclides which a e p oduced in conc e e mainly by he mal neu on ac i a ion. This is he
case, o ins ance, o 36Cl. Chlo ine is only ound in ace amoun s in conc e e, mainly caused
by he Cl-p esence in wa e . Typical chlo ide concen a ions in conc e e a e in he ange o
100 ppm. The much lowe p esence o chlo ine in conc e e, ne e heless, is compensa ed by
he ela i ely high he mal neu on cap u e c oss sec ion, o 41.8 ±1.2 b [Sims and Juhnke,
1969], his is, 2 o de s o magni ude highe han he one o 40Ca. In any case, e en assuming
chlo ide concen a ions nea limi s, 36Cl specific ac i i ies would no exceed 35 Bq/g, which is
well below he limi o exemp ma e ial, 104Bq/g [CSN,2003;IAEA,2012].
5.3.3 Resul s om he la e al samples
Samples om he le side o he block 4 we e no sen o be measu ed by AMS a CNA since
hese would be equi alen o hose om he igh side o he block 3. A scheme o he loca ion
o hese samples was p esen ed in Figu e 5.6. The heigh s o hese loca ions, ela i e o he
cen e o he eac o co e, we e +360 cm, +280 cm, +200 cm, +120 cm, -260 cm and -300 cm.
The dep hs we e 22.5 cm and 67.5 cm o sub-blocks A and B, and 35 cm and 105 cm o
sub-block E.
The 41Ca/40Ca esul s in hese loca ions is p esen ed in Figu e 5.9. The main unce ain y
o he in e p e a ion o hese esul s is he possibili y o c oss-con amina ion du ing he
bioshield segmen a ion p ocess. Ne e heless, in mos o he cases, an expec ed indi ec
ela ionship be ween he e ical dis ance o he co e and he 41Ca/40Ca a io is ound; so
is be ween dep h and 41Ca/40Ca. Taking in o accoun ha hese samples a e ela i ely a
om he a eas o maximum neu on fluence, he exemp ions o hese indi ec ela ionships
a e no unexpec ed. These exemp ions may be caused by he he mal neu on di usion p o-
cesses o he p esence o highe as neu on fluences nea hese a eas which a e mode a ed
by he conc e e, as i happened in he case o he d ill co es. I is, ne e heless, qui e unex-
pec ed ha he esul s om he 3 sides a e no equi alen since, because o he geome y
101
CHAPTER 5. 41CA MEASUREMENT IN CONCRETE SAMPLES FROM THE BIOSHIELD OF A
NUCLEAR POWER PLANT
o he eac o co e, he neu on spa ial dis ibu ion should ha e a 45º o a ional symme y.
On he o he hand, his o a ional symme y should only be expec ed on he neu on fluence
o an homogeneous ma e ial. As i was p e iously s a ed, his is no he case o conc e e
walls. One o he elemen s whe e his he e ogenei y is expec ed o be high is hyd ogen, since
addi ional wa e is ypically inco po a ed du ing he ha dening p ocess, and in he di e en
uel changes du ing he ope a ion o he NPP. Hyd ogen p esence is key o he mal neu on
di usion and as neu on mode a ion.
The minimum 41Ca specific ac i i y in his samples is ound o be close o 0.15 Bq/g. Fo mos
o hese la e al samples he maximum mass a ailable was o se e al hund ed o millig ams in
o de o keep he sample package below 137Cs exemp ion limi s. On he o he hand, he mos
sensi i e s udies o 41Ca measu emen s by LSC epo de ec ion limi s o 0.1 Bq. Acco ding
o his, 41Ca cha ac e iza ion by LSC would ha e no been possible, and AMS becomes neces-
sa y. Besides, e en in he case whe e samples could ha e been measu ed by LSC, as hose
om he d ill co es, he sample mass which should be des oyed and s a is ical unce ain ies
would ha e been much highe .
Figu e 5.9: 41Ca/40Ca esul s, in ×10-9 uni s, om he la e al samples. The -I and -D makes e e ence
o which o bo h sides o he block he samples come om, as seen om he eac o ca i y: I, o le
side; D, o igh side. The dashed line ep esen s he heigh whe e he cen e o he eac o co e was.
102
Chap e 6
Summa y and conclusions
This hesis has ulfilled i s main goal: he se ing-up o he 41Ca measu emen s wi h he 1 MV
AMS sys em a CNA, SARA. Mos o he esul s ob ained wi h his sys em can also be applied
o o he 1 MV AMS sys ems om HVE. Besides, he da a collec ed om he di e en mea-
su emen s pe o med du ing his hesis on his sys em and he 600 kV AMS sys em a ETH
Zu ich, TANDY, ha e p o ided in o ma ion use ul o any 41Ca AMS measu emen , specially a
low ene gy sys ems. Rou ine AMS measu emen s o 41Ca wi h SARA ha e been implemen ed
a CNA, applied o he cha ac e iza ion o low le el nuclea esidues.
In chap e 1, an ini ial idea abou he di e en applica ions o 41Ca measu emen s was p e-
sen ed, oge he wi h he basic ideas ela ed o he AMS echnique. The p esen a ion o he
SARA and TANDY sys ems in chap e 2 se ed also o desc ibe he physical p ocesses in ol ed
in he di e en elemen s o AMS sys ems. These desc ip ions ocus on he ecen changes
on bo h sys ems and in hose di e ences which a e impo an o he 41Ca AMS pe o mance.
Di e en s udies ela ed o he 41K in e e ence, which is he main challenge in 41Ca AMS a
low ene gies, a e p esen ed in chap e 3. The expe imen s in o de o se ing-up and op-
imize he pe o mance o 41Ca measu emen s a SARA a e de ailed in chap e 4. The fi s
applica ion o 41Ca AMS a CNA, he s udy o he 41Ca spa ial dis ibu ion in he bioshield o a
NPP, can be ound in chap e 5.
41Ca AMS a low ene gies
E en when he 41K in e e ence can be es ima ed by sequen ially measu ing he o he s able
iso ope o po assium, 39K (K-co ec ion), i is impo an o assess he possible ways by which
his in e e ence can a y.
103
CHAPTER 6. SUMMARY AND CONCLUSIONS
The exis ence o he (41K57Fe)-ion and i s ela ionship wi h he 41K in e e ence has been
p o en hanks o se e al expe imen s pe o med a TANDY du ing he s ay a ETH LIP. This is
he cause o he ela i ely high 41K ion a es when me allic elemen s a e spu e ed. The main
consequences, which ha e been obse ed in bo h sys ems, SARA and TANDY, a e:
•The dependence o he 41K in e e ence on he ma e ials used in he ca hode, like he
a ge holde o he binde . Fo ins ance, highe 41K/40Ca has been obse ed when Al
a ge holde s a e used, ins ead o Ti o Cu. Al ma e ials ypically p esen highe Fe
ace con en .
•The 41K/40Ca is no cons an in ime. Because o his, he K-co ec ion mus be applied
sequen ially du ing he da a analysis. A de ailed s udy o his e olu ion a SARA is p e-
sen ed, showing how he 41K in e e ence beha io di e s e en o di e en a ge s
om he same sample.
Applying his K-co ec ion manually can be ime consuming. The e o e, wo p og ams ha e
been w i en in FORTRAN code o he 41Ca measu emen s a SARA and TANDY. The use o
hese p og ams means a huge educ ion o he ime dedica ed o his analysis, om a ound
one hou o se e al minu es.
41Ca measu emen s a SARA
A de ailed s udy o he pe o mance pa ame e s o 41Ca measu emen s a SARA has been
pe o med h oughou his hesis. The main pa ame e s s udied ha e been:
•Ioniza ion efficiency: mixing he calcium fluo ide samples wi h sil e in a CaF2:Ag
weigh a io o 1:4, he (40CaF3)-ou pu cu en o he HVE SO-110B ion sou ce is s able
and be ween 50 and 150 nA. The ioniza ion efficiency in his case is a ound 0.1%.
•S ippe ansmission: he cha ge s a e wi h he highes ansmission in he He s ip-
pe is he 2+ s a e. Using a He gas p essu e o 0.03 mba , his is, a He mass hickness o
∼0.075 µg·cm-2, ansmission o his cha ge s a e is 40%, and a ull sup ession o he
molecula backg ound is achie ed. The possibili ies o he cha ge s a e 3+, wi h a lowe
ansmission o 20%, a e being s udied because o he possibili y o educing he final
backg ound.
•GIC de ec ion efficiency: he compac GIC p esen s de ec ions efficiencies highe han
90% o 41Ca a es lowe han 1000 s-1. Fo highe a es, he efficiency dec eases lin-
ea ly.
104
•41Ca/40Ca backg ound: while he ypical 41K/40Ca a ios a e in he 10-11 ange, he
41Ca/40Ca in blank samples, a e applying he K-co ec ion, is one o de o magni ude
lowe .
•HE side op ical ansmission: hanks o he independence o he objec o he HEM
on he cha ge s a e, due o he quad upole iple a he exi o he accele a o , he
41Ca beam is op imally anspo ed h ough he HE spec ome e . The measu emen s
o 41Ca s anda ds show ha he op ical ansmission in his sec o is in he 90-100%
ange.
The compa ison o he pe o mance pa ame e s a SARA wi h hose a TANDY show ha he
capabili ies o bo h sys ems o 41Ca AMS a e equi alen . SARA is pe ec ly fi o pe o m,
o ins ance, 41Ca measu emen s o biomedical applica ions and o cha ac e iza ion o low
le el nuclea esidues.
41Ca in he bioshield om a nuclea powe plan
The fi s applica ion o he 41Ca measu emen s a SARA has been he s udy o he 41Ca spa ial
dis ibu ion in he bioshield o he Jos´
e Cab e a NPP. The 41Ca/40Ca a io in his s uc u e is
di ec ly ela ed o he he mal neu on fluence su e ed by i . A adiochemical me hod o
ea la ge numbe o conc e e samples has been de eloped and applied o samples om a
huge a ie y o dep hs and heigh s o Jos´
e Cab e a’s bioshield. Only 100-500 mg pe sample
a e needed o be des oyed in his ea men , one o de o magni ude lowe han he sample
mass needed o LSC measu emen s. Besides, AMS no only imp o es he limi o de ec ion,
bu also imp o es he p ecision o his measu emen s.
The esul s om he 41Ca measu emen s a SARA show ha , because o he complexi y o he
geome y and he he e ogenei y o he ma e ial, he beha io o he he mal neu on fluence
becomes less ideal he a e he samples a e om he posi ion o maximum neu on fluence.
Consequen ly, neu on anspo simula ions may be no enough o es ima e he in en o ies
o he adionuclides p oduced by he neu on adia ion.
105

CHAPTER 6. SUMMARY AND CONCLUSIONS
Rele an publica ions wi hin his PhD hesis
Two a icles ha e been published ela ed o his hesis:
The s udies o he ela ionship wi h he 41K in e e ence o he i on p esence and he ime
e olu ion, and he specific ea u es o he measu emen analysis ela es o he K-co ec ion,
de ailed in chap e 3, a e p esen ed in
C. VIVO-VILCHES, J.M. L´
OPEZ-GUTI´
ERREZ, M. GARC´
IA-LE´
ON, C. VOCKENHUBER
Fac o s ela ed o 41K in e e ence on 41Ca AMS measu emen s
Nuclea Ins umen s and Me hods in Physics Resea ch Sec ion B: Beam In e ac ions wi h
Ma e ials and A oms, in p ess
doi: 10.1016/j.nimb.2018.04.031
The main esul s and expe imen s p esen ed on chap e 4 abou he op imiza ion o he
pe o mance pa ame e s o he 41Ca measu emen s wi h SARA can be ound in
C. VIVO-VILCHES, J.M. L´
OPEZ-GUTI´
ERREZ, M. GARC´
IA-LE´
ON, C. VOCKENHUBER, T. WALZCYK
41Ca measu emen s on he 1 MV AMS acili y a he Cen o Nacional de Acele ado es
(CNA, Spain)
Nuclea Ins umen s and Me hods in Physics Resea ch Sec ion B: Beam In e ac ions wi h
Ma e ials and A oms 413 (2017) 13-18
doi: 10.1016/j.nimb.2017.10.003
106
Appendix A
GICOSY files
Figu es 2.18 and 2.20 show simula ions o he ion op ics in he HE spec ome e s o SARA and
TANDY, pe o med wi h GICOSY [Weick,n.d.]. The geome y da a om bo h sys ems we e
ob ained om Jacob [2001] and Chamizo [2009]. The meaning o he di e en codes can be
ound in he manual, which is in D . Helmu Weick’s webpage. Simula ion om Figu e 2.15
has also been pe o med wi h GICOSY.
Below, he files w i en o he HE spec ome e s o SARA and TANDY a e p esen ed.
A.1 SARA HE spec ome e
S N SARA-HEsec o ;
R P 0.395 41 2 ;
P X 0.00 0.014 ;
P Y 0.00 0.014 ;
D P 0.005 0.0025 ;
;
VQ2=7.8;
VQ1=-VQ2;
;
SS;
;
D L 0.24 ;
;
FF0;
107
APPENDIX A. GICOSY FILES
A C 0.8 1.1375 0.0475;
FF0;
;
D L 0.095 ;
;
F F 1;
;
E Q 0.059 VQ1 0.025;
;
F F 1;
;
D L 0.0405 ;
;
F F 1;
;
E Q 0.132 VQ2 0.025;
;
F F 1;
;
D L 0.0405 ;
;
F F 1;
;
E Q 0.059 VQ1 0.025;
;
F F 1;
;
D L 0.52 ;
;
F F 3 25.6 0.0 ;
;
M S 0.85 90 0.0215 0 0 0 0 0;
;
FF3310.0;
;
D L 1.094 ;
;
108
A.2. TANDY HE SPECTROMETER
D H E (X,Y) 100000 0.003 0.003 1 10 ;
;
D L 0.375 ;
;
FF100;
;
E S 0.65 120 0.0125 1 -1 1;
;
FF100;
;
D L 0.375 ;
;
D H E (X,Y) 100000 0.003 0.003 1 10 ;
;
D L 0.375 ;
;
D S 0.025 10 100 1 3 1 3;;
DB0.0350.03511313131;
END ;
;
A.2 TANDY HE spec ome e
S N TANDY-HEsec o ;
R P 0.23 41 2 ;
P X 0.00 0.014 ;
P Y 0.00 0.014 ;
D P 0.025 0.0025 ;
;
SS;
;
D L 0.38 ;
;
FF0;
A C 0.24 2.083 0.0475;
FF0;
109
APPENDIX C. FORTRAN PROGRAMS FOR 41CA MEASUREMENT ANALYSIS
Numbe o cycles (=seq)/ un:
5
min. 40Ca HE (nA):
20
max. 39K/40Ca in un:
1.25e-9
max a e (c s/s):
3000.0
Igno e i s un? (yes/no=1/0)
0
Sample sca e (%):
1
Cha ge s a e:
2
Numbe o s anda ds o in e es :
3
S anda d 1 nom. al, numbe o samples, posi ions:
1.0181e-8
2
3
15
S anda d 2 nom. al, numbe o samples, posi ions:
1.0479e-9
2
4
16
116

C.2. OUTPUT FILES
S anda d 3 nom. al, numbe o samples, posi ions:
1.0520e-10
2
5
17
Numbe o blanks:
4
Posi ion:
2
7
14
19
Pulse (1/0=yes/no)
1
Pulse equency (Hz)
100.131
C.2 Ou pu files
C.2.1 ou -K-co ec ion
This file is w i en while pe o ming he K-co ec ion. The in o ma ion p o ided is:
•39K/40Ca and 41M/40Ca a ios om he di e en uns om all he blank samples, which
a e he (x,y) alues o he linea eg ession.
•The K-co e ion ac o α, his is, he slope om he linea eg ession.
•The ela i e unce ain y o α.
•The R2 alue o he linea eg ession.
117
APPENDIX C. FORTRAN PROGRAMS FOR 41CA MEASUREMENT ANALYSIS
C.2.2 ou - uns
The file p o iding in o ma ion abou he di e en uns om all he samples is w i en be-
ween applying he K-co ec ion and he o he impo an co ec ions: blank and s anda d
co ec ion.
The in o ma ion p o ided o each un is:
•The numbe o sequences which passed he disca d c i e ia du ing ha un.
•The o al numbe o 41M coun s de ec ed in ha un.
•The a e age 40Ca cu en .
•The a e age 39K/40Ca a io.
•The K-co ec ed 41Ca/40Ca a io.
•The ela i e unce ain y o his a io.
•The ela i e s anda d de ia ion o his a io.
In he case o measu emen s a TANDY, his file also p o ides in o ma ion abou he s ippe
ansmission.
C.2.3 ou -blank
This file is w i en while he backg ound is being calcula ed. I jus shows:
•The a e age 41Ca/40Ca om each blank sample a e K-co ec ion.
•The 41Ca/40Ca backg ound, his is, he a e age om all he blank samples.
•The ela i e s anda d de ia ion om his backg ound.
C.2.4 ou -s ds
This file is w i en while calcula ing he nominal/expe imen al ac o om he s anda d sam-
ples. I shows, o each di e en s anda d ma e ial:
•The nominal 41Ca/40Ca a io om ha ma e ial.
118
C.2. OUTPUT FILES
•The a e age o he measu ed 41Ca/40Ca in all he ca hodes om ha ma e ial.
•The ac o be ween his nominal and measu ed a ios.
•The ela i e s anda d de ia ion om his ac o .
I also shows ha a e age o he nominal/measu ed ac o om di e en s anda d ma e ials,
and i s final ela i e unce ain y.
C.2.5 ou -final- a ios
In his file, he final esul s om each sample a e p esen ed.
The in o ma ion p o ided o each sample is:
•The numbe o uns om ha sample which passed he disca d c i e ia.
•The a e age K-co ec ed 41Ca/40Ca a io.
•The ela i e unce ain y om ha a io.
•The ela i e s anda d de ia ion om ha a io.
•The in eg al o he χ2dis ibu ion which is used o selec he unce ain y o he blank-
co ec ed a io.
•The blank-co ec ed 41Ca/40Ca a io.
•The ela i e unce ain y om ha a io.
•The final s anda d-co ec ed 41Ca/40Ca a io.
•The ela i e unce ain y om ha a io.
•The o al numbe o 41M coun s de ec ed in all he uns o m ha sample.
•The a e age 39K/40Ca a io.
•The a e age 40Ca cu en .
In he case o measu emen s a TANDY, his file also p o ides in o ma ion abou he s ippe
ansmission.
119
APPENDIX C. FORTRAN PROGRAMS FOR 41CA MEASUREMENT ANALYSIS
120
Bibliog aphy
Akhmadalie , S., Helle , R., Han , D., Rugel, G., Me chel, S., (2013). The new 6 MV AMS-
acili y DREAMS a D esden. Nuclea Ins umen s and Me hods in Physics Resea ch Sec ion
B: Beam In e ac ions wi h Ma e ials and A oms 294, 5–10.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X12000961
Al on, G., (1993). Ion sou ces o accele a o s in ma e ials esea ch. Nuclea Ins umen s
and Me hods in Physics Resea ch Sec ion B: Beam In e ac ions wi h Ma e ials and A oms
73, 221–288.
h p://linkinghub.else ie .com/ e ie e/pii/0168583X9395738Q
A nold, M., Me chel, S., Bou l`
es, D. L., B auche , R., Benede i, L., Finkel, R. C.,
Aumaˆı e, G., Go dang, A., Klein, M., (2010). The F ench accele a o mass spec om-
e y acili y ASTER: Imp o ed pe o mance and de elopmen s. Nuclea Ins umen s and
Me hods in Physics Resea ch Sec ion B: Beam In e ac ions wi h Ma e ials and A oms 268,
1954–1959.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X10002028
Baba, T., Funaya, N., Miyazaki, N., Takahashi, T., Kaiku a, T., Seki, R., Sueno, S., Jiang,
S., Nagashima, Y., (1997). De ec ion o 41Ca wi h a 7Li37Cl molecula pilo beam. Nuclea
Ins umen s and Me hods in Physics Resea ch Sec ion B: Beam In e ac ions wi h Ma e ials
and A oms 123, 183–185.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X96007847
Benne , C. L., Beukens, R. P., Clo e , M. R., Go e, H. E., Liebe , R. B., Li he land, A. E.,
Pu se , K. H., Sondheim, W. E., (1977). Radioca bon Da ing Using Elec os a ic Accele a-
o s: Nega i e Ions P o ide he Key. Science 198, 508–510.
h p://www.sciencemag.o g/cgi/doi/10.1126/science.198.4316.508
Be gq is , I., D ake, D., McDaniels, D., (1974). Radia i e cap u e o as neu ons by 40Ca.
Nuclea Physics A 231, 29–44.
h p://linkinghub.else ie .com/ e ie e/pii/0375947474902917
121

BIBLIOGRAPHY
BMJV, (n.d.). Ve o dnung ¨
ube den Schu z o Sch¨
aden du ch ionisie ende S ahlen.
h ps://www.gese ze-im-in e ne .de/s lsch _2001/BJNR171410001.h ml
Chamizo, E., Enamo ado, S. M., Ga c´ıa-Le´
on, M., Su e , M., Wacke , L., (2008)a. Plu o-
nium measu emen s on he 1 MV AMS sys em a he Cen o Nacional de Acele ado es
(CNA). Nuclea Ins umen s and Me hods in Physics Resea ch Sec ion B: Beam In e ac ions
wi h Ma e ials and A oms 266, 4948–4954.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X08009506
Chamizo, E., L´
opez-Gu i´
e ez, J. M., Ruiz-G´
omez, A., San os, F., Ga c´ıa-Le´
on, M., Maden,
C., Alfimo , V., (2008)b. S a us o he compac 1 MV AMS acili y a he Cen o Nacional
de Acele ado es (Spain). Nuclea Ins umen s and Me hods in Physics Resea ch Sec ion B:
Beam In e ac ions wi h Ma e ials and A oms 266, 2217–2220.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X08002218
Chamizo, E., (2009).Medida de is´
o opos de plu onio, 239Pu y 240Pu, median e espec ome
´
ıa
de masas con acele ado es de baja ene g
´
ıa. Ph.D. hesis, Uni e sidad de Se illa.
Chamizo, E., Ch is l, M., Fifield, L., (2015)a. Measu emen o 236U on he 1 MV AMS sys em
a he Cen o Nacional de Acele ado es (CNA). Nuclea Ins umen s and Me hods in Physics
Resea ch Sec ion B: Beam In e ac ions wi h Ma e ials and A oms 358, 45–51.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X15004796
Chamizo, E., L´
opez-Lo a, M., Villa, M., Casacube a, N., L´
opez-Gu i´
e ez, J. M., Pham,
M. K., (2015)b. Analysis o 236U and plu onium iso opes, 239,240Pu, on he 1 MV AMS sys-
em a he Cen o Nacional de Acele ado es, as a po en ial ool in oceanog aphy. Nuclea
Ins umen s and Me hods in Physics Resea ch Sec ion B: Beam In e ac ions wi h Ma e ials
and A oms 361, 535–540.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X15001913
Chamizo, E., San os, F. J., L´
opez-Gu i´
e ez, J. M., Padilla, S., Ga c´ıa-Le´
on, M., Heine-
meie , J., Schnabel, C., Scognamiglio, G., (2015)c. S a us epo o he 1 MV AMS acili y
a he Cen o Nacional de Acele ado es. Nuclea Ins umen s and Me hods in Physics Re-
sea ch Sec ion B: Beam In e ac ions wi h Ma e ials and A oms 361, 13–19.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X15001391
Chamizo, E., L´
opez-Lo a, M., B essac, M., Le y, I., Pham, M., (2016). Excess o 236U in he
no hwes Medi e anean Sea. Science o The To al En i onmen 565, 767–776.
h p://linkinghub.else ie .com/ e ie e/pii/S0048969716308452
Chamizo, E., L´
opez-Lo a, M., (in p ess). Accele a o mass spec ome y o 236U wi h He
s ipping a he Cen o Nacional de Acele ado es. Nuclea Ins umen s and Me hods in
122
BIBLIOGRAPHY
Physics Resea ch Sec ion B: Beam In e ac ions wi h Ma e ials and A oms.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X18302659
Ch is l, M., Vockenhube , C., Kubik, P. W., Wacke , L., Lachne , J., Alfimo , V., Synal, H.-
A., (2013). The ETH Zu ich AMS acili ies: Pe o mance pa ame e s and e e ence ma e ials.
Nuclea Ins umen s and Me hods in Physics Resea ch Sec ion B: Beam In e ac ions wi h
Ma e ials and A oms 294, 29–38.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X12001577
C ans on, F., Whi e, D., (1971). The mal neu on cap u e c oss sec ions in calcium. Nuclea
Physics A 169, 95–100.
h p://linkinghub.else ie .com/ e ie e/pii/0375947471905628
CSN, (2003). INSTRUCCI´
ON IS/05, de 26 de eb e o de 2003, del Consejo de Segu idad Nu-
clea , po la que se definen los alo es de exenci´
on pa a nucleidos seg´
un se es ablece
en las ablas A y B del anexo I del Real Dec e o 1836/1999. Bole
´
ın Oficial del Es ado 86,
14221–14229.
h ps://www.boe.es/busca /doc.php?id=BOE-A-2003-7520
CSN, (n.d.). Gene al in o ma ion - Jos´
e Cab e a.
h ps://www.csn.es/en/cen al-nuclea /jose-cab e a
Denk, E., Hillegonds, D. J., Vogel, J., Synal, A., Geppe , C., Wend , K., Fa inge , K., Hen-
nessy, C., Be glund, M., Hu ell, R. F., Walczyk, T., (2006). Labeling he human skele-
on wi h 41Ca o assess changes in bone calcium me abolism. Analy ical and Bioanaly ical
Chemis y 386, 1587–1602.
h p://link.sp inge .com/10.1007/s00216-006-0795-5
Denk, E., Hillegonds, D. J., Hu ell, R. F., Vogel, J., Fa inge , K., H¨
auselmann, H. J., K aen-
zlin, M., Walczyk, T., (2007). E alua ion o 41Calcium as a New App oach o Assess Changes
in Bone Me abolism: E ec o a Bisphosphona e In e en ion in Pos menopausal Women
Wi h Low Bone Mass. Jou nal o Bone and Mine al Resea ch 22, 1518–1525.
h p://doi.wiley.com/10.1359/jbm .070617
Dillmann, I., Domingo-Pa do, C., Heil, M., K¨
appele , F., Wallne , A., Fo s ne , O., Golse ,
R., Ku sche a, W., P ille , A., S eie , P., Mengoni, A., Gallino, R., Paul, M., Vockenhu-
be , C., (2009). De e mina ion o he s ella (n,γ) c oss sec ion o 40Ca wi h accele a o
mass spec ome y. Physical Re iew C 79, 065805.
h ps://link.aps.o g/doi/10.1103/PhysRe C.79.065805
123
BIBLIOGRAPHY
D¨
obeli, M., Ko le , C., S ocke , M., Weinmann, S., Synal, H.-A., G ajca , M., Su e , M.,
(2004). Gas ioniza ion chambe s wi h silicon ni ide windows o he de ec ion and iden i-
fica ion o low ene gy ions. Nuclea Ins umen s and Me hods in Physics Resea ch Sec ion
B: Beam In e ac ions wi h Ma e ials and A oms 219-220, 415–419.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X04001211
Dunai, T. J., (2010). Cosmogenic nuclides : p inciples, concep s and applica ions in he ea h
su ace sciences. Camb idge Uni e si y P ess.
Elmo e, D., Phillips, F. M., (1987). Accele a o Mass Spec ome y o Measu emen o Long-
Li ed Radioiso opes. Science 236, 543–550.
h p://www.sciencemag.o g/cgi/doi/10.1126/science.236.4801.543
Elmo e, D., Bha acha yya, M. H., Sacco-Gibson, N., Pe e son, D. P., (1990). Calcium-41
as a long- e m biological ace o bone eso p ion. Nuclea Ins umen s and Me hods in
Physics Resea ch Sec ion B: Beam In e ac ions wi h Ma e ials and A oms 52, 531–535.
h p://linkinghub.else ie .com/ e ie e/pii/0168583X90904716
ENRESA, (n.d.). Disman ling o he Jose Cab e a Nuclea Powe Plan .
h p://www.en esa.es/eng/index/ac i i ies-and-p ojec s/disman lin
g-and-en i onmen al- es o a ion/disman ling-o - he-jose-cab e a-n
uclea -powe -plan
Field, K., Remec, I., Pape, Y. L., (2015). Radia ion e ec s in conc e e o nuclea powe plan s
–Pa I: Quan ifica ion o adia ion exposu e and adia ion e ec s. Nuclea Enginee ing and
Design 282, 126–143.
h p://linkinghub.else ie .com/ e ie e/pii/S0029549314005603
Fifield, L. K., (2008). Accele a o mass spec ome y o he ac inides. Qua e na y
Geoch onology 3, 276–290.
h p://linkinghub.else ie .com/ e ie e/pii/S1871101407000581
Fifield, L., Tims, S., Fujioka, T., Hoo, W., E e e , S., (2010). Accele a o mass spec ome y
wi h he 14UD accele a o a he Aus alian Na ional Uni e si y. Nuclea Ins umen s and
Me hods in Physics Resea ch Sec ion B: Beam In e ac ions wi h Ma e ials and A oms 268,
858–862.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X09010982
Fink, D., Paul, M., Hollos, G., Theis, S., Vog , S., S ueck, R., Engle , P., Michel, R., (1987).
Measu emen s o 41Ca spalla ion c oss sec ions and 41Ca concen a ions in he g an me-
eo i e by accele a o mass spec ome y. Nuclea Ins umen s and Me hods in Physics
124
BIBLIOGRAPHY
Resea ch Sec ion B: Beam In e ac ions wi h Ma e ials and A oms 29, 275–280.
h p://linkinghub.else ie .com/ e ie e/pii/0168583X87902497
Fink, D., Middle on, R., Klein, J., Sha ma, P., (1990).41Ca: Measu emen by accele a o
mass spec ome y and applica ions. Nuclea Ins umen s and Me hods in Physics Re-
sea ch Sec ion B: Beam In e ac ions wi h Ma e ials and A oms 47, 79–96.
h p://linkinghub.else ie .com/ e ie e/pii/0168583X9090049Z
Fink, D., Klein, J., Middle on, R., Vog , S., He zog, G. F., (1991).41Ca in i on alls, G an and
Es he ille: p oduc ion a es and ela ed exposu e age calcula ions. Ea h and Plane a y
Science Le e s 107, 115–128.
h p://linkinghub.else ie .com/ e ie e/pii/0012821X9190048M
F eeman, S., Se ass, R., King, J., Sou hon, J., Fang, Y., Woodhouse, L., Bench, G., McAn-
inch, J., (1995). Biological sample p epa a ion and 41Ca AMS measu emen a LLNL. Nu-
clea Ins umen s and Me hods in Physics Resea ch Sec ion B: Beam In e ac ions wi h Ma-
e ials and A oms 99, 557–561.
h p://linkinghub.else ie .com/ e ie e/pii/0168583X95003428
F eeman, S. P., King, J. C., Viei a, N. E., Woodhouse, L. R., Ye gey, A. L., (1997). Human
calcium me abolism including bone eso p ion measu ed wi h 41Ca ace . Nuclea Ins u-
men s and Me hods in Physics Resea ch Sec ion B: Beam In e ac ions wi h Ma e ials and
A oms 123, 266–270.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X9600701X
G´
omez-Guzm´
an, J. M., (2010).Medida de 129I en el sis ema de Espec ome
´
ıa de Masas
con Acele ado es de 1 MV del Cen o Nacional de Acele ado es. Aplicaci´
on a p oblemas
medioambien ales. Ph.D. hesis, Uni e sidad de Se illa.
G´
omez-Guzm´
an, J. M., L´
opez-Gu i´
e ez, J. M., Pin o-G´
omez, A., Holm, E., (2012).129I mea-
su emen s on he 1 MV AMS acili y a he Cen o Nacional de Acele ado es (CNA, Spain).
Applied Radia ion and Iso opes 70, 263–268.
h p://linkinghub.else ie .com/ e ie e/pii/S0969804311004040
G´
omez-Guzm´
an, J. M., Bishop, S., Faes e mann, T., Famulok, N., Fimiani, L., Hain, K.,
Jahn, S., Ko schinek, G., Ludwig, P., Rod igues, D., (2015). Acc e ion a e o ex a e es-
ial 41Ca in An a c ic snow samples. Nuclea Ins umen s and Me hods in Physics Resea ch
Sec ion B: Beam In e ac ions wi h Ma e ials and A oms 361, 620–626.
h p://linkinghub.else ie .com/ e ie e/pii/S0168583X15004875
Go e, H. E., Ful on, B. R., Elmo e, D., Li he land, A. E., Beukens, R. P., Pu se , K. H.,
Naylo , H., (1979). Radioiso ope De ec ion wi h Tandem Elec os a ic Accele a o s. IEEE
125