The Alpha Magnetic Spectrometer (AMS) on the international space station: Part II – Results from the first seven years

M. Aguilar, L. Ali Cavasonza, G. Ambrosi, L. Arruda, N. Attig, F. Barao, L. Barrin, A. Bartoloni, S. Ba¸se˘gmez-du Pree, J. Bates, R. Battiston, M. Behlmann, B. Beischer, J. Berdugo, B. Bertucci, V. Bindi, W. de Boer, K. Bollweg, B. Borgia, M.J. Boschini, M. Bourquin, E.F. Bueno, J. Burger, W.J. Burger, S. Burmeister, X.D. Cai, M. Capell, J. Casaus, G. Castellini, F. Cervelli, Y.H. Chang, G.M. Chen, H.S. Chen, Y. Chen, L. Cheng, H.Y. Chou, S. Chouridou, V. Choutko, C.H. Chung, C. Clark, G. Coignet, C. Consolandi, A. Contin, C. Corti, Z. Cui, K. Dadzie, Y.M. Dai, C. Delgado, S. Della Torre, M.B. Demirköz, L. Derome, S. Di Falco, V. Di Felice, C. Díaz, F. Dimiccoli, P. von Doetinchem, F. Dong, F. Donnini, M. Duranti, A. Egorov, A. Eline, J. Feng, E. Fiandrini, P. Fisher, V. Formato, C. Freeman, Y. Galaktionov, C. Gámez, R.J. García-López, C. Gargiulo, H. Gast, I. Gebauer, M. Gervasi, F. Giovacchini, D.M. Gómez-Coral, J. Gong, C. Goy, V. Grabski, D. Grandi,

M. Graziani, K.H. Guo, S. Haino, K.C. Han, R.K. Hashmani, Z.H. He, B. Heber, T.H. Hsieh, J.Y. Hu, Z.C. Huang, W. Hungerford, M. Incagli, W.Y. Jang, Yi Jia, H. Jinchi, K. Kanishev, B. Khiali, G.N. Kim,

Th. Kirn, M. Konyushikhin, O. Kounina, A. Kounine, V. Koutsenko, A. Kuhlman, A. Kulemzin, G. La Vacca, E. Laudi, G. Laurenti, I. Lazzizzera, A. Lebedev, H.T. Lee, S.C. Lee, C. Leluc, J.Q. Li, M. Li, Q. Li, S. Li, T.X. Li, Z.H. Li, C. Light, C.H. Lin, T. Lippert, Z. Liu, S.Q. Lu, Y.S. Lu, K. Luebelsmeyer, J.Z. Luo, S.S. Lyu, F. Machate, C. Mañá, J. Marín, J. Marquardt, T. Martin, G. Martínez, N. Masi, D. Maurin, A. Menchaca-Rocha, Q. Meng, D.C. Mo, M. Molero, P. Mott, L. Mussolin, J.Q. Ni, N. Nikonov, F. Nozzoli, A. Oliva, M. Orcinha, M. Palermo, F. Palmonari, M. Paniccia, A. Pashnin, M. Pauluzzi, S. Pensotti, H.D. Phan, V. Plyaskin, M. Pohl, S. Porter, X.M. Qi, X. Qin, Z.Y. Qu, L. Quadrani, P.G. Rancoita, D. Rapin, A. Reina Conde, S. Rosier-Lees, A. Rozhkov, D. Rozza, R. Sagdeev, S. Schael, S.M. Schmidt, A. Schulz von Dratzig, G. Schwering, E.S. Seo, B.S. Shan, J.Y. Shi, T. Siedenburg, C. Solano, J.W. Song, R. Sonnabend, Q. Sun, Z.T. Sun, M. Tacconi, X.W. Tang, Z.C. Tang, J. Tian, Samuel C.C. Ting, S.M. Ting, N. Tomassetti, J. Torsti, C. Tüysüz, T. Urban, I. Usoskin, V. Vagelli, R. Vainio, E. Valente, E. Valtonen, M. Vázquez Acosta, M. Vecchi, M. Velasco, J.P. Vialle, L.Q. Wang, N.H. Wang, Q.L. Wang, S. Wang, X. Wang, Z.X. Wang,

Received date : 5 September 2020 Accepted date : 9 September 2020

Please cite this article as: M. Aguilar, L.A. Cavasonza, G. Ambrosi et al., The Alpha Magnetic Spectrometer (AMS) on the international space station: Part II – Results from the first seven years, Physics Reports(2020), doi:https://doi.org/10.1016/j.physrep.2020.09.003.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The Alpha Magnetic Spectrometer (AMS)

1

on the International Space Station:

2

Part II – Results from the First Seven Years

3

M. Aguilar,

30

_{L. Ali Cavasonza,}

1

_{G. Ambrosi,}

36

_{L. Arruda,}

28

_{N. Attig,}

24

_{F. Barao,}

28

4

L. Barrin,

15

_{A. Bartoloni,}

42

_{S. Ba¸segmez-du Pree,}

18,a

_{J. Bates,}

21

_{R. Battiston,}

39, 40

5

M. Behlmann,

10

_{B. Beischer,}

1

_{J. Berdugo,}

30

_{B. Bertucci,}

36, 37

_{V. Bindi,}

20

_{W. de Boer,}

25

6

K. Bollweg,

21

_{B. Borgia,}

42, 43

_{M.J. Boschini,}

32

_{M. Bourquin,}

16

_{E.F. Bueno,}

18

_{J. Burger,}

10

7

W.J. Burger,

39

_{S. Burmeister,}

26

_{X.D. Cai,}

10

_{M. Capell,}

10

_{J. Casaus,}

30

_{G. Castellini,}

14

8

F. Cervelli,

38

_{Y.H. Chang,}

47, 48

_{G.M. Chen,}

6, 7

_{H.S. Chen,}

6, 7

_{Y. Chen,}

16

_{L. Cheng,}

22

9

H.Y. Chou,

48

_{S. Chouridou,}

1

_{V. Choutko,}

10

_{C.H. Chung,}

1

_{C. Clark,}

10, 21

_{G. Coignet,}

3

10

C. Consolandi,

20

_{A. Contin,}

8, 9

_{C. Corti,}

20

_{Z. Cui,}

22, 23

_{K. Dadzie,}

10

_{Y.M. Dai,}

5

_{C. Delgado,}

30

11

S. Della Torre,

32

_{M.B. Demirk¨oz,}

2

_{L. Derome,}

17

_{S. Di Falco,}

38

_{V. Di Felice,}

44,b

_{C. D´ıaz,}

30

12

F. Dimiccoli,

39

_{P. von Doetinchem,}

20

_{F. Dong,}

34

_{F. Donnini,}

44,b

_{M. Duranti,}

36

_{A. Egorov,}

10

13

A. Eline,

10

_{J. Feng,}

10

_{E. Fiandrini,}

36, 37

_{P. Fisher,}

10

_{V. Formato,}

44,b

_{C. Freeman,}

20

14

Y. Galaktionov,

10

_{C. G´amez,}

30

_{R.J. Garc´ıa-L´opez,}

27

_{C. Gargiulo,}

15

_{H. Gast,}

1

_{I. Gebauer,}

25

15

M. Gervasi,

32, 33

_{F. Giovacchini,}

30

_{D. M. G´omez-Coral,}

20

_{J. Gong,}

34

_{C. Goy,}

3

_{V. Grabski,}

31

16

D. Grandi,

32, 33

_{M. Graziani,}

36, 37

_{K.H. Guo,}

19

_{S. Haino,}

47

_{K.C. Han,}

29

_{R.K. Hashmani,}

2

17

Z.H. He,

19

_{B. Heber,}

26

_{T.H. Hsieh,}

10

_{J.Y. Hu,}

6, 7

_{Z.C. Huang,}

19

_{W. Hungerford,}

21

18

M. Incagli,

38

_{W.Y. Jang,}

13

_{Yi Jia,}

10

_{H. Jinchi,}

29

_{K. Kanishev,}

39

_{B. Khiali,}

44,b

_{G.N. Kim,}

13

19

Th. Kirn,

1

_{M. Konyushikhin,}

10

_{O. Kounina,}

10

_{A. Kounine,}

10

_{V. Koutsenko,}

10

_{A. Kuhlman,}

20

20

A. Kulemzin,

10

_{G. La Vacca,}

32, 33

_{E. Laudi,}

15

_{G. Laurenti,}

8

_{I. Lazzizzera,}

39, 40

_{A. Lebedev,}

10

21

H.T. Lee,

46

_{S.C. Lee,}

47

_{C. Leluc,}

16

_{J.Q. Li,}

34

_{M. Li,}

1

_{Q. Li,}

34

_{S. Li,}

1

_{T.X. Li,}

19

_{Z.H. Li,}

6

22

C. Light,

20

_{C.H. Lin,}

47

_{T. Lippert,}

24

_{Z. Liu,}

16

_{S.Q. Lu,}

19

_{Y.S. Lu,}

6

_{K. Luebelsmeyer,}

1

23

J.Z. Luo,

34

_{S.S. Lyu,}

19

_{F. Machate,}

1

_{C. Ma˜}

_n´a,

30

_{J. Mar´ın,}

30

_{J. Marquardt,}

26

_{T. Martin,}

10, 21

24

G. Mart´ınez,

30

_{N. Masi,}

8, 9

_{D. Maurin,}

17

_{A. Menchaca-Rocha,}

31

_{Q. Meng,}

34

_{D.C. Mo,}

19

25

M. Molero,

30

_{P. Mott,}

10, 21

_{L. Mussolin,}

36, 37

_{J.Q. Ni,}

19

_{N. Nikonov,}

1

_{F. Nozzoli,}

39

26

A. Oliva,

8

_{M. Orcinha,}

28

_{M. Palermo,}

20

_{F. Palmonari,}

8, 9

_{M. Paniccia,}

16

_{A. Pashnin,}

10

27

M. Pauluzzi,

36, 37

_{S. Pensotti,}

32, 33

_{H.D. Phan,}

10

_{V. Plyaskin,}

10

_{M. Pohl,}

16

_{S. Porter,}

21

28

X.M. Qi,

19

_{X. Qin,}

10

_{Z.Y. Qu,}

47

_{L. Quadrani,}

8, 9

_{P.G. Rancoita,}

32

_{D. Rapin,}

16

_{A. Reina}

29

Conde,

27

_{S. Rosier-Lees,}

3

_{A. Rozhkov,}

10

_{D. Rozza,}

32, 33

_{R. Sagdeev,}

11

_{S. Schael,}

1

30

S.M. Schmidt,

24

_{A. Schulz von Dratzig,}

1

_{G. Schwering,}

1

_{E.S. Seo,}

12

_{B.S. Shan,}

4

_{J.Y. Shi,}

34

31

T. Siedenburg,

1

_{C. Solano,}

10

_{J.W. Song,}

23

_{R. Sonnabend,}

1

_{Q. Sun,}

23

_{Z.T. Sun,}

6, 7

32

M. Tacconi,

32, 33

_{X.W. Tang,}

6

_{Z.C. Tang,}

6

_{J. Tian,}

36, 37

_{Samuel C.C. Ting,}

10, 15

33

S.M. Ting,

10

_{N. Tomassetti,}

36, 37

_{J. Torsti,}

49

_{C. T¨}

_uys¨

_uz,

2

_{T. Urban,}

10, 21

_{I. Usoskin,}

35

34

V. Vagelli,

41, 36

_{R. Vainio,}

49

_{E. Valente,}

42, 43

_{E. Valtonen,}

49

_{M. V´azquez Acosta,}

27

35

M. Vecchi,

18

_{M. Velasco,}

30

_{J.P. Vialle,}

3

_{L.Q. Wang,}

23

_{N.H. Wang,}

23

_{Q.L. Wang,}

5

36

S. Wang,

20

_{X. Wang,}

10

_{Z.X. Wang,}

19

_{J. Wei,}

16

_{Z.L. Weng,}

10

_{H. Wu,}

34

_{R.Q. Xiong,}

34

37

W. Xu,

22, 23

_{Q. Yan,}

10

_{Y. Yang,}

45

_{H. Yi,}

34

_{Y.J. Yu,}

5

_{Z.Q. Yu,}

6

_{M. Zannoni,}

32, 33

_{C. Zhang,}

6

38

F. Zhang,

6

_{F.Z. Zhang,}

6, 7

_{J.H. Zhang,}

34

_{Z. Zhang,}

10

_{F. Zhao,}

6, 7

_{Z.M. Zheng,}

4

39

H.L. Zhuang,

6

_{V. Zhukov,}

1

_{A. Zichichi,}

8, 9

_{N. Zimmermann,}

1

_{and P. Zuccon}

39, 40

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Jour

nal

Pr

e-pr

oof

2

_{Department of Physics, Middle East Technical University (METU), 06800 Ankara, Turkey}

43

3

_{Univ. Grenoble Alpes, Univ. Savoie Mont Blanc,}

44

CNRS, LAPP-IN2P3, 74000 Annecy, France

45

4

_{Beihang University (BUAA), Beijing, 100191, China}

46

5

_{Institute of Electrical Engineering (IEE), Chinese}

47

Academy of Sciences, Beijing, 100190, China

48

6

_{Institute of High Energy Physics (IHEP), Chinese}

49

Academy of Sciences, Beijing, 100049, China

50

7

_{University of Chinese Academy of Sciences (UCAS), Beijing, 100049, China}

51

8

_{INFN Sezione di Bologna, 40126 Bologna, Italy}

52

9

_{Universit`a di Bologna, 40126 Bologna, Italy}

53

10

_{Massachusetts Institute of Technology (MIT), Cambridge, Massachusetts 02139, USA}

54

11

_{East–West Center for Space Science, University}

55

of Maryland, College Park, Maryland 20742, USA

56

12

_{IPST, University of Maryland, College Park, Maryland 20742, USA}

57

13

_{CHEP, Kyungpook National University, 41566 Daegu, Korea}

58

14

_{CNR–IROE, 50125 Firenze, Italy}

59

15

_{European Organization for Nuclear Research (CERN), 1211 Geneva 23, Switzerland}

60

16

_{DPNC, Universit´e de Gen`eve, 1211 Gen`eve 4, Switzerland}

61

17

_{Univ. Grenoble Alpes, CNRS, Grenoble INP, LPSC-IN2P3, 38000 Grenoble, France}

62

18

_{Kapteyn Astronomical Institute, University of Groningen,}

63

P.O. Box 800, 9700 AV Groningen, The Netherlands

64

19

_{Sun Yat–Sen University (SYSU), Guangzhou, 510275, China}

65

20

_{Physics and Astronomy Department, University of Hawaii, Honolulu, Hawaii 96822, USA}

66

21

_{National Aeronautics and Space Administration Johnson}

67

Space Center (JSC), Houston, Texas 77058, USA

68

22

_{Shandong Institute of Advanced Technology (SDIAT), Jinan, Shandong, 250100, China}

69

23

_{Shandong University (SDU), Jinan, Shandong, 250100, China}

70

24

_J¨

_{ulich Supercomputing Centre and JARA-FAME,}

71

Research Centre J¨

ulich, 52425 J¨

ulich, Germany

72

25

_{Institut f¨}

_{ur Experimentelle Teilchenphysik, Karlsruhe}

73

Institute of Technology (KIT), 76131 Karlsruhe, Germany

74

26

_{Institut f¨}

_{ur Experimentelle und Angewandte Physik,}

75

Christian-Alberts-Universit¨at zu Kiel, 24118 Kiel, Germany

76

27

_{Instituto de Astrof´ısica de Canarias (IAC), 38205 La Laguna, and Departamento}

77

de Astrof´ısica, Universidad de La Laguna, 38206 La Laguna, Tenerife, Spain

78

28

_{Laborat´orio de Instrumenta¸c˜ao e F´ısica Experimental}

79

de Part´ıculas (LIP), 1649-003 Lisboa, Portugal

80

29

_{National Chung–Shan Institute of Science and}

81

Technology (NCSIST), Longtan, Tao Yuan, 32546, Taiwan

82

30

_{Centro de Investigaciones Energ´eticas, Medioambientales}

83

y Tecnol´ogicas (CIEMAT), 28040 Madrid, Spain

84

31

_{Instituto de F´ısica, Universidad Nacional Aut´onoma}

85

de M´exico (UNAM), Ciudad de M´exico, 01000 Mexico

86

32

_{INFN Sezione di Milano–Bicocca, 20126 Milano, Italy}

87

33

_{Universit`a di Milano–Bicocca, 20126 Milano, Italy}

88

34

_{Southeast University (SEU), Nanjing, 210096, China}

89 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Jour

nal

Pr

e-pr

oof

35

_{Sodankyl¨a Geophysical Observatory and Space Physics and}

90

Astromony Research Unit, University of Oulu, 90014 Oulu, Finland

91

36

_{INFN Sezione di Perugia, 06100 Perugia, Italy}

92

37

_{Universit`a di Perugia, 06100 Perugia, Italy}

93

38

_{INFN Sezione di Pisa, 56100 Pisa, Italy}

94

39

_{INFN TIFPA, 38123 Povo, Trento, Italy}

95

40

_{Universit`a di Trento, 38123 Povo, Trento, Italy}

96

41

_{Agenzia Spaziale Italiana (ASI), 00133 Roma, Italy}

97

42

_{INFN Sezione di Roma 1, 00185 Roma, Italy}

98

43

_{Universit`a di Roma La Sapienza, 00185 Roma, Italy}

99

44

_{INFN Sezione di Roma Tor Vergata, 00133 Roma, Italy}

100

45

_{National Cheng Kung University, Tainan, 70101, Taiwan}

101

46

_{Academia Sinica Grid Center (ASGC), Nankang, Taipei, 11529, Taiwan}

102

47

_{Institute of Physics, Academia Sinica, Nankang, Taipei, 11529, Taiwan}

103

48

_{Physics Department and Center for High Energy and High Field}

104

Physics, National Central University (NCU), Tao Yuan, 32054, Taiwan

105

49

_{Space Research Laboratory, Department of Physics and}

106

Astronomy, University of Turku, 20014 Turku, Finland

107

Abstract

108

The Alpha Magnetic Spectrometer (AMS) is a precision particle physics detector on the

Inter-109

national Space Station (ISS) conducting a unique, long-duration mission of fundamental physics

110

research in space. The physics objectives include the precise studies of the origin of dark matter,

111

antimatter, and cosmic rays as well as the exploration of new phenomena. Following a 16-year

112

period of construction and testing, and a precursor flight on the Space Shuttle, AMS was installed

113

on the ISS on May 19, 2011. In this report we present results based on 120 billion charged cosmic

114

ray events up to multi-TeV energies. This includes the fluxes of positrons, electrons, antiprotons,

115

protons, and nuclei. These results provide unexpected information, which cannot be explained

116

by the current theoretical models. The accuracy and characteristics of the data, simultaneously

117

from many different types of cosmic rays, provide unique input to the understanding of origins,

118

acceleration, and propagation of cosmic rays.

119 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Jour

nal

Pr

e-pr

oof

CONTENTS

120

Introduction

6

121

I. AMS on the ISS

8

122

A. Permanent Magnet

10

123

B. Tracker

11

124

1. Tracker coordinate measurements

11

125

2. Tracker rigidity measurement

15

126

3. Determination of tracker absolute rigidity scale and alignment

17

127

4. Charge-sign identification

20

128

5. Tracker charge measurement

22

129

C. Transition Radiation Detector (TRD)

25

130

D. Time of Flight Counters (TOF)

28

131

E. Anticoincidence Counters (ACC)

32

132

F. Ring Imaging Cherenkov Counter (RICH)

33

133

G. Electromagnetic Calorimeter (ECAL)

34

134

1. ECAL shower reconstruction

36

135

2. ECAL energy reconstruction

37

136

3. Proton rejection using ECAL

39

137

H. Trigger and Data Acquisition

41

138

II. Origins of Cosmic Positrons

45

139

III. Origins of Cosmic Electrons

61

140

IV. Cosmic Protons

74

141

V. Cosmic Antiprotons

77

142

VI. Properties of Cosmic Elementary Particles

82

143

VII. Nuclear Cross Section Measurements

90

144

VIII. Primary Helium, Carbon, and Oxygen Fluxes

96

145

IX. Proton-to-Helium Flux Ratio

100

146

X. Secondary Lithium, Beryllium, Boron Fluxes

147

and Secondary to Primary Ratios

102

148

XI. Properties of Cosmic Helium Isotopes

111

149

XII. Cosmic Nitrogen Flux

116

150

XIII. Primary Neon, Magnesium, and Silicon Fluxes

124

151

XIV. Strangelets

133

152

XV. Time-Dependent Proton and Helium Fluxes

137

153 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Jour

nal

Pr

e-pr

oof

XVI. Time-Dependent Electron and Positron Fluxes

143

154

XVII. Summary

151

155

Acknowledgments

152

156

References

153

157 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Jour

nal

Pr

e-pr

oof

INTRODUCTION

158

The Alpha Magnetic Spectrometer (AMS) is a precision particle physics detector on the

159

International Space Station (ISS) conducting a unique, long-duration mission of fundamental

160

physics research in space. The physics objectives include precise studies of the origins of dark

161

matter, antimatter, and cosmic rays as well as the exploration of new phenomena. Prior

162

to the main mission, a precursor flight of AMS, AMS-01, was flown on the Space Shuttle

163

Discovery in June 1998. AMS-01 results were published in Physics Reports, “The Alpha

164

Magnetic Spectrometer (AMS) on the International Space Station: Part I – results from the

165

test flight on the space shuttle” [

1

]. With the experience accumulated from AMS-01, a new

166

state of the art detector, known as AMS-02 or AMS on the ISS, was built and launched on

167

the Space Shuttle Endeavour and installed on the ISS on May 19, 2011 (see Figure

1

). The

168

ISS is orbiting the Earth at an altitude of

_{∼ 410 km with an orbit inclination of 52 degrees.}

169

AMS

FIG. 1. AMS is a unique precision magnetic spectrometer on the ISS. AMS will operate on the ISS for the Station’s lifetime. It is mounted on the ISS with a 12 degree angle to the zenith to prevent that the rotating ISS solar arrays are in the AMS field of view.

There are two kinds of cosmic rays in space:

170

1. Neutral cosmic rays (light rays and neutrinos) that have been studied by many

satel-171

lite (COBE [

2

], EGRET [

3

], WMAP [

4

], Planck [

5

], ROSAT [

6

], Fermi [

7

],

AG-172

ILE [

8

], Chandra [

9

], INTEGRAL [

10

], the Hubble Space Telescope [

11

] and the coming

173 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Jour

nal

Pr

e-pr

oof

James Webb Telescope [

12

], . . . ) and ground based (ARGO-YBJ [

13

], HAWC [

14

],

174

H.E.S.S. [

15

], MAGIC [

16

], IceCube [

17

], Tibet ASgamma [

18

], LHAASO [

19

], . . . )

175

experiments. Studies of neutral cosmic rays have provided fundamental information

176

about the universe;

177

2. Charged cosmic rays are absorbed in the Earth’s atmosphere and therefore, in the

178

energy range up to multi-TeV, their properties such as charge magnitude and sign,

179

mass, and rigidity can only be studied in space. In the past there have been many

180

excellent experiments with balloons (ATIC [

20

], BESS [

21

], CAPRICE/WiZard [

22

],

181

CREAM [

23

], . . . ) and with satellites (CRIS [

24

], HEAO [

25

], PAMELA [

26

], . . . ).

182

Recently, precision non-magnetic, calorimeter experiments in space (CALET [

27

],

ISS-183

CREAM [

28

], and DAMPE [

29

]) have also begun to provide important results.

184

There are excellent ongoing and proposed experiments that study cosmic rays at the

185

highest energies. This includes innovative experiments EUSO-SPB [

30

] and POEMMA [

31

],

186

as well as ground-based experiments – the Pierre Auger Observatory [

32

], H.E.S.S. [

15

],

187

LHAASO [

19

], KASCADE-Grande [

33

], TA [

34

], CTA [

35

], and others.

188

AMS is the first long duration (about two 11-year solar cycles), large acceptance precision

189

magnetic spectrometer to measure the sign and value of the charge, the momentum, the

190

rigidity, and the flux of elementary particles (positrons, electrons, antiprotons, protons),

191

nuclei, and anti-nuclei directly in space.

192 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Jour

nal

Pr

e-pr

oof

I. AMS ON THE ISS

193

As seen in Figure

2

, the AMS detector [

36

] consists of a permanent magnet and an

194

array of particle detectors to measure the velocity β = v/c, momentum P , charge Z, and

195

rigidity R = P/Z of traversing particles and nuclei. Within the magnet bore and above and

196

below the magnet are a total of 9 precision silicon tracker layers, L1 to L9. The Transition

197

Radiation Detector (TRD) is located at the top of the AMS. Above the magnet bore are

198

two orthogonal planes of Time of Flight counters (TOF), the Upper TOF, and below the

199

bore are another two orthogonal planes, the Lower TOF. The Anti-Coincidence Counters

200

(ACC or Veto), surround the tracker within the magnet bore. Below the Lower TOF is the

201

Ring Imaging Cherenkov counter (RICH) and below that the Electromagnetic Calorimeter

202

(ECAL). Each of these detector elements is described below.

203 TRD Upper TOF Tr a c k e r Lower TOF RICH ECAL 1 2 7-8 3-4 9 5-6 TRD identify e+_{, e} -Silicon Tracker measure Z, P ECAL measure E of e+_{, e}

-Upper TOF measure Z, E

Magnet identify ±Z, P

RICH measure Z, E

Lower TOF measure Z, E x

y z

B

ACC reject side particles

FIG. 2. The AMS detector showing the main elements and their functions. AMS is a TeV pre-cision, multipurpose particle physics magnetic spectrometer. It identifies particles and nuclei by their charge (Z), energy (E) and momentum (P ) or rigidity (R = P/Z), which are measured inde-pendently by the Tracker, TOF, RICH and ECAL. The ACC counters, located in the magnet bore, are used to reject particles entering AMS from the side. The AMS coordinate system, concentric with the magnet, is also shown. The x axis is parallel to the main component of the magnetic field and the z axis is pointing vertically.

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Jour

nal

Pr

e-pr

oof

The separation of the TRD and the ECAL by a magnetic field ensures that most of the

204

secondary particles generated in the TRD are swept away by the magnet and do not enter

205

into the ECAL. In this way, the rejection powers of the TRD and the ECAL are independent.

206

Before launch to the ISS, AMS was tested extensively at the CERN test beam with

207

electrons, positrons, protons, and pions. Pion beams were used to simulate transition

radi-208

ation effects in the TRD for high energy protons. In total, more than 2000 combinations of

209

particles, energies, incident angles, and locations were tested.

210

AMS on the ISS has functioned reliably and the properties of the detector are continuously

211

monitored. Minute changes compared to the original calibration at the CERN accelerator

212

before launch are corrected in the data analysis. This ensures the quality and accuracy of

213

the data.

214

During more than 10 years of AMS construction, a large international group of physicists

215

have developed a comprehensive Monte Carlo simulation program for AMS. This program is

216

based on the geant4 package [

37

] and it is constantly being improved with the AMS data

217

obtained on the ISS. This Monte Carlo program simulates electromagnetic, hadronic, and

218

nuclear interactions of particles and nuclei in the material of AMS, namely:

219

1. Electromagnetic interactions including

220

• Ionization losses

221

• Bremsstrahlung

222

• Pair Production

223

• Multiple and single Coulomb scattering

224

2. X-ray generation for the TRD detector

225

3. Cherenkov photon generation and propagation for the RICH detector

226

4. Elastic hadronic and nuclear scattering including

227

• Hadron elastic scattering

228

• Elastic nuclear scattering for ions

229

5. Inelastic hadronic and nuclear interactions from theoretical models tuned according

230

to our nuclear cross section measurements (see Section

VII

).

231

The trigger conditions and digitization of the signals are simulated according to the

232

measured characteristics of the electronics. The simulated events then undergo the same

233

reconstruction as used for the data.

234

The AMS coordinate system is concentric with the center of the magnet. The x-axis is

235

parallel to the main component of the magnetic field and the z-axis is parallel to the magnet

236 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Jour

nal

Pr

e-pr

oof

A. Permanent Magnet

241

The permanent magnet [

38

], shown in Figure

3

a, was flown on the AMS engineering

242

flight in 1998 and was described in detail in the AMS-01 Physics Report [

1

]. It is made of

243 244

64 high-grade Nd-Fe-B sectors assembled in a cylindrical shell 800 mm long with an inner

245

diameter of 1115 mm. This configuration produces a field of 1.4 kG at the center of the

246

magnet and negligible dipole moment and field leakage outside the magnet (see Figure

3

b).

247

This is important in order to eliminate the effect of torque on the ISS and to ensure safety

248

of astronauts. The detailed 3-dimensional field of the magnet was mapped in 2010 and is

249

shown in Figure

3

c. The field was measured in 120,000 locations to an accuracy of better

250

than 1%. Comparison with the measurements performed with the same magnet in 1997,

251

before the engineering flight AMS-01, shows that the field did not change within 1%, limited

252

by the accuracy of the 1997 measurement, as shown in Figure

3

d.

253

On the ISS, slight temperature induced changes in the field are constantly monitored and

254

corrected in the analysis.

255 1 2 3 4 5 6 7 89 10 1112 1314 15 Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α Α

b)

a)

Phi Angle (deg) 0 100 200 300 Rad_ius (c m) 0 10 20 30 40 F ie ld (k G a u s s ) 0 0.5 1 1.5 2 0 0.5 1 1.5 2

c)

Phi Angle (deg)

0 100 200 300 Rad ius _(c m) 0 10 20 30 D e v ia ti o n (% ) 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1

d)

FIG. 3. (a) The magnet being prepared for the AMS mission on the ISS. The magnet is 800 mm long with an inner diameter of 1115 mm. (b) The arrangement of the AMS magnet showing the field directions of the 64 permanent magnet sectors resulting in negligible dipole moment and field leakage outside the magnet. (c) The magnet field measured in 2010 at z = 0. (d) The deviation between the field measurements in 1997 and 2010. The Radius and Phi Angle are cylindrical coordinates in the x_{− y plane, such that Phi=0 corresponds to the x-axis.}

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Jour

nal

Pr

e-pr

oof

B. Tracker

256

The tracker, together with the magnet, accurately determines the trajectory and charge

257

Z of cosmic rays by multiple measurements of the coordinates (x and y) and energy loss. It is

258

composed of 192 ladders, each containing double-sided silicon sensors, readout electronics,

259

and mechanical support [

38

,

39

]. The AMS tracker has nine layers, L1–L9, and 196,608

260

readout channels. As shown in Figure

4

a, three planes of aluminum honeycomb with carbon

261

fiber skins are equipped with ladders on both sides. The layers on these planes are numbered

262

L3–L8. Another three planes are equipped with one layer of ladders each, numbered L1,

263

L2, and L9. As also indicated in Figure

4

a, L1 is located on top of the TRD, L2 is above

264

the magnet, and L9 is between the RICH and the ECAL. L9 covers the ECAL acceptance.

265

L2–L8 constitute the inner tracker. The total lever arm of the tracker from L1 to L9 is 3.0

266

m. From the trajectory and the magnetic field map, the tracker measurement directly yields

267

the rigidity or momentum per unit charge, R = P/Z.

268

Positions of the planes of the inner tracker are held stable by a special carbon fiber

269

structure [

1

]. It is monitored using 20 IR laser beams, which penetrate through all planes of

270

the inner Tracker and provide micron level accuracy position measurements. The positions

271

of L1 and L9 are aligned using cosmic ray protons every 2 minutes. As seen in Figures

4

b

272

and

4

c, they are stable to 2–3 µm.

273 L1 L5 L6 L3 L4 L7 L8 L9 L2 L1 L2 L9 L3−L4 L5−L6 L7−L8 -50 0 50 100 m) μ Residual in Y ( -20 -10 0 10 20 -20 -10 0 10 20 En tr ie s En tr ie s 0 1000 2000 3000 4000 m μ = 2.2 σ Ma y J u l Se p N o v J a n Ma r Ma y J u l Se p N o v J a n Ma r Ma y J u l Se p N o v J a n Ma r Ma y J u l Se p N o v J a n Ma y Ma r J u l Se p N o v J a n Ma r Ma y J u l Se p N o v J a n Ma r Ma y J u l Se p N o v J a n Ma r Ma y m ) μ R e s id u a l in Y ( m ) μ R e s id u a l in Y ( -50 0 50 100 0 1000 2000 3000 σ= 2.3 μm 2011 2012 2013 2014 2015 2016 2017 2018 b) Layer 1 c) Layer 9 a)

FIG. 4. (a) The 9 layers of the AMS silicon tracker and their locations within the detector. The alignment stability of (b) Layer 1 and (c) Layer 9 over seven years.

1. Tracker coordinate measurements

274 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Jour

nal

Pr

e-pr

oof

of the rigidity (or momentum) [

40

]. This provides spatial resolution of 5

_{− 10µm in the}

279

bending plane and 13

_{− 20µm in the non-bending plane.}

280

We present the method of determination of the y coordinate. The method of

determina-281

tion of the x coordinate is similar.

282

When a charged particle crosses a layer of the silicon tracker, its coordinate d (either

283

in x or y) is determined by taking the ratio between the signals induced on the two strips

284

between which the particle passed (see Figure

5

).

285

The amplitude of the induced signals should be proportional to Z

2

_{. For high Z, the}

286

amplitudes gradually saturate and become non-linear. This causes coordinate resolution

287

degradation.

288

We have developed an optimal technique to correct for the non-linear effect. As illustrated

289

in Figure

5

, we identify a function K(d) which will restore the linearity of the amplitudes

290

by taking into account measured amplitudes A

0

, A

1

and the corrected position d:

291

d = 1/(1 + (A

0

+ K(d))/A

1

).

(1)

{

{

A

₁

A

₀

K(d)

_Corrected

Uncorrected

0

1

_d

FIG. 5. Schematic of particle coordinate measurement in the silicon tracker. The solid red arrow indicates the point where the particle intersected the silicon sensor, the two adjacent strips are positioned at 0 and 1 in “strip” units. The maximum amplitude is defined as A0 (at 0) and the

next largest adjacent strip as A1 (at 1). For an ideal tracker, the ratio between amplitude A0

and A1 gives the particle coordinate, d = 1/(1 + A0/A1). To correct for the non-linear effect, the

correction function K(d) is added to the measured non-linear amplitude A0. The dashed arrow

shows the uncorrected coordinate measurement. The corrected coordinate is at its true position within the measured accuracy of few microns.

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Jour

nal

Pr

e-pr

oof

To determine K(d), we used the fact that cosmic rays are uniform and isotropic, so for

292

an ideal tracker, without nonlinearity, we should see a uniform event position density (see

293

Figure

6

a), while in the non-linear case the event density is distorted (see Figure

6

b).

294

The function K(d) was found to be different for different Z. Figure

6

c shows the event

295

position density distribution for carbon nuclei before and after correction. With this

correc-296

tion, the observed coordinate accuracy in the bending plane is 6.5 µm for helium, 5.1 µm

297

for carbon, and 6.3 µm for oxygen.

298

0.5

-0.5

0

d

Tracker linear response

Tracker non-linear response

Uniform Event Position Density

Distorted Event Position Density

0.5

-0.5

0

d

a)

b)

Position -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 Ev e n ts D e n s ity 0 0.01 0.02 0.03 0.04

A1

A0

A1

c)

FIG. 6. (a) The uniform event position density for an ideal tracker. (b) The distorted event position density for the tracker with non-linear amplitude response. (c) The event position density before correction (black curve) and after correction (red curve) for carbon. The strips are positioned at 0 for the strip with the amplitude A0, and at 1 or -1 for A1.

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Jour

nal

Pr

e-pr

oof

Figure

7

shows the coordinate measurement accuracy for nuclei from Z = 2 to Z = 26

299

for two different track geometries: L1–L9 and L1–L8. Note that, the L1–L9 geometry has

300

tracks with almost normal incidence and therefore has the best coordinate resolution, while

301

the L1–L8 geometry has tracks with larger inclination angles. Due to the design of the

302

tracker readout amplifier, the maximum non-linearity occurs for Z

_{∼ 9.}

303 2 4 6 8 10 12 14 16 m] μ [ σ R e si d u a ls 3 4 5 6 7 8 9 10 11 L1-L9 Geometry Z 5 10 15 20 25 m] μ [ σ R e si d u a ls 3 4 5 6 7 8 9 10 11 L1-L8 Geometry a) b) Z

FIG. 7. AMS tracker residuals σ as functions of nuclei charge Z for (a) the L1–L9 geometry and (b) the L1–L8 geometry. The residuals were obtained by comparison of the differences of the coordinates measured in layers L3 or L5 to those obtained from the track fit using the measurements from L1, L2, L4, L6, L7, and L8 in the rigidity range R > 50 GV. Due to the design of the tracker readout amplifier, the maximum non-linearity occurs for Z _{∼ 9.}

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Jour

nal

Pr

e-pr

oof

2. Tracker rigidity measurement

304

Together, the tracker and the magnet determine the rigidity R of charged cosmic rays

305

by measuring the particle trajectory in the magnetic field. To find the particle trajectory

306

we use a track-finding algorithm based on cellular automaton for finding the track segments

307

and then constructing the track, as illustrated in Figure

8

. This improves the track finding

308

efficiency and rejection of spurious hits in the detector [

41

].

309

FIG. 8. Schematics of the track-finding algorithm. (a) Event with original hits. (b) Event with all track segments. (c) Correct segments chosen based on minimization of angles between segments. (d) Chose one track as the result of using a χ2-like track quality estimator.

This algorithm is particularly important for heavy nuclei events, in which additional hits

310

or track segments are often present due to delta-ray generation and nuclei interactions with

311

the tracker materials.

312

Once a track has been found, its rigidity is determined using a track-fitting algorithm

313

based on the Kalman filtering technique. It accurately accounts for energy losses and

mul-314

tiple scattering by charged particles [

42

]. For low rigidities (< 20 GV) the ∆R/R is 0.1.

315

This is particularly important for the measurements of nuclear isotopes. The maximum

de-316

tectable rigidity (MDR) with this algorithm is 2.0 TV for protons, 3.2 TV for helium, 3.7 TV

317

for carbon, 3.4 TV for oxygen, and 3.7 TV for iron.

318

Test beam data are important in understanding the tracker performance. Figure

9

a

319

shows the comparison between data and the Monte Carlo simulation of the inverse rigidity

320

measured by the tracker for 400 GeV/c protons from the CERN test beam. As seen, the

321

Monte Carlo simulation describes not only detector resolution effects (the central part of

322

the distribution) but also the effects of interactions with the detector materials including

323

multiple, large angle, elastic, and quasi-elastic scattering (the tails of the distribution).

324

Figure

9

shows that the agreement between the data and the Monte Carlo simulation extends

325

over five orders of magnitude. To study the tracker performance beyond the test beam

326

momentum (400 GeV/c proton beam), we used the data from the ISS to compare the rigidity

327 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Jour

nal

Pr

e-pr

oof

-0.02

-0.01

0

0.01

0.02

Ev

e

n

ts

10

2

10

3

10

4

10

5

10

6

10

400 GeV/c Test Beam

400 GeV/c Simulation

1

Rigidity

400

[GV ]

1

-1

]

-1

[TV

L2−9

R

1

-

L1−8

R

1

-20

-15

-10

-5

0

5

10

15

20

E

v

e

n

ts

1

10

2

10

3

10

1130 < R < 1800 GV

_{p data}

p MC

a)

b)

FIG. 9. (a) Comparison between data and the Monte Carlo simulation of the inverse rigidity measured by the tracker for 400 GV test beam protons. As seen, the agreement between the data and the Monte Carlo simulation extends over five orders of magnitude. (b) The difference of the inverse rigidities measured with the upper (Layers L1–L8) and the lower (Layers L2–L9) parts of the tracker for the cosmic ray proton data collected on the ISS and for the Monte Carlo simulation in the rigidity range [1130-1800] GV.

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Jour

nal

Pr

e-pr

oof

3. Determination of tracker absolute rigidity scale and alignment

332

In AMS, for all Z, the largest systematic error in the determination of the fluxes at the

333

highest energies is due to the uncertainty of the absolute rigidity scale. The AMS tracker

334

alignment and the absolute rigidity scale determination were performed before launch using

335

the CERN test beam data, as shown in Figure

9

a. Vibrations and accelerations on the ISS

336

due to movement of solar arrays, the attitude change, docking and undocking of the visiting

337

vehicles, astronaut activities, and during the AMS launch into space as well as outgassing

338

of the carbon fiber supporting structure in vacuum, may change the ladder positions of

339

the inner tracker at the sub-micron level, and therefore cause shifts in the absolute rigidity

340

scale. Note, a coherent shift in the inner tracker layers of less than 0.5 microns is sufficient

341

to create an absolute rigidity scale shift of 10% at 1 TV.

342

The in-flight rigidity scale shift s and its uncertainty were obtained by the comparison of

343

the inverse absolute rigidity 1/

_{|R|, measured by the tracker, with the inverse energy 1/E,}

344

measured by the electromagnetic calorimeter, for positron events and electron events [

43

].

345

• First, the electron events and positron events were split into 72 energy bins from 2 to

346

300 GeV, with bin widths chosen according to the calorimeter energy resolution.

347

• Next, probability density function (PDF) for each bin were calculated from the 1/|R|−

348

1/E electron distributions. Figure

10

shows the 1/

_{|R| − 1/E distribution for electron}

349

events and its parametrization together with the positron events in the 11 to 13 GeV

350

bin.

351

• Each PDF is then parametrized by the sum of a Gaussian function and an

exponen-352

tially modified Gaussian function.

353

• The bin-by-bin PDF distributions are parametrized as functions of energy.

354

• The resulting energy dependent PDF, f(1/|R| − 1/E, E), is then used in an unbinned

355

likelihood fit of the rigidity scale shift parameters, namely s

+

_{for positrons and s}

−

_for

356

electrons.

357

• The likelihood is defined as

P

+

logf (1/

_{|R| + s}

+

_{− 1/E, E) +}

P

−

_{logf (1/}

_{|R| + s}

−

₋

358

1/E, E), where the summations include all positron events or electron events,

respec-359

tively.

360

• The absolute rigidity scale shift is then evaluated as s = (s

+

_{− s}

−

_)/2.

361 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Jour

nal

Pr

e-pr

oof

]

-1

[TV

E

1

-

|R|

1

-20

-10

0

10

20

30

40

Eve

n

ts

100

11<E<13 GeV

+

e

11<E<13 GeV

-e

parameterization

-e

11<E<13 GeV

+

e

11<E<13 GeV

-e

parameterization

-e

20

200

FIG. 10. The 1/|R| − 1/E distribution for the energy bin 11 to 13 GeV for electrons (red data points) and positrons (blue data points). The distribution for electrons is parametrized by the sum (red curve) of a Gaussian and an exponentially modified Gaussian with a χ2/d.o.f. = 65/68.

As mentioned above, various effects at launch and in space may modify rigidity scale

362

and bias flux measurements. To determine the time-dependent rigidity shift correction, the

363

electron-positron data from the first 7 years of operations is divided into four time intervals.

364

Figure

11

shows the time dependence of the rigidity scale correction for the period from May

365

2011 to May 2018. With this time dependent correction, the accuracy of the rigidity scale

366

shift is found to be within 0.033 TV

−1

_{or 3% at 1 TV, limited mostly by available positron}

367

statistics.

368 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Jour

nal

Pr

e-pr

oof

]

-1

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

Year

2013

2015

2017

R

ig

id

ity

s

c

a

le

c

o

rr

e

c

ti

o

n

[

T

V

FIG. 11. The time dependence of the rigidity scale correction for the period from May 2011 to May 2018.

The positions of L1 and L9 are dynamically aligned every 2 minutes according to

extrap-369

olations from the inner tracker (layers L2 to L8). Therefore, the rigidity scale of the full

370

span tracker with layers from L1 to L9 follows the rigidity scale of the inner tracker. To

371

verify this, the difference in the rigidity scale shifts between the full span tracker and the

372

inner tracker, s(1/R

19

)

− s(1/R

28

), has been estimated using high energy cosmic ray proton

373

and helium events. Both proton and helium samples yield similar results. The difference

374

is found to be s(1/R

19

)

− s(1/R

28

) =

−0.019 ± 0.011 TV

−1

independent of rigidity [

43

].

375

The corresponding small shifts in the L1 and L9 positions were corrected and the error of

376

0.011 TV

−1

_{is added in quadrature to the total error of the rigidity scale.}

377

Overall, the tracker rigidity scale was measured with an accuracy of

_{±0.034 TV}

−1

_(i.e

378

quadratic sum of 0.033 TV

−1

and 0.011 TV

−1

). This corresponds to the determination of a

379

coherent displacement of layers L2–L8 by less than 0.2 microns.

380

To verify that after these corrections rigidity scale is time independent, we have studied

381

the time dependence of the measured fluxes. Figure

12

shows the ratio of the proton and

382 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Jour

nal

Pr

e-pr

oof

11

Rigidity[GV]

1 10 102 103

Ratios of Proton Fluxes

0.85 0.9 0.95 1 1.05 1.1 1.15 May 2011-Jan 2013 Feb 2013-Oct 2014 Nov 2014-Jul 2016 Aug 2016-May 2018 Rigidity[GV] 10 ₁₀2 ₁₀3

Ratios of Helium Fluxes

0.85 0.9 0.95 1 1.05 1.1 1.15 May 2011-Jan 2013 Feb 2013-Oct 2014 Nov 2014-Jul 2016 Aug 2016-May 2018 a) b)

FIG. 12. The ratios of the (a) proton and (b) helium fluxes measured for each of the four 21-month periods to the flux measured over 7 years. At low rigidities, the fluxes are time dependent due to solar modulation. At high rigidities, above 100 GV, the fluxes are consistent within measurement errors.

4. Charge-sign identification

387

Charge–sign identification is a critical property of the spectrometer. It allows AMS

388

to distinguish negatively charged cosmic rays from positively charged cosmic rays. It is

389

quantified by the amount of charge confusion. For example, in the analysis of positrons

390

it is important to distinguish positrons from charge-confusion electrons, i.e. those electrons

391

reconstructed with the positive charge sign due to the finite tracker resolution or interactions

392

in the detector materials. To this end, a charge confusion estimator Λ

e

CC

is defined using the

393

Boosted Decision Trees technique [

44

,

45

]. The estimator combines several measurements:

394

the ratio of the energy from the calorimeter to the momentum from the tracker, E/p, the

395

track χ

2

_{/d.o.f., momenta reconstructed with different combinations of tracker layers, the}

396

number of hits in the vicinity of the track, and the charge measurements in the TOF and in

397

the tracker. With this method, positrons, which have Λ

e

CC

∼ +1, are efficiently separated

398

from charge confusion electrons, which have Λ

e

CC

∼ −1.

399

After selection of a sample of positron and electron events with the TRD and ECAL,

400

the charge confusion in data is obtained using the template fitting technique with the Λ

e CC

401

distribution [

45

]. From the fit to the positive rigidity sample, we obtain the number of

402

positron N

e+

events, charge confusion electron background N

_ec.c.₋

events, and proton

back-403

ground events. From the fit to the negative rigidity sample, we obtain the number of

elec-404

tron N

_e−

events, charge confusion positron background N

_ec.c.+

events, and proton background

405

events. The charge confusion fraction in the electron data is calculated as N

c.c.

e−

/(N

_ec.c.−

+N

e−

)

406

and similarly for positrons.

407

The charge confusion in the Monte Carlo simulation is directly calculated as the fraction

408

of electrons being reconstructed with positive rigidity after the same selection cuts as used

409

in data. As an illustration, the comparison between the electron charge confusion fraction

410

in the data and in the Monte Carlo simulation is presented in Figure

13

. As seen, the charge

411

confusion is well reproduced by the Monte Carlo simulation. The charge confusion fraction

412

is <8% in the energy range up to 1 TeV.

413 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Jour

nal

Pr

e-pr

oof

Energy [GeV]

1

10

10

2

10

3

C

h

a

rg

e

c

o

n

fu

s

io

n

fr

a

c

ti

o

n

-6

10

-5

10

-4

10

-3

10

-2

10

-1

10

MC

Data

FIG. 13. The charge confusion fraction as a function of energy measured with electrons. The charge confusion fraction is <8% in the energy range up to 1 TeV. As seen, the charge confusion is well reproduced by the Monte Carlo simulation.

Similar methods are used to differentiate antiprotons and anti-deuterons from proton and

414

deuteron backgrounds.

415 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Jour

nal

Pr

e-pr

oof

5. Tracker charge measurement

416

The nine tracker layers independently measure the charge

_{|Z| of cosmic rays. The}

ion-417

ization energy losses deposited in the silicon are proportional to Z

2

_{and they are measured}

418

with strips on both the x and y sides of a silicon sensor.

419

The energy loss deposition is collected by several strips on both x and y sides. The

420

relative signal on either side is related to the transverse distance d between the nearest strip

421

and the location where the particle crosses the silicon, as shown in Figure

14

. The strip

422

amplitudes are first corrected for electronics gain factors and then for the dependence on

423

the particle position d and inclination angle θ. Details of this technique as well as the effects

424

of non-linear electronics response for high Z nuclei are described in Ref. [

46

].

425

COG

-2

-1

0

1

2

ADC

0

100

200

300

400

1

Signal Response of strips in Tracker

!

Angle of inclination

Position d

A

₄

A

₂

A

₀

A

₁

A

₃

A

_#

($=1-5) are the strips with five largest signal responses.

A

₁

is called “

seed strip

”

FIG. 14. Schematic of ionization energy deposition signals per strip, in ADC counts, at the incident particle position d (defined in Eq.1, see also Figure 5) and the inclination angle θ of the particle (red arrow). A0 is the strip with the highest amplitude (that is, the strip nearest to the crossing

point), the next highest adjacent strip A1, the other adjacent strip A2, etc...

426 427 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Jour

nal

Pr

e-pr

oof

The charge measurement is obtained by combining the charge determined by both x and

428

y strips. Figure

15

shows the charge resolution of tracker layers L2–L8 for nuclei from Z = 1

429

to Z = 28. This resolution enables us to perform the precision measurement of all nuclei

430

fluxes up to and beyond nickel (Z = 28).

431

Z

Charge

0

5

10

15

20

25

30

Charge Resolution

∆

Z

/Z

0

0.01

0.02

0.03

0.04

0.05

He

C

Ne

Si

Ca

Fe Ni

FIG. 15. The inner tracker (layers L2–L8 combined) charge resolution ∆Z/Z. The solid line is to guide the eye.

432 433 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Jour

nal

Pr

e-pr

oof

Figure

16

shows the charge measured simultaneously by the TOF (see section

I D

) and

434

the tracker.

435

ToF

Ch

arge

0

5

10

15

20

25

30

Tracke

r Charg

e

0

5

10

15

20

25

30

E

ve

n

ts

2

10

3

10

4

10

5

10

6

10

7

10

8

10

9

10

H Li He Be B C O Ne Na Mg Al Si P S Cl Ar K Ca Sc Ti V Cr Mn Fe Co Ni F N

FIG. 16. The charge measurement by the TOF (see Figure22) and the tracker.

436 437 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Jour

nal

Pr

e-pr

oof

C. Transition Radiation Detector (TRD)

438

The Transition Radiation Detector (TRD) [

47

] is located at the top of the AMS as shown

439

in Figure

2

. Its main purpose is to identify electrons and positrons by transition radiation

440

while rejecting protons at a level of 10

3

_{. The TRD also provides an independent tracking}

441

capability and determination of the charge value of the nuclei by measuring their rate of

442

energy loss (dE/dx).

443

The TRD consists of 5248 proportional tubes of 6 mm diameter with a maximum length

444

of 2 m arranged side-by-side in 16–tube modules. The 328 modules are mounted in 20

445

layers. As shown in Figure

17

a, the assembly of the TRD layers is supported in a tapered

446

octagonal carbon fiber structure with a very low coefficient of thermal expansion. Such a

447

structure ensures the minimum relative movement of the TRD elements with the variation

448

of the ambient temperature.

449

As shown in Figure

17

b, each layer is interleaved with a 20 mm thick fiber fleece

ra-450

diator, LRP375BK, with a density of 0.06 g/cm

3

_{. There are twelve layers of proportional}

451

tubes along the y axis located in the middle of the TRD and, along the x axis, four layers

452

located on top and four on the bottom. The tubes are filled with a 90:10 Xe:CO

2

gas

mix-453

ture. Xenon efficiently captures the transition radiation X-rays generated in the radiator

454

layers. CO

2

ensures stable operation of the proportional tubes. An anode wire in each straw

455

tube measures the signal from the resulting ionization due to the captures of the transition

456

radiation photons as well as the ionization signal of the traversing charged particle (see

457

Figure

17

c).

458 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Jour

nal

Pr

e-pr

oof