The SuperCam Instrument Suite on the NASA Mars 2020 Rover: Body Unit and Combined System Tests

Space Sci Re (2021) 217:4
h ps://doi.o g/10.1007/s11214-020-00777-5
The Supe Cam Ins umen Sui e on he NASA Ma s
2020 Ro e : Body Uni and Combined Sys em Tes s
Roge C.Wiens
1·Syl es e Mau ice2·Sco H. Robinson1·An hony E. Nelson1·
Philippe Cais3·Pe nelle Be na di4·Raymond T. Newell1·Sam Clegg1·
Shi K. Sha ma5·S e en S o ms1·Jona han Deming1·Da el Beckman1·
Ann M. Ollila1·Oli ie Gasnaul 2·Ryan B. Ande son6·Y es And é7·
S. Michael Angel8·Go ka A ana9·Elizabe h Auden1·Pie e Beck10 ·
Joseph Becke 1·Ka im Benze a a11 ·Syl ain Be na d11 ·Oli ie Beyssac11 ·
Louis Bo ges1·B uno Bousque 12 ·Ke y Boyd1·Michael Ca ey1·
Je ey Ca lson13 ·Kepa Cas o9·Jo den Celis1·Bap is e Chide2,14 ·Ke in Cla k13 ·
Edwa d Clou is15 ·Elizabe h C. Co doba13 ·Agnes Cousin2·Magdalena Dale1·
Lau en De lo es13 ·Do o hea Delapp1·Mu iel Deleuze7·Ma hew Di mye 1·
Ch is ophe Donny7·Gilles D oma 16 ·M. Geo ge Du an1·Miles Egan5·
Joan E in13 ·Cecile Fab e17 ·Amau y Fau11 ·Woodwa d Fische 18 ·Oli ie Fo ni2·
Thie y Fouche 4·Reuben F esquez1·Jens F yden ang19 ·Denine Gasway1·
I ai Gon ijo13 ·John G o zinge 18 ·Xa ie Jacob20 ·Sophie Jacquinod4·
Je ey R. Johnson21 ·Robe a A. Klisiewicz1·James Lake1·Nina Lanza1·
Ja ie Lase na22 ·Je emie Lasue2·S éphane Le Mouélic23 ·Ca ey Lege IV1·
Richa d Le eille24 ·E ic Lewin10 ·Guille mo Lopez-Reyes25 ·Ralph Lo enz21 ·
E ic Lo igny7·S e en P. Lo e1·B iana Luce o1·Juan Manuel Mada iaga9·
Mo en Madsen19 ·So en Madsen13 ·Nicolas Mangold23 ·Jose An onio Man ique25 ·
J.P. Ma inez1·Jesus Ma inez-F ias26 ·Ke in P. McCabe1·
Timo hy H. McConnochie27 ·Jus in M. McGlown1·Sco M. McLennan28 ·
Nou eddine Melikechi29 ·Pie e-Y es Meslin2·John M. Michel1·Da id Mimoun14 ·
Anupam Mis a5·Gilles Mon agnac16 ·F anck Mon messin30 ·Vale ie Mousse 7·
Naomi Mu doch14 ·Ho on Newsom31 ·Logan A. O 1·Zacha y R. Ousname 13 ·
Lau en Pa es2·Yann Pa o 2·Ra al Pawluczyk32 ·C. Glen Pe e son1·Paolo Pille i2·
Pa ick Pine 2·Gab iel Pon 7·F ancois Poule 33 ·Che yl P o os 32 ·
Benjamin Que ie 3·Hea he Quinn1·William Rapin11 ·Jean-Michel Reess4·
Amy H. Regan1·Ad iana L. Reyes-Newell1·Philip J. Romano13 ·Clemen Roye 33 ·
Fe nando Rull25 ·Benigno Sando al1·Joseph H. Sa ao1·Violaine Sau e 11 ·
Ma cel J. Schoppe s13 ·Susanne Sch öde 34 ·Daniel Sei z1·Te a Shephe d1·
Pablo Sob on35 ·B uno Dubois36 ·Vishnu S idha 13 ·Michael J. Toplis2·
Imanol To e-Fdez9·Ian A. T e el13 ·Ma k Unde wood13 ·And es Valdez1·
Jacob Valdez1·Dawn Venhaus1·Pe e Willis13
Recei ed: 16 May 2020 / Accep ed: 27 No embe 2020 / Published online: 21 Decembe 2020
© The Au ho (s) 2020
The Ma s 2020 Mission
Edi ed by Kenne h A. Fa ley, Kenne h H. Willi o d and Ka h yn M. S ack
BR.C. Wiens
[email p o ec ed]
1Los Alamos Na ional Labo a o y, Los Alamos, NM, USA
4Page 2 o 87 R.C. Wiens e al.
Abs ac The Supe Cam ins umen sui e p o ides he Ma s 2020 o e , Pe se e ance, wi h
a numbe o e sa ile emo e-sensing echniques ha can be used a long dis ance as well
as wi hin he obo ic-a m wo kspace. These include lase -induced b eakdown spec oscopy
(LIBS), emo e ime- esol ed Raman and luminescence spec oscopies, and isible and in-
a ed (VISIR; sepa a ely e e ed o as VIS and IR) e lec ance spec oscopy. A emo e
mic o-image (RMI) p o ides high- esolu ion colo con ex imaging, and a mic ophone can
be used as a s and-alone ool o en i onmen al s udies o o de e mine physical p ope ies
o ocks and soils om shock wa es o lase -p oduced plasmas. Supe Cam is buil in h ee
pa s: The mas uni (MU), consis ing o he lase , elescope, RMI, IR spec ome e , and
associa ed elec onics, is desc ibed in a companion pape . The on-boa d calib a ion a ge s
a e desc ibed in ano he companion pape . He e we desc ibe Supe Cam’s body uni (BU)
and es ing o he in eg a ed ins umen .
The BU, moun ed inside he o e body, ecei es ligh om he MU ia a 5.8 m op i-
cal ibe . The ligh is spli in o h ee wa eleng h bands by a demul iplexe , and is ou ed
ia ibe bundles o h ee op ical spec ome e s, wo o which (UV and iole ; 245–340 and
385–465 nm) a e c ossed Cze ny-Tu ne e lec ion spec ome e s, nea ly iden ical o hei
coun e pa s on ChemCam. The hi d is a high-e iciency ansmission spec ome e con ain-
ing an op ical in ensi ie capable o ga ing exposu es o 100 ns o longe , wi h a iable delay
imes ela i e o he lase pulse. This spec ome e co e s 535–853 nm (105–7070 cm−1Ra-
man shi ela i e o he 532 nm g een lase beam) wi h 12 cm−1 ull-wid h a hal -maximum
peak esolu ion in he Raman inge p in egion. The BU elec onics boa ds in e ace wi h
he o e and con ol he ins umen , e u ning da a o he o e . The mal sys ems main ain a
wa m empe a u e du ing c uise o Ma s o a oid con amina ion on he op ics, and cool he
de ec o s du ing ope a ions on Ma s.
2Ins i u de Reche che en As ophysique e Plane ologie (IRAP), Uni e si é de Toulouse, UPS,
CNRS, Toulouse, F ance
3Labo a oi e d’as ophysique de Bo deaux, Uni . Bo deaux, CNRS, Bo deaux, F ance
4Labo a oi e d’E udes Spa iales e d’Ins umen a ion en As ophysique, Obse a oi e de Pa is,
Meudon, F ance
5Uni e si y o Hawaii, Manoa, HI, USA
6U.S. Geological Su ey As ogeology Science Cen e , Flags a , AZ, USA
7Cen e Na ional d’E udes Spa iales, Toulouse, F ance
8Uni e si y o Sou h Ca olina, Columbia, SC, USA
9Uni e si y o Basque Coun y, UPV/EHU, Bilbao, Spain
10 Ins i u de Plané ologie e d’As ophysique de G enoble, Uni e si é G enoble Alpes, G enoble,
F ance
11 Ins i u de Miné alogie, Physique des Ma é iaux e Cosmochimie, CNRS, Museum Na ional
d’His oi e Na u elle, So bonne Uni e si é, Pa is, F ance
12 Cen e Lase s In enses e Applica ions, Uni e si y o Bo deaux, Bo deaux, F ance
13 Je P opulsion Labo a o y/Cal ech, Pasadena, CA, USA
14 Ins i u Supé ieu de l’Aé onau ique e de l’Espace (ISAE), Toulouse, F ance
15 Uni e si y o Winnipeg, Winnipeg, Canada
16 Uni Lyon, ENSL, Uni Lyon 1, CNRS, LGL-TPE, 69364 Lyon, F ance
17 GeoRessou ces, Uni e si é de Lo aine, Nancy, F ance
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 3 o 87 4
Resul s ob ained wi h he in eg a ed ins umen demons a e i s capabili ies o LIBS,
o which a lib a y o 332 s anda ds was de eloped. Examples o Raman and VISIR spec-
oscopy a e shown, demons a ing clea mine al iden i ica ion wi h bo h echniques. Lumi-
nescence spec a demons a e he u ili y o ha ing bo h spec al and empo al dimensions.
Finally, RMI and mic ophone es s on he o e demons a e he capabili ies o hese sub-
sys ems as well.
Keywo ds Pe se e ance o e ·LIBS ·Raman spec oscopy ·In a ed spec oscopy ·
Mic ophone on Ma s ·Supe Cam ·Jeze o c a e ·Ma s
1 In oduc ion
NASA’s Ma s o e s ha e used a ious emo e-sensing ins umen s o e he las wo and
a hal decades. The Sojou ne o e was ou i ed o emo e sensing wi h only image s
(Golombek e al. 1999). The Ma s Explo a ion Ro e s (MER) we e equipped wi h Minia-
u e The mal Emission Spec ome e s (Mini-TES; Ch is ensen e al. 2003) in addi ion o
s e eo mul ispec al image s (Pancam; Bell e al. 2003). Mini-TES p o ided he i s com-
posi ional emo e sensing om a Ma s o e beyond imaging il e wheels. Howe e , he
su ace o Ma s is co e ed by dus , which limi s he abili y o passi e emo e-sensing de-
ices o make obse a ions o he mine alogy o chemis y o he unde lying ocks. The
Chemis y and Came a (ChemCam) ins umen on he Cu iosi y o e o e comes his chal-
lenge by using a lase o abla e he dus and addi ionally enabling emo e dep h p o iles o
se e al hund ed µm o unde s and he su ace condi ions o he ocks (Mau ice e al. 2012;
18 Cali o nia Ins i u e o Technology, Pasadena, CA, USA
19 Uni e si y o Copenhagen, Copenhagen, Denma k
20 Ins i u de mécanique des luides de Toulouse (CNRS, INP, Uni . Toulouse), Toulouse, F ance
21 Johns Hopkins Uni e si y Applied Physics Labo a o y, Lau el, MD, USA
22 Uni e sidad de Malaga, Malaga, Spain
23 Labo a oi e de Plané ologie e Géodynamique, Uni e si é de Nan es, Uni e si é d’Ange s, CNRS
UMR 6112, Nan es, F ance
24 McGill Uni e si y, Mon eal, Canada
25 Uni e si y o Valladolid, UVA, Valladolid, Spain
26 Agencia Es a al Consejo Supe io de In es igaciones Cien i icas, Mad id, Spain
27 Uni e si y o Ma yland, College Pa k, MD, USA
28 S a e Uni e si y o New Yo k, S ony B ook, NY, USA
29 Uni e si y o Massachuse s, Lowell, MA, USA
30 Labo a oi e A mosphè es, Milieux, Obse a ions Spa iales, Pa is, F ance
31 Uni e si y o New Mexico, Albuque que, NM, USA
32 Fibe Tech Op ica, Ki chene , ON, Canada
33 Ins i u d’As ophysique Spa iale (IAS), O say, F ance
34 Deu sches Zen um ü Lu - und Raum ah (DLR), Ins i u e o Op ical Senso Sys ems, Be lin,
Ge many
4Page 4 o 87 R.C. Wiens e al.
Wiens e al. 2012). ChemCam uses lase -induced b eakdown spec oscopy (LIBS) o ob-
ain semi-quan i a i e elemen al abundances om as e s o small obse a ion poin s 350–
550 µm in diame e (Mau ice e al. 2012). While ChemCam is limi ed mos ly o chemical
composi ions a he han mine alogy, i s abili y o de ec and quan i y hyd ogen is impo an
o unde s anding he hyd a ion s a e o he soils and o iden i ying some hyd a ed mine -
als (Sch öde e al. 2015; Rapin e al. 2016,2018,2019; Thomas e al. 2020). ChemCam’s
chemis y is complemen ed by isible- ange (“VIS”) e lec ance spec oscopy o ∼850 nm
ha allowed Johnson e al. (2015,2017) o cons ain he mine alogy o i on-bea ing ma-
e ials (e.g., hema i e, oli ine, and e ic sul a es). Howe e , his passi e spec al ange is
no diagnos ic o phyllosilica es and ca bona es, which a e impo an o unde s anding he
his o y o Ma s’ hyd a ion, clima e, and habi abili y. Because o Cu iosi y’s ela i e lack
o emo e mine al-iden i ica ion capabili ies, he Ma s 2020 Science De ini ion Team man-
da ed ha he nex NASA o e should possess he abili y o obse e mine al composi ions
by emo e sensing (Mus a d e al. 2013).
The Supe Cam ins umen is a esponse o his equi emen o emo e mine alogy while
p ese ing he abili y o emo e dus p io o making obse a ions o nea by a ge s, and
p o iding he same o be e chemis y and high- esolu ion imaging as ChemCam. This new
ins umen esul ed om a happy collision o ideas om p e ious mission p oposals. I was
ecognized yea s ago ha lase -induced b eakdown spec oscopy (LIBS) and emo e Raman
spec oscopy bo h equi ed a lase , a elescope, and an op ical spec ome e (e.g., Wiens e al.
2005), and membe s o he Supe Cam eam sough o make ha a eali y o e he yea s. The
i s a emp was o he ExoMa s o e (Cou eges-Lacos e e al. 2007), bu he LIBS was
descoped ea ly in he ExoMa s de elopmen (Rull e al. 2017). A emo e Raman-LIBS com-
bina ion much close in design o Supe Cam was de eloped du ing he Venus Su ace and
A mosphe e Geochemical Explo e (SAGE) mission which only p oceeded h ough Phase-
A de elopmen (Clegg e al. 2014).
I was well unde s ood ha he Ma ian su ace would bene i e y s ongly om a com-
bina ion o emo e Raman and isible- o-in a ed (VISIR) e lec ance spec oscopy. These
wo mine alogy echniques a e highly complemen a y, as Raman signals occu as a esul
o a change in pola izabili y o a molecule wi h a omic ib a ions, whe eas in a ed spec-
oscopy is sensi i e o a change in he dipole momen s. Raman spec oscopy is sensi i e
o a symme ic s e ch bu in a ed spec oscopy is no . On he o he hand, Raman can be
insensi i e o asymme ic s e ches o which in a ed spec oscopy is sensi i e. The powe
o Raman spec oscopy o igina es om he ac ha he ac i i y o Raman modes depends
bo h on he o m o a ib a ional ha monic and he s e eochemis y o he molecule in ques-
ion. Gi en he unexpec edly high abundances o eldspa s (no gene ally ecognized by nea
in a ed spec oscopy) in igneous loa ocks, conglome a es, and sedimen a y ou c ops in
Aeolis Palus a Gale c a e (e.g., Sau e e al. 2015,2016), i is impo an o iden i y hese
mine als, a capabili y p o ided by Raman spec oscopy. Sho -wa e in a ed spec oscopy
is highly selec i e and diagnos ic o phyllosilica es and o he (pola /nonpola ) mine als, so
Supe Cam includes bo h spec oscopic echniques.
In addi ion, wo mo e echniques we e added. The ha dwa e used o pulsed-lase Raman
spec oscopy also enables ime- esol ed luminescence (TRL) spec oscopy. Luminescence
is a gene al e m ha encompasses he as , spin-allowed ansi ions e e ed o as luo es-
cence, and he slowe , spin- o bidden ansi ions called phospho escence. Supe Cam can
35 SETI Ins i u e, Moun ain View, CA, USA
36 Uni e si é de Toulouse; UPS-OMP, Toulouse, F ance
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 5 o 87 4
Table 1 Lis o echniques employed by Supe Cam o emo e sensing. See ex o explana ions
Technique Pu pose Dis anceaFoo p in
LIBS Quan i a i e elemen al abundances To 7 m 0.25–0.45 mm
Raman Iden i ica ion o Raman-b igh mine als, o ganics To ∼7 m 0.74 m adb
TRL Iden i ica ion o o ganics, mine als, REEscTo ∼7 m 0.74 m ad
VISIRdIden i ica ion o mine als; a mosphe ic s udies To km VIS: 0.74 m ad
IR: 1.2 m ad
RMI Rock ex u es, con ex s To km 18.8 m ad
Mic ophone Physical p ope ies o ocks, a mosphe ic s udies To ∼4m N/A
aMinimum dis ance, all op ical cases: 1.05 m om he MU, posi ioned ∼2 m abo e la g ound
bm ad =milli adians
cREEs = a e-ea h elemen s
dVIS spec al ange is 0.40–0.85 µm; IR ange is 1.3–2.6 µm; o he anges a e gi en in he ex
de ec , dis inguish, and cha ac e ize bo h luo escence and phospho escence using TRL. Fi-
nally, acous ic spec al sensing was added o emo ely de e mine he physical p ope ies o
ocks and o assess a mosphe ic p ope ies (e.g., Mu doch e al. 2019; Chide e al. 2019).
Desc ip ion o he Supe Cam ins umen , shown in Fig. 1, is di ided among se e al pa-
pe s. The science goals, mos o he echnical equi emen s, and he desc ip ion o he Mas
Uni (MU) a e in a companion pape (Mau ice e al. 2020). He e we p o ide a desc ip ion
o he Body Uni (BU) and o he in eg a ed es ing p io o launch. An o e iew o he
o e calib a ion a ge s is in ano he companion pape (Man ique e al. 2020). Mo e de ail
on subsys ems and calib a ions a e o will be p o ided in o he pape s published sepa a ely
(Roye e al. 2020, and u u e pape s).
2 Ins umen O e iew
We begin wi h a b ie o e iew o he en i e ins umen (Fig. 1) be o e ocusing on he Body
Uni . Table 1p o ides a sho lis o he echniques employed by Supe Cam.
LIBS p o ides a omic emission spec a o ma e ial abla ed om small spo s on ock o
soil a ge s, leading o quan i a i e elemen al composi ions o majo , mino , and ace ele-
men s. ChemCam is able o de ec and quan i y ∼25 elemen s (e.g., Mau ice e al. 2016),
and Supe Cam is expec ed o achie e he same as, o sligh ly be e pe o mance han Chem-
Cam, wi h he same dis ance capabili y (Table 1). The analy ical oo p in is necessa ily
small, as he op ical powe densi y mus be maximized o c ea e a plasma on he a ge . The
lase used o achie e he plasmas p o ides up o 14 mJ and >10 MW/mm2o 1064 nm
pho ons pe pulse (Mau ice e al. 2020). As desc ibed la e , he use o he ansmission
spec ome e in he g een o ed spec al ange allows ime ga ing and in ensi ica ion o he
signal, which may be used o special LIBS s udies, such as o ampli y an o he wise weak
emission line.
Supe Cam employs he i s use o g een-lase Raman spec oscopy in space, and sha es
he dis inc ion o he i s plane a y Raman spec ome e wi h SHERLOC (Bha ia e al.
2020, his jou nal). Supe Cam achie es Raman spec oscopy a emo e dis ances o ∼7m
by using a pulsed lase — he same one as o LIBS, equency doubled o 532 nm—and an
in ensi ied, ga ed de ec o coupled o a ansmission spec ome e . The g een Raman lase

4Page 6 o 87 R.C. Wiens e al.
Fig. 1 The Supe Cam ins umen , consis ing o he Mas Uni (MU), shown in (a), he Body Uni (BU),
shown in (b), and he Supe Cam Calib a ion Ta ge (SCCT), shown in (c). As pa o he MU (a), he lase can
be seen o he le , p o uding om behind he elec onics box. The elescope is a he a end, a he cen e
o which he pe iscope mi o o he g een lase beam is moun ed. The co esponding pe iscope mi o
can be seen jus pas he elec onics box, acing away om he came a. A he nea end o he elec onics
box, a hea ing pad is jus o he le o he connec o s. The Mas Uni is moun ed on insula ing ee , and is
shown he e es ing on a handling ix u e. The BU (b) shows a ansmission spec ome e es ing behind wo
iden ical e lec ion spec ome e s, all moun ed on op o he elec onics box. Th ee op ical ibe bundles can
be seen wi h hei p o ec i e shields nea he uppe le ; hese ans e ligh o he spec ome e s om he
demul iplexe . The only pa o he demul iplexe ha is isible is he ibe connec o , p o uding a he le
cen e . This is whe e he ligh om he MU en e s he BU. One o h ee se s o he moelec ic coole s is
seen in he lowe cen e , iden i ied by wo isible hea pipes ha un unde he spec ome e s o cool hei
de ec o s. On he SCCT (c), wen y-nine ci cula a ge s and se e al o he calib a ion a ge s a e moun ed.
The i anium pla e a he uppe igh is used o wa eleng h calib a ion ia LIBS spec a. Imaging a ge s and
a Ma s me eo i e sample line he le side. Dimensions o all h ee Supe Cam uni s a e gi en in Table 2
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 7 o 87 4
beam is collima ed a he han ocused, and i o e laps wi h he 0.74 m ad ield o iew
(FOV) o he spec ome e , de ined by he elescope and op ical ibe ha ans e s he ligh
om he MU o he BU (Mau ice e al. 2020). This oo p in hus anges be ween 1.5 mm
diame e when looking s aigh down a he g ound om he elescope’s heigh o 2 m, o
∼5 mm diame e when obse ing a a dis ance o 7 m. Raman spec a a e always dim,
and collec ing hese spec a emo ely esul s in a limi ed numbe o pho ons (Sec . 7.3.3).
Supe Cam’s design is a comp omise o enable all o he echniques, which limi ed he abili y
o op imize he pho on h oughpu , wi h he esul ha Supe Cam obse es mine als and
o ganic ma e ials ha a e s ong Raman emi e s. Among he bes a e ca bona es, sul a es,
and phospha es, bu Supe Cam also expec s o iden i y qua z and plagioclase eldspa , he
la e o which a e no , o a e only poo ly, de ec ed by nea -in a ed e lec ance spec oscopy.
The pulsed lase and ime-ga ed spec ome e also p o ide he capabili y o TRL.
P omp luo escence is emi ed by o ganic ma e ials, being emi ed and decaying wi hin
nanoseconds o s imula ion (Lakowicz 2006). This p omp o ganic luo escence is he bane
o Raman spec oscopy on Ea h, as e en ime- esol ed Raman spec oscopy wi h nanosec-
ond lase pulses and he as es -ga ed de ec o s gene ally do no disc imina e agains i .
Howe e , on Ma s, i may be a powe ul ool o disco e ing concen a ions o o ganic
ma e ials. Supe Cam’s in ensi ie ga e can be delayed up o milliseconds wi h a empo al
esolu ion o 10 ns, p o iding a second dimension o cha ac e izing mine al luo escence.
Using he ime dimension, he p esence o ce ain a e-ea h elemen s (REEs) and ansi ion
me als can be iden i ied (Ga e al. 2015; Ollila e al. 2018).
Supe Cam’s VISIR e lec ance spec oscopy is he i s o co e he spec al ange o 0.4
o 2.6 µm om he su ace o Ma s. I does so u ilizing se e al spec ome e s. The main
phyllosilica e iden i ica ion egion, in he nea in a ed, has been success ully used by he
o bi ing spec ome e s, speci ically he Compac Reconnaissance Imaging Spec ome e o
Ma s (CRISM; Mu chie e al. 2007) and Obse a oi e pou la Mine alogie, l’Eau, les Glaces
e l’Ac i i é (OMEGA; Bib ing e al. 2004). On Supe Cam his spec al ange is p o ided by
a wa eleng h-scanning spec ome e (1.3–2.6 µm) in he MU, while obse a ions in he VIS
ange (0.4–0.85 µm) ake ad an age o he BU spec ome e s ha a e also used o LIBS,
Raman, and TRL spec oscopies. The e is a gap in he spec al co e age be ween 0.85 and
1.3 µm (Fig. 41 o Mau ice e al. 2020). The VIS ange (co e ed by he iole and ans-
mission spec ome e s) has a somewha smalle a ge oo p in han he IR spec ome e
(Table 1). On Ma s 2020, Supe Cam’s poin spec al obse a ions a e well complemen ed
by Mas cam-Z, which can image a disc e e wa eleng hs om 0.44 o 1.0 µm (Bell e al.
2020, his jou nal).
The emo e mic o-image (RMI) and mic ophone comple e he lis o echniques a ail-
able wi h Supe Cam (Table 1). High- esolu ion con ex images o he analysis a eas a e c i i-
cal o hei in e p e a ion, especially gi en he small oo p in o he analysis echniques. The
RMI p o ides his con ex ia Baye - il e colo images wi h a esolu ion ≤80 µ ad (de ined
as a line pai wi h mo e han 20% con as ), allowing 160 µm g ains o be esol ed a 2 m
dis ance om he ins umen , e.g., di ec ly in on o he o e . (The RMI esolu ion is no
pixel limi ed, as i s ins an aneous ield o iew, o IFOV, is 9.2 µ ad, no conside ing he
Baye il e .) The mic ophone p o ides acous ic signals om he LIBS shock wa e. S udies
(e.g., Chide e al. 2019) ha e shown ha a combina ion o ock ha dness and densi y can be
ob ained by he a e o dec ease o acous ic ene gy wi h inc easing numbe o lase pulses a
he same loca ion. The mic ophone will also be use ul o a mosphe ic s udies (Chide e al.
2020), pe haps including phenomena o which we a e no ye awa e. As subsys ems o he
MU, he RMI and mic ophone a e desc ibed in Mau ice e al. (2020).
Figu e 2shows a schema ic diag am o he Supe Cam ins umen . Supe Cam is di ided
in o wo majo uni s, he BU and he MU. The MU esides a he op o he o e ’s mas and
4Page 8 o 87 R.C. Wiens e al.
Fig. 2 Schema ic diag am showing he majo uni s and subcomponen s o he Supe Cam ins umen sui e.
The Mas Uni (MU) consis s o he main lase which p o ides wo wa eleng hs using wo Galilean beam
expande s, he elescope, a con inuous-wa e lase (CWL) o ocusing, and a mic ophone. The op ical box
(OBOX) also includes he in a ed (IR) spec ome e and he Remo e Mic o-Image (RMI), he de ec o o
which is a complemen a y me al oxide semiconduc o (CMOS). An elec onics box (EBOX) con ols and
powe s he a ious subsys ems in he MU. Acquisi ion o he a ge is p o ided by he o e mas azimu hal
(AZ) and ele a ion (EL) mo ions. Elec ical cables and an op ical ibe connec he Body Uni (BU) o he
MU. The ibe ca ies ligh in he 245–853 nm ange o he demul iplexe (labeled Demux) in he BU, which
dis ibu es he ligh o h ee spec ome e s co e ing ul a iole (UV), iole (VIO), g een, o ange, and ed
spec al anges. The la e a e cha ac e ized by a ansmission spec ome e , which uses an in ensi ie d i en
by a high- ol age powe supply (HVPS). All h ee BU spec ome e s collec ligh wi h cha ge-coupled de-
ices (CCDs) cooled by he moelec ic coole s (TECs). The elec onics box (EBOX) in he BU ope a es he
ins umen , p o ides powe o he BU spec ome e s and he MU, and communica es wi h he o e h ough
he con ol and da a handling (C&DH) boa d. A se o calib a ion a ge s is moun ed on he back o he o e
o acili a e calib a ion while on Ma s
con ains he lase , elescope, Remo e Mic o-Image , In a ed Spec ome e , Mic ophone,
and associa ed elec onics. The MU was designed, buil , and es ed in F ance unde he sup-
po and di ec ion o he Cen e Na ional d’E udes Spa iales (CNES), wi h in eg a ion a he
Ins i u de Reche che en As ophysique e Plane ologie (IRAP). The BU con ains an op ical
demul iplexe , op ical spec ome e s o LIBS, Raman, and passi e VIS spec oscopy, and
associa ed elec onics o con ol he BU and MU and in e ace wi h he o e . The BU was
designed, buil , and es ed a Los Alamos Na ional Labo a o y, in he US. A hi d pa o
he ins umen , onboa d calib a ion a ge s we e p o ided by an in e na ional wo king g oup
wi hin he Supe Cam eam and consis mos ly o sin e ed pelle s (e.g., Mon agnac e al.
2018). Cha ac e iza ion o he a ge s was pe o med by a eam o Eu opean scien is s; he
Uni e sidad de Valladolid was esponsible o in eg a ion and en i onmen al es ing o he
a ge assembly (Man ique e al. 2020).
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 9 o 87 4
Fig. 3 Loca ions o he Supe Cam uni s on he o e . The igh side shows he o e body in e ed, wi h he
Body Uni ci cled. I is nex o he RSM side o he o e o minimize he leng h o he ibe ha ans e s
he op ical signal om he mas uni . The o e ’s ins umen and elec onics bay is 1181 ×1106 mm (leng h,
le - igh , x wid h)
Table 2 Supe Cam physical
p ope ies
aIncludes ope a ion o he
he moelec ic coole s. BU powe
when idling is 12 W
Mass
(kg)
Dimensions
l×w×h(mm)
Max powe
(W)
Mas Uni 6.11 383 ×201 ×163 27
Body Uni 4.44 221 ×157 ×205 43a
Cal. Ta ge 0.25 110 ×100 ×17 0
TOTAL 10.80 70
Figu e 1shows he comple ed Supe Cam ins umen , which was designed o con o m
o he space p o ided by he o e p ojec (Fig. 3). The a ailable olume was qui e sim-
ila o ha o ChemCam (Mau ice e al., Wiens e al. 2012), and so om bo h he i age
and mechanical/ he mal- equi emen s andpoin s, Supe Cam looks e y simila o i s p e-
decesso . Supe Cam’s elescope di e s om ChemCam’s, as he newe ins umen ea-
u es a pe iscope mi o ha di ec s Raman lase ligh o he in ended sample (Fig. 1).
The BU’s op ical demul iplexe and wo e lec ion spec ome e s—ul a iole and iole
(UV, VIO)—a e nea ly iden ical o ChemCam’s, bu i s hi d e lec ion spec ome e was
eplaced by a ime-ga ed, in ensi ied, high- h oughpu ansmission spec ome e o enable
Raman spec oscopy. The calib a ion a ge assembly (Fig. 1; Man ique e al. 2020) is as ly
expanded and imp o ed o e ChemCam’s (Fab e e al. 2011; Vaniman e al. 2012;Wiens
e al. 2012).
Table 2p esen s some o Supe Cam’s physical p ope ies. O e all, Supe Cam has almos
exac ly he same mass as ChemCam. The MU and calib a ion a ge s a e sligh ly hea ie ,
bu he BU is ligh e . The di e ence is la gely due o ChemCam’s he moelec ic coole
(TEC) assembly, which— hanks o a JPL eam–was added o ha ins umen wi hin he i-
nal 18 mon hs o i s de elopmen , when mass was no a conce n (Wiens e al. 2012). The
TECs on Supe Cam we e planned om he beginning, and so he design o he associa ed
cooling sys em is ligh e . Ano he di e ence a ec ing he mass o Supe Cam is he eplace-
men o be yllium by i anium o he BU spec ome e s. The mal expansion o ChemCam’s
spec ome e s equi es ca e ul compensa ion o changes in he wa eleng h calib a ion wi h
espec o he ins an aneous empe a u e o he spec ome e s on Ma s, which change by
20 ◦C diu nally. Ti anium has a lowe coe icien o he mal expansion (CTE) han Be, so
main aining wa eleng h calib a ion will be easie . To i Ti spec ome e s in o he igh mass
4Page 16 o 87 R.C. Wiens e al.
Fig. 9 Typical inspec ion images o demul iplexe end (a) and spec ome e end (b) o ibe bundles, showing
he one eeding he ansmission spec ome e . The co e o each ibe is 50 µm diame e . Fibe s a e backli
o he inspec ions. The ela i e in ensi y o each ibe may be a unc ion o he posi ion o he lamp, and so
does no indica e ela i e h oughpu in his case. Inse shows a magni ied image o he linea a ay o ibe s
a he spec ome e end be o e (c)anda e (d) bonding a 28.7 µm sli . Inse (e) shows he mapping o ibe s
in he a ay, wi h cen al ibe s in he ci cle mapped o cen al ibe s in he line
Table 5 Modeled heo e ical maximum ansmission h ough he ibe bundle and sli assemblies, and mea-
su ed ansmissions o a ious wa eleng hs
Spec ome e Sli wid h
(µm)
% T ansmission @ Wa eleng h (nm) Theo e ical
maximum
300 405 565 660 780
UV 20.7 234242443651
VIO 21.0 224242423552
T ansmission28.7 325153534669
a LANL o unc ionali y and h oughpu , and o ensu e p ope e mina ions, ci cula i y and
linea i y o he a angemen s, and mapping om one end o he o he .
A e cha ac e iza ion, sli s p ocu ed om Na ional Ape u e we e ins alled on he bes
bundles. The sli s a e 1.16 mm long, p oduced in 13 µm hick pla es ha we e blackened.
The sli s we e inspec ed and measu ed a LANL p io o ins alla ion, and pe o mance was
checked a e ins alla ion. The sli wid hs used in each spec ome e a e gi en in Table 5,
along wi h he measu ed op ical e iciency o he comple ed ibe assemblies wi h sli s a -
ached. No e ha he epo ed ansmission alues include losses bo h om he sli (gi en
unde “Theo e ical maximum”) and he losses om he packing ac ion o he nine een
50 µm ibe s.
As wi h any spec ome e , selec ing a sli wid h in ol es a ade be ween spec al e-
sol ing powe and op ical e iciency. To ensu e he op imal choice, we buil and es ed ibe
bundle assemblies wi h se e al sli wid hs and on each we measu ed he op ical h oughpu
and esol ing powe , de ined as he a e age ull wid h a hal maximum (FWHM) o se e al
a omic emission lines om a neon lamp. The esul s, seen o he ansmission spec ome e
in Fig. 10, led us o selec he 28.7 µm wide sli assembly o ligh . The neon emission
lines ha we e used do no co e he mos challenging spec al egion, om 535–555 nm,
whe e spec al esolu ion is wo se. Wi h his sli , he ansmission spec ome e jus mee s i s

The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 17 o 87 4
Fig. 10 Compa ison o op ical h oughpu e iciency (a) and spec al esolu ion (b) as unc ions o en ance
sli wid h o ansmission spec ome e . The heo e ical maximum op ical ansmission is shown as a do ed
line in (a). As expec ed, a wide sli allows mo e ligh bu also inc eases he FWHM o spec al ea u es. The
28.7 mic on sli was selec ed o ligh . Al hough no shown in he igu e ( he lowes -wa eleng h Ne line used
was a 576 nm), his sli jus mee s he 12 cm−1 esolu ion equi emen a he sho -wa enumbe end o he
spec um
FWHM esolu ion equi emen o 12 cm−1e e ywhe e. Simila es ing led o he selec ion
o he 20.7 and 21.0 µm sli s o he UV and VIO spec ome e s, espec i ely (Table 5).
3.1.3 Re lec ion Spec ome e s
As shown in Fig. 5, wo o he h ee spec ome e s a e c ossed Cze ny-Tu ne e lec ion
spec ome e s. Table 3gi es speci ica ions in e ms o wa eleng h ange and esolu ion o
he spec ome e s. The /4 op ical design is essen ially iden ical o ha o ChemCam (Wiens
e al. 2012). The wo spec ome e s a e iden ical o each o he wi h he excep ion o he
mi o s, g a ings, and g a ing angles. Figu e 11 shows a ende ing o he in e io o he
e lec ion spec ome e s. The g a ings, manu ac u ed by Richa dson, a e 2400 lines pe mm
(lpmm) on Scho N-SF8 glass subs a es wi h aluminum coa ings. The UV g a ing is 240 nm
blaze wa eleng h, while he VIO g a ing is 300 nm. The mi o s ha e sphe ical conca e
su aces wi h 100 mm adii o cu a u e, wi h dielec ic su aces o maximum e lec ance
in he espec i e spec al anges. Mi o s we e manu ac u ed by OPCO, and ha e e lec i i y
>94% be ween 240–340 nm o he UV spec ome e , and >98% be ween 380–500 nm
o he VIO spec ome e .
The di ac ed images o he spec ome e sli s a e p ojec ed on o cha ge-coupled de ices
(CCDs). All h ee spec ome e s ha e iden ical CCD assemblies wi h he excep ion o he
CCD su ace coa ings, which we e selec ed o op imize ligh collec ion a he espec i e
anges (Table 3). As wi h ChemCam, hey a e E2V 42-10 CCDs, 13.5 µm squa e pixels in
an a ay o 2048 ×515 wi h 50 addi ional blind se ial- egis e pixels on each side. The ac-
i e a ea is 27.6×6.9 mm. Ex ensi e es ing was ca ied ou on he CCDs and on eplica es
om he same ac o y lo . In addi ion o es ing and cha ac e iza ion a he manu ac u e
(Table 3), ead noise, da k cu en , image and se ial pixel well capaci ies, and sensi i i y o
inpu ol ages we e all es ed a LANL, and he bes among se e al CCDs o each ype we e
assigned o he ligh ins umen . A blemish was ound in he ligh VIO CCD a e in eg a-
ion on o he CCD boa d, in which he sensi i i ies o pixels in ows be ween 240 and 250
a columns 1323–1326 a e signi ican ly educed. This was no no ed by he manu ac u e .
The LIBS spec al image om andesi e s anda d JA-3 is shown in Fig. 12.In he e-
lec ance spec ome e s, he ligh is sp ead ac oss up o 180 cen al ows. (The VIO blemish
is no seen.) ChemCam in eg a es he signal om 200 e ical ows o maximize he collec-
ion (Wiens e al. 2012). Fo Supe Cam we plan o use se e al di e en e ical in eg a ion-
ow se ings on he e lec ion spec ome e s (Table 3). Ope a ion o he Supe Cam CCDs
and hei iming wi h espec o he lase a e discussed in Sec s. 3.3.1 and 4.2.
4Page 18 o 87 R.C. Wiens e al.
Fig. 11 Rende ing o a Supe Cam e lec ion spec ome e . Ligh en e s om he ibe bundle and sli assem-
bly a he uppe igh . I is collima ed by he ci cula mi o a lowe le . The g a ing a uppe igh p o ides
spec al dispe sion, he i s o de o which is ocused by he ec angula mi o a he uppe le on o he de-
ec o assembly a he lowe igh . Ba les can be seen along he uppe and lowe sides and nex o he g a ing,
used o abso b highe o de e lec ions om he g a ing. The cylinde p o uding a he le is a he mal swi ch
3.1.4 T ansmission Spec ome e
The ansmission spec ome e is a signi ican inno a ion ela i e o ChemCam; i enables
he emo e Raman and TRL spec oscopy echniques, while p o iding enhanced capabili ies
o LIBS and also collec ing passi e VIS e lec ance spec a. The p ima y unc ions a e
o p o ide high ansmission and in ensi ica ion a ela i ely high esolu ion o he weak
Raman signal, and o p o ide ime ga ing o 100 ns o minimize noise om backg ound
ambien ligh and luo escence. The ansmission spec ome e design is based e y loosely
on ea lie comme cial Raman ansmission spec ome e s such as he Kaise Holospec, bu
wi h a long sepa a e de elopmen pe iod, i s a U. Hawaii wi h modi ied comme cial pa s
(e.g., Sha ma 2007) and hen a LANL wi h cus om pa s, s a ing a ound 2009.
Figu e 5shows he basic a chi ec u e o he spec ome e , while Fig. 13 shows an op i-
cal ay ace. Th ee di ac ed bands a e used o simul aneously co e he wa eleng h ange
equi ed o LIBS (535–853 nm) while achie ing he esolu ion needed o plane a y g een-
lase Raman spec oscopy (12 cm−1) o e i s ange (∼105 o 4000 cm−1; 535–676 nm).
Because he LIBS spec al esolu ion and ligh collec ion equi emen s a e no as s ingen
in his spec al ange, he ansmission spec ome e was op imized o e he na owe Ra-
man ange while s ill co e ing he b oade LIBS spec al ange. The ea u e ha limi s he
esolu ion is he in ensi ie a he han he CCD.
AsshowninFig.5and Fig. 13, ligh passes h ough he sli and is collima ed be o e en-
coun e ing a dich oic beam spli e assembly, which sepa a es a ed op ical band. Two pa al-
lel beams a el o sepa a e ansmission g a ings. A compound g a ing (one di ec ly behind
he o he , wi h an angula sepa a ion) sepa a ely di ac s g een and o ange ligh , while an
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 19 o 87 4
Fig. 12 LIBS ligh om andesi e s anda d JA-3 (sho in e es ial a mosphe e) as seen by Supe Cam’s
de ec o s. Shown a e ull images o he CCDs o UV, iole (VIO), and ansmission (TS) spec ome e s.
These images a e shaded as he na u al loga i hm o in ensi y. Rows and columns a e indica ed, as a e he
wa eleng hs. Two lines ex ending o he op o he VIO image a e Hg emission lines om oom ligh s. In
he ansmission spec ome e , he edge o he in ensi ie can be pe cei ed by he edges o he con inuum
emission
adjacen g a ing di ac s he ed band. All h ee bands a e ocused on he in ensi ie . As a
inal s ep, in ensi ied ligh is e- ocused on he CCD in he elay sec ion. A ende ing o he
spec ome e is shown in Fig. 14.
The dich oic beam spli e (Figs. 14,15a) ecei es ligh om he collima ing lens and
e lec s i down owa d he g a ings. Below he e lec ing mi o , a cus om dich oic mi o
om Op oSigma, posi ioned a a 45◦angle, passes >93% o ligh om 720–850 nm di ec ly
down o he ed-band g a ing while e lec ing o he side >95% o ligh om 530–700 nm.
This ligh o ms he g een and o ange bands. I is e lec ed by ano he 45◦mi o s aigh
down owa d he compound g een-o ange g a ing.
4Page 20 o 87 R.C. Wiens e al.
Fig. 13 Ray aces in a model o he ansmission spec ome e , showing collima ion, di ac ion, and o ma-
ion o sli images on he image in ensi ie en ance window and elay lens ca ying in ensi ied image o CCD
senso . Rays a e colo ed by wa eleng h; 534 nm o 853 nm o he in ensi ie , and 545 nm om he in ensi ie
phospho o he CCD
Fig. 14 Cu away ende ing o he ansmission spec ome e showing he in e nal layou
The wo pa allel beams a e di ec ed on o h ee ansmissi e di ac ion g a ings (Figs. 14,
15b); he long wa eleng hs >715 nm go o one g a ing, and he sho e wa eleng hs o 530–
715 nm go o a duplex compound g a ing. The di ac i e elemen in he 715–855 nm beam
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 21 o 87 4
Fig. 15 Subsec ions o he
ansmission spec ome e . The
spec ome e sec ion (a)isshown
wi h he ibe connec o acing o
he ea . The dich oic beam
spli e is in ocus. The g a ing
assembly (b) is shown wi h he
inpu side isible, and elay lens
is seen in i s housing (c), looking
om he CCD side. O he
componen s on he able a e
isible h ough he elay lens
is a single 1800 lpmm g a ing. The compound g a ing con ains wo di ac i e elemen s: a
2480 lpmm g a ing op imized o he 530–620 nm ange (nominal cen e : 570 nm), and a
2110 lpmm g a ing op imized o he 610–700 nm ange (nominal cen e a 670 nm). The
ulings in he wo g a ings a e o a ed 1.8 deg ees om each o he , so ha he di ac ed
spec a a e spa ially sepa a ed on he in ensi ied CCD (ICCD) inpu window. The single
1800 lpmm g a ing is o a ed by an addi ional 1.8 deg ees so ha i s di ac ed ou pu is
spa ially sepa a ed om he o he wo. Volume-phase (VPH) ansmission g a ings om
Wasa ch Pho onics Inc. we e p e iously ligh quali ied as pa o he ExoMa s/RLS ins u-
men (Rull e al. 2017). The ou pu o he h ee VPH di ac ion g a ings is a se o h ee
spec al bands s acked in he spa ial di ec ion on he ace o he ICCD (see Fig. 16). An
ach oma ic lens assembly, simila o a scan lens in design, ocuses each spec al componen

4Page 22 o 87 R.C. Wiens e al.
Fig. 16 Map o spec ome e sli
images a a ious wa eleng hs (in
nm) wi h he ield o iew o he
in ensi ie (ci cle) d awn o
scale. The g een band (middle,
530–618 nm) s a s pa -way in
om he CCD’s edge in o de o
op imize he esolu ion. The
o ange band (598–720 nm) is
p ojec ed o he le , and he ed
band (707–853 nm) o he igh .
The CCD is ead-ou om he
igh side in his o ien a ion.
Compa e wi h he bo om panel
o Fig. 12
on o he pho oca hode o he ICCD sub assembly. The images o he sli on he ICCD ace
a e magni ied 1.22 imes by he scan lens assembly.
The ICCD subassembly consis s o h ee main elemen s: in ensi ie ube, CCD senso ,
and elay lens (Fig. 15) o ca y ligh om he in ensi ie o he CCD. The in ensi ie ube
p o ides wo essen ial unc ions o Raman and TRL spec oscopies: I ampli ies he weak
op ical signal e u ned om he sample and allows e y as ( equi emen o ≤100 ns)
ga ing. This ga ing is essen ial o isola e he b ie Raman signal om he delayed and much
longe -li ed luminescence gene a ed by he sample (Sec . 7.4) and o s udy ime- esol ed
luminescence.
P ima y ICCD design conside a ions a e op ical gain, esolu ion, spec al esponsi i y,
ga ing abili y, and ligh quali ica ion. Wo king closely wi h ITT Exelis in Roanoke VA
(now pa o Elbi Sys ems), we iden i ied an in ensi ie which mee s all equi emen s. The
in ensi ie ea u es a con inuously a iable gain o e 45 dB pho on/pho on, esolu ion ex-
ceeding 48 lpmm, and a pho oca hode wi h quan um e iciency (QE) o e 30% om 530 nm
o ∼900 nm. Gain is adjus ed by selec ing he ol age ac oss he ube’s mic ochannel pla e,
and signals a e ime-ga ed by ga ing he ol age.
A high-pe o mance elay lens assembly couples ligh om he in ensi ie phospho ou -
pu o he CCD. A cus om double-Gauss lens design is used o mee equi emen s o high
coupling e iciency ( /2.7), >30◦FOV and compac size. Lens design akes ad an age o
he ac ha ligh exi ing he in ensi ie is nea ly monoch oma ic a 545 nm. The in ensi ie ’s
P43 phospho does gene a e mino ou pu s a o he wa eleng hs, bu hese a e supp essed
by coa ings on he elay lens su aces. The esol ing powe o his elay lens assembly su -
passes ha o he in ensi ie ube. The pa axial magni ica ion o he elay lenses is −1.26.
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 23 o 87 4
Fig. 17 Side-by-side ende ing o ChemCam and Supe Cam Body Uni s. Supe Cam’s BU is 345 g ligh e
and occupies a sligh ly smalle olume
A 20 µm ea u e on he in ensi ie maps o 25.2 µm on he CCD, which co esponds o 1.87
pixels.
3.2 Mechanical and The mal Desc ip ion
3.2.1 Mechanical
The Supe Cam BU mechanical design uses ChemCam as i s s a ing poin (Fig. 17), bu in-
co po a es many e inemen s and design changes d i en by he speci ic Supe Cam equi e-
men s. The BU consis s o i e majo sub-assemblies (Fig. 4). These a e he elec onics box,
demul iplexe , e lec ion spec ome e s, ansmission spec ome e , and connec o in e ace
b acke . The design is modula , allowing o each sub-assembly o be assembled and ini ially
es ed and aligned be o e inal in eg a ion. This modula i y was also in ended o minimize
he impac upon o he modules in case disassembly was needed o ouble shoo ing.
The elec onics box (EBOX) (Fig. 18) is an aluminum s uc u e ha con ains he spec-
ome e elec onics (SE) p in ed ci cui boa d (PCB), command and da a handling (C&DH)
PCB, and he low- ol age powe supply (LVPS). I also unc ions as a s able pla o m o he
o he sub-assemblies moun ed on i s op panel. Addi ionally, he EBOX se es as he he -
mal in e ace o he RAMP; he in e ace is desc ibed in Sec . 3.2.2. The EBOX design uses
a backplane/daugh e -ca d a chi ec u e. Each daugh e ca d can be easily and independen ly
ins alled o emo ed wi hou signi ican impac o he o e all assembly, and wi h no im-
pac o he op o-mechanical assemblies. This is an imp o emen o e he ChemCam design.
All elec ical connec ions be ween boa ds a e h ough he backplane using nano-D connec-
o s. Each daugh e boa d is moun ed o an aluminum ame, which ac s as a he mal sink
and p o ides s i ness o he boa d in ib a ion. Each module slides in o he housing and
is clamped in place wi h wedgelocks. The wedgelock join s p o ide a easonable he mal
in e ace be ween he housing and he module ame. The LVPS ame also makes up he
4Page 24 o 87 R.C. Wiens e al.
Fig. 18 An exploded iew showing he majo componen s o he elec onics box (EBOX). I s back-
plane/daugh e -ca d a chi ec u e allows easie emo al and ins alla ion o ci cui boa ds du ing assembly
and es ing. The EBOX ac s as he suppo s uc u e o he op ical sub-assemblies
base panel o he EBOX. On he op o he EBOX is a pa ch panel o dis ibu ing powe o
he decon amina ion hea e sys ems. This panel uses nano-D connec o s o minimize mass
and olume, and allowed easy elec ical connec ion o hea e s du ing assembly o he majo
op ical subsys ems on o he EBOX.
The demul iplexe uses he same o e all a chi ec u e as ChemCam’s bu was op imized
o Supe Cam. This included educing he o e all leng h by ∼25 mm o allow mo e oom
o he ansmission spec ome e , and in oduc ion o an imp o ed ibe cable suppo
b acke o be e suppo he ibe cables in ib a ion and o p o ec he ibe op ic cables
du ing handling. The modi ica ions also included accommoda ions o he e ined op ical el-
emen s. Figu e 7shows he op o-mechanical design o he demul iplexe . The lenses used o
ocus he ligh on o he ibe ips a e moun ed in i anium ba els wi h an i-backlash sp ings.
The ocal dis ance was adjus ed du ing assembly by u ning he ba els which we e easily
accessible wi h he demul iplexe co e emo ed. Two-dimensional alignmen o he ibe s
was accomplished by ansla ion o he ibe ip and connec o ac oss he op su ace o he
demul iplexe body. Once aligned and ocused, he ba els and ibe s we e locked in place
o p e en mo emen in ib a ion. The demul iplexe is also equipped wi h decon amina ion
hea e s, he mos a s, and a ligh - igh en ing ba le.
The e lec ion spec ome e s we e based on he same basic op ical design as ChemCam
bu he mechanical designs we e signi ican ly e ined (Fig. 11). The Supe Cam e lec ion
spec ome e s a e composed o wo mechanically iden ical spec ome e assemblies bol ed
oge he . Fo ChemCam, he h ee e lec ion spec ome e housings we e manu ac u ed om
be yllium. Due o he heal h- ela ed complica ions o using be yllium and o imp o e he -
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 25 o 87 4
mal s abili y, i was decided o use i anium o Supe Cam’s spec ome e s. Ti anium has a
signi ican ly lowe CTE han be yllium and 97% lowe he mal conduc i i y which helps
he mally isola e he op ics. Ti anium is 2.4 imes dense han be yllium, and he e o e he
o e all design equi ed signi ican op imiza ion o keep he mass as low as possible. In he
end, each Supe Cam spec ome e had almos he same mass as hose on ChemCam. The
op ical moun s we e also e ined o educe mass, imp o e bonding o op ical componen s,
and make op ical adjus men easie han on ChemCam. Each op ic is moun ed on compac
kinema ic moun s ha can be adjus ed om he ex e io o he housing. Once adjus ed and
s aked, a co e was placed o e he moun ing egion. This co e p o ides a ligh igh seal as
well as a moun ing loca ion o decon amina ion hea e s. Each spec ome e housing en s
h ough a ligh - igh ba le equipped wi h an 18 mic on il e mesh o p o ec om pa icula e
con amina ion.
The ansmission spec ome e (Fig. 14) is an en i ely new design and accoun ed o a
la ge ac ion o he de elopmen al wo k o c ea e a compac obus package ha could i
wi hin he alloca ed Supe Cam olume and mass limi a ions. The lenses a e moun ed in
Ti ba els wi h inge sp ings and e aining ings. Ti anium was chosen o be e ma ch he
CTE o he lenses o minimize he mal s esses on he lenses and allow o igh e ole -
ances. A numbe o he op ical elemen s equi ed he abili y o adjus a e assembly. These
adjus men s hen equi ed locking o p e en mo ion in ib a ion and he mal cycling. Some
o he ba els could be adjus ed o ocus om he ex e io o he spec ome e and hen
locked in place. The old mi o s and di ac ion g a ings we e moun ed in Ti lexu es ha
allowed ip/ il adjus men . Below he ansmission spec ome e is he high- ol age powe
supply (HVPS) (Fig. 14) ha powe s he in ensi ie . The HVPS is a ached o he in ensi ie
assembly only by he elec ical leads; i is isola ed he mally and s uc u ally o elimina e
any he mal leaks o induced s esses in o he op ical assembly. Packaging he HVPS in he
small space below he ansmission spec ome e equi ed special ca e.
All h ee spec ome e s ha e mechanically iden ical CCD modules moun ed o he ea
o he spec ome e housings; hey a e he mally isola ed om he housings (Fig. 19). The
modules con ain he CCDs and CCD PCBs as well as decon amina ion hea e s and he -
mos a s which in e ace wi h he cooling sys em on he EBOX. The CCD boa ds connec o
he SE boa d ia lex cables ou ed o main ain mo e han he minimum bend adius o he
cables. Addi ionally, wo o he CCD modules ha e he mos a s ha measu e he ex e nal
housing empe a u es o he spec ome e uni s and epo his in o ma ion in he s a e o
heal h da a.
Due o he e y con ined space Supe Cam occupies in he o e , a connec o in e ace
b acke was added o p o ide easie access o he elec ical connec o s du ing in eg a ion
wi h he o e (Fig. 3). Flex ci cui s ou e he connec ions om he EBOX PCBs o he op
o he connec o b acke .
Despi e signi ican inc eases in capabili y and pe o mance, he inal ligh Supe Cam
Body Uni ended up 345 g ligh e han ChemCam’s, occupying a sligh ly smalle olume.
3.2.2 The mal
The Supe Cam BU he mal design consis s o wo ac i e sys ems o CCD cooling and
o decon amina ion, and passi e design elemen s o cooling o elec onics and he mal
s abili y o op ical sys ems. In all cases, he majo he mal in e ace o he ins umen is
he RAMP, which is a luid-loop-con olled moun ing su ace o a ionics and o he ins u-
men a ion. The RAMP p o ides an in e ace empe a u e wi hin −40 o +50 ◦C. Based on
expe ience wi h he MSL o e , which has he same o e all he mal design, he RAMP in-
e ace o he BU is expec ed o be be ween 0 and +35 ◦C on Ma s. In o de o maximize
4Page 32 o 87 R.C. Wiens e al.
Fig. 25 Schema ic diag am o he BU spec ome e elec onics boa d. ADC =analog- o-digi al con e e ;
DAC =digi al- o-analog con e e ; Op Amp =ope a ional ampli ie ; MOSFET =me al-oxide semiconduc-
o , ield-e ec ansis o . See main ex o he meaning o o he abb e ia ions
second (Mbps) o commands ( om RCE o Supe Cam) and 7.5 Mbps o eleme y ( om
Supe Cam o RCE). Rese and boo bank selec ion is p o ided ia a se o RS-422 ini ial-
iza ion disc e e signals.
The C&DH also ac s as he digi al in e ace o he MU o e a 6-me e lex cable be ween
he BU and he MU. An LVDS HSS in e ace is used o ansmi RMI and Mic ophone
da a p oduc s om he MU o he BU a up o 10 Mbps. In he C&DH FPGA, his HSS
in e ace is mapped o one o he Leon3FT’s ou SpaceWi e in e aces a 30 MHz. An
LVDS uni e sal asynch onous ecei e - ansmi e (UART) in e ace a up o 9600 baud is
he p ima y command and eleme y in e ace be ween he wo uni s, and a se o LVDS
disc e e signals a e used as a lase igge , an in ensi ie igge , and a ese o he MU.
A se o h ee SpaceWi e links ope a ing a 30 MHz in e ace be ween he SE module’s
h ee CCD in e aces and he C&DH, which allows he use o di ec memo y access (DMA)
o he C&DH SDRAM. A numbe o disc e e signals o con ol he HVPS also o igina e on
he C&DH and a e passed h ough o he SE.
The C&DH con ols wo powe swi ches on he LVPS boa d, used o con ol powe o
he MU and o he BU TECs, espec i ely. SOH ci cui y is also p o ided on he C&DH o
moni o an a ay o empe a u es and ol ages.
The SE module’s p ima y unc ion is o clock and ead he e2 CCDs. A schema ic dia-
g am is shown in Fig. 25. A se o h ee ChemCam-he i age 14-bi Maxwell/DDC analog-
o-digi al con e e s (ADCs) a e used. ChemCam-he i age CCD clock and ol age condi-
ioning ci cui s a e implemen ed on-boa d. A Mic osemi FPGA is used o house he CCD
clocking logic and SpaceWi e i mwa e. The CCD pixels a e digi ized a 14 bi s, bu a e
o e -sampled o 16 bi s, aking he mean o ou eadings (Fig. 26). Addi ionally, a digi al

The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 33 o 87 4
Fig. 26 Oscilloscope aces illus a ing he co ela ed quad uple sampling used o imp o e he signal–
o-noise a io. ORST is ou pu ese pulse; R3 is eadou egis e phase-3 clock pulse
co ela ed double-sampling echnique is employed o sub ac he CCD e e ence ol age
om he ac i e pixel. These echniques oge he imp o ed he de ec o ead noise by a ac-
o o o e wo e sus ChemCam. One side e ec is ha a sa u a ed pixel can ead sligh ly
below 216 digi al numbe s (DNs) due o he a e aging and sub ac ion ha occu s simul a-
neously wi h he digi iza ion. Ano he side e ec is ha he maximum ans e a e using
his me hod is 400 kHz, which is abou 20% slowe han used by ChemCam. The esul is
somewha highe da k noise due o longe ans e imes (Sec . 4.2). Nominally, he CCDs
a e ead in a ow-summed one-dimensional (1D) mode. A wo-dimensional (2D) diagnos ic
mode is also a ailable, and can be used o image all ows o he CCDs (e.g., Fig. 12).
The SE boa d in e aces wi h he HVPS by p o iding he supply ol age om he LVPS,
along wi h igge signals om he C&DH and a gain ol age supplied by a 12-bi digi al-
o-analog con e e (DAC). Boa d-le el SOH ci cui y is also p o ided on he SE o moni-
o ing o SE, CCD, spec ome e , and HVPS empe a u es, ol ages, and cu en s.
The CCDs a e each indi idually moun ed o iden ical ci cui boa ds (Fig. 27), which
con ain on -end signal-condi ioning elec onics. These modules a e physically moun ed in
he h ee BU spec ome e s, and a e elec ically connec ed o he SE ia lex connec o s.
A numbe o ins umen empe a u e senso s a e used o moni o he LVPS, C&DH,
he CCD de ec o s, he HVPS, and he spec ome e housings. These senso s a e only ac i e
when he ins umen is powe ed on. They we e calib a ed du ing ins umen he mal- acuum
es s.
3.3.2 High-Vol age Powe Supply, In ensi ie , and Timing wi h Respec o he Lase
A cus om high ol age powe supply (HVPS) powe s he ansmission spec ome e ’s op ical
in ensi ie . A simpli ied schema ic is shown in Fig. 28. The HVPS applies h ee bias ol ages
o he in ensi ie : −600 V om he pho oca hode o he mic ochannel pla e on , +1200 V
ac oss he mic ochannel pla e on o ea , and +3800 V om he mic ochannel pla e ea
o he phospho sc een. Gain is adjus able ia 0–5 V inpu om he SE boa d using a 12-bi
DAC. This adjus men changes he bias ol age o he in ensi ie . Resolu ion is main ained
ac oss he ange o gains used o his ins umen .
The HVPS ci cui y consis s o a low- ol age side and a high- ol age side. The high-
ol age side is po ed o p e en a cing in he hin Ma s a mosphe e. The mal-cycling li e
4Page 34 o 87 R.C. Wiens e al.
Fig. 27 CCD boa d schema ic
diag am. Op Amp =ope a ional
ampli ie ; Temp = empe a u e.
See main ex o o he
abb e ia ions
Fig. 28 Simpli ied schema ic diag am o he high- ol age powe supply ha ope a es he in ensi ie depic ed
in Fig. 14.MCP=mic o-channel pla e
es s we e pe o med on se e al HVPS uni s o ensu e ha hey could su i e a la ge numbe
o diu nal cycles as will be expe ienced wi hin he o e body.
The C&DH boa d p o ides ga e-open and -close pulses ha a e synch onized wi h he
lase and wi h he CCD and i s eadou p ocess. Diagnos ic es ing o he in ensi ie ’s ise
ime and s abili y was pe o med by i ing he lase a a Raman-b igh a ge and making
sepa a e collec s wi h di e en delay imes. The lase pulse (and hence he Raman signal)
du a ion is 4 ns (Mau ice e al. 2020, his jou nal). The p ecision o he lase - o-in ensi ie
iming is app oxima ely ±14 ns, as he wo componen s a e ope a ed by di e en FPGAs in
hei sepa a e pa s o he o e (MU clock equency is 20 MHz and he BU clock speed is
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 35 o 87 4
Fig. 29 Desc ip ion o es ing he du a ion o he in ensi ie ga e by gene a ing a Raman signal on a a ge
mul iple imes wi h di e en delays o each collec . The echnique is illus a ed schema ically in (a). The
esul is shown in (b) o a 100 ns ga e. Each poin ep esen s da a collec ion a he delay gi en on he x-axis.
The y-axis ep esen s he Raman signal peak in ensi y o a gi en delay se ing. The ising po ion o he cu e
on he le o he plo (b) comes om signals ecei ed a he end o he in eg a ion pe iod (a, op); he alling
po ion on he igh in (b), a longe delays, ep esen s signals ecei ed a he beginning o he in eg a ion
pe iod, as shown a he bo om o (a)
50 MHz; iming uses he hal - equency s eps o double he accu acy). Figu e 29 shows he
esul s o a 100 ns in ensi ie ga e, collec ed om an ensemble o eadings a di e en ga e
imes. The ise and all imes o he signal a e 30–40 ns, a li le longe han he sum o he
iming unce ain y and he lase pulse du a ion, indica ing ha he in ensi ie goes om ull
o o ull on in a ew (10–20) nanoseconds. The minimum ecommended ga e ime o his
sys em is 100 ns. Sho e ga es can be used, howe e , he ansis o - ansis o logic (TTL)
pulses s a o o e lap, which can s ess he sys em, and he gain cu e becomes comp essed.
Gi en he iming unce ain ies and a iable obse a ion dis ances, he peak o he gain cu e
may no coincide wi h he a i al o he ligh a he in ensi ie i sho e ga es we e o be
used.
The op o he cu e in Fig. 29b shows ha s abili y is wi hin ±10%. The sligh dips
a 640 and 670 ns a e likely pa o a ixed pa e n. Tes s wi h longe cables be ween he
in ensi ie and he HVPS showed lowe s abili y due o inging caused by sligh impedance
misma ches in he cables. Gi en he expec ed ep oducibili y o he pa e n, s abili y be ween
obse a ions made unde he same condi ions a e gene ally be e han he s anda d de ia ion
ac oss he pla eau in he igu e.
The ze o poin o he iming uni s on he x-axis in Fig. 29 is a bi a y. The lase ac ually
i es a a ound =660 ns in e e y case, and he delay is adjus ed o mo e he 100 ns window
a ound he ime he ligh a i es a he de ec o . Fo a a ge ∼2 m dis an , i akes ∼33 ns
o he ligh o a el om he lase o he a ge , back o he elescope, and down he 6 m
op ical ibe o he BU spec ome e . A delay se ing o 690 ns esul s in jus ca ching he
Raman signal as he ga e opens comple ely. A he o he end, a delay o 590 ns esul s in he
in ensi ie obse ing he Raman signal jus as he 100 ns ga e is s a ing o close. The BU
con ols he i ing o he lase , and he ze o poin o he in ensi ie delay ( =0 ex apola ion
om Fig. 29) was designed o occu be o e he lase i es.
A numbe o de ails a e co e ed du ing he sequence con aining he lase pulse and he
ICCD ga e. Fi s , he in ensi ie HVPS mus be powe ed on, he CCDs mus be powe ed
on, and he lase capaci o s mus be cha ged. I he CCD is no ye in eg a ing (i.e., o he
4Page 36 o 87 R.C. Wiens e al.
Fig. 30 Ope a ing s a es o he
Supe Cam BU as commanded by
he o e . POR =powe -on ese ,
which is i s achie ed by
p o iding powe o he
ins umen
i s o se e al lase pulses o be collec ed on a single CCD exposu e, o using single-sho
exposu es), hen he pixels a e con inuously being dumped o he dump d ain nea he se ial
egis e . The CCD exposu e is s a ed and a “ i e- he-lase ” pulse is sen on a disc e e line o
he lase . The signal is ecei ed a he MU, which s a s lase pumping and hen igge s he
Q-swi ch Pockels cell (Mau ice e al. 2020). Simul aneously he MU sends a “lase sync”
signal on a disc e e line o he BU, which s a s he delay ime o he in ensi ie , opening
he ga e by pulsing he ca hode ol age a he p ope ime. The in ensi ie delay uses 24
bi s; he ga e uses 32, each bi ep esen ing a 10 ns ime ick. These esul in a maximum
in ensi ie exposu e o 42.95 s and a maximum delay o 167 ms. The in ensi ie ol age is
e eshed e e y millisecond o exposu es ha exceed 1 ms.
Tes s wi h calib a ed lamps showed excellen s abili y be ween successi e spec a, wi h
s anda d de ia ions o 0.5–1.0% o he e lec ion spec ome e s and in he 1.5–3.0% ange
o he ansmission spec ome e o 30 spec a aken a oom empe a u e. In eg a ion imes
o he e lec ion spec ome e s we e 5 ms, in eg a ing ac oss he cen e 16 e ical ows.
The ga e o he ansmission spec ome e was 10 µs, wi h a gain se ing o 2500. The
ligh sou ce was an Ene ge iq EQ-99 lamp posi ioned a a dis ance o 5 m, as desc ibed in
Sec . 7.1.1.
3.3.3 Powe Consump ion
The Supe Cam BU consumes 12.0 W when idling. Addi ional powe is used by he CCDs,
he HVPS, and he in ensi ie when hey a e ope a ing, b inging he maximum BU powe o
19.1 W. The TECs consume an addi ional 24.0 W.
3.4 So wa e
The BU ligh so wa e consis s o a VxWo ks 6.7 ope a ing-sys em ke nel and he cus-
om Supe Cam applica ion. The e a e wo copies o he ke nel and applica ion s o ed in
adia ion- ole an magne o esis i e andom-access memo y (MRAM). A boo loade esides
in a adia ion-ha d p og ammable ead-only memo y (PROM) ha con ols he s a up o he
ins umen ligh so wa e. Figu e 30 gi es a s a e diag am o he ins umen . The Supe Cam
BU so wa e s a s by execu ing a boo loade con ained in a PROM. Based upon he ini-
ializa ion signal sen o he ins umen , he boo loade selec s one o he wo ke nels and
applica ions s o ed in MRAM. The boo loade pe o ms in eg i y checks on he so wa e,
loads i in o synch onous dynamic andom-access memo y (SDRAM), and s a s execu ing
he ligh so wa e. Once he ligh so wa e begins execu ion, i ini ializes i sel , pe o ms
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 37 o 87 4
a buil in sel - es , hen eaches he idle s a e awai ing commands om he o e compu e
elemen (RCE). When a command is ecei ed, he ins umen goes o an ope a e s a e and
execu es he command. Once he command is comple e i e u ns o he s andby s a e. Only
one command can be execu ed a a ime and he RCE mus wai un il he ins umen is in he
s andby s a e be o e sending ano he command. The only excep ion is he abo command.
The ins umen ligh so wa e is designed o handle o -nominal si ua ions. The ins u-
men handles e o s in commanding, such as in alid pa ame e s o in alid commands, using
eply condi ion codes as de ined by he p ojec . Ins umen -speci ic o -nominal si ua ions
a e epo ed using condi ion codes and s a us lags as de ined in he ins umen command
dic iona y. The ins umen main ains wo logs in non- ola ile memo y: an ins umen e o
log and a command his o y log. I a so wa e p ocessing e o occu s, a imes amp and an
e o code a e eco ded ha indica e he na u e o he e o ha occu ed. Bo h logs can
be placed in o he ins umen ’s da a bu e and ansmi ed o he RCE using speci ic dump
commands and he ansmi da a command as de ined in he command dic iona y.
Du ing he mission, i will be possible o upload new so wa e, including new ke nels, o
applica ions, using he LOAD_MEMORY command as de ined in he ins umen command
dic iona y. I is possible o w i e o any a ea o memo y using his command.
3.4.1 Command Handling and Ins umen S a us
Two key esponsibili ies o he BU so wa e a e o ecei e and p ocess commands com-
ing om he RCE, and o p o ide eleme y and da a o he RCE. Some commands like
CONFIGURE_CCD_REGIONS a e execu ed quickly and only a command eply is sen o
he RCE. O he commands like DO_FOCUS, which ake much longe o execu e, esul in
he BU sending a command eply ollowed some ime la e by a science da a ame. Bo h
esponses o he RCE con ain s a us lags and a condi ion code o communica e he ins u-
men ’s s a e.
The e a e i een pe sis en s a us lags ha he so wa e main ains and a e communica ed
o he RCE as pa o e e y ins umen ans e ame. These include lags o MU powe ,
MU di ec cu en (DC) con e e s, IR spec ome e eady, con inuous-wa e lase (CWL)
eady, lase equency double eady, shu e s a us, lase s ack eady, lase HVPS on, sun
sa e y, BU HVPS s a e, and CCDs almos eady and eady. Some o he s a us lags a e o
MU unc ions ha a e explained in he companion pape (Mau ice e al. 2020, his jou nal).
I he ins umen so wa e de ec s a aul , he so wa e will ale he RCE by se ing he
SEND_EVR, SYS_ERROR, o SHUTDOWN s a us lags depending on he se e i y o he
aul .
In addi ion o s a us lags, he e a e 30 condi ion codes ha he ins umen so wa e can
use o communica e s a us wi h he RCE. These ange om indica ing he e was co up ion
o he ecei ed command, o an e o communica ing wi h he spec ome e elec onics, o
an e o communica ing he MU.
Ins umen commands ep esen one le el in a la ge se o nes ed le els. Figu e 31 illus-
a es he ac ha he o e uses a highe le el o (spacec a ) commands, such ha a space-
c a command may pa se se e al Supe Cam ins umen commands. Highe le els exis in
he command s uc u e such ha a pe son using he Componen -based Campaign, Planning,
Implemen a ion, and Tac ical (COCPIT) ool o planning he o e ope a ions may be able
o use a empla e o ca y ou a signi ican po ion o he emo e-sensing ac i i y o a gi en
ime ame.
The ins umen BU so wa e only esponds o ins umen le el commands. Those com-
mands a e combined o comple e a highe le el unc ion. As an example, he ollowing
simpli ied sequence o ins umen commands could be used o pe o m a LIBS collec :

4Page 38 o 87 R.C. Wiens e al.
Fig. 31 Illus a ion o he nes ed na u e o he commands ha ope a e Supe Cam, spanning om ins umen-
-le el commands (I-cmds) o spacec a -le el commands (S-cmds) up o componen s, which ope a e mo e
han one ins umen
•COLLECT_SOH
•MU_SEND_HOUSEKEEPINGS (housekeeping da a gene ally consis s o SOH and s a-
us lags)
•CONFIGURE_CCD_VERT_TIMING ( his command and he nex wo con igu e he e -
ical and ho izon al iming o he CCD exposu es, and he speci ic ows om which o
collec da a)
•CONFIGURE_CCD_HORZ_TIMING
•CONFIGURE_CCD_REGIONS
•CONFIGURE_HVPS
•CONFIGURE_INTENSIFIER
•MU_DC_DC_CHAIN_ON ( u ns on he DC con e e o enable a ious MU unc ions)
•COLLECT_SOH
•MU_SEND_HOUSEKEEPINGS
•MU_CONFIGURE_LASER
•DO_SPECTRA ( i es he lase and collec s he spec a using he speci ied lase , in ensi-
ie , and CCD con igu a ions)
•COLLECT_SOH
•MU_SEND_HOUSEKEEPINGS
•XMIT_DATA (sends he esul ing da a o he o e )
The BU so wa e has been designed such ha all commands ecei ed and all da a eplies
like s a e o heal h (SOH) o spec a a e assembled in o a single gene ic bu e in he in-
s umen . Da a ma ke s and da a leng hs a e used o iden i y each segmen and make he
gene ic bu e pa sable. The en i e gene ic bu e is ans e ed a once o he RCE using
he XMIT_DATA ins umen command. Sec ion 4.4 p o ides de ail on he o ma ing o he
da a p oduc gene a ed by he ins umen .
3.4.2 MU Commanding, Focus Managemen , and Moni o ing
O he esponsibili ies o he BU ligh so wa e include powe ing and commanding o he
MU, ocus managemen , and moni o ing o he MU. Ins umen -le el commands om he
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 39 o 87 4
RCE in ended o he MU a e in many cases jus a pass- h ough o he BU so wa e. How-
e e , he e a e imes when he BU mus modi y command a gumen s be o e o wa ding he
command, and he e a e imes such as he DO_FOCUS command whe e, as pa o he com-
mand execu ion, he BU so wa e will in e nally gene a e an MU-speci ic command. The
commands and MU eplies a e sen o e a UART da a link be ween he BU and MU. Sim-
ila o he in e ace wi h he RCE, he MU indica es s a us in i s command and da a eplies
o he BU.
The BU ligh so wa e is esponsible o main aining he s a e o he MU ocus posi ion.
This includes keeping ack o he ocus posi ion in mo o s eps ( able posi ion), con e ing
he able posi ion o dis ance o he a ge in millime e s, con e ing command a gumen s
in millime e s o mo o s eps, calcula ing sligh ocus o se s needed o op imally pe o m
ei he RMI o spec oscopy, and main aining he s a e o he sun-sa e lag, indica ing ha he
ocus posi ion is no nea in ini y (Mau ice e al. 2020), which would be dange ous o he
op ics i he sun passed h ough he FOV. The ocus mo o posi ion is held in he BU’s non-
ola ile memo y. When he ocus mo o posi ion changes ei he due o a manual mo e o an
au o ocus, he MU communica es he change o he BU. I he ocus posi ion main ained by
he BU is deemed o be inaccu a e, he p ope posi ion can be ees ablished by d i ing he
mo o o a limi swi ch.
The e a e se en componen s on he MU ha he BU moni o s o ensu e hey a e wi hin
he allowable ligh empe a u e (AFT): lase s ack, lase doubling c ys al, ocus able, CWL
lase , in a ed spec ome e acous o-op ic unable il e (IRS AOTF), and IRS TEC ho side,
and LPVS. The BU checks he empe a u es e e y ime an MU_SEND_HOUSEKEEPINGS
command is sen , and, i he e is no o he command being execu ed, he BU will au oma i-
cally send he command e e y en seconds. I a empe a u e is ound o be ou o AFT, he
BU will se an e o s a us lag and communica e i o he RCE when he nex command is
ecei ed.
4 De ails o Ope a ion
4.1 O e iew
In gene al, Supe Cam is designed o ope a e in a simila manne o ChemCam, which akes
images and pe o ms spec al analyses on a se ies o poin s a anged in a line o g id, e e ed
o as a “ as e .” I uses an au o ocus ou ine and akes an RMI image a he beginning o
each as e and occasionally he ea e , as equi ed o main ain good ocus and o co e he
as e a ea wi h con ex imaging. Upon close look, he e a e many mo e de ails in ol ed
in Supe Cam’s ope a ions due o he la ge numbe o ypes o spec al obse a ions and
he inc eased complexi y o some o hem. Figu e 32 shows a gene al low diag am o
ope a ion o Supe Cam. The igu e and he desc ip ion below includes all o he ypes o
spec oscopy, bu obse a ions using only one ype o spec oscopy (o only imaging), o a
di e en combina ion o a di e en o de o obse a ions, a e all equally easible.
Ta ge acquisi ion, p epa a ion, and RMI imaging: A a ge is selec ed by he science
eam, and i s coo dina es a e uplinked o he o e as pa o he sequence o commands. The
a ge could also be selec ed by AEGIS (Au onomous Explo a ion o Ga he ing Inc eased
Science), an onboa d algo i hm cu en ly being used on MSL o au onomously selec ing
ChemCam a ge s (F ancis e al. 2017). I is also expec ed o be able o pe o m closed-
loop poin ing based on pa e n ecogni ion by he o e compu e ope a ing on Na cam o
Supe Cam images. Ano he possible op ion is o use blind a ge ing, in which he ins umen
4Page 40 o 87 R.C. Wiens e al.
Fig. 32 Flow diag am o
ypical Supe Cam ope a ions.
This illus a ion includes all o
he spec al echniques, hough i
is no necessa y o use all
echniques on a gi en
obse a ion, no o use he o de
o aking spec a shown he e. See
ex o desc ip ion
is poin ed a a ixed loca ion on he g ound in o e coo dina es wi hou any knowledge
o he e ain (Mau ice e al. 2016). This was done on ChemCam p io o AEGIS. In any
case, he o e is commanded o u n Supe Cam on and o send i he p ope commands
o he sequence o analyses. Once Supe Cam is on, i is commanded o do se e al hings
in p epa a ion o analysis (Fig. 32): Assuming ha LIBS, Raman, o VISIR spec oscopy
will be done, he TECs will be u ned on o s a p e-cooling he CCDs. A p e-cooling ime
o 15–20 minu es is expec ed o be used p io o he s a o aking spec a. I LIBS o
Raman spec a a e planned, he Nd:YAG lase will be wa med o i s ope a ing empe a u es
o −15 ◦C o he s ack and −10 ◦C o he equency-doubling c ys al (Mau ice e al.
2020). Finally, i he CWL will be used o au o ocus, i needs o be wa med o −10 ◦C.
The p e e ed au o ocus me hod uses a se ies o RMI images, o which no wa ming is
equi ed (Mau ice e al. 2020). Du ing he mal equilib a ion, he RSM u ns owa d he
p ope coo dina es o he a ge . Au o ocus is he nex s ep, ollowed by an RMI image
cen e ed a he i s poin o he as e . A e he RMI da a ha e been ans e ed, spec a a e
aken on he i s poin .
LIBS: Typically, i LIBS plus o he spec a a e planned, we expec o pe o m LIBS i s ,
as he LIBS shock wa e emo es dus om he a ge , clea ing he su ace o he o he
echniques. The lase pi is much smalle in diame e (∼250–400 µm; Mau ice e al. 2020,
his jou nal) han he FOVs o he Raman, VIS (bo h 0.74 µ ad), and IR (1.15 µ ad) obse -
a ions. A a a ge dis ance o 2.5 m, he a ea o he LIBS lase pi ep esen s ∼2.5% o
he FOV o Raman and VIS obse a ions, and ∼1% o he FOV o he IR obse a ions.
Because o his, he damage caused by he lase is expec ed o ha e negligible impac on
he o he obse a ions (Fau e al. 2019). Ins umen commands p epa e he lase and spec-
ome e s. Bo h ac i e (wi h lase ) and da k (backg ound, wi hou lase ) LIBS spec a a e
aken a 3 Hz; he da ks will be sub ac ed om he ac i e spec a on he g ound. We plan
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 41 o 87 4
o use ∼30 lase sho s o each s anda d LIBS obse a ion. A sepa a e spec um is e u ned
o each lase sho , while we downlink only s a is ics (mean, median, and s anda d de ia ion
o each channel) om he 30 da k collec s. The sepa a e ac i e spec a po en ially eco d
changes as he lase p o iles up o ∼100 µm in o he sample (assuming 30 lase sho s). Mi-
c ophone eco ding o he LIBS plasmas is op ional; acous ic da a a e sen o he o e a e
he eco ding is inished. A e he LIBS da a ha e been ans e ed o he o e , he nex
spec al ac i i y can begin.
Raman/TRL: Fo Raman and TRL spec oscopies, changes in he MU include deploying
a beam s op o block he LIBS lase pa h and p epa ing o use he Pockels cell o cause
he lase beam o be equency doubled, o p oduce a g een 532 nm beam (Mau ice e al.
2020). Ins umen commands p epa e he ansmission spec ome e . Bo h ac i e and da k
spec a a e aken, usually using 100–200 lase sho s a 10 Hz. Fo Raman spec oscopy, he
ac i e spec a can be a e aged on boa d, as no changes o he sample ake place as a esul o
successi e lase pulses. In ha case, he mean, median, and s anda d de ia ion a e e u ned
o ac i e and da k collec s, each. Fo TRL, sho bu s s o lase sho s a e i ed, each bu s
wi h a di e en ime delay be ween he lase and in ensi ie (see Sec s. 4.2.2 and 7.4) o
eco d a ime spec um o he in ensi y o he lase -induced luminescence. Once he Raman
o TRL obse a ion is comple e, he LIBS lase beam s op mus be emo ed; he e is a ime
limi o wo minu es o i s deploymen (Mau ice e al. 2020).
VISIR: Fo IR obse a ions, he IR TEC in he MU mus cool he IR pho odiode ∼50 ◦C
below he MU empe a u e be o e he obse a ion. VIS spec a a e aken using he VIO and
ansmission spec ome e s in he BU, ollowed immedia ely by IR spec a (Mau ice e al.
2020). The Supe Cam IRS is wa eleng h-scanning, and ypically akes <80 s o acqui e a
spec um plus da ks. The VIS spec al acquisi ion is done in se e al seconds a mos .
Typically, once hese spec a a e aken, he RSM mo es o a new poin on he same a ge ,
de eloping a as e pa e n, usually wi h poin - o-poin spacing o 0.5 o 2 m ad (=1.25 o
5 mm a a ypical dis ance o 2.5 m). Planned as e s include 1×5and1×10 line scans, and
2×2and3×3 g id pa e ns. As he as e p oceeds, an addi ional au o ocus may be needed
o ensu e p ope ocus is main ained, especially in he case o a e ical as e , whe e he
dis ance o he a ge is mo e likely o change. Ano he RMI image is aken a he conclusion
o a as e , and depending on he angula dis ance ha he as e co e s, an addi ional RMI
image may need o be inse ed in he middle o ensu e imaging documen a ion o all as e
poin s. Typically an RMI image should be aken e e y 8 m ad o less o keep he spec al
obse a ion poin s wi hin i s FOV (Table 1).
This p ocess can be epea ed o successi e a ge s. Se e al clean-up ac i i ies a e pe -
o med o conclude he obse a ions o one o se e al a ge s, shown in Fig. 32.Thesein-
clude mo ing he ocus s age back o he sun-sa e posi ion, and pe o ming a con olled
wa m-up o he IRS TEC. A e ha he ins umen is powe ed o .
The h ee c i ical esou ces o Ma s ope a ions a e ime, ene gy, and da a olume. T ack-
ing and op imizing each o hese is impo an o e icien ope a ions. A no ional powe
p o ile is p o ided in Fig. 33. In his example, he BU TECs a e powe ed on o 12 min-
u es be o e he MU lase wa ming is s a ed. Ac i i ies ha ollow a e CWL au o ocus, RMI
imaging, LIBS +Mic ophone, Raman, and VISIR, ollowed by a inal RMI image, and end-
ing wi h con olled shu -down o he ins umen . Jus one poin is obse ed in his example,
bu using se e al spec al echniques. The u n-on and he mal equilib a ion akes o e 15
minu es, while he analysis po ion o he sequences akes a li le unde ou minu es. Gi en
he ela i ely long ime i akes o u n-on and he mal equilib a ion, i is use ul o spend
enough ime o make mo e obse a ions once he e o has been made o p epa e he ins u-
men . The MU is equi ed o be able o ope a e wo hou s a a ime. A e ha , i needs o
4Page 48 o 87 R.C. Wiens e al.
da a, and auxilia y da a (e.g. he posi ion coo dina es o he RSM) o he command his o y.
EDRs con aining science da a will be p ocessed and calib a ed by a speci ic pipeline, and
he esul ing p oduc s a e called clean da a eco ds (CDRs).
5 Model De elopmen and En i onmen al Tes ing
The o e all plan o de elopmen o Supe Cam emphasized ea ly alida ion o new ech-
nologies, pa icula ly he ansmission spec ome e and i s HVPS as applied o Raman
spec oscopy, ollowed by an en i onmen al quali ica ion model (EQM) and ligh model
(FM). To acili a e he ea ly alida ion, an enginee ing de elopmen uni (EDU) was de el-
oped and es ed. I s MU consis ed o a lase and associa ed elec onics, elescope, RMI, and
an in a ed spec ome e ha was ibe -coupled o he elescope. The EDU BU consis ed o
elec onics, demul iplexe , e lec ion spec ome e s, and a ansmission spec ome e . TEC
cooling o he EDU CCDs was done wi h comme cial uni s ia a non- ligh coppe ba
om he CCDs. The EDU ansmission spec ome e had only wo spec al bands ins ead o
he h ee in he EQM and FM, and so i had low esolu ion. In eg a ed es ing o he EDU
Supe Cam ins umen was done in he sp ing o 2016 a LANL, and was used o e i y a
numbe o de ails, including (a) o e all coupling o he BU and MU, including he iming
be ween he lase and spec ome e s; (b) pe o mance wi h he e lec ion spec ome e s ha
was compa able o ChemCam; (c) basic Raman and TRL unc ionali y and h oughpu ; and
(d) scien i ic s udies o TRL spec a (Ollila e al. 2017). The la e was easible because he
luminescence peaks a e o en b oad, and a e hus ela i ely insensi i e o he esolu ion o
he spec ome e , which was no op imized a his ea ly de elopmen s age. The EDU did no
use ligh -quali ied elec onic pa s, and i was no buil o en i onmen al es ing. En i on-
men al es ing was ins ead only ca ied ou on new sub-assemblies, such as he ansmission
spec ome e , be o e in eg a ion in o he BU.
A e he EQM was buil , he MU EDU was s ipped o i s lase and in a ed spec om-
e e . A special BU simula o was econs uc ed om spa e pa s, wi h no op ics. The wo
uni s we e deli e ed o JPL in 2017 as he Supe Cam Tes Uni (TU), which suppo ed e -
i ica ion and alida ion (V&V) ac i i ies and ope a ions aining. This uni will be moun ed
on he ehicle sys em es bed (VSTB) a JPL and will be main ained by he eam o he
du a ion o he mission on Ma s. The BU EDU is planned o be used as a calib a ion model
in he labo a o y. A new MU will be ebuil o ha .
The mos signi ican change be ween he EDU and EQM BU was he ansmission spec-
ome e (Sec . 3.1.4), which was edesigned wi h he dich oic beam spli e and compound
g a ing o yield h ee wa eleng h bands ins ead o he wo ha he EDU had. This imp o ed
he esolu ion o mee he 12 cm−1 equi emen . The EQM BU unde wen en i onmen al
es ing in LANL, while he EQM MU unde wen es ing in Toulouse be o e in eg a ing he
uni s. Tes ing o he BU iden i ied h ee issues: sligh mo ion o he e lec ion spec ome e
op ics du ing ib a ion and shock, unde -pe o mance o he TECs, and a componen ailu e
in he HVPS a e i s cold he mal cycle. The e lec ion spec ome e s we e only sligh ly
edesigned om ChemCam. Fo bo h ins umen s he op ics a e held in ension agains a
h ee-poin moun . The p oblem was ne e seen on ChemCam, and o Supe Cam, he only
occu ences we e a he quali ica ion le els, wi h ex emely mino mo emen a he ligh -
accep ance ins umen ib a ion, and no mo emen du ing o e shock o ib a ion. The
HVPS componen ailu e led o e-analysis o he mechanical s esses on he componen s,
and some e-posi ioning o componen s. The TECs we e no modi ied on he FM ela i e o
he EQM, bu a numbe o imp o emen s we e made o he ligh TEC cooling sys em o

The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 49 o 87 4
insula e i u he om he en i onmen and o sp ead he hea o he RAMP be e . Du ing
es ing o he EQM BU, he wa eleng h dependence on empe a u e o all h ee spec ome-
e s was shown o be much less on Supe Cam (Sec . 7.1.3) han on ChemCam (Wiens e al.
2012).
The EQM is ully unc ional. The EQM BU and MU equi ed some ewo k pos -FM-
deli e y o ma ch he FM pe o mance. They will be used as a calib a ion model, pa icula ly
o complemen he LIBS spec al da abase o known composi ions, and o acqui e a lib a y
o ime- esol ed Raman and luminescence signa u es. Thus he EQM is c i ical o achie ing
Supe Cam’s o e all scien i ic mission.
The ligh BU has e y ew changes ela i e o he o iginal EQM o he han hose men-
ioned abo e. The mal, Ma s-p essu e es ing o he FM iden i ied one new issue: The in-
ensi ie c ea ed lashes o ligh a andom imes when a Ma s p essu e. This was due o an
impe ec ion in he mul i-channel pla e and su ounding po ing. Po ing o he ubes was
mos success ul on he EQM ube, and so ha ube is being used o ligh . I was also ound
ha he EQM in ensi ie ube had a small endency o c ea e lashes a Ma s p essu e in N2
gas. Ni ogen is he es gas o choice, as i is easie and sa e o use in a he mal chambe ,
and ep esen s an o e - es ela i e o he Ma ian a mosphe e, o which b eakdown does
no occu as easily. The lashes we e likely due o cha ge buildup on he ou pu window o
he in ensi ie due o elec ic ield leakage h ough he uni . Ex ensi e es ing was ca ied
ou in Ma s gas, showing ha he e we e no lashes unde Ma s condi ions using no mal
ope a ing p ocedu es.
As pa o he equi emen s o deli e y he FM BU unde wen a se ies o en i onmen al
quali ica ion es s including andom ib a ion and he mal acuum, which we e conduc ed
using Los Alamos acili ies. Random ib a ion es ing was pe o med in each o he h ee
axes a p o o ligh le els o one minu e pe axis. A e ib a ing each axis a isual inspec ion
was pe o med and he accele ome e da a was e iewed o assess any mechanical change
o he uni . A comp ehensi e unc ional es was also conduc ed be o e he es , be ween
axes, and a e he es o de e mine any pe o mance change a e exposu e o he ib a ion
en i onmen .
The BU hen unde wen 204 hou s o he mal acuum es ing. The chambe suppo ed
bulkhead eed h ough connec o s ha allowed powe , command and eleme y signals, and
an op ical ibe o be coupled om suppo equipmen ou side he chambe o he uni in
he chambe . When ins alled on he o e he Supe Cam BU has i s basepla e a ached o
he unde side o he RAMP such ha he he mal sink is abo e he CCDs. The o ien a ion
a ec s he pe o mance o he hea pipes, so we needed o mimic he BU o ien a ion in he
o e o ealis ic he mal es s. Fo his, a seconda y pla en was moun ed on legs inside he
he mal acuum chambe wi h i s moun ing su ace acing down. The basepla e o he BU
was hen bol ed o his su ace. To comple e he es se up a numbe o he mocouples we e
a ached o he uni .
Du ing he mal acuum es ing he BU comple ed h ee he mal cycles ac oss i s
p o o ligh empe a u e ange o +60 C o −50 C. The i s he mal cycle was conduc ed
a high acuum below 1 ×10−5To o eplica e c uise condi ions. The emaining wo cy-
cles we e conduc ed a Ma s p essu e (∼7 To ) using gaseous ni ogen. Each he mal cycle
included long soaks a he maximum and minimum empe a u es. Du ing he soaks and
a a ious in e media e empe a u es, es ing was conduc ed o e i y he unc ionali y o
he elec onics and he e ec i eness o he TECs o cool he CCDs, and o assess he pe -
o mance o he spec ome e s. Du ing he high- acuum cycle he decon amina ion hea e s
we e es ed o e i y hei abili y o main ain he op ics a he desi ed empe a u e.
Ea lie in he p ojec , py o-shock es ing was pe o med on he EQM BU using Los
Alamos acili ies. Resul s om his es we e used o app o e he BU FM wi hou es ing,
4Page 50 o 87 R.C. Wiens e al.
limi ing po en ial mechanical s ess on he ligh model. Addi ionally, an elec omagne ic
compa ibili y and elec omagne ic in e e ence (EMC/EMI) es was pe o med using he
EQM BU coupled wi h he EQM MU a a JPL acili y. Resul s om he EMC/EMI es ing
showed ha he BU exceeded maximum adia ed signal le els in a couple c i ical equency
bands. This was add essed on he FM by adding some addi ional EMI shielding o speci ic
a eas o he uni . Final es ing o he ins umen on he o e e i ied ha Supe Cam mee s
equi emen s, and can be used du ing UHF communica ions, which is an imp o emen o e
ChemCam.
6 Pe o mance Tes ing
The es and alida ion campaign o Supe Cam was cons ained by he a ailabili y and capa-
bili ies o he uni s and models, as desc ibed below. Table 8shows a summa y o he es ing
o a ious models and con igu a ions. The EDU was use ul o es ing o unc ionali y, de-
spi e he ac ha some o he EDU subsys ems we e no ully ligh -like, and some we e
no p esen . Fo example, he mic ophone was no pa o he EDU, and as men ioned, he
ansmission spec ome e only p o ided low esolu ion. The EQM was a be e es bed, as i
was ligh -like in nea ly e e y sense. I p o ided a s able pla o m o s udies o Raman pe -
o mance, al hough a ew aspec s o i s pe o mance (alignmen , ocus a cold empe a u es)
we e no as good as ha o he FM. Thus, Raman and TRL s udies could be made a close
dis ance (e.g., 2–2.5 m), bu no a dis ances >5 m. Raman and TRL obse a ions we e also
ca ied ou wi h he FM uni s, o en e- es ing samples analyzed wi h ea lie models.
Fo he ligh model, a p oblem a ose in he all o 2018 ha equi ed a comple e e-
build o he MU (Mau ice e al. 2020, his jou nal). Because o ha , he ligh MU was no
eady un il a e he equi ed deli e y da e o he BU o he o e . Valida ion o he BU in a
ealis ic en i onmen , de elopmen o a LIBS spec al lib a y, and ini ial calib a ion o he
ins umen esponse unc ion (sensi i i y s. wa eleng h, o IRF) o he sys em had o be
done on a combina ion o FM BU +EQM MU. To alida e c i ical MU unc ions such as
he au o ocus o se s (o se s as a unc ion o empe a u e o di e en ela i e ocus poin s
o LIBS, RMI, and he CWL au o ocus mode; Mau ice e al. 2020), a e deli e y o he FM
BU o he o e , he EQM BU was sen o Toulouse o suppo es ing in he con igu a ion
o EQM BU +FM MU. Finally, as shown on he igh side o Table 8, he las ligh po ion
o he ins umen was he FOC, he ligh e sion o which ne e le JPL. So when he BU
and la e he MU we e in eg a ed on o he o e , all o he ligh pa s o he ins umen
we e inally oge he . Because o he sho ime a ailable in he inal ligh con igu a ion,
esul s (Sec . 7) will be p esen ed om a ious con igu a ions, as app op ia e o ep esen
he o e all pe o mance o he ligh ins umen . Because we could c oss-check he FM uni s
wi h he EQM uni s, and because he EQM uni s we e gene ally e y simila o he ligh
uni s, we belie e he di e ences in cha ac e iza ion ela i e o he comple e ligh uni we e
ela i ely mino and limi ed o hose aspec s no ed he e.
6.1 Fligh -Model Body Uni +Enginee ing Quali ica ion Model Mas Uni
Tes ing a LANL
F om 6 o 23 Ap il 2019, he FM BU was es ed in i s inal con igu a ion wi h he EQM MU
o pe o mance and so wa e alida ion, and o de elop spec al lib a ies. Bo h he BU and
MU we e housed in a he mal chambe wi h d y ni ogen a ambien p essu e, and a −10 ◦C
o op imize he lase ene gy. Fo LIBS, a Ma s sample chambe p essu ized o 5.8±0.2To
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 51 o 87 4
Table 8 Summa y o es ing o he Supe Cam models
BU MU Loca ion
o es ing
#Da a
iles
S udy
Raman, TRL
Valida e
unc ionali y
C oss
calib a ion
LIBS spec al
lib a y
Au o ocus
o se s
Realis ic he mal
en i onmen
Fligh
ibe
EDU EDU LANL TRL X
EQM EQM LANL 5534 Raman X
FM EQM LANL 3100 X X X
EQM FM IRAP 743 X X X
FM FM JPL 1119 X X X X
4Page 52 o 87 R.C. Wiens e al.
Table 9 Numbe s o unique samples o ocks, mine als, and mine al g oups obse ed by he FM BU and
EQM MU du ing he alida ion pe iod
Rock/mine al ype LIBS Mine al/ ype Raman VISIR
Andesi e 14 Ca bona e 17 5
T achy e 4 Silica e 12 6
Ano hosi e 6 Plagioclase- eldspa 9 1
Basal 33 Sul a e 8 4
Py oxene 7 Oli ine 6 1
Plagioclase- eldspa 3 Phospha e 5
Oli ine 6 Phyllosilica e 4 9
Dole i e 3 Py oxene 3 1
Gabb o 8 Se pen ine 2 3
No i e 5 Mica 1
Syeni e +g ani e 5 Amphibole 1
Ca bona e 8 Diamond 1
Sul a e 7 Fluo i e 1
Fe, Ti, oxides, banded i on 8 Me al oxide 1 9
Mn oxides 13 Sul ide 1
Me amo phic, A chean sedimen s 49 Ni a e 1
Clas ic sedimen s including shale 91 Pe chlo a e 1 1
Phyllosilica e (kaolini e, smec i e) 6
Miscellaneous o he 31 O ganic, mine al mix 51 15
SCCT 25 SCCT 14 13
TOTAL 332 TOTAL 138 68
wi h CO2was used o simula e he Ma s a mosphe e, which a ec s he plasma p ope ies.
Du ing obse a ions o samples in he Ma s chambe , he window o he ins umen he mal
chambe was emo ed, and a pipe was ins alled be ween he ins umen he mal chambe
and he window o he sample chambe . In his way, he e was a single window be ween
he ins umen and he samples, as will be he case on Ma s. Raman, TRL, RMI, and VISIR
obse a ions we e made wi h he ins umen -chambe window in place and he samples in
open ai .
The p ima y pu pose o he LIBS obse a ions was o es ablish a lib a y o spec a o a
di e se sui e o geological ma e ials, mos ly a one dis ance, and hen a smalle lib a y a
se e al dis ances. Expe ience wi h ChemCam has shown ha a la ge numbe o s anda ds
a e needed o quali y calib a ion using mul i a ia e me hods o all o he majo elemen s
and a mix u e o me hods o mino and ace elemen s (Wiens e al. 2013; Clegg e al.
2017; Pay é e al. 2017). Dep h p o iles we e also p oduced using a la ge numbe o lase
sho s (Sec . 7.2.4). Table 9gi es a summa y o he LIBS obse a ions. Th ee obse a ions
we e made o each s anda d, wi h each obse a ion consis ing o 30 lase pulses. Each lase
obse a ion has a co esponding da k spec um o 30 collec s. S anda ds we e loaded in
he sample chambe on a u e ha holds app oxima ely i een 30-mm diame e p essed
powde pelle s, which is he physical o m o he LIBS s anda ds. Wi h each u e load,
a small i anium pla e was ins alled and obse ed. C oss compa ison o he Ti obse a ions
om one u e load o ano he can be used as a check agains p oblems wi h ocus, clipping
o he lase beam in he pipe, lase ene gy, o ambien p essu e. All o he LIBS a ge s we e
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 53 o 87 4
obse ed a 2.85–3.00 m dis ance. The SCCTs we e obse ed again a 1.55 m ( he dis ance
o he SCCTs on he o e ) and a 4.25 m. A ull lis o he s anda ds and he calib a ion
pa ame e s de i ed om he lib a y will be published elsewhe e.
Raman and VISIR spec a we e aken o alida e he p o ocols and o ob ain ep esen a-
i e spec a om he ins umen . Table 9lis s he di e en unique a ge s ha we e obse ed
by hese echniques ha we e ca ied ou wi h he FM BU +EQM MU. A ew TRL spec a
we e also eco ded. In addi ion o pu e mine als, some mine al mix u es we e es ed, as well
as mix u es o o ganic species wi h inac i e binde ma e ials. As men ioned abo e, many
mo e such obse a ions we e ca ied ou wi h he comple e EQM ins umen , especially o
TRL spec oscopy (Ollila e al. 2017). The esul s a e p esen ed in Sec . 7.
6.2 Enginee ing Quali ica ion Model Body Uni +Fligh -Model Mas Uni
Tes ing in Toulouse
The o e schedule equi ed he BU o be ins alled i s , lea ing se e al mo e weeks o com-
ple e and deli e he MU. Du ing his ime he EQM BU was sen o F ance o acili a e inal
es ing o he FM MU. The p ima y objec i es we e o e i y he communica ion be ween
he wo uni s and o pe o m sa e y checks by he BU so wa e ha included moni o ing o
MU empe a u es and se ing he co ec s a us lag i ou o ange, and e i ying ha he
BU se he app op ia e mo o speed, numbe o s eps, dis ance o limi swi ches, pumping
cu en , and lase es ime (30 s) be ween consecu i e bu s s. We also checked ha he BU
epo ed he co ec condi ion codes, and he sun-sa e s a us in all con igu a ions, e en i
communica ion wi h MU was los .
The seconda y objec i es we e o e i y ins umen unc ionali y, and o cha ac e ize
some key pa ame e s ha a e unique o he FM MU. These include CWL- o-LIBS and CWL-
o-RMI ocus-o se algo i hms (Table 8) ha we e subsequen ly coded in o he FM BU
so wa e. The es also checked he quali y o he LIBS ocus and he numbe o spec ome e
CCD ows o use o a oid sa u a ion a sho dis ance. Some mic ophone noise es s we e
also pe o med.
Finally, his con igu a ion p o ided an oppo uni y o acqui e some addi ional calib a ion
da a which could be used along wi h da a om he EQM MU +FM BU es s a LANL (see
Sec . 6.1). LIBS spec a we e acqui ed a 1.56 m and 3 m, and compa ed o ChemCam spec-
a on Ma s, con i ming he compa a i e esul s o be p esen ed in Sec . 7.1. A ew es s a
7 m showed be e Supe Cam pe o mance han ChemCam a ha long dis ance, bu he ac-
ual long-dis ance capabili ies will ha e o be shown on Ma s. Time- esol ed Raman spec a
we e also ob ained om gypsum. TRL spec a we e ob ained om apa i e and showed REE
and Mn2+ de ec ions a ∼100 ppm le els o a 0.5 ms in eg a ion window.
A e he success ul conclusion o hese b ie es s, he FM MU was deli e ed o he
o e .
6.3 Tes ing on he Ro e
Ex ensi e pe o mance es ing o he end- o-end FM Supe Cam ins umen on he o e ook
place in se e al campaigns. The gene al de ails a e gi en in Table 10. Essen ially all o he
unc ions o Supe Cam we e es ed a leas once (excep RMI z-s ack). Alignmen es s we e
done be o e and a e dynamic and he mal en i onmen al es ing o e i y ha he FOV o
he ansmission spec ome e is well aligned (a he p ope empe a u e) wi hin he g een
lase beam, as needed o Raman spec oscopy (Mau ice e al. 2020). These we e done using
a Spi icon came a and neu al densi y il e s. Checks we e also made o he CWL and LIBS

4Page 54 o 87 R.C. Wiens e al.
Table 10 Tes ing o Supe Cam on he o e in 2019 o ea ly 2020; see ex o explana ion
Campaign July Augus STT EMI/EMC Decembe Janua y
Tes s CWL au o ocus RMI au o ocus Au o ocus Ins um. on Code upda e Da k collec s
RMI au o ocus IRF LIBS RMI Alignmen RMI HDR
Alignmen The mal con ol LIBS +MIC IRF IRS side B
LIBS (3 m) TRL sweep (5 m) MIC alone LIBS +MIC
(pulsed mode)
Raman (3 m) Passi e VIS Raman
RMI (came a
model 2, 3, 5 m)
98-poin IR scan VISIR
RMI
2D spec a
Addi ional es ing (su i al only) included o e -le el ib a ion and RSM- elease shock
lase beam posi ions ela i e o he RMI FOV. Obse a ions we e made o suppo modeling
o he poin ing o Supe Cam ela i e o Mas cams and Na cams (pa allax and o se ). This
came a model is c i ical o being able o command he co ec posi ion o hi a a ge wi h
Supe Cam based on he loca ion o he a ge in he Na cam o Mas cam images, all om
a gi en o e posi ion. RMI images aken in suppo o he came a model and o he es ing
e ealed a small numbe o unexpec ed pa icles nea he ocal plane o he RMI image .
In es iga ion indica ed ha hese pa icles educe he ligh only sligh ly in small a eas o he
image, and hey can be emo ed om he images as pa o he la - ield co ec ion.
The sys em he mal es (STT; Table 10) co e ed ∼2 weeks in Oc obe 2019 wi h he
o e in a la ge he mal/ acuum chambe a a ange o empe a u es o simula e Ma s. Mos
o he es ing was done a ∼10 To o N2 o simula e Ma s he mal condi ions, wi h some
es ing o simula e he c uise condi ions ( acuum). Al hough he RTG powe supply was
no ins alled on he o e , he RAMP was hea ed as i i we e p esen , o p o ide ealis-
ic empe a u es o he ins umen s and o he componen s in he o e . Two se s o pla es
wi h geological a ge s we e moun ed in he chambe . One was on he g ound nea he o e
(2.6 m om Supe Cam) and ano he pla e was 4.6 m om he ins umen , hanging on he
wall. The SCCTs on he back o he o e cons i u ed a hi d se o a ge s. In addi ion, Su-
pe Cam ook one RMI image o he a ge s ins alled by he SHERLOC eam. STT p o ided
he i s ealis ic es en i onmen o he IR spec ome e , and also o he LIBS +mic o-
phone combina ion. All o he o he main obse a ion modes we e ca ied ou in STT: LIBS,
Raman/TRL, VISIR, and RMI. Las ly, all end- o-end he mal con ol loops and associa ed
ha dwa e we e con i med in STT.
The o e -le el elec omagne ic in e e ence and compa ibili y (EMI/EMC) es s com-
plemen ed o he EMI/EMC es ing done ea lie a he ins umen le el. The es alida ed
ha Supe Cam does no cause in e e ence wi h he ul a-high equency (UHF) an enna,
and so he ins umen can be ope a ed a he same ime as UHF uplink and downlink.
Decembe 2019 es ing o Supe Cam consis ed o a mino so wa e upda e and epea
obse a ions o IRF and alignmen . The la e wo we e impo an o e i y ha o e i-
b a ion, shock, and he mal es ing had no ad e sely a ec ed he op ical sys em. The inal
IRF is gi en in Sec . 7.1.1. In Janua y 2020, ue “da ks” we e ob ained by acqui ing da k
spec a wi h he lid on. Finally, we es ed he backup IRS pho o-diode, he RMI HDR mode
and he mic ophone pulsed mode (Table 10).
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 55 o 87 4
7Resul s
This sec ion desc ibes esul s o he in eg a ed Supe Cam ins umen . Some addi ional e-
sul s in e ms o IR spec a, RMI images, and acous ic esul s can be ound in he companion
pape (Mau ice e al. 2020), as hese echniques a e comple e a he MU le el and did no
equi e an in eg a ed ins umen .
7.1 O e all Op ical P ope ies
We p esen some o he o e all op ical p ope ies he e. Radiome ic in o ma ion on he RMI
is p esen ed in he companion pape (Mau ice e al. 2020).TheIRFo heIRspec ome e
is gi en in (Roye e al. 2020).
7.1.1 Ins umen Response Func ion (IRF), BU Spec ome e s
Two di e en calib a ed lamps we e used o de e mine he IRF ac oss he spec al ange
om 245–853 nm. Fo he iole , g een, o ange, and ed spec al anges o he ansmission
and VIO e lec ion spec ome e s, a calib a ed Labsphe e in eg a ing sphe e and lamp as-
sembly was used. This lamp p o ided uni o m adiance gene ally a he 2% le el o be e
ac oss i s ape u e (∼100 mm) and o easonable angles o a ew deg ees. The same lamp
was used o calib a e ChemCam (Wiens e al. 2012). As a second sou ce, a calib a ed En-
e ge iq EQ-99 lase plasma discha ge lamp was used ac oss he en i e spec al ange. The
ad an ages o his lamp include a signi ican ly b igh e ou pu a all wa eleng hs compa ed
o he Labsphe e, and a su icien signal in he UV o allow he calcula ion o IRF. O e all, i
is a good b oadband sou ce, bu i s ou pu con ains a ew emission peaks, such as in he ed
ange, and also ∼455 nm in he VIO ange, ha a e no ully desc ibed in he manu ac u e ’s
calib a ion da a. Use o he Labsphe e allowed an independen check o hese egions. O he
han he plasma emission peaks, c oss checking o esul s ob ained wi h he Labsphe e and
wi h he EQ-99 showed close ag eemen , wi hin ∼2.5–3% ac oss all bu he UV spec al
ange, which is no co e ed by he Labsphe e.
Figu e 36 shows he p elimina y IRF o he BU spec ome e s. This measu emen was
made wi h Supe Cam moun ed in he o e , wi h he sys em a oom empe a u e (de ec o s
a 29–32 ◦C). IRF obse a ions we e made sho ly a e in eg a ion on o he o e , and again
nea he end o he o e es pe iod. IRF measu emen s we e also made on he ins umen
es se -up a LANL, o he EQM MU and he FM BU, and also o he all-EQM ins umen .
A LANL, a calib a ed ibe - ed Ocean Insigh DH-2000 lamp was also used o calib a e he
FM BU +EQMMUUV ange.
As seen in Fig. 36, mul iple obse a ions we e made using di e en numbe s o in eg a-
ion ows o he e lec ion spec ome e s (UV, VIO; see Table 3and Fig. 12). The educ ion
o in eg a ion ows om 200 o 40 and 16 allows he ins umen o a oid sa u a ion o
nea by, b igh LIBS a ge s, especially he o e calib a ion a ge s. The EQ-99 lamp was
used o calib a ion o he UV ange wi h he ins umen on he o e . Howe e , i was oo
b igh o he 200- ow in eg a ion, and so his cu e was cons uc ed om a scaling ac o
de e mined in he ea lie IRF measu emen s ha occu ed sho ly a e in eg a ion. In his
way, he 16- and 40- ow cu es we e scaled up o 200 ows as shown in Fig. 36a. Fo he
VIO and ansmission spec ome e s, he Labsphe e lamp was used o o e calib a ion, as
i a oided alignmen challenges incumben wi h he EQ-99 lamp. Fo he da a in Fig. 36c,
he ed window o he ansmission spec ome e (∼715–853 nm) used a educed se o 70
4Page 56 o 87 R.C. Wiens e al.
Fig. 36 P elimina y ins umen
op ical esponse unc ions o he
Supe Cam BU UV (a), VIO, (b),
and ansmission (c)
spec ome e s. The uni s a e
digi al numbe s (DN) pe pho on
inciden a he elescope ape u e.
Da a we e aken wi h he
ins umen moun ed in he o e .
Fo he e lec ion (UV, VIO)
spec ome e s, ead-ou o ewe
CCD ows esul s in a lowe
op ical esponse ha can be used
o a oid sa u a ion o nea by
a ge s. Fo he ansmission
spec ome e (c), a la ge ange in
esponse is needed o
accommoda e bo h b igh LIBS
signals and weak Raman signals.
A log scale is used o p esen he
gain se ings planned o LIBS
(2500) and Raman spec oscopy
(3200). The esponse cu es o
he ansmission spec ome e
clea ly show he h ee di e en
(g een, o ange, ed) op ical
windows, wi h dips in esponse a
he ansi ion egions. The exac
posi ion o he ansi ion was
selec ed o a oid any impo an
LIBS emission peaks.
ChemCam’s ins umen esponse
(Wiens e al. 2012)isshown o
compa ison. ChemCam is limi ed
o 14-bi numbe s, in con as o
Supe Cam’s 16 bi s
ows o in eg a ion. This ade accep s sligh ly lowe IRF o be e op ical esolu ion in
his egion.
The igu e also shows ha di e en in ensi ie gain DAC se ings we e used o Lab-
sphe e lamp obse a ions wi h he ansmission spec ome e (Fig. 36c). The wo cu es
a e o he gain se ings planned o be used o LIBS (2500) and Raman and TRL spec a
(3200). Obse a ions we e made a o he gain se ings, including 2100, 2300, 2600, 2900,
3200, and 3500, spanning a ange o o e h ee o de s o magni ude in op ical gain (see
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 57 o 87 4
Fig. 37 Gain p oduced ac oss he Supe Cam in ensi ie as a unc ion o he digi al- o-analog coun alue
(0–4095). The ypical DAC se ing o LIBS is 2500; Raman spec oscopy uses a DAC se ing o 3200, bu
hese can be adjus ed, depending on condi ions, e.g., dec eased o a oid sa u a ion o inc eased o highligh a
weak ea u e
also Fig. 37). O he han scaling ac o s, hese IRF cu es a di e en gain se ings appea
iden ical, indica ing ha he IRF changes in magni ude only, and no in spec al dis ibu ion.
Gi en ha he elescope, FOC, and e lec ion spec ome e s a e simila o ChemCam’s,
a compa ison wi h ha ins umen is in o de . ChemCam’s IRF was measu ed using 200
ows in all h ee spec ome e s. Tha ins umen ’s highes sensi i i y in he UV ange is 8 ×
10−4DN/pho on be ween 280 and 300 nm (Fig. 36a; Wiens e al. 2012). Supe Cam’s mos
sensi i e egion is a a li le longe wa eleng h, be ween 295 and 310 nm, achie ing 1.7×
10−3DN/pho on using 200 ows o in eg a ion (Fig. 36a). The image pixels a e e ec i ely
iden ical be ween he wo uni s, bu he digi al ep esen a ion o he pixel in ensi y is di ided
in o 65536 s eps on Supe Cam ins ead o 16384 (see Sec . 3.3.1). A he low-wa eleng h
end o he UV ange, bo h ChemCam’s and Supe Cam’s sensi i i ies d op simila ly. In he
VIO egion, ChemCam achie es jus o e 1e-3 DN/pho on in he 415–440 nm ange, while
Supe Cam’s IRF eaches a maximum o ∼3.3e-3 DN/pho on be ween 380 and 400 nm
(Fig. 36b), he esul o e o s o p o ide highe e iciency a sho e wa eleng hs. The
maximum sensi i i y in Supe Cam’s UV and VIO spec ome e s a e sho o he expec ed
ac o o ou highe in he digi ized signal. The Supe Cam demul iplexe was edesigned
o a o he Raman spec al ange a he sligh expense o all sho e wa eleng hs, so his is
no su p ising. A sligh dip a ∼425.5 nm indica es he p esence o he VIO CCD blemish,
gi ing a maximum d op in ins umen esponse o ∼4%. Because he signal is in eg a ed
o e many e ical pixels o any gi en wa eleng h, a e co ec ion o ins umen esponse,
we do no expec o see any e ec on he p ocessed da a.
While he Supe Cam ansmission spec ome e is e y di e en om he ChemCam e-
lec ion spec ome e ha co e s he same spec al ange, we can s ill compa e he IRF a
he planned LIBS HVPS se ing o 2500 wi h he IRF o ChemCam in his ange. The la e
achie es ∼1.1×10−3DN/pho on, wi h a maximum a 670 nm. Supe Cam’s IRF has wo
peaks a ound 1×10−2DN/pho on, one a ound 580 nm and one nea 660 nm (Fig. 36c). The
4Page 64 o 87 R.C. Wiens e al.
Fig. 42 To al emission om a ge JA-1 as a unc ion o dis ance
o op imum pe o mance. The ligh ins umen was only in eg a ed a JPL, and he o e ’s
sys em he mal es — he only ime he in eg a ed ins umen ’s lase was cooled—did no
acili a e a 7 m obse a ion. A se o obse a ions was made a inc easing dis ances wi h
he EQM ins umen in a he mal chambe a −10 ◦C using a lase cu en o 140 A. The
a ge , JA-1, was in ai ins ead o in a Ma s a mosphe e. Focus was pe o med manually, as
a z-s ack. The esul s in e ms o he change in o al emission (sum o all channels) is shown
up o 6 m ( he longes dis ance pe o med, due o he size o he oom) in Fig. 42.Some
o he peaks sa u a ed a dis ances close han 2.8 m using all 200 ows o he e lec ion
spec ome e s and a nominal gain o 2500 on he in ensi ie , al hough he o e all emission
was no s ongly a ec ed, since much o he emission comes om he con inuum wi h he
a ge a ambien p essu e. The o e all end wi h dis ance is close o −3,whe e is he
dis ance. The end is s eepe han he −2 end expec ed o passi e obse a ions due o he
loss o he peak lase powe densi y a longe dis ances (Mau ice e al. 2020). Peaks we e
s ill ela i ely s ong (some >3000 DN) a 6 m, and he FM MU +EQM BU demons a ed
LIBS a 7 m (Sec . 6.2), so ha achie ing he 7 m equi emen on Ma s is well assu ed.
Unpublished expe ience wi h ChemCam indica es ha he limi ing ea u e o de ec ion a
long dis ances is he abili y o he lase o ocus su icien ly well o op ically couple wi h he
a ge and p oduce a spa k. I a spa k is p oduced, he sys em is always sensi i e enough o
acqui e i s associa ed emission spec um. (On Ma s, ChemCam has succeeded in ob aining
weak spec a om some i on me eo i e a ge s as dis an as 9.2 m; Johnson e al. 2020).
Addi ionally, on Supe Cam, o he spec al egion o he ansmission spec ome e , he
gain can be boos ed, which may inc ease he abili y o de ec peaks in he g een, o ange,
and ed po ion o he spec um a long dis ances. In e ms o quan i ica ion, co ec ion o
a dis ance e ec in he calib a ion is needed o p o ide easonably accu a e obse a ions o
7m.
7.2.3 Time-Resol ed LIBS
Supe Cam o e s he abili y o use ime esolu ion o LIBS spec a in he ansmission
spec ome e , hough no in he sho e wa eleng h anges. The la ges ime di e ence wi hin
LIBS spec a is be ween a omic and molecula spec a. Ou s udies indica ed ha essen ially
all a omic emission is inished wi hin 1 µs, while molecula emission con inues o e a longe

The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 65 o 87 4
Fig. 43 Example o he abili y o Supe Cam o use ime esolu ion o s udy LIBS p ocesses. Shown a e
ends in he in ensi y o he CaF molecula peak measu ed a 603 nm as a unc ion o he exposu e du a ion
up o 10 µs (a), and wi h up o 40 µs delay, on a log scale (b). In (a), all exposu es s a a a delay o 800 ns.
In (b), all exposu es ha e a du a ion o 10 µs
pe iod o ime. To check he iming o he CaF molecula peak (Fig. 41c), LIBS was obse ed
om a basal a ge doped wi h nea ly 50% CaF2a a dis ance o ∼2.3m; his es used he
EQM ins umen . Two se s o da a we e aken. In he i s se , he in ensi ie ga e was se o
10 µs (long) and he delay was inc eased o see how much o he CaF peak emained a e
di e en imes. The o he da a se used a cons an delay o 800 ns ( oo long o cap u e all o
he a omic emission) and di e en exposu e du a ions.
The esul s a e shown in Fig. 43. The molecula emission peak pe sis s o >40 µs (las
da a poin in Fig. 43b) bu mos o he emission is gone much ea lie ; 6% o he emission
emains a 5 µs and only 1% a 10 µs ( ou h and i h da a poin s in Fig. 43b). Based
on hese da a, i was decided o he mission o use a LIBS exposu e du a ion o 10 µs
wi h he ansmission spec ome e . A ela i ely sho LIBS exposu e wi h he ansmission
spec ome e dec eases he con ibu ion om ambien ligh . On ChemCam, sub ac ion o
ambien e lec ed ligh is a signi ican ac o in his spec al ange, and some sola ea u es,
especially he Balme line sub ac ion o H a 656 nm, can cause p oblems i he su ace
albedo changes due o he exca a ion o a (da k) hole in soil (e.g., Sch öde e al. 2015).
Rela i e o ChemCam, which has an e ec i e ∼8 ms exposu e, he ambien ligh ob ained
wi h he ansmission spec ome e is educed by nea ly h ee o de s o magni ude, so he no-
lase (“da k”) spec a will ha e much lowe signal o be sub ac ed om he ac i e spec um.
(The LIBS exposu e du a ions o he e lec ion spec ome e s a e 1 ime ick =34.133 µs,
bu he pixels a e s ill ac i e while being ead, esul ing in he same e ec i e du a ion as
ChemCam. Howe e , ambien ligh in ensi y in he VIO and UV anges a e much lowe
han in he g een, o ange, and ed anges co e ed by he ansmission spec ome e .)
Se e al o he ime- esolu ion expe imen s we e ca ied ou on LIBS spec a wi h he
ansmission spec ome e . One ques ion was whe he i is easible o emo e he con inuum
by s a ing he exposu e sligh ly la e . Mos o he con inuum is gone wi hin ∼100 ns.
Howe e , some o he mos apidly-quenching peaks s a o disappea wi hin ha ime scale,
including sul u peaks. Some spec a could be ob ained showing S peaks and almos no
con inuum. Howe e , Supe Cam has some unce ain ies on he o de o 20 ns due o he
wo FPGAs ha espec i ely ope a e he lase and he in ensi ie (Fig. 29 and Sec . 3.2).
Because o his, o Supe Cam LIBS da a i is bes no o y o emo e he con inuum wi h
he in ensi ie ga e. O he expe imen s may be o in e es , such as empo ally sepa a ing he
H emission line a 656 nm om an o e lapping C line (Sch öde e al. 2018) o s udying
molecula emission lines o CaCl (Vog e al. 2018) o CaO. Finally, i should be possible o
collec a omic emissions la gely o he exclusion o molecula emissions. This is po en ially
4Page 66 o 87 R.C. Wiens e al.
Fig. 44 Measu ed pi dep hs as a unc ion o he numbe o lase sho s o basal (a) and dolomi e (b)made
wi h he EQM lase a a dis ance o 2.86 m in a simula ed Ma s a mosphe e. E o ba s show he s anda d
de ia ions o h ee pi s a each numbe o sho s. Inse s show he samples. Rows o pi s can be seen in he
dolomi e sample (b)
use ul o obse ing one o he s onge P emission lines which is no mally hidden wi hin
he CaF s uc u e o mine als like apa i e.
7.2.4 LIBS Dep h P o iles
LIBS has a special ad an age o emo ely p o iling in o ocks and soils, so ha su ace
wea he ing o o he su ace coa ings can be de ec ed and cha ac e ized (e.g., Lanza e al.
2012,2015,2016). Se e al a ge s we e p o iled using a iable numbe s o sho s pe loca ion
wi h he EQM MU and FM BU; addi ionally, one dep h p o ile was pe o med in ATLO wi h
he ligh con igu a ion. The esul s may di e be ween he wo because he EQM elescope’s
p ima y mi o had a nickel laye be ween he s uc u e and he e lec i e aluminum coa ing
ha was la e disco e ed o cause de o ma ion o he mi o a cold empe a u es (Mau ice
e al. 2020). The EQM lase dep h p o ile was pe o med wi h he ins umen a −5 o
−10 ◦C, so he ocus may ha e been somewha poo e han on he FM, which does no ha e
he Ni laye . No e ha ChemCam has a Ni laye (Mau ice e al. 2012) because he issue was
no disco e ed un il la e . In any case, he EQM dep h p o ile should ep esen a wo s case
in e ms o he Supe Cam ligh dep h-p o ile capabili ies. We p esen he EQM esul s i s .
Two a ge s we e used: a basal and a sandy dolomi e (Fig. 44, inse s). The basal sample
is om he ∼1.5 Ma Black Poin La a Flow, which is pa o he San F ancisco olcanic
ield ∼40 miles no h o Flags a , AZ (Ul ich and Bailey 1987); his basal was also used
o he LIBS dep h p o ile expe imen s desc ibed in Lanza e al. (2012,2015). The dolomi e
sample is om he Moenkopi o ma ion, Wupa ki membe , which composes he sedimen a y
bed ock ha p eda es he la a low in he same a ea (S ewa e al. 1972). Bo h samples we e
p epa ed as sawed bille s ha we e no polished bu we e ela i ely la . A se ies o LIBS
sho s we e pe o med on each sample o p oduce pi s wi h 5, 10, 20, 30, 50, 100, 150, 200,
300, 400, and 500 sho s pe loca ion (3 pi s each, e.g., Fig. 44b, inse ). These we e done a
a s ando dis ance o 2.85 m unde a simula ed Ma s a mosphe e. Pi dep hs and olumes
we e hen assessed using a Keyence VK-X100 3D Lase Scanning Con ocal Mic oscope
wi h a VK-X150 con olle . Scanning heigh s we e manually de e mined by ocus a he
su ace o he sample h ough o he bo om o he c a e . The Keyence Mul iFile Analyze
was used o de e mine c a e olume.
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 67 o 87 4
Fig. 45 P o ile o a lase pi made wi h 500 sho s in dolomi e, and a side iew o a slice nea he su ace.
(a) P o ile o he lase pi . (b) Backsca e ed elec on image o he same dolomi e sample in hin sec ion; he
na u al ex e io o he ock is seen nea he op o he image, wi h he ou line o he LIBS pi om (a) in black
( he pi was no made a his loca ion, bu he scale o ea u es can be compa ed). The wea he ed su ace
o he e es ial dolomi e is a di e en composi ion han he in e io ock, and LIBS dep h p o iles sample
and di e en ia e be ween ex e io and in e io composi ions. The LIBS pi is app oxima ely cone shaped bu
shows some a ia ion due o he p ope ies o he beam
Bo h he pi dep h and he olume o he pi inc ease apidly a i s , wi h he a e o abla-
ion dec easing wi h inc easing numbe s o lase pulses (Fig. 44). This is expec ed because
he su aces in he pi a e no longe no mal o he lase beam as a gene ally cone-shaped
pi de elops (Fig. 45). The dolomi e abla ed mo e eadily han he basal , p oducing deepe
pi s. Pi s made wi h 30 sho s ( ypically used o ChemCam a ge s; Wiens and Mau ice
2015; Mau ice e al. 2016) anged om 84–107 µm deep in basal and 230–296 µm deep in
dolomi e. Pi s made wi h 500 sho s a e aged 280 µm deep in basal , while 500-sho pi s in
dolomi e we e nea ly 1 mm deep (920–980 µm), wi h one ou lie poin a 400 sho s ha ga e
a g ea e dep h o 1100 µm. These esul s con i m ha he ma e ial p ope ies o a a ge , in
pa icula i s ha dness, play a la ge ole in he o al abla ion dep h ha may be achie ed wi h
LIBS dep h p o ile analyses. Con e sely, he ha dness o a a ge may be in e ed by he el-
a i e pi size p oduced by LIBS analysis. The ChemCam FM achie ed compa able esul s
(Wiens e al. 2012). On he same samples as Supe Cam epo s he e, he ChemCam labo a-
o y uni achie ed compa able esul s unde simila analysis condi ions, p oducing pi s wi h
dep hs ∼120 µm wi h 300 sho s and ∼350 µm a 900 sho s in he basal , and ∼370 µm
wi h 300 sho s and ∼1000 µm a 900 sho s in he dolomi e (see images in A idson e al.
2014).
Dep h p o ile esul s on Ma s may be a ec ed by he s abili y o he o e and he RSM.
Fo ChemCam dep h p o iles ha used >150 spec a, a pause o o e a minu e was needed
a e 150 sho s o ans e he spec al da a o he o e . I was no ed ha in a leas some
cases in which dep h p o iles o >150 sho s we e made, he i s ew spec a a e he pause
showed s onge emission, hough o be due o a sligh mo emen o he cen e o he beam
on he a ge , possibly only by a ew mic ons. Images o he esul an pi s also sugges ed
ha he e had been a e y sligh mo emen du ing he pause, causing a sligh o se in he
analysis loca ion be ween 150 sho bu s s. Supe Cam will be able o comple e 500 sho s in a
single bu s , which will allow g ea e dep h p o iling capabili ies han ChemCam ega dless
o mechanical s abili y.
One dep h p o ile was es ed while he FM Supe Cam ins umen was moun ed on he
o e , du ing he sys em he mal es , wi h a p essu e o 8.1 To o N2gas. The a ge
(“ClinQzO h”) consis s o 49% clinop iloli e, 32% o hoclase, and 19% qua z, and was a
4Page 68 o 87 R.C. Wiens e al.
a dis ance o 2.6 m. The lase was a −15 ◦C and used 140 A. A bu s o 150 pulses was used,
bu i was a he same loca ion as an obse a ion using 30 pulses ha occu ed immedia ely
be o e, so he pi was made om a o al o 180 pulses. The maximum c a e dep h, measu ed
wi h he Keyence mic oscope, is 194 µm, wi h a c a e olume o 1.67 ×106µm3.The
bo om o he pi slan s sligh ly o one side; i is no clea i ha is due o RSM mo emen
du ing he dep h p o ile o o he p ope ies o he a ge o he lase beam. No e ha he
EQM-lase dep h p o ile shown in Fig. 45 appea s simila .
7.2.5 Quan i a i e Elemen al Abundances om LIBS
Calib a ion o Supe Cam LIBS will be p esen ed in ano he pape . He e a b ie o e iew
is p o ided. P e-p ocessing consis s o sub ac ing a backg ound (non-lase ) spec um, de-
noising, wa eleng h calib a ion, con inuum emo al, s i ching o he o e lapping spec al
egions om he ansmission spec ome e , and dis ance co ec ion o p o ide adiance
in pho ons/second/cm2/s /µm as a unc ion o wa eleng h. Fu he p ocessing is done by
masking and no malizing he spec a. Quan i ica ion will be done using ei he mul i a ia e
o uni a ia e eg ession. Based on ChemCam expe ience, i is likely ha a mul i a ia e ap-
p oach will be used o majo elemen s and a uni a ia e app oach will be used o mino
and ace elemen s. Quan i a i e elemen al composi ions will be epo ed o eigh majo
elemen s and a leas ou ace elemen s. A spec al lib a y consis ing o 332 s anda ds was
de eloped wi h he FM BU +EQM MU, wi h a ge s a 2.86 m dis ance in a simula ed
Ma s a mosphe e. A e p e-p ocessing (in p og ess), he eam will op imize and compa e
mul i a ia e eg ession models o each elemen , simila o he me hods desc ibed in Wiens
e al. (2013)andClegge al.(2017). Fo he ace elemen s, we expec o ollow p ocedu es
ou lined in Pay é e al. (2017) and Cousin e al. (2020).
Main aining calib a ion o e a la ge ange o dis ances is challenging (Melikechi e al.
2014; Mezzacappa e al. 2016). Unpublished s udies wi h ChemCam da a indica e he need
o dis ance co ec ions o da a aken beyond 3.5 m wi h ha ins umen . Wo k is ongoing
o apply dis ance co ec ions o he Supe Cam da a.
The o e calib a ion a ge s play an impo an ole. Once he o e is on Ma s, c oss
compa ison be ween spec a om Ma s and equi alen spec a o eplica e a ge s sho on
Ea h wi h a labo a o y clone will allow channel-wise compa isons o he esul s. An “Ea h-
o-Ma s” co ec ion will be applied o he spec a collec ed on Ea h, as i has been o
ChemCam (Clegg e al. 2017), o emo e any di e ences be ween he ins umen s ha e-
main a e co ec ing o he ins umen esponse. Addi ionally, he o e calib a ion a ge s
allow checks o he accu acy o he quan i ica ion.
7.3 Raman Spec a
7.3.1 O e all Raman Resul s
The Supe Cam Raman echnique ope a es emo ely ia a elescope using a pulsed lase and
a ga ed de ec o se o 100 ns exposu es. These ac o s dis inguish Supe Cam om ypi-
cal comme cial labo a o y ins umen s which use a mic oscope and hus ope a e a sho
dis ances (i.e. ∼mm) wi h small spo sizes (i.e. ∼µm) and use a con inuous-wa e (CW)
lase . Due o he la ge spo size o 0.74 m ad ( he FOV o he spec ome e h ough he ele-
scope), Supe Cam’s Raman mode will likely p obe se e al mine al phases simul aneously,
depending on g ain size. The Raman signal (li e ime ≈10−15 s) is e ec i ely only p oduced
du ing he exci a ion pulse, while o he signals like luminescence (li e imes om mine als
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 69 o 87 4
Fig. 46 Raman spec a o pu e mine als aken using he FM BU and EQM MU. Spec a a e a e ages om
100 lase pulses aken wi h a delay o 650 ns, a ga e wid h o 100 ns, and a gain o 3200. Lase ene gy was
measu ed a ∼8 mJ. The dis ance was 2.77 m. The alc is om an unknown locali y while he apa i e is
om Du ango, Mexico, he diopside om Tanzania, he oligoclase om On a io, Canada, he gypsum om
Du ango, Mexico, he qua z om Minas Ge ais, B azil, he calci e om Ma o G osso do Sul, B azil, he
ba i e om Cumb ia, UK, he hyd omagnesi e om I an, and he oli ine om San Ca los, AZ, USA. Spec a
o he oligoclase, oli ine, and alc ha e been mul iplied by 4 o enla ge he peaks
a e gene ally >10−9s) a e exci ed by he pulse bu con inue o decay long a e he lase
pulse is inished. The 100 ns ga e synch onized wi h he lase pulse allows e icien ejec ion
o in e e ing signals like mine al luminescence o dayligh en e ing he elescope, he eby
op imizing he collec ion o he Raman signal. Compa ed o CW ins umen s, Supe Cam
Raman bene i s om signal in ensi ica ion and il e ing in he ime-domain bu uses con-
side ably less exci a ion, and ecei es a much smalle ac ion o he signal because o he
dis ance.
Rep esen a i e Raman spec a ob ained wi h he Supe Cam EQM MU and FM BU a e
shown on Fig. 46. Va ious mine al a ge s we e es ed including silica es, phospha es, sul-
a es, ca bona es and accesso y phases (e.g., oxides). Mos a ge s we e used as aw single
c ys als bu some we e p epa ed as powde pelle s. Se e al o ganic and na u al ock a ge s
we e also es ed du ing he campaigns. Supe Cam’s spec a a e gene ally high-quali y o
phospha es, sul a es, and ca bona es wi h clea de ec ion o in e nal molecula modes and
la ice ib a ions. No ably, (i) polymo phs can be unambiguously dis inguished by s udy-
ing la ice ib a ions (e.g. calci e s. a agoni e), (ii) OH/H2O is clea ly de ec ed in ele an
phases (e.g. gypsum, hyd omagnesi e, and alc) and (iii) spec al esolu ion is su icien o
de ec di e ences in sub le Raman shi s o he main peaks due o composi ional a ia ions
(e.g. Mg s. Ca ca bona es). Some silica es (e.g. oli ine, qua z, diopside, and oligoclase)

4Page 70 o 87 R.C. Wiens e al.
yield well-de ined spec a as well, wi h simila in o ma ion, while o he s a e challenging o
analyze due o low signal (e.g. some phyllosilica es). Accesso y phases like opaque mine -
als a e no de ec able. Di icul y in de ec ion is mos ly due o he high alue o a mine al’s
op ical abso p ion coe icien ha p e en s olume ic analysis and/o due o poo Raman
e iciency (e.g., Fau e al. 2019); his is consis en wi h labo a o y ins umen s.
The spec a shown in Fig. 46 we e aken unde somewha mo e a o able condi ions han
on Ma s. Fi s , he mine als we e in many cases gem quali y o collec o ’s e sions. Min-
e als o his size and quali y will no likely o en be ound on Ma s. Howe e , o challenge
any conce ns abou hese c ys als being o non- ep esen a i e sizes, g ain-size s udies a e
epo ed below. Secondly, he ins umen was main ained a −10 ◦C. On Ma s, he de ec o s
will mo e likely be close o 0 ◦C. Addi ionally, he in ensi ie will likely be be ween 10 and
30 ◦C. I is no as sensi i e o empe a u e as he CCDs, bu we expec some inc eased elec-
on backg ound om he in ensi ie a hese empe a u es. Bo h o hese ac o s will add o
he noise loo o he ins umen .
7.3.2 G ain Size E ec s
The in ensi y o Raman spec a has long been known o be a ec ed by he g ain size o
he a ge (e.g., Sch ade e al. 1991) and he mix u es o g ains ha a e in con ac wi h
each o he . Remo e Raman spec oscopy is mo e sensi i e o his de ail han in-si u Raman
spec oscopy, whe e plen y o signal is usually ob ained (e.g., Pas e is and Beyssac 2020;
To e-Fdez e al. 2020). Se e al g ain-size s udies we e ca ied ou wi h Supe Cam p io o
deli e y; we p esen he esul s o wo such s udies he e.
Seleni e om he Gli e Mine, U ah USA, was cleaned wi h alcohol, pa ially c ushed in
a mo a and sie ed h ough di e en mesh sizes. Reagen g ade Mg sul a e was dissol ed
in wa e which was subsequen ly e apo a ed o p oduce la ge epsomi e c ys als. These we e
d ied and pa ially c ushed in a mo a and sie ed h ough di e en mesh sizes. G ains we e
p essed in o pelle s and we e obse ed a a dis ance o 2.25 m om Supe Cam’s EQM wi h
he ins umen a <−10 ◦C o maximum lase ene gy. Figu es 47 and 48 show he e ec s
o g ain size. In bo h cases, he ac ion wi h sizes >500 µm yielded lowe signal han he
ac ion be ween 250 and 500 µm. G ain sizes <45 µm showed clea ly lowe signal han he
la ge g ain sizes, and in he case o seleni e, a ac ion wi h g ain sizes <25 µm showed he
lowes signal. I hese ela ionships hold o o he mine als, i sugges s ha he bes Raman
esul s will be ob ained on sands ones wi h medium o ine g ains, al hough coa se-g ained
sands ones may pe o m well oo; sil s ones and muds ones will gi e poo e Raman signals
on Ma s (e.g., Pas e is and Beyssac 2020).
7.3.3 De ec ion Limi s and Dis ance E ec s
An impo an de ail in Supe Cam’s Raman de ec ion capabili ies is he ac ha he lase
beam co e s a signi ican ly la ge a ea han he 0.74 m ad FOV o he elescope and spec-
ome e o a ge s a close dis ance. As dis ance inc eases, losses due o he dec easing
solid angle o he elescope om he a ge become signi ican and empe a u e-dependen
co-alignmen o he lase beam and spec ome e FOV a ec s he signal (Mau ice e al.
2020). He e we discuss he numbe o Raman pho ons de ec ed om a es made a di e en
dis ances using he EQM ins umen .
The a e a which Raman pho ons a e gene a ed by a gi en sample is go e ned by (i) he
di e en ial Raman sca e ing c oss-sec ion (i.e. cm2s −1molecule−1) o a a ge molecule,
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 71 o 87 4
Fig. 47 Supe Cam Raman spec a o p essed-powde pelle s o di e en g ain sizes o seleni e gypsum
(Gli e Mine, U ah), obse ed wi h he EQM a 2.25 m dis ance in eg a ing 100 lase pulses. Wa enumbe s
a e app oxima e. Inse s show close de ail o he inge p in (a) and wa e egions (b) o he spec um
Fig. 48 Supe Cam Raman spec a o p essed-powde pelle s o di e en g ain sizes o epsomi e, obse ed
wi h he EQM a 2.25 m dis ance in eg a ing 100 lase pulses. Wa enumbe s a e app oxima e. Inse s show
close de ail o he inge p in (a) and wa e egions (b) o he spec um
4Page 72 o 87 R.C. Wiens e al.
(ii) he numbe o molecules o ha species ha a e illumina ed, and (iii) he ansmi -
ed ene gy o he exci a ion sou ce. To gauge he numbe o Raman pho ons collec ed a
Supe Cam’s en ance ape u e and a he pho oca hode o Supe Cam’s in ensi ie , a c oss-
calib a ion s udy was pe o med a he Uni e si y o Hawaii a M¯
anoa o he Raman sca e -
ing c oss-sec ion o cyclohexane and Raman e iciency o a p essed pelle sample o gypsum
powde . A p essed pelle o powde ed gypsum sample was measu ed by Supe Cam and by
UH’s Raman spec ome e . In he c oss-calib a ion s udy, UH’s emo e Raman spec ome e ,
desc ibed p e iously (Gasda e al. 2015), was adiome ically calib a ed by a Labsphe e cal-
ib a ed lamp in a manne analogous o ha u ilized o calib a e Supe Cam (Sec . 7.1.1). The
adiome ic calib a ion was e i ied by measu ing he di e en ial Raman-sca e ing c oss
sec ion o cyclohexane’s 801.3cm
−1Raman mode. Measu emen s we e made wi h bo h
1 cm and 1 mm pa h-leng h S a na cells, which ag eed well. The measu ed di e en ial Ra-
man sca e ing c oss-sec ion was ound o be 4.58 ×10−30 cm2s −1molecule−1,whichwas
wi hin 1% o he published alues (T ulson and Ma hies 1986; Acos a-Maeda e al. 2017)
o ha di e en ial Raman c oss-sec ion. A ield-s op, placed a he sample loca ion 6.11
me e s om he p ima y mi o o UH’s Raman spec ome e , ci cumsc ibed he p ojec ion
o he spec ome e ’s en ance sli a he sample loca ion so ha he lase ene gy wi hin he
FOV o he spec ome e a he sample loca ion could be measu ed di ec ly. A e e i y-
ing he accu acy o he IRF, he Raman e iciency o gypsum’s ν1emission a 1008 cm−1
was de e mined o be 2.04 ×10−8, which is in gene al ag eemen wi h published alues o
gypsum samples (S opa e al. 2004,2005).
The gypsum sample was obse ed by Supe Cam as a unc ion o dis ance using he EQM
ins umen . The Raman spec um o gypsum was co ec ed om coun s (DN) a he CCD o
pho ons a he ape u e by applica ion o he IRF. Following he IRF co ec ion, he spec a
we e co ec ed by geome ical ac o s o con e pho ons a he en ance ape u e o Raman
pho ons emi ed a he sou ce, assuming Lambe ian sca e ing in o πs . The esul s o his
expe imen a e p esen ed in Fig. 49. Th ee egions o he plo a e eadily dis inguishable.
Fi s , he lase pho ons pe pulse hi ing he a ge (and also he Raman pho ons p oduced a
he a ge ) wi hin he FOV o he ansmission spec ome e inc eases wi h dis ance as he
FOV encloses a g ea e po ion o he lase beam (Mau ice e al. 2020). Second, he numbe
o lase pho ons hi ing he a ge wi hin he FOV pla eaus as he po ion o he lase beam
wi h he highes ene gy densi y appa en ly s a s o all ou side o he spec ome e FOV.
Thi d, he lase ene gy declines as his e ec ou weighs he expanded spec ome e FOV.
The sys em is sensi i e o e y sligh (i.e. <0.35 m ad) lase -beam misalignmen s esul ing
om elescope empe a u e a ia ions (Mau ice e al. 2020), causing he lase spo o de ia e
ela i e o he spec ome e FOV, so mo e o less ligh is cap u ed by he spec ome e FOV.
As he dis ance inc eases, he coupling e iciency o he lase spo o he FOV dec eases,
depending on he amoun o mis-alignmen .
Fo highly ampli ied sys ems, as is he case wi h he ansmission spec ome e when he
in ensi ie gain is a a high se ing o Raman spec oscopy, i is use ul o es ima e he numbe
o pho ons being collec ed o unde s and he limi a ions o he echnique. This numbe is no
gi en by he IRF, which also includes he ampli ica ion o he pho ons be ween he in ensi ie
ca hode and he CCD. To de e mine sho noise and SNR, he numbe o pho ons coun ed a
he ca hode is he ele an pa ame e . To de e mine his om he numbe o pho ons in he
FOV a he a ge (Fig. 49), he numbe o pho ons en e ing he ins umen ape u e was
calcula ed om he solid angle sub ended by he ins umen , and he ansmission om
he ape u e o he in ensi ie pho oca hode was es ima ed om he ansmission o each
op ic. Depending on he wa eleng h bin o in e es , one o six pe cen o ligh en e ing he
elescope ape u e eaches he pho oca hode, wi h nea 6% e iciency a he wa enumbe
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 73 o 87 4
Fig. 49 Numbe o pho ons gene a ed a he a ge wi hin he spec ome e ield o iew (FOV) as a unc ion
o dis ance, o he ν1mode o gypsum powde using he EQM Supe Cam ins umen . The esul is based
on da a aken a hese dis ances and knowledge o he spec ome e FOV con ol ed wi h an independen
de e mina ion o he Raman e iciency o he sample. Fo a gi en sample, he numbe o Raman pho ons
p oduced wi hin he FOV depends on he alignmen o he lase beam and he spec ome e FOV, which
can deg ade beyond a ce ain dis ance. The numbe o pho ons collec ed and de ec ed by he ins umen is a
con olu ion o his cu e wi h he 1/ 2losses wi h dis ance, as p esen ed in Table 12
Table 12 P opaga ion o Raman
pho ons om sou ce o de ec o
o gypsum ν1emission
Uni s a e pho ons be ween he
FWHM o he Raman peak pe
lase pulse. Fo he sou ce
emission, only pho ons p oduced
wi hin he spec ome e FOV a e
coun ed (as in Fig. 49), esul ing
in an inc ease be ween 1.8 and
3.8 m. Numbe s a e based on
measu emen s made wi h he
EQM. The FM will ha e di e en
esul s based on alignmen a he
ime o measu emen
Dis ance
(m)
A
sou ce
A
ape u e
A
ca hode
1.8 1.8e7 4820 310
2.3 4.0e7 4080 260
2.8 7.5e7 3480 220
3.3 1.0e8 2400 150
3.8 1.3e8 1710 110
4.3 1.3e8 1080 68
4.8 1.3e8 690 43
6.0 1.1e8 250 14
used in his expe imen . Table 12 shows he numbe o pho ons wi hin he spec ome e
FOV a he sou ce, he ape u e, and he in ensi ie pho oca hode. A minimum o one o
en pho ons pe channel is necessa y o achie e de ec ion. Supe Cam’s de ec ion limi is
wa eleng h-dependen , equi ing he collec ion o 50 o 150 pho ons a he ape u e. Each
peak is a leas i e channels wide, and Raman obse a ions a e expec ed o use a leas i y
lase pulses, so s a is ics a e buil up. This calcula ion was pe o med wi h he EQM; he
alignmen and beam p o ile is belie ed o be be e wi h he FM, al hough he angula size
o he lase beam is he same. This exe cise, which used a ela i ely s ong sul a e sca e ing
spec al ea u e, illus a es he challenges o obse ing Raman signals a long dis ance.
4Page 80 o 87 R.C. Wiens e al.
Supe Cam’s image (RMI) has a small ield o iew compa ed o he o he came as on
boa d, bu i s con ibu ion is pa amoun o he success o he in es iga ion. As eques ed by
he SDT, i is essen ial o documen he con ex (mo phology, ex u e) o each chemis y and
mine alogy obse a ion, and o place hem in o he b oade pe spec i e gi en by he Na cam
(Maki e al. 2020, his jou nal) and Mas cam (Bell e al. 2020, his jou nal) image s.
One aspec o Supe Cam is la gely ye o be ealized wi hin his Ma s 2020 mission,
and ha is he syne gy ha comes om usion o he esul s om he di e en echniques.
This da a usion can be applied wi hin he Supe Cam in es iga ions, o a he le el o all
echniques on he Ma s 2020 o e (Fa ley e al. 2020). S udies o da e include one ea ly
ield s udy by a small eam ep esen ing a ious ins umen s (Ma in e al. 2020), an exe cise
on a “mys e y ock” wi hin he Supe Cam eam (Ollila e al. 2019), and se e al emo e
ope a ions and science eam aining (ROASTT) e en s o ganized wi hin he Ma s 2020
p ojec (Lawson e al., in p epa a ion). All o hese we e much mo e limi ed in scope han
is expec ed o be expe ienced wi h he Pe se e ance o e co e ing mul iple ou c ops o e a
la ge ield a ea (Sun and S ack 2020). We expec he syne gy expe ienced a bo h he human
le el, and wi h ools such as machine lea ning, will p o ide su p ising disco e ies ha would
no be possible wi hou combining he esul s o he indi idual echniques.
Acknowledgemen s Many people con ibu ed o his p ojec in addi ion o he co-au ho s, and we a e mos
g a e ul o hei suppo . This p ojec was suppo ed in he US by he NASA Ma s Explo a ion P og am, and
in F ance by CNES, CNRS, and local uni e si ies. Suppo in Spain was p o ided by he Spanish Science
Minis y. Supe Cam bene i ed om LANL labo a o y-di ec ed esea ch and de elopmen unding which
p o ided ea ly p o o ypes o he new echnologies inco po a ed in he Supe Cam BU. J. Bell, A. Yings , and
K. Benne a e hanked o e iewing his manusc ip ; edi o ial suppo by K. Willi o d is also g a e ully
acknowledged. SDG.
Au ho s’ con ibu ions All au ho s con ibu ed o ei he he p oposal o he de elopmen and es ing o
he Supe Cam ins umen as desc ibed in his pape .
Funding Was p o ided in he US by NASA’s Ma s Explo a ion P og am. Funding in F ance was p o ided
by CNES and CNRS. Funding in Spain was p o ided by he Spanish Science Minis y. Some unding o da a
analyses a LANL was p o ided by Labo a o y-Di ec ed Resea ch and De elopmen unds.
Da a a ailabili y Da a p esen ed in he Resul s sec ion o his pape a e being made a ailable o he Plan-
e a y Da a Sys em Geosciences Node unde Ma s 2020/Supe Cam.
Code a ailabili y No applicable.
Con lic s o in e es /Compe ing in e es s The au ho s decla e ha he e a e no con lic s o in e es o
compe ing in e es s.
Publishe ’s No e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps
and ins i u ional a ilia ions.
Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License,
which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long as
you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons
licence, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle a e in-
cluded in he a icle’s C ea i e Commons licence, unless indica ed o he wise in a c edi line o he ma e ial.
I ma e ial is no included in he a icle’s C ea i e Commons licence and you in ended use is no pe mi ed
by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om he
copy igh holde . To iew a copy o his licence, isi h p://c ea i ecommons.o g/licenses/by/4.0/.

The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 81 o 87 4
Appendix o Abb e ia ions, Ac onyms, and Sho De ini ions
ADC Analog- o-digi al con e e
AEGIS Au onomous explo a ion o ga he ing inc eased science: a so wa e package
o analyze Na cam images onboa d and selec Supe Cam a ge s
AFT Allowable ligh empe a u e
AOTF Acous o-op ic uning il e , pe o ms wa eleng h selec ion o he IR spec om-
e e
APG Annealed py oly ic g aphi e
Be Be yllium
BU Body uni
C&DH Con ol and da a handling (elec ical boa d)
CCD Cha ge coupled de ice
CDR Calib a ed da a eco d
CMOS Complemen a y me al oxide senso
CNES Cen e Na ional d’E udes Spa iales
CTE Coe icien o he mal expansion
CWL Con inuous-wa e lase
DAC Digi al- o-analog con e e
DC Di ec cu en
Decon Decon amina ion
DN Digi al numbe
EDR Expe imen al da a eco d
EDU Enginee ing de elopmen uni
EMI/EMC Elec omagne ic in e e ence and elec omagne ic compa ibili y
EQM Enginee ing quali ica ion model
FITS Flexible image anspo sys em ( ile o ma )
FPGA Field p og ammable ga e a ay
FOC Fibe op ic cable
FOV Field o iew
FM Fligh model
FWHM Full wid h a hal maximum
GOR G een, o ange, and ed spec al egions co e ed by he ansmission spec om-
e e
HGA High-gain an enna
HSS High-speed se ial
HVPS High- ol age powe supply
Hz He z
ICER Lossy p og essi e wa ele image comp ession
IR In a ed
IRAP Ins i u de Reche che en As ophysique e Plane ologie, he lead Supe Cam in-
s i u ion in F ance
IRS In a ed spec ome e
J Joules
JPL Je P opulsion Labo a o y
kHz Kilohe z
LANL Los Alamos Na ional Labo a o y
LIBS Lase -induced b eakdown spec oscopy
Lpmm Lines pe millime e
4Page 82 o 87 R.C. Wiens e al.
LVDS Low- ol age di e en ial signal
LVPS Low- ol age powe supply
Mbps Million bi s pe second
MHz Megahe z
M ad Milli adians
MRAM Magne o esis i e andom-access memo y
MSL Ma s Science Labo a o y
MU Mas uni
mV Milli ol
NA Nume ical ape u e
Na cam Na iga ion came a o he Ma s 2020 o e
Nd:YAG Neodymium-doped y ium-aluminum ga ne
P43 Phospho sc een used in he image in ensi ie
QE Quan um e iciency
PCB P in ed ci cui boa d
PROM P og ammable ead-only memo y
RAMP Ro e accesso y moun ing pla e, he main moun ing su ace o ins umen s
and elec onics in he in e io o he o e body
RCE Ro e compu e elemen (compu e )
RMI Remo e mic o-image
RSM Remo e-sensing mas
RTG Radio he mal iso ope gene a o
RTV Room- empe a u e ulcanizing
SAGE Venus Su ace and A mosphe e Geochemical Explo e
SCCT Supe Cam calib a ion a ge s
SDRAM Synch onous dynamic andom-access memo y
SE Spec ome e elec onics (boa d)
SHERLOC Scanning Habi able En i onmen s wi h Raman and Luminescence o O ganics
and Chemicals, a UV Raman spec ome e moun ed on he Pe se e ance o e ’s
a m
SOH S a e o heal h
S S e adian
STT Sys em he mal es , o e -le el
SUROM S a -up ead-only memo y
TEC The mal-elec ic coole
Ti Ti anium
TRL Time- esol ed luminescence
TTL T ansis o - ansis o logic
UART Uni e sal asynch onous ecei e - ansmi e
UV Ul a iole spec al ange (∼245–340 nm) co e ed by one o Supe Cam’s wo
e lec ion spec ome e s
V Vol s
VIO Viole spec al ange (∼385–465 nm) co e ed by one o Supe Cam’s wo spec-
ome e s
VIS Visible spec al ange encompassing he iole , g een, o ange, and ed spec al
anges, used o passi e e lec ance spec oscopy
VISIR Visible and in a ed passi e e lec ance spec oscopy
W Wa s
XDR Ex ac ed da a eco d
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 83 o 87 4
Re e ences
T.E. Acos a-Maeda, A.K. Mis a, J.N. Po e , D.E. Ba es, S.K. Sha ma, Appl. Spec osc. 71, 1025–1038
(2017). h ps://doi.o g/10.1177/0003702816668531
R.E. A idson, P. Bellu a, F. Cale , A.A. F aeman, J. Ga in, O. Gasnaul , J. G an , J.P. G o zinge , V. Hamil-
on, M. He e ly, K.A. Iagnemma, J. Johnson, N. Lanza, S. Le Mouelic, N. Mangold, D. Ming, M. Meh a,
R.V. Mo is, H. Newsom, N. Renno, D. Rubin, J. Schiebe , R. Sle en, A.R. Vasa ada, J. Vizcaino, R.C.
Wiens, J. Geophys. Res., Plane s 119, 1322 (2014). h ps://doi.o g/10.1002/2013JE004605
J. Ba ba and, M. Pagel, Am. Mine al. 86, 473 (2001)
J. Bell III, S.W. Squy es, K.E. He kenho , J.N. Maki, H.M. A neson, D. B own, S.A. Collins, A. Dingizian,
S.T. Ellio , E.C. Hage o , A.G. Hayes, M.J. Johnson, J.R. Johnson, J. Joseph, K. Kinch, M.T. Lem-
mon, R.V. Mo is, L. Sche , M. Schwoche , M.K. Shepa d, G.H. Smi h, J.N. Sohl-Dicks ein, R.J.
Sulli an, W.T. Sulli an, M. Wadswo h, J. Geophys. Res. 108, 8063 (2003). h ps://doi.o g/10.1029/
2003JE002070
J. Bell e al. (2020, his jou nal). h ps://doi.o g/10.1007/s11214-020-00755-x
M.Y. Be ezin, S. Achile u, Chem. Re . 110, 2641 (2010)
O. Beyssac, Elemen s 16, 117 (2020)
R. Bha ia e al. (2020, his jou nal). SPAC-D-20-00105R1
J.-P. Bib ing, A. Sou lo , M. Be hé, Y. Lange in, B. Gondo , P. D ossa , M. Bouyé, M. Combes, P. Puge ,
A. Seme y, G. Bellucci, V. Fo misano, V. Mo oz, V. Ko so , G. Bonello, S. E a d, O. Fo ni, A. Gend in,
N. Nanaud, F. Poule , G. Poulleau, T. En enaz, T. Fouche , R. Melchio i, F. Al ie i, P. Pine , B. Schmi ,
C. So in, E. Haube , H. Ho mann, R. Jaumann, U. Kelle , R. A idson, J. Mus a d, F. Fo ge , ESA SP
1240, 37 (2004)
S.A. B okus, D.K. Sille i, J. Ma k Lunde be g, P.A. DeYoung, G.F. Peaslee, D.E. Ca pen e , J. Buscaglia,
Am. Mine al. 100, 451 (2015)
E. Can ela , G. Li an e, T. Calde on, R. Melend ez, A. Millan, M.A. Al a ez, M. Ba boza-Flo es, J. Alloys
Compd. 323–324, 851 (2001)
D.M. Chew, M.G. Babechuk, N. Cogné, C. Ma k, G.J. O’Sulli an, I.A. Hen ichs, D. Doepke, C.A. McKenna,
Chem. Geol. 435, 35 (2016)
B. Chide, S. Mau ice, N. Mu doch, J. Lasue, B. Bousque , X. Jacob, A. Cousin, O. Fo ni, O. Gasnaul , P.-Y.
Meslin, J.-F. F on on, M. Bassas-Po us, A. Cadu, A. Sou nac, D. Mimoun, R.C. Wiens, Spec ochim.
Ac a B 153, 50 (2019). h ps://doi.o g/10.1016/j.sab.2019.01.008
B. Chide, S. Mau ice, D. Mimoun, N. Mu doch, R.C. Wiens, Luna Plane . Sci. LI, 1366 (2020)
P.R. Ch is ensen e al., J. Geophys. Res. 108(E12), 8064 (2003). h ps://doi.o g/10.1029/2003JE002117
S.M. Clegg, R.C. Wiens, A.K. Mis a, S.K. Sha ma, J. Lambe , S. Bende , R. Newell, K. Nowak-Lo a o, S.
Sm eka , M.D. Dya , S. Mau ice, Spec ochim. Ac a 68, 925 (2014)
S.M. Clegg, R.C. Wiens, R.B. Ande son, O. Fo ni, J. F yden ang, J. Lasue, A. Pille i, V. Pay e, T. Bouche ,
M.D. Dya , S.M. McLennan, R.V. Mo is, T.G. G a , S.A. Me zman, B.L. Ehlmann, S.C. Bende ,
R.L. Toka , I. Belgacem, H. Newsom, R.E. McIn oy, R. Ma inez, P. Gasda, O. Gasnaul , S. Mau ice,
Spec ochim. Ac a B 129, 64 (2017)
S.M. Clegg, R.B. Ande son, W. Rapin, B.L. Ehlmann, D.E. Ande son, N.H. Thomas, P. Gasda, J. F yden ang,
O. Fo ni, H. Newsom, D. Blaney, W. Goe z, S. Mau ice, R.C. Wiens, 50 h Luna and Plane a y Science
Con e ence (2019), p. 2768
E.A. Clou is, P.M. Ashe , S.A. Me zman, J. Geophys. Res. 107 (2002). h ps://doi.o g/10.1029/2000JE1467
E.A. Clou is, F.C. Haw ho ne, S.A. Me zman, K. Renn, M.A. C aig, D. Ma cino, M. Me ho , J. S ong, J.F.
Mus a d, D.L. Blaney, J.F. Bell III, F. Vilas, Ica us 184, 121 (2006)
E.A. Clou is, M.A. C aig, R.V. K uzelecky, W.R. Jam oz, A. Sco , F.C. Haw ho ne, S.A. Me zman, Ica us
195, 140 (2007). h ps://doi.o g/10.1016/j.ica us.2007.10.028
G.B. Cou eges-Lacos e, B. Ahle s, F. Rull Pe ez, Spec ochim. Ac a A 68, 1023 (2007)
A. Cousin, O. Fo ni, P.-Y. Meslin, S. Clegg, A. Ollila, O. Gasnaul , S. Mau ice, R.C. Wiens ( he ChemCam
Team), 51s Luna and Plane a y Science Con e ence (2020), p. 2160
M.C. De Sanc is, E. Ammanni o, H.Y. McSween, A. Raponi, S. Ma chi, F. Capaccioni, M.T. Cap ia, F.G.
Ca ozzo, M. Cia niello, S. Fon e, M. Fo misano, A. F ige i, M. Gia dino, A. Longoba do, G. Magni,
L.A. McFadden, E. Palomba, C.M. Pie e s, F. Tosi, F. Zambon, C.A. Raymond, C.T. Russell, Science
355, 719–722 (2017)
B.L. Ehlmann, J.F. Mus a d, S.L. Mu chie, F. Poule , J.L. Bishop, A.J. B own, W.M. Cal in, R.N. Cla k, D.J.
Des Ma ais, R.E. Milliken, L.H. Roach, T.L. Roush, G.A. Swayze, J.J. W ay, Science 322, 1828 (2008)
A. El Ali, V. Ba bin, G. Calas, B. Ce elle, K. Ramseye , J. Bou oulec, Chem. Geol. 104, 189 (1993)
C. Fab e, S. Mau ice, A. Cousin, R.C. Wiens, O. Fo ni, V. Sau e , D. Guillaume, Spec ochim. Ac a B 66,
280 (2011). h ps://doi.o g/10.1016/j.sab.2011.03.012
K. Fa ley e al. (2020, his jou nal). SPAC-D-20-00057
4Page 84 o 87 R.C. Wiens e al.
A. Fau, O. Beyssac, M. Gau hie , P.-Y. Meslin, A. Cousin, K. Benze a a, S. Be na d, J.C. Boullia d, O.
Gasnaul , O. Fo ni, R.C. Wiens, C. D oue , M. Mo and, P. Rosie , Y. Ga ino, S. Pon , S. Mau ice,
Spec ochim. Ac a 160, 105687 (2019)
O. Fo ni, M. Ga , M. Toplis, S.M. Clegg, V. Sau e , S. Mau ice, R.C. Wiens, N. Mangold, O. Gasnaul , V.
Sau e , S. Le Mouelic, P.-Y. Meslin, M. Nachon, R.E. McIn oy, A.M. Ollila, A. Cousin, J.C. B idges,
N.L. Lanza, M.D. Dya , Geophys. Res. Le . 42, 1020 (2015). h ps://doi.o g/10.1002/2014GL062742
R. F ancis, T. Es lin, G. Do an, S. Johns one, D. Gaines, V. Ve ma, M. Bu l, J. F yden ang, S. Mon ano, R.C.
Wiens, S. Scha e , O. Gasnaul , L. De lo es, D. Blaney, B. Bo ns ein, Sci. Robo . 2, eaan4582 (2017)
N. F ay, A. Ba dyn, H. Co in, K. Al wegg, D. Baklou i, C. B iois, L. Colangeli, C. Eng and, H. Fische , A.
Glasmache s, E. G ün, G. Hae endel, H. Henkel, H. Hö ne , K. Ho nung, E.K. Jessbe ge , A. Koch, H.
K üge , Y. Lange in, H. Leh o, K. Leh o, L. Le Roy, S. Me ouane, P. Modica, F.-R. O hous-Daunay,
J. Paque e, F. Raulin, J. Rynö, R. Schulz, J. Silén, S. Siljes öm, W. S eige , O. S enzel, T. S ephan, L.
Thi kell, R. Thomas, K. To ka , K. Va muza, K.-P. Wanczek, B. Zap udin, J. Kissel, M. Hilchenbach,
Na u e 538, 72–74 (2016). h ps://doi.o g/10.1038/na u e19320
M. Ga , R. Reis eld, G. Pancze , P. Blank, G. Boulon, Spec ochim. Ac a A 54, 2163 (1998)
M. Ga , G. Pancze , R. Reis eld, I. Shinno, J. Alloys Compd. 300–301, 267 (2000)
M. Ga , R. Reis eld, G. Pancze , Mode n Luminescence Spec oscopy o Mine als and Ma e ials (Sp inge ,
Be lin, 2005), p. 356
M. Ga , R. Reis eld, G. Pancze , Mode n Luminescence Spec oscopy o Mine als and Ma e ials, 2nd edn.
(Sp inge , Be lin, 2015), ISBN-13: 978-3319247632, 606 pp.
P.J. Gasda, T.E. Acos a-Mayeda, P.G. Lucey, A.K. Mis a, S.K. Sha ma, G.J. Taylo , Appl. Spec osc. 69, 173
(2015). h ps://doi.o g/10.1366/14-07483
M.P. Golombek, R.C. Ande son, J.R. Ba nes, J.F. Bell III, N.T. B idges, D.T. B i e al., J. Geophys. Res.
104, 8523 (1999)
T.A. Goudge, J.F. Mus a d, J.W. Head, C.I. Fasse , S.M. Wiseman, J. Geophys. Res. 120, 775 (2015).
h ps://doi.o g/10.1002/2014JE004782
T.A. Goudge, R.A. Milliken, J.W. Head, J.F. Mus a d, C.I. Fasse , Ea h Plane . Sci. Le . 458, 357 (2017)
K.P. Hand, A.E. Mu ay, J.B. Ga in, W.B. B inke ho , B.C. Ch is ne , K.S. Edge , B.L. Ehlmann, C.R.
Ge man, A.G. Hayes, T.M. Hoehle , S.M. Ho s , J.I. Lunine, K.H. Nealson, C. Pa anicas, B.E. Schmid ,
D.E. Smi h, A.R. Rhoden, M.J. Russell, A.S. Temple on, P.A. Willis, R.A. Yings , C.B. Phillips, M.L.
Cable, K.L. C a , A.E. Ho man, T.A. No dheim, R.P. Pappala do ( he P ojec Enginee ing Team).
Repo o he Eu opa Lande De ini ion Team, pos ed 2017 (2017)
P. Hoess, K. Flede , P oc. SPIE 4128, Image In ensi ie s and Applica ions II (2000). h ps://doi.o g/10.1117/
12.405878
B.H.N. Ho gan, R.B. Ande son, G. D oma , E.S. Amado , M.S. Rice, Ica us 339, 113526 (2020).
h ps://doi.o g/10.1016/j.ica us.2019.113526
J.R. Johnson, J.F. Bell III, S. Bende , D. Blaney, E. Clou is, L. DeFlo es, B. Helmann, O. Gasnaul , B. Gonde ,
K. Kinch, M. Lemmon, S. Le Mouelic, S. Mau ice, M. Rice, R. Wiens ( he MSL Science Team), Ica us
249, 74 (2015). h ps://doi.o g/10.1016/j.ica us.2014.02.028
J.R. Johnson, C. Achilles, J.F. Bell III, S. Bende , J. Be ge , E. Clou is, B. Ehlmann, A. F aeman, O. Gas-
naul , V.E. Hamil on, S. Le Mouelic, S. Mau ice, P. Pine , L. Thompson, D. Welling on, R.C. Wiens,
J. Geophys. Res., Plane s 122, 2655 (2017). h ps://doi.o g/10.1002/2016JE005187
J.R. Johnson, P.-Y. Meslin, J.F. Bell III, R.C. Wiens, S. Mau ice, O. Gasnaul , Luna Plane . Sci. LI, 1136
(2020)
M. Kayama, H. Nishido, T. Sekine, T. Nakaza o, A. Gucsik, K. Ninagawa, J. Geophys. Res. 117, E09004
(2012)
M. Kayama, H. Nishido, S. Toyoda, K. Komu o, A.A. Finch, M.R. Lee, K. Ninagawa, Am. Mine al. 99,65
(2014)
A. Kiely, M. Klimesh IPN P og ess Repo 42-155 (2003). h ps://ipnp .jpl.nasa.go /p og ess_ epo /42-155/
155J.pd
J.R. Lakowicz, P inciples o Fluo escence Spec oscopy, 3 d edn. (Sp inge , Be lin, 2006), p. 954
N.L. Lanza, S.M. Clegg, R.C. Wiens, R.E. McIl oy, H.E. Newsom, M.D. Deans, Appl. Op . 51, B74 (2012)
N.L. Lanza, A. Ollila, A. Cousin, R.C. Wiens, S. Clegg, N. Mangold, N. B idges, D. Coope , M. Schmid , J.
Be ge , R. A idson, N. Melikechi, H. Newsom, R. Toka , C. Ha dg o e, A. Mezzacappa, R. Jackson,
B. Cla k, O. Fo ni, S. Mau ice, M. Nachon, R.B. Ande son, J. Blank, M. Deans, D. Delapp, R. Le eillé,
R. McIn oy, R. Ma inez, P.-Y. Meslin ( he MSL eam), Ica us 249, 43 (2015)
N.L. Lanza, R.C. Wiens, R.E. A idson, B.C. Cla k, W.W. Fische , R. Gelle , J.P. G o zinge , J.A. Hu owi z,
S.M. McLennan, R.V. Mo is, M.S. Rice, J.F. Bell III, J.A. Be ge , D.L. Blaney, N.T. B idges, F. Cale
III, J.L. Campbell, S.M. Clegg, A. Cousin, K.S. Edge , C. Fab e, M.R. Fisk, O. Fo ni, J. F yden-
ang, K.R. Ha dy, C. Ha dg o e, J.R. Johnson, J. Lasue, S. Le Mouelic, M.C. Malin, N. Mangold,
J. Ma in-To es, S. Mau ice, M.J. McB ide, D.W. Ming, H.E. Newsom, S. Sch öde , L.M. Thompson,
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 85 o 87 4
A.H. T eiman, S. VanBommel, D.T. Vaniman, M.-P. Zo zano, Geophys. Res. Le . 43, 7398 (2016).
h ps://doi.o g/10.1002/2016GL069109
S. Le Mouelic, O. Gasnaul , K.E. He kenho , N.T. B idges, Y. Lange in, N. Mangold, S. Mau ice, R.C.
Wiens, P. Pine , H.E. Newsom, R.G. Deen, J.F. Bell III, J.R. Johnson, W. Rapin, B.L. Ba aclough, D.L.
Blaney, L. De lo es, J. Maki, M.C. Malin, R. Pe ez, M. Saccoccio ( he MSL Science Team), Ica us 249,
93 (2015). h ps://doi.o g/10.1016/j.ica us.2014.05.030
C. Lenz, D. Talla, K. Ruschel, R. Skoda, J. Gö ze, L. Nasdala, Mine al. Pe ol. 107, 415 (2013)
C. Lenz, L. Nasdala, D. Talla, C. Hauzenbe ge , R. Sei z, U. Koli sch, Chem. Geol. 415, 1 (2015)
J.N. Maki, D. G uel, C. McKinney, M.A. Ra ine, M. Mo ales, D. Lee, R. Willson, D. Copley-Woods,
M. Val o, T. Goodsall, J. McGui e, R.G. Sella , J.A. Scha ne , M.A. Caplinge , J.M. Shamah, A.E.
Johnson, H. Ansa i, K. Singh, T. Li win, R. Deen, A. Cul e , N. Ruo , D. Pe izzo, D. Kessle ,
C. Basse , T. Es lin, F. Alibay, A. Nelessen, S. Alge missen, Space Sci. Re . 216, 137 (2020).
h ps://doi.o g/10.1007/s11214-020-00765-9
J. Man edi e al. (2020, his jou nal). SPAC-D-20-00067R1
N. Mangold, G. D oma , V. Ansan, F. Salese, M.G. Kleinhans, M. Massé, C. Quan in, K.M. S ack, As obi-
ology 20, 994–1013 (2020). h ps://doi.o g/10.1089/as .2019.2132
J.A. Man ique, G. Lopez-Reyes, A. Cousin, F. Rull, S. Mau ice, R.C. Wiens, M.B. Madsen, J.M. Mada iaga,
O. Gasnaul , J. A amendia, G. A ana, P. Beck, S. Be na d, P. Be na di, M.H. Be n , A. Be ocal, O.
Beyssac, P. Caïs, C. Cas o, K. Cas o, S.M. Clegg, E. Clou is, G. D oma , C. D oue , B. Dubois,
D. Esc ibano, C. Fab e, A. Fe nandez, O. Fo ni, V. Ga cia-Baonza, I. Gon ijo, J. Johnson, J. Lase na,
J. Lasue, S. Madsen, E. Ma eo-Ma i, J. Medina, P.-Y. Meslin, G. Mon agnac, A. Mo al, J. Mo os,
A.M. Ollila, C. O ega, O. P ie o-Balles e os, J.M. Reess, S. Robinson, J. Rod iguez, J. Saiz, J.A.
Sanz-A anz, I. Sa d, V. Sau e , P. Sob on, M. Toplis, M. Vene anda, Space Sci. Re . 216, 138 (2020).
h ps://doi.o g/10.1007/s11214-020-00764-w
P. Ma in, B.L. Ehlmann, N.H. Thomas, R.C. Wiens, J.J.R. Hollis, L.W. Beegle, R. Bha ia, S.M. Clegg, D.L.
Blaney, Ea h Space Sci. 7, e2019EA000720 (2020). h ps://doi.o g/10.1029/2019EA000720
S. Mau ice e al. (2020, his jou nal). SPAC-D-20-00069R1
S. Mau ice, R.C. Wiens, M. Saccoccio, B. Ba aclough, O. Gasnaul , O. Fo ni, N. Mangold, D. Ba a oux,
S. Bende , G. Be ge , J. Be na din, M. Be hé, N. B idges, D. Blaney, M. Bouyé, P. Cais, B. Cla k, S.
Clegg, A. Cousin, D. C eme s, A. C os, L. DeFlo es, C. De ycke, B. Dingle , G. D oma , B. Dubois,
M. Dupieux, E. Du and, L. d’Us on, C. Fab e, B. Fau e, A. Gabo iaud, T. Gha sa, K. He kenho , E.
Kan, L. Ki kland, D. Kouach, J.-L. Lacou , Y. Lange in, J. Lasue, S. Le Mouélic, M. Lescu e, E. Lewin,
D. Limonadi, G. Manhes, P. Mauchien, C. McKay, P.-Y. Meslin, Y. Michel, E. Mille , H.E. Newsom,
G. O ne , A. Paille , L. Pa es, Y. Pa o , R. Pe ez, P. Pine , F. Poi asson, B. Que ie , B. Sallé, C. So in,
V. Sau e , H. Se an, J.J. Simmonds, J.-B. Si en, R. S iglich, N. S eibig, J.-J. Thoca en, M. Toplis, D.
Vaniman, Space Sci. Re . 170, 95 (2012). h ps://doi.o g/10.1007/s11214-012-9912-2
S. Mau ice, S. Clegg, R.C. Wiens, O. Gasnaul , W. Rapin, O. Fo ni, A. Cousin, V. Sau e , N. Mangold, L. Le
Dei , M. Nachon, R.B. Ande son, N. Lanza, C. Fab e, V. Pay é, J. Lasue, P.-Y. Meslin, J.-B. Si en, N.
Melikechi, S. Le Mouelic, J. F yden ang, A. Vasa ada, J.C. B idges, S.C. Bende , S. Sch öde , R. F an-
cis, P. Pine , H. Newsom, A. Ollila, K. He kenho , M.B. Madsen, G. D oma , P. Beck, E. Lewin, J.-L.
Lacou , Y. Lange in, B. Gonde , L. d’Us on, G. Be ge , M. Toplis, J.R. Johnson, M.D. Dya , N. B idges,
D. Vaniman, B. Ba aclough, J. Anal. A . Spec om. (2016). h ps://doi.o g/10.1039/c5ja00417a
N. Melikechi, A. Mezzacappa, A. Cousin, N.L. Lanza, S.M. Clegg, R.C. Wiens, G. Be ge , S. Mau ice, S.
Bende , O. Fo ni, D. Delapp, J. Lasue, O. Gasnaul , H. Newsom, A.M. Ollila, E. Lewin, E.A. B e es,
M.D. Dya , J. F yden ang, D. Blaney, B.C. Cla k ( he MSL Science Team), Spec ochim. Ac a B 96,
51 (2014). h ps://doi.o g/10.1016/j.sab.2014.04.004
A. Mezzacappa, N. Melikechi, A. Cousin, R.C. Wiens, J. Lasue, S.M. Clegg, R. Toka , S. Bende , N.L.
Lanza, S. Mau ice, G. Be ge , O. Fo ni, O. Gasnaul , M.D. Dya , T. Bouche , E. Lewin, C. Fab e ( he
MSL Science Team), Spec ochim. Ac a B 120, 19 (2016). h ps://doi.o g/10.1016/j.sab.2016.03.009
G. Mon agnac, G. D oma , P. Beck, F. Me cie , B. Reyna d, A. Cousin, S. Mau ice, R.C. Wiens, J. Raman
Spec osc. (2018). h ps://doi.o g/10.1002/j s.5406
S. Mu chie, R. A idson, P. Bedini, K. Beisse , J.-P. Bib ing, J. Bishop, J. Bold , P. Ca ende , T. Choo, R.T.
Clancy, E.H. Da ling on, D. Des Ma ais, R. Espi i u, D. Fo , R. G een, E. Guinness, J. Hayes, C. Hash,
K. He e nan, J. Hemmle , G. Heyle , D. Humm, J. Hu cheson, N. Izenbe g, R. Lee, J. Lees, D. Loh ,
E. Mala e , T. Ma in, J.A. McGo e n, P. McGui e, R. Mo is, J. Mus a d, S. Pelkey, E. Rhodes, M.
Robinson, T. Roush, E. Schae e , G. Seag a e, F. Seelog, P. Sil e gla e, S. Sla ney, M. Smi h, W.-J.
Shyong, K. S ohbehn, H. Taylo , P. Thompson, B. Tossman, M. Wi zbu ge , M. Wol , J. Geophys.
Res. 112, E05S03 (2007). h ps://doi.o g/10.1029/2006JE002682
N. Mu doch, B. Chide, J. Lasue, A. Cadu, A. Sou nac, M. Bassas-Po s, J. Me ison, J.J. I e sen, C. Mo e o,
C. Valsco, L. Pa es, A. Hynes, V. Godi e , R.D. Lo enz, P. Cais, P. Be na di, S. Mau ice, R.C. Wiens,
D. Mimoun, Plane . Space Sci. 165, 260 (2019). h ps://doi.o g/10.1016/j.pss.2018.09.009

4Page 86 o 87 R.C. Wiens e al.
J.F. Mus a d, M. Adle , A. Allwood, D.S. Bass, D.W. Bea y, J.F. Bell III, W.B. B incke ho , M. Ca , D.J. Des
Ma ais, B. D ake, K.S. Edge , J. Eigenb ode, L.T. Elkins-Tan on, J.A. G an , S.M. Milko ich, D. Ming,
C. Moo e, S. Mu chie, T.C. Ons o , S.W. Ru , M.A. Seph on, A. S eele, A. T eiman, Repo o he Ma s
2020 Science De ini ion Team, 154 pp., pos ed July, 2013, by he Ma s Explo a ion P og am Analysis
G oup (MEPAG) a h p://mepag.jpl.nasa.go / epo s/MEP/Ma s_2020_SDT_Repo _Final.pd (2013)
A. Ollila, R.C. Wiens, R. Pe ez, A. Nelson, M. Bodine, S. Mau ice, S.K. Sha ma, P.J. Gasda, J.-M. Reess, T.
Fouche , E. Clou is, J. Johnson, S. Bende , F. Mon messin, P. Be na di, J. F yden ang, K. McCabe, P.
Cais, Luna Plane . Sci. 48, 2339 (2017)
A.M. Ollila, N.L. Lanza, O. Beyssac, M. Gau hie , S. Clegg, A. Mis a, S. Sha ma, R.C. Wiens, S. Mau ice,
O. Gasnaul , Luna Plane . Sci. 49, 2786 (2018)
A.M. Ollila, R.C. Wiens, S. Mau ice, A. Cousin, R. Ande son, O. Beyssac, L. Bonal, P. Beck, S. Clegg, B.
Chide, L. De lo es, G. D oma , W. Fische , O. Fo ni, T. Fouche , O. Gasnaul , J. G o zinge , J. Johnson,
J. Lasue, J. Lase na, J.M. Mada iaga, M. Madsen, N. Mangold, T. Nelson, R. Newell, J. Ma inez-F ias,
S. McLennan, A. Mis a, F. Mon messin, S. Robinson, S. Sha ma, F. Rull, D. Venhaus, P. Be na di,
J.-M. Reese, A. Reyes-Newell, F. Poule , N. Lanza, I. To e-Fdez, J. A amendia, R. Pe ez, in Nin h
In e na ional Ma s Con e ence #6352, Pasadena, Cali o nia (2019)
J.D. Pas e is, O. Beyssac, Elemen s 16, 87 (2020)
V. Pay é, C. Fab e, A. Cousin, V. Sau e , R.C. Wiens, O. Fo ni, O. Gasnaul , J. Lasue, A. Ollila, W. Rapin,
P.-Y. Meslin, S. Mau ice, M. Nachon, L. Le Dei , N. Lanza, S.M. Clegg, J. Geophys. Res. 122, 650
(2017). h ps://doi.o g/10.1002/2016JE005201
F. Pos be g, N. Khawaja, B. Abel, G. Choble , C.R. Glein, M.S. Gudipa i, B.L. Hende son, H.-W. Hsu,
S. Kemp , F. Klenne , G. Mo agas-Klos e meye , B. Magee, L. Nölle, M. Pe y, R. Re iol, J.
Schmid , R. S ama, F. S olz, G. Tobie, M. T ielo , J.H. Wai e, Na u e 558, 564 (2018). h ps://doi.o g/
10.1038/s41586-018-0246-4
S. Po in, O. B issaud, P. Beck, B. Schmi , Y. Magna d, J.-J. Co eia, P. Rabou, L. Jocou, Appl. Op . 57, 8279
(2018). h ps://doi.o g/10.1364/AO.57.008279
W. Rapin, P.-Y. Meslin, S. Mau ice, D. Vaniman, M. Nachon, N. Mangold, S. Sch öde , O. Gasnaul , O.
Fo ni, R.C. Wiens, G.M. Ma inez, A. Cousin, V. Sau e , J. Lasue, D. A che ( he MSL science eam),
Ea h Plane . Sci. Le . 452, 197 (2016). h ps://doi.o g/10.1016/j.epsl.2016.07.045
W. Rapin, B. Chau i é, T. Gab iel, A. McAdams, P.-Y. Meslin, B.L. Ehlmann, B. Rondeau, E. Dehouck, N.
Mangold, C. Ha dg o e, S. Mau ice, R.C. Wiens, O. Gasnaul , A. Cousin, O. Fo ni, J. F yden ang, S.
Sch öde , J. Geophys. Res., Plane s (2018). h ps://doi.o g/10.1029/2017JE005483
W. Rapin, B.L. Ehlmann, G. D oma , J. Schiebe , N.H. Thomas, W.W. Fische , V.K. Fox, N.T. S ein, M.
Nachon, B.C. Cla k, L.C. Kah, L. Thompson, H.A. Meye , T.S.J. Gab iel, C. Ha dg o e, N. Man-
gold, F. Ri e a-He nandez, R.C. Wiens, A.R. Vasa ada, Na . Geosci. (2019). h ps://doi.o g/10.1038/
s41561-019-0458-8
C. Roye , F. Poule , J.-M. Reess, C. Pilo ge , V. Hamm, T. Fouche , S. Mau ice, O. Fo ni, R.C. Wiens, Re .
Sci. Ins um. 91, 063105 (2020). h ps://doi.o g/10.1063/1.5145390
F. Rull, S. Mau ice, I. Hu chinson, A. Mo al, C. Pe ez, C. Diaz, M. Colombo, T. Belengue , G. Lopez-Reyes,
A. Sansano, O. Fo ni, Y. Pa o , N. S eibig, S. Woodwa d, C. Howe, N. Ta cea, P. Rod iguez, L. Seoane,
A. San iago, J.A. Rod iguez-P ie o, J. Medina, P. Gallego, R. Canchal, P. San ama ia, G. Ramos, J.L.
Vago, As obiology 17, 627 (2017). h ps://doi.o g/10.1089/as .2016.1567
M.R. Sal a o e, T.A. Goudge, M.S. B amble, C.S. Edwa ds, J.L. Band ield, E.S. Amado , J.F. Mus a d, P.R.
Ch is ensen, Ica us 301, 76 (2018)
V. Sau e , M. Toplis, A. Cousin, O. Fo ni, R.C. Wiens, C. Fab e, O. Gasnaul , M. Fisk, S. Mau ice, A. Ollila,
N. Mangold, L. Le Dei , P. Beck, W. Rapin, P. Pine , J. Blank, S. Clegg, P.-Y. Meslin, M.D. Dya , H.
Newsom, N. B idges, N. Lanza, S. Le Mouelic, D. Vaniman, Na . Geosci. 8, 605 (2015). h ps://doi.o g/
10.1038/NGEO2474
V. Sau e , M.J. Toplis, P. Beck, N. Mangold, R.C. Wiens, P. Pine , A. Cousin, S. Mau ice, L. LeDei , R.
Hewins, O. Gasnaul , C. Quan in, O. Fo ni, H. Newsom, P.-Y. Meslin, J. W ay, N. B idges, V. Pay e,
Li hos 254–255, 36 (2016)
B. Sch ade , A. Ho mann, S. Kelle , Spec ochim. Ac a 47A, 1135 (1991)
S. Sch öde , P.-Y. Meslin, O. Gasnaul , S. Mau ice, A. Cousin, O. Fo ni, N. Mangold, W. Rapin, S. Le
Mouelic, A. Ollila, M. Nachon, J. Lasue, M.D. Dya , S. Clegg, R. Jackson, R.C. Wiens, Ica us 249,43
(2015). h ps://doi.o g/10.1016/j.ica us.2014.08.029
S. Sch öde , K. Rammelkamp, D.S. Vog , S. F ohmann, A. Cousin, O. Gasnaul , H.-W. Huebe s, Luna
Plane . Sci. Con . 49, 1962 (2018)
S.K. Sha ma, Spec ochim. Ac a A 68, 1008 (2007)
J.-L. S a ck, F. Mu agh, As on. As ophys. 288, 343 (1994)
J.-L. S a ck, J. Fadili, F. Mu agh, IEEE T ans. Image P ocess. 16(2), 297 (2007)
The Supe Cam Ins umen Sui e on he NASA Ma s 2020 Ro e ... Page 87 o 87 4
J.H. S ewa , F.G. Poole, R.F. Wilson, R.A. Cadigan, U. S. Geol. Su . P o . Pap. 691 (1972). h ps://doi.o g/
10.3133/pp691
J.D. S opa , P.G. Lucey, S.K. Sha ma, A.K. Mis a, H.W. Hubble, P oc. SPIE 5163, 99 (2004). h ps://doi.o g/
10.1117/12.505090
J.D. S opa , P.G. Lucey, S.K. Sha ma, A.K. Mis a, G.J. Taylo , H.W. Hubble, Spec ochim. Ac a A 61, 2315
(2005). h ps://doi.o g/10.1016/j.saa.2005.02.030
V.Z. Sun, K.M. S ack, Geologic Map o Jeze o C a e and he Nili Planum Region, Ma s. U. S. Geological
Su ey Scien i ic In es iga ions Map 3464 (2020). h ps://doi.o g/10.3133/sim3464
N.H. Thomas, B.L. Ehlmann, W. Rapin, F. Ri e a-He nandez, N.T. S ein, J. F yden ang, T. Gab iel, P.-Y.
Meslin, S. Mau ice, R.C. Wiens, J. Geophys. Res., Plane s 125, e2019JE006289 (2020). h ps://doi.o g/
10.1029/2019JE006289
I. To e-Fdez, K. Cas o, O. Beyssac, A.M. Ollila, G. A ana, S. Be na d, P. Be na di, P. Cais, S. Clegg,
M. Egan, O. Fo ni, O. Gasnaul , I. Gon ijo, J.M. Mada iaga, J.A. Man ique, S. Mau ice, A. Mis a, G.
Mon agnac, T. Nelson, R. Newell, P. Pille i, S. Robinson, F. Rull, S.K. Sha ma, R.C. Wiens, P. Willis
( he Supe Cam Science Team), Luna Plane . Sci. 51, 1868 (2020)
M.O. T ulson, R.A. Ma hies, J. Chem. Phys. 84, 2068 (1986)
G.E. Ul ich, N.G. Bailey, Geologic map o he SP Moun ain pa o he San F ancisco olcanic ield, no h-
cen al A izona. U. S. Geological Su ey, Miscellaneous Geologic In es iga ions Map MF-1956, scale
1:50000 (1987)
D. Vaniman, M.D. Dya , R.C. Wiens, A. Ollila, N. Lanza, J. Lasue, M. Rhodes, S.M. Clegg, Space Sci. Re .
170, 229 (2012). h ps://doi.o g/10.1007/s11214-012-9886-0
D.S. Vog , K. Rammelkamp, S. Sch öde , W.H. Hübe s, Ica us 302, 470 (2018)
G.A. Waychunas, Re . Mine al. Geochem. 78, 175 (2014). h ps://doi.o g/10.2138/ mg.2014.78.5
D.C. Wells, E.W. G eisen, R.H. Ha en, As on. As ophys. Suppl. 44, 363 (1981)
R.C. Wiens, S. Mau ice ( he ChemCam and MSL Science Teams), Elemen s 11, 33 (2015)
R.C. Wiens, S.K. Sha ma, J. Thompson, A. Mis a, P.G. Lucey, Spec ochim. Ac a A 61, 2324 (2005)
R.C. Wiens, S. Mau ice, B. Ba aclough, M. Saccoccio, W.C. Ba kley, J.F. Bell III, S. Bende , J. Be na din,
D. Blaney, J. Blank, M. Bouye, N. B idges, P. Cais, R.C. Clan on, B. Cla k, S. Clegg, A. Cousin, D.
C eme s, A. C os, L. DeFlo es, D. Delapp, R. Dingle , C. D’Us on, M.D. Dya , T. Ellio , D. Enema k,
C. Fab e, M. Flo es, O. Fo ni, O. Gasnaul , T. Hale, C. Hays, K. He kenho , E. Kan, L. Ki kland, D.
Kouach, D. Landis, Y. Lange in, N. Lanza, F. LaRocca, J. Lasue, J. La ino, D. Limonadi, C. Linden-
smi h, C. Li le, N. Mangold, G. Manhes, P. Mauchien, C. McKay, E. Mille , J. Mooney, R.V. Mo is, L.
Mo ison, T. Nelson, H. Newsom, A. Ollila, M. O , L. Pa es, R. Pe ez, F. Poi asson, C. P o os , J.W.
Rei e , T. Robe s, F. Rome o, V. Sau e , S. Salaza , J.J. Simmonds, R. S iglich, S. S o ms, N. S eibig,
J.-J. Thoca en, T. T ujillo, M. Uliba i, D. Vaniman, N. Wa ne , R. Wa e bu y, R. Whi ake , J. Wi , B.
Wong-Swanson, Space Sci. Re . 170, 167 (2012). h ps://doi.o g/10.1007/S11214-012-9902-4
R.C. Wiens, S. Mau ice, J. Lasue, O. Fo ni, R.B. Ande son, S. Clegg, S. Bende , B.L. Ba aclough, L. De-
lo es, D. Blaney, R. Pe ez, N. Lanza, A. Ollila, A. Cousin, O. Gasnaul , D. Vaniman, M.D. Dya , C.
Fab e, V. Sau e , D. Delapp, H. Newsom, N. Melikechi ( he ChemCam eam), Spec ochim. Ac a B 82,
1 (2013). h ps://doi.o g/10.1016/j.sab.2013.02.003
R.C. Wiens, P.J. Gasda, A.K. Mis a, T.E. Acos a-Maeda, S.K. Sha ma, H. Quinn, K. Ganguly, R. Newell, S.
Clegg, B. Sando al, L. O , S. Mau ice, C. Vi mon ois, Luna Plane . Sci. 51, 1780 (2020)