dark matter searches - slac · cdms has maintained high dark matter discovery po-tential by...

Dark Matter Searches

Andrew WagnerSASS Presentation

April 8, 2009

Andrew Wagner April 2009 SASS Presentation2

Outline

• If its Dark how do we know its there?• What could it be? (What is it?)• How could it be directly detected?• What are present limits set by two experiments.• What will we know in the next few years.


This page was copied from Nick Strobel's Astronomy Notes. Go to his site at www.astronomynotes.com for the updated and corrected version.

Pin Wheel, Pin Wheel Spinning Around ...


Double Vision

Hot Gas (X-ray)

Inferred Distribution of matter from lensing


What could it be...

• Massive Compact Halo Objects (MACHOS):Brown Dwarfs, 100M Black holes ...

• Axions: Needed to solve Strong CP problem could be massive enough to explain dark matter• Lightest Particles: Under R or K parity conservation the lightest SUSY or KK particle would be stable and only weakly interacting. Collectively these are known as WIMPS (weakly interacting massive particles)


ER

e-h, e-i

Phonons

ZEPLIN, XENON,

WArP, ArDM

CRESST

ROSEBUDCDMS

EDELWEISS

DAMA/LIBRA,

KIMS, ANAIS,

XMASS, CLEAN

DRIFT,DM-TPC,

COUPP

NEWAGE,MIMAC

COUPP,

PICASSO

Direct Detection Experiments

Photons

Ge, CS2, C3F8,He3 Xe, Ar

CaWO4, BGO

ZnWO4, Al2O3 ..

NaI, CsI,Xe, Ar

Ge, SiER

e-h, e-i

Phonons

ZEPLIN, XENON,

WArP, ArDM

CRESST

ROSEBUDCDMS

EDELWEISS

DAMA/LIBRA,

KIMS, ANAIS,

XMASS, CLEAN

DRIFT,DM-TPC,

COUPP

NEWAGE,MIMAC

COUPP,

PICASSO


Photons


CaWO4, BGO

ZnWO4, Al2O3 ..

NaI, CsI,Xe, Ar

Ge, Si

ER

e-h, e-i

Phonons

ZEPLIN, XENON,

WArP, ArDM

CRESST

ROSEBUDCDMS

EDELWEISS

DAMA/LIBRA,

KIMS, ANAIS,

XMASS, CLEAN

DRIFT,DM-TPC,

COUPP

NEWAGE,MIMAC

COUPP,

PICASSO


Photons


CaWO4, BGO

ZnWO4, Al2O3 ..

NaI, CsI,Xe, Ar

Ge, Si

A tale of two technologies

Bolometers: Semiconducting crystals which detect lattice vibrations (phonons) from WIMPinteractions.

Noble Liquid Scintillators: Tanks of noble liquids which detect scintillation (photons) from WIMP interactions.


Dan Bauer

Fermilab

May 6, 2004252Cf Neutron & Gamma calibration data

Upper red dashed lineare +/- 2 ! gammaband

Lower red dashed lineare +/- 2 ! nuclearrecoil band

Phonon non-uniformitycorrected with highstatistics gammacalibrations

Bands and cutsdetermined withcalibration data aswas the analysisthreshold energy

Recoil Energy (keV)

Ioni

zatio

n Y

ield

= Io

niza

tion

Ene

rgy/

Rec

oil E

nerg

y

Bolometers

Gamma Calibration Ba(133)Neutron Calibration Cf(252)


Bolometers....brrrr


Oh’ my darling... Oh’ my darling


Two Phase Xe/Ar Time Projection Chamber

(XENON,ZEPLIN,WArP,ArDM)

It’s a Gas! ... and a Liquid!


Background Estimate and Sensitivity Reach

Geant4 model

Based on the measured activity of the ma-

terials used to build the detector, a de-

tailed simulation of the gamma and neu-

tron backgrounds has been carried out.

The expected gamma background in a 50

kg fiducial volume is estimated to be less

than 0.01 events/keV/kg/day while the

neutron background is expected to be less

than 0.9 n/year. Assuming the same back-

ground rejection power and threshold as

XENON10, the new detector should be

background free for about 2 months, corresponding to a sen-

sitivity reach of ! 2 " 10#45 cm2 for a 100 GeV WIMP.WIMP Mass [GeV/c

2]

WIM

P!

nucl

eon c

ross!

sect

ion [

cm2]

CDMSII (2005)

XENON10 (2008)

CDMSII (2008)

XENON100 (2009)

101

102

103

10!45

10!44

10!43

10!42

XENON100 projected sensitivity

Screening Facility Shield

Gator HPGe detector

A dedicated facility for screen-

ing materials used in the con-

struction of XENON100 has

been built and consists of an

ultra-low background, 100% ef-

ficient (2 kg) HPGe spectrom-

eter enclosed in a 5 cm OFRP

Cu and 20 cm Pb outer layer

shield. The LNGS screening

facility has also been used for

many of the XENON100 sam-

ples essaying. XENON100 cryostat in the shield

Data AcquisitionThe XENON100 data acquisition system is composed of 31 CAEN

V1724 14 bit 100 MHz flash ADCs to digitize the 242 PMTs signals.

The V1724 permits operation in deadtime-less mode where data is

written to a circular buffer and where multiple events can be stored

before they are read via the VME bus. The digitized signals are

“zero length encoded” by the V1724 FPGA, i.e. only the relevant

signal portions are transfered from the ADCs to the data acquisition

computer, to allow faster event transfer rates (> 60 Hz).

XENON100 DAQ

TPC & Meshes

XENON100 TPC Resistors

The inner volume of the XENON100 TPC, defined by 24

interlocking PTFE panels, has a radius of 15 cm and a drift

length of 30 cm. The uniformity of the drift field is ensured

by a set of 40 field shaping wires, mounted inside and out-

side the PTFE structure. The spectroscopic performance

of different mesh designs has been simulated and the final

detector will be equipped with hexagonal meshes for the

proportional scintillation region.

PMTsThe top array is composed of 98 tubes (QE !23%) disposed in circular patterns to enable good XYposition resolution while minimizing the number of tubes required. The bottom array is composed of

80 high QE (!33%) tubes arranged on a square grid to maximize light collection. The top (bottom)

shield arrays each have 32 tubes arranged in alternating inward and down (up) directions to allow

them to view simultaneously the top, bottom and side portions of the active LXe shield.

XENON100 top PMT array XENON100 bottom PMT array XENON100 bottom shield array

Design

The XENON100 detector

The XENON100 detector is an evolution of the first

prototype, aiming at a dramatic improvement in sen-

sitivity through a factor of 100 reduction in gamma-

background and a factor of 10 increase in fiducial mass.

The XENON100 cryostat was

designed to fit in the existing

XENON10 passive shield, to enable

a rapid deployment of the experi-

ment, paying however attention to

the requirement for low background.

To this end, XENON100 uses a

novel cryogenics design with the

pulse tube refrigerator (PTR) located

far from the detector and outside its

shielded cavity, along with signal

and high-voltage feedthroughs,

eliminating their contribution to the

background.

For effective background reduction, XENON100 also uses an

active LXe shield for a total of 105 kg viewed by 64 PMTs,

surrounding the inner target with 65 kg of Xe. The TPC is in-

strumented with 178 PMTs. The PMTs are of the same type de-

veloped for XENON10, but with lower radioactivity and higher

quantum efficiency (QE).

Laboratori Nazionali del Gran SassoLike XENON10, the new XENON100 experiment is located

underground in the Gran Sasso National Laboratory (LNGS)

in Italy. The average rock coverage of 1.4 km (3100 mwe),

provides a factor of 106 reduction of the surface muon flux.

XENON100

XENON10

The XENON10 TPC had a total active mass of 15 kg of LXe. To de-

tect the direct and proportional scintillation light, compact, metal channel

1” square PMTs (Hamamatsu R8520-06-Al). An array of 41 PMTs was

located below the cathode, fully immersed in LXe, to efficiently detect the

direct scintillation light while an array of 48 PMTs, in the gas, were used

to detect the proportional light and provided the X-Y event location in the

active volume, with a precision of a few millimeters. The drift time mea-

surement provided the Z-coordinate, with a precision of a few hundred

microns.

From October 6th 2006 until February 20th 2007, the

XENON10 detector was operated in WIMP-search mode at

the Gran Sasso underground laboratory and recorded about

1800 events in the 4.5 to 29.6 keVr energy range, a priori des-

ignated as the signal region. Out of these 1800 events, 10 were

observed in the WIMP window after all cuts. By considering

all ten observed events, with no background subtraction, and

using the “maximum gap” method [Phys. Rev. D 66, 032005

(2002)], the experiment placed in 2007 the best limit on the

spin independent WIMP-nucleon cross-section.

Phys. Rev. Lett. 100, 021303 (2008)

WIMP Mass [GeV/c2]

WIM

P!

nu

cleo

n c

ross!

sect

ion

[cm

2]

101

102

103

10!44

10!43

10!42

CDMS!II (2004 + 2005)

XENON10 (58.6 live days)

Roszkowski, Ruiz & Trotta (2007) CMSSM

Ellis et. al (2005) CMSSM

The XENON ProjectThe XENON project aims to detect Galactic WIMPs through their elastic scattering with Xe nuclei in

a 1-ton scale liquid xenon detector (XENON1T) placed deep underground, with a sensitivity to both

spin independent and spin dependent WIMP-nucleon coupling.

The detector is a Time Projection Chamber (TPC) operated in dual phase (liquid/gas), self-shielded

by an active veto of pure LXe scintillator with event-by-event discrimination provided by the simul-

taneous measurement of ionization and scintillation. 3D event localization and adequate shielding

further reduce the background. The first prototype detector (XENON10) was deployed underground,

at the Gran Sasso National Laboratory (LNGS) during 2006. With 136 kg · days exposure, this first

experiment reported in 2007 the best sensitivity to WIMP-nucleon spin independent cross-section.

The current phase of the project involves a new detector (XENON100), currently under commission-

ing at LNGS. The projected background rate, based on careful materials screening, and the expected

exposure will allow to reach a sensitivity of !2 " 10#45 cm2.

Liquid Xenon to Detect Dark Matter WIMPsThe advantages of using liquid xenon (LXe) for dark matter direct detection are numerous: its high

stopping power (Z = 54, ! = 3g/cm3) allows for a compact self-shielding geometry, its large A(!131) makes it attractive for spin independent interactions (" ! A2) and the presence of !50%

odd isotopes (12954

Xe, 13154

Xe) also makes it good for spin dependent interactions, it has no long lived

radioactive isotopes, and it is also an efficient and fast scintillator with a wavelength (!175 nm) that

enables direct readout by PMTs.

The XENON100 Dark Matter ExperimentD. Aharoni6, E. Aprile1 (spokesperson), K. Arisaka6, F. Arneodo2, A. Askin3, L. Baudis3, J. M. R. Cardoso4, B. Choi1, D. B. Cline6, L. C. C. Coelho4, S. Fattori2, A. D. Ferella3, L. M. P. Fernandes4,

K. L. Giboni1, A. Kish3, K. Lim1, J. A. M. Lopes4, Y. Mei5, K. Ni1, U. Oberlack5, G. Plante1, D. Rubin1, R. Santorelli3, J. M. F. dos Santos4, M. Schumann5, P. Shagin5, E. Tziaferi3, H. Wang6

1Department of Physics, Columbia University, New York, USA 2INFN Laboratori Nazionali del Gran Sasso, Assergi, Italy3Physics Institute, University of Zurich, Zurich, Switzerland 4Department of Physics, University of Coimbra, Coimbra, Portugal

5Department of Physics and Astronomy, Rice University, Houston, USA 6Department of Physics, University of California, Los Angeles, USA

16

the

170 kg LXe(70 kg target)

XENON100: The TPC Assembly

• x10 fiducial mass of XENON10

• x100 less back than XENON10

A Veto ... Actively


4

0 20 40 60 80 1000

0.2

0.4

0.6

Recoil energy (keV)

Ion

iza

tio

n y

ield

0

0.2

0.4

0.6

Ion

iza

tio

n y

ield

FIG. 3: Top: Ionization yield versus recoil energy in all de-tectors included in this analysis for events passing all cutsexcept the ionization yield and timing cuts. The signal re-gion between 10 and 100 keV recoil energies was defined usingneutron calibration data and is indicated by the curved lines.Bulk-electron recoils have yield near unity and are above thevertical scale limits. Bottom: Same, but after applying thetiming cut. No events are found within the signal region.

Figure 4 shows the Poisson 90% C.L. upper limit onthe spin-independent WIMP-nucleon cross section de-rived from this data set (upper solid curve), based onstandard assumptions about the galactic halo [7]. Theminimum lies at 6.6!10!44 cm2 for a 60GeV/c2 WIMP.

Our previous data from Soudan [10, 11] have been re-analyzed [17] yielding a slight improvement in sensitiv-ity over our previous publications (upper curve in Fig-ure 4). A combined limit from all Soudan data (lowersolid curve in Figure 4), using Yellin’s Optimum Intervalmethod [18] to account for observed events, gives an up-per limit of 4.6!10!44 cm2 at 90% C.L. for a WIMP massof 60GeV/c2, a factor of "3 stricter than our previouslypublished limit.

We also analyzed our data in terms of spin-dependentWIMP-nucleon interactions. Under the assumption ofspin-dependent coupling to neutrons alone and using theGe form factor given in [23], we find a minimum upperlimit of 2.7 ! 10!38 cm2 (1.8 ! 10!38 cm2) at 90% C.L.for this data set (combined Soudan data).

CDMS has maintained high dark matter discovery po-tential by limiting expected backgrounds to less thanone event in the signal region. These results from ourSoudan measurements set the best WIMP sensitivity forspin-independent WIMP-nucleon interactions over a widerange of WIMP masses. Our new limits cut significantlyinto previously unexplored regions of the central param-eter space predicted by supersymmetry.

The CDMS collaboration gratefully acknowledges Pa-trizia Meunier, Daniel Callahan, Pat Castle, Dave Hale,Susanne Kyre, Bruce Lambin and Wayne Johnson fortheir contributions. This work is supported in part

WIMP mass [GeV/c2]

Sp

in!

ind

epen

den

t cr

oss

sec

tion

[cm

2]

101

102

103

10!44

10!43

10!42

10!41

Baltz Gondolo 2004

Roszkowski et al. 2007 95% CL


CDMS II 1T+2T Ge Reanalysis

XENON10 2007

CDMS II 2008 Ge

CDMS II Ge combined

FIG. 4: Spin-independent WIMP-nucleon cross-section up-per limits (90% C.L.) versus WIMP mass. The upper curve(dash-dot) is the result of a re-analysis [17] of our previouslypublished data. The upper solid line is the limit from thiswork. The combined CDMS limit (lower solid line) has thesame minimum cross-section as XENON10 [19] (dashed) re-ports, but has more sensitivity at higher masses. Parame-ter ranges expected from supersymmetric models describedin [20] (grey) and [21] are shown (95% and 68% confidencelevels in green and blue, respectively). Data courtesy of [22].

by the National Science Foundation (Grant Nos. AST-9978911, PHY-0542066, PHY-0503729, PHY-0503629,PHY-0503641, PHY-0504224 and PHY-0705052), by theDepartment of Energy (Contracts DE-AC03-76SF00098,DE-FG02-91ER40688, DE-FG03-90ER40569, and DE-FG03-91ER40618), by the Swiss National Foundation(SNF Grant No. 20-118119), and by NSERC Canada(Grant SAPIN 341314-07).

[1] D. N. Spergel et al. (WMAP Collaboration) Astrophys.J. Suppl. 170, 377 (2007); M. Tegmark et al., (SDSSCollab.) Phys. Rev. D 69, 103501 (2004).

[2] G. Steigman and M.S. Turner, Nucl. Phys. B253, 375(1985).

[3] B.W. Lee and S. Weinberg, Phys. Rev. Lett. 39, 165(1977); S. Weinberg, Phys. Rev. Lett. 48, 1303 (1982).

[4] G. Jungman, M. Kamionkowski, and K. Griest, Phys.Rep. 267, 195 (1996); G. Bertone, D. Hooper, and J. Silk,Phys. Rep. 405, 279 (2005).

[5] M. W. Goodman and E. Witten, Phys. Rev. D 31, 3059(1985).

[6] R. J. Gaitskell, Ann. Rev. Nucl. Part. Sci. 54, 315 (2004).[7] J.D. Lewin and P.F. Smith, Astropart. Phys. 6, 87

(1996).[8] E.A. Baltz, M. Battaglia, M.E. Peskin and T. Wizansky,

4

0 20 40 60 80 1000

0.2

0.4

0.6

Recoil energy (keV)

Ion

izati

on

yie

ld

0

0.2

0.4

0.6Io

niz

ati

on

yie

ld

FIG. 3: Top: Ionization yield versus recoil energy in all de-tectors included in this analysis for events passing all cutsexcept the ionization yield and timing cuts. The signal re-gion between 10 and 100 keV recoil energies was defined usingneutron calibration data and is indicated by the curved lines.Bulk-electron recoils have yield near unity and are above thevertical scale limits. Bottom: Same, but after applying thetiming cut. No events are found within the signal region.

Figure 4 shows the Poisson 90% C.L. upper limit onthe spin-independent WIMP-nucleon cross section de-rived from this data set (upper solid curve), based onstandard assumptions about the galactic halo [7]. Theminimum lies at 6.6!10!44 cm2 for a 60GeV/c2 WIMP.

Our previous data from Soudan [10, 11] have been re-analyzed [17] yielding a slight improvement in sensitiv-ity over our previous publications (upper curve in Fig-ure 4). A combined limit from all Soudan data (lowersolid curve in Figure 4), using Yellin’s Optimum Intervalmethod [18] to account for observed events, gives an up-per limit of 4.6!10!44 cm2 at 90% C.L. for a WIMP massof 60GeV/c2, a factor of "3 stricter than our previouslypublished limit.

We also analyzed our data in terms of spin-dependentWIMP-nucleon interactions. Under the assumption ofspin-dependent coupling to neutrons alone and using theGe form factor given in [23], we find a minimum upperlimit of 2.7 ! 10!38 cm2 (1.8 ! 10!38 cm2) at 90% C.L.for this data set (combined Soudan data).

CDMS has maintained high dark matter discovery po-tential by limiting expected backgrounds to less thanone event in the signal region. These results from ourSoudan measurements set the best WIMP sensitivity forspin-independent WIMP-nucleon interactions over a widerange of WIMP masses. Our new limits cut significantlyinto previously unexplored regions of the central param-eter space predicted by supersymmetry.

The CDMS collaboration gratefully acknowledges Pa-trizia Meunier, Daniel Callahan, Pat Castle, Dave Hale,Susanne Kyre, Bruce Lambin and Wayne Johnson fortheir contributions. This work is supported in part

WIMP mass [GeV/c2]

Sp

in!

ind

epen

den

t cr

oss

sec

tion

[cm

2]

101

102

103

10!44

10!43

10!42

10!41

Baltz Gondolo 2004



CDMS II 1T+2T Ge Reanalysis

XENON10 2007

CDMS II 2008 Ge

CDMS II Ge combined

FIG. 4: Spin-independent WIMP-nucleon cross-section up-per limits (90% C.L.) versus WIMP mass. The upper curve(dash-dot) is the result of a re-analysis [17] of our previouslypublished data. The upper solid line is the limit from thiswork. The combined CDMS limit (lower solid line) has thesame minimum cross-section as XENON10 [19] (dashed) re-ports, but has more sensitivity at higher masses. Parame-ter ranges expected from supersymmetric models describedin [20] (grey) and [21] are shown (95% and 68% confidencelevels in green and blue, respectively). Data courtesy of [22].

by the National Science Foundation (Grant Nos. AST-9978911, PHY-0542066, PHY-0503729, PHY-0503629,PHY-0503641, PHY-0504224 and PHY-0705052), by theDepartment of Energy (Contracts DE-AC03-76SF00098,DE-FG02-91ER40688, DE-FG03-90ER40569, and DE-FG03-91ER40618), by the Swiss National Foundation(SNF Grant No. 20-118119), and by NSERC Canada(Grant SAPIN 341314-07).

[1] D. N. Spergel et al. (WMAP Collaboration) Astrophys.J. Suppl. 170, 377 (2007); M. Tegmark et al., (SDSSCollab.) Phys. Rev. D 69, 103501 (2004).

[2] G. Steigman and M.S. Turner, Nucl. Phys. B253, 375(1985).

[3] B.W. Lee and S. Weinberg, Phys. Rev. Lett. 39, 165(1977); S. Weinberg, Phys. Rev. Lett. 48, 1303 (1982).

[4] G. Jungman, M. Kamionkowski, and K. Griest, Phys.Rep. 267, 195 (1996); G. Bertone, D. Hooper, and J. Silk,Phys. Rep. 405, 279 (2005).

[5] M. W. Goodman and E. Witten, Phys. Rev. D 31, 3059(1985).

[6] R. J. Gaitskell, Ann. Rev. Nucl. Part. Sci. 54, 315 (2004).[7] J.D. Lewin and P.F. Smith, Astropart. Phys. 6, 87

(1996).[8] E.A. Baltz, M. Battaglia, M.E. Peskin and T. Wizansky,

5

WIMP Mass [GeV/c2]

WIM

P!

nucl

eon c

ross!

sect

ion [

cm2]

101

102

103

10!44

10!43

10!42

CDMS!II (2004 + 2005)

XENON10 (136 kg!d)

Roszkowski, Ruiz & Trotta (2007) CMSSM

Ellis et. al (2005) CMSSM

FIG. 5: Spin-independent WIMP-nucleon cross-section upperlimits (90% C.L.) versus WIMP mass. Shown curves are forthe previous best published limit (upper, blue) [26] and thecurrent work (lower, red). The shaded area is for parametersin the constrained minimal supersymmetric models [6, 27].

volume (see Fig. 3) where anomalous events due to theLXe around the bottom PMTs happen more frequently,

as discussed above. Second, the anomalous S1 hit pat-tern cut discussed earlier for the primary blind analysiswas designed to be very conservative. An independentsecondary blind analysis performed in parallel with theprimary analysis, used a more sophisticated cut to iden-tify anomalous hit patterns in S1 and rejected 3 (No.’s6, 8, 10) of these 4 candidate events as being multiplescatter events with one hit vertex outside the active vol-ume. Third, the expected nuclear recoil spectrum forboth neutrons and WIMPs falls exponentially with en-ergy, whereas the candidate events appear preferentiallyat higher energy.

This new result further excludes some parameter spacein the minimal supersymmetric models [5] and theconstrained minimal supersymmetric models (CMSSM)(e.g. [6, 27]).

This work is supported by the National Science Foun-dation under grants No. PHY-03-02646 and PHY-04-00596, and by the Department of Energy under Con-tract No. DE-FG02-91ER40688, the CAREER GrantNo. PHY-0542066, the Volkswagen Foundation (Ger-many) and the FCT Grant No. POCI/FIS/60534/2004(Portugal). We thank the Director of the Gran SassoNational Laboratory, Prof. E. Coccia, and his sta! forsupport throughout this e!ort. Special thanks go to theLaboratory engineering team, P. Aprili, D. Orlandi andE. Tatananni, and to F. Redaelli of COMASUD for theircontribution to the XENON10 installation.

[1] D. N. Spergel et al., (WMAP Collaboration), ApJ, inpress (2007); arXiv:astro-ph/0603449v2.

[2] M. Tegmark et al., (SDSS Collaboration), Phys. Rev. D74 (2006) 123507.

[3] D.N. Schramm and M. S. Turner, Rev. Mod. Phys. 70,303 (1998); R.H.Cyburt, B.D. Fields, and K.A.Olive,Phys. Lett. B 567, 227 (2003).

[4] G. Steigman and M.S. Turner, Nucl. Phys. B 253, 375(1985).

[5] A.Bottino et al., Phys. Rev. D 69, 037302 (2004).[6] J. Ellis et al., Phys.Rev.D 71, 095007 (2005).[7] H.C. Cheng, J. L. Feng, and K.T. Matchev, Phys. Rev.

Lett. 89, 211301 (2002); GServant and T. M.Tait, NewJ. Phys. 4, 99 (2002).

[8] A.Birkedal-Hanson and J. G.Wacker, Phys. Rev. D 69,065022 (2004).

[9] G. Jungman, M. Kamionkowski, and K. Griest, Phys.Rep. 267, 195 (1996); G. Bertone, D. Hooper, and J.Silk, Phys. Rep. 405, 279 (2005).

[10] M. W. Goodman and E. Witten, Phys. Rev. D 31, 3059(1985); R. J. Gaitskell, Annu. Rev. Nucl. Part. Sci. 54,315-59 (2004).

[11] A. I. Bolozdynya, Nucl. Instrum. Methods Phys. Res.,Sect. A 422, 314 (1999).

[12] M. Yamashita et al., Astropart. Phys. 20, 79 (2003).[13] E. Aprile et al., Phys. Rev. Lett. 97, 081302 (2006).

[14] T. Shutt et al., Nucl. Instrum. Methods Phys. Res., Sect.A in press (2007); arXiv:astro-ph/0608137v2.

[15] http://www.lngs.infn.it/[16] E. Aprile et al. (XENON Collaboration), “The

XENON10 Dark Matter Search Experiment”, in prepa-ration for Phys. Rev. D.

[17] M. Yamashita et al., Nucl. Instrum. Methods Phys. Res.,Sect. A 535, 692 (2004).

[18] T. Haruyama et al., Adv. Cryog. Eng. 710, 1459 (2004).[19] J. Angle et al. (XENON Collaboration), “Sensitivity of

the XENON10 WIMP search to spin-dependent interac-tions”, in preparation.

[20] E. Aprile et al., Phys. Rev. D 72, 072006 (2005).[21] V. Chepel et al., Astropart. Phys. 26, 58 (2006).[22] P. Sorensen et al., Proc. 6th Intl. Workshop for Identifi-

cation of Dark Matter, World Scientific (2006).[23] M. Yamashita et al., Proc. 6th Intl. Workshop for Iden-

tification of Dark Matter, World Scientific (2006).[24] S. Yellin, Phys. Rev. D 66, 032005 (2002).[25] J.D. Lewin and P.F. Smith, Astropart. Phys. 6, 87

(1996).[26] D. S. Akerib et al. (CDMS Collaboration), Phys. Rev.

Lett. 96, 011302 (2006).[27] L. Roszkowski, R. Ruiz de Austri and R. Trotta,

arXiv:0705.2012 [hep-ph].

4

while the S1 signal from a normal event in the active vol-ume is distributed more evenly over the PMTs (smallerS1RMS). A large fraction of events that leaked intothe WIMP-signal window are of this type of backgroundand could be removed by the cuts discussed above. Thecut acceptance !c for single-scatter nuclear recoil events,based on AmBe fast neutron calibration data, is listed inTable I.

FIG. 3: Position distribution of events in the 4.5 to 26.9 keVnuclear recoil energy window, from the 58.6 live-days ofWIMP-search data. (+) Events in the WIMP-signal regionbefore the software cuts. (!) Events remaining in the WIMP-search region after the software cuts. The solid lines indicatethe fiducial volume, corresponding to a mass of 5.4 kg.

The 3D position sensitivity of the XENON10 detec-tor gives additional background suppression with fiducialvolume cuts [22]. Due to the high stopping power of LXe,the background rate in the central part of the detectoris lower (0.6 events/keVee/kg/day) than that near theedges (3 events/keVee/kg/day). The fiducial volume isdefined to be within 15 to 65 µs (about 9.3 cm in Z, outof the total drift distance of 15 cm) drift time windowand with a radius less than 8 cm (out of 10 cm) in XY ,corresponding to a total mass of 5.4 kg (Fig. 3) [23]. Thecut in Z also removes many anomalous events due to theLXe around the bottom PMTs, where they happen morefrequently compared to the top part of the detector.

After all the cuts were finalized for the energy windowof interest, we analyzed the 58.6 live-days of WIMP-search data. From a total of about 1800 events, tenevents were observed in the WIMP search window aftercuts (Fig. 4). We expect about seven statistical leakageevents (see Table I) by assuming that the !Log10(S2/S1)distribution from electron recoils is purely Gaussian,an assumption which is statistically consistent with theavailable calibration data. However, the uncertainty ofthe estimated number of leakage events for each energy

FIG. 4: Results from 58.6 live-days of WIMP-search in the5.4 kg LXe target. The WIMP search window was definedbetween the two vertical lines (4.5 to 26.9 keV nuclear recoilequivalent energy) and blue lines (about 50% nuclear recoilacceptance).

bin in the analysis of the WIMP search data is currentlylimited by available calibration statistics. To set conser-vative limits on WIMP-nucleon cross sections, we con-sider all ten observed events, with no background sub-traction. Figure 5 shows the 90% C.L. upper limits onWIMP-nucleon cross sections as a function of WIMPmass calculated using the “maximum gap” method in[24] and using the standard assumptions for the galactichalo [25]. The current work gives a WIMP-nucleon crosssection 90% C.L. upper limit of 8.8 ! 10!44 cm2 at aWIMP mass of 100 GeV/c2, a factor of 2.3 lower thanthe previously best published limit [26]. For a WIMPmass of 30 GeV/c2, the limit is 4.5!10!44 cm2. We haveused a constant 19% nuclear recoil scintillation e"ciencyto derive the limit. The result varies by ±20%(±35%) formass 100 (30) GeV/c2 WIMPs when varying the nuclearrecoil scintillation e"ciency Leff over a range of 12%to 29%, corresponding to the lowest energy data pointsmeasured in [20] and in [21]. The measured single scatternuclear recoil spectrum from the AmBe calibration datais consistent at the 20% level with the Monte Carlo pre-dicted spectrum, both in absolute event rate and spectralshape, when Leff is taken as 19% over the energy rangeof interest.

Although we treated all 10 events as WIMP candi-dates in calculating this limit, none of the events arelikely WIMP interactions. !Log10(S2/S1) values for 5events (compared with 7 predicted) are statistically con-sistent with the electron recoil band. These are labeledas No.’s 3, 4, 5, 7, 9 in Fig. 3 and Fig. 4. As shown inTable I these leakage events are more likely to occur athigher energies. A posteriori inspection of event No. 1shows that the S1 coincidence requirement is met be-cause of a noise glitch. Event No.’s 2, 6, 8, 10 are notfavored as evidence for WIMPs for 3 main reasons. First,they are all clustered in the lower part of the fiducial

The Amazing Race


WIMP Mass [GeV/c2]

Cro

ss-s

ecti

on

[cm

2]

(no

rmal

ised

to

nu

cleo

n)

080816094101

http://dmtools.brown.edu/ Gaitskell,Mandic,Filippini

101

102

103

10-48

10-46

10-44

10-42

10-40

XENON100: Projected Sensitivity

19

Exposure Time with Zero Bkg [month]

0 5 10 15 20 25

]2

(9

0%

C.L

. u

pp

er

lim

it)

[cm

-p!"

-4510

-4410

-4310

XENON100 Sensitivity Projection for 100 GeV WIMPs )2

cm

-44

= 1

0-p!"

Ex

pe

cte

d #

of

WIM

P E

ve

nts

(

1

10

210

XENON100 Sensitivity Projection for 100 GeV WIMPs

XENON100 Projected (2009)

SuperCDMS Projected (2012)

XENON10 (2007)

CDMS (2008) 30 kg x 7 month(zero BKG)~Aug 2009

signal discovery

sensitivity

XENON1T Projected (2013)

Sensitivity reach is !SI~2x10-45 cm2 after 7 months data with zero BKG

Two years of data would give ~20 WIMP events if !SI~2x10-44 cm2 for 100 GeV

Back to the Future

dark matter searches - slac · cdms has maintained high dark matter discovery po-tential by...

Documents