Upload
others
View
4
Download
0
Embed Size (px)
Citation preview
Let’sTalkAboutNeutrons
arXiv:1006.5255 [hep-ph]
In seeking new physics, ``discovery'' experiments find new phenomena, which violate a symmetry, or contradict a sharp
feature of the old model.
The premises of dark matter searches are unique. Experiments define ``signals'' to be
``events not accounted for by known backgrounds.'' The events themselves are very ordinary.
Events come from myriad causes.
No experiment can rule out the hypothesis of particle dark matter: it is constructed with great freedom of mass, flux, and interactions. Not detecting anything has no effect. The
null hypothesis of ``no signal'' of dark matter is poorly defined. It is tantamount to a model of the background.
thebackgroundISthehypothesis
“theundiscussedproblems”of...
CalibrationandbackgroundsviaSIGNALmodel
NEGLECTofRESONANTprocesses
...andthereturnoftheNEUTRON
Techniques for nuclear and particle physics experiments: a how-to approach By William R. Leo
What’spublishedforDMexperiments:
theElasticBilliardBallModel
Elastic2->2scatteringa“law”,unlessenoughenergytoexciteanuclearlevel
M.GoodmanandE.Witten,PRD31,1985
theDMliteraturesaysthatneutronsarelikewimps,
butwithsmallermass...
BasicMisconceptionsII:
M. Goodman and E. Witten, PRD 31,1985
...and so, theory notions ofwimps are cited for background reality....
“low energy cross sections are constant (in energy,
angle, etc) ”
``One line of defense against the muon-induced (underground) neutrons is to moderate the neutrons
below detector threshold before they reach the detector. Note than an 18 KeV neutron has a
maximum energy deposition on germanium of 1 KeV. ''
P. Barnes CDNS, 96 Dissertation, early expressed:
(...and THERMAL energy is defined as 0.024 eV)
Thehappyconvenienceofbilliardballneutrons
famousquotations,intinyfontwhile Ge and Si have similar scattering rates per nucleon for neutrons, Ge is 5–7 times more efficientthan Si for coherently scatteringWIMPs CDMS Phys.Rev.D68:082002,2003
As in the previous experiment, the propagation of these neutrons wassimulated accurately, as confirmed by comparison with veto-coincident and calibration-source neutrons CDMS Phys.Rev.Lett.102:011301,2009
Over 600,000 events were recorded using the 252Cf source during five separate periods throughout the runs, including more than 105 nuclear recoils used to characterize WIMP acceptance. CDMS Phys.Rev.Lett.102:011301,2009
Neutrons induced by radioactive processes or bycosmic-ray muons interacting near the apparatus cangenerate nuclear-recoil events that cannot be distinguishedfrom possible dark matter interactions on anevent-by-event basis. Monte Carlo simulations of thecosmic-ray muons and subsequent neutron productionand transport have been conducted with FLUKA [13],MCNPX [14] and GEANT4 [15] to estimate this cosmogenicneutron background. CDMS Phys.Rev.Lett.102:011301,2009
In order to provide nuclear-recoil events that mimicWIMP interactions, a 252Cf-fission neutron source is placed on the top face of the scintillator veto. Because the neutrons emitted by this source have such low energies (see e.g. [54]), the top layers of polyethylene insidethe shield are removed to permit the neutrons to penetrate to the cryostat. With the source and shielding in this configuration, the data set is dominated by neutrons, making the total event rate about 3 times higher than during low-background data-taking. In all other ways, the data-taking conditions are as usual. The source activity is known to ∼5% accuracy, so the absolute normalization of the spectrum is well determined Phys.Rev.D66:122003,2002.
The energy deposited in the detector by an interactingparticle is called “recoil energy” ER. If the particleinteracts with an electron or electrons (e.g. by Comptonscattering, K-capture, etc.), the event is called an electronrecoil; if the particle interacts with a nucleus (e.g.by WIMP-nucleus or neutron-nucleus elastic scattering),the event is a nuclear recoil. Most of the recoil energyis converted almost immediately into phonons, Phys.Rev.D66:122003,2002.
Two methods are used to measure this flux of unvetoed external neutrons. The first method involves comparing the rate of nuclear-recoil events in the Ge detectors with the rate in the Si detector, since Ge is more sensitive to WIMPS and Si is more sensitive to neutrons. The second method is to count the number of events consisting of nuclear recoils in two or more detectors Phys.Rev.D66:122003,2002.
For a low-mass WIMP, estimates of the neutron background have no effect Phys.Rev.D66:122003,2002.
!E ! EX2mT mX
(mT + mX)2(1" cos�).
Everyone’sfavoritebilliardballNEUTRONCALIBRATIONfollows...
DAMA/LIBRA - calibrate at accelerator 2.45 MeV n beam
COGENT - calibrate with monochromatic n beam, 24 KeV
(1! cos!) " 2mT
mn
10KeV
En# 0for10KeV,selecttheangle:
Chagani˙NaIrecoils˙idm2006
Phys.Rev.Lett.102:011301,2009, Phys.Rev.D66:122003,2002.
JCAP 0709:009,2007; NIM A 574 (2007) 385
CDMS - calibrate with 252 Cf source, MeV n peak
CALIBRATIONS!!
XENON - many papers, 1-3 MeV beam
Unfortunately,NeutronsMisbehave
NeutronCrossSections
100 1000 10000 100000.0.1
1
10
100
1000
10000
Ge 70,72,73,74
an important background regionnot reportedly calibrated
!tot (barns !)
252 Cf
1 MeV
here’s
the
calibratio
n point
x
neutron energy E (eV)nuDat
0.0001
0.01
1
100
10 4 10 710 110 -210 -5
En (eV)
σ(n, γ)(barns) Germanium
Silicon
0.001
0.1
10
1000
σ(n, γ)(barns)
10 4 10 710 110 -210 -5
En (eV)
Xenon
NeutronCrossSections
Nucleus % !thermal(tot) !thermal(n, ") (barns) !̄epithermal(n, ") (barns) !elastic (barns)
23Na ! 100 3.84 0.53 " 0.32 1.6; (2.45 MeV)28Si 92.23 2.13 0.16 " 0.084 ! 3 ? (MeV)29Si 4.67 2.70 0.12 0.08 ”30Si 3.1 2.56 0.11 0.54 ”70Ge 21.23 16.9 3.10 2.60 ! 6 ? (MeV)72Ge 27.66 9.75 0.89 0.90 ”73Ge 7.73 19.5 14.7 " 62.974Ge 35.94 7.69 0.52 0.6576Ge 7.44 8.52 0.15 1.35124Xe 0.09 150 150 3190126Xe 0.089 8.53 3.46 66.5128Xe 1.90 12.0 5.19 12.66129Xe 26.4 24.6 21.0 245130Xe 4.07 11.0 4.78 4.91131Xe 21.2 91.2 90.0 " 882 ”132Xe 26.9 4.21 0.45 5.27 ! 2 ? (2-4 MeV)134Xe 10.4 4.76 0.26 0.47 ”136Xe 8.86 8.48 0.26 0.14 ”127I ! 100 9.86 6.14 " 159.6 3.2; (2.45 MeV)
afewneutroncrosssectionsinDMmaterials
arXiv:1006.5255 [hep-ph]arrows = max( abundance! !̄epithermal )
NeutronsMisbehaveALot
0.001 0.1 10 1000 100000.0.001
0.1
10
1000
I 127
0.00001 0.01 10 10000 1.!1071
10
100
1000
10000
Xe131 red, 132 blue
100 1000 10000 100000.
10
100
1000
10000close up, 23 Na , I 127
!tot (barns !) !tot (barns !)
!tot (barns !) !tot (barns !)
DAMA signal region
100,000 barns
E (eV) E (eV)
E (eV)E (eV)
MeVcalibration point
region not
100 1000 10000 100000.0.1
1
10
100
1000
10000
Ge 70,72,73,74!tot (barns !)
x
E (eV)
data: nuDat
Processesnotreported,forreasonswecan’texplain
not just captures, but prompt gammasby the score...
...and nuclear levels don’t predict the resonances
“compound nucleus” ...is notpredictableevenin
principle
the“compoundnucleus”isacircularterm-notatheory-torenametheresonancesthatarenotpredicted
bynuclearlevels...whichismostofthem
whatever happens, it happens
Germaniumisacomplicatedsubstancevisa-visthermalneutrons
415 gammas in Budapest set. 831 gammas in ENSDF
70Ge Sigma=3.15 16 b %Abundance=21.23 4 72Ge Sigma=0.98 9 b %Abundance=27.66 3 73Ge Sigma=15.0 20 b %Abundance=7.73 1 74Ge Sigma=0.34 8 b %Abundance=35.94 2
76Ge Sigma=0.060 10 b %Abundance=7.44 2
these are not “capture gammas”these are
“prompt gammas”,dammit!
( and each isotopeis different)
(low energycutoff is
due to detectorsand internalconversion...not an end to
spectrum)
``The set is not complete, missing about 28\% of the total energy
and 74\% of the gamma rays from the capture level.'' Reedy
“The EGAF database is often incomplete because continuum gamma -rays can comprise up to 90% of the spectrum. “ RB Firestone et al,
data: iaea PGAA
what’s reported for neutron backgrounds? DAMA/LIBRA: ``In fact, environmental neutrons would induce
the reaction $^{23}Na(n; \gamma)^{24}Na$ with 0.1 barn cross-section and the reaction $^{23}Na(n; \gamma)^{24m}Na with
0.43 barn cross-section''.
CDMS: determined by simulations. Cannot in principle discriminate against neutrons
COGENT : can’t find a mention of neutron cross sections or rates
NIM A 592 (2008) 297
astro-ph /1002.4703v2
JCAP 0709:009,2007; NIM A 574 (2007) 385
Calibrationbybilliards ...is done
Neutrons induced by radioactive processes or bycosmic-ray muons interacting near the apparatus cangenerate nuclear-recoil events that cannot be distinguishedfrom possible dark matter interactions on anevent-by-event basis. Phys.Rev.Lett.102:011301,2009
ActivationonEarthsurface ...is mentioned
“0.53 barns”THERMAL!
Nomentionfoundofresonantprocesses...consequences....
calibrations ...being based on billiard balls... don’t cover energy range of experiment
quenching factors -for neutrons - are unknown.
backgrounds are unknown? why not ?
rates of activation known ? how and why?
plus, annual variations are everywhere. Evenmuonshowit!
Dama’s discussed process ofneutron capture and activation...
23Na levels...look safe! NuDat-BNL
“0.53 barns”THERMAL! (THIS
isdiscussed...)
430 KeVgap.
Safe !
.... no mention found of Iodine, withepithermal sigma = 160 barns; 24.99 minutes later, 128I decays
data Nudat-BNL
yacan’tvetothis
dama sigma region
Dama’s undiscussed
problem: IODINE
COGENT 2009lists 11.4 day 71Ge decay
and veto-able68Ge
Not all activationand conversioncan be vetoed
COGENT’sundiscussedproblem:internalconversion
data Nudat-BNL
COGENT 2010
Papp 2003 8.4 KeV x-ray beam
Ni electrodes
Ge M internal conversion
“signal”
“Prudence and past experience prompt us to continue work to exhaust less exotic possibilities. We extend an invitation to other researchers in this field to proceed with the same caution.”
COGENT signal (blue)
atop Papp x-ray beam
M-internal conversion
(red)
unexplained signal region
AnnualVariationsEverywhereicarus TM/03-01 reports 5%
annual variation of underground “neutron fluxes”
(for Soudan, see M. Goodman 98)
radon in bedrooms in England...
G. Bruno, Journal of Physics: Conference Series 203 (2010) 012091
Radon, Gran Sasso Hall A
MACRO, Astropart Phys 7, 109 (1997) measures annual variation of undergound muons
0 50 100 150 200 250 300 350
0
-0.06 -0.04 -0.02
0.020.040.06
day number
ICARUSneutron
time dependence...
predictsDAMA/LIBRA
time dependence...
curve: icarus phasepoints: Dama/Libra data
arXiv:1006.5255 [hep-ph]
The existing hypotheses about backgrounds stated by
experiments, accepted at face value and as published, make possible a variety of neutron-induced events to
be registered as dark matter signals
Summary:
If the problemsare known to a few expertsinside collaborations....
...whyaren’ttheyappearinginconferencepresentationsandjournalarticle?
(backgroundsarenotTHETHEORIST’Sburdenofproof)
PositiveSuggestions
Why not calibrate everything all beams full energy range at accelerators, reactors, sources, multiples. Stop assuming elastic recoil model for backgrounds
X-rays help calibrate sub-KeV region wherehpge detectors perform for 30 years. S/N>>1. Why not try it?
Check out the limitations of GEANT, FLUKA, MCNP MCNP-PoliMi etc Re: neutrons. Explore the unknowns of neutrons. There’s less known than you think.
PositiveSuggestionsCurrent stategies are under-determined, hinge on “if not backgound we know, must be dark !”. Unacceptable ! Discovery needs over-determined multiple-detection consistency. DAMA has developed logic of multiply over-determination fail-safe signal strategy....almost.
To control ubuitous environmental annual effects, why not duplicate detector in southern hemisphere?
ICEDAMA It’s only money.
Lead is a SOURCE of neutrons, almost the worst shield. Cd stops thermals, TRANSMITS > eV. Activation, Auger, internal conversion need to be explored and divulged.
Divulge !
acknowlegments
under which JPR takes responsibility for his own
misunderstanding,if any
discussions/emails withRichard Firestone, Alejandro Sonzogni, S. F. Mughabghab, Boris Pritychenko, G. A. Miller, Blayne Heckel, Craig Roberts, James
Vary, Graham Wilson, and J. D. Bowman,plus
Rita Bernabei, Phil Barbeau, Durdana Balakishiyeva,
it’salongroad.Let’sworkfordiscovery!
Michael Attisha, Brown University dissertation (2006)
Accidental exposure to a neutron source