FLASH

FLASHFLASHFLuorescence in Air from SHowers

(SLAC E-165)

Pisin ChenSLAC

Report to DOE HEP ReviewSLAC, June 15, 2005

Fluorescence from Air in Showers(FLASH)

J. Belz1, D. Bergman5, Z. Cao2, F.Y. Chang4, P. Chen3*, C.C. Chen4, C.W. Chen4, C. Field3, C. Hast3, P. Huentemeyer2, W-Y. P. Hwang4, R. Iverson3, C.C.H. Jui2,

G.-L. Lin4, E.C. Loh2, K. Martens2, J.N. Matthews2, J.S.T. Ng3, A. Odian3, K. Reil3, J.D. Smith2, S. Schnetzer5, P. Sokolsky2*, R.W. Springer2,

S.B. Thomas2, G.B. Thomson5, D. Walz3, A. Zech5

1University of Montana, Missoula, Montana2University of Utah, Salt Lake City, Utah

3Stanford Linear Accelerator Center, Stanford University, CA4Center for Cosmology and Particle Astrophysics (CosPA), Taiwan

5Rutgers University, Piscataway, New Jersey

* Collaboration Spokespersons

ULTRA HIGH ENERGY COSMIC RAYS

• Flux measurements of HiRes and AGASA differ significantly.

• Especially different conclusions on GZK cutoff (see next page).

~10 events vs~1 event(comparable total exposure)

UHECR: From Source to DetectorUHECR: From Source to Detector

CMB γ

FLASH Motivation

• Yield of number UV photons per charged particle sets fluorescence energy scale. How well is fluorescence yield known?– Y = photons per meter per charged particle.

• What is the spectrum of the light?– λ-4 Rayleigh scattering causes longer wavelengths to dominate

signal at larger distances, and therefore higher energy events.

• Does the fluorescence light follow shower development? Do shower codes correctly predict shower curve?– shower profile via fluorescence IN GAS vs. shower ionization

has never been directly measured!

Importance of Spectral DistributionImportance of Spectral Distribution

• At large distances of up to 30 km, which are typical of the highest energy events seen in a fluorescence detector, knowing the spectral distribution of the emitted light becomes essential due to the λ-4 attenuation from Rayleigh scattering.

Bunner (1967)

ObjectivesObjectives

• Spectrally resolved measurement of fluorescence yield to ~10%.

• Investigate effects of electron energy.• Study effects of atmospheric impurities.• Observe shower development of electron

pulses in air equivalent substance (Al2O3) with energy equivalents around 1018 eV.

Why Measure Fluorescence at SLAC?Why Measure Fluorescence at SLAC?

• Extensive Air Showers (EAS) are predominantly a superposition of EM sub-showers.

• FFTB beam-line can provide energy equivalent showers from ~1015 to ~1020eV.– 107-1010 electrons/pulse at 28.5 GeV.

THIN TARGET EXPERIMENT

• Pass electron beam through a thin-windowed air chamber.– Measure the yield over wide range of

pressures at and below atmospheric.– Measure the total fluorescence yield in air.– Measure emission spectrum using narrow

band filters and spectrometer.

– Effects of N2 concentration, from pure N2 to air. Also H2O effect.

T-461 (6/02), Run I (9/03) and Run III (7/04)

Fluorescence Measurements

• Use SLAC FFTB e- beam.

• Use narrow and wide band filters to measure total and spectral line strength of fluorescence light produced from e-

beam.

Thin Target Experiment

FLASH Experimental DesignThin Target Stage

• Electron beam passes (5x107-5x109 e-/pulse) through a chamber of air. 1x1 – 2x2 mm beam spot.

• HiRes PMTs are used to measure the fluorescence signal.

• 1 cm gap well defined by interior tubes.

• Interior blackened and baffled.

LED LED

Pres

PM

T

e-

The dE/dx Curve

(FLASH)

(2004)

(1996)

T-461 Run (6/02)

FLASH: Fluorescence Measurement

• No “large” discrepancy from expected total yield or spectral shape.

• Some details of spectral shape still need to be resolved.

~4.0 photons per meter per e-.Paper soon…

THICK TARGET EXPERIMENT

• Shower the e- beam to confirm fluorescence signal follow shower development.– 30 GeV beam– 1010 e–

– 3x1020 eV shower!

• Radiation Protection would only allow us to “dump” around 107 e– (3x1017 eV).

• Observe fluorescence signal at various shower depth.

Run II (6/04) and Run III (7/04)

Fluorescence, Particle Count and Shower Simulation

Curve shows • predicted

particle counts (Monte Carlo),

• measured particle counts (Ion Chamber)

• measured fluorescence signal (PMTs).

Paper soon…

FLASH: Status and Prospects• Publications in preparation:

– T-461 (June 2002 data): total yield, pressure dependence, effect of impurity.

– Thin-Target (2003 and 2004 data): precision spectrally resolved yield measurements; humidity dependence.

– Thick-Target (2004 data): fluorescence and charged particle yields as a function of shower depth and comparison with shower Monte Carlo simulations.

• Current involvement– Thin-Target and Thick-Target data analysis and publication preparation.

• Future Prospects – Detailed bench studies of calibration and systematic issues; Expected to

be completed summer 2005.

– Based on these studies and by consulting with other groups involved in such work (Nagano/TA/Auger/EUSO…) to determine what improvement should be made. Decision made by end of Aug. on future run request.

PUBLICATIONS• AN EXPERIMENT TO MEASURE THE AIR FLUORESCENCE YIELD IN

ELECTROMAGNETIC SHOWERS. By P. Huntemeyer et al. (FLASH Collaboration) SLAC-PUB-11254 (2005). 4pp. Presented at 28th International Cosmic Ray Conferences (ICRC 2003), Tsukuba,

Japan, 31 Jul - 7 Aug, 2003. Published in AIP Conf. Proc. 698: 341-344 (2004); Also in *New York 2003,

Intersections of particle and nuclear physics* 341-344 (2004).

• MEASUREMENT OF PRESSURE DEPENDENT FLUORESCENCE YIELD OF AIR: CALIBRATION FACTOR FOR UHECR DETECTORS.

By K. Reil et al. (FLASH Collaboration), submitted May 2005. SLAC-PUB-11068 (2005).

• MEASUREMENT OF AIR FLUORESCENCE AT 28.5 GEV AND IN ELECTROMAGNETIC SHOWERS.

By K. Reil et al. (FLASH Collaboration), in Proceedings of 22nd Texas Symposium on Relativistic Astrophysics, Stanford University, submitted March, 2005.

• K. Reil et al., Thin Target results…, in preparation.

• J. Belz et al., Thick Target results…, in preparation.

FLASH Summary

1. The fluorescence yield has been well measured. The foundation of the fluorescence technique remains strong.

2. The total amount of fluorescence light is proportional to shower size (as was always assumed).

3. Further studies into systematic issues with both fluorescence and ground array techniques are needed.

The discrepancy in UHECR flux rate does not appear to be caused by the fluorescence yield!