Systematic studies on the rate capability of MWPC operated in Xe/CO2

D. Gonzalez-Diaz for the CBM-TRD groupMarch, 2008

The CBM experiment at FAIR

TRD stations

The TRD. specs

-> Goal of the TRD:

● Electron identification to reconstruct J/ψ and ψ'.

● Searching for such rare probes requires of high interaction rates up to 10 MHz. In the region closer to the beam axis and for the first TRD station this means rates of 100 kHz/cm2.

-> Requirements for a MWPC as active detector:

● High TR absorption, a low ageing figure and high rate capability. This imposes the use of narrow gap chambers ~ 3 mm operated in Xe/CO2 mixtures.

Experimental setup for systematic studies on rate capability

s=4 mm, h=3 mm.

baVM ln

First step. Determination of the maximum gain before glow discharge

glow discharge

Maximum gain before glow discharge

maximum gaindepends on the fraction of quencher

Mmax~2 105

s= 3 mm s= 4 mm

Rate capability measurements. An example

bM

Mr

C

shnqaMM

L

oeo ln2

exp2

2

Fit parameters: μ(CO2+ in noble gas) and μ(CO2+ in CO2)

combined by assuming the Blanc's additive law (with f being the fraction of noble gas):

)(

1

)(

1

222 COinCO

f

gasnobleinCO

f

Global fit to the Mathieson formula

baVM ln

An example from Argon

An example from Xenon

Rate capability from the Mathieson model and comparison with direct measurement

Comparison of the estimated mobilities with existing data

μCO2+ (in CO2) [mm2 V-1s-1] 108 ± 22 109(1)

μCO2+ (in CO2) [mm2 V-1s-1] 126 ± 25 109

μCO2+ (in CO2) [mm2 V-1s-1] 116 ± 23 109

μCO2+ (in Ne) [mm2 V-1s-1] 340 ± 70 694(2)

μCO2+ (in Ar) [mm2 V-1s-1] 229 ± 46 214(3)

μCO2+ (in Xe) [mm2 V-1s-1] 74 ± 15 52(3)**

(1) G. Schultz, G. Charpak, F. Sauli, Rev. Phys. Appl. 12(1977)67(2) μo from Ellis et al. Transport Properies of Gaseous Ions (II)(3) μo from Ellis et al. Transport Properies of Gaseous Ions (I)

** Not available experimentally. Represented is the mobility μo of Xe+ in Xe

Rate capability for Xenon mixtures at M=104

Results from GSI test beam

M~104

10 100

rate [kHz/cm2]

protons p = 2 GeV/c (~mips)

M. Petris

consistentpicture!

Conclusions.

● Maximum rate of Φ = 100 kHz/cm2 at 5% gain drop achieved with protons of p = 2 GeV/c (~mips) in Xe/CO2(85/15) is fully consistent with the systematic measurements performed in this work.

●Feasibility of operation in Xe/CO2 mixtures shown at the

required CBM rates. Adequate choice of the concentration of Xe/CO2 and a reduction of a factor 2 in operating gain would result in a safe increase of a factor 4 in the rate capability.

●The gain at sustained glow discharge is M = 2x105 for Xe/CO2, independent of the fraction of quencher. Operation of the chamber at gains M = 0.5-1x104 provides therefore a safe margin.

●A global fit to the data taken based on the Mathieson formula was achieved, showing great predictive power. The obtained mobilities are in good agreement with mobilities at zero field for Argon and CO2. For Neon a disagreement by a factor 2 observed. For Xenon no data is available, but the scale seems correct.

Appendix

Beam size dependence

r [mm]2 4 6 8 10

oV

rV

)(

0

5.0)(

o

eff

V

rV

A

A

Preliminary

A. Kalweit

Therefore: the rate capability is overestimated by the same factor (an independent measurement with de-focused beam points also to a factor around 0.6).

Gap h (~1/h2)

Wire pitch s (~1/s)

Gain drop F (~F)

Gain Mo (~Mo)

F

sh

C

nqaMF

bMFr L

oeo

o

1ln

21

1

1ln22

After obtaining the parameters of the fit (2x3 mobilities), the rate capability r at arbitrary gain drop F can be re-obtained as:

Rate capability.

Experimental procedure for gain measurements.

● The rate was monitored with a CAMAC scaler and corrections for the system dead-time applied. In these measurements this correction is at most 13%.

● The current was measured with a power supply of 1 nA resolution. At low gains, a Keithley picoamperemeter with resolution below pA was used.

● The pressure and temperature were monitored continuously, and the operating voltage re-normalized to the corresponding voltage at P=1000 bar and T=20 oC, to account for changes in density.

● The rate over the chamber was kept at the level of 10 kHz for gains M>105 and at the level of 120 kHz for gains M<105 (this reduces the space-charge at high gains, simplifying the theoretical description). This was achieved for all the gas mixtures studied, by accommodating the different X-ray conversion probabilities with changes in the current of the tube.

●Measurements for s = 2, 3, 4 mm and gas mixtures of Ne,Ar,Xe/CO2 at concentrations of noble gas f = 0, 10, 20, 40, 60, 80, 90, 99 were performed

X-ray tube vs (reference) Fe55 source (I).

The energy distributions are very similar but, in the case of the source, the measured distribution is dominated by the detector resolution and in the case of the tube it is dominated by the shape of the primary X-ray distribution.

Ar/CO2

(80/20)

escapepeak

SO: Once calibrated in energy, the tube was used for the whole data taken

X-ray tube vs (reference) Fe55 source (II).

Good mutual agreement (below 5%) between both!

The glow discharge.

Photon or ion feedback may result in re-generation of electrons close to the cathode that causes a self-sustained current in the detector (glow-discharge).

A. Battiato

In this kind of mixtures and at this temperature and pressure conditions, it is the main cause of gain limitation of the detectors

Comparison with models (Magboltz).

Meta-stable Ar* always appear during the avalanche process. If Ar* finds a CO2 molecule, it can be de-excited upon ionization of it (Penning effect).

The fraction of Ar* atoms that undergoes Penning or ‘Penning fraction’ is a free parameter, but can be constrained by our data since it depends only on the mixture.

C. Garabatos

Ar/CO2

(80/20)

PRELIMINARY!

Experimental procedure for rate capability measurements.

● The gain was measured as a function of the primary rate. The dead-time of the system was estimated (and corrected) by making use of the proportionality between the current of the X-ray tube and the primary rate, and using a phenomenological model with only 1 free parameter.

● The area of illumination has been measured by exposing a Polaroid film.

● The current was measured through the power supply.

● The rate was measured witha NIM scaler.

Rate capability. The ‘standard model’ (I).

E. Mathieson, NIM A 249(1986)413

In the particular case where , the dependence of the gain with rate is given by the transcendental equation:

baVM ln

)])(ln[2

exp(1

12

MMVC

shbanqMM b

loeo

Where no is the initial number of clusters, φ the primary flux in [Hz/cm2], Cl is the capacitance per unit length, μ the ion mobility, s the wire pitch and h the gap between anode and cathode.

Slopes of the gain curve

The onset of the glow discharge