Blum W., Riegler W., Rolandi L. Particle Detection with Drift Chambers

Подождите немного. Документ загружается.

432 12 Drift-Chamber Gases

the space between the anode and the cathode. In Malter’s experiment the positive

surface charges were produced by an external beam of particles; when this beam was

removed, the ﬁeld emission was not immediately interrupted, but decayed slowly

because the surface charge took an appreciable time to leak away.

In a wire chamber, some thin insulating coating on the cathodes is charged up by

the positive ions that are produced in the avalanches and drift to the cathodes. The

surface charge density depends on the balance between the neutralisation rate and

the ion collection rate, the latter being proportional to the rate of charge multiplica-

tion on the anode. If the surface charge density increases above some critical value,

secondary electron emission from the cathode into the gas via the Malter effect be-

comes possible. The secondary electrons reach the anode and produce avalanches,

thus increasing the positive-ion production rate. If the secondary-electron emission

rate is sufﬁciently high, the process becomes self-sustaining and the dark current

remains, even if the source of the primary radiation is removed from the chamber.

If one stops the avalanches by turning down the anode voltage, the dark current

is obviously interrupted, but it does not immediately reappear when the anode volt-

age is subsequently turned up again. A certain intensity of radiation on the area in

question is required to reestablish the current.

An efﬁcient way of preventing the development of cathode deposits is to keep

the electric ﬁeld at the cathode surfaces low. Plane cathodes and wires with larger

diameters show smaller deposits.

In chambers with appreciable standing currents, the addition of a small amount of

water in the gas has often been found to suppress the currents [WOR 86, DAN 89].

An explanation is that the cathode deposits are made conductive so as to increase

the ion leakage rate through the insulating ﬁlm. It is also generally known that small

amounts of alcohol have a tendency to prevent or delay the appearance of dark

currents.

Another cause of standing currents and of the breakdown of chambers is the so-

called whiskers, ﬁne strands of material growing on the cathode or the anode. Glow

discharges may appear owing to the large ﬁelds they create.

The growth of whiskers is primarily a question of the gas composition. For ex-

ample, a change from the gas composition 1 (Ar(89%) + CO

(10%) + CH

(1%)) to

the similar gas composition 2 (Ar(90%) + CO

(5%) + CH

(5%)) produced whiskers

which grew from the cathode wires in the ﬁeld direction, according to a test by the

JADE group [KAD 86]. Upon going back to composition 1, the whiskers shrank

and ﬁnally disappeared. In another test [FOS 86] it was shown that whisker growth

on cathode wires is supported by gas mixtures containing the lower alkanes, which

presumably form (CH

)

-type polymers. On the other hand, no mixtures of car-

bon dioxide and argon support whisker growth; rather they make existing whiskers

retreat and disappear. Table 12.12 summarises these tests, which were done under

the extreme conditions of glow discharge. Although such extreme conditions do not

usually apply to chambers that are working well, they may develop temporarily at

some imperfect spot in a chamber.

Let us add to this list of causes of dark currents the photoeffect discussed in

Sect. 4.1. Depending on the photoabsorption properties of the gas, photons emitted

12.6 Deterioration of Chamber Performance with Usage (‘Ageing’) 433

Table 12.12 Whisker growth measured in glow discharges produced on a cathode wire [FOS 86]

Gas mixture Whisker growth Comments

(percentages) allowed?

Argon (50) + ethane (50) Yes Whiskers grow down to

concentrations of ethane

of 5% or more

Argon (48) + ethane (48) Yes Growth is suppressed by adding

+ ethanol (4) ethanol

Argon (40) + ethane (40) Yes

+ carbon dioxide (20)

Argon + carbon dioxide No No mixture of Ar and CO

supports whisker growth

Argon (70) + carbon No

dioxide(25) + ethanol (5)

Argon (90) + methane (10) Yes

Argon (95) + methane (5) Yes

Argon (50) + isobutane (50) Yes Threshold for whisker growth about

Argon (60) + isobutane (40) twice as high as with argon–ethane

Argon (92) + isobutane (8) mixtures

in an avalanche may create electrons in the gas or on the cathode which produce

new avalanches. Whether this feedback situation can lead to limited dark currents

depends on the presence of some damping mechanism.

12.6.3 Ageing Tests in the Lower-Flux Regime

The process of ageing has been investigated in numerous laboratory experiments.

With the help of intense radioactive sources, the ageing process occurs within

shorter periods of time. Although it is established that the measured fatal dose is

not independent of this intensity (it comes out larger at intensities that are too high

[VAV 86]), the method is useful for the comparison of different factors that inﬂu-

ence the ageing process. A list complete up to 1985 is contained in [WOR 86]. Later

work may be followed by consulting [ATA 87] and [KAD 90].

A ﬁrst line of investigation concerns the nature of the quenching agent, which,

either alone or in combination with a noble gas (typically Ar), would serve as the

counting gas. Perhaps some quenching agents age later than others? One often mea-

sures the spectrum of an Fe-55 source and observes its peak and width as functions

of the charge collected on the sense wires. Figure 12.7 shows the results of Turala

and co-workers [TUR 86]; the relative pulse height as well as the peak of the Fe-55

signal began to deteriorate at lifetimes (total charge collected) that were different by

an order of magnitude or more as isopentane was exchanged for carbon dioxide or

nitrogen.

434 12 Drift-Chamber Gases

Fig. 12.7 Energy resolution and peak position of the

Fe signal in various gas mixtures as functions

of the total collected charge, measured by [TUR 86]

It is generally reported that the addition of a small percentage of some alco-

hol (CH

OH, (CH

OH)

,(CH

)

CHOH) to the counting gas extends the

lifetime of chambers signiﬁcantly, even if it does not prevent the development of

deposits.

Another line of enquiry starts from the observation that chemical elements found

in the wire deposits are occasionally foreign to the chemical compounds of the gas

or the wires. Silicon, a dominant signal in X-ray ﬂuorescence or Auger-electron-

spectrometry analyses of some wire deposits could only have been contained in

the gas at the low concentrations that are permitted by the tight tolerances of the

gas-purity speciﬁcations. In addition there may have been some parts of the cham-

ber installation where silicon-containing compounds reached the gas at very low

vapour pressure. In one instance, some silicon oil present in the counting gas with

a vapour pressure of 10

−9

Torr was observed to cause rapid growth of polymers

on the anode [HIL 86]. This implies a surprisingly high silicon collection efﬁciency

of the avalanche process. Could it be that other substances ‘poison’ the gas in a

similar manner? And if so, under what circumstances? In Fig. 12.8, the surpris-

ing results of a test performed by Kotthaus are depicted. The input gas stream of

argon–ethane was conducted through clean steel tubes; when the integrated col-

lected charge reached a certain value marked in the ﬁgure, a 10-m-long hose of soft

PVC was introduced into the input gas stream, resulting in an immediate deteriora-

tion of the test-tube performance, as measured by the pulse height and the resolution

of the Fe-55 spectrum. When the hose was removed, the performance did not return

12.6 Deterioration of Chamber Performance with Usage (‘Ageing’) 435

Fig. 12.8 Effect of a soft, 10-m-long, PVC gas hose. Dots (referring to the left-hand-side scale)

indicate the position of the

Fe peak relative to the (unaged) reference position. The crosses (refer-

ring to the right-hand-side scale) show the width of the

Fe distribution relative to its peak position

in the radiation damaged wire region. The lower drawn-out curve (referring to the right-hand scale)

represents the same quantity, but measured at the (unaged) reference position. The period of time

in which the PVC hose was put into the input gas stream is also indicated [KOT 86]

to its original values, but the loss of gain persisted. It is also remarkable that the

resolution changed quite signiﬁcantly on a piece of wire that had not aged (outside

the area of exposure), but only when in the presence of the PVC hose. Such obser-

vations support the idea that the process of polymerisation, which later continues

on its own, is started by certain dangerous substances, in this case perhaps one of

the organic softening agents that are mixed into the PVC giving it its characteristic

odour. In his summary [VAV 86], Va’vra makes a list of contaminations that, accord-

ing to experience, should be avoided in order to achieve extended chamber lifetimes;

the list includes halogens, oil traces (also from bubblers), rubber (especially silicon

rubber), polyurethane adhesive, PVC and teﬂon tubing, soft epoxies and adhesives,

aggressive solder, any unknown organic materials, and large amounts of G10 (glass-

ﬁbre-enforced material containing Si). Cold traps have a beneﬁcial effect [HIL 86].

12.6.4 Ageing Tests in the High-Flux Regime

Here we refer to the new high-rate experiments (HERA-B, LHC-experiments) in

which the entire charge collected on the sense wires will probably go up to 10

Coulomb per cm of wire length or more, two or three orders of magnitude higher

than in the experiments of the previous generation. Here one comes into a new

regime of ageing.

436 12 Drift-Chamber Gases

The consequences are clear: all hydrocarbons as quenchers are out of the ques-

tion because of their tendency to polymerise; compare Table 4.1 in which only CO

is useable. Even the slightest contamination, especially of silicon compounds, has

to be excluded, requiring extremely clean gas supply systems and well-planned val-

idation tests for the materials of all components in contact with the gas. Lists of

materials and their outgassing rates exist [NASA], [BOU96], [CAP01].

At the high-charge doses studied, new phenomena became apparent and a new

quench gas, CF

, was introduced. Carbon tetraﬂuoride is non-ﬂammable, non-

polymerising and in gas mixtures with a noble gas makes fast counters even at

moderate ﬁelds. The reaction products formed in avalanches are quite aggressive

towards materials such as electrodes and gas ﬁlters, but also against previously ex-

isting polymer deposits, which they seem to be able to remove. There are the two

competing processes caused by the radicals formed in avalanches: polymerisation

and etching. This competition was demonstrated in an experiment by Openshaw

et al. [OPE 91]. Working with a gas consisting of a mixture of CF

and isobutane,

they were able to shift the balance between the two processes towards more poly-

merisation or towards more etching by changing the relative gas concentrations.

As a modern example we note the tubes of the ATLAS transition radiation tracker

(TRT) [AKE 03], which have perhaps the most heavily irradiated proportional coun-

ters among the modern experiments, as their sense wires will draw up to 0.15

A/cm

of wire length, corresponding to approx. 10 C/cm in 10 years of LHC operation.

Based on the experience described above, the ﬁrst prototypes were operated with a

gas containing CF

: 70%Xe + 20%CF

+ 10%CO

, the xenon being required for

the transition radiation photon conversion. The presence of CF

proved in the end

to be too aggressive against some chamber components, notably glass. The com-

mercial price of xenon necessitates a closed gas circulation system. The group has

developed certiﬁcation methods for cleaning components and has demonstrated that

the deterioration caused by the unavoidable presence of minute traces of silicon –

Fig. 12.9 Anode wire after only 0.08 C/cm accumulated charge. The deposits were caused by

traces of Si from a lubricant used in the early straw production; they lead to a gas gain drop of

about 30%

12.6 Deterioration of Chamber Performance with Usage (‘Ageing’) 437

Fig. 12.10 Surface of a wire observed after 0.5 C/cm of accumulated charge, for the TRT standard

gas (see text) with 1.2% of water and 1.5% of oxygen added. Amorphous deposits of WO are

observed on top of the gold-plated tungsten wire. An energy dispersive X-ray (EDX) spectrum

reveals the presence of tungsten and oxygen

an example is shown in Fig. 12.9 – can be healed by etching away the deposits, a

process which will occur automatically during the experiment.

High charge doses on the wire can cause a new phenomenon even more dis-

astrous than whisker growth: the gold plating on the tungsten wire may peel off

(Fig. 12.12) under the inﬂuence of gas components such as H

OorO

.Alsothe

wire may swell to twice its diameter forming tungsten oxide, (Fig. 12.10). Under

similar circumstances, crystalline tungsten oxide layers form deposits on the wire

(Fig. 12.11).

An excellent summary of the state of the art of ageing is the article the editors

of the DESY workshop [WOR 01] wrote for the IEEE Transactions on Nuclear

Science [TIT02].

Fig. 12.11 Surface of a wire carrying crystalline deposits of WO under experimental conditions

similar to those of Fig. 12.10

438 12 Drift-Chamber Gases

Fig. 12.12 Various manifestations of damage to the gold-plating after 2-6 C/cm of accumulated

charge for diffferent current densities (1-4

A/cm). No gold-plating damage was observed when

the water content was below 0.1%. All photographs from [AKE 03]

References 439

The requirement to eliminate all hydrocarbons from the gas of proportional

chambers dedicated to very-high-rate experiments changes the situation in a drastic

way. Whereas previously we carefully optimised the gas properties by choosing the

right gas composition from a good variety of possible components, now there are

only a few left. Important gas properties such as the drift velocity and its depen-

dence on electric and magnetic ﬁelds, diffusion, Lorentz angle, quenching strength,

etc. (cf. Sects. 12.1, 12.5) must be taken as they come, if the chambers are to sur-

vive the high dose awaiting them in the new experiments. If, for example, the drift

velocity cannot be made insensitive to the electric ﬁeld, any space charge in the drift

region will limit the measurement accuracy and open a new frontier for the control

and, perhaps, correction of such effects.

A thorough understanding of the ageing process must involve the chemical

processes in the plasma of the avalanches. There are certainly several competing

mechanisms at work; this is also apparent from the wide variety of manifestations

of the deposits as well as from the fact that drastically different element abundances

are measured in different spots along the wires [KOT 86].

At present we do not even know whether a better understanding of the plasma

polymerisation on our wires will give us the means to substantially prolong the

lifetime of the chambers. Thus for the time being we are left with a few rules –

to keep the gas and the wires extremely clean, to add some alcohol or water if

necessary, and to keep the cathode wires thick. Since it is the total collected charge

which destroys the wire, it is useful to keep the ampliﬁcation low.

References

[AKE 03] T. Akeson et al., Aging studies for the ATLAS Transition Radiation Tracker (TRT),

in [WOR01], p. 166

[ATA 87] M. Atac, Wire chamber aging and wire material, IEEE Trans. NS 34, 476 (1987)

[BEC 92] U. Becker, private communication (1992)

[BEI 88a] S.P. Beingessner, T.C. Meyer and M. Yvert, Inﬂuence of chemical trace additives on

the future ageing of the UA1 central detector, Nucl. Instrum. Methods A 272, 669

(1988)

[BEI 88b] S.P. Beingessner, T.C. Meyer, A. Norton, M. Pimi

a, C. Rubbia and V. Vuillemin, First

running experience with the laser calibration system of the UA1 central detector, in

Proc. Int. Conf. on Advanced Technology and Particle Physics, Villa Olmo (Como),

Italy, 1988

[BEN 89] A.C. Benvenuti et al., A nonﬂammable gas mixture for plastic limited streamer tubes.

Nucl. Instrum. Methods Phys. Res. A 284, 339 (1989)

[BIA 89] S.F. Biagi, A multiterm Boltzmann analysis of drift velocity, diffusion, gain and mag-

netic effects in argon-methane-water-vapour mixtures, Nucl. Instrum. Methods Phys.

Res. A 283, 716 (1989)

[BIN 86] D.M. Binnie, Experience with the TASSO chambers, in [WOR 86], p. 213

[BOU96] R. Bouclier et al., xxxx, Nucl. Instrum. Methods A 381, 289 (1996)

[BRE 74] A. Breskin, G. Charpak, B. Gabioud, F. Sauli, N. Trautner, W. Duinker, G. Schultz,

Further results on the operation of high-accuracy drift chambers, Nucl. Instrum.

Methods 119, 9 (1974)

440 12 Drift-Chamber Gases

[BRO 79] F. Brown, N. Hadley, P. Miller, M. Pettersen, D. Phan, G. Przybylski and P. Rohrisch,

List of poisoning times for materials tested in the materials test chamber, LBL Berke-

ley note TPC-LBL-79-8 (unpublished)

[BUR 48] J.H. Burgoyne and G. Williams-Leir, Limits of inﬂammability of gases in the pres-

ence of diluants, a critical review, Fuel 27, 118 (1948)

[CAP01] M. Capeans, Aging and materials: lessons for detectors and gas systems, in [WOR01]

[CAT 89] A. Cattai, H.G. Fischer, A. Morelli, Drift velocity in Ar + CH

+ water vapours,

DELPHI internal note 89-63, 28 July 1989, and H.G. Fischer, private communication

[CER 96] Technical Inspection and Safety Commission (TIS) (ed.), Flammable Gas Safety

Manual 1996, Annex to Safety Code G, CERN, Geneva 1996

[DAN 89] M. Danilov, V. Nagovitsin, V. Shibaev, I. Tichomirov, E. Michel, W. Schmidt-

Parzefall, A study of drift chamber ageing with propane, Nucl. Instrum. Methods

Phys. Res. A 274, 189 (1989)

[DRY 86] S.L.T. Drysdale, K.W.D. Ledingham, C. Raine, K.M. Smith, M.H.C. Smyth,

D.T. Stewart, M. Towrie and C.M. Houston, Detection of toluene in a proportional

counter gas by resonant two-photon ionization spectroscopy, Nucl. Instrum. Methods

A 252, 521 (1986)

[EST 86] P. Estabrooks, Aging effects in a large drift chamber in the Fermilab Tagged-Photon

Spectrometer, in [WOR 86], p. 231

[FOS 86] B. Foster, Whisker growth in test cells, in [WOR 86], p. 227, and private communi-

cation

[GUS 84] E.M. Gushchin, A.N. Lebedev and S.V. Somov, Tracks produced by an N

laser in

a streamer chamber, Instrum. and Exp. Tech. (Translation of Prib. i. Tekh. Eksp.) 27,

546 (1984)

[HIL 86] H.J. Hilke, Detector calibration with lasers, a review, Nucl. Instrum. Methods A 252,

169 (1986)

[HIL 86a] H.J. Hilke, Summary of ageing studies in wire chambers by AFS, DELPHI and EMC

groups, in [WOR 86], p. 153

[HUB 85] G. Hubricht, K. Kleinknecht, E. M

uller, D. Pollmann and E. Teupe, Ionization of

counting gases and ionizable gaseous additives in proportional chambers by UV

lasers, Nucl. Instrum. Methods A 228, 327 (1985)

[HUB 86] G. Hubricht, K. Kleinknecht, E. M

uller, D. Pollmann, K. Schmitz and C. St

urzl,

Investigation of UV laser ionisation in naphtalene and phenol vapours added to pro-

portional chamber gases, Nucl. Instrum. Methods A 243, 495 (1986)

[HUK 88] M. Huk, P. Igo-Kemenes and A. Wagner, Electron attachment to oxygen, water and

methanol in various drift chamber gas mixtures, Nucl. Instrum. Methods Phys. Res.

A 267, 107 (1988)

[KAD 86] H. Kado, Performance of the JADE Vertex Detector, in [WOR 86], p. 207

[KAD 90] J. Kadyk, J. Wise, D. Hess and M. Williams, Anode wire aging tests with selected

gases, IEEE Trans. NS 37, 478 (1990)

[KOT 86] R. Kotthaus, A laboratory study of radiation damage to drift chambers, in [WOR 86],

p. 161

[LED 84] K.W.D. Ledingham, C. Raine, K.M. Smith, A.M. Campbell, M. Towrie, C. Trager

and C.M. Houston, Laser induced ionization in proportional counters seeded with

low-ionization-potential vapours, Nucl. Instrum. Methods A 225, 319 (1984)

[LED 85] K.W.D. Ledingham, C. Raine, K.M. Smith, M.H.C. Smyth, D.T. Steward, M. Towrie

and C.M. Houston, Wavelength dependence of laser-induced ionization in propor-

tional counters, Nucl. Instrum. Methods A 241, 441 (1985)

[LEH 89] I. Lehraus, R. Matthewson and W. Tejessy, Materials test facility for the ALEPH

TPC, CERN Note EF 89-8, also as ALEPH 89-89 (unpublished)

[LEW 61] B. Lewis and G. von Elbe, Combustion, Flames and Explosions of Gases, 2nd edn

(Academic, New York London 1961)

[MAL 36] L. Malter, Thin ﬁlm ﬁeld emission, Phys. Rev. 50, 48 (1936)

References 441

[NASA] The National Aeronautics and Space Administration makes available a databas with

thousands of materials and their outgassing properties, reachable via the NASA

homepage on the internet.

[OPE 91] R. Openshaw, R.S. Henderson, W. Faszer, M. Salomon, Etching of anode wire de-

posits with CF

/isobutane (80:20) avalanches, Nucl. Instrum. Methods in Phys. Res.

A 307, 298 (1991)

[PAL 75] V. Palladino and B. Sadoulet, Application of classical theory of electrons in gases to

drift proportional chambers, Nucl. Instrum. Methods 128, 323 (1975)

[PIL 86] P.H. Pile, Radiation damage control in the BNL Hypernuclear Spectrometer drift

chamber system, in [WOR 86], p. 219

[RIE 72] F.F. Rieke and W. Prepejchal, Ionization cross-sections of various gaseous atoms and

molecules for high-energy electrons and positrons, Phys. Rev. A 6, 1507 (1972)

[SAU 86] F. Sauli, When everything was clear, in [WOR 86], p. 1

[SCH 86] B. Schmidt, Drift und Diffusion von Elektronen in Methan und Methan-Edelgas-

Gemischen, Dissertation, Univ. Heidelberg (1986)

[SCH 92] Private communication J. Schmid, CERN (1992)

[SPI 86] W. Haberichter, H. Spinka, Wire chamber degradation at the Argonne ZGS, presented

by H. Spinka, in [WOR 86], p. 99

[TIT02] M. Titov, M. Hohlmann, C. Padilla, N. Tesch, Summary and Outlook of the Inter-

national Workshop on Aging Phenomena in Gaseous Detectors (DESY, Hamburg,

October 2001), IEEE Transactions on Nucl. Sc. 49, pp. 1609–1621

[TOW 86] M. Towrie, J.W. Cahill, K.W.D. Ledingham, C. Raine, K.M. Smith, M.H.C. Smyth,

D.T.S. Stewart and C.M. Houston, Detection of phenol in proportional-counter gas

by two-photon ionization spectroscopy, J. Phys. A (At. Mol. Phys.) 19, 1989 (1986)

[TUR 86] A. Dwurazny, Z. Hajduk, M. Turala, ageing effects in gaseous detectors, and search

for remedies, presented by M. Turala, in [WOR 86], p. 113

[ULL 86] OMEGA and SFM Collaborations, The OMEGA and SFM experience in intense

beams, presented by O. Ullaland, in [WOR 86], p. 107

[WOR 86] Proceedings of the Workshop on Radiation Damage to Wire Chambers, held at

Berkeley, January 1986 (J. Kadyk workshop organizer), Lawrence Berkeley Labo-

ratory, University of California, Berkeley, LBL 21 170 (1986) (unpublished)

[WOR 01] M. Hohlmann, C. Padilla, N. Tesch, M. Titov (eds.), Aging Workshop 2001, Pro-

ceedings of the International Workshop on Aging Phenomena in Gaseous Detectors,

DESY, Hamburg, Germany, October 2-5, 2001, Nucl. Instr. Meth. A in Phys. Res. 515

(2003)

[VAV 86] J. Va’vra, Review of wire chamber ageing, in [WOR 86], p. 263

[YVE 86] S. Beingessner, T. Meyer, V. Vuillemin, M. Yvert, Our aging experience with the

UA1 central detector, presented by M. Yvert, in [WOR 86], p. 67

[ZAB 65] M. Zabetakis, Flammability characteristics of combustible gases and vapours, Bul-

letin 627 of the Bureau of Mines, United States Department of the Interior 1965. The

curves are generally in agreement with the numbers of [BUR 48], except for CH

/He