1: %\documentstyle[12pt,epsfig,rotate]{article}

2: \documentstyle[]{IEEEtran}

3: \def\figurename{\large Figure}

4: \def\abstractname{\large Abstract}

5: %\def\tablename{\large Table}

6: %\pagestyle{empty}

7: 

8: %\textwidth=166mm

9: %\textheight=238mm

10: %\oddsidemargin=-4.4mm

11: %\topmargin=-0.4mm

12: %\headheight=0mm

13: %\headsep=0mm

14: 

15: % A useful Journal macro

16: \def\Journal#1#2#3#4{{#1} {\bf #2}, #3 (#4)}

17: 

18: % Some useful journal names

19: \def\NCA{\em Nuovo Cimento}

20: \def\NIM{\em Nucl. Instrum. Methods}

21: \def\NIMA{{\em Nucl. Instrum. Methods} A}

22: \def\NPB{{\em Nucl. Phys.} B}

23: \def\PLB{{\em Phys. Lett.}  B}

24: \def\PRL{\em Phys. Rev. Lett.}

25: \def\PRD{{\em Phys. Rev.} D}

26: \def\ZPC{{\em Z. Phys.} C}

27: 

28: \newcommand{\ra}{\rightarrow}

29: \newcommand{\aoh}{\mbox{$ Ar/CO_2/CH_4 $}}

30: \newcommand{\afh}{\mbox{$ Ar/CF_4/CH_4 $}}

31: \newcommand{\afo}{\mbox{$ Ar/CF_4/CO_2 $}}

32: 

33: 

34: \begin{document}

35: 

36: 

37: \begin{titlepage}

38: 

39: \pagebreak

40: 

41: \vspace*{0.1cm}

42: \begin{center}

43: {\huge Institute of Theoretical }

44: \end{center} 

45: \begin{center}

46: {\huge and Experimental Physics }

47: \end{center} 

48: \begin{center}

49: \vspace{0.5cm}

50: {\huge Preprint  43-00 }

51: \end{center} 

52: \vspace{3.0cm}

53: \begin{center}

54: {\large \bf M.Danilov, L.Laptin, I.Tichomirov, M.Titov, Yu.Zaitsev}

55: \end{center} 

56: 

57: \begin{center}

58: \vspace{2.cm}

59: {\large\bf Aging tests of the proportional wire chambers}

60: \end{center}

61: \begin{center}

62: %\vspace{1.0cm} 

63: {\large\bf using $\afh$(74:20:6), $\afh$(67:30:3)}

64: \end{center}

65: \begin{center}

66: %\vspace{1.0cm} 

67: {\large\bf and $\afo$(65:30:5) mixtures}

68: \end{center}

69: \begin{center}

70: %\vspace{1.0cm} 

71: {\large\bf for the HERA-B Muon Detector}

72: \end{center}

73:  

74: \vspace{7.5cm}

75: 

76: \begin{center}

77: {\large\bf Moscow 2000}

78: \end{center} 

79: 

80: \end{titlepage}

81: 

82: \newpage

83: 

84: %\vspace*{7.5cm}

85: 

86: \begin{abstract}

87: \large 

88: 

89: The Muon Detector of the HERA-B experiment at DESY is a gaseous 

90: detector that provides muon identification in a high-rate hadronic

91: environment. 

92: We present our studies on the properties  of several fast gases,

93: $\afh$ (74:20:6), $\afh$ (67:30:3) and $\afo$ (65:30:5), 

94: which have been found to fulfill muon detection requirements.

95: 

96: The severe radiation environment of the HERA-B experiment leads to the maximum

97: charge deposit on a wire, within the muon detector, of 200 mC/cm per year.

98: For operation in such an environment, the main criteria for the choice of gas

99: turned out to be stability against aging.

100:   An overview of aging results from

101: laboratory setups and experimental detectors for binary and ternary 

102: mixtures of $Ar$, $CH_4$, $CF_4$ and $CO_2$ is presented and

103: the relevance of the various aging results is discussed.

104: Since it is not clear how to extrapolate aging results from small

105: to large areas of irradiation, the lifetime of 

106: aluminum proportional chambers was studied under various conditions.

107: In this paper we provide evidence

108: that aging results depend not only upon the 

109: total collected charge.   

110: It was found that the aging rate for irradiation with $Fe^{55}$

111: $X$-rays and 100~MeV $\alpha$-particles may differ by more than

112: two orders of magnitude.

113: 

114: \end{abstract}

115: \large

116: \clearpage

117: 

118: \pagenumbering{arabic}

119: \setcounter{page}{1}

120: 

121: \section {Introduction.}

122: 

123:  The HERA-B experiment is a hadronic B-factory at DESY, 

124: operating presently under LHC-like conditions, 

125: five years before startup of LHC.

126: The muon detection system in HERA-B is made up of three different types 

127: of gas proportional chambers: tube, pad and pixel~\cite{Proposal,TDR}. 

128: The readout of the chambers is based on the ASD-8

129: amplifier shaper discriminator chip developed by the University

130: of Pennsylvania~\cite{electronics}.

131: Strong restrictions on the gas choice are imposed by

132: the harsh operating conditions.

133:  In addition to the high radiation load, resulting in an  

134: accumulated charge on a wire of up to 200 mC/cm per year, the  

135: muon chambers have to withstand the presence of heavily 

136: ionizing particles, neutrons, and low energy gammas.

137:  The first level trigger  requires high hit efficiency and 

138: fast signal collection within the 96~ns time interval

139: between two consecutive bunch crossings.

140: 

141: In section 2, the requirements and limitations of the choice 

142: of gases for the muon system are explained and the chamber

143: characteristics studied with several fast $Ar/CF_4$-based mixtures

144: are discussed. 

145: %However, the main criteria for the choice of the gas

146: %turned out to be the stability against the aging.

147:  The overview of aging results from

148: laboratory setups and experimental detectors for binary 

149: and ternary mixtures of $Ar$, $CH_4$, $CO_2$ and $CF_4$ is 

150: presented in section 3.  

151: Since it is not clear how to extrapolate aging results 

152: from laboratory setups, when irradiating a small region of the wire,

153: to large areas of irradiation with hadronic particles, 

154: we have studied aging properties

155: of aluminum proportional chambers operated with $\afh$ and $\afo$

156: mixtures with $Fe^{55}$ and $Ru^{106}$ sources and in 

157: 100~MeV $\alpha$-beam. 

158: The results of these aging tests performed in a variety of conditions

159: are summarized in sections 4 and 5.

160:  

161: %Nowdays, two large area tube chambers 

162: %are being irradiated in the HERA-B high-rate environment.

163: %In this paper we provide an evidence,

164: %that aging results depend not only from the total 

165: %collected charge, but, in addition, 

166: %could be also affected by

167: %the mode of operation and size of the irradiation area.

168: 

169: 

170: \section{ Selection of fast gas mixtures.}

171: 

172: Large area gaseous detectors operated under 

173: high intensity particle fluxes impose 

174: new requirements for gas selection. 

175: The HERA-B muon system, with a total gas volume of 8~$m^3$,

176: puts rather stringent constraints

177: on the gaseous medium to be used: extremely low aging, 

178: high sparkproofness, good electrical properties (high drift 

179: velocity $v_d$, convenient operating electric field E/p),

180: good chemical properties (non-flammable, non-poisonous).

181:  Since it is necessary to replenish part of the gas to avoid 

182: contamination by air, the cost of the component gases is also an

183: important issue for gas selection.

184: 

185: For a minimum working voltage, noble gases are usually chosen

186: since they require the lowest electric field intensities

187: for avalanche formation.

188:  However, photons limit the attainable gas amplification  

189: in the $Ar$-operated counters to $\sim 10^{4}$

190: without encountering a permanent discharge induced

191: by the photon-feedback mechanism  taking place at the cathode.

192:  This problem can be remedied by the addition of gases,

193: such as hydrocarbons, alcohols, $CO_2$, $H_2 O$, $N H_3$.

194:   These molecular additives act like quenchers and improve 

195: the operational stability 

196: of the counter by absorption of ultraviolet photons from the 

197: multiplication process, de-excitation of $Ar$ atoms~\cite{scint},

198: and by the charge transfer 

199: mechanism prevent the neutralization of argon ions at the cathode,

200: thus reducing electron emission 

201: through predissociation of quench gas ions~\cite{start,co2}.

202:  Some of the polyatomic gases will also effectively increase 

203: the total ionization by the Penning effect.     

204: 

205:  Moreover, in pure $Ar$ the electron drift velocity is $\sim 30$

206: times lower than in $CH_4$, although their elastic cross 

207: sections are nearly identical. However, the addition of even

208: small quantities ($\sim$0.1$\%$) of a molecular gas increases the  

209: drift velocity in argon threefold and also alters the shape dependence

210: of $v_d$ on $E/P$~\cite{N2,book1}.

211: %  In addition, in pure $Ar$ the drifting electrons obtain a rather

212: %low velocities in moderate electric fields, however, even small

213: %quantities (0.001??-10 $\%$) of molecular additives 

214: %($H_2$,$CO_2$,$N_2$) increases the

215: %drift velocity in argon threefold and also alters the shape dependence

216: %of $V_d$ on E/p~\cite{book1,N2}.

217:   In general, the drift velocity $v_d$ varies inversely with the 

218: product of the total scattering cross-section $\sigma_{sc}$

219: and the square root of the mean electron energy 

220: $\overline{\epsilon} $~\cite{fast1,fast2}.

221:  Traditionally, the fast gas mixtures were achieved by 

222: adding to $Ar$ a polyatomic gas~(usually a hydrocarbon or $CO_2$) 

223: which effectively shifts the low energy part of the 

224: electron distribution function into the region of the

225: Ramsauer-Townsend~(R-T) minimum, located at $\sim$ 0.5~eV,

226: of the momentum transfer and total electron scattering

227: cross-section of $Ar$.

228: The cooling of the electron swarm occurs as the energetic

229: electrons quickly transfer their energy to rotational and vibrational 

230: modes of the polyatomic gas, whose scattering cross-section has a

231: large inelastic component to the right of the R-T minimum. 

232: Furthermore, the cooling effect of a given gas in a high electric field 

233: gas amplification region need not be the same as it's cooling effect

234: in the lower electric field drift region. As was shown

235: in~\cite{armit}, the very similar effects observed 

236: in the $Ar/CH_4$ and $Ar/CO_2$ mixtures 

237: in the avalanche region are contrasted 

238: with different properties of these gases in the lower electric fields:

239: $CO_2$ is a much cooler gas than $CH_4$ in the drift region,

240: as is evidenced by the measured diffusion coefficients~\cite{cern}. 

241: These results open the

242: possibility of choosing a gas mixture to separately optimize

243: the desired drift and gain behavior in a particular application.

244: 

245:  Almost twenty years ago, $CF_4$ has been proposed for 

246: high rate environments as a most effective additive 

247: to raise the drift velocity in noble gases.

248: This gas obtains its large drift velocity because of the

249: sizeable Ramsauer-Townsend dip in the elastic cross-section which coincides

250: with a very large vibrational inelastic cross-section~\cite{biagi}.

251:  For the $Ar/CF_4$(90:10) mixture the 

252: electron drift velocity was found to exceed 10~cm/$\mu s$, 

253: which is twice as fast as for the conventional mixture

254: $Ar/CH_4$(90:10)~\cite{fast1}.

255:  Further increase of the $CF_4$ fraction 

256: in $Ar/CF_4$ mixtures tends to push 

257: the maximum drift velocity to the higher field region, 

258: but when the percentage of $CF_4$ becomes larger than 20~$\%$,

259: the $v_d$ is the same whatever the mixture,

260: indicating that it's mainly controlled by the $CF_4$ 

261: cross section~\cite{cf42,mix1,transport}.

262:  The lower vibrational modes, as well as  

263: the lower-lying negative ion states of $CF_4$ as compared to 

264: $CH_4$, may account for the superior drift properties 

265: of the $Ar/CF_4$ mixtures compared to  

266: $Ar/CH_4$~\cite{christ,book2}.

267: 

268: However, all binary $Ar/CF_4$ mixtures have rather poor energy 

269: resolution, compared to $Ar/CH_4$, due to the production 

270: of negative ions~\cite{fast1,attach}.

271: The dissociative attachment processes in $CF_4$ are the only mechanisms

272: leading to negative ion ($F^{-}$, $CF_3^{-}$) formation, via two

273: broad and overlapping resonances located between 4.5 and 

274: 10~eV~\cite{cf42,cf41}, and can cause serious inefficiencies in the

275: collection of electrons and degrade the position resolution.

276: The main evidence for the electron attachment is the direct measurement

277: of the effective ionization coefficient $\overline{\alpha} = \alpha - \eta$;

278: where $\alpha$ and $\eta$ are, respectively, the electron impact ionization 

279: and electron attachment coefficients. The negative values 

280: of $\overline{\alpha}$ were observed in the drift region

281: in pure $CF_4$, $Ar/CF_4$(80:20) and

282: $Ar/CF_4/H_2 O$(80:18:2) mixtures. Very interestingly,

283: the electron thermalizing 

284: properties effected by substituting 10~$\%$ of $CF_4$ in the

285: $Ar/CF_4$(80:20)

286: by 10~$\%$ of $CO_2$ virtually eliminates the electron attachment

287: in $Ar/CF_4/CO_2$(80:10:10) mixture~\cite{christ}.

288: 

289:  Moreover, due to the small $CF_4$ quenching cross-sections 

290: of metastable $Ar$-states~\cite{quench1,quench2},

291: the $Ar/CF_4$ mixtures are not sufficiently

292: self-quenching and show an intolerable level of afterpulsing  

293: even at moderate gas amplifications~(see also section 2.1). 

294:  This is not the case for the $Ar$/hydrocarbon mixtures, where

295: no significant afterpulsing was observed so far at moderate gas gains.

296: At the same time, some of the $Ar/CO_2$ mixtures are known to suffer from a low 

297: sparkproofness~\cite{becker}, 

298: since the quenching ability of $CO_2$ is smaller than that of hydrocarbons.

299: 

300: %\setlength{\unitlength}{1mm}

301: \begin{figure}[ph]

302: \setlength{\unitlength}{1mm}

303:  \begin{picture}(160,200)

304:  \put(-10.0,80.0){\special{psfile=fig1.eps vscale=58 hscale=56 } }

305:  \put(75.0,80.0){\special{psfile=fig2.eps vscale=58 hscale=56 } }

306:  \put(-10.0,-30.0){\special{psfile=fig3.eps vscale=58 hscale=56 } }

307:  \put(75.0,-30.0){\special{psfile=fig4.eps vscale=58 hscale=56 } }

308:  \put(20,120){ Figure 1: Electron drift velocity }

309:  \put(30,114){ as a function of $E/P$ }

310:  \put(103,120){Figure 2: Avalanche size curves as a }

311:  \put(104,114){ function of HV for 5.9~keV $X$-rays }

312:  \put(20,10){Figure 3: Tube chamber efficiency }

313:  \put(35,4){ as a function of HV }

314:  \put(103,10){Figure 4: Singles counting rates for a}

315:  \put(103,4){ tube chambercell as a function of HV}

316:  \end{picture}

317: %\caption[kk]{\em }

318: %  \label{fig:point2}

319: \end{figure}

320: 

321: The advantage of the enhanced drift velocity of $CF_4$ 

322: could be realized either in binary $CF_4$/quencher

323: or in ternary $Ar/CF_4$/quencher mixtures.

324: The choice of $CH_4$ among the hydrocarbons is 

325: dictated by the requirements that the drift velocity in the

326: lighter hydrocarbons is larger than that in the heavier  

327: ones~\cite{mix1}, especially in the  low-field region.

328: Fig. 1 shows the drift velocity for the $Ar/CF_4/CH_4$(74:20:6),

329: $CF_4/CH_4$(90:10), $Ar/CF_4/CO_2$(65:30:5),  $Ar/CF_4/CO_2$(80:10:10)

330:  and $CF_4/CO_2$(90:10) mixtures, as a function of the electric

331: field strength at atmospheric pressure. These results are taken from 

332: Refs.~\cite{christ,grimm,beck,mix2}. As can be seen from the curves,

333: the $CF_4/CH_4$ mixtures are significantly faster than 

334: $CF_4/CO_2$ for any given value of $E/P$.

335:   The use of $\afh$ could be more preferable for 

336: high-rate proportional chamber applications than $CF_4/CH_4$ mixtures, 

337: since the former shows fast rises in the drift velocity for 

338: the relatively low fields, with respect to the latter,

339: and the mobilities of $CH_4^+$ ions in $Ar$

340: are much higher than those in $CF_4$,

341: which is desirable in terms of space charge effects~\cite{mix1,mix2}.

342:  An additional disadvantage of the $CF_4/CH_4$ mixture is the necessity

343: to operate at high electric fields $E/P$ for a sufficient gas gain,

344: resulting in exceedingly high voltages in the chamber. 

345: This could introduce a problem  of electrostatic

346: instability of anode wires in large chambers.

347: Furthermore, considerations of cost and safety 

348: led us to consider $Ar/CF_4/CH_4$ mixtures as a

349: reasonable compromise between many experimental requirements.

350:  An optimization of the exact proportions of $Ar$, $CF_4$, and $CH_4$  in

351: the mixture was performed using the Magboltz and Garfield simulation programs.

352: As a result of simulation, $Ar/CF_4/CH_4$ (74:20:6) and another mixture

353: with a reduced $CH_4$ content $Ar/CF_4/CH_4$ (67:30:3) were chosen

354: for muon detector operation.

355: Unfortunately, mixtures with hydrocarbons have a tendency to 

356: cause polymerization effects, while

357: $CO_2$ quencher has been proven to give a much longer lifetimes.

358: However, addition of $CO_2$ appreciably changes the drift properties 

359: of the $Ar/CF_4/CO_2$ mixture, compared  

360: to the case of $Ar/CF_4/CH_4$;

361: reduces the maximum drift velocity, and shifts it to the higher 

362: field region.

363: In practice, we chose $Ar/CF_4/CO_2$(65:30:5) mixture, 

364: as a fall back gas, if the aging criterion becomes dominant.

365: 

366: Finally, in the  $Ar/CF_4/CH_4$ (74:20:6), 

367: $Ar/CF_4/CH_4$ (67:30:3) and $Ar/CF_4/CO_2$ (65:30:5) mixtures,

368: electron attachment processes may complicate gas performance.

369: However, the primary ionization in the chamber cell

370: is large enough and more than compensates for any losses  due to  

371: electron attachment, especially since only hit information is processed 

372: in the muon detector.

373: 

374: %\setlength{\unitlength}{1mm}

375: %\begin{figure}[ph]

376: % \begin{picture}(160,180)

377: % \put(-20.0,50.0){\special{psfile=fig1.eps vscale=58 hscale=56 } }

378: % \put(65.0,50.0){\special{psfile=fig2.eps vscale=58 hscale=56 } }

379: % \put(-20.0,-60.0){\special{psfile=fig3.eps vscale=58 hscale=56 } }

380: % \put(65.0,-60.0){\special{psfile=fig4.eps vscale=58 hscale=56 } }

381: % \put(5,90){ Figure 1: Electron drift velocity }

382: % \put(15,84){ as a function of $E/P$ }

383: % \put(88,90){Figure 2: Avalanche size curves as a }

384: % \put(89,84){ function of HV for 5.9~keV $X$-rays }

385: % \put(5,-20){Figure 3: Tube chamber efficiency }

386: % \put(20,-26){ as a function of HV }

387: % \put(88,-20){Figure 4: Singles counting rates for a}

388: % \put(88,-26){ tube chambercell as a function of HV}

389: % \end{picture}

390: %\caption[kk]{\em }

391: %  \label{fig:point2}

392: %\end{figure}

393: 

394: 

395: \subsection{ Tube Chamber Characteristics. }

396: 

397:  In this paper, we address the problem of gas selection 

398: and present the measured data only for tube 

399: chambers.

400:  The tube chamber is a closed-cell proportional wire chamber 

401: made from an aluminum profile~(wall thickness 2~mm)

402: with a drift cell 14~$\times$~12~mm$^2$ in cross section.

403:  Fig. 5 shows the schematic drawing of the tube chamber.

404: A gold-plated tungsten wire of 45~$\mu$m diameter and a length 

405: of nearly 3~m is stretched inside each cell and fixed 

406: mechanically with pins at the end of the chamber.

407:  In order to prevent efficiency losses due to dead spaces between cells, 

408: the tube chamber consists of two layers, each having 16 drift cells, 

409: shifted with respect to each other by half a cell size.

410: 

411: \begin{figure}[bth]

412: \setlength{\unitlength}{1mm}

413:  \begin{picture}(100,50)

414: \put(40.0,8.0){\special{psfile=tubeb1.ps

415:    voffset=0 hoffset=0  hscale=50 vscale=50 angle=0}}

416:  \put(30,6){ Figure 5: Schematic drawing of the tube chamber. Dimensions are }

417:  \put(40,0){ indicated in units of mm. }

418:  \end{picture}

419: \end{figure}

420: 

421:  Firstly, we have collected the $Fe^{55}$ spectra for each mixture

422: under study: $\afh$ (74:20:6), $\afh$ (67:30:3) and $\afo$ (65:30:5)

423: with a goal of inferring the response to minimum ionizing particles.

424: % The $Fe^{55}$ spectra were collected for each mixture

425: %under study: $\afh$ (74:20:6), $\afh$ (67:30:3) and $\afo$ (65:30:5).

426: Measured pulse heights were converted to avalanche sizes,

427: which correspond to the total number of electrons in the 

428: avalanche. The gas gain value

429: can be obtained  by dividing the observed pulse size by $\sim$ 200, 

430: the average number of primary electrons,

431: if the mixture has no electron attachment.

432:  Fig. 2 shows the total avalanche size increasing in an exponential 

433: manner with the high voltage, within the limits of the proportional region.

434: %All avalanche curves show good exponential behaviour

435: %within the limits of the proportional region, indicating a proper

436: %chamber operation.

437: It has been shown previously~\cite{gain1} that the onset 

438: of secondary gas effects such as space charge saturation

439: may be detected as a break in the expected scaling 

440: of the gas gain with respect both to anode surface field and

441: to gradient of the electric field near the wire.

442:  It should be noted that for these data taken with an $Fe^{55}$

443: source with its fixed initial charge ($\sim$200 electrons),

444: a particular value of gas gain is equivalent to a definite value of the 

445: total charge in the avalanche. Space charge effects depend on the 

446: total avalanche size and distribution of the positive ion cloud. Hence,

447: these numerical results cannot be automatically applied to other 

448: ionization sources with different magnitudes or distributions of the

449: initial charge.

450: 

451: Secondly, the tube chamber characteristics have been 

452: studied on the 3~GeV electron test beam at DESY.

453: The beam intensity was $\sim$~1~$kHz/cm^{2}$.

454:  The trigger was initiated by a coincidence of 

455: four beam-defining scintillator counters,

456: between which the chamber was positioned.

457: %Using a coincidence between the counters as a trigger, 

458: Then, the number of events with at least one hit in the double layer

459: chamber, within the 96~ns, versus the number of triggered events

460: yields an efficiency.

461: Fig. 3 shows the measured efficiency 

462: for the mixtures being studied, as a function of high voltage. 

463: This figure demonstrates, that the actual efficiency is approximately 

464: 100~$\%$ for the double layer chamber. 

465: However, for a single layer the usage of $\afo$(65:30:5) mixture  

466: results in about 7~$\%$ of signals registered outside the 96~ns  

467: (for perpendicular tracks), while in $\afh$ mixture nearly all

468: signals are collected within the 96~ns. 

469:  Fig. 4 presents singles (untriggered) counting rates for a 

470: tube chamber cell as a function of high voltage. 

471: From these curves, wide high voltage plateaus for a stable 

472: and efficient operation 

473: with  the $\afh$(74:20:6), $\afh$(67:30:3) and $\afo$(65:30:5) gases

474: are observed, where the counting rate is roughly uniform. 

475: Real afterpulses and multiple hits of ASD-8 per signal both 

476: contribute to the rise in singles rate curves, above the 

477: nominal beam intensity, with increasing  high voltage. 

478:  In comparison, $Ar/CF_4$(70:30) mixture

479: will not guarantee stable and efficient operation over

480: sufficient high-voltage range in the tube chambers.

481: The difference in singles rate curves, and correspondingly 

482: in detection efficiency,

483: between $Ar/CF_4$(70:30) and $\afh$(67:30:3) 

484: may be attributed to the enhanced gas electron production efficiency

485: and smaller electron attachment for the latter mixture.

486: In addition, onset of secondary avalanches in $Ar/CF_4$

487: limit the maximum operating voltage to 2.5~kV~(see Fig. 4), thus

488: causing a drastic decrease in the counting rate plateau width.

489:  

490: The results presented here justify 

491: the choice of $\afh$ (74:20:6), $\afh$ (67:30:3) and $\afo$ (65:30:5)

492: gases for muon detection with high efficiency.

493: %The radiation hardness of these mixtures will be discussed in sections 4-5.

494: 

495: 

496: \section{ General characteristics of aging processes. }

497: 

498:   Aging effects in proportional wire chambers,

499: a permanent degradation of operating

500: characteristics under sustained irradiation, has been and remains

501: the main limitation to their use in high-rate experiments~\cite{charpak}.

502:  Many processes are expected to occur simultaneously in the

503: gaseous discharges surrounding the wire

504: and it is nearly impossible to obtain

505: a quantitative chemical description in each particular case.

506:  There is much experimental information, excellently

507: summarized in~\cite{workshop,kadyk,vavra,cern1}, that suggests the 

508: wire chambers lifetime could be affected, and in some cases 

509: dramatically, by the nature and purity

510: of the gas mixture, materials used in contact with the gas,

511: by different additives, and trace contaminants.

512: 

513: While the specific reactions responsible for wire chamber aging 

514: are extremely complex, 

515: a qualitative description of the aging phenomena

516: in different gases could be obtained

517: from similarities between chemical processes in wire chamber

518: avalanches and those

519: that occur in the better-understood low-pressure 

520: rf-plasmas~\cite{kadyk,vavra,gcms,chemmod,hess}.

521: During wire avalanches many molecules break up 

522: by collisions with electrons, de-excitation of atoms, 

523: and UV-photon absorption processes. 

524: Whereas most ionization processes require electron energies 

525: greater than 10~eV,

526: the breaking of the chemical bonds in molecules and formation of free

527: radicals requires only 3-4~eV, and could lead to a large

528: abundance of free radicals over the ions in the  wire avalanches.

529: Similar to plasma polymerization~\cite{yasuda}, 

530: a free-radical polymerization 

531: seems to be the most appropriate mechanism of the wire chamber aging.

532:  Since free radicals are chemically very active they will either

533: % Once free radicals formed, they

534: %are chemically very active and will either

535: recombine back to original molecules or other volatile species,

536: or form new cross-linked molecular structures, 

537: which could react further to produce chain-like polymers

538: of increasing molecular weight, thus lowering

539: the volatility of the resulting product.

540:  When the growing of polymerized chain becomes large enough

541: for condensation to occur, it will diffuse to the

542: electrode surface.

543: The polymer deposition mechanism can be 

544: conceived as a phenomenon that

545: occurs whenever the gaseous species fails to bounce back on 

546: collision with a electrode surface, including a surface of particles 

547: already formed in the gas discharges.

548: Initially the polymer could be held to the surface very weakly,

549: %and could be easily blown away by a large gas flow, 

550: unless 

551: some additional chemical reaction takes place between the polymer 

552: atoms and atoms of the wire material. 

553: Furthermore, many free radicals are expected to have 

554: permanent or induced dipole moments; therefore,

555: electrostatic attraction to a wire could

556: play a significant role in the polymer deposition process.

557: %especially since the rates of some reactions 

558: %in the vicinity of the wire may increase

559: %quadratically with the high free radical concentration.

560: Thus, aging effects in gaseous detectors would lead to 

561: deposition of thin, very high quality polymers on the anode wires, 

562: and would result in a loss of the gas gain and degradation

563: of ionization measurements.

564: Deposits on cathodes can induce discharges by secondary

565: electron emission (or Malter effect~\cite{malter}), and would lead to 

566: chamber breakdown.

567: 

568: 

569: Traditionally, the aging rate was parameterized as a normalized gas gain 

570: loss~\cite{kadyk}:

571: 

572: \[ R = - \frac{1}{G} \frac{dG}{dQ} ( \%~per~C/cm ) \]

573: 

574: where $G$ is an initial gas gain, 

575: $dG$ is the loss of gas gain after collected charge $Q$ per

576: unit length.

577:  The parameterization used above implicitly assumes that the gain drop

578: depends linearly on the charge collected per unit length of the wire, and 

579: that the aging rate is only a function of the total collected charge,

580: independent of gas gain and radiation intensity. In reality, this

581: assumption is not fulfilled.

582:  Several examples in the literature~\cite{juri,kott,opensh1,algeri,cern1} 

583: clearly indicate that the degree of aging for a given 

584: radiation dose could also depend upon the mode of operation,

585: being larger for smaller current densities, other conditions

586: being held constant. These observations raise a question on the

587: relevance of aging results, obtained in accelerated aging tests

588: with extremely high current densities, extrapolated to  

589: normal conditions.

590:   Moreover, aging studies of ATLAS muon drift tubes,

591: %irradiated over a large area, 

592: have shown a strong dependence of the aging

593: rate on the mode of operation (irradiation rate, high voltage)

594: and size of the irradiated area~\cite{dissert}.

595: Furthermore, it has been found that

596: the aging rate in the HERA-B honeycomb tracker chambers varies

597: substantially, when changing from $\gamma$ irradiation

598: %by orders of magnitude, when changing from $\gamma$ irradiation 

599: to heavy ionizing particles~(see section 3.1). 

600:  This paper presents evidence  that the 

601: aging rate for irradiation with 5.9~keV $X$-rays and 

602: 100~MeV $\alpha$'s may differ by more than two orders of 

603: magnitude~(see sections 4 and 5). 

604: Furthermore, it was recently reported in~\cite{lyon} and 

605: it will be shown in a subsequent paper~\cite{hadr}, that

606: the rate of polymer deposition could also be affected by the 

607: mode of operation, size of the irradiated area and 

608: water addition.

609: 

610: 

611: % This paper also presents an evidence  that the 

612: %rate of polymer deposition could be affected by the 

613: %mode of operation and by the size of irradiation area~(see sections 4,5,6). 

614: 

615:  These results exclude the possibility to extrapolate aging

616: rates obtained in the laboratory setups by irradiating

617: a small region of the wire to the large areas of

618: irradiation in hadronic environment. Therefore, for each 

619: particular case, the final decision for long-term 

620: stability of operation should be taken only after obtaining similar 

621: performance in conditions as close as possible to real ones.

622: 

623: \subsection{ Results from wire chambers operation. }

624: 

625: Over the last few decades, an impressive variety of

626: experimental data has been accumulated and it is not feasible

627: to go through all published results. In addition, new 

628: results make doubtful the quantitative comparison of

629: aging properties, realized at very different conditions.

630: Therefore, we will present below only a brief summary of the most recent

631: data, focusing on the binary and ternary mixtures, containing

632: $Ar$, $CH_4$, $CO_2$ and $CF_4$.

633: 

634: There are a lot of reports 

635: that clearly indicate premature aging in $Ar/CH_4$ filled

636: wire chambers exposed to intense 

637: radiation~\cite{juri,tsarnas,smith,kollef,kwong}.

638:  The polymerization of $CH_4$ is attributed to the

639: hydrogen deficiency of radicals and their ability to make bonds 

640: with hydrocarbon molecules~\cite{kadyk,yasuda}, leading 

641: to the buildup of deposits

642: on the anode wires, which usually contain carbon and lighter

643: elements. 

644: 

645: Attempts were made to replace the organic quenchers with more

646: stable ones, such as $CO_2$. 

647: However, the gradual decomposition of $CO_2$ in $Ar/CO_2$ mixtures

648: also takes place and stable operation 

649: was found to be possible up to charges of 

650: 0.5-0.7~$\frac{C}{cm~wire}$~\cite{vavra,atlas}. 

651: Furthermore, carbon dioxide is also known for its

652: carbon buildup, which occurs specifically at the 

653: cathode~\cite{co2}, however, this carbon layer

654: did not affect the performance of the drift

655: tubes~\cite{atlas}.

656: Another effect, which is enhanced in $Ar/CO_2$ mixtures 

657: compared to gases containing hydrocarbon quenchers,

658: is the spurious counting rate~(noise), that presumably comes

659: from the spikes on the wires. This noise

660: can be cured by the

661: addition of water vapor to the drift gas~\cite{atlas}.

662: 

663: 

664: The aging studies in the ternary mixtures $Ar/CH_4/CO_2$

665: indicate the dependence of the aging rate on the 

666: particular set of operating conditions,

667: in some cases being smaller for larger $CH_4$ content 

668: in the mixture~\cite{kott,atlasnote,sadr}. 

669: The stable operation far beyond 0.6~$\frac{C}{cm~wire}$ 

670: was observed for $Ar/CH_4/N_2/CO_2$ (94:3:2:1) 

671: + 1200~$ppm$ of $H_2 O$ mixture, however,

672: some conditions where found were only a tenth of 

673: nominal lifetime was achieved~\cite{kollef}. Moreover,

674: for the $Ar/CH_4/N_2/CO_2$ mixture, measurements have shown 

675: that their sensitivity to aging decreases with 

676: decreasing $CH_4$ and increasing $CO_2$ content.

677: 

678: The use of gases having $CF_4$ as a component was found to be  

679: attractive in the plasma processes, since the $CF_4$

680: is an ideal source for a variety of reactive neutral and ionic 

681: fragment atoms and molecules,

682: and especially neutral fluor, which is a  

683: desirable active species in etching processes~\cite{mogab}.

684: The $CF_4$ molecule is 

685: %a attractive gas component because it is 

686: relatively inert in its electronic ground state, but

687: all excited electronic states of $CF_4$ and $CF_4^{+}$ ion

688: lead to dissociation.

689: Below the onset of electronic excitations, which is rather

690: high - 12.5~eV, collisions of electrons with $CF_4$ molecule 

691: lead to elastic scattering, vibrational excitation and 

692: dissociative attachment. Above this energy, 

693: dissociation of the $CF_4$ into neutrals or charged

694: fragments becomes significant~\cite{cf42,cf41}.

695: Actually, in plasma environment $CF_4$-based gases are used for 

696: both etching and deposition processes, 

697: the distinction being made by the gas and its amount with which $CF_4$ is mixed. 

698: The balance between these processes is called 'competitive 

699: ablation and polymerization'.

700: In general, the addition of

701: oxygenated species shifts the chemistry of $CF_4$ plasmas toward

702: etching, while the addition of hydrogenated species shifts the 

703: chemistry toward polymerization~\cite{chemmod,yasuda,kushner,winters}.

704:  For example, the polymer formation by $CF_4$ plasmas 

705: in the presence of $CH_4$ was observed~\cite{yasuda}.

706: 

707:  In the wire chamber operation, the usage of $CF_4$ is 

708: also quite controversial. 

709:  Many studies have demonstrated excellent aging properties, 

710: up to 10~$\frac{C}{cm~wire}$,

711: of $CF_4/i C_4 H_{10}$ (80:20) avalanches~\cite{kadyk1,opensh}, 

712: which also 

713: has an ability to etch both silicon-based~\cite{kadyk,opensh} and

714: hydrocarbon~\cite{opensh} deposits from the previously aged 

715: gold-plated wires. 

716: However, another proportions of $CF_4$ 

717: and isobutane could totally change the chemical reactions on the surface

718: of the anode wire.

719: Heavy carbonaceous deposits were formed on the gold-plated

720: wires irradiated in the mixtures $CF_4/i C_4 H_{10}$ (95:5) and

721: $CF_4/i C_4 H_{10}$ (20:80). The absence

722: of fluorocarbon deposits on the anode wires 

723: indicates that the deposits were formed only from $i C_4 H_{10}$,

724: without incorporation of $CF_x$ fragments.

725: Details of the chemical model and aging results in the 

726: $CF_4/i C_4 H_{10}$ are presented elsewhere~\cite{chemmod}.

727: Another gas observed to form heavy carbonaceous anode deposits

728: was $CF_4/C_2 H_4$(95:5)~\cite{chemmod}. At the same time,

729: the 4 $\%$ addition of $CF_4$ has been shown to inhibit anode damage

730: in $Ar/C_2H_6$(50:50) mixture up to the doses of

731: 1.5~$\frac{C}{cm~wire}$~\cite{opensh1}. Aging studies have been

732: also performed with the $CF_4/CH_4$ (90:10) gas mixture 

733: %for the D0 muon test chamber and 

734: and no deterioration was found up to 1.9~C/cm of the wire~\cite{d0}.

735:   Finally, it should be noted, that most of the aging studies in the

736: $CF_4$/hydrocarbon mixtures, were performed by irradiating a small region of the wire 

737: under well controlled laboratory conditions.

738:  At the same time, several painful experiences with aging 

739: are known from the high energy experiments.

740: The D0 muon drift chambers, 

741: filled with $Ar/CF_4/CH_4$(80:10:10) undergo a fast aging when

742: operated in a radiation environment. 

743: Vapors from a glue used in the construction were deposited on the wires

744: in a sheath, with the deposition rate proportional to accumulated charge~\cite{d0note}.

745: Furthermore, operation in the high-rate hadronic HERA-B environment

746: showed that the aging rate could depend on the type of 

747: ionization and size of the irradiated area~\cite{hohlmann,kolanoski}.

748: The severe anode and cathode aging effects

749: were observed in the honeycomb chambers tested in the

750: HERA-B conditions with $CF_4/CH_4$(80:20) and $Ar/CF_4/CH_4$(74:20:6) mixtures

751: after the radiation dose of several $mC$ per $cm$ of the wire.

752: This effect was surprising since the

753: chambers were proved to be immune a very large $X$-ray doses, up to

754: the 5~C/cm of the wire. 

755:  

756:  The aging properties of the gas mixtures 

757: $Ar(Xe)/CF_4/CO_2$, which by analogy with plasma experiments

758: should provide etching environments in the wire avalanches,  

759: have been also widely investigated over the past years.

760: The radiation tests of the ATLAS straw tubes using 

761: $Xe/CF_4/CO_2$ mixture showed no signs of aging 

762: for a integrated charge of up to 8~$C/cm$~\cite{akesson,straw}. 

763: However, the

764: detailed examination of anode wires and straw walls (cathode)

765: after irradiation had shown evidence for the etching

766: processes on both surfaces~\cite{romaniouk}.

767:  Another interesting phenomena was found in straw tubes

768: irradiated under exceedingly high current densities in $Xe/CF_4/CO_2$ mixture.

769: %with a $Sr^{90}$ source in $Xe/CF_4/CO_2$ mixture

770: %under exceedingly high current densities.

771: After an accumulated charge of 9~$C/cm$ the 

772: `anode swelling effect' was observed,

773: where destruction of gold coating of the gold-plated tungsten wires, 

774: followed by the tungsten oxidation~\cite{kriv1,kriv2}.

775:  Recently, it was also reported~\cite{hohlmann,kolanoski}, that 

776: under a high radiation intensity in $\afo$ (65:30:5) mixture, 

777: gold-plating of tungsten wires

778: can be etched away completely exposing the tungsten. 

779: However, the addition of 

780: water to the $\afo$ (65:30:5) was found to reduce the rate of 

781: etching processes, and the honeycomb chambers were able to survive 

782: radiation doses of up to 2~$\frac{C}{cm~wire}$.

783: The chemical etching processes in the $\afo$ 

784: are likely to be responsible for the

785: enhanced removal of the diamond coating observed in 

786: MSGC-GEM chambers~\cite{beauty}. 

787: 

788: Amazingly, extremely rapid aging has been observed 

789: for pure $CF_4$ and in $Ar/CF_4/O_2$(50:40:10) mixture,

790: which were expected to have strong etching abilities~\cite{chemmod,coyle}.

791:  In the former case, the aging process was related to chemical processes 

792: at the cathode, where trace fluorocarbon deposits were found.

793:  This type of aging phenomena was found to be cathode-material-dependent,

794: and resulted in a loss 

795: of gas gain rather than in a self-sustained discharge.

796: For the latter case, although $C$, $O$, and $F$ elements were observed 

797: on the anode wire, it is likely that the rapid aging was as a result of

798: the same (cathode phenomenon) observed in pure $CF_4$. 

799: The fluorination 

800: of the anode wires, found after exposure in $Ar/CF_4/O_2$(50:40:10) avalanches,

801: can cause anode aging in addition to cathode effects~\cite{chemmod}.

802:  A further phenomena due to the consistency of the cathode surfaces

803: was observed in ATLAS aluminum drift tubes operated in 

804: $Ar/CF_4/N_2/CO_2$ (94.5:0.5:2:3). Here, no change in gas gain 

805: for the tubes was found up to an accumulated charge of 

806: 5~$\frac{C}{cm~wire}$.

807: However, the increased count rate at the beginning of the tests,

808: due to afterpulses in this gas at high gains

809: decreased with the time to the level, when no significant

810: afterpulsing was observed. 

811: This reduction was suggested to be due to the increased layer

812: of fluor and oxygen found at the cathode~\cite{dissert}.

813:  Similar effects were also found during the long-term irradiation of drift tubes 

814: with stainless steel cathodes in pure $CF_4$~\cite{denisov}.

815: 

816: These aging results clearly demonstrate that

817: the presence of large amounts of $CF_4$ in the mixture does not

818: necessarily ensure good aging properties. 

819: Moreover, perfluorocarbons represent the most extreme case of ablation

820: competing with polymer formation.

821: Therefore, all attempts

822: to improve the aging performance of the mixtures by adding $CF_4$

823: must be tested in the actual environment.

824: 

825: An abundance of literature exists describing the effect of certain additives

826: with oxygen-containing groups, like water, alcohol, methylal, oxygen.

827: These components, added to the mixture in small concentrations,

828: usually (but not always) extend the chamber lifetime,

829: especially against the polymerization of hydrocarbons.

830:  Of special interest is the addition of water vapors,

831: that has been found to suppress effectively polymerization in

832: wire chambers~\cite{kollef,charm}, to prevent Malter breakdowns~\cite{straw1},

833: or even to restore the original operation in aged 

834: counters~\cite{algeri,argus}; a few hundred to a few thousand $ppm$ of water

835: are generally used. The most natural mechanism by which water impedes Malter 

836: breakdowns and significantly improves the analog response of the

837: aged counters comes from the fact that water tends to increase the

838: conductivity of the damaged electrodes~\cite{kadyk,algeri}. 

839: In addition, water could

840: effectively suppress the secondary photon-mediated phenomena, if

841: the UV photons from carbon excitations are responsible for the

842: photoionization at the cathode~\cite{vavra2}. At the same time, the

843: exact mechanism by which water stops or reduces the aging rate 

844: in wire chambers is not clearly understood. One possible explanation

845: could be gained from the similarities with plasma chemistry,

846: where water acts like a blocking agent of the chain growth

847: mechanism, reacting with polymer precursors, thereby decreasing the

848: concentration of free radicals available for polymerization~\cite{yasuda}.

849:  Another explanation is that water, due to its large dipole moment,

850: could effectively slow-down electrons in the avalanches 

851: making the mixture `cooler'. In this case, aging effects resulting from free

852: radical polymerization would be reduced, since fewer

853: radicals would be formed in the wire avalanches. 

854:  Finally, it is well established, that

855: aging effects in the wire chambers, even when they are entirely due to

856: polymerization of gas mixture components can be accelerated by the

857: discharges~(glow discharges of sparks) 

858: and photoelectric feedback~\cite{vavra2,atac}. 

859: These effects are strongly dependent on the quality and type of the 

860: electrodes: surfaces of anode wire or cathode could be rather 

861: imperfect, thus, `triggering' polymerization in wire chambers. 

862: Known remedies for aging in this case are also water and alcohols,

863: which may prolong the lifetime considerably.

864: Furthermore, aging can be 

865: initiated by the trace contaminants such as $Si$, halogens~(other than $CF_4$),

866: sulfur, phthalats, benzine rings, which are either initially present in the 

867: gas, or result from outgassing of solid materials in contact with gas.

868: The detailed summary of the influence of

869: commonly used materials on aging properties 

870: may be found in ~\cite{kadyk,romaniouk}. 

871: When constructing large systems, 

872: one should before to certify the gas purity and chemical reactivity

873: of various materials in order to avoid the presence of 'bad' molecules 

874: in contact with the active gas volume.

875: %Therefore, for the new high-intensity experiments, all precautions should be taken 

876: %in order to avoid the presense of 'bad' malecules or materials in contact

877: %with the active gas volume.

878: 

879: 

880: 

881: \section{ Aging results from laboratory tests.}

882: 

883:  We have carried out a laboratory study of the aging properties

884: of $\afh$ (74:20:6) mixture, when exposed to an intensive $Fe^{55}$ source, 

885: and of $\afh$ (67:30:3) and $\afo$ (65:30:5) mixtures, both irradiated

886: with a $Ru^{106}$ source, up to 

887: a total radiation dose of 2~$C/cm$ of the wire.

888:  In each test, a single cell  

889: of the proportional tube chamber was irradiated through 

890: a hole in the cathode wall, closed  with thin $Al$ foil,

891: in the center of the 50~cm long cell.

892:  The irradiated area was limited by a collimator

893: to about 30~mm of the wire for the $Fe^{55}$ source 

894: and to 10~mm for the $Ru^{106}$ source.

895: While the $Fe^{55}$ 5.9~keV $X$-rays produce

896: an ionization cluster in the chamber cell, 

897: corresponding to $\sim$~200 primary electrons, the 

898: $Ru^{106}$ source provides 3.54~MeV electrons

899: from the decay of $Rh^{106}$, which leaves an ionization track

900: with $\sim$~90 primary electrons per cm. 

901:  The degree of aging  was determined quantitatively by

902: comparing $Fe^{55}$ spectra, measured at the irradiated and reference

903: positions of the wire, using the same electronics chain.

904: This method removed problems 

905: arising from the $Fe^{55}$ peak position movement due to changes in 

906: the pressure, the temperature, and gas mixture itself.

907: %since the measurements are performed at both positions under identical

908: %conditions.

909:  The total collected charge was

910: determined by integrating continuously the recorded anode current.

911: The gas flow rate was 1~liter/hour for all tests.

912: 

913: \setlength{\unitlength}{1mm}

914: \begin{figure}[bth]

915:  \begin{picture}(160,110)

916:  \put(-15.0,-20.0){\special{psfile=fig5.eps vscale=58 hscale=56 } }

917:  \put(70.0,-20.0){\special{psfile=fig6.eps vscale=58 hscale=56 } }

918:  \put(11,20){Figure 6: Ratio of $Fe^{55}$ peak positions }

919:  \put(12,14){ at irradiated and reference spots as a}

920:  \put(14,8){ function of the accumulated charge. }

921:  \put(102,20){Figure 7: Measured dependence of a}

922:  \put(106,14){self-sustained dark current as a }

923:  \put(106,8){ function of gas flow, after the }

924:  \put(104,2){ removal of the irradiation source. }

925:  \end{picture}

926: %\caption[kk]{\em }

927: %  \label{fig:point2}

928: \end{figure}

929: 

930: A standardized set of conditions was established for the $Fe^{55}$

931: irradiation of the $\afh$ (74:20:6) mixture, corresponding to a current of

932: 0.7~$\mu A/cm$ at a HV=2.65~kV. 

933: %and leading to a mean avalanche size of $Q \sim 3 \times 10^{8}$.

934: For irradiation with the $Ru^{106}$ source, the high voltages for

935: operation in $\afh$ (67:30:3)~($\afo$ (65:30:5)) mixtures

936: were set to 2690~V (2790~V), corresponding to current densities

937: of 1.2~$\mu A$ (0.9~$\mu A$).

938: % and leading to mean avalanche sizes

939: % $Q \sim 6 \times 10^{7}$ ($Q \sim 4 \times 10^{8}$) respectively.

940:  The different current densities for the 

941: irradiation of $\afh$ (67:30:3) and $\afo$ (65:30:5) mixtures

942: is attributed to the reduced source strength for the latter case,

943: due to the larger thickness of $Al$ foil used 

944: as a cathode above the irradiated region of the wire.

945:  The high voltages used during the aging studies already exceed

946: the proportional mode of gas amplification, 

947: therefore, events with higher pulse heights than those in the

948: proportional mode also appear at these voltages.

949: % Moreover, the higher space-charge 

950: %effects induced by the 5.9~keV $X$-rays,

951: %compared with electrons from $Ru^{106}$,

952: %could result in a larger fraction of the the registered events 

953: %already in the self-quenching streamer~(SQS) mode

954: %for the irradiation with $Fe^{55}$ source.

955: % The difference in chamber behaviour ....

956: 

957: % The higher space-charge effects created by the $Fe^{55}$ source

958: %compared with $Ru^{106}$ promotes the transition from the

959: %proportional to the self-quenching streamer~(SQS) mode, and could

960: %result in a larger number of the the registered events 

961: %already in the SQS mode for the former case.

962: 

963: Fig. 6 shows the ratio of $Fe^{55}$ peak positions, measured 

964: at irradiated and reference spots as a function of the accumulated charge. 

965:  For the $\afh$ (74:20:6) mixture,

966: the relative mean amplitude $\frac{A_{irrad}}{A_{reference}}$

967: monotonously dropped to about 85~$\%$ of the initial value,

968: after the radiation dose of 2~$C/cm$ of the wire,

969: corresponding to $R \sim 8~$ $\%$ /~$C/cm$. 

970: %Despite these degradings,

971: Such a rate of gas gain losses could be acceptable  for the long-term

972: operation of the muon chambers in HERA-B.

973: However, after a collected charge $\sim$ 1~$C/cm$ 

974: significant self-sustained dark current 

975: appeared in addition to those normally associated with radiation.

976:  This dark current persisted even after the removal of

977: the radiation source, but it 

978: was totally quenched, either by a large increase in the gas flow rate, 

979: from 1~liter/hour to 10~liters/hour~(see Fig. 7), 

980: or by turning off the high voltage.

981:  This phenomena is associated with the polymer coating

982: and subsequent charge buildup on the cathode surface,

983: by a mechanism similar to the well known Malter discharge~\cite{malter}.

984: 

985: Irradiation with the $Ru^{106}$ source of 

986: $\afh$ (67:30:3) and $\afo$ (65:30:5) mixtures

987: led to $R$ values consistent with  zero~(see Fig. 6).

988: Such promising aging behavior make these gases suitable candidates

989: for use in the tube chambers. Unfortunately, as will be reported 

990: in section 5, the aging results, obtained in the high-rate

991: hadronic environment when irradiating a large area chamber differ 

992: significantly from those described above.

993: 

994: \section{ Aging studies in a 100~MeV $\alpha$-beam }

995: 

996: \subsection{ Aging in $\afh$(74:20:6) gas mixture.}

997: 

998:  After the tests with radioactive sources,

999: aging studies with the  $\afh$ (74:20:6) mixture were carried out 

1000: by irradiating a single layer tube chamber 

1001: in a 100~MeV $\alpha$-beam in Karlsruhe~(see Fig. 8).

1002:  Due to the small range of the 100~MeV $\alpha$'s  

1003: in aluminum, the thickness of the $Al$-wall 

1004: was reduced to $\sim$ 200~$\mu m$.

1005:  The $\alpha$-beam intensity was 

1006: almost uniform over the beam area 8~$\times$~8~$cm^2$.

1007:  This allowed a constant

1008: irradiation of four 

1009: wires fully, and, in addition, two

1010: wires were exposed over half of their cell width. 

1011: All wires were equipped for analog readout.  

1012: Furthermore, for one of wires, which we will call a `reference wire',

1013: two holes were made in the chamber wall - within the region

1014: of irradiation and outside of it. Both holes

1015: were closed with kapton foil.

1016: For the `reference wire', it was possible

1017: to measure the degree of aging by comparing the $Fe^{55}$

1018: spectra at the two positions of the same wire.

1019:  

1020: A premixed gas mixture of $\afh$(74:20:6) 

1021: was transported by stainless-steel tubes connected directly

1022: to the tube chamber inlet and outlet. The gas flow was in serial

1023: inside a chamber from one cell to another, 

1024: with the flow rate about 6~liters/hour.

1025:  The operating voltage was set to 2.35 kV, corresponding to a

1026: gas gain $\sim 4 \times 10^{4}$, measured with $Fe^{55}$ $X$-rays.

1027:  The radiation intensity varied during the aging studies 

1028: by a factor of 3, thus excluding any possibility to

1029: investigate aging behavior as a function of rate.

1030:  Therefore, the current densities were in the range 

1031: $\sim$~250~nA/cm up to 750~nA/cm.

1032: By monitoring the chamber current, the total collected charge

1033: was determined.

1034: 

1035:  Since the average energy losses of $\alpha$-particles 

1036: in the chamber cell are much larger than those of $X$-rays, 

1037: the same avalanche size as that for $X$-rays is obtained 

1038: at a lower voltage. 

1039: In practice, at an applied voltage of 2.35~kV,

1040: the total charge released from $\alpha$-particles  

1041: was measured to be a factor of 10-15 

1042: larger than that corresponding to a 5.9~keV $X$-ray

1043: localized energy deposition in proportional mode.

1044: Moreover, the signals from $\alpha$'s at this voltage

1045: were also accompanied by self-quenching streamer discharges, 

1046: observed in several percent of the events at this voltage.

1047:  This may be explained by the fact that since the transformation

1048: from proportional to streamer mode depends upon the primary

1049: ionization density~\cite{gang}, heavily ionizing particles should enter 

1050: the self-quenching streamer mode at lower voltages than $Fe^{55}$ $X$-rays

1051: and minimum ionizing particles.

1052: 

1053: \setlength{\unitlength}{1mm}

1054: \begin{figure}[bth]

1055:  \begin{picture}(160,110)

1056:  \put(-40.0,-37.0){\special{psfile=fig7.eps vscale=80 hscale=80 } }

1057:  \put(75.0,-20.0){\special{psfile=fig8.eps vscale=58 hscale=56 } }

1058:  \put(11,20){Figure 8: Sketch of the tube chamber }

1059:  \put(12,14){ used in the $\alpha$-beam in Karlsruhe. }

1060: % \put(0,-2){ function of the accumulated charge. }

1061:  \put(102,20){Figure 9: Singles rate curves for the  }

1062:  \put(102,14){ `reference wire' in the irradiated region  }

1063:  \put(102,8){ (stars), after exposure to 60~$\frac{mC}{cm~wire}$,}

1064:  \put(108,2){ and outside of it (circles). }

1065:  \end{picture}

1066: %\caption[kk]{\em }

1067: %  \label{fig:point2}

1068: \end{figure}

1069: 

1070:  

1071: After the first 24 hours of irradiation, which resulted in a 

1072: collected charge of $\sim$~50~$\frac{mC}{cm~wire}$,

1073: the status of the `reference wire' was checked 

1074: by recording the $Fe^{55}$ pulse-height spectra.

1075: Severe anode aging was observed for this wire

1076: in the irradiated area, corresponding to a normalized gas gain

1077: loss value $R \sim 750~\%~per~C/cm $, while the $Fe^{55}$

1078: spectra in the non-irradiated region remained unchanged.

1079: The downward shift of the peak value was accompanied by a

1080: distortion and broadening of the $Fe^{55}$ pulse-height spectrum.

1081:  After the next 6 hours of exposure, when an additional 

1082: 10~$\frac{mC}{cm~wire}$ were collected,  

1083: a steady decrease of the  

1084: gas gain for the `reference wire' has been found, and the $R$ value 

1085: became even slightly higher $\sim 850~\%~per~C/cm $.

1086:  Furthermore, during these hours,

1087: sizeable increase in the single counting rate,

1088: above the actual $\alpha$-beam intensity,

1089: was observed from several of the irradiated wires. 

1090: Since that time, the irradiation of the `reference wire' was 

1091: halted, but, the aging studies with all other wires were continued 

1092: until the collected charge 

1093: had reached a value of 280~$\frac{mC}{cm~wire}$.

1094: Two wires, irradiated over half of their cell width, accumulated

1095: half of this charge - 140~$\frac{mC}{cm~wire}$.

1096: The aging behavior of the wires during irradiation was quite similar:

1097: large increase in the single counting rate, as a result of

1098: spark discharges,

1099: was observed from all wires at an applied voltage of

1100: 2.35~kV, together with the monotonic

1101: drop in the chamber current.

1102: 

1103: %\setlength{\unitlength}{1mm}

1104: %\begin{figure}[bth]

1105: % \begin{picture}(160,110)

1106: % \put(-50.0,-37.0){\special{psfile=fig7.eps vscale=80 hscale=80 } }

1107: % \put(65.0,-20.0){\special{psfile=fig8.eps vscale=58 hscale=56 } }

1108: % \put(-3,20){Figure 8: Experimental setup at the }

1109: % \put(-2,14){ $\alpha$-beam in Karlsruhe. }

1110: % \put(0,-2){ function of the accumulated charge. }

1111: % \put(88,20){Figure 9: Singles rate curves for the  }

1112: % \put(87,14){ `reference wire' in the irradiated region  }

1113: % \put(88,8){ (stars) and outside of it (circles) }

1114: % \put(90,2){ after exposure to 60~$\frac{mC}{cm~wire}$. }

1115: % \end{picture}

1116: %\caption[kk]{\em }

1117: %  \label{fig:point2}

1118: %\end{figure}

1119: 

1120: \setlength{\unitlength}{1mm}

1121: \begin{figure}[ph]

1122:  \begin{picture}(160,200)

1123:  \put(40.0,160.0){\special{psfile=fig10.ps vscale=45 hscale=48  } }

1124:  \put(40.0,85.0){\special{psfile=fig11.ps vscale=45 hscale=48 } }

1125:  \put(10.0,15.0){\special{psfile=fig12.ps vscale=36 hscale=36 } }

1126:  \put(93.0,3.0){\special{psfile=fig13.ps vscale=30 hscale=35 } }

1127:  \put(15,153){Figure 10: SEM of the 'reference wire' after the accumulated charge of 60~$\frac{mC}{cm~wire}$.}

1128:  \put(15,147){ $C$ and $F$  

1129: elements were identified together with Au signal from the wire material.}

1130: % \put(0,138){ together with Au signal from the wire material.}

1131:  \put(15,83){Figure 11: SEM of the cathode surface composition in the irradiated region }

1132:  \put(35,77){ of the wire after exposure to 280~$\frac{mC}{cm~wire}$. }

1133:  \put(13,8){Figure 12: Typical deposits on the   }

1134:  \put(9,2){ anode wire after exposure to 280~$\frac{mC}{cm~wire}$.}

1135: % \put(0,-12){ 240~$\frac{mC}{cm~wire}$. }

1136:  \put(98,8){Figure 13: Wire irradiated over half }

1137:  \put(98,2){ of its cell width. Deposits built}

1138:  \put(98,-4){up on one side of the wire towards }

1139:  \put(98,-10){all of the primary electrons drifted. }

1140:  \end{picture}

1141: %\caption[kk]{\em }

1142: %  \label{fig:point2}

1143: \end{figure}

1144: 

1145: After the aging studies in Karlsruhe, the chamber characteristics

1146: have been studied on the 3~GeV electron beam with intensity

1147: $\sim$~1~$kHz/cm^{2}$.

1148: Fig. 9 shows the singles rate curves for the `reference wire'

1149: in the irradiated region, after exposure to 

1150: 60~$\frac{mC}{cm~wire}$, and outside of it.

1151:  From these curves it is seen that, as result of irradiation, 

1152: there is a shift upwards of the 

1153: lower end of the efficiency plateau~(showing a decrease of gain

1154: at a given voltage) with a

1155: consequent reduction in effective plateau length. 

1156:  All other irradiated wires were also studied on the electron beam.

1157: Wires that have been exposed to a charge

1158: of 280~$\frac{mC}{cm~wire}$ were found to be heavily damaged,  

1159: showing an efficiency at the level of ~$\sim$1~$\%$-10~$\%$

1160: for the nominal operating high voltage 2.35~kV.

1161:  In addition, for all damaged wires the breakdown knee

1162: in the singles rate curves significantly displaced towards lower voltages.

1163: This effect can also be seen for the `reference wire' in Fig. 9. 

1164: For the damaged region of this wire,

1165: the breakdown point occurs

1166: at HV=2.45~kV, where the single counting rate increases abruptly.

1167: At the same time, whenever the electrons illuminates regions of the 

1168: wire that have not been previously irradiated, no actual difference

1169: in the singles rate curves (before and after irradiation) shows up.

1170: 

1171: %In addition, whenever damaged regions of the wires were illuminated with 

1172: %electrons, the spurious counting rates (`noise'), ranging from 50~kHz to 500~kHz,

1173: %were observed for all aged wires. 

1174: %This effect can be seen for the `reference wire' in Fig. 9. For this wire, 

1175: %starting from HV=2.45~kV, spurious spark discharges 

1176: %dominate in the single counting rate,

1177: %thus, shifting the breakdown knee towards lower voltages.

1178: % The onset of spurious counting rates

1179: %was every time associated with the appearance of the chamber current 

1180: %$\sim~0.1~-~0.2~\mu A$/wire. 

1181: %At the same time, 

1182: %in the absence of electron beam or whenever the electrons illuminated 

1183: %regions of the wire that have not been previously damaged,

1184: %noise counts together with currents were not observed. 

1185: 

1186: 

1187: %\setlength{\unitlength}{1mm}

1188: %\begin{figure}[ph]

1189: % \begin{picture}(160,200)

1190: % \put(50.0,145.0){\special{psfile=fig10.ps vscale=25 hscale=30  } }

1191: % \put(30.0,55.0){\special{psfile=fig11.ps vscale=40 hscale=48 } }

1192: % \put(10.0,-5.0){\special{psfile=fig12.ps vscale=30 hscale=30 } }

1193: % \put(90.0,-10.0){\special{psfile=fig13.ps vscale=35 hscale=35 } }

1194: % \put(0,148){Figure 10: SEM of the 'reference wire' after the accumulated charge of 60~$\frac{mC}{cm~wire}$.}

1195: % \put(0,142){ $C$ and $F$  

1196: %elements were identified together with Au signal from the wire material.}

1197: % \put(0,138){ together with Au signal from the wire material.}

1198: % \put(0,78){Figure 11: SEM of the cathode surface composition in the irradiated region }

1199: % \put(20,72){ of the wire after exposure to 280~$\frac{mC}{cm~wire}$. }

1200: % \put(-3,0){Figure 12: Typical deposits on the   }

1201: % \put(-4,-6){ anode wire after exposure to 280~$\frac{mC}{cm~wire}$.}

1202: % \put(0,-12){ 240~$\frac{mC}{cm~wire}$. }

1203: % \put(88,0){Figure 13: Wire irradiated over half }

1204: % \put(88,-6){ of its cell width. Deposits built}

1205: % \put(88,-12){up on one side of the wire towards }

1206: % \put(88,-18){all of the primary electrons drifted. }

1207: % \end{picture}

1208: %\caption[kk]{\em }

1209: %  \label{fig:point2}

1210: %\end{figure}

1211: 

1212: After the beam tests, 

1213: the chamber was opened for inspection and both

1214: anode wires and cathodes surfaces were analyzed, using 

1215: microscope and later scanning electron microscopy~(SEM).

1216:  Observation of all damaged wires   

1217: under a microscope show vertically structured black deposits  

1218: or 'whiskers'~(see Fig. 12), distributed

1219: randomly within the irradiated area only,

1220: and of up to lengths of 50~$\mu m$ from the wire surface. 

1221: The 'whiskers' were found to be very fragile,

1222: and easily broken if touched.

1223:  Of special interest, there were two wires at the boundaries of the

1224: irradiated area where only half of the cell width was exposed 

1225: to $\alpha$'s. These wires showed significant asymmetries in 

1226: the 'whiskers'~(see Fig. 13) with 

1227: deposits built up on one side of the anode wire towards

1228: all of the primary electrons drifted. 

1229:  The observed asymmetries in the anode deposits indicate that 

1230: even for high ionization densities produced by $\alpha$-particles,

1231: the operating parameters (high voltage, gas mixture) 

1232: led rather to avalanches  

1233: well localized on one side of the anode wire rather than to avalanches 

1234: whose spread was large enough to surround the wire completely. 

1235: 

1236:  Examination of all damaged wires by SEM

1237: revealed deposits, containing carbon and fluorine 

1238: as the only detectable elements,

1239: thus confirming the presence of a polymer coating. 

1240: Moreover, carbon and fluorine signals were also identified in the

1241: irradiated regions free of 'whiskers',

1242: together with Au signal from the wire material.

1243:  Fig. 10 shows, as an example, the SEM of the `reference wire' after the 

1244: accumulated charge of 60~$\frac{mC}{cm~wire}$.

1245:  New wires and wires which were exposed to the irradiated gas,

1246: but not irradiated themself, showed peaks only from Au.

1247: Unfortunately, this method is unable to detect hydrogen,

1248: so hydrocarbons registered as a carbon alone and are also 

1249: trapped in the polymer coating.

1250:  SEM analysis of the cathode surface composition  

1251: in the irradiated region of the wire after exposure to

1252: 280~$\frac{mC}{cm~wire}$ was also performed.

1253: Translation of peak areas 

1254: into atomic abundances indicates that the top layer of the 

1255: cathode consists mainly

1256: of $Al$($>80~\%$) with some traces of carbon and oxygen~(see Fig. 11).

1257: 

1258:  A direct means of investigating the radiation-initiated spark discharges

1259: was to separate anode and cathode effects by combining new and aged

1260: portions of the tube chamber. When the cathodes of the tubes

1261: that had been previously aged were restrung with the new anode wires,

1262: the breakdown knee in the singles rate displaced to the higher voltages.

1263: %there was no longer an abrupt transition to spurious spark discharges.

1264: Therefore,

1265: one reason for the discharges are some local

1266: deposits on the anode wires~(conductive `whiskers').

1267: %the noise counts can be explained 

1268: %as being due to the spikes on the anode wires.

1269: It was reported by Holland~\cite{book3}, that the

1270: properties of polymer films can be electrically

1271: conductive or insulating according  to the carbon to hydrogen ratio

1272: and a degree of cross-linking.

1273:  If a coating is primarily carbon, 

1274: rather than normal polymers, the conductivity will be much higher.

1275: At the same time, the energy and 

1276: amount of particle bombardment on the surface can very easily change

1277: the composition and therefore conductivity. 

1278: Hydrogen can be easily

1279: removed by ion bombardment, so heavy bombardment leads to a more

1280: carbonaceous than polymer-like deposit, which may be relevant

1281: for the case of irradiation with $\alpha$-particles.

1282: 

1283: % During the  process of polymer deposition, 

1284: %the massive channeling of polymer chains to the least

1285: %resistive regions would produce vertically structured deposits

1286: %that would be identified as 'whiskers'.

1287: 

1288: % The results presented in sections 4 and 5, 

1289: %demonstrate that the aging rate for irradiation with 5.9~keV $X$-rays and 

1290: %100~MeV $\alpha$'s may differs by two orders of magnitude.

1291: % Despite the negligible gas gain loss  

1292: %measured after the long-term irradiation with $Fe^{55}$

1293: %$X$-rays, rapid aging effects, observed with $\alpha$-particles, 

1294: %completely ruled out $\afh$ (74:20:6) mixture as a possible 

1295: %candidate for the operation in the HERA-B high radiation environment, 

1296: %where, at least, part of the ionization

1297: %may be deposited by the heavily ionizing particles.

1298: % However, as long as the data from different aging tests can not be combined

1299: %properly, it is very difficult to give an extrapolation about the 

1300: %lifetime in the real detector.

1301: % Therefore, in order to find a link between 

1302: %results in different setups, we also performed an 

1303: %aging studies in the HERA-B environment,

1304: %under conditions as similar as possible to real ones.

1305: % In a subsequent paper we will present the results

1306: %

1307: %The results from these tests are discussed in section 6.

1308: 

1309: 

1310: %\section{ Aging results in a HERA-B environment.}

1311: 

1312: %\subsection{ Aging setup. }

1313: 

1314: %\subsection{ Aging in $\afh$(67:30:3) gas mixture.}

1315: 

1316: %\subsection{ Aging in $\afo$(65:30:5) gas mixture.}

1317:  

1318: %\section{ Discussion of results. }

1319: 

1320: %\section{ Further investigations. }

1321: 

1322: 

1323: \section{ Summary and outlook. }

1324: 

1325: 

1326:  Experimental conditions at HERA-B impose very strong

1327: requirements for the gaseous detectors.

1328:  The high radiation load of the HERA-B experiment results in an accumulated

1329: charge of up to 200~$\frac{mC}{cm~wire}$ in the tube chambers. The 

1330: chamber signals, which are used to form the trigger, must be collected 

1331: in less than the 96~ns between bunch crossings.

1332: 

1333:  A short description of the criteria which are relevant for the choice of 

1334: gases in the HERA-B muon detector are discussed and 

1335: the tube chamber characteristics 

1336: studied for several $Ar/CF_4$-based mixtures are reported.

1337: However, the main criteria for the choice of the gases are the aging properties.

1338:  In this paper, the brief summary of the most recent data

1339: focusing on the binary and ternary mixtures, containing 

1340: $Ar$, $CH_4$, $CO_2$ and $CF_4$ is presented.

1341: Although, $CF_4$ gas has been widely used over the last ten years,

1342: the results from it's usage in the drift tubes are quite controversial.

1343:  The aging results reported in this paper 

1344: demonstrate that the aging rate for irradiation with 5.9~keV $X$-rays and 

1345: 100~MeV $\alpha$'s may differ by more than two orders of magnitude.

1346:  Despite the negligible gas gain loss  

1347: measured after the long-term irradiation with $Fe^{55}$

1348: $X$-rays, rapid aging effects observed with $\alpha$-particles 

1349: completely ruled out $\afh$ (74:20:6) mixture as a  

1350: candidate for operation in the HERA-B high radiation environment, 

1351: where at least part of the ionization

1352: is deposited by heavily ionizing particles.

1353:   Moreover, 

1354: from these results it is evident that from the accumulated charge alone, 

1355: it is not possible to combine the data from the different radiation sources 

1356: into one consistent model.

1357:  Therefore, as long as the data from different aging tests can not be combined

1358: properly, it is very difficult to give an extrapolation about the 

1359: lifetime in the real detector.

1360:  In order to find a link between 

1361: results in different setups, we also performed

1362: aging studies with $\afh$ (67:30:3) 

1363: and $\afo$ (65:30:5) mixtures in the HERA-B environment,

1364: under conditions as similar as possible to real ones.

1365:  In a subsequent paper~\cite{hadr} we will report on the results from these tests, 

1366: which indicate that the aging rate could also be affected by

1367: the mode of operation, size of the irradiated area and water addition.

1368: 

1369: 

1370: %\section{ Further investigations. }

1371: 

1372: 

1373: 

1374: 

1375: \section*{Acknowledgments.}

1376: 

1377:  We express our special appreciation for discussions with Dr. J. Va'vra.

1378:  We would like to thank Dr. G. Bohm and Wildau

1379: Politechnik Institute for the possibility to analyse the wires.

1380:  We thank to K. Reeves and S. Aplin for reading and correcting this 

1381: manuscript.

1382: 

1383:  This work was supported by the 

1384: Deutsches Elektronen-Synchrotron (DESY), 

1385: by the Alexander von Humboldt-Stiftung and Max Planck Research Award.

1386: 

1387: 

1388: 

1389: %\newpage

1390: 

1391: \begin{thebibliography}{99}

1392: 

1393: \bibitem{Proposal} T. Lohse $et~al.$, HERA-B collaboration,

1394: An Experiment to Study CP Violation in the B System Using an 

1395: Internal Target at the HERA Proton Ring, Proposal,

1396: {\bf DESY-PRC 94/04} (1994). 

1397:  

1398: \bibitem{TDR} E. Hartouni $et~al.$, HERA-B collaboration,

1399: An Experiment to Study CP Violation in the B System Using an 

1400: Internal Target at the HERA Proton Ring, Design Report,

1401: {\bf DESY-PRC 95/01} (1995). 

1402: 

1403: \bibitem{electronics} M. Buchler $et~al.$,

1404: IEEE Trans. Nucl.Sci., {\bf NS-46 } (1999) 126-132.

1405: 

1406: \bibitem{scint} T.J. Sumner {\it et al},

1407: IEEE Trans. Nucl. Sci., {\bf NS-29 (5)} (1982) 1410-1414.

1408: 

1409: \bibitem{start} S.A. Korff and R.D.Present, 

1410: Phys. Rev. {\bf Vol.65 (9)} (1944) 274-282.

1411: 

1412: \bibitem{co2} V. Bawdekar,

1413: IEEE Trans. Nucl. Sci {\bf NS-22 } (1975) 282-285.

1414: 

1415: \bibitem{N2} L. Colli, U. Fracchini,

1416: Rev. Sci. Instr. {\bf Vol.23 (1)} (1952) 39-42.

1417: 

1418: \bibitem{book1} L.G. Christophorou, Atomic and Molecular 

1419: Radiation Physics, (Wiley, New York, 1971)

1420: 

1421: \bibitem{fast1} L.G. Christophorou {\it et al}, 

1422: Nucl. Instr. and Meth. {\bf A 163} (1979) 141-149.

1423: 

1424: \bibitem{fast2} L.G. Christophorou {\it et al}, 

1425: Nucl. Instr. and Meth. {\bf A 171} (1980) 491-495.

1426: 

1427: \bibitem{armit} J. C. Armitage {\it et al},

1428: Nucl. Instr. and Meth. {\bf A 271} (1988) 588-496.

1429: 

1430: \bibitem{cern} A. Peisert, F. Sauli,

1431: CERN 84-08 (1984).

1432: 

1433: \bibitem{biagi} S. Biagi,

1434: Nucl. Instr. and Meth. {\bf A 421} (1999) 234-240.

1435: 

1436: \bibitem{cf42} M.C. Borgade {\it et al}, 

1437: J. Appl. Phys, {\bf Vol.80, (3)} (1996) 1325-1336.

1438: 

1439: \bibitem{mix1} T.Yamashita {\it et al}, 

1440: Nucl. Instr. and Meth. {\bf A 317} (1992) 213-220.

1441: 

1442: \bibitem{transport} S.R.Hunter {\it et al}, 

1443: J. Appl. Phys, {\bf Vol.58, (8)} (1985) 3001-3015.

1444: 

1445: \bibitem{christ} L.G. Christophorou {\it et al}, 

1446: Nucl. Instr. and Meth. {\bf A 309} (1991) 160-168.

1447: 

1448: \bibitem{book2} L.G. Christophorou (ed.), Electron-Molecule

1449: Interactions and Their Applications, vols. 1 and 2, 

1450: (Academic Press, New York, 1984)

1451: 

1452: \bibitem{attach} W.S. Anderson {\it et al}, 

1453: Nucl. Instr. and Meth. {\bf A 323} (1992) 273-279.

1454: 

1455: \bibitem{cf41} L.G. Christophorou {\it et al}, 

1456: J. Phys. Chem. Ref. Data, {\bf Vol.25, No.5} (1996) 1341-1388.

1457: 

1458: \bibitem{quench1} L.G. Piper {\it et al}, 

1459: J. Chem. Phys, {\bf Vol.59, (6)} (1973) 3323-3340.

1460: 

1461: \bibitem{quench2} J.E. Velazco {\it et al}, 

1462: J. Chem. Phys, {\bf Vol.69, (10)} (1978) 4357-4373.

1463: 

1464: \bibitem{becker} U. Becker {\it et al}, 

1465: Nucl. Instr. and Meth. {\bf A 315} (1992) 14-20.

1466: 

1467: \bibitem{grimm} O. Grimm, Diplomarbeit, 

1468: University Hamburg, July (1998) (in german)

1469: 

1470: \bibitem{beck} M. Beck, Dissertation, 

1471: University Rostock, June (1999) (in german)

1472: 

1473: \bibitem{mix2} T.Yamashita {\it et al}, 

1474: Nucl. Instr. and Meth. {\bf A 283} (1989) 709-715.

1475: 

1476: \bibitem{gain1} S. Beingessner, R. Carnegie,

1477: Nucl. Instr. and Meth. {\bf A 260} (1987) 210-220.

1478: 

1479: \bibitem{charpak} G. Charpak {\it et al},

1480: Nucl. Instr. and Meth. {\bf A 99} (1972) 279-284.

1481: 

1482: \bibitem{workshop} Proc. Workshop on Radiation Damage to Wire Chambers, 

1483: Lawrence Berkeley Laboratory (Jan. 1986) LBL-21170.

1484: 

1485: \bibitem{kadyk} J. A. Kadyk,

1486: Nucl. Instr. and Meth. {\bf A 300} (1991) 436-479.

1487: 

1488: \bibitem{vavra} J. Va'vra, ref.~\cite{workshop}, pp. 263-294.

1489: 

1490: \bibitem{cern1} R. Bouclier {\it et al},

1491: Aging of microstrip gas chambers: problems and solutions,

1492: CERN-PPE 96-33 (1996).

1493: 

1494: \bibitem{gcms} J. Wise {\it et al},

1495: IEEE Trans. Nucl. Sci {\bf NS-37 (2) } (1990) 470-477.

1496: 

1497: \bibitem{chemmod} J. Wise {\it et al}, 

1498: J. Appl. Phys, {\bf Vol.74, (9)} (1993) 5327-5340.

1499: 

1500: \bibitem{hess} D. W. Hess, ref.~\cite{workshop}, pp. 15-24.

1501: 

1502: \bibitem{yasuda} H. Yasuda, Plasma Polymerization,

1503: (Academic Press, 1985)

1504: 

1505: \bibitem{malter} L. Malter

1506: Phys. Review, {\bf Vol. 50} (1936) 48-58.

1507: 

1508: \bibitem{juri} I. Juricic, J. Kadyk, ref.~\cite{workshop}, pp. 141-159.

1509: 

1510: \bibitem{kott} R. Kotthaus, ref.~\cite{workshop}, pp. 161-193.

1511: 

1512: \bibitem{opensh1} R. Openshaw {\it et al}, 

1513: IEEE Trans. Nucl. Sci {\bf NS-36 (1) } (1989) 567-571.

1514: 

1515: \bibitem{algeri} A. Algeri {\it et al},

1516: Nucl. Instr. and Meth. {\bf A 338} (1994) 348-367.

1517: 

1518: \bibitem{dissert} V. Pashhoff, 

1519: Dissertation, University Freiburg, October (1999)

1520: 

1521: \bibitem{lyon} M. Titov, 

1522: A Gaseous Muon Detector at the HERA-B experiment,

1523: talk given at Nuclear Science Symposium and Medical Imaging Conference,

1524: 15-20 October 2000, Lyon, France 

1525: 

1526: \bibitem{hadr} M. Danilov, L. Laptin, I. Tichomirov, M.Titov, Yu. Zaitsev,  

1527: Aging of gaseous detectors under high-rate irradiation with 

1528: hadronic particles, in preparation.

1529: 

1530: \bibitem{tsarnas} N. Spielberg, D. Tsarnas

1531: Rev. Sci. Instr. {\bf Vol.46 (8)} (1975) 1086-1091.

1532: 

1533: \bibitem{smith} A. Smith, M. Turner

1534: Nucl. Instr. and Meth. {\bf A 192} (1982) 475-481.

1535: 

1536: \bibitem{kollef} M. Kollefrath {\it et al},

1537: Nucl. Instr. and Meth. {\bf A 419} (1998) 351-356.

1538: 

1539: \bibitem{kwong} K. Kwong,

1540: Nucl. Instr. and Meth. {\bf A 238} (1985) 265-272.

1541:  

1542: \bibitem{atlas} V. Paschhoff, M.Spegel, Aging studies for the

1543: ATLAS MTD's using $Ar/CO_2$(90:10), ATLAS muon note-019 (1999).

1544: 

1545: \bibitem{atlasnote} I. Boyko {\it et al}, Aging of aluminum

1546: drift tubes filled with $Ar/CO_2/CH_4$ (92:5:3), 

1547: ATLAS muon note-089 (1995).

1548: 

1549: \bibitem{sadr} H. Sadrozinski, ref.~\cite{workshop}, pp. 121-129.

1550: 

1551: \bibitem{mogab} C. Mogab {\it et al}, 

1552: J. Appl. Phys, {\bf Vol.49, (7)} (1978) 3796-3803.

1553: 

1554: \bibitem{kushner} M.J. Kushner,

1555: J. Appl. Phys, {\bf Vol.53, (4)} (1982) 2923-2938.

1556: 

1557: \bibitem{winters} H.F. Winters {\it et al}, 

1558: J. Appl. Phys, {\bf Vol.48, (12)} (1977) 4973-4983.

1559: 

1560: \bibitem{kadyk1} J. Kadyk {\it et al},

1561: IEEE Trans. Nucl. Sci {\bf NS-37 (2) } (1990) 478-486.

1562: 

1563: \bibitem{opensh} R. Openshaw {\it et al},

1564: Nucl. Instr. and Meth. {\bf A 307} (1991) 298-308.

1565: 

1566: \bibitem{d0} G. Alexeev {\it et al},

1567: Technical design Report for the D0 Forward Muon Tracking

1568: Detector Based on Mini-Drift Tubes, D0 Note 3366, (1997)

1569: 

1570: \bibitem{d0note} B. Baldin {\it et al},

1571: Technical design of the central muon system, D0 Note 3365, (1997)

1572: 

1573: %\bibitem{stegmann}  C. Stegmann, The Outer Tracker of HERA-B, 

1574: %Proceedings of 7th International Conference on Instrumentaion for 

1575: %Colliding Beam Physics, Hamamatsu'99, Japan (in press)

1576: 

1577: \bibitem{hohlmann} M. Hohlmann, The Outer Tracker of HERA-B, 

1578: Proceedings of 8th Pisa Meeting on advanced detector,

1579: Isola d'Elba'00, Italy (in press)

1580: 

1581: \bibitem{kolanoski}  H. Kolanoski, Investigation of Aging

1582: in the HERA-B Outer Tracker Drift Tubes,

1583: talk given at Nuclear Science Symposium and Medical Imaging Conference,

1584: 15-20 October 2000, Lyon, France 

1585: 

1586: \bibitem{akesson} T. Akesson {\it et al},

1587: Nucl. Instr. and Meth. {\bf A 361} (1995) 440-456.

1588: 

1589: \bibitem{straw} ATLAS Technical Design Report, CERN (1997)

1590: 

1591: \bibitem{romaniouk} A. Romaniouk,

1592:  Choice of materials for the constructio of TRT,

1593: ATLAS Internal Note, INDET-98-211 (1998)

1594: 

1595: \bibitem{kriv1} G. Gavrilov {\it et al},

1596: Aging investigation of ATLAS TRT straws, PNPI-preprint-2328 (1999)

1597: 

1598: \bibitem{kriv2} T. Ferguson {\it et al},

1599: Possible new mechanism for anode wire aging

1600: in gas filled detectors, PNPI-preprint-2331 (1999)

1601: 

1602: \bibitem{beauty} T. Zeuner, The MSGC-GEM Inner Tracker for HERA-B

1603: Nucl. Instr. and Meth. A (Beauty'99 proceedings, in press).

1604: 

1605: \bibitem{coyle} J. Va'vra  {\it et al},

1606: Nucl. Instr. and Meth. {\bf A 324} (1993) 113-126.

1607: 

1608: \bibitem{denisov} D.S. Denisov, 

1609: On using $CF_4$ as a working gas for drift tubes (in russian),

1610: IHEP-preprint-90-16 (1990)

1611: 

1612: \bibitem{charm} J. DeWulf  {\it et al},

1613: Nucl. Instr. and Meth. {\bf A 252} (1986) 443-449.

1614: 

1615: \bibitem{straw1} J.Kadyk {\it et al},

1616: Nucl. Instr. and Meth. {\bf A 300} (1991) 511-517.

1617: 

1618: \bibitem{argus} M. Danilov {\it et al},

1619: DESY Internal Note 88-090, DESY (1988)

1620: 

1621: \bibitem{vavra2} J. Va'vra, 

1622: Aging of gaseous detectors, SLAC-PUB-5207 (1990)

1623: 

1624: \bibitem{atac} M. Atac, ref.~\cite{workshop}, pp. 55-66.

1625: 

1626: \bibitem{gang} AN Ji-Gang {\it et al},

1627: Nucl. Instr. and Meth. {\bf A 267} (1988) 396-407.

1628: 

1629: \bibitem{book3} L. Holland, Thin Film Microelectronics 

1630: (Chapman and Hall Ltd, 1965) p. 157.

1631: 

1632: 

1633: \end{thebibliography}

1634: 

1635: \end{document}

1636: 

1637: