Vacuum and beam transport lines: main principles

Vacuum/volume

48/number IO/pages 793 to 802/1997 0 1997 Elsevier Science Ltd All rights reserved. Printed in Great Britain 0042-207x/97 $17.00+.00

Pergamon PII: s0042-207x(97300022-5

Vacuum and beam transport lines: main principles J PivarE,* J PivarEt and K D Tumanov, Joint lnstitute Reactions, 14 1980 Dubna, Moscow region, Russia received 9 February

for Nuclear

Research,

Flerov Laboratory

of Nuclear

1997

Some details of vacuum system designs of beam transport lines are presented. The effect of gas desorption on vacuum equilibrium and stability and ion-induced pressure instability in the beam transport lines are discussed. Main vacuum relationships are also given as a basis for understanding the vacuum system configuration of beam transport lines. Residual gas spectra together with specific outgassing rates of main materials used for construction of beam transport lines are shown. 0 1997 Elsevier Science Ltd. All rights reserved

Introduction

l

The vacuum system of the beam transport line is one of the main components of the transport line. Whereas, operating pressure ranges for vacuum systems of the ECR (Electron Cyclotron Resonance) ion sources are 5 x 10~2-10~5 Pa,’ the required operating pressure range for the vacuum system of the beam transport line is 10-4-10~9 Pa. Basically, it consists of stainless steel and copper. The beam tubes are pumped with turbomolecular, cryosorption, getter-Ti-sublimation, NEG (Non Evaporable Getter) pumps’ combined with sorption and rotary pumps. Other suitable combinations of the vacuum pumps as well as the pressure measuring gauges are shown in Figure 1. This paper is giving a view on the gas desorption processes especially ion-induced pressure instability influencing on the operating vacuum inside the beam pipe of the transport system. Scaling relationships are also pointed out as a basis for understanding the vacuum configurations of beam transport lines as well as molecular conductances and beam pipe outgassing. The outgassing rates at room temperature are also presented for materials such as Viton (Russia and Roumania versions), ‘0’ rings (Rubber 7889 and TU MCHP 9024) molybdenum VNII NP-257 and Apiezon ‘L’ greases, Ramsay grease,3 rotary pump oil VM-4 and glassfibre samples, currently used at our laboratory.

l l l l

Outgassing due to the desorption of weakly bound molecules on the walls of the vacuum system; The ions induced by extracted and accelerated ions; The diffusion of hydrogen from the walls of the vacuum system; The neutral gas produced inside the transport beam lines; The desorption of molecules generated by hard X-ray and bremstrahlung.

To establish the required pressure for a given pumping speed the average thermal outgassing rate and the total average desorption rate must be below certain definite values.

PUMPS

VACUUM GAUGES

lo* “--l--T

10

2 1

Desorption of gases The pressure from 10~4-10-“Pa must be maintained transport beam lines and its beam pipe despite: l

The thermal

outgassing

inside the

of surfaces;

*Permanent address: Institute of Physics, Slovak Academy of Sciences, Dubravska cesta 9, SK-842 28 Bratislava, Slovak Republic. tpermanent address: J. Pivari: (Jr), Comenius University, MPF, Dept. of Solid State Physics, Mlynska dolina, SK-842 15 Bratislava, Slovak Republic,

Figure 1. A list of pumps and gauges suitable for pumping measurement

in beam transport

and pressure

lines. 793

J Pivare et al: Vacuum

and beam transport

lines

Ion-induced pressure instability The ions induced by extracted and accelerated beam can produce desorption of strongly bound molecules. The desorption flow rate II,, Q, can be expressed by’

ni = ysL(I/e)n

Q, = qcrL(Z/e)p

(I)

where 4 is the molecular desorption yield (molecules ion-‘), 0 is the ionization cross section of extracted and accelerated ions (m’) (for example for high energy protons cr = 1.2 x lo-** rn’ and for CO,4 L is the length of beam section (m), I is the beam intensity (A), r is the electron charge (A s), n is the number of molecules in unit volume (m-‘) and p is the pressure (Pa). Note. if ions are taken from the residual gas. then q will represent a net desorption yield. For q < 0 ‘a beam pumping’ can be observed, i.e. the system acts like an ion pump. In the presence of the ion induced desorption q > 0 and the equilibrium pressure can be expressed from the equilibrium equation

where Q,/L is defined by eqn 1, Q/L is the specific gas flow defined by eqn 8, Q,/L is the total specific gas flow which can be expressed by equation Q,/L = pS,, and S,, is the ‘effective linear pumping speed’ defined by eqn 14. Then, it holds that the equilibrium pressure p

p =

Q,/LS,,

=

y”(z’e)p +qA X,,

(4

which gives

gives efficient sputtering and desorption of strongly adsorbed gas molecules. On the other hand, the addition of O2 produces CO and CO: compounds from carbon contaminants on the surface which can easily be pumped out. However. the ion-induced pressure desorption rate dQ’!dt is given by

dQi

dt

= Q,r7

if the molecular desorption rate q = 1 molecules ion-‘, the ionization cross-section 4 0 = 1.2 x 10m2’m’, the length L = 1m, the beam current I = 10e3 A. the electron charge e = 1.60220 x IO-“As, and the pressurep = 10m4Pa. Main principles The highest obtainable average pumping speed of the beam transport line vacuum system strongly depends on the beam pump conductance. To illustrate this important limitation caused by the finite conductance, let us consider the system shown in Figure 2. In the molecular flow regime the flow of molecules along the vacuum pipe to the nearest pump is expressed by the equation,

where Q is the gas flow (Pa m3 ss’), 1~ is the specific molecular conductance (m4s-‘) ()u = LC), C is the conductivity (m’ SK’), p is the pressure inside the pipe (Pa), A is the specific surface area (m) (A = F/L), F is the surface area (m’) and q is the specific outgassing rate (uniform) (Pams-‘). These equations can be combined to give

d2p

IV-=

dx2

where p,, = qAIS,,r. By introducing

Solution ditions

dp dx r=L?

we obtain

p=

x lo-“Pam3sp’

po

-Aq. of this simple equation

= 0

depends

on the boundary

con-

and

A@

PI.,=0 = s

111.

’ - WL,

Hence, the ‘critical cannot exceed

current’

product

for S + E (see eqn 14)

which follow from the evident symmetry considerations. By the first integration of the eqn 9 using the first boundary condition we get

dp -= dx Therefore, the pressure p is a function of the beam current I. The higher the current I, the higher the equilibrium pressure p. The ratio C/L is also important parameter. The higher the ratio C/L, the higher the ‘critical current’ I,,,,. The pressure instability caused for example by the unexpected increase in pressure in the vacuum system of the beam line can be reduced by adding cryogenic, getter-Ti sublimation, or NEG pumps, respectively. With a 3OO’C bakeout and Ar+O, glow discharge cleaning of the source stages and beam pipes the molecular desorption yield 4 can be reduced. The required number of ions bombarding the beam pipe is typically 1022mm2 and results in that 4 < 0. The Ar glow discharge 794

4

--x+c, IV

Figure 2. Outline of the beam transport line pumping system. S is the pumping speed of the pump in points ‘0’. ‘A’, ‘B’ (m’s_‘), L is the distance between pumps reference point ‘0’ (m).

(m) and x is the distance

measured

from the

J Pivari: et al: Vacuum and beam transport lines

where C, = (Aq/++*)(L/2). After the second integration using the second boundary condition, it holds that

w = 0.07 rn4sY1

of eqn 9

where CZ = (AqL/S). Finally, the general solution of eqn 9 is a well known parabolic pressure profile along the beam pipe

(11) From equation dp/d.u = 0 we can calculate the position of the maximum pressure. It is possible to show that the maximum occurs at x, = L/2, it means at the midpoint between pumps. After substitution x, to eqn 11 we get

(12) What concerns of the average pressure for the beam pipe, which is more relevant, we can calculate it from equation

10-l

2

3

100

2

3

10'

L (ml

Figure 3. Effective linear pumping speed S,, as a function of pump

(13) It is convenient s

eff

to define an ‘effective linear pumping

LL+r

= (

12p1:

s

speed’

distance L for various sizes of pumps, a vacuum beam pipe conductance w = 0.07m“s-’ and different pumping speeds S = 0.02, 0.04, 0.06, 0.12, 0.24, 0.5 and CCm3s-‘, respectively.

(14) )

so thatp,, = &I&,, orpa&I = Q/L. It is evident that S,,cannot exceed 12tv/LZ, irrespective of how large pumps are used, Equally, the lowest achievable average pressure is limited to AqL’/l2tia. Since the conductance is generally determined by the apertures in electrodes and screens (screening of high frequency), the only parameter is the interpump distance L. Obviously, many small pumps at short distance are preferable to a few and large pumps. Figure 3 illustrates this effect for a vacuum beam pipe conductance of 0.07 m4 s-l and different pumping speeds S (0.02, 0.04, 0.06, 0.12, 0.24, 0.5, cc m3s-‘), respectively. It is shown that the effective pumping speed is practically the ‘linear function’ of the pump distance for various sizes of pumps. Such a pumping structure may be installed along the vacuum system and hence the analysis provides a ‘linear pumping speed’ inside the beam pipe. The real pressure distribution and therefore the average pressure also follow from equation similar to the previous one

(15) where it’ is the specific molecular

conductance

(IV = f. C) (m”s-‘),

p is the pressure inside the pipe (Pa), x is the distance measured from the reference point ‘0’ (see Figure 2) (m), A is the specific surface area (A = F/L) (m), F, is the surface area (m*), L is the distance between pumps (see Figure 3) (m), s is the specific pumping speed defined as s = SjL (m2s-‘), S is the pumping speed of the pump in points ‘O’, ‘A’, ‘B’ (see Figure 2) (m’s_‘) and q is the specific outgassing rate (Pam s-‘). By solution of differential equation (15) using the boundary conditions (10) we find other pressure profiles along the beam pipe. At the beginning we have to solve the differentia1 equation

in the form of p,(x) 2 err where r $ 0. It is possible to show that r’ = S/M’,and Y,.~= ~(s/Pv)“-‘. Then, one soiution of differential equation may be expressed as P,(X) = CI e’l-‘+ C2 e’z-’ where C,, CZ are calculated through (10). Then, we get two equations C,erlL’*

_ C, e-r1L.2

By solution

=0

of equations

and

the boundary

C,+C,=,-

conditions

A@

we get

Finally, the general solution of differential eqn 15 we can expressed as p(x) = p,(x)+pz wherep, = Aq/s. If I’] + r, therefore one obtains equation

p(x) = C, e’“+C,e~‘“+

4

-

s

(16)

where

and

c, = -A& s (1 te-y. 795

J Pivarti et al: Vacuum

and beam transport AVERAGE

10-3E

1

’

!

lines

PRESSURE

’

1

1

I

’

fore the flux of gas Q (Pa m’ s- ‘) is proportional gradient Ap (Pa). It holds that5

4

to the pressure

Q = CAp

(20)

where the conductance C can be expressed for orifice, long cylindrical tube, long tube with elliptic section, at the molecular flow regime, by means of relations’.“:

0.04 0.06

(a) Orifice of area F (ml); 0.24

;m3

C = 36.4F

0.5

(b) long cylindrical

10-6’

0

2

tube, radius r (m), length L (m);

(22)

’ 6

4

(21)

s-r

J

8

10

12

L Cm)

Figure 4. Plot of the average pressure p.” vs distance

L between pumps speeds S = 0.02,0.04,

for Aq = 3 x lo-‘Pam’s_’ and different pumping 0.06.0.12. 0.24.0.5 and cc, m3s-‘, respectively.

(c) long tube with elliptic section (semi-axes &?

c=

Usually, the vacuum system can be designed with respect to s < ~1 and (S/W)“’ L < 1. In this configuration one can obtain (17) Therefore,

The average pressure

$

(18)

poZ after integration

of P(X) is determined

by pa,, = i

fp(x) dx = 2 (erL -eprL) 0

It is again convenient then

431

(23)

L@+@“’

where T is the absolute weight.

temperature

The molecular flow through with Monte Carlo programs

(K) and M is the molecular

complex systems can be computed adapted to the specific geometry.’

Outgassing of the beam pipe

eqn 16 may be reduced to

p(x) = C,(e”+e-‘“)+

a, b), a (m), b (m);

+

3. s

require in order to be I’ < 1. In this case,

Finally, the average pressure may be expressed

For every vacuum system the size of the required pumps is directly related to the outgassing. The first important source is the static and thermal outgassing of weakly adsorbed molecules and diffusion of H, from the bulk of the material. The standard procedure to reduce the thermal outgassing is the well known bakeout of the beam pipe. The pressure inside an unbaked system is mainly determined by water vapours. In a clean and well baked system Hz will be the dominant residual gas constituent. Typically, total specific outgassing rate q for unbaked and baked beam pipe made of stainless steel is given in Table 1. Therefore, the thermal outgassing rate dQ,/dt of the surface area F = 1 m2 is given by

dQ-r= Fq = __

10-6pam3s-‘.

dt

by (19)

Thus, it is also evident that the lowest achievable pressure depends only on the ratio of 2AqL/S. The average pressure vs the distance between pumps L for Aq = 3 x lO_‘Pam’ ss’ and different pumping speeds S (0.02, 0.04, 0.06, 0.12, 0.24. 0.5 and CJ m3 ss’) is shown in Figure 4.

The second important source of gas in the beam line is the socalled ‘dynamic’ outgassing ’ in the presence of the beam. Here,

Table 1. Total specific outgassing pipe made of stainless steel.4

rate q for unbaked

Process

Temperature

(K)

and baked

q (Pa m s-‘)

Molecular conductance In high vacuum the flow of molecules

is limited by wall collisions, i.e. the mean free path 1 of the gas molecules is greater than the characteristic dimensions of the vacuum system (D/i < 1; D is the diameter of the beam pipe). In this molecular flow regime, the conductance C (m’ SK’) is independent of the pressure, there796

beam

Unbaked 1OOh

293

lomh

Pumpdown Baked 30-l 50 h at 3OO’C

573

IO-9

J fivak

efal: Vacuum and beam transport lines COLD TRAP KITH COOLANT RESERVOIR

characteristics

of several materials

GCUq 53

RESERVOIR

MATERIALS

Figure 5. Outgassing

PG

used in vacuum

technology.’

strongly bound molecules can be desorbed which do not otherwise contribute to the thermal desorption. The outgassing characteristics of several materials used in vacuum technology are shown in Figure 5.’ The quantity of the outgassing rate very strongly depends on the finished treatment with materials (untreated; degreased: solvent, vapour; polished: mechanical, chemical, blast, electro; baked: non-metals at SO100°C up to 24 h, metals at 300400-C up to 100 h). It is also shown that the best metals used in vacuum technology are Al, Cu and stainless steel. Then, in order to obtain a specific outgassing rate less than 10mh Pa m SK’ the vacuum systems are not using Rubber, polyamids, epoxy and PTFE-Teflon materials. The outgassing characteristics of materials exposed to vacuum are also published in papers8-“’

Residual gas species results Outgassing characteristics of several materials were studied in vacuum at room temperature for periods as long as 24 h. Samples were placed in a stainless steel vacuum chamber which was evacuated by a 0.5 m3 SY’ turbomolecular pump. The vacuum chamber volume and surface area were approximately 0.0025m3 and 0.12 rn’. respectively. Pressures were measured from 0.1 MPa to 1O--‘Pa by using a combination of thermocouple, Pirani, BayardAlpert and Penning gauges. Typically, the test chamber and connecting lines were conditioned between test runs using a vacuum bakeout at about 150°C at a base pressure of about 10p4Pa. Concurrently, the RGA (Residual Gas Analyser) probe was baked out at 150’ C. A minimum of 10 h was allowed for the chamber to cool to room temperature prior to loading the next test samples. The amount of residual gases which were evolved from the samples under test were determined using the method of peak height summation and by subtracting the spectra measured of the empty test chamber from the spectra measured with the samples. Vacuum stand Scheme of the stand is shown in Figure 6. It is designed to pump the vacuum chamber with diagnostic elements and gauges to the

Figure 6. Scheme of the stand used for measuring RP is rotary pump, TMP current feedthrough. DV spectrometer gauge, MDV magnetic drive valve, PG TG is thermocouple gauge, control unit.

of the RCA spectra. is turbomolecular pump, ECF is electrical is dosing valve, RGSG is residual gas is manual drive valve, EMD is electrois Penning gauge, IG is ionization gauge, PIG is Pirani gauge. and GCU is gauge

pressure of 10-6Pa. The entire system including the holder of the samples is constructed of stainless steel, combined with copper and steel materials, respectively. One rotary pump is used in order to produce the vorvacuum (BL 90).” The rotary and turbomolecular pumps continuously pump the whole system. The rotary pump is also used for slow-acting evacuation of the tested chamber before connection of the turbomolecular pump. Proceeding from the dimensions of the vacuum system an effective speed S,,r is calculated for the center of the stand chamber. There was found to be: $, = 0.1 mi s-’ for air at 10d5 Pa. The reduced pumping speed of the turbomolecular pump is mainly due to the small diameter of the used pipes and trap. Results Outgassing rates are summarized in Table 2. It is seen that the VITON ‘0’ ring is better than Rubber 7889 by a factor of 15

Table 2. Specific outgassing rates q for some materials beam transport line vacuum technology. Sample

Exposure

‘0’ ring-VITON Russia ‘0’ ring-VITON Roumania ‘0’ ring-RUBBER-7889 ‘0’ ring-TU MCHP-9024 MO grease APIEZON ‘L’ RAMSAY grease Rotary pump oil-VM-4 Glass-fibre

24 24 24 20 20 20 20 20 20

(h)

Pressure 7 x Io-b 4x lo-5 lo-4 2x 10-j 4x 10-h 5x 10-7 1o--4 7x lo-i 5 x loms

used in the FLNR

(Pa)

q (Pa m s 6.5 x lo-’ 2 x IOmh lo-’ 8 x 10-j 9 x 10-j 2 x Io-i 2 x 10-j 7 x 10-2 lo-”

‘)

J PivarE et &Vacuum

and beam transport

lines

18 ____~--OF THE CHAMBER

BACKGROUND

/

-

2

17

A/q

VITON SKF - 26

28

32

PUMPING TIME (h)

24

t (“Cl

20

PRESSURE(Pa) DESOR. RATE i(Pa m s-I)

7 x IO -6 6,5 x ,o-5

I A/q Figure 7. Spectrum

of ‘0’ Ring Viton SKF-26 and background of the chamber from the plasma of the RCA spectrometer MX 7304 (monopole). rateMeasuring conditions were: ‘0’ Ring Viton-pumping time ~ 24 h; temperature ~ 20’C; pressure ~ 7 x 10 mePa and desorption 6.5 x 10-j Pam s- ‘. Background of the chamber-pumping time ~ 72 h; temperature-20,C: pressure-8 x lo-‘Pa and leakage-l.3 x lo-” Pa m3 s- ‘.

times and Ramsay grease is worse than APIEZON ‘L’ grease by a factor of 100 times. The worst material is the Rubber 7889 and the worst grease is the Ramsay grease. The outgassing rates presented for the VITON are the same as those reported in Ref. “. Similarly, our results again indicate not to recommend the use of the Rubber 7889 and Ramsay grease for clean interior surfaces of the beam transport line vacuum systems. The RGA spectra are shown in Figures 7-13. The spectra are measured by the RGA spectrometers MX 7304 (monopole) ” respectively. The spectrometer and IPDO-2A (omegatron)14, gauges are placed directly on the measuring chamber of the stand (Figure 6). By spectra analysis one can see the peak sets as 27, 29; 39, 41, 43; 55, 57, 58: 69; 77, 78; 85, 87; 101, 103, 105; 130, 132. 134, 136; 149. 151. 153, 155; 198; corresponding to the characteristic clusters of the n-C,H, hydrocarbons, or m-C,Cl,,F, freons. These fragments arise also during the ionization processes from oils used in the backing pumps, some diffusion pumps and turbomolecular pumps. In addition to the spectra, the expressive water vapor contamination is illustrated but there are also the peaks at 135, 137, 138, 140, 148, 155, 157 and 160 which can give rise to some difficulties in interpretation, mainly if the vacuum chamber is made of stainless steel and if it is divided from the vacuum pumps by two liquid nitrogen traps. Nevertheless, by means of the spectral catalogue, where the spectra of the pure substances are standardized, it is possible to find the appropriate substances in the majority of cases. 798

Conclusion The present work describes not only the main relations used at the construction of the beam transport line vacuum systems but also provides the new data in the region of the outgassing rates and the RGA spectra for the different solid, grease and liquid materials. Taken together, all these data show that it may be impossible to obtain the suitable operation vacuum (10-j Pa) and the necessary cleanliness of the beam lines if the vacuum exposed surfaces have higher specific outgassing rates than IO-‘Pam s -‘. We recommend to make the beam tubes of the transport beam lines with stainless steel, steel with stable structure and with a low relative magnetic permeability. Aluminium and copper can also be successfully used because of good mechanical properties. availability and very low desorption rates. It is practically impossible to separate the effects of outgassing, diffusion. and permeation which are manifold higher in polymers than in metals.” The use polymers for clean interior surfaces of the transport beam line vacuum systems is not recommended.

Acknowledgements The authors wishes to thank V B Kutner, head of the FLNR ISS for his understanding in this work, L Holland, editor of Vacuum for his recommendations concerning the degassing characteristics of oils, rubbers and greases I6 ” and S P Mishchenko for her help concerning the language correction of the paper.

J PivarC et al: Vacuum and beam transport lines xl .

‘O’-RING

RUBBER Russia

90 97

118

Figure 8. Spectrum of O-ring seal Rubber 7889-to A/q = 145 (Russia) sample from the plasma of the RGA Measuring conditions were: Pumping time-24 h; temperature-20’C; pressure-lO_‘Pa and desorption rate -

‘O’-RING

1

z z w FS - 4,

spectrometer MX 7304 (monopole). lO_‘Pam SK’.

/

i

2

16

26

39

49 54

64

‘O’-RING 1%

132

VITON Roumania

t

b

7889

t= 4

75 A/q

RUBBER

92 103 7889

148

160

Russia

2 N

2 175 187 A/q

209

Figure 9. Spectrum

of O-ring seal Rubber 7889-from A/q = 145 (Russia) and O-ring seal VITON (Roumania) samples from the plasma of the RGA spectrometer MX 7304 (monopole). Measuring conditions were: O-ring seal Rubber-pumping time-24 h; temperature-20°C; pressure - 10m4Pa and desorption rate---IO-‘Pam SKI and O-ring seal VITON-pumping time-24 h; temperature-20’C; pressure--lx lO_‘Pa and desorption rate2x 10-6Pamsm’. 799

J Pivarc et al: Vacuum

and beam transport

lines

‘0’

1oc I

-3

t

54 c

u P L ; z 2 54-

MCHP 9024

Russia

18

z3 z

F;

RING-TU

28

w

IIIII

z2

3039

56

42

17 26=g 32

4453

I

61

102

A/q

118

i

MO GREASE

Russia

28 18

2-

32 38 44

I

I

rl II

17

27

37 41

100

78 I

II

1

56

74

84

148 1

110

Ah Figure 10. Spectrum of O-ring seal TU MCHP 9024 (Russia) and molybdenum grease VNII NP-257 (Russia) from the plasma of the RGA spectrometer IPDO-2A (omegatron). Measuring conditions were: O-ring seal TU MCHP 9024-pumping time-20 h; temperature-20°C; pressure-2 x IF5 Pa and desorption rate-8 x 10e4 Pam ss’ and molybdenum grease-pumping time-20 h; temperature-20°C; pressure4 x 10m6Pa and desorption rate 9 x 10m-5Pam ss’.

APIEZON

‘L’

Eneland

RAMSAY

Russia

1828 I

42 144

DESOR. (Pa m

56

1

18 25 20

27

39

58

2 x 10-S

100

70

Es938

RATE s-l)

lli 68 74

150

Ah

Figure 11. Spectrum of Apiezon ‘L’ (England) and Ramsay (Russia) greases from the plasma of the RGA spectrometer IPDO-2A (omegatron). Measuring conditions were: Apiezon ‘L’-pumping time-20 h; temperature-20°C; pressure-5 x IO-‘Pa and desorption rate-2 x lO~‘Pam s-’ and Ramsay-pumping time-20 h; temperature-20°C; pressure-lUm4 Pa and desorption rate-2 x IO-‘Pa m s-l. 800

J PivarE et al: Vacuum

and beam transport

lines

ROTARY VM-4

56 42

PUMP OIL Russia

86

72

102 112

130

62 26 28

lo-

1

621

II44 16 171g2732

II

160

A/q

4o

Figure 12. Spectrum of rotary pump oil VM-4 (Russia) conditions were: Pumping time-20h; temperature-20’C;

-

C

samples from the plasma of the RGA spectrometer IPDO-2A pressure-7x IO-‘Pa and desorption rate-7x lO_‘Pa m SY’.

(omegatron).

Measuring

GLASSFIBRE 58

I()()

74 85

PRESSURE( DESOR. RATE 1 (Pa m s-l)

5 x lo->

1 j

1 I

! ~

10-4

115 I

I

128

I

1571

1

26 20 27

17

I

40 42

155

Figure 13. Spectrum of glassfibre samples from the plasma of the RGA spectrometer time-20 h; temperature-20°C; pressure-5 x lO_‘Pa and desorption rate-lO_‘Pam

(omegatron).

Measuring

conditions

were: Pumping

3. Eschbach,

References 1. Pivarc, J., J. Var. Sci. Technol., 1994, A12(5), 2716. 2. Doni, F., Boffito, C. and Ferrario, B., J. Vat. Sci. Technol., A4(6), 2447.

IPDO-2A SK’.

1986,

L., Prakticheskie scedenva po oakuumnoj technikc, Izdatelstvo ‘Energia’, Moskva-St. Petersburg (1966) in Russian. Translation from the Praktikum der Hochvakuumtechnik, Erzeugung und Messung Niedriger Drucke. Akademische Verlagsgesellschaft. 4. Grobner, O., In Proceedings qf‘the CAS CERN Accelerator School 801

J PivarE et al: Vacuum

and beam

transport

lines

General Accelerator PhJ,sics. Gif-sin-Yvette, Paris, 2, 3314 September 1984,489, eds P. Bryant and S. Turner. Report CERN 85-19, Geneva (1985). Deshman, S.. Nauchnye osnoc~ cakuumnoj techniky, Izdatelstvo ‘MIR’, Moskva 1964 in Russian. Translation from the S. Dushman, Scientific Foundation qf VacuumTechnique, Second Edition. John Willey. New York, 1962. Pivarc. J.. In Proceedings ofthe 1993 Particle Accelerator Conference, Washington D.C.. 17-20 May 1993, Vol. 5, ed. Steven T. Corneliussen. IEEE Service Center, 445 Hoes Lane, Piscataway, NJ 088544150, 1993, p. 3894. Roth. A., Vacuum Technobgy. North-Holland Publishing Company, Amsterdam-New York-Oxford. 1976. Erikson, E. D.. Beat, T. G., Berger, D. D. and Frazier, B. A.. J. Vat. Sci. Technol., 1984, A2, 206.

802

9. Lee, G.. IEEE Trans. on Nucl. Sc.. 1985, NS-32, 3806. 10. Egelmann, G. and Genet, M., Wahl, W., J. Vat. Sri. Technol.. 1987. A5, 2337. 11. Rotary pump BL 90, Zaklad Techniki Prozniowej. Koszalin, ul. Przemyslowa 113, Poland. 12. Craig, R. D., Vacuum. 1970, 21. 139. 13. RGA spectrometer MX7304 (monopole), SPO ‘ELEKTRON’, 244030 Sumy, ul. Komsomolskaja 68 a, Ukraine. 14. RGA spectrometer IPDO-2A (omegatron), SPO ‘ELEKTRON’, 244030 Sumy, ul. Komsomolskaja 68 a, Ukraine. 15. Peacock. R. N., J. Vat. Sci. Technol., 1980, A17, 330. 16. Holland. L., Vacuum, 1971, 21, 45. 17. Holland, L.. Physics Bulletin, 1972, 23, 581,