Abstract Particle Backgrounds at the LEP Detectors.

Particle Backgrounds at the LEP Detectors.High Beam EnergiesG. von Holtey, SL Division, CERN, Geneva, SwitzerlandAbstractObserved background rates from synchrotron radiation photons and off-energy electrons at the LEP experiments with 92 GeV beams in LEP are reported. Possible performance limitations due to background at higher beam energies and luminosities are discussed.1. INTRODUCTIONMain sources of particle backgrounds at the LEP detectors are photons from synchrotron radiation (SR) in the fields of straight section quadrupoles and off-energy electrons and positrons from bremsstrahlung of beam particles with rest-gas molecules along the straight sections and the first part of the bending arc. The most important limit on photon background comes from ageing of gas tracking detectors, while electron background can cause high spurious trigger rates and can degrade the luminosity measurement.The updated background protection system [1], in particular the SR-masks installed within the experimental vacuum chambers, works very well and allows to shield the detectors against the several orders of magnitude higher SR-photon rate produced at the W-energy as compared to rates at the Z o-peak [2]. The collimation of off-momentum particles, however, is not efficient, as these particles stay within the beam envelope up to the last strong low-beta quadrupoles [3]. The only handle against this background is the strict control of the vacuum pressure in the straight sections around the experiments. Measured background rates at 92 GeV during 1997 physics running are reported in section 2. Possible limitations on the physics performance due to backgrounds at the detectors are discussed in section 3.2. BACKGROUND RATES AT 92 GeV Machine induced particle background rates at the LEP detectors with colliding beams of 92 GeV did not present serious limitations to the physics performance of LEP in 1997.Synchrotron radiation photon rates stayed well below acceptable limits at all four experiments up to the highest beam currents reached. The same is true for off-energy particle rates, except for IP6 (OPAL), where high rates of off-momentum electrons were reaching the detectors because of high vacuum pressure bumps in the IP6 insertion. The effect became particularly disturbing after a cavity window broke in September 1997 and the vacuum pressure in this region had risen by a factor over 100, reaching values of 3x10-8 Torr [4]. In order that the experiment could continue to take data, the field in the affected cavity had to be lowered, leading to a slow conditioning rate and consequently persistent high vacuum pressure levels for the rest of the year.These statements are illustrated by a selection of experimental data from the Collaborations. SR-photon and off-energy background rates as function of the total beam current measured in the SAMBA detector from ALEPH are shown in Figures 1 [5]. Rates observed during a two week period of physics running with the (90o/60o)-optics in LEP are plotted in Fig.1a, rates with the (102o/90o)-optics during the last few days of 1997 running are shown in Fig.1b.0.51.01.52.01 1.52 2.53 3.54 4.55Itot[mA]N[kHz]Electrons31.07. - 14.08. 183 GeV1.02.03.04.05.06.01 1.52 2.53 3.54 4.55Itot[mA]N[kHz]Photons31.07. - 14.08. 183 GeVFigure 1a: Electron and photon background rates versus total beam current at 91.5 GeV and (90/60)-optics. Rates are measured by the ALEPH SAMBA detector. One unit of ‘figure-of-merit’ corresponds to an electron rate of 625 Hz and a photon rate of 1.25 kHz.A stronger than linear increase of all rates with beam current is observed, in particular for the off-energy electrons. However, all rates stay well below acceptable levels of 5 in “figure-of-merit” units, even for the highest currents reached. Photon rates that scatter above thegeneral band of measurements are mostly correlated with trips of SC-cavities. Similar levels of background rates were seen in the DELPHI detectors [6]. Both experiments observed a reduction of photon and electron rates at similar beam currents for the (102/90)-optics (Fig.1b).0.51.01.52.011.522.533.544.55I tot [mA ]N [k H z ]Electrons02.11. - 09.11. 184 GeV0.51.01.52.011.522.533.544.55I tot [mA ]N [k H z ]Photons02.11. - 09.11. 184 GeVFigure 1b: Electron and photon background rates versus total beam current at 92 GeV and (102/90)-optics. Rates are measured by the ALEPH SAMBA detector. One unit of ‘figure-of-merit’ corresponds to an electron rate of 625 Hz and a photon rate of 1.25 kHz.A typical energy distribution of positrons seen in the ALEPH SICAL detector is shown in Fig. 2. The off-energy positron signal between 26 and about 70 GeV is clearly separated from the much larger Bhabha-peak at beam energy.010020030040050060070020406080100120E [GeV ]E n t r i e sLEP FILL 3970 10-Aug-1997 183 GeVE cut = 26 GeVE beamFigure 2: Energy distribution of positrons detected in the ALEPH SICAL calorimeter at 91.5 GeV and (90/60)-optics. An energy cut is applied at 26 GeV.The following figures show background measurements from the OPAL detectors [7]. Rates of synchrotron radiation photons and off-energy electrons in the OPAL Vertex Drift Chamber (CV) for runs with different beam energies and optics during 1997 physics data taking are shown as function of the beam current in Fig. 3. While photon rates stay in all cases safely below 5 units of ‘figure-of-merit’, off-electron rates with the (90/60)-optics at 91.5 GeV rise as high as 5 units. As the OPAL detectors could not be switched on until electron rates had been fallen down to these levels, actual rates at the beginning of fills with high currents were even higher.When comparing results from the two different optics, it appears that SR-photons rates are about equal, but a much higher scatter is observed with the (102/90)- optics (open circles in Fig. 3). Off-energy electron rates with the (102/90)-optics are systematically below those with the (90/60)-optics (open squares), however, it is not clear whether this is due to optics changes or due to tighter vertical collimation introduced at the last part of (90/60)-optics running.Figure 3: Rates of hits per bunch crossing of synchrotron radiation photons and off-energy electrons measured in the OPAL Vertex Drift Chamber (CV) are shown as function of the total current in the two crossing bunches.The different symbols belong to different beam energies and LEP optics. One unit of ‘figure-of-merit’ corresponds to a photon hit rate of about 0.6/BX and an electron hit rate of about 0.9/BX.The next figure shows energy distributions of electronsand positrons as seen in the OPAL SW luminometer before (Fig. 4a) and after (Fig. 4b) the moment when a cavity window broke on the right side of IP6 and the vacuum pressure locally had risen by a factor of over 100.Figure 4: Energy distributions of electrons and positons measured with the OPAL SW calorimeter. The upper part shows the distributions in the left and right detector before the vacuum accident, the lower part after the event. Note the increase in scale by a factor of three for the lower figures.When comparing the two sets of graphs the effect of a large vacuum pressure bump at about 210 m from IP6 becomes evident. The large tails of low energy particles, which were present already before the accident, are nearly doubled in height, as compared to the Bhabha-peak at beam energy. In addition a new strong family is appearing, which is peaked at about 60% of the beam energy and reaches twice the height of the Bhabha signal. The very high rate of low energy particles, which are also seen in the central detectors, leads to a dramatic increase of the rate of spurious triggers in the experiment.3. POSSIBLE LIMITATIONSWhile the particle background at the LEP experiments could be controlled to within tolerable limits at 92 GeV running in 1997, the protection against SR-photons can rapidly degrade with very large beam emittances and/or strong non-gaussian tails in the transverse beam distributions. Off-energy particle rates can become a serious problem if the vacuum pressure rises considerably in one or several sections in an even insertion.3.1 Synchrotron radiationThe strength, but also the limits, of the protection against synchrotron radiation photons by the SR-masks is clearly demonstrated in the measurement of Fig. 5. Withthe ‘mask protecting’ collimators closed to 148x , theminimum allowed value if apertures are set at 128x ,photon rates at the OPAL and DELPHI experiments are small and 10 times higher beam currents could be tolerated. However, photon rates at the experiments would rise exponentially if these collimators could not be closed to below 178xin IP8 (and IP4, which has the same optics) or below 258xin IP6 (and IP2). The role of these collimators, which are placed close to a waist of the horizontal beam size at about 55m from the IP’s, is to protect the downstream SR-masks from direct impacts of photons radiated in upstream quadruples QS5 to QS11.The shadowing of the masks is to first order independent of beam parameters and is in all four IP’s obtained with collimator settings of less than ±20 mm. As the beta-values at the collimators are quite different in the two types of straight sections - -x= 46.8 m in IP4&8 against -x= 20.8 m in IP2&6 - the maximum opening, that still allows protection of the SR-masks, is 1.5 times smaller in IP4&8 than in IP2&6, if measured in units of rms beam size.10152025300.00.51.01.5collimator half opening (x/σx) BGD1/mAOPALDELPHIsettingFigure 5: “Figure-of-merit” for SR-photons (BGD1) measured in DELPHI and OPAL as function of the opening of horizontal collimators COLH.QS3 and COLH.QS4, respectively. (92 GeV, (102/90)-optics,2.6 mA total beam current, Jx= 1.28 and 0x= 30 nm).Experience has shown, that background collimators around the experiments must be opened to at least 28x and 58yabove the respective aperture settings, in order not to scrape beam halo particles. With this rule in mind, the maximum safe opening of ±20 mm for the mask-protecting collimators implies a maximum allowed setting of the aperture limiting collimators in IP5.x σ()ap ≤x col =20mmεx βcol−2This limit is shown in Fig. 6, where the maximum horizontal aperture that still allows to protect the SR-masks - expressed in units of rms beam size - is plotted for both types of IP’s as function of the beam emittance.As consequence of the larger horizontal beam waist in the QS3 doublet region of IP4 and IP8, the limit to the machine aperture due to background considerations stems from these two IP’s. However, the masks can still be protected, even in IP4&8, for emittances as large as 60 nm, if the physical machine aperture can be limited to <108x in IP5.103050705.010.015.020.0emittance εx (nm)r e q u i r e d a p e r t u r e (x /σ)a pIP2&6IP4&8Figure 6: Maximum allowed horizontal aperture setting to protect SR-masks in IP4&8 (full line) and IP2&6 (broken line) as function of the beam emittance.For beams with gaussian or near-to-gaussian density distributions a 108x limit leaves sufficient space to maintain beam lifetimes of well above 10 hours.However, strong non-gaussian beam tails in the horizontal plane have been observed [8] with colliding beams in the (102/90)-optics, in which case the required space might not always permit sufficient beam life times.In order to allow comparison of the requested aperture limitations due to background considerations with measured (or simulated) “dynamic” aperture values in units of normalized aperture, the former limits are transformed into a A t=10h units in Fig.7.A x τ≥10h ()=x ap τ≥10h ()βx ,ap(mm ∗m −1/2)If these curves are compared to measured or estimated (preliminary) values of ten hour life time apertures for the different optics summarized in Table 1 [8], it becomes clear that background conditions can pose conflicting requirements to the machine performance in the presence of non-gaussian beam tails.2040602.03.04.0emittance εx (nm)√A x (10-3 m 1/2)IP2&6IP4&8Figure 7: Required horizontal aperture to protect SR-masks in units of a A 9=10h as function of the gaussian beam core emittance.LEP opticsDynamic aperture [10-3a m]a A xa A y 90°/ 60°,Gauss 2.1-Gauss+tails 2.1 1.4102°/ 90°,Gauss 1.7-Gauss+tails2.61.5Table 1: Measured (estimated) physical apertures for a 10hour beam lifetime for different optics at high energy [8].Aperture limiting collimators in the vertical plane in LEP are conventionally set around a 258y value of a ‘nominal’ vertical emittance, chosen to be 10% of thenatural horizontal emittance. The corresponding settingsin units of a A y are shown in Figure8 (solid curve).Vertical background collimators around the experiments can then be closed, if required, to 308y .2040600.51.52.5horizontal emittance εx (nm )√A y (10-3 √m )y ap =20σyy ap =25σyy ap =30σyFigure 8: Vertical aperture collimator settings in units of a A y as function of the horizontal natural beam emittance.This strategy has in most cases resulted in a good protection against backgrounds from the vertical plane [1]. Under certain conditions, in particular at 45.6 GeV at the beam-beam limit, larger 308y openings were required for lifetime reasons, at other moments, with large bunch train bump amplitudes, smaller 208y openings were successfully used, to limit off-momentum backgrounds.Again, when compared to values of Table 1, conflicts between background and lifetime arguments might occur,with very small horizontal emittances and/or non-gaussian tails in the vertical density distributions.A detailed study of the physical aperture, needed to obtain sufficiently long lifetimes with colliding beams of high currents at the beam-beam limit, becomes therefore important in order to test the validity of the (102/90)-optics for high beam energies and high luminosity’s in LEP.3.2 Off-energy electronsThe main source for off-energy electron and positron background at the LEP detectors is from bremsstrahlung of beam particles with rest-gas molecules. The scattering probability does, to first order, not depend on beam energy and is proportional to the gas pressure P(z) at thepoint of production z. The vacuum pressure has two components, the static pressure P stat (z) and the dynamic pressure P dyn (E 0,z), expressed in Torr/mA, which is proportional to the beam current i, and can depend, via synchrotron radiation, on the beam energy E 0. The number of off-energy particles N off-E , that reach the experiment, is proportional to the beam current and the vacuum pressureN off −E ∝i ∗F (E ,z )∗P (z )dz ∫0E 0∫dEandP (z )=P stat (z )+i ∗P dyn (E 0,z ){},where F(E,z) is the transmission function for an off-energy particle with energy E/E 0 from its point of creation z through the LEP lattice into the detector. While N is normally proportional to the beam current i, the background rate rises with the square of the beam current,if the vacuum pressure is dominated by a dynamic component, e.g. due to high synchrotron light power.This effect has clearly been observed at high beam energies in LEP.-300-200-1000.000.010.020.03z(m)F (z )Figure 9: Probability of off-energy electrons to reach the experiment as function of the distance z of its origin from IP8. The probability function F(z) is normalized to 1, anda constant vacuum pressure of P(z)=10-9Torr has been used.The transmission function F(z), integrated over all energies, for a typical LEP insertion is shown in Figure 9.It was obtained by simulation using a modified version of the MC code TURTLE DECAY [9]. An off-energy particle was classified as background if its distance from the beam axis at the location of the downstream SR-mask exceeded 42 mm. A constant vacuum pressure over the whole length was assumed.Off-energy particles from distances of up to 300 m from the IP can reach the detectors, but the transmission probability is not a very strong function of the distance to the IP. F(z) falls by only a factor of 2.5 from close to the IP to the end of the straight section at 250 m. Therefore, a high vacuum pressure bump occurring at any distance up to 300 m from the IP has about the same effect on the off-momentum background rate and dominates the total rate if the local pressure rises by more than a factor of 10 over the average pressure. The energy of electrons that reach the experiment, however, is a strong function of the production distance z. It stays at a constant value ofE/E 0 =0.65±0.05 from distances between -300m< z <-150m and then falls linearly down to values below 0.1 at close distance to the IP. As high energy background particles do more harm to the detectors than low energy ones, vacuum pressure bumps at large distances from the IP are most harmful.4. CONCLUSIONNo serious limitations of the physics performance dueto particle backgrounds were encountered at 92 GeV running during 1997. The improved protection system,including synchrotron radiation masks, effectively shields the experimental detectors against synchrotron radiation photons. Off-energy electron and positron rates were acceptably small in all detectors, except in OPAL, were high electron rates, due to large vacuum pressure bumps around IP6, led to increased trigger rates. The two LEP optics used perfored about equally, with some preference for the (102/90)-optics.It is expected that the photon background at the detectors can be sufficiently controlled up to highest beam energies and beam currents as long as lifetime arguments allow to close aperture collimators to below required threshold values. A possible limitation arises from insertions IP4 and IP8, where the horizontal beam waist in the QS3 doublet is 1.5 times larger than the waist at the corresponding position in IP2 and IP6. In case that colliding beams with high currents should produce large non-gaussian beam tails and require more space to maintain useful lifetimes, an optics solution aiming at reducing the beam waists in IP4 and IP8 might be the only remedy.There is a definite risk to limit the physics performance in case the vacuum pressure substantially rises in one or several sections of an insertion and causes excessive trigger rates in the experiment.REFERENCES[1]G. von Holtey et al., “Study of Beam-InducedParticle Backgrounds at the LEP Detectors”, CERN SL/97-40 (EA), June 1997, to be published in NIM [2]G. von Holtey, “Simulations of Photon Backgroundsto LEP Experiments at 90 GeV Beam Energy”,CERN SL/92-54 (1992)[3]E. Falk,V. Hedberg, G. von Holtey, “Simulation ofoff-energy Electron Background in DELPHI”,CERN SL/97-04 (1997)[4]N. Hilleret, “1997, a hot year for the vacuum system:a review”, these proceedings[5]K. Affholderbach (ALEPH), private communication,1998 [6]L. di Ciaccio (DELPHI), private communication,1998 [7]A. H. Ball and P. Estabrooks (OPAL), privatecommunication, 1998 [8]H. Burkhardt, I. Reichel, private communication,1998 [9]K.L. Brown and C. Iselin, CERN 74-2 (1974)。