Ulrich J. Becker, “L3 at LEP” - LNS46 Symposium: On the Matter of Particles
BECKER: L3 at LEP is really a matter of set zero particles. And set zero particles decay into everything we think is truly elementary today, with one exception, namely, the top quark. LEP is the machine in Geneva, Switzerland, which with 27 kilometer circumference has four interaction regions. L3 being here, where electrons and positrons collide with energies up to 100 GV.
L3 surrounds the intersection region with a vertex chamber and a huge facet eye of 12,000 crystals of BGO to detect photons and electrons. Hydrones go into the 400 tons of uranium color emitter and muons will penetrate everything and be bent under the influence of the 0.5 Tesla field and registered and analyzed in this [INAUDIBLE].
L3 also is a collaboration of 500 scientists from 42 institutions led by Professor Ting. Technically L3 is optimized for position in muons and photons and electrons. Such that the resolution is 1% in dimuon mass and delta P over P 2.5%. Since resolution goes linear with B, that's quadratic with L a very large conventional magnet was constructed. The very good unique photon resolution was achieved by new crystals bismuth germinate.
The magnet itself has 140 turns to generate a five kilo [INAUDIBLE] field and that necessitates a large coil with a hefty wire carrying 30,000 amps. The magnetic field return is done by 8,000 tons of iron, making it the largest electromagnet at the present moment, according to the book of records from Guinness. You see it here in the moment that the support tube was cantilevered in, which carries all other components of the detector. The muon detector, each of its wheels is 95 tons, gently glides on air cushioned so the alignment of 30 microns is not upset. Here you
see it in reality. And behind this is an alignment, which the muons coming from the intersection region are measured to 30 microns accuracy and they're bent. This was checked and under better conditions from set decays where you know that the muon should have the momentum of the beam. And indeed, from almost 9,000 events now observed a resolution of 2.5% as anticipated. Inside the huge support tube is the Hadron calorimeter, which stops the hadrons and measures the energy. And inside that, in turn, is this huge array of BGO crystals, shown here, one half finished one half still being loaded with crystals into the carbon fiber structure.
These crystals for the first time were grown in equality and length so that the electromagnetic showers could be recorded with 1% accuracy above 2 GV. It was done in China. The performance can best seen from Pi zero to gamma gamma decays, which is resolved to 7 MEV. The sigman to lambda gamma with wi 4 MEV. And the eta, which just was addressed, was 16 MEV.
In 1990, the machine started in '98 in 1990 130,000 said zeros got collected, which gave rise to 35 publications. In 1991, another 300,000 said zeros got recorded. So the total is almost half a million now, which gives rise to a determination of precise parameters of the standard model and measures lifetimes very accurately, all in all 60 papers.
Let me address just a few of them. We need, as a matter of fact, very accurate luminosity measurement to do precision measurements, which is done from small angled Bhabha scattering. You can see that the measurements as points and theory agrees very well. So that we can say the luminosity is known presently to 0.6% and will be even better in the future. With that, we can measure as a function of energy of the machine, the hadronic events and obtain a line shape of a Breit-Wigner here, which, if magnified, is incredibly accurate. This shows the differential of the values. If you look at events which have strong energy deposits back to back in the BGO crystals, it's E plus E minus. And you can plot those. They also follow a line shape of the said 0 after T channel contributions are removed. If you look for back to back muons in the muon system here, again, you can observe the line shape, and the peak of this distribution gives you the partial width.
Down here you see towers decaying into an electron, the other one and two three pions, which would enable you to measure the lifetime, which was done. They also followed the same line shape. So that we have no very accurate results of this partial [INAUDIBLE] from electrons, muons and taus down here. Hadronic widths is measured here. The widths of the said zero was fitted. And a very accurate measurement of the mass has been done.
Now we can ask what is missing. And the missing channels, if you attribute them to neutrinos, you'll find there are three neutrino species in this world, unless they are heavier than 45 GV. So dark matter has to be searched elsewhere. One also can measure with electrons, muons and taus The electrons, the forward-backward asymmetry that an electron scatters forward or backward. That symmetry is shown here as a function of energy going through the said zero mass. The reason this is important is because this is asymmetry, if you calculate it from the vector and axial [INAUDIBLE] coupling constants. It expresses itself in this term. And you can extract it from this measurement and a partial width measurement we have just shown.
They are related to the standard model of coupling constants with the Weinberg angle. The result of this is shown here. GV GA on a very enlarged scale, GA is close to the expected value of minus 0.5. These are the one sigma and two sigma contours. And from this we can evaluate the sine squared to the effective. And from refinement of the calculation of the partial widths, we can conclude that the top mass must be around 196 GV.
Let me come to a test of QED. From E plus E minus goes to gamma gamma, which has nothing to do with the zed 0. In fact, the gammas are strongly peaked towards the beamline and very well described by QED with cutoff parameters which are very high and let you conclude that electron radius is less equal then to the minus 18 meters. Also, if you consider gammas from the e star exchange, there is no room for it and the mass must be bigger than 122 GEV, if it exists.
Let me switch from QED to QCD, which was now calculated completely to second order. Looking at three jet events and looking at the smallest energy jet scaled to the beam energy, we see a strongly falling distribution, which happens to agree to the QCD calculation for a gluon spin one. If the gluon spin was zero this distribution wouldn't fit at all. In QCD, we should have flavor independence of the coupling constant. It shouldn't depend on which kind of working quark you use. We can single out the heaviest quarks known now. Beak works by large PT muons and measure for those events of S and it turns out to be the same then for all the others, which is a necessary condition.
The thrust distribution, which is the projecting of energies on an axis, is known over four decades now and agrees with the theory very well. This allows to extract this coupling constant and the value you find this way 0.118. However, if you use other methods you can get higher results. Therefore we give an average of 0.125 and consider a theoretical uncertainty of 0.08.
Search for the standard Higgs is done by looking for Zed 0 decaying into a Higgs and a virtual Zed 0, which in turn decays into a neutrino anti-neutrino quark anti-quark lepton anti-leptor. This is a process which is completely calculatable. So if the Higgs mass were 40 GEV you should find so many of this kind of decays and of neutrino decays. So this is the prediction for the various channels. No such event was seen. Therefore one can say up to the 95% confidence level that the Higgs mass must be bigger than 52.3 GEV.
There is an event which is bigger than 52.3 GEV and which fulfills the topology which we expect from such an event, in that it has a mu plus, a mu minus, and to rather broad jets, which could be attributed to a BB bar and they would form a mass of about 70 GEV. But clearly, the likelihood of such a thing occurring is low and it needs more events. Charge takes were searched for in the tau channels and CS bar. The result is that up to 45 GEV nothing was found.
The following things are unique to the L3 detector and its features. The eta was always a little mysterious, as it was mentioned before. It can be observed with a BGO with very good resolution and high certainty. The spectrum of eta is observed in Zed 0 decays. However, as a function of the relative energy, follows exactly to what you expect and therefore is nothing mysterious. Three gammas can be detected very well and Zed zeros should not decay into it, unless there is a signature for new physics.
One event of this type was seen. 0.5 are expected from QED. That lets us set a branching ratio of 10 to the minus 5 for this process not to happen. Hard photons will explore the details of the interaction at short distances. If you compare the zero decay and 2 QQ bar with a gluon emission, the color string lines will lead to the gluon. If you have QQ bar where the gamma emitted, the color line goes from the quark to the quark. So you expect more hadronic energy to be ejected in this part here. And this is an energy flow diagram. And the purple line gives you this process and indeed there is more hadronic energy ejected for this gamma events than for gluon events. And in the other half it's the other way around, as it should be. So this gives insight into the clustering structure perhaps.
Physics of heavy quarks. They can be singled out by muons having a high transverse momentum with respect to the general jet axis of the order of a quarter of the mass. That lets us select very well B mesons from other things. Because a cut at one and a half GV and PT leaves you only with this red staff, which is prompt B decays. That's true for muons as well for electrons.
Taking only those events with a high PT muon, you can again plot their occurrence. And they follow bright richner shape, again, of the Zed zero with a partial width of 8 at 385 MEV as expected. Re-extrapolate the muon, the decay muon towards the beam, will since it's a decay, miss the exact beam point by a impact parameter delta. Evaluating that gives you a measure for the lifetime, which is 1.3 picoseconds. From this lifetime, and knowing the branching ratio on how often B's decay to leptons, together with low energy data from Cleo, we can evaluate the transition matrix element from B's to C's to be 0.046.
Looking into the zero decays with MU plus MU minuses, the invariant mass peaks at 3.1 GEV showing the J resonance. And a branching ratio for this process is 4 times 10 to the minus 3. Re-extrapolating the MU plus MU minus to the decay point gives us a measure of the decay time by this distance. This is a different method, which agrees exactly to the first one in giving 1.3 picoseconds. Taking this B event and looking by the charge of the muon for the forward backward asymmetry, we find one, which is 0.08, which would allow to extract the coupling constants for B mesons. B mesons can mix first normally, if we have a B meson it will decay only in a MU minus and a E minus, not a MU plus emulous. The B zero bar however, will decay into MU plus E plus. But what may happen with a probability chi of mixing is that this B turns into B zero bar. In that case, it ought to decay into MU plus as well as the other one. So like sine dileptons indicate that mixing took place.
Just counting the events of normal occurrence plus minus combinations, and those which have equal sine and therefore could be due to mixing, lets us determine the mixing parameter to be 11%. Lepton flavor violation in Z zero decays would signify themselves by Zed decaying to say a MU plus to a minus, or vise versa. In that case, if one looks for muons only, one should see muons which peak at a Beam energy. And due to the good muon resolution, they should be within this red indicated diagram. As you can see, you only see decay muons from tau events. And therefore, we can set an upper limit of 10 to the minus 5 for this process not to happen.
Tau polarization. Recording top plus tau minus events, the taus are polarized between helecity minus one and helecity plus one stage. The measurement of the polarization is important because it determines the Weinberg angle directly. The measured spectrum of taus decaying into a Pi's and K's is given by this point here. The analytic helicity ones amplitudes, shown down here and minus one here, they together explain the data and measure the polarization.
So from tau to pi decays the polarization is minus 15%. But you can do it also from tau to E and tau to MU decays with the following results averaged to a minus 13% polarization.
Other topic is a model independent analysis of the number of neutrinos, which relies on the [? following ?] effect. Although, you create a Z 0 which decays in neutrino neutrino bar. It may have emitted a gamma beforehand. And that gives you a signature that this process took place. The spectrum of this gammas is given by this blue histogram and calculation. And agrees exactly with the data. If the number of neutrinos is three. So this gives an independent confirmation.
This is a very interesting event we have two of, which was a hard muon, two hard muons and two hard gammas, which are not close to them. What is surprising about these two events, measured with an extremely good gamma gamma resolution, the invariant gamma gamma mass for both of them is close to 60 GEV. And then there is nothing before you have here, QED corrections, a long distance away.
Let me summarize. Major results from L3 based on half million Zed zero events, number of neutrinos is three. Lepton radius measured to be smaller than 10 to the minus 17 centimeters, and quarks as well. Mass of Higgs must be bigger than 53 GEV. The M zero mass is known to a very high accuracy. So is the partial leptonic widths. The properties of heavy quarks have been measured very precisely. The decay times of B mesons. The mixing of B mesons has been determined and a precise determination of the stern coupling constant to second order has been achieved.
Unique to L3 are the results that the B to C transition matrix element was measured in four different methods. The sinus of the [INAUDIBLE] Weinberg angle was determined. And independent confirmation that the number of light neutrinos is indeed-- three was given. Flavor violation of Zed decays does not take place. The flavor independence of QCD is manifested by B's. The [INAUDIBLE] spectrum does not show any anomalies. The unexpected decays like this, indeed, don't exist. To hard photon emotion from quarks agrees very well with expectations and confirms that also quarks are very pointlike. The branching ratio into J particles was measured as 4 times 10 to the minus 3. And the mass limit for excited electrons set to over 122 GEV.
What will be next? Six million Zed zero events, which allow to search for new phenomenon, as the last mysterious events I showed you. Precision tests of the electoral week theory. In 1994 the energy will raise to 200 GEV or above. And enable W plus W minus protection with this luminosity in three years. To test electroweak theories further. Make a Higgs search up to masses of 80 to 90 GEV. And search for new phenomenon.
The W plus W minus production is interesting in that it can proceed with the following three diagrams, of which this one is specific to the non [INAUDIBLE] nature of the standard model. Together, they give the very [INAUDIBLE] cross-section shown here. If this graph wouldn't exist they would raise to very high values. This is even demonstrated more in the angular distribution, which the decay muons from these Ws would have with respect to the beam axis, cosine here. And the standard model you expect a very steeply rising angular distribution. Whereas, if this graph doesn't exist, it would be very much flatter. And you see many more events in the back [INAUDIBLE] hemisphere.
After all these successors were the standard model, let me remind you that the standard model has 20 undetermined parameters, which have to be taken from experiment. Namely, the coupling constant, electron masses and the quark masses and the mixing parameters. And 13 of those come from the so-called Higgs sector, which has to be explored in future. And that I will leave to [INAUDIBLE]. Because if you plot these masses on a strongly logarithmic scale, you'll see that they are all raising upwards. And something ought to happen in this region. Thank you.