Written by Todd Huffman for JPhys+, based on the paper Tagging b quarks at extreme energies without tracks in JPhysG.

In the summer of 2015 the centre of mass energy of the Large Hadron Collider (LHC) at CERN increased to 13 TeV from 8 TeV  (1 TeV is 10¹² electron-volts of energy). My research is to search for massive states of matter that decay into two quarks or gluons producing exceptionally high energy jets of particles in their wake. One such event is shown in Figure 1 where you can easily see two back-to-back jets of particles streaming from the collision point at the centre of the detector.

Figure 1. Shown is the end result of a proton-proton collision within the ATLAS detector at the LHC. Two “jets” of particles are the result of the collision in this case. ATLAS Experiment © 2016 CERN

There are many theories that allow for the existence of such massive particles, but my primary interest is driven by experiment. Regardless of any theory, a detector like the ATLAS detector shown in Figure 1 is constructed such that we would see such massive particles if they do in fact exist.

The backgrounds though are considerable, so we have to look for something distinctive in high-energy jets.  It turns out that in many theories there are exotic particles that preferentially decay into B hadrons, sub-atomic particles containing the bottom quark (which is the fermion partner to the top quark). B hadrons survive for about 1.5 picoseconds (1.5 x 10^-12 s) before decaying themselves into less exotic hadrons which can then be detected as part of the jets like those in Figure 1.

1.5 picoseconds does not seem like a long time, but that is only the average decay time when the hadron is at rest. Within a high energy jet the B hadron, which has a mass around 5 GeV/c², will have an energy higher than 500 GeV.  This great energy means the B hadrons are travelling very close to the speed of light and therefore, to us in the lab, their average lifetime increases more than 100-fold. 150 pico-seconds might not seem like a long time either, but it means that the B hadron can travel several cm away from the collision before it decays.  A great deal of effort in the ATLAS, CMS, and LHCb experiments goes into finding these B hadrons within such jets by looking for the vertex from the decay of these hadrons.  This is called “b-tagging” the jet or the event.

The problem with such exceptionally high speeds, and slow decay times, is that the B hadrons can traverse the inner layers of the detector before they decay. This is illustrated in Figure 2 where a particle comes in from below, crosses a detector which registers a hit, and then decays between detector layers; so the next detector further out registers an increased number of hits from the many decay products of the B hadron decay.

Traditional b-tagging methods try to use the hits obtained to reconstruct the particle tracks, but if the B hadron decayed after the first layer or two there will be missing layers or even worse, the tracks will use false hits from inner layers which were not actually from the correct particle. As a result of this effect and also other relativistic effects, the efficiency of track-based b-tagging starts to drop as the energy increases.

Figure 2. Shown schematically here is a particle traversing a pixel layer from the lower left and decaying before the next layer, causing multiple hits to appear. For this technique to be effective the particle should decay into many daughter particles. B hadrons have this property. © IOP Publishing, All Rights Reserved

Our idea was to see if, rather than even trying to form tracks from the dense environment within a high energy jet, we could identify B hadrons by only looking to see if there is an increase in hit pixels between detector layers.

We studied this using a simulation of a 4-layer pixel detector and a set of exceptionally high energy jets, up to 5 TeV per jet. One pixel layer is essentially the same type of detector that is used in your mobile phone camera, but specially hardened to survive the radiation environment of the LHC. Figure 2 also illustrates the principle between two of the layers. The earlier layer has only one hit in Figure 2, while the next layer out has multiple hits. This shows up as a “multiplicity jump” in the number of hits. By using the difference in the number of hits divided by the number of hits in the inner layer, we can look for multiplicity jumps between any two pixel layers in the neighbourhood of previously found jets.

The study optimised the size of the multiplicity jump used and the results are shown in Figure 3 as the efficiency of successfully tagging a jet containing a B hadron. This efficiency is calculated with respect to B hadrons which did decay within the region where it was possible to detect them. So a value of 50% means that ½ of all B hadrons that could have been detected were detected. Not only is this a fairly high efficiency, it remains viable out to much higher energy. We think the slow drop-off starting at 1.5 TeV is actually due to B hadrons with such high energy, thus living so long, that they survived the entire trip through all 4 layers of our detector!

Figure 3. The fiducial efficiency for finding B hadrons using the multiplicity jump method with an optimised cut. Also shown is the probability that a light-quark jet would pass the same cut. © IOP Publishing, All Rights Reserved

There are many backgrounds which can dilute this technique. One of the biggest we looked into was the possibility that a perfectly regular hadron (not a B hadron) would strike an atomic nucleus within the detector and produce a shower of particles that would then appear as multiple hits in the next layer. This effect does occur and appears in Figure 3 as the “efficiency” of mis-tagging a jet which does not have a B meson.

We are currently engaging with the ATLAS collaboration to improve our study with a more realistic detector simulation and also to include other effects that we were unable to simulate for the paper. Should these studies also prove successful we will have, at a stroke, increased the sensitivity of these detectors to ultra-high energy B hadrons dramatically.

Read the full paper
Tagging b quarks at extreme energies without tracks

The Authors

Prof. Todd Huffman, Mr. Charles Jackson, and Prof. Jeff Tseng

Prof. B. Todd Huffman is an experimental particle physicist at Oxford University and tutorial fellow at Lady Margaret Hall.

Mr. Charles Jackson is a master’s student, specializing in areas of Condensed Matter and Particle Physics, at the University of Oxford Department of Physics.

Prof. Jeff Tseng is an experimental particle physicist at Oxford University and tutorial fellow at St. Edmund Hall.

This research was carried out at the Particle Physics sub-department of Oxford University with support from the UK government through STFC and HEFCE.

This work is licensed under a Creative Commons Attribution 3.0 Unported License

Figure 1. ATLAS supplementary data. ATLAS Experiment © 2016 CERN, reused with permission.

Figure 2 and 3. From Tagging b quarks at extreme energies without tracks B Todd Huffman et al 2016 J. Phys. G: Nucl. Part. Phys. 43 085001. Copyright IOP Publishing.

Categories: Journal of Physics G: Nuclear and Particle Physics

Tags: ATLAS, CERN, Hadron physics, High energy physics, Instrumentation, LHC, Particle collisions, Particle physics, Particle physics phenomenology, Simulation, Standard model