1 / 59

Tracking particles in space and time

"Explore new physics by tracking particles with advanced timing technology. Learn about 4D tracking, sensor variations, and the importance of timing for LHC experiments."

laguilar
Download Presentation

Tracking particles in space and time

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Tracking particles in space and time Besides a few indirect signals of new physics, particle physics today faces an extraordinary drought. We need to cross an energy- cross section desert to reach the El-dorado of new physics. Very little help in the direction of this path is coming from nature, the burden is on the accelerator and experimental physicists to provide the means for this crossing. Timing is one of the enabling technologies to cross the desert People of Silicon N. Cartiglia, INFN, Torino – 4DTracking, Bologna. The journey to new physics across the LHC desert

  2. The effect of timing information • The inclusion of track-timing in the event information has the capability of changing radically how we design experiments. • Timing can be available at different levels of the event reconstruction, in increasing order of complexity: • Timing in the event reconstruction Timing layers • this is the easiest implementation, a layer ONLY for timing • Timing at each point along the track  4D tracking • tracking-timing • Timing at each point along the track at high rate  5D tracking • Very high rate represents an additional step in complication, very different read-out chip and data output organization N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  3. One sensor does not fit all • Silicon sensors for tracking come in many shapes, fitting very different needs: • Spatial precision: from a few microns to mm (pixels, strips) • Area: from mm2 up to hundred of square meter • Radiation damage: from nothing to >1E16 neq/cm2 (3D, thin planar, thick planar) • Likewise, Silicon sensors for time-tracking are being developed to fit different needs with respect of time and space precision. The geometries above are combined with: • Very high time precision ~ 30-50 ps per plane • Good time precision ~ 50-100 ps per plane N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  4. Preamble: simulator Weightfield(2) Started at HEPHY http://www.hephy.at/en/innovation/detector-development/detector-simulation/ Further developed in Torino Available at: http://personalpages.to.infn.it/~cartigli/Weightfield2/Main.html It requires Root build from source, it is for Linux and Mac. It will not replace TCAD, but it helps in understanding the sensors response N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  5. Current situation at LHC: no real need for timing N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  6. Is timing really necessary at HL-LHC? The research into 4D tracking is strongly motivated by the HL-LHC experimental conditions: 150-200 events/bunch crossing N. Cartiglia, INFN, Torino – 4DTracking, Bologna. • According to CMS simulations: • Time RMS between vertexes: 153 ps • Average distance between two vertexes: 500 um • Fraction of overlapping vertexes: 10-20% • Of those events, a large fraction will have significant degradation of the quality of reconstruction At HL-LHC: Timing is equivalent to additional luminosity

  7. One extra dimension: tracking in 4Dimension N. Cartiglia, INFN, Torino – 4DTracking, Bologna. Timing complements tracking in the correct reconstruction of the events

  8. 4D tracking: Timing at each point • Massive simplification of patter recognition, new tracking algorithms will be faster even in very dense environments • Use only “time compatible points” N. Cartiglia, INFN, Torino – 4DTracking, Bologna. protons Z- Vertex distribution protons Timing

  9. 3+1 tracking: tracker + timing layer N. Cartiglia, INFN, Torino – 4DTracking, Bologna. Dedicated Layer(s) in the tracking Dedicated detector

  10. Silicon time-tagging detector (a simplified view) N. Cartiglia, INFN, Torino – 4DTracking, Bologna. Time is set when the signal crosses the comparator threshold The timing capabilities are determined by the characteristics of the signal at the output of the pre-Amplifier and by the TDC binning. Strong interplay between sensor and electronics

  11. Good time resolution needs very uniform signals Signal shape is determined by Ramo’s Theorem: Drift velocity Weighting field N. Cartiglia, INFN, Torino – 4DTracking, Bologna. The key to good timing is the uniformity of signals: Drift velocity and Weighting field need to be as uniform as possible Basic rule: parallel plate geometry

  12. Time resolution total current total current Subleading, ignored here parallel plate geometry electron current electron current hole current hole current Time walk: Amplitude variation, corrected in electronics Shape variations: non homogeneous energy deposition Usual “Jitter” term Here enters everything that is “Noise” and the steepness of the signal N. Cartiglia, INFN, Torino – 4DTracking, Bologna. Need large dV/dt

  13. Signal formation in silicon detectors We know we need a large signal, but how is the signal formed? What is controlling the slew rate? N. Cartiglia, INFN, Torino – 4DTracking, Bologna. • A particle creates charges, then: • The charges start moving under the influence of an external field • The motion of the charges induces a current on the electrodes • The signal ends when the charges reach the electrodes

  14. What is the signal of one e/h pair? (Simplified model for pad detectors) Let’s consider one single electron-hole pair. The integral of the current is equal to the electric charge, q: + + - - However the shape of the signal depends on the thickness d: thinner detectors have higher slew rate N. Cartiglia, INFN, Torino – 4DTracking, Bologna. Thin detector i(t) Thick detector D t • One e/h pair generates higher current in thin detectors d Weighting field

  15. Large signals from thick detectors? (Simplified model for pad detectors) Thick detectors have higher number of charges: d + + + + + + + - - - - - - - However each charge contributes to the initial current as: N. Cartiglia, INFN, Torino – 4DTracking, Bologna. The initial current for a silicon detector does not depend on how thick (d) the sensor is: D Number of e/h = 75/micron velocity • Initial current = constant Weighting field

  16. Summary “thin vs thick” detectors (Simplified model for pad detectors) Thin detector d + + + + + + + - - - - - - - S Thick detector N. Cartiglia, INFN, Torino – 4DTracking, Bologna. i(t) D Thick detectors have longer signals, not higher signals We need to add gain

  17. Gain needs E ~ 300kV/cm. How can we do it? 1) Use external bias: assuming a 50 micron silicon detector, we need Vbias = ~ 600 - 700 V Difficult to achieve N. Cartiglia, INFN, Torino – 4DTracking, Bologna. 2) Use Gauss Theorem: E = 300 kV/cm  q ~ 1016 /cm3 Need to have 1016/cm3 charges !!

  18. Gain in Silicon detectors • Gain in silicon detectors is commonly achieved in several types of sensors. It’s based on the avalanche mechanism that starts in high electric fields: V ~ 300 kV/cm • Gain definition: • = itis the inverse of a distance, strong function of E DV ~ 300 kV/cm N. Cartiglia, INFN, Torino – 4DTracking, Bologna. Concurrent multiplication of electrons and holes generate very high gain - - + + + - - + - + + - - - - - + - - + - - + + + - + + - - - • Silicon devices with gain: • APD: gain 50-500 • SiPM: gain ~ 104

  19. Electric fields in Silicon sensors Gain happens when the Efield is near the critical values, 300 kV/cm 3 methods to increase Efield: Doping in the bulk Doping in the gain layer Bias N. Cartiglia, INFN, Torino – 4DTracking, Bologna. • The “low gain avalanche diode” offers the most stable situation • Gain due to interplay between gain layer and bias

  20. Standard vs Low Gain Avalanche Diodes N. Cartiglia, INFN, Torino – 4DTracking, Bologna. The LGAD sensors, as proposed and manufactured by CNM (National Center for Micro-electronics, Barcelona): High field obtained by adding an extra doping layer E ~ 300 kV/cm, closed to breakdown voltage

  21. Gain layer: a parallel plate capacitor with high field • The doping of the gain layer is equivalent to the charge on the plates of the capacitor. • Bias adds additional E field n implant Gain layer p implant Gain: exp(field * distance) High Efield Drift space Shallow gain layers work at higher E field. Q2 E2 E1 Q1 N. Cartiglia, INFN, Torino – 4DTracking, Bologna. HPK HPK FBK d2 Shallow GL Deep GL n implant d1 Doping density [a.u] Gain layer p implant Depth [a.u.] • Examples of gain layer shapes from a few of our samples. • GL differs for depth and width: both parameters are important.

  22. Mean free paths in different situations Multiplication in the Bulk: ~ 170-200 kV/cm Deep Gain Layer: ~ 290 – 300 kV/cm Shallow Gain Layer: ~ 380 – 400 kV/cm Massey model: The position of the GL determines the field working point: the deeper it is, the lower the field is Bulk N. Cartiglia, INFN, Torino – 4DTracking, Bologna. Deep Gain Layer (DGL) Shallow Gain Layer (SGL)

  23. How gain shapes the signal Gain electron: absorbed immediately Gain holes: long drift home + - Initial electron, holes + - + - N. Cartiglia, INFN, Torino – 4DTracking, Bologna. • Electrons multiply and produce additional electrons and holes. • Gain electrons have almost no effect • Gain holes dominate the signal •  No holes multiplications

  24. Gain current vs Initial current  Go thin!! (Real life is a bit more complicated, but the conclusions are the same) Full simulation (assuming 2 pF detector capacitance) N. Cartiglia, INFN, Torino – 4DTracking, Bologna. 300 micron: ~ 2-3 improvement with gain = 20 !!! Significant improvements in time resolution require thin detectors

  25. Gain and Signal current i(t) medium N. Cartiglia, INFN, Torino – 4DTracking, Bologna. thick thin t3 t2 t1 t The rise time depends only on the sensor thickness ~ 1/d

  26. Effect of Temperature: Gain Layer data 50V 90V N. Cartiglia, INFN, Torino – 4DTracking, Bologna. Using the data above, the coefficient b can be determined: It takes a larger Bias increase for FBK UFSD (shallower gain layer, lower dl/dE) to compensate for a change in temperature than for HPK UFSD

  27. Ultra Fast Silicon Detectors UFSD are LGAD detectors optimized to achieve the best possible time resolution Specifically: Thin to maximize the slew rate (dV/dt) Parallel plate – like geometries (pixels..) for most uniform weighting field High electric field to maximize the drift velocity Highest possible resistivity to have uniform E field Small size to keep the capacitance low Small volumes to keep the leakage current low (shot noise) N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  28. Physical limit to time precision: Non-Uniform Energy deposition • Fluctuations in ionization cause two major effects: • Amplitude variations, that can be corrected with time walk compensation • For a given amplitude, the charge deposition is non uniform. • These are 3 examples of this effect: N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  29. UFSD time resolution summary • The UFSD advances via a series of productions. • For each thickness, the goal is to obtain the intrinsic time resolution • Achieved: • 20 ps for 35 micron • 30 ps for 50 micron Resolution without gain N. Cartiglia, INFN, Torino – 4DTracking, Bologna. UFSD1 UFSD2, 3

  30. UFSD time resolution UFSD from Hamamatsu: 30 ps time resolution, Value of gain ~ 20 N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  31. Non-Uniform charge deposition • Consider a configuration with a given gain, for example gain = 10 • The time resolution changes “within” the same Landau. • Larger signals in a Landau have: • more variation in the energy deposition • Less uniform signals • Worse time resolution N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  32. UFSD group: FBK – Trento Uni – INFN-To UFSD1: 300-micron. First LGAD production at FBK. Gain layer study, edges UFSD2: 50-micron. Gain layer doping: Boron, Gallium, Boron + Carbon, Gallium+Carbon UFSD3: 50-micron, produced with the stepper, many Carbon levels, small dead space UFSD31: 50-micron, exploration of edge termination RSD1: 50-micron, readout in AC mode UFSD3 UFSD2 UFSD1 N. Cartiglia, INFN, Torino – 4DTracking, Bologna. RSD1 UFSD31

  33. Irradiation effects • Irradiation causes 3 main effects: • Decrease of charge collection efficiency due to trapping • Doping creation/removal • Increased leakage current, shot noise • We need to design a detector that is able to survive large fluences, up to ~ 1E15 neq/cm2 N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  34. Acceptor removal Unfortunate fact: irradiation de-activate p-doping removing Boron from the reticle Boron Radiation creates Si interstitial that inactivate the Boron: Si_i + B_s  Si_s + B_i N. Cartiglia, INFN, Torino – 4DTracking, Bologna. Two possible solutions: 1) use Gallium, 2) Add Carbon Gallium is substitutional From literature, Gallium has a lower possibility to become interstitial Carbon is substitutional Interstitial Si interact with Carbon instead of with Boron and Gallium

  35. Effect of acceptor removal Acceptor removal, Gain layer deactivation N. Cartiglia, INFN, Torino – 4DTracking, Bologna. To some extent, the gain layer disappearance might be compensated by increasing the bias voltage

  36. Impurity engineering of radiation resistance more radiation resistance: B+C N. Cartiglia, INFN, Torino – 4DTracking, Bologna. less radiation resistance Carbon addition works really well, increasing by a factor of 2-3 the radiation hardness Gallium is actually is not more rad-hard than Boron

  37. Acceptor Removal: the full story N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  38. Acceptor removal data Acceptor removal coefficient The removal of acceptors depends on the acceptors density the removal is less important for higher densities N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  39. Electronics: What is the best pre-amp choice? Energy deposition in a 50 mm sensor • Fast slew rate • Higher noise • Sensitive to Landau bumps Charge Sensitive Amplifier Current signal in a 50 mm sensor N. Cartiglia, INFN, Torino – 4DTracking, Bologna. Current Amplifier • Slower slew rate • Quieter • Integration helps the signal smoothing

  40. UFSD performance N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  41. Noise in irradiated sensors Time resolution in LGAD is determined by jitter and charge non uniformity: The jitter term contains electronic noise and Current noise: N. Cartiglia, INFN, Torino – 4DTracking, Bologna. Jitter • Current noise: noise due to the combination of • High leakage current  Shot Noise • Randomness of multiplication mechanism  Excess noise factor

  42. Noise as a function of fluence and gain Goal: the noise from Silicon current should stay below that of the electronics N. Cartiglia, INFN, Torino – 4DTracking, Bologna. Data and model look similar.

  43. From a pad to a tracker: CMS ETL Endcap Timing Layer N. Cartiglia, INFN, Torino – 4DTracking, Bologna. 7 m2 of sensors on each side A circle obtained with long staves • ~ 16000 sensors: • 2x4 cm2--- small sensors • Thickness of active area: 40-50 microns • Pad size: 1.3 x 1.3 mm2 (512 pads)

  44. Fill factor The gap is due to two components: • 1) Adjacent gain layers need to be isolated (JTE & p-stop) • 2) Bending of the E field lines in the region around the JTE area • Both under optimization Different junction termination/p-stop design • CMS Goal: 30 micron gap = 96% fill factor N. Cartiglia, INFN, Torino – 4DTracking, Bologna. Gain JTE JTE+p-stop @20 micron: 99% of full gain @15 micron: 90% total signal JTE Different gain doping p-stop pad isolation Very aggressive design: <10 micron per side

  45. Fill factor solution: trenches • Trenches (the same technique used in SiPM): • No pstop, • No JTE  no extra electrode bending the field lines No gain area Current version R&D goal N. Cartiglia, INFN, Torino – 4DTracking, Bologna. JTE + p-stop design Trench design

  46. AC-LGAD with a resistive n++ electrode  E and Ew fields are very regular Segmentation is achieved via AC coupling AC coupling n++ electrode N. Cartiglia, INFN, Torino – 4DTracking, Bologna. gain layer p++ electrode The AC read-out sees only a small part of the sensor: small capacitance and small leakage current.

  47. Signal formation: fast signal + slow discharge Fast signal: same shape as the equivalent DC sensor. The shape depends on the weighting field. Slow discharge: depends on the read-out RC. Small RC have larger and shorter positive lobes short RC long RC Slow discharge Fast signal e- N. Cartiglia, INFN, Torino – 4DTracking, Bologna. h+

  48. 5D tracking: 4D tracking + very high rate One last twist of complication: 4D tracking at very high rate requires multiple TDC per bin, very high data transfer and a lot of power. Unfortunately, as soon as you say: “we can do 4D tracking”, the community asks for high rate too… N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  49. Summary and outlook Timing layers, 4D- and 5D- tracking are being developed for the next generation of experiments It is a challenging and beautiful developments, that requires a collective effort to succeed. There is no “one technology fits all”: depending on segmentation, precision, radiation levels and other factors the best solution changes. It would be great if in our journey we stumble upon a highway, to take us out of the desert Full bibliography: http://personalpages.to.infn.it/~cartigli/NC_site/UFSD_References.html N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

  50. Extra N. Cartiglia, INFN, Torino – 4DTracking, Bologna.

More Related