ALICE 2 detector systems (see legend and text for details).

Accumulation of integrated luminosity over time for different trigger types in pp (left) and \PbPb (right) collisions during LHC Run 2.

Time frame and heartbeat frame structure in continuous and triggered mode. HeartBeat (HB) triggers are issued in continuous and triggered modes to all upgraded detectors. Physics triggers can be sent to upgraded detectors in triggered mode and are sent to non-upgraded detectors in all modes. HBF and TF rates are programmable with the following nominal values; HBF: 1 every orbit, $\sim$89.4 µs/$\sim$10 kHz, TF: 1 TF every 128 HBFs/$\sim$11 ms/$\sim$100 Hz.

ALICE readout and control system architecture.

Block diagram of the common readout unit (CRU): the CRU forms the interface between the first-level processors (via PCIe), the central trigger system (via TTS), and the detectors (via TTS and FE).

Picture of a CRU; bottom left, FPGA cooling radiator, bottom right, power mezzanine; top row; 3 out of 8 Minipods installed; top left, fiber optics cable to MPO connector on front panel; bottom left, SFP transceivers.

Schematic cross-section of a pixel cell.

Architecture of the ALPIDE chip.

Block diagram of the ALPIDE chip.

Diagrams of the ITS2 inner barrel and outer barrel modules.

ALPIDE sensor chip detection efficiency and fake-hit rate vs global threshold setting. Beam test results (\SI{6}{\giga eV/\it{c}} pions, orthogonal incidence). ALPIDE substrate reverse bias: \SI{-3}{\volt}.

ALPIDE chip hit-position resolution and average cluster size as a function of global threshold setting. Beam test results with \SI{6}{\giga eV/\it{c}} pions with perpendicular incidence. ALPIDE substrate reverse bias: ${\rm -3 V}$.

Block diagram of the SAMPA ASIC.

Block diagram of the front-end implemented in the SAMPA ASIC.

Block diagram of the SAMPA SAR ADC.

Diagram of digital signal processing chain in the SAMPA.

SAMPA bare die (left) and TFBGA packaged chip (right).

Example response curve for 4\,mV/fC configuration.

Schematic layout of the ITS2. The three innermost layers form the inner barrel, the middle and outer layers form the outer barrel.

Layout of the staves of the inner and outer barrels.

Inner HIC seen from the sensor side. The green tabs are used for fixing and handling, and are removed before mounting the HIC.

Top view and bottom view of an outer barrel HIC. The yellow cables connect the HIC to the power bus.

Pictures of the ITS2 assembly. Top left: A view of the MAM used for chip inspection and HIC assembly. Top right: Photo of an outer barrel stave, with power bus cables opened on the two sides. Bottom: Photo of an inner barrel stave, with detailed views shown in the insert.

Space frame and cold plate cooling scheme (Left) Inner Barrel; (Right) Outer Barrel.

Azimuthal distribution of single contributions to the material budget of an inner (left) and outer (right) barrel stave layer. The relative contribution of each component to the total material budget is quoted.

Overview of the mechanical structure of the ITS2. The upper panel shows the Inner Barrel, the Outer Barrel staves and the MFT in the back. The lower panel shows the IB and OB conical structural shells supporting the respective services.

Schematic design and interconnections of the ITS2 readout unit.

Schematic description of the stave production workflow.

ITS2 in the clean room during on-surface commissioning. The lower left shows a zoomed-in view of the half barrels of the outer barrel (OB) and inner barrel (IB) type.

Fake-hit rate of an inner half-barrel as a function of the number of masked pixels.

Top: Outer barrel surrounding the beam pipe with the Muon Forward Tracker (MFT) in the background. Bottom: ITS2 inner barrel bottom half-barrel next to the beam pipe, outer barrel and MFT in the background.

Event display of a cosmic muon traversing all layers of ITS2 twice, no magnetic field.

Longitudinal distribution of the primary vertex positions from ITS2 tracks reconstructed online during the LHC pilot beam in October 2021.  .

Schematic view of the Muon Forward Tracker (left) and its integration with the central barrel (right).

Detailed view of the elements composing the MFT detector.

Example of an assembled MFT ladder. Upper picture: back side of the FPC with the glue spots on which the sensor are glued. Bottom picture: front view of assembled ladder.

Exploded view of a half-disk (D00-01).

Left: Half-disk during ladder gluing. Right: Glue deposition pattern.

Half-cone structure with air ducts in light blue.

MFT PSU boards with mezzanine assembled on their support.

Noise occupancy, measured in hit/pixel/event as a function of the number of masked pixels over the whole MFT detector.

Display of the reconstructed muon tracks in the MFT from a TED shot event.

Schematic view of the ALICE TPC.

Schematic view of a stack with four GEM foils. The baseline settings for the voltages across the four GEMs, the transfer fields between the GEMs, and the induction field between GEM\,4 and the pad plane are indicated as well.

Energy resolution $\sigma(^{55}\rm{Fe})$ as a function of ion backflow (IBF) in a 4-GEM stack (S-LP-LP-S) in Ne-CO$_2$-N$_2$ (90-10-5) The gas gain is kept at 2000 in all measurements by adjusting the voltages on GEM\,3 and GEM\,4 at a fixed ratio of 0.8 or 0.95. Figure from  .

Exploded view of an IROC with chamber body components and GEM frames. Figure taken from .

GEM design details. The detail on top shows the 200\,$\mu$m separation between the copper sectors and an additional clearance of 100\,$\mu$m for the GEM holes from the copper edge Figure taken from .

Detailed powering scheme of a GEM stack Each subsequent high-voltage channel is stacked on top of the lower-lying channel The ground reference is defined by a separate line connected to the ground of the detector The line for GEM\,4 top is shunted with a resistor ($R_{\text{shunt}}$) inside the high-definition current meter Each line is connected to the detector through a decoupling resistor ($R_{\text{dec}}$) The signal from a calibration pulser is coupled via a capacitor ($C_{\text{pulser}}$) to the line for GEM\,4 bottom Individual loading resistors ($R_{\text{load}}$) are mounted on all segments on the top sides of the GEMs Figure taken from .

Schematic view of the TPC readout system Five SAMPA chips amplify, shape, and digitize the current signals picked up on the connected pads Two GBTx ASICs multiplex the digitized data GBTx\,0 forwards the data from two and a half SAMPA chips to a VTRx GBTx\,1 forwards the data from the other two and a half SAMPAs to one VTTx module (two optical transmit channels) GBTx\,0 also receives configuration data and the reference clock through the VTRx The reference clock is distributed to the other components A GBT-SCA chip is used for monitoring and configuration Figure taken from .

Layout of the final TPC FEC PCB Rev.\,1a The components are mounted on both sides of the board The figure shows the top side with five SAMPAs, two GBTx, one GBT-SCA, one VTRx, one VTTx and some other components On the bottom side a few additional small components and the connectors to the detector are placed. Figure taken from  .

TPC d$E$/d$x$ as a function of momentum $p$. Online plot from quality control during pilot run with 900\,GeV collisions in October 2021. No track selection criteria were applied The magnetic field value was of 0.5\,T.

Measured shape of the ion tailin data recorded with laser tracks.

Visualization of the ion tailand of the common-mode effect for one readout channel zoomed around the baseline region The data are from a simulation with 30\% pad occupancy and without noise from theelectronics The common-mode effect shifts the baseline to below zero. The ion tail is visible forsignals with large amplitude Both effects are corrected in the CRU FPGAs.

View of the FIT detectors illustrating the relative sizes of each component. From left to right FDD-A, FT0-A, FV0, FT0-C, and FDD-C are shown. Note that FT0-A and FV0 have a common mechanical support. FT0-A is the small quadrangular structure in the centre of the large, circular FV0 support. Note that all detectors are planar with the exception of FT0-C, which has a concave shape centered on the IP. The inset table lists the distance from the interaction point and the pseudorapidity coverage for each component.

Photograph of one half of the FV0. The optical fibers connect the scintillators to the PMTs on the rim of the support structure, the black structure seen here. The center wall has been removed to show the scintillator, the surface matrix structure, and the optical fibers.

Schematic diagram of the FIT readout electronics. The MicroChannel Plates (MCP) are described in the FT0 section.

Left: Station 2 of the tracking system; the readout electronics are distributed on the surface of a quadrant. Right: Stations 4 and 5 of the tracking system; the readout electronics are mounted along the top and bottom edges of the slats.

The Muon Tracking readout scheme.

The two geometries of the DualSAMPA boards (DS12 on the left and DS345 on the right), with the white connector plug socket on PCB and on the other side the black connector connecting to SOLAR boards.

Left: Flex mounted on a slat connecting five DualSAMPA cards linking through a green flat ribbon cable to the readout board. Right: Large electronic PCBs covering the surface of a quadrant.

The FLEX scheme.

The SOLAR board: functional diagram (left panel) and photo of the board itself (right).

Data flow scheme.

CRU scheme.

Overview of one MID half-plane in open position.

FEERIC ASIC architecture (top) and FEERIC card picture (bottom).

Wireless threshold distribution scheme (left) and master or node electronics card (right).

MID readout architecture (left) and readout cards (right picture) with local (LOC) (top right), regional (REG) (top left) and J2 bus (bottom) between local and regional.

The status of TRD supermodules concerning capacitors in the high-voltage distribution.

Induced common-mode signal with and without capacitors for the anode high-voltage. The baseline of pads in the same high-voltage segment as a cosmic-ray particle with an integrated signal between 10\,000 and 12\,500 ADC counts is shown before (red) and after (blue) the removal of the capacitors. For comparision, the baseline from pads in a high-voltage segment without hits is shown in green.

Left: $\Delta\alpha$ versus impact angle for a typical TRD chamber in Run 2, having a fixed uncalibrated drift velocity. The quoted values refer to the Run 2 calibration procedure (upper row) and to the new calibration scheme (lower row). Right: Average $\Delta\alpha$ versus impact angle for all TRD chambers after the calibration was applied. The red band shows the RMS of the distribution.

(left) HPTDC programming in Run 1 and 2 operations (top arrow) and in Run 3 (bottom arrow). The three trigger levels L0, L1 and L2a are replaced by a periodic trigger with a given frequency, mimicking a continuous readout. All hits (black lines) are read out and can be associated to physical events at a later stage. (right) Possible selection of parameters (fixed trigger frequency $f_\mathrm{T}$ and matching window width $m_\mathrm{w}$) to realize a continuous readout. The green circle corresponds to the chosen point of operations.

Hit time within the orbit of randomized hits sent at fixed rate to HPTDC inside a TOF crate operated in continuous readout mode. No holes are seen in the distribution, which means that random hits are received in all 3564 bunch crossings through the whole LHC orbit.

The DRM2 card: on the left the VTRX transceiver and the GBTx ASIC (covered by a heat dissipating panel). On the right the ARM piggy-back card is visible. The additional optical receivers for the SCL and the LHC clock are in the middle of the front panel.

From left to right: HMPID Front-End Electronics (FEE), Readout Control Board (RCB) with the readout FPGA, the TTCRx and the Source Interface Unit (SIU). On the right, the C-RORC cards installed on O$^{2}$ FLP computers.

HMPID full DAQ structure. Four C-RORC boards are installed on two First Level Processor (FLP) computers of the O$^2$ data acquisition environment. Fourteen optical links connect the 14 RCBs, two per RICH module.

Event rate as a function of the detector occupancy, which is on average about 0.17\% and 2\% in \pp and \PbPb collisions, respectively.

Expected relative statistical uncertainty on the anti-deuteron absorption cross section in \pPb and \PbPb collisions for the full Run 3 data sample and for one half of the Pb--Pb sample.

Schematic view of the EMCal, consisting of two disjunct detector segments, in the top-left hemisphere and the bottom-right hemisphere (DCal), covering approximately opposite locations in azimuth. The PHOS calorimeter inside the DCal segment is indicated in brown.

SRU readout rate as a function of interaction rate for different Multi-Event Buffering (MEB) schemes. The left panel shows simulated data, while the right panel shows measured performance.

PHOS detection element consisting of a PbWO$_4$ crystal and APD with preamplifier (left) and a fraction of a cell matrix of one PHOS module (right).

Digitized signal waveform from one channel of the PHOS FEE board. Sampling time is 100\,ns, digitization of sampled amplitude is 10 bits.

PHOS time resolution dependence on photon transverse momentum during the Run 2 data taking.

Performance of the digitizer during Pb--Pb 2018 data taking in the operating conditions chosen for Run 3. On the left plot: the lower part of the triggered spectrum of ZNC common photomultiplier in Pb--Pb collisions where the emission of a single $\SI{2.76}{\tera\electronvolt}$ neutron and multiples are visible. The spectrum is fitted to a superposition of gaussian functions whose peak positions $\mu_i$ are related to the neutron multiplicity by the relation $\mu_i=\mu_{1n}\times i$ and their widths by the relation $\sigma_{i}=\sigma_{1n}\sqrt{i}$, where $i$ is the neutron multiplicity and $\mu_{1n}$ and $\sigma_{1n}$ are the mean and the r.m.s. of the single neutron peak, respectively. The autotrigger algoritm effectively rejects pedestal events. On the right plot: the arrival time of ZNC common photomultiplier signals w.r.t. the reference ALICE L0 trigger signal.

The cage: a support structure for beam pipe, ITS2, and MFT. Shown is the cage (magenta) with the bottom half of ITS and MFT as well as the beam pipe already installed.

Beam pipe installed for the ALICE 2 detector with an outer diameter of \SI{36}{\mm}.

Overview of the components of the O$^2$ data read out and processing systems and the main data flows.

Synchronous reconstruction workflow.

Simultaneous dataflows inside the FLP from the CRUs to the DDR memories and from the memory to the Infiniband network to the EPN farm.

O$^2$ Quality Control design.

AliECS design.

The O$^2$ computing system monitoring design.

The ALICE CR0 data centre which houses the EPN farm.

Network diagram of the Run 3 O$^2$ facility.

Data distribution software framework.

Photograph of a CTP module.

Trigger system overview.

Access to the hardware implemented in ALF-FRED mechanism.

Transverse (solid circle markers) and longitudinal (open square markers) impact parameter resolution in \PbPb{} collision data with ALICE 1 (blue points) and simulations the upgraded detector, ALICE 2 (red points).

Performance projections for measurements of the nuclear modification factor $R_{\rm AA}$ and elliptic flow coefficient $v_{2}$ (see text for details) for charm and beauty hadrons .

Performance projection for the ratio of beauty baryon and meson yields in central \PbPb{} collisions with the total expected integrated luminosity of 10 nb$^{-1}$ (at full magnetic field) for Run 3 and 4. From .

Projection of the invariant mass distribution of electron-positron pairs before (left panel) and after (right panel) subtraction of the contribution from decays of light and heavy flavour mesons, except the $\rho$ meson. From .

Projected yield ratio of $\psi$(2S) and J$/\psi$ in pp and \PbPb{} collisions, measured with the MFT and muon arm . Model calculations from .

Projected limits on jet energy loss from the measurement of the yields of jets recoiling from high-\pt{} trigger particles in different collision systems. From .