Inclusive-cluster \sigmalongPb\ distribution as a function of \pt\ in data for \pp (bottom left frame) and \PbPb collisions for two different centrality classes 0--10\% central (top left frame) and 30--50\% semi-central (top right frame). The dotted line corresponds to the tight value of the upper selection limit for single photon candidate clusters (narrow clusters)and the dashed line corresponds to a looser photon upper selection used in \PbPb collisions below 18 \GeVc. The dotted-dashed line corresponds to the narrow cluster's lower limit.

Inclusive-cluster \sigmalongPb\ distribution in data (black bullets) and PYTHIA 8 simulation (jet--jet+$\gamma$--jet processes, blue squares). The four panels display these distributions for two selected cluster \pt\ ranges, $12\pt14$ \GeVc on the left and $40\pt60$ \GeVc on the right, and two collision systems: \pp (top) and \PbPb 0--10\% central (bottom). The simulation is decomposed in its different particle origins: prompt $\gamma$ ($\gamma^{\rm prompt}$, green line), not merged decay $\gamma$ ($\gamma^{\rm decay}$, blue area), merged decay photon clusters ($\gamma\gamma$) from \piz (red area) or $\eta$ (brown area) The threshold value $\sigmalongPb = 0.3$ or 0.392 (corresponding to $\sigma_{\rm max}^{2}$(13 \GeVc)) is shown on all plots as a dotted or dashed vertical line, respectively.

Schematic view of the UE estimation areas considered in the analysis and for the estimation of the associated systematic uncertainty. The radius gap $\Delta R_{\rm UE gap}$ (see text) is also illustrated.

$\rho^{\rm UE}$ distributions normalised by the number of events in each centrality class in data calculated in the $\eta$-band with $\Delta R_{\rm UE gap}=0.1$ out of an isolation cone centred at inclusive clusters with $\pt > 10$ \GeVc. Left: for \PbPb collisions and the cone radius $R=0.2$ as a function of centrality. Right: for \pp and \PbPb collisions for different centrality classes for the cone radius $R=0.4$.

\ptIsoCh\ distribution for narrow clusters with $0.1\sigmalongPb0.3$ for $\pt > 16$ GeV/$c$,in \pp (bottom left frame) and \PbPb collisions in two centrality classes, 0--10\% central (top left frame) and 70--90\% peripheral (top right frame), with $R=0.2$ (black bullets) and $R=0.4$ (blue squares),in data (full markers) and simulated PYTHIA 8 $\gamma$--jet (open markers), embedded into data in the considered centrality class for the \PbPb collision case.

Purity of the isolated-photon sample as a function of \ptg\ calculated using Eq. \eqref{eq:ABCDpurityMC} and with the statistical and systematic uncertainty discussed in Sect. \ref{sec:sys_unc} for $R=0.2$ and $0.4$. The curves (plain for low \ptg, dashed for high \ptg) are the two sigmoid functions as defined in Eq. \eqref{eq:sigmoid}, obtained by fitting the points as explained in the text.

Contributions from reconstruction, identification, and isolation to the total efficiency calculated using Eq. \eqref{eq:efficiency}, as a function of the reconstructed photon \ptg for \pp (bottom left frame) collisions and \PbPb collisions for two centrality classes: 0--10\% (upper left frame) and 70--90\% (upper right frame). Green markers: reconstruction efficiency $\varepsilon^{\mathrm{rec}}$ Red or brown markers: efficiency additionally due to the photon identification by shower shape selection $\varepsilon^{\mathrm{rec}} \times \varepsilon^{\mathrm{id}}$ or the isolation criterion $\varepsilon^{\mathrm{rec}} \times \varepsilon^{\mathrm{iso}}$ Blue markers: efficiency due to the isolation criterion and shower shape selection $\varepsilon^{\mathrm{rec}} \times \varepsilon^{\mathrm{id}} \times \varepsilon^{\mathrm{iso}}$ Black markers: fraction $\kappa^{\mathrm{iso}}$ of generated photons which are isolated The efficiency is obtained from PYTHIA 8 simulations of \pp collisions $\gamma$--jet processes, embedded into data in the considered centrality class for the \PbPb collision case.

Total isolated-photon efficiency as a function of \ptg\ calculated using Eq. \eqref{eq:efficiency} with the systematic uncertainty discussed in Sect. \ref{sec:sys_unc} for $R=0.2$ and $0.4$, for \pp collisions and \PbPb collisions for five centrality classes. The efficiency is obtained from PYTHIA 8 simulations of \pp collisions $\gamma$--jet processes, embedded into data in the considered centrality class for the \PbPb collision case.

Left: \tr trigger efficiency for inclusive clusters and isolated-narrow clusters with $R=0.2$ (similar for $R=0.4$) obtained with PYTHIA 8 simulations combining $\gamma$--jet and jet--jet processes, embedded in data in the considered centrality class for \PbPb collisions and considering the corresponding trigger thresholds ($E_{\rm trig}$) in each system. Right: \tr trigger rejection factor calculated by applying the trigger efficiency for \pp collisions and for each of the \PbPb centrality classes considered. Solid lines over points result from a constant fit, values given in Table \ref{tab:Lumi}, dashed lines indicate the fit uncertainty obtained with the procedure explained in Sect. \ref{sec:sys_unc}. For clusters above 12 \GeVc and peripheral \PbPb collisions, the rejection factor and trigger efficiency are calculated for the sum of the two triggered samples with thresholds at 5 and 10 GeV.

Contributions to the systematic uncertainty of the isolated-photon purity and their quadratic sum as a function of \ptg for $R=0.2$, in \pp collisions (bottom left frame) and two \PbPb collisions centrality classes: 0--10\% (top left frame) and 30--50\% (top right frame). The statistical uncertainty is also shown and may appear larger than the total uncertainty as a result of the smoothing done by the fit over the purity (see text).

Contributions to the systematic uncertainty of the isolated-photon cross sectionand their quadratic sum as a function of \ptg for $R=0.2$ in \pp collisions (bottom left frame) and two \PbPb collisions centrality classes: 0--10\% (top left frame) and 30--50\% (top right frame). Statistical uncertainty is also shown.

Left: Isolated-photon differential cross section measured in \pp and \PbPb collisions at \snnfive for five \PbPb centrality classes for $R=0.2$. Error bars and boxes are the statistical and systematic uncertainties, respectively. The bands correspond to NLO pQCD calculations with JETPHOX, for \PbPb collisions calculated for the 0--100\% centrality class and scaled by \Ncoll. Right: Ratio of data over JETPHOX NLO pQCD calculations. The bands centred at unity correspond to the JETPHOX pQCD calculations, their width represents the scale (blue) uncertainty and PDF (orange) uncertainty. The normalisation uncertainties are not included in the left panel but they are shown in the right panel as a grey box on the left of each of the frames around unity.

Left: Isolated-photon differential cross section measured in \pp and \PbPb collisions at \snnfive for five \PbPb centrality classes for $R=0.4$. Error bars and boxes are the statistical and systematic uncertainties, respectively. The bands correspond to NLO pQCD calculations with JETPHOX, for \PbPb collisions calculated for the 0--100\% centrality class and scaled by \Ncoll. Right: Ratio of data over JETPHOX NLO pQCD calculations. The bands centred at unity correspond to the JETPHOX pQCD calculations, their width represents the scale (blue) uncertainty and PDF (orange) uncertainty. The normalisation uncertainties are not included in the left panel but they are shown in the right panel as a grey box on the left of each of the frames around unity.

Ratio of isolated-photon cross section measured with $R=0.4$ over $R=0.2$ for \PbPb and \pp collisions at \snnfive. Each panel for each \PbPb collisions centrality class, bottom right panel for \pp collisions. Error bars and boxes are the statistical and systematic uncertainties, respectively. The violet band corresponds to pQCD calculations with JETPHOX, the width represents the PDF uncertainty, and the blue vertical bars indicate the statistical uncertainty from the Monte Carlo approach.

Isolated-photon cross section ratio of pp collisions at \sseven from  (left) and \sthirteen from  (right) over \sfive in data and NLO calculation from JETPHOX for $R=0.4$. Error bars and empty boxes are the data statistical and systematic uncertainties, respectively. Filled boxes represent the theory scale (orange) and PDF (pink) uncertainties.

Nuclear modification factor \raa for isolated photons at \snnfive for isolation-cone radii $R=0.2$ (black) and $R=0.4$ (blue). Error bars and boxes are the statistical and systematic uncertainties, respectively. The isolated-photon \raa is compared to that of charged particles  and charged pions  from ALICE, and to the ratio of the \zz-boson yield in each centrality class to the 0--90\% class measured by CMS . The bands correspond to pQCD calculations with JETPHOX for \PbPb collisions (nPDF) for 0--100\% centrality over pp collisions (PDF). The width of each band corresponds to the scale and PDF uncertainties The normalisation uncertainties are represented as a red box centred at unity The solid line in the most peripheral centrality class 70--90\% at \raa = 0.82 corresponds to the HG-PYTHIA model expectation .

Nuclear modification factor \raa for isolated photons measured by ALICE for isolation-cone radii $R=0.2$ and $R=0.4$ and CMS  for isolation-cone radius $R=0.4$ at \snnfive and for four centrality classes. Error bars and boxes are the statistical and systematic uncertainties, respectively. The solid line in the peripheral centrality class 50--90\% at \raa = 0.91, is the result of the HG-PYTHIA model . The ALICE normalisation uncertainties are represented as a red box centred at unity. For CMS, the normalisation uncertainties are displayed as a violet box for the integrated luminosity and a green box for the nuclear overlap function $\langle T_{\rm AA}\rangle=\Ncoll / \sigma_{\rm NN}^{\rm INEL}$.