Midrapidity dip and limiting fragmentation of charged hadrons from RHIC to LHC energies

The charged-hadron pseudorapidity density $\dndeta$ in central heavy-ion collisions has a midrapidity depression that deepens with collision energy. Whether it reflects structure in the rapidity density $\dndy$ or only the kinematics of the $y\toη$ transformation has mostly been argued with effective or model-based Jacobians. We test this without an event generator, with a $y\toη$ kernel built from measured identified-hadron spectra. In central Au+Au and Pb+Pb collisions from $\snn=19.6$~GeV to 5.36~TeV the data do not require a dip in $\dndy$: the decisive null-hypothesis family gives $p\ge\pMinAll$ in all variants. For dips up to one rapidity unit wide, the depth giving $p<0.05$ in half the pseudo-experiments is \rng{\dminLHClo}{\dminLHChi}\% at the LHC and \rng{\dminRHIClo}{\dminRHIChi}\% at 62.4--200~GeV. Generator closure tests show a false dip at top RHIC energy caused by the forward region. As the LHC data barely constrain the fragmentation region, an energy sum rule calibrated on RHIC data, with the RHIC shape of that region, gives best-fit LHC/RHIC ratios of \rng{\sLoE}{\sHiE}, \rng{\sLoF}{\sHiF} and \rng{\sLoG}{\sHiG} at 2.76, 5.02 and 5.36~TeV for three shape families, with 95\% intervals over the families of \rng{\sNFLoE}{\sNFHiE}, \rng{\sNFLoF}{\sNFHiF} and \sNFG{} that include unity, although at 2.76 and 5.02~TeV the fits prefer \rng{\pullLoEF}{\pullHiEF}$\,σ$ more energy than calibrated. Over plausible RHIC covariances and kernels the best-fit ratios span \rng{\sSystAllLo}{\sSystAllHi}, and single variants exclude unity at \systUnionAbs: limiting fragmentation is neither established nor excluded.