This is Hua Pei. I am glad to have a chance to introduce our on-going ZDC detector R&D to you. About the simulation work on high-charge fragments, here is the basic workflow we use;

1) First, the lower-energy (relative to RHIC) AA collisions are generated using the isospin-dependent quantum molecular dynamics (IQMD) model. Large charge (Z>5) fragments are produced by mechanism similar to coalescence, i.e. nucleons are grouped into big ions if they are close in coordinate and momentum space. The parameters we use now are if \Delta position < 3.5 fm and \Delta momentum < 250 MeV/c. Meanwhile, only such fragments composed of nucleons, including single protons and neutrons, are saved in the output of IQMD model in our case to simplify the simulation work, since pion yield is trivial in the forward region.

* Introductions of the IQMD model can be found in papers including: 
"Modelling the many-body dynamics of heavy ion collisions: Present status and future perspective", Eur. Phys. J. A 1, 151 (1998). doi:10.1007/s100500050045, arXiv:nucl-th/9811015
and applications of the IQMD model can be found in papers like:
"Azimuthal-sensitive three-dimensional HBT radius in Au–Au collisions at Ebeam=1.23A GeV by the IQMD model", Eur. Phys. J. A 58, 81 (2022). doi:10.1140/epja/s10050-022-00722-w

2) Then, we have the option to de-excite these fragments and let them decay into smaller(-charge) fragments, which we usually did. Before this de-excitation, there are often two large charge fragments in each event, one from projectile and one from target, and usually one of them fly forwards into the inner ring of ZDC acceptance. After this de-excitation, these large charge fragments mostly decay into smaller charge fragments (but still Z>5), and they flying direction are more diverse into both inner and outer rings of ZDC.

* Please note that as Yaping already described, our current ZDC geometry is not really "zero degree", but rather cover from a few degree (inner ring) until 10~15 degree (outer ring) around the projectile beam axis.

3) Then, we make these fragments, including single protons and neutrons, the input of GEANT4 and run with plastic scintillator. Currently the GEANT4 is being run within the framework of FairROOT 18.0.6, and GEANT4 10.4.1. We modified the running script to manually add all fragments (Z<=92, Z<=A<=3Z) into a list of "FairIon", and then manually add them into FairROOT by "FairRunSim->AddNewIon". Currently I don't modify the default list of interactions in the FairROOT.

* By looking at the log file, I can find the following mechanism are applied during GEANT running: hadronic processes for pi/K/\Lambda/gamma/electron, and proton/neutron/deuteron/triton/He3/Alpha, plus their anti-particles. But nothing specific for Z>2? 

4) In the output of GEANT4, the energy loss of each ion in the plastic scintillator, was well fit by a paranoid curve, as E ~ Z^2. Note this is calculated before any quenching and energy-to-photon efficiency is manually implemented.

* The quenching effect is now being estimated by work like RAA 2019 Vol. 19 No. 3, 47(10pp) doi: 10.1088/1674–4527/19/3/47, Fig. 15.
* The energy-to-photon efficiency are considered only of intra-channel discrepancy, mostly depending on the positions of energy deposition and their distance to the PMT window. The inter-channel discrepancy is expected to be largely eliminated by calibration.


Hope my introduction of ZDC can do some help. Please don't hesitate to ask if any text is not clear. We also prepared a few documents describing the reconstruction of centrality and event plane, and Yaping shall send them to you soon.


Cheers
Hua Pei