Asked Simone Pagan Griso Simulations of BSM convener

Asked Simone Pagan Griso, Simulations of BSM convener for Snowmass, and she put me in touch with some useful contacts to simulate our BSM particle reach. Simon Knapen, Jose Zurita, Brian Shuve, Dave Curtin, Zhen Liu, Laura Jeanty. All responded except Zhen Liu. Typical letter I sent: Dear Laura, Some of us are developing a proposal to add a new subdetector to CMS to search for longer-lived or even stable new particles in the very forward region. We submitted the attached Eo. I to Snowmass. We believe this is superior in some regions of {coupling, mass} space to other searches, e. g. Central detectors, FASER, MATHUSLA, NA 62 etc. A very strong point is a 20 m long 1 m diameter vacuum decay volume followed by precision tracking. Would you be able to help us with such a plot or plots? Simone Griso suggested that this might fit with your plans for the Snowmass process. Basically, as you can see in the note, we potentially have three modes, see Fig. 2: A) Production angle at LHC-14 Te. V between 1 mrad and 4 mrad, all momenta. Penetrating 50 m of steel (300 lambda_int) B) and decaying in vacuum between z = 96 - 116 m. All decay modes possible and any masses (probably up to about 25 Ge. V? ). B) Production angle 0 - 1 mrad, all momenta, vacuum all the way, decaying between z = 82 m and 116 m but the decay products have to have radial distance from beam line at z = 116 m >12. 5 cm to be in the detectors (all phi between R = 12. 5 cm and 50 cm. ) C) Production angle at LHC-14 Te. V between 1 mrad and 4 mrad, all momenta. Penetrating 50 m of steel (300 Lambda_int) and not decaying but interacting in calorimeter 26 lambda_int (e. g. ). So semi-strongly interacting, maybe 10^-3 hadronic. Are any such particles ruled out? Could be stable or very long lived. Time-of-flight can help reject the neutron etc background. We assume zero background for A and B (C will have backgrounds we need to study - if it is worth looking). Thanks for any simulations you can help us with. For m(A') > 1 Ge. V so above pi 0 and eta decays what are the main expected production mecahnisms? q-qbar annihilation, quark bremsstrahlung? Other? Best regards, Mike

From: Brian Shuve Assistant Professor of Physics Harvey Mudd College Thanks for your email! This seems like an interesting proposal for a subdetector, and could be a great addition to existing experiments or proposals! I will be honest - I don’t know that I’m the right person for this. Part of it is that I don’t have a lot of expertise with simulations in the forward region and am not tooled up for this particular thing. Also, my faculty position is at an undergraduate college with a heavy teaching load (I am currently teaching the equivalent of 3 classes) – as you can imagine, between moving classes online, and some ongoing family situations, I have little capacity to take on something new. In terms of suggestions, I know that Jared Evans is an expert in long-lived particles and has studied some forward processes. Likewise, Brian Batell might know some folks who can help because he was involved in simulations for projections for MATHUSLA from B meson decays, even though he’s not affiliated with the experiment (I don’t think). Tien-Tien Yu had a student who was doing some simulations for a sub-detector near ATLAS - I’m not sure if anything came of that, but she might have suggestions for theorists who could help you out. Or I would be happy to find others if that would be useful. Regarding production modes, my understanding is that Drell-Yan is the dominant mechanism above a few Ge. V and bremsstrahlung is subdominant, but the Drell-Yan estimates become really sketchy below a few Ge. V because the PDFs aren’t really well constrained there (nor is that necessarily the right way to be modelling the p p collisions). See, for example, https: //arxiv. org/pdf/1804. 00661. pdf. I know that LHCb has gotten around this using a model-independent way of determining the A’ production cross section - essentially, they extract the A’ rate from the dilepton spectrum in the vicinity of the A’ (but away from the peak), using the fact that both the photon and A’ couple to the EM current. The method is based on this paper: https: //arxiv. org/pdf/1603. 08926. pdf. I wonder if you can try something similar? In any event, I’m happy to chat more if it would be useful. Thanks for thinking of me, and good luck with your study! Brian

From Jose Francisvo Zurita, KIT, Germany Many thanks for your e-mail and your Lo. I. I briefly flicked through it, but I think this is a great idea and I would be very happy to help. Full disclosure: #1: I have worked on similar environments (e. g; LHe. C, LHCb) but never looked in thevery forward direction of LHC or a beam-dump experiment (e. g: SHi. P). Yet this should not be a problem. #2: I am a full member of MATHUSLA, so I would need to be extra careful with statements of the order of “We will be much better than X, Y, Z” (I would like to avoid conflicts like those with ANUBIS) Regarding the possible models, there are 4 “portal” models people normally consider: scalars (from Higgs), pseudoscalars (axions), fermions (heavy neutral leptons) and vectors (dark photon). There are zillions of plots like these in the PBC report (or any talk by Gaia Lanfranchi). Let me focus on scalars and vectors for the moment. On scalars: main production will come from H(125) produced at LHC, and then H->SS decay. I guess we would only pick one S decay in the fiducial volume. The S decays are a bit model dependent, but the simplest example is to mimic a SM-like Higgs of that mass (see Fig. 4 of https: //arxiv. org/pdf/1809. 01876. pdf). so we would decay to bb above 10 Ge. V, and to cc / tau mostly. However one can consider leptophobic examples, or up-type couplings only, to get rid of some other bounds(e. g: LHCb rare decays). On the good side: Higgs cross section is “fixed” to about 60 pb, then we need to invoke the geometrical acceptance, let’s say 0. 003 mrad, so we have about 1 E-3 geometrical acceptance (if production were isotropic, which is not true, but let’s keep things simple). HL-LHC would do 3000 fb-1, so 2 E 8 Higgs, so we would have 2 E 5 events, which means (requiring sth like 20 events) that we could test a exotic BR of Higgs of about 1 E-4. ATLAS + CMS can only do 2. 5 % if I recall correctly. So we are on business (LHCb could be a contender here, but this is a function of the mass). The vectors are normally produced from qq, and their decay modes are also model dependent, namely one can charge SM particles under the new U(1) [or not]. The simplest option is using only photon mixing, so one would look for pair of charged particles (ee, mu mu, tau, qq. hadrons, etc). In both cases I think the window of opportunity would be for very low masses, because that is where ATLAS / CMS don’t have a good reconstruction. I hope this is useful to start a dialogue. We can discuss at some point if you want, or get me in touch with other members of the collaboration. Best Regards, José

From Laura Jeanty, response also to Simone Pagan Griso Dear Simone, Michael, Thanks for initiating the communication! Indeed, I plan to generate some dark photon samples to use as a benchmark to study the ILC’s sensitivity to LLPs. More specifically, to have any chance of being competitive, I’m interested in the dark Higgs / Higgs portal mode of dark photon production, and I’m aiming for the ~10 Ge. V-ish range for where the ILC might be able to extend beyond existing forward detectors / beam-dump experiments. I attach my LOI here in case it’s useful. I asked Simone if anyone else had a working setup or events I could borrow, but it seems we are the two groups working on dark photons under the LLP snowmass umbrella, so I guess we have to figure it ourselves. . . As a first outline, my plan is to see if I (or my student) can use David Curtin’s madgraph setup here: http: //insti. physics. sunysb. edu/~curtin/hahm_mg. html to generate some events, and then I’ll see if I can put them through the full ILC simulation. I’m happy to share anything useful I learn and/or produce along the way, but I’m by no means an expert on the generation side, and progress will be slow from my side due to other commitments. best, Laura

From Dave Curtin: Hi Michael, The Dark photon simulations you describe are somewhat nontrivial unless you’re already tooled up for them – anything involving material interactions or dark photons below 12 gev takes some thought and possibly simulations beyond running plug and play madgraph in our HAHM dark photon model. Unfortunately I’m completely overwhelmed and I really cannot take on any more work right now. Perhaps you could reach out to the faser or codex people, they’ve focused more on dark photons than I have and might be tooled up to do some of this in a time efficient manner. But I know they’re very busy too. All that being said, this stuff is ultimately not hard so if you find a grad student who has time to spend on this that should work too. Best of luck, David

From Simon Knapen: Dear Mike, Good to hear from you. I am interested, but unfortunately I am also very overloaded at the moment, and I hesitate to make any promises. This would be a minimum of a few days to a week of work for me I think, and I don't have this amount of time in the near future I am afraid. For the MC, was the script I sent you in August helpful in any way? If you have a student or postdoc who could implement the simulation, I could meet with them to train them. The production modes and lifetimes are model dependent. For a dark photon above 1 Ge. V, indeed annihilation and brehmstrallung appear to be most important. This is described in detail in the FASER papers. For a dark scalar, B decays can be very important, as well as Higgs decays, though the latter may be too central for your setup. For option C, I suppose you have a short range interaction in mind? It depends on the mass of the particle, but I suspect it will be very difficult to build a model that can do this without running into problems with exotic meson decays and/or early universe cosmology. Best regards, Simon