LC calorimeter R and D EoI reviews (2002) -- Dhiman Chakraborty

Overview of LC Calorimeter R&D Priorities

Perhaps the most important issue for LC detectors is that of energy flow (EF) and how agressively to pursue it. The new challenge for LC calorimeters is to improve jet resolution to about 0.3/sqrt(Ejet). The TESLA group has shown that this can be achieved using a highly segmented ECal and a highly segmented HCal. The techologies chosen to implement the segmentation are Si-W for the ECal (both TESLA and SD), "digital" HCal (both TESLA and SD), and scint.-tile HCal (TESLA).

It is fair to say that the resolution quoted above, obtained for q-qbar final states, is not a robust result and it is not clear how it extrapolates to more complicated final states. Hence, simulations are critical now for a number of reasons:

to verify that jets can be decomposed into charged and neutral components in the calorimeter; and to apply the EF to a variety of final states
to use these results to feedback to detector design parameters
to determine if EF can be effectively applied to alternative hardware designs
to improve the methodology toward the theoretical limit of jet resolution of 0.15/sqrt(Ejet)
evaluate physics benchmarks; identify key processes with jet final states; when are beam constraints of limited value?

To attack this set of issues systematically requires a sustained effort with a certain critical mass of researchers. Therefore, it is important to encourage a number of groups to become involved in addressing the points above.

For groups we wish to advance a technology different from the highly segmented approach, it is important that they first demonstrate a comparable level of performance for jet physics. Otherwise, one might be tempted to judge that it is more a technology R&D than one which will advance the LC effort. Of course, another argument might be that the highly segmented approach is too expensive, in which case proponents should still demonstrate the performance.

On the hardware side, a full calorimeter of the type envisioned by TESLA and SD for implementing the agressive EF has never been built. There is one top-priority challenge for each technology:

for the Si-W ECal -- how does one integrate the order 50 M detector elements with a readout in a reasonable way
for the digital HCal -- what detector technology to use
for the scint.-tile HCal -- how to readout the tiles efficiently and without breaking the bank

These points are being addressed in Europe and now also in N. America. The last point is also being addressed in Asia. One implicitly assumes here that the EF will be shown to be robust and widely applicable, in which case the technologies above should presently be considered the leading ones.

Radiation should not be an issue, but all designs need to address being immersed in a 3 to 5 T field.

There are some generic calorimeter issues which need to be addressed and so far have not received sufficient attention:

timing requirement -- effect of pileup background
forward calorimetry (the masks and within) and hermeticity requirements
interplay with tracker and tracking group on energy flow

LCRD EOIs

Contact	Title/Comments	FY03 request
Onel	Cerenkov compensated calorimetry	89,000
	The idea is an interesting one: To provide event-by-event compensation by exploiting two different measurements of shower development, one by ionization and one by Cerenkov light. The Cerenkov measurement would allow one to correct for fluctuations in the neutral/charged ratio of hadronic showers and thereby provide a response correction to the primary ionization-based measurement. The technique is not a proven one in practice, but has shown promise. The main issue for this proposal is demonstrating the potential use of this technique for the LC environment, where one plans to use the tracker to measure the charged hadrons. In this case, the issue of improving hadronic energy resolution in the calorimeter is reduced. The proposers suggest 3 applications for the LC. I find (1) to be interesting: To provide a correction for the hadrons which interact in a crystal ECal, which I see to be the main drawback of crystals. In effect, on might hope this could compensate for the lack of longitudinal segmentation. CMS has shown (Kunori at Calor02) that the transverse segmentation of its crystal array might be sufficient to utilize energy flow. As discussed in point (3), there may be need for good timing resolution which can be provided by a Cerenkov readout, depending of course on the other technologies involved. My suggestion is that the proponents concentrate on applications of this technique which are clearly value added to current LC plans. Since several groups in the US are now proposing a crystal ECal R&D, then point (e) in the proposal seems appropriate, as discussed above. Otherwise, it is important to be clear whether the proponents plan to improve by modification detector ideas which are presently under consideration, such as tile-scint HCal designs, or to propose a new detector concept. If the latter, it is important to provide a plausible scenario for the overall improvement one could achieve in performance for jet resolution and/or cost. Note: Indirect costs are not (explicitly) included in the budget.
Mallik	Exploring Crsyal Calorimetry for a Linear Collider Detector	49,000
	So far, crystals have not been proposed for the LC. This is presumably because an important physics case for excellent EM resolution has not been apparent, as is H->2 gamma at LHC. However, it is interesting to explore the potential tradeoffs for crystals for jet measurement at the LC, basically improved photon resolution vs. limited segmentation. In the proposal, it would be good to include information (eg in Table 1) on the relative light yield and the nuclear interaction length. The latter is an important parameter for determining how important is the loss of fine longitudinal segmentation. This is a reasonable and welcome proposal. The proponents are leading experts in this area. My main suggestion is that more emphasis be placed on the physics simulation. Most of us are rather familiar with the performance of crystals per se (even though the long. segmentation studies are interesting), but I'm not sure anyone really knows what kind of jet reconstruction and resolution is achievable. CMS has done studies applying energy flow which showed big improvements, but the LC is an entirely different jet environment. Since the cost of crystals per volume unit (X0 times cm2) is comparable to silicon/tungsten, it is important to perform simulations on benchmark processes like WW,ZZ -> jets and compare to Si/W to determine the relative merits. Note: indirect costs are not (explicitly) included in the budget.
Butler	Study of Energy-Flow Algorithms for Linear Collider Calorimetry	39,426
	The proposal is to begin simulation studies involving energy flow reconstruction. Additional manpower is badly needed for this effort, which is a key issue for the LC detector. If the proponents can concentrate effort on this problem, then this is a very welcome contribution. The proposal is pretty vague and there are some misunderstandings, which hopefully will quickly be cleared up pending some study, involvement in working group meetings, etc. An example is the statement that the granularity for LC detectors is to "accurately characterize shower shape," while it is primarily intended to separate energy depositions. The initial goal to "demonstrate the specified jet energy resolution" for energy flow is not a trivial one. My suggestion is to get involved with the working groups and to find an important and rewarding piece of the simulation/reconstruction as an initial goal. Since the group has background in muons, it might be interesting to look at the issue of how to track MIPs through the (ECal and) HCal as part of energy flow. We note that the proponents are also included in the RPC EOI (Repond, et al.). The suggestion is to also collaborate with these folks on the simulation front. Note: Indirect costs are included (ouch).
Frey	Development of a silicon-tungsten ECal test module for an electromagnetic calorimeter	35,000
	The proponents wish to study the Si-W option for the ECal. From the energy-flow point of view, the small Moliere radius of tungsten and the fine segmentability of silicon makes it a highly desirable combination for an ECal. Indeed, it is the preferred choice of TESLA and the LC SD detector. The main drawbacks of the choice are the high cost of silicon and the difficulty of reading out the very large number of pixels (~50M). The proponents believe that these challenges can be met by choosing the largest commercially available silicon wafers (~1000 pixels/wafer) and by bump-bonding to the wafer a readout chip that will perform both analog measurements and digitization for all the channels on that wafer in-situ. Production of a test-beam ready prototype module is envisaged in 3 years. Detailed simulation and algorithm developent for single particles and partons, as well as benchmark physics processes are needed to justify the technology choice and to optimize the detector parameters. This is another integral part of this well-rounded proposal. The proposal is balanced and well thought out. The proponents have vast experience in e+e- collider detectors and are among the most active members of the American LC community. Availability of alternative funding for much of the proposed activities adds to its strength. The funds requested are very modest considering of the weight of the task. Note: No indirect costs are included. No comments from Ray Frey.
Repond	Study of Resistive Plate Chambers as Active Medium for the HCal	32,000
	A good and welcome proposal. RPCs are the "obvious" detector to be used with a digital HCal... if only thay can be made to operate reliably. Perhaps glass RPCs are the answer in the relatively low-rate LC environment. Clearly this is important to evaluate. While a group in Europe is also evaluating this, there are enough parameters and nuances to RPC implementation that it is important to have parallel efforts. The proponents are ramping up as RPC experts. Some suggestions for the proposal. Explain how the engineers/technicians will be used. Will there be flexibility in the type of test module built, depending upon the availability of other LC detector prototypes? Will the prototype electronics include analog for comparison to digital? Please include a simple study which looks at projected LC singles rates and compare to required charge-up times and compare to existing applications -- show that a glass RPC makes sense (on paper) at the LC. Note: Indirect costs not (explicitly) included.
White	Digital Hadron Calorimetry for the Linear Collider using GEM Technology	52,000
	A highly segmentable, robust detector is key to the digital HCal approach being pursued in Europe and N. America. The idea of using GEMs for this is unique to this proposal at this time. Hence, this is a timely and important proposal. I think the outcomes of this work will be followed by many. The EOI plan is well thought out. For the proposal it would be nice to see more detail on what is expected for the electronics in terms of ease of integration in a small space, cost, heat load, etc. This is important since this is an area where RPCs have an apparent advantage. It is nice to see the explicit inclusion of a simulation component in the proposal. The development of expertise and techniques for energy flow are necessary for the entire effort. The proponents are already well integrated with LC working groups. Note: Indirect costs not (explicitly) included.
Nauenberg	Study and Develop Scintillator-Fiber Readout Scintillator Calorimetry with High Spatial Resolution	54,484
	The goal here is to determine if scintillating tiles can provide sufficient segmentation to do energy flow. The problem to date has been that tiles less than about 5 cm on a side have been seen to be difficult to fabricate and efficiently read out. The idea in this proposal is to stagger alternate tile layers to effectively increase segmentation. This element of the proposal is unique. Presumably the thrust of this proposal is for the ECal, with the tiles seen as a (hopefully) lower cost alternative to silicon. The segmentation proposed here does not approach what is achievable with silicon, still it may be sufficient for an ECal for a large-radius detector (~2 m ECal). The proposal includes simulation studies to examine the efficacy of this approach in the energy flow framework -- good. On the detector side, examining light collection as a function of fiber positioning on the tiles, tile thickness, and readout device will be useful. I encourage the proponents to have close ties with the efforts for the TESLA tile HCal, the JLC tile ECal, and the HCal tile proponents from NIU/NICADD. Also, I encourage the development of a model for estimating the cost of a full ECal using this approach, as it is likely to turn out to be useful. On the simulation side, it will be good to obtain parametrizations of performance as a function of segmentation. This is a good proposal. It is nice to see the combination of detector and simulation efforts, and the contributions from the university. Note: Indirect costs are included.
Onel	Micro-machined Vacuum Photodetectors	125,000
	This proposal should clearly be redirected to the ADRD program. It is truly generic in nature, and should be evaluated as such. If successful, it might find application to the LC, as well as elsewhere. Note: Indirect costs are not included.
Magill	Optimization of the HCal for a Future e+e- Linear Collider with Energy Flow Jet Reconstruction	13,000
	The proponents wish to (1) optimize HCal properties and parameters by evaluating energy flow reconstruction as a function of these parameters and (2) develop the energy flow algorithms themselves. This EOI shows a sophisticated approach to these points, reflecting in part the fact that the proponents' work in these areas is already underway. The work so far has been very positive and LCRD should encourage its continuation. Note: Indirect costs are not included.

UCLC EOIs

Contact	Title/Comments	FY03 request
Zutshi	Simulations, Prototyping and Energy Flow Algorithms for a Scintillating Digital Hadron Calorimeter	67,500 +53,500
	This EOI actually contains two elements, budgeted separately. The overall goal is to study hadron calorimetry for the LC, in particular to evaluate an approach using scintillating tiles. The two proposal elements are "prototyping" and "algorithm development". The proponents have considerable experience in both these areas, albeit not in e+e-, and have a program which is already underway. The idea of using scintillating tiles for a LC HCal is not unique, as it is being pursued for TESLA and for the JLC detector. What is unique here is the idea to use tiles as the active medium for a digital (or near-digital) HCal, where one uses only one (or a few) readout bits per tile to save on cost and complexity in order to provide a highly segmented device. Wisely, I believe, they choose to remain flexible in the tradeoff between readout bits and segmentation. Prototyping: The key issue here is readout. Can one establish a cheap readout scheme commensurate with the digital scheme? Or is the electronics chain able to provide additional bits "for free"? In the latter case, the project significantly overlaps with the TESLA tile HCal option. The question of readout width vs segmentation will likely be settled by a combination of simulation and prototyping efforts. Cost, particularly for the readout, will presumably be included in the evaluation. Algorithm development: The proposed work on energy flow algorithm development is key to the overall LC effort and is welcomed. The work here is already well underway and represents a significant contribution. It is good to see both parts of this proposal going on in parallel. This is a good proposal and the work proposed will have a long-term impact on LC calorimetry. The simulation and algorithm work is already well integrated with the LC working groups. Presumably, the scintillator prototyping will also benefit with cooperation with other groups pursuing scintillating tiles. Note: Indirect costs are included. No comments from Dhiman Chakraborty.
Karmgard	Fast Response Tile Scintillation Development for Calorimetry and Tracking in NLC Detectors	54,327
	This EOI is primarily one of generic R&D, rather than one specifically targeted to the LC. This is particularly the case for the second part of the proposed work (fiber optic waveguide optimization). On the other hand, the first part, involving tile scintillator readout, is possibly directly applicable to the LC, since there are many tile calorimeter designs being considered. My suggestion is to (1) redirect this proposal to the ADRD program and (2) determine where collaboration with group(s) developing scintillating tile prototypes would wish to include the ideas proposed here. Possibilities include NIU/NICADD, Colorado, and TESLA (Korbel). (Presumably the JLC tiles are too thin for this approach?) Note: Indirect costs are included.
Oreglia	Optimization of LC detector elements for physics analysis	15,000
	This EOI consists of two distinct parts: Simulation studies and development of readout electronics for calorimeter prototypes. The latter effort is also listed in the RPC proposal (Repond) and would not begin until year 2. Both parts are very focussed and well considered. The simulation effort will strive to make a sort of generic parameterized calorimeter which will allow one to compare a range of choices of segmentation (transverse and longitudinal) and detector material for a standard physics benchmark process. Presumably, these parameters could also include detector radius and B field. One issue not mentioned on the EOI is how one includes showers shape information, which presumably is important for the segmentation studies. The existing LCD fast simulation (eg available in root) includes average shapes with randomly thrown parameters. Some comparisons similar to the proposed studies were begun by Iwasaki. Systematic studies along these lines will be very interesting. Both parts of the EOI would represent excellent contributions. Is the manpower sufficient? I note that other groups have proposed simulation studies which are similar, but with different emphases. These include Illinois-Chicago (Varelas) and Kansas (Wilson). The SLAC group would presumably also be interested. Note: Indirect costs are included.
Varelas	Calorimeter R&D and physics studies for the LC detector	49,500
	This EOI proposes to examine detector parameter optimizations by way of Standard Model physics processes. In spirit, the proposed work is similar to that from U. Chicago (Oreglia), but the process here is quite vague. The year 1 activity of developing a GEANT4 package with general calorimeter configuration is unclear. Which calorimeter? Is the package different than that developed in the LCD root software, or from that being developed by NIU? I suggest the EOI be sharpened for the proposal after consultation with other groups working in this area (Chicago, NIU, SLAC). It is presumably known that the year 2 activity of developing prototype readout electronics also overlaps with that of Oreglia, et al. This collaboration should be mentioned by both groups. Note: Indirect costs are included.
Wilson	Investigation and Design Optimization of a Compact Sampling Electromagnetic Calorimeter with High Spatial, Timing, and Energy Resolution	52,000
	This EOI demonstrates a good grasp of the issues for an ECal at the LC. The range of issues discussed is quite broad, perhaps too broad, although it is noted that the interests of other groups overlap in some of the areas discussed, for example in examining detector optimization in terms of resolution (Chicago). However, with the possible exception of Colorado, this EOI is unique in proposing to examine timing and hermiticity requirements. Studies of pileup vs timing, for example, would be quite valuable. The proposed work also includes examining a scintillator/silicon hybrid ECal, where thin (1 mm) tile layers are ganged longitudinally and spatial resolution is augmented by silicon layer(s). This approach coincides rather closely with that of the JLC calorimeter, which also includes a segmented layer -- in their case either scintillating strips or silicon, although the JLC design requires a compensating Pb/scintillator sandwich. It is an interesting idea and worth pursuing with simulation studies, as proposed. Given the scope of possible activities, the year one plan is well focussed. The proposal should include the detail of how the performance comparisons will be carried out (e.g. a full GEANT4 model? based on existing models?) Again, a unique and probably generically useful contribution would be a timing requirement study. (One could also hope for a hermiticity study, too, since that is not obviously covered elsewhere.) Such activities should be supported. Note: Indirect costs are included.