Objective for this project is to perform calculations that will be base for a NASA proposal on improvement of data acquisition and analysis for the existing high-energy linearly polarized photon polarimeters for astrophysics applications and for development of two new prototypes that will have an analyzing power and efficiency of 10% for E up to 250 MeV; one based on pair production on the nucleus and another on triplet production on electrons.
Motivation for this project are our previous results obtained on the polarimeter that we already built and the strength of the team that we have to explore fundamental challenges of pair production and detection and their implications for the development of a medium to high energy photon polarimeter. Dr Hunter is a well known NASA expert who designed several medium to high energy polarimeter prototypes.i,ii,iii,iv B. Wojtsekhowski and B. Vlahovic already developed a high energy photon polarimeter which was successfully implemented at Jefferson National Laboratory nuclear experiments. That polarimeter has the highest analyzing power ever achieved for the energies of few MeV up to few GeV.v The concept of the polarimeter and the theoretical work is shown in our NIM papervi.
Proposed Research: a) Improvement of the existing polarimeters, data acquisition and data analysis: We calculated the cross section as a function of the φ± angle, which is the angle between the polarization plane and a vector between a positron and an electron crossing the detector plane, normal to the photon momentum. (See Figures 1 and 2 below.) Our numerical integration shows that the obtained asymmetry in terms of φ± is considerably larger than in terms of φ+- which is used with all other methods. The calculation of the production cross section as a function of that angle shows that the asymmetry is larger by factor of 1.4 than the asymmetry from any other known polarimeter method. It is also important to note that in this formalism the asymmetry does not depend on the beam profile and the position of the vertex can be unknown. The analyzing power is about 0.26 for the thin radiator and equal energies of electron and positron. During our test run at the Spring-8 facility, the analyzing power was determined to be 19.2% for equal-energy pairs and 11.9% for all events.
The figure-of-merit of the polarimeter is the product of the efficiency and the second power of its analyzing power. We are proposing to reanalyze previously collected data using our method which will increase figure-of-merit by at least a factor of 2. We are also proposing to rewrite the algorithm for data acquisition for further data taking. b) A microstrip polarimeter for linearly polarized medium energy photons: We considered a polarimeter made of many layers of Si micro strip detectors. In the proposed polarimeter each layer is an active converter, and a veto counter, and a tracker. Preliminary estimation suggests that this polarimeter will have an analyzing power and efficiency of 10% for an Eof 250 MeV.
Efficiency of the pair polarimeter is proportional to thickness of the converter and could be estimated as the thickness of the converter in units of radiation length of its material. Typical converter efficiency used in our experiment is 0.1 %. Analyzing power averaged over energy sharing between electron and positron for very thin converter was found to be of 0.14. Scattering in the converter washes out the correlation between the photon polarization plane and the pair plane. Analyzing power is slowly reduced for the large converter thickness which practically compensates the increase of converter efficiency. This effect was calculated for the photon energy up to 2 GeV and is energy independent.
Figure 3. The layout of the photon polarimeter. The photon arrives from the top. Inter layer distance is 20 mm. The system of microstrip detectors serves as a veto, converter, and a tracker. Calorimeter is providing a trigger for the DAQ.
Event selection and DAQ logic: The polarimeter scheme is presented in Figure 3. The figure illustrates the event when the track of two particles is traced to layer #3, in which the track presented only in one narrow cluster. Layer #2 does not have a signal, which used a veto and identification of the photo-induced event. Analysis of the track geometry with layers #3, #4 and #5 will be used for the determination of the photon polarization. The signals from all strips will be digitized and searched for tracks that ended in the calorimeter modules #4 and #6, and initialized a trigger. The track whose origin is inside a stack of layers is indicating a candidate of electron-positron pair produced by a photon.
Triplet events: A small portion (on order of 10%) of pairs will be produced from interaction with atomic electrons. Because the momentum transfer to the electron is very small, most of these events are impossible to detect. When the recoil energy reaches 1 MeV the recoil electron (triplet) is moving at angle 20-40 degree with the produced pair and can be identified. Triplet events could be confused with production of delta electrons by components of the pair, however, it is easy to resolve by tracing the vertex of triplet and pair tracks. Even if the number of triplet events will be very small they will provide an independent polarimetry method within the same detector system.
c) A Triplet Polarimeter for Linearly Polarized Photons: We considered a polarimeter prototype based on triplet pair production. Preliminary estimation suggests that a polarimeter based on liquid Ne that will have an analyzing power of 14% and an efficiency of 0.3% could be constructed.
Concept: Use of triplet photon polarimetry was proposed in 1972.i Several groups tried to develop an experimental device for polarization measurements of the beam and one test result was reported.ii We present here very briefly a scheme of the polarimeter prototype that we propose to build, which is in some aspects similar to the MEGA project polarimeter.
We considered the detector made of multi-layer TPC chamber which is filled with liquid Ne (radiation length of 24 cm, density is 0.9 g/cm3). Most likely Ar is also a possible option. Liquid methane should also be investigated. A total of 48 cm of Ne will provide 85% efficiency for photon detection. Each Ne layer has thickness of 4 cm and the readout planes at both ends. The ionization drift of 2 cm occurs before a particle reaches a readout plane. The liquid plays the role of an active converter and also a veto counter, and a tracker.
The analyzing power for the triplet process was found to be 0.141±0.006.iv The efficiency of the triplet process was often considered to be 1/Z times less than for pair polarimeter. In an actual device it is even lower, because the majority of recoil electrons have very small energy. This means that the cross section is 0.02 re2 per electron, or 3% of the total for triplet electrons.
Triplet vs Pair Polarimeter: Presented above is likely close to the best possible realization of the triplet option of the photon polarimeter. Similar efficiency of 0.3% could be obtained by the pair polarimeter with just 5 layers of Si MSD. When MSD are spaced by 20 cm the required two-track resolution is 100 m which is well achievable with 50 m interstrip distance. For 1m2 detector of MSD we will have 103x20x10(groups)x5(planes)x3(xyz directions) = 3x106 readout channels.
This system will have an analyzing power of about 10%, which corresponds to the figure-of-merit twice lower than for the proposed liquid triplet polarimeter. Please note that the proposed Xe gas polarimeter that is not cryogenic based will have about 0.4 radiation lengths which will reduce the efficiency by a factor of 5. Use of Xe will further lower the efficiency by an additional factor of 5 which will make the figure-of-merit lower by a factor of 25 in comparison with the proposed liquid polarimeter.
Described is the concept of a polarimeter for any photons with energy above 10 MeV. The polarimeter with analyzing power 14% and 0.3% efficiency should include 12 layers of the liquid Ne detectors with two-track resolution of 200 m. Such a system will be a reliable instrument in energy range which up to now was not investigated at all. The parameters of the proposed system are within existing technological capabilities.