22. Detail Attachments

22. Detail Attachments

c)Purpose

i)Develop the technology to track particles with fractional millimeter spatial resolution, in three-dimensional space, using the “Time Projection” technique in noble liquids.

ii)Develop applications of this tracking technology, focusing on low energy neutrino interactions, and specifically on the low energy solar neutrinos from reactions that dominate energy production in the Sun but have never been observed in a real-time experiment.

iii)Perform measurements in an existing 0.1m cryostat suitable for noble liquids down to 2K, to enable development of practical readout techniques.

iv)Pursue engineering studies for a cubic meter class detector that serves as an engineering prototype for studying the readout issues in all noble liquids, and will give adequate rate, when filled with liquid neon, to measure with moderate accuracy the neutrinos from the proton-proton (“pp”) fusion reaction in the Sun.

This work effort may support at a minimum level or concurrently, as appropriate, the Technology Transfer and Science Education missions of the Department of Energy.

d)Approach

Scientific Focus of the Detector Development

The first measurements in our small test cryostat have progressed far enough that we are able to formulate scientific goals and specify technical approaches that seem feasible to develop. Our goal is to achieve the first measurements of the low energy neutrinos from the reaction that dominates energy production in the Sun, the pp reaction. This has been a challenging task for many years, since although the rate is relatively large due to the large number of neutrinos, the energy of the neutrinos is very low, averaging about 200keV, and a very special detector is required to register the electrons scattered by the neutrinos in the detector medium. Very low backgrounds must be achieved in an energy region where many electrons are usually found due to Compton scattering of background photons. We address these problems by the high performance of our detector for low energy electrons, and the extraordinary purity of the cryogenic liquids neon and helium.

Solar Neutrinos from the pp Reaction

Several important experiments have studied the interactions of neutrinos from the Sun over the last several decades, beginning with Ray Davis’ pioneering Homestake experiment, which first detected a deficit in the number of electron neutrinos, and which led to his 2002 Nobel Prize. Most recently, the SNO experiment has provided compelling evidence that a fraction of the electron neutrinos produced in the Sun oscillate into other types of neutrinos during their eight minute journey to Earth, confirming the deficit observed by Davis some thirty years earlier.

Neutrino oscillation can only occur if neutrinos have non-zero masses. The unambiguous observation of oscillations from solar neutrinos, and now also from atmospheric and reactor-produced neutrinos, has provided us with the first tangible evidence of phenomena beyond the “Standard Model” of particle physics, which assumes that neutrinos are massless and do not oscillate.

As a result of these discoveries, neutrino physics is now one of the highest priority programs, both within the U.S and internationally. A precision, real-time measurement of the neutrinos produced by the pp reaction in the Sun is an important component of this program. The 2004 Multi-Division APS Study “The Neutrino Matrix”, for example, lists as the third of its three summary recommendations “the development of an experiment to make precise measurements of the low-energy neutrinos from the Sun”. The recommendation continues: “So far, only the solar neutrinos with relatively high energy, a small fraction of the total, have been studied in detail. A precise measurement of the low-energy neutrino spectrum would test our understanding of how solar neutrinos change flavor, probe the fundamental question of whether the Sun shines only through nuclear fusion, and allow us to predict how bright the Sun will be tens of thousands of years from now.”

A measurement of the pp solar neutrino flux, in comparison with existing precision measurements of the high energy 8B neutrino flux, will demonstrate the transition between vacuum and matter dominated oscillation, quantitatively testing a fundamental prediction of the standard scenario of neutrino flavor transformation. The pp experiment will also permit a significantly improved determination of θ12 and, together with other solar neutrino measurements, either a measurement of θ13 or a constraint a factor of two lower than existing bounds. A precise measurement of the pp neutrino flux, including a comparison with the solar luminosity measured via photons, will also provide stringent tests on the theory of stellar evolution.

Tracking in Three Dimensions

Our aim is to observe particle tracks with fractional millimeter spatial resolution, in three-dimensional space. The tracks are created as charged particles move through the detection medium, producing a trail of ionization charge along their path. There is a relation between the length of the track, the “range”, and the particle energy. A track may be as long as the scale of the detector, meters, or as short as one resolution element, if the range given by energy loss in the medium is of that order. The physical limit to the precision of this measurement is set by diffusion.

The “Time Projection” Technique

In an electric field the charge moves with a constant velocity, or “drifts”. Under an applied field, tracks in the detector volume move bodily along the field, finally arriving at a readout plane where the charge is measured at each point, or pixel, as a function of time. This technique reduces the three-dimensional tracking problem to a two-dimensional problem. The drift time is used to store the signals, so that the readout is serialized. The low particle flux in the underground laboratory allows the drift time to be long. This reduces the bandwidth of the system; reducing cost and power dissipation. The track can be distorted by macroscopic motion of the liquid. This motion is often dominated by convection caused by thermal gradients. We have made simulations of this effect, and find that it can be largely suppressed by establishing a thermal gradient aligned with the direction of gravity.

Liquid Neon and Liquid Helium as the Detection Medium

We are interested in detectors with a large mass, in order to achieve adequate rate for the pp neutrino reactions. This favors high density. However, the desire to observe tracks of very low energy particles favors low density. The liquid form of inert, or noble, elements is the optimal choice, although a number of non-polar liquids such as hydrogen, methane, etc. are also options.

The first problem encountered in designing a device depending on a long drift distance, a meter or more, is reducing the level of electron attachment on impurities in order to attain a sufficiently long electron lifetime. For noble liquids, chemical purification is easier, but the operating temperature is an equally important consideration. At the ultra-low temperatures of liquid neon and liquid helium, impurities have negligible vapor pressure, and there are no known limits to the electron lifetime, solving one of the biggest challenges in a liquid Time Projection detector.

A second advantage offered by these liquids is very low spatial diffusion as the tracks are drifted through the medium. The diffusion decreases as a function of the temperature of the charge carrier, and for liquid neon and liquid helium this permits fractional millimeter resolution even after drifting over several meters.

Unlike most materials, the work function for electrons in liquid neon and helium (also hydrogen) is negative. For a free electron inside these substances, the liquid is repelled and the electron is localized inside a cavity of vacuum, which is often called an electron bubble, or “e-Bubble”. This is a nanoscale object, displaying both classical and quantum aspects. The e-Bubble moves in an electric field at a speed given by Stokes’ Law, meaning that it remains in thermal equilibrium with the liquid, and allows the excellent spatial resolutions we desire. Since the e-Bubble is a mesoscopic object, it drifts at speeds of order a thousand times slower than are typical for electrons in other liquids. For the low rate, low background applications we are interested in, this is an advantage and allows us to serialize the readout as described below. The e-Bubble also displays the quantum states calculated for an electron in a box, with bound states and a well-defined photoionization threshold.

In order to go down to KeV detection energies, some charge amplification of the ionization signal is needed. While we wish to use the liquid phase for our detector medium, past efforts to obtain useful gain by avalanches in liquids have not been very successful. If we eject the charge into a lower density vapor phase however, significant gain can be obtained, using the Gas Electron Multiplier (GEM) devices for example. As an electron drifts towards a liquid-vapor interface in a two phase system, it comes to rest just below the surface, at a point where the image charge force balances the force from the applied electric field. The electron is “stored” here, until it diffuses or tunnels into the vapor, on a timescale determined by the temperature and electric field. For the liquids of interest, this timescale can be as long as seconds, and allows charges to be held at the surface, with the possibility of a gated release into the vapor for detection. This may be utilized to implement a scan in the readout plane that converts one coordinate to a time measurement, with only the orthogonal direction to be read out by segmentation of electrodes in that dimension. In this concept we are using time to express two coordinates: possible only if they are scanned on time scales differing by at least three orders of magnitude. We intend to use drift speeds in the liquid less than 0.1m/s and drift over distances of a meter or more. To give clean images, the rate of tracks in the detector must then be small. In practice, this is feasible only in the environment of an underground laboratory.

We are investigatinga number of schemes to implement the scan across the liquid surface, and have so far considered three options: a light beam in the form of a line focused in the readout plane, which photo-ionizes the localized electron at the liquid-vapor interface; a localized electric field pulse on a grid of wires straddling the surface;and an acoustic wave traveling nearly parallel to the surface that perturbs the electron-surface system and allows the electron to escape.

Radioactive Backgrounds

The purity of cryogenic liquids that allow a long electron lifetime is equally important for the reduction of natural radioactive backgrounds. Penetrating photons from the walls of the enclosure are usually addressed by shielding with purified water, but radiation from impurities in the detector material itself is particularly problematic. A powerful method to reject false events due to penetration of external photons is to recognize the “Compton clusters” of scattered electrons they produce in the detector medium. This relies on the good spatial resolution we aim for in our detector, but is not provided in most other ideas for extending neutrino measurements down to keV energies. We have been studying the experience of the low background counting community, and will profit from the common knowledge to design our detector with materials that have low radiogenic emissions. At energies below about 10keV, the detector becomes “self-shielding”, in that the absorption length is small compared to the detector dimensions, and the only backgrounds that matter are due to radioactivity in the liquid itself.

The Readout Plane

We have been studying the readout of the charge when it arrives at the readout plane. The performance of this detector component is likely to have the greatest influence on both cost and performance of the system. To this end, we have been doing R&D at BNL, Columbia and Novosibirsk with a liquid helium/neon prototype at the 0.1m scale.

Our first aim was to find a method of amplifying the charge on the track, in order to reach the lowest possible electron energies, and we chose the technique of Gas Electron Multipliers pioneered by Sauli as the most promising avenue. We have found that a mixture of neon vapor with a fraction of a per cent of hydrogen gives large gain with the standard GEM planes available from CERN. We used three GEM planes in series, the most commonly used configuration. It is far from evident that these devices will perform at temperatures of 2-30K in the same way that they do at room temperature, but in fact we observe gain in helium below 3K and much larger gain where a molecular gas can be added, as in neon vapor.

Inspired by these results, we are focusing our current plans on this readout option, though we are actively considering a number of other options. Many measurements in our small cryostat are needed to validate this option. For example, the neutrino interactions take place in the liquid, which has a high density, but the GEM operates in the vapor phase, with a relatively low density.The ionization electrons pass through the liquid-vapor interface and enter the GEM, which is immediately above it. The process of transition has been studied in past years by low-temperature physicists, but we need to extend such measurements to the conditions appropriate to our detector.

Our results to date in dense vapors of heliumand neonindicate that the charge gain for stable performance is higher in neon with a small admixture of hydrogen than in pure helium or neon, and we accordingly choose this mixture as our tentative baseline.

The charge could be collected on a two dimensional array of amplifiers following the GEM, but though electronics with good performance below 2K are available, they do not have the elegant fabrication power of standard silicon circuits. We are focusing at this time on the readout of the light generated in the avalanche. The camera that reads out this light can be at higher temperature, above 50K, where silicon circuits, for example charge coupled devices, work at their maximum performance. Note that there is abundant visible light emission from the avalanche region, in contrast to the far ultra violet emission from the primary ionization. This is due to electron exchange reactions in the avalanche region.

An important feature of the GEM geometry is that the avalanches are confined to a very small volume, occupying just the central portion of a small hole in the plastic film forming the GEM, about 25 microns in transverse diameter. The light comes from a few percent of the surface of the GEM, and if a phase-space matching system can be devised to transport the light to a camera, the fraction of the light captured can be much larger than that naively expected. We have calculated the properties of a system using micro-lenses, also called lenslets. These are small lenses molded on the surface of a plastic sheet. In our calculations, they are used as field lenses, transporting light efficiently to the camera lens (see figure below). A calculation by Peter Takacs of BNL Instrumentation Division, including many relevant optical considerations, yields an optical capture efficiency of a few per cent, much larger than in a simple system. The lenslets are at the low temperature of the liquid, as is the camera lens, but the camera plane itself is at 50K, the optimum temperature for the CCD. The temperature drop is carried out in vacuum in the optics of the main lens, a system with which we gained experience in the large hydrogen-neon bubble chambers with fish-eye lenses at low temperature, and film at room temperature.

Schematic showing GEM plane (Grid 1) Schematic optical readout of the cubic meter engineering prototype,

and lenslet array (Object 2). showing CCD cameras viewing GEM+lenslet array above the interface.

Measurement Program

Our present 0.1m scale cryostat is very well instrumented with five widows, precise and flexible temperature control and wide pressure range and feedthroughs with high voltage ratings as well as numerous signal feedthroughs. It is accessible rapidly for changes in configuration. As mentioned above, we have a number of physical measurements to make in the near future. In a number of cases, the results are of general interests for technical and physics reasons.

This FWP is for FY 2006-8, but the collaboration is funded in FY 2004-5 at Nevis Labs, enabling the work to go forward during the whole period 2004-8.

We expect that our first campaign of measurements in the test chamber will be completed in the next twelve months. This should include:

1. Complete the measurements of the charge gain available from the GEMs in pure vapors of neon and helium, and in vapors with small admixtures of hydrogen.
2. Measurements of the light yield from the GEMs in pure vapors of neon and helium, and in vapors with small admixtures of hydrogen.
3. Measurements of the charge storage time at the interface in neon, where there have been no measurements to date, and in helium, over a much wider range of parameters than available in the literature.
4. Measurements of parameters that have not yet been published, such as the free ion yield for different ionization density.
5. Depending on results from ii., investigations of light yield from alternative devices in vapor, (e.g. MicroMEGAS).
6. First trials of gated charge release from the interface, using pulsed electric fields or infra-red laser photo-ionization.

Engineering Prototype

Our plans are based on the results we have obtained recently on GEM detectors at low temperatures. They have allowed us to develop a baseline concept that should us to obtain the event rate and background suppression needed to obtain a first measurement of the low energy solar neutrinos. It is based on a detector with one cubic meter of liquid neon, giving an event rate of about two hundred events per year. This needs dimensions of the order of one meter. We have started preliminary work on the design of this device, with the help of designers who have experience with large systems. Fortunately, the requisite engineering team is in place at BNL, led by Jack Sondericker and Margareta Rehak, who will soon finish their present obligations. We will take advantage of recent experience in the use of low background materials such as copper and plastics. The cryostat required for this prototype will be procured in industry, with delivery anticipated in 2007. One year will be required for assembling and commissioning the prototype. Some initial physics measurements could begin in 2008. The modest size of the prototype should permit road transportation, and an expedited installation in an underground laboratory.