The Wondrous New World of Modern Particle AstrophysicsAksel Hallin, University of Alberta, Edmonton AB, Canada

Doug Hallman, Laurentian University, Sudbury ON, Canada

T o investigate the frontiers of particle physics, physicists and engineers are building detec-tors and making measurements in unusual

settings from outer space to far-flung regions of the Earth. In the past several decades, laboratories have been set up deep underground in working mines or mountain tunnels to look at subatomic particles from our Sun and to search for the strange dark matter particles needed to explain the evolution of galaxies. This paper describes important current developments in astroparticle physics, the SNOLAB underground laboratory in Sudbury, Canada, and several of the ex-periments that are being developed in that facility.

The past decades have seen a remarkable expan-sion in our understanding of the overall structure of the universe—experiments have provided power-ful evidence that subatomic particles not present in normal matter have a major role in the structure and evolution of the universe. Astroparticle physics con-centrates on understanding these particles and their astrophysical effects. In this paper, we concentrate on neutrinos and dark matter.

Neutrinos, which are second only to photons in numerical abundance in the universe, are low mass, stable, uncharged, spin-1/2 particles. Through ac-celerator-based experiments, we know that there are three types (or flavors): electron neutrinos (ne), muon neutrinos (nm), and tau neutrinos (nt), associated with electrons, muons, and tau particles, respectively. On Earth, electron-neutrinos are most often created by radioactive decay. They only interact with normal matter via the weak nuclear force and therefore are

very penetrating —able to pass through many light-years of material with only a tiny chance of absorp-tion. The nuclear reactions that fuel the Sun generate copious quantities of electron-neutrinos; almost 3% of the Sun’s energy is emitted in the form of neutrinos. On Earth, this works out to 35 W/m2 or 60 billion neutrinos/cm2.s. Solar neutrinos travel directly from the core of the Sun to the Earth in eight minutes. Heat energy takes about 40,000 years to propagate from the core to the surface of the Sun, where it radiates as vis-ible energy. Thus, a comparison of the measured heat flux and the measured neutrino flux compares histori-cal energy production in the Sun to the current energy production.

Although neutrinos account for only a small frac-tion of the energy released from a star like our Sun, a supernova explosion, which marks the collapse and death of a star, emits almost all of its energy in the form of neutrinos. Since the neutrinos are generated early in the collapse and penetrate the outer layers of the star, neutrinos can be detected up to a day before light from a supernova. It is thought that all the chem-ical elements in our universe that are heavier than iron were created in such supernovae explosions. We are still unable to determine the details of such explosions and it is hoped that measurements of their neutrinos may provide important clues to understanding super-novae. Although neutrinos are stable and copious in the universe, they possess several properties that are not understood. On the basis of recent experiments, we know that neutrinos do have non-zero masses, but we do not know the individual masses of each type

274 DOI: 10.1119/1.3116835 The Physics Teacher ◆ Vol. 47, May 2009

or species of neutrino. We also do not know whether neutrinos are identical to or distinct from their an-tiparticles. There are indications that the particle properties of neutrinos may explain, through a process called leptogenesis, the fact that the universe is domi-nated by matter and not antimatter.

A large body of evidence indicates that the universe is filled with dark matter—invisible mass that is not part of stars, interstellar dust, or molecules—and that there is significantly more dark matter than normal matter. While the gravitational effects of this dark matter have been measured in several experiments, none of these measurements are able to distinguish between different constituent particles. There are no known particles that have the right properties to com-prise the dark matter. Particles need to be stable and interact very weakly, but be massive enough to clump together into galaxies. A leading candidate is a class of particles called WIMPs (Weakly Interacting Mas-sive Particles), which interact as weakly as neutrinos but have a mass about 100 times that of the proton (the mass of an atom of tin). In our galaxy, we would expect these particles to be gravitationally bound and form a spherical halo. The kinetic energy of these WIMPs is limited by the galactic escape velocity of a few hundred km/s.

Astrophysical Neutrino ExperimentsUnderground experiments to investigate neutrinos

from the Sun began with the pioneering work of Ray Davis and his collaborator in the Homestake mine in Lead, SD, starting in the mid 1960s. By developing a technique to detect tiny amounts of argon produced by neutrino absorption in the chlorine atoms of a tank of carbon tetrachloride, Davis was able to measure the flux of neutrinos of a limited energy region from the Sun. This measurement proved the hypothesis that nuclear fusion reactions do indeed generate the energy in the Sun. Several other experiments followed, including GALLEX/GNO (in Italy) and SAGE (in Russia) based on neutrino absorption in gallium and the water-Cherenkov detectors in Japan (Kamiokande and Super-Kamiokande) and North America (the IMB experiment in a salt mine near Cleveland, OH, and the Sudbury Neutrino Observatory in Sudbury, Canada, described below).

The Physics Teacher ◆ Vol. 47, May 2009 275

A Brief History of the Sudbury Neutrino Observatory

The Sudbury Neutrino Observatory (SNO) was proposed by an international group of scientists in the mid 1980s to address the solar neutrino problem —a puzzling discrepancy between the number of neu-trinos from the Sun’s core measured by Davis and all the other early experiments, compared to the number predicted by highly developed models of solar burning by John Bahcall1 and others. Only about one-third to one-half the expected neutrino number was seen in these early experiments. With the agreement of Vale Inco Ltd. to provide a deep underground site and of Atomic Energy of Canada to loan the heavy water re-quired in the detector’s core, SNO was approved and funded by research agencies in Canada, the United States, and Great Britain.

SNO Data Taking and DiscoveriesThe SNO detector is located in a huge rock cavern

—22 meters in diameter and 30 meters high, two kilometers underground (see Fig. 1) at the Vale Inco Creighton Mine, west of Sudbury, Ontario. The rock above the detector filters out cosmic rays that would otherwise produce photons in the detector, masking the tiny signals from neutrinos. Neutrinos (symbol

Fig. 1. Diagram of the SNO Detector. (Art by Don Foley© 2006 National Geographic Society.)

n) are detected in 1000 metric tons (106 kg) of heavy water contained in a clear plastic spherical tank at the center of the detector. Three types of reactions for neutrinos are seen in the deuterium in the heavy water: the charged current (CC), neutral current (NC) and elastic scattering (ES) reactions, as shown in Fig. 2. In SNO, approximately 13 neutrino induced events were detected each day (6 CC, 6 NC and 1 ES). These three reactions end up producing energetic electrons directly or after a neutron is absorbed. Visible photons are emitted by these energetic electrons as they travel through water by the Cherenkov effect and detected by SNO’s 9600 ultrasensitive photo tubes surrounding the plastic tank.

The CC reaction allows us to measure the number and energy spectrum of electron-neutrinos, while the NC reaction allows us to measure the flux of all types of neutrinos. The elastic scattering reaction is less like-ly than the other two, but it has the nice property that the electron points back at the Sun. Neutrino interac-tions lead to flashes of light; in a typical event (shown in Fig. 3) about 50 phototubes detect single photons and flash.

The outer part of the detector is filled with ultra-pure normal water, which supports the central plastic tank and shields the core against tiny amounts of ra-dioactivity in the surrounding rock. Any dust from the mine that enters the detector water can interfere with

neutrino measurements, so SNO has been operated as a cleanroom laboratory from the time of its construc-tion. Ultrapure materials are also used throughout the detector to minimize interferences.

In close to seven years of data taking, beginning in 1999, SNO has seen thousands of neutrinos through all three reactions. Careful modeling of each neutrino event allows the energy, location, and direction of each event to be determined.

By comparing neutrino reaction rates, SNO an-nounced in May 2001 that it had strong evidence that neutrinos, which start out as electron neutrinos in the Sun’s core, oscillate from this original type to other types as they travel from the core of the Sun to the detector. Two-thirds of the neutrinos from the Sun have changed from their original type to one of the other two types. This flavor change can only happen if neutrinos have mass, and SNO’s result proves that neutrino type is not a conserved quantum number. Since these other neutrinos could not be identified by earlier experiments, SNO’s total neutrino numbers are much higher than those reported earlier, and now are in excellent agreement with the predictions of solar theories—the solar neutrino problem has been solved. SNO’s papers2,3 detailing these results were the most cited in the worldwide physics literature in 2002, with more than 1000 references each.

276 The Physics Teacher ◆ Vol. 47, May 2009

Fig. 2. Neutrino reactions in SNO.

Fig. 3. A neutrino elastic scattering signal in SNO. Each green dot represents a phototube that has received Cherenkov light sent out in a cone-shaped burst from the interaction site.

Neutrino Reactions on Deuterium

Charged-Current

np

np

ve

ve

vx

neutrino deuteronprotons

Cherenkov electron

Neutral-Current

pp

neutrino deuteron

neutrino

neutron

proton

p

n

neutrinoneutrino

Cherenkov electron

electron

Elastic Scattering

SNOLAB – A New International Facility for Underground Science

Building on the success of the SNO experiment and the expertise of the SNO researchers and staff, an expanded underground facility, SNOLAB—four times larger than the original SNO site—has just been completed in early 2009. A research center was also constructed in 2004-2005 near the mine entrance with offices and cleanroom laboratories to support new underground experiments.

SNOLAB’s facilities, shown in Fig. 4, include three new experimental halls—the Cube Hall, the Cryopit, and the Ladder Labs—to house a group of experi-ments of different sizes and requirements. Rooms for detector development, material purity assays, and personnel are also included. Because the host rock at 2-km depth has an interior temperature of 41oC, huge air conditioning units are required to cool air and detectors, and an extensive network of air handlers and filters provides cleanroom air. The new SNOLAB facilities are fully linked with the SNO laboratory through a new personnel and equipment and supplies entry area. Experiments will be separately operated by collaborative groups, with common support services provided by SNOLAB personnel. Visit the website http://www.snolab.ca for further information.

The SNOLAB Research ProgramPrototypes for two dark matter experiments—the

PICASSO bubble detector and the DEAP liquid ar-gon detector—are already in operation in the SNO laboratory. The SNO detector is also being refurbished and configured for the SNO+ experiment, which uses a liquid scintillator. A supernova neutrino detec-tor, HALO, is also being developed for installation at SNOLAB in the near future. Plans for the installation of other new experiments and of larger more sensitive detectors for several existing ones are in progress.

The SNO+ Experiment – A New Detec-tor from SNO

Following the completion of the measurement pro-gram for SNO in 2006, the SNO detector and cavity are being renovated and refurbished to establish a new experiment—SNO+. The SNO+ detector is designed to make several measurements:

• To search for neutrinoless double beta decay. Double beta decay is a very rare process, but occurs in nuclei such as 150Nd, which has a half-life of 7.9 x 1018 years, and in a number of other nuclei. The process is rare because it requires two neu-trons to simultaneously decay into two protons, two electrons and two anti-neutrinos. However, if neutrinos are identical to their own antiparticles, the neutrinos can annihilate. In this case, the decay is termed neutrinoless double beta decay and the two emitted electrons carry away the energy (the daughter nucleus receives an insignificant amount

The Physics Teacher ◆ Vol. 47, May 2009 277

Fig. 4. The SNOLAB laboratory layout. The facility – the world’s deepest large lab – is 2000 m (more than five Empire State Buildings) underground, where cosmic ray interferences are reduced by a factor of close to 100,000 from those at Earth’s surface.

Fig. 5. The SNO+ detector. A rope net will be installed over the detector to keep the sphere submerged within the light water shield.

of energy because it is very massive). The rate of neutrinoless double beta decay depends upon the precise mass of the neutrino.

• To measure low-energy neutrinos from the Sun, and in particular the so-called pep and CNO neutrinos. The neutrinos measured by SNO can be calculated to an accuracy of only about 15%. Lower energy pep neutrinos (created by the reaction p + p + e- → ne + d in the Sun’s core) can be calculated with 10 times this accuracy, and a comparison of the measured rate to the calculation will be a sensitive probe of nonstandard neutrino physics. CNO neutrinos, created by the Sun’s car-bon-nitrogen-oxygen fuel cycle, have never been observed, and may offer new insight into the work-ing of the Sun and energy generation in large stars (in which this is the dominant energy cycle).

• To measure geo-neutrinos from the Earth. Neutrinos are the only known particles that pen-etrate the Earth, and consequently people have speculated that with larger, improved neutrino detectors one might ultimately make a neutrino “x-ray” of the Earth’s interior. The early application of geo-neutrino measurements is to understand the distribution of uranium and thorium throughout the Earth’s interior, and to measure the amount of heat generated in the Earth by radioactive decay.

• To watch for neutrinos from a supernova with-in our galaxy. This would provide additional data to understand the mechanisms of stellar collapse.

• To measure antineutrinos from nuclear reac-tors, and to test the models and previous measure-ments of reactor neutrino oscillations (from one type to another).

Since the scintillator has a specific gravity of 0.86, there is a buoyancy force of about 1.2 x 106 N on the sphere, and a rope hold-down net is needed to keep the detector in place. A new purification facility for the scintillator also needs to be installed, along with a system to maintain the gas above the system free of radon. The hold-down system is shown in Fig. 5. For further information visit http://www.snoplus.phy.queensu.ca.

The DEAP/CLEAN ExperimentAs detectors pass through the WIMP halo that is

thought to fill our galaxy, the WIMP particles will oc-

casionally collide with nuclei in the detector. In detec-tors that contain a liquefied noble gas, a flash of a few hundred visible photons is generated. A prototype liq-uid argon gas detector (DEAP-1) was installed in the SNO laboratory in 2007, and has provided proof of concept and background measurements as well as de-tector design information for a larger detector. The re-cently formed DEAP/CLEAN collaboration is build-ing two experiments to search for dark matter—an initial 100-kg detector (called MiniCLEAN) that can be filled with either liquid argon or liquid neon, fol-lowed closely by a 1000-kg detector that will use liquid argon (DEAP-3600). Liquid argon is a very good dark matter detector for several reasons:

• Argon is a noble gas and therefore can be purified relatively easily to minimize background signals from impurities.

• Argon generates a large amount of light when ion-izing radiation passes through it.

• Argon is relatively inexpensive and has been previ-ously used in large detectors.

• There is a big difference in the time evolution of light from normal background radioactivity versus that for WIMP-induced activity, allowing the dis-crimination of backgrounds from candidate events.

The background constraints are severe, since the experiment is being designed to detect one recoil signal per year. The DEAP-3600 detector consists of 3600 kg of cryogenic liquid argon inside an acrylic sphere, sur-

278 The Physics Teacher ◆ Vol. 47, May 2009

rounded by approximately 260 photomultiplier tubes (PMTs) that run at room temperature (Fig. 6). The acrylic light guides and spacers serve as heat insulation, neutron shielding, and permit light transmission from the inner detector to the working PMTs. The entire detector is encapsulated in a water shield to further re-duce background signals from natural radioactivity in the surrounding rock. See http://www.deapclean.org for more information.

The PICASSO ExperimentThe PICASSO project (Project In CAnada to

Search for Supersymmetric Objects) is a dark matter search experiment. The sensitive material of PICASSO detectors is the superheated liquid C4F10, which is dispersed in the form of 50-µm to 100-µm diameter droplets in a viscous medium. If a dark matter particle hits a fluorine atom in a droplet, the recoiling atom deposits its kinetic energy on its track in the surround-ing liquid. A tiny proto-bubble forms and grows ex-plosively until the entire droplet is transformed into a vapor bubble. An acoustic pulse is produced and de-tected by sensors. Events can be localized by GPS-like triangulation analysis. PICASSO’s latest modules have 80 g of active mass of C4F10 and nine piezoelectric sensors. These detectors are shown in Fig. 7. In the fall of 2008 all of the 32 detectors of the ongoing Phase I detector were installed and are in operation. The PICASSO collaboration reported4 an improved upper limit for WIMP spin-independent interactions with nuclei. No evidence for a WIMP signal was seen for an exposure of 2-kg days—this result provides a new upper limit to WIMP interactions in a portion of the WIMP mass region. A Phase II detector with a 100-kg to 1000-kg active volume is planned in the future. See http://www.picassoexperiment.ca.

An Exciting Decade of Plans, Projects, and Progress

The facilities and experiments outlined above are representative of the large number of new astropar-ticle physics initiatives in underground laboratories around the world. In the United States, planning is well advanced for DUSEL—The national Deep Un-derground Science & Engineering Laboratory—at the Homestake Mine site in Lead, SD. Experiments planned at DUSEL are outlined at http://www.lbl.

gov/nsd/homestake/ and at http://www.sandfordun-dergroundlaboratoryathomestake.org. All of this activ-ity speaks well for opportunities for particle astrophys-ics graduate study, and for engineering and technical program graduates. Experiments in the next decade should reach sensitivities of neutrino mass, dark mat-

The Physics Teacher ◆ Vol. 47, May 2009 279

Fig. 6. A conceptual drawing of the central vessel of the DEAP-3600 experiment.

Fig. 7. The Phase 1 Picasso detector with its array of 32 sensor modules.

ter interaction cross sections, and rare decay rates that, current theories suggest, could lead to absolute mass assignments for neutrinos, actual detection of dark matter particles, and confirmation of the rare neutri-noless double beta decay process. Watch for updates on these exciting quests over the years ahead.

Educational Activities and Opportunities

There is a great interest on the part of students and members of the general public in the astrophysics of our universe and the particles that are a part of its composition and dynamics. At SNO and SNOLAB, we have developed and maintained an educational outreach program in partnership with Sudbury’s sci-ence center, Science North, from the early days of planning for the facility and experiments. The labora-tory location—in a working mine 2 km below the Earth’s surface—greatly limits the number of tours and visits to the site. Thus, we planned and developed displays and an Object Theatre at Science North (http://www.sciencenorth.ca), so that visitors could be informed of the science of the lab and experiments, see representative parts of the detectors, and monitor experiment progress. Lecture series at the center have also been held along with outreach presentations to groups and school classes. Websites and materials were developed to highlight construction progress and the experiment operation phases and their results.

Particle astrophysics educational resources are quite widely available (often on the web), but the level of difficulty is an issue. In Canada, a widely used grade 12 textbook5 features the Sudbury Neutrino Observa-tory and links this topic to an introduction to particles and detection fundamentals. We conclude this brief look at an exciting field with some references6-12 to representative articles that may be of interest.

References1. J. Bahcall, Neutrino Astrophysics (Cambridge University

Press, Cambridge, 1989). See also the excellent website http://www.sns.ias.edu/~jnb/.

2. Q. Ahmad et al., “Measurement of the rate of ne + d –> p + p + e- interactions produced by 8B solar neutrinos at

the Sudbury Neutrino Observatory,” Phys. Rev. Lett. 87, 071301 (2001).

3. Q. Ahmad et al., “Direct evidence for neutrino flavor transformation from neutral current interactions in the Sudbury Neutrino Observatory,” Phys. Rev. Lett. 89, 011301 (2002).

4. M. Barnabé-Heider et al. (Picasso Collab.), “Improved spin dependent limits from the PICASSO Dark Matter Search Experiment,” Phys. Lett. B624, 186-194 (2005).

5. Nelson Physics 12, student text, National Ed. (Nelson Education Ltd., Scarborough, ON, Canada, 2003), Unit 5, Chap. 13; ISBN-13: 9780176259884.

6. Arthur B. McDonald, Joshua R. Klein, and David Wark, “Solving the solar neutrino problem,” Sci. Am. 288 (4), 40–49 (April 2003). Also in “Frontiers of Phys-ics,” special edition of Sci. Am. 15 (3), 22–31 (2005).

7. John N. Bahcall and Edwin E. Salpeter, “Stellar energy generation and solar neutrinos,” Phys. Today 58 (10), 44–47 (Oct. 2005).

8. Francis Halzen and Spencer R. Klein, “Astronomy and astrophysics with neutrinos,” Phys. Today 61 (5), 29–35 (May 2008).

9. Garth Huber, “Contemporary nuclear physics: From the core of matter to the fuel of stars,” Phys. Canada 63 (3), 121–126 (2007).

10. Alain Bellerive, “Probing the quantum universe,” Phys. Canada 62 (4), 185–190 (2006).

11. John N. Bahcall, Frank Calaprice, Arthur B. McDon-ald, and Yoji Totsuka, “Solar neutrino experiments: The next generation” Phys. Today 49 (7), 30–36 (July 1996).

12. Bertram Schwarzschild, “Cosmic-ray showers provide strong evidence of neutrino flavor oscillation,” Phys. To-day 51 (8), 18–20 (Aug. 1998).

PACS codes: 10.00.00, 95.00.00

Aksel Hallin is a professor and Canada Research Chair in astroparticle physics at the University of Alberta in Edmonton, Canada. He is working on the SNO+ and DEAP/CLEAN experiments, as well as continuing the final data analysis of SNO.aksel.hallin@ualberta.ca Doug Hallman is a professor emeritus in the Department of Physics at Laurentian University, Sudbury, Canada. He is Director of Communications for SNO and researches ultrapure materials and analysis methods for the SNO+ experiment, as well as the establishing and maintaining of laboratory and detector cleanroom conditions.dhallman@laurentian.ca

280 The Physics Teacher ◆ Vol. 47, May 2009