Nuclear astrophysics brings together the latest developments in astronomy and theoretical and experimental nuclear physics in a quest to understand the origins and evolution of all the naturally occurring chemical elements in the universe, without which the world as we know it would not exist. Nuclear astrophysics requires an intimate knowledge of the inner workings of stars, particularly either those that die in energetic explosions such as supernovae or undergo cataclysmic thermonuclear blasts, such as novae and X-ray bursts. All the chemical elements except the very light hydrogen, helium, and lithium were created in nucleosynthesis processes in hot stellar environments such as stars, novae, and supernovae. The underlying processes that govern the evolution of these objects are the interactions between atoms, and the microscopic properties of individual nuclei.
The field of nuclear astrophysics aims to solve the mystery of the origins of the chemical elements and to understand the physics and evolution of cataclysmic variable stellar systems such as novae and X-ray bursts. Sophisticated models are used to predict and reproduce the observations seen with the latest generation of astronomical observational tools. Crucially, the nuclear physics input to the models is based on laboratory measurements, making these models as close to reality as current technology and techniques allow. Most of the key nuclear reactions that are important to the study of these environments involve short-lived radioactive nuclei.
The ISAC facility at TRIUMF is the ideal location to study these nuclei and their reactions because of its combination of beams of short-lived nuclei, variable-energy accelerators, and a suite of world-class experimental facilities. The nuclear beams, the accelerators, and experimental facilities have been optimized for studying reactions of astrophysics interest.
Results and Progress
Current research indicates that the observed abundances of the chemical elements, measured in carbonaceous chondrites (meteorites) and the solar atmosphere, have arisen via a series of nucleosynthesis processes occurring in the quiescent burning phases of stars and various explosive burning scenarios. The taxonomy of theses processes reflects the geographical landscape of the resulting isotopic abundances, with the A > 74 stable isotopes on the neutron-deficient side of stability being denoted the “p-nuclei” and are produced via a so-called “p-process”. The majority of nuclei in valley of stability are produced by the “s-process,” a series of slow neutron captures on stable seed nuclei with a nucleosynthesis path that involves stable nuclei mostly. The neutron-rich and the heaviest nuclei are produced by the “r-process,” a series of rapid neutron captures.
While some general details of these nucleosynthesis processes are known, the detailed picture remains shrouded in mystery. The s-process, what could be described as the least complex of the three processes, is fairly well described by the canonical distribution, an “astrophysics-free” model resulting from the exposure of an iron-rich initial composition to a parametric neutron-flux distribution.
The r-process in contrast, is one of the greatest mysteries in astrophysics. The abundance peaks seen in the isotopic distributions were quickly realized to have arisen due to the presence of closed neutron shells, and that the process must involve a series of rapid neutron captures way out into unexplored neutron-rich territory, competing at every turn with photodissociation reactions and beta decay. The global and specific nuclear properties of the nuclei involved in this path are needed in order to construct a realistic model of the r-process. Required properties of the nuclei include the ground-state masses (in order to calculate neutron-separation energies), beta decay half-lives and branching ratios, nuclear excited state properties, and radiative-capture reaction rates.
Thus the measurement of these properties in neutron-rich nuclei constitutes the major justification for the proposal to develop neutron-rich nuclear beams at TRIUMF. Another very important thrust has been focused on questions related to processes on the neutron-deficient side of stability due to the availability of accelerated radioactive ISOL beams at intensities unique in the world. The large range of experimental facilities at ISAC have exploited these beams to delve into the realms of explosive hydrogen and helium burning in sites such as supernovae, X-ray bursts, and classical novae.
Explosive Hydrogen Burning and Characteristic Gamma Emitters
A large effort exists to measure the reactions involved in explosive hydrogen burning, which is thought to occur in hydrogen-rich hot places such as accreting binary systems, both X-ray binaries and classical novae, and supernovae. In these systems, reaction rates of proton capture and other charged-particle reactions are required not only to construct viable physical models of the systems, but also to enable predictions and reproductions of observables such as luminosity curves, ejected isotopic abundances and, most relevantly, yields of characteristic γ-ray emitting radioisotopes observed by the latest generation of space-based telescopes. The γ-ray emitters, such as 18F, 22Na, 26Al, 60Fe, and 44Ti, have long been sought after as diagnostic tools of explosive stellar models. Gamma rays from 26Al and 60Fe have now been observed in the bulk of the interstellar medium in our galaxy, while γ-rays from 44Ti have been observed in an individual supernova remnant.
The most hoped-for observational signal in γ-ray astronomy, for classical novae, is the 1,275 keV γ-ray from the decay of 22Na (t1/2=2.6yr). Because model predictions indicate that it is made prolifically in the set of reactions of the thermonuclear runaway in a nova explosion, its decay can provide a detectable flux of γ-rays for such space-based observatories as the European Space Agency’s International Gamma-Ray Astrophysics Laboratory (INTEGRAL) satellite. Sodium-22 is synthesized within the NeNa cycle via a reaction sequence involving many stable and radioactive isotopes. Estimates predict a detectable 22Na signal from novae within a distance of around 1000 parsecs of Earth. Peak temperatures in the thermonuclear runaway reach 0.4 x 109 K in some O-Ne nova models, and hence resonant reactions dominate these reaction pathways in a centre-of-momentum energy regime of ~100 to ~1000 keV for proton capture reactions. Present theoretical estimates of the relevant nuclear structure have insufficient accuracy to enable reliable reaction rate calculations. Vital experimental information on the strengths and positions of resonances is required to enable credible model predictions of sufficient accuracy to satisfy the capabilities of the observational satellite-based instruments.
Until very recently, several reactions contributing to the formation and destruction of 22Na were unknown experimentally. Amongst them, 21Na(p,γ)22Mg was known to have a large influence on the synthesized 22Na yield. In addition, a brute-force, γ-ray spectroscopy measurement recently revealed a previously unknown resonance in the 22Na(p,γ)23Mg reaction, which threw that reaction rate into considerable uncertainty.
Measurement of the 21Na(p,γ)22Mg Reaction
At ISAC-I, a large campaign has been waged to determine experimentally the 21Na(p,γ)22Mg reaction rate. A direct measurement of all the known resonances in the reaction was made (fig 1) with the DRAGON facility using an accelerated 21Na beam of energy range 200A–1500A keV impinging on a windowless re-circulating hydrogen gas target. Each of these resonances was measured down to the 20% uncertainty (1σ) level required for the nova models [J.M. D’Auria et al., Phys. Rev. C 69 (2004)]. In parallel, the complimentary TUDA facility performed resonant elastic scattering studies with the 21Na beam impingent on a CH2 target, in the centre-of-momentum energy range 500–1500 keV, identifying states in the 22Mg compound nucleus, both known and previously unknown, and using R-Matrix fits to attempt spin-parity assignments, partial width, and resonance energy measurements (fig 2). This data, enabled an ordering of the 22Mg level scheme, which determined that all the contributing resonances to the 21Na reaction had been measured directly by DRAGON. Consequently, the 21Na(p,γ)22Mg reaction is considered the most well-measured reaction rate involving a rare-isotope nucleus and is often cited as a textbook example of how such measurements are made.
Fig 1 - Temperature-dependent reaction rates for the 21Na(p,γ)22Mg reaction, showing contributions from some of the resonances measured at DRAGON. Inset: Thick-target scan over the narrow, dominant 206 keV resonance.
The 21Na(p,γ)22Mg reaction rate determined at TRIUMF was stronger than the limited theoretical estimates, leading to a faster destruction of 22Na during the thermonuclear runaway and consequently a smaller ejected yield. Taken alone, this would somewhat increase the detectability distance of a classical nova, an important consideration when attempting to observe a nova with a γ-ray observatory. However, accurate estimates will not be possible until all the contributing reactions have been put on a firm experimental basis as has the 21Na(p,γ)22Mg reaction.
Fig 2 - TUDA excitation function from 21Na(p,p) and 21Na(p,p') scattering, fitted with a multichannel R-matrix.
One such reaction, and the next target for the TRIUMF astrophysics program concerning the production of 22Na, is the measurement of a newly discovered 22Na(p,γ)23Mg resonance. This reaction can best be studied by exploiting the massive intensities of 22Na produced by ISAC-I’s high power silicon carbide targets to implant 22Na targets and to perform a traditional prompt γ-ray measurement. The DRAGON group has successfully implanted several of these targets and designed a dedicated high-vacuum chamber, which has been constructed and installed at the Center for Experimental Nuclear Physics and Astrophysics (CENPA) at the University of Washington, Seattle. These TRIUMF targets will be used in conjunction with the intense and high quality tandem-produced proton beam to measure the resonance to high accuracy, further slashing the uncertainties of the nova 22Na production rate and giving reliable estimates of 22Na fluxes for astronomers. This measurement is now underway.
Measurement of the 26gAl(p,γ)27Si Reaction
Also of significant importance to explosive hydrogen burning scenarios is the synthesis of the radioisotope 26Al. This isotope, with its characteristic γ-ray at 1,809 keV, has long been a target for γ-ray astronomers. Because of its relatively long half-life [t1/2=(7.2±0.2)x105 yr], it provides insufficient flux given the ejected yields from stellar objects to enable detection from an individual source. However, the bulk of the 26Al produced in our galaxy from all sources does provide enough γ-ray flux to be measured (fig 3).
Fig 3 - COMPTEL All-Sky map of the Galactic distribution of 26Al, as determined by the flux of the characteristic 1809 keV gamma ray.
Since its first detection by the NASA High Energy Astronomy Observatory HEAO-3 satellite, its distribution in our galaxy has been extensively studied, most recently using the INTEGRAL satellite, which was able to show that the material was co-rotating with the visible matter in the galactic plane using Doppler-shift studies of the 1809 keV line. The observed distribution of 26Al is correlated with the 83 GHz microwave “free-free” map denoting the ionized interstellar medium, suggesting that its concentration in regions of high star formation points to massive stars as progenitors. Indeed, recent massive star models, including core-collapse supernovae and the Wolf-Rayet phases of more massive stars, can produce a total amount of 26Al that is commensurate with observations when incorporated into Galactic chemical evolution models. However, significant contributions from other sources such as Asymptotic Giant Branch (AGB) stars and classical novae cannot be ruled out, and past nova models have predicted that up to 20% of the total observed 26Al could come from novae. This percentage is, however, at odds with observation, and the nova models, to be creditable, must be based on experimental reaction rates that give the correct ejected yields of 26Al.
Of the reactions that affect 26Al production in the thermonuclear runaway of a classical nova, the 25Al(p,γ)26Si and 26gAl(p,γ)27Si reactions are particularly important (here the suffix g denotes the ground state of 26Al as opposed to the 6-second lifetime isomeric state at 226 keV, which is not significantly thermally populated at nova temperatures). The 25Al(p,γ)26Si reaction is an ISAC-I high-priority approved experiment, which required additional research and development for producing an intense 25Al beam. However, an intense 26gAl beam is possible because of the long half-life of the isotope and the substantial production factors in an ISAC-I high-power silicon carbide target. The uncertainty in the 26gAl(p,γ)27Si reaction lay in the dominant, isolated, narrow resonance at 184 keV, being previously assigned with a strength of 65 μeV from shell model estimates with significantly large uncertainty. An unpublished direct measurement of this resonance in normal kinematics yielded a value of 55 μeV. The uncertainty in the reaction rate led to a large uncertainty in the predicted 26Al ejected yield from nova, so it was considered vital that the unpublished measurement was confirmed using an independent measurement utilizing a different experimental technique.
Using the TRIUMF Resonant Laser Ion Source (TRILIS), peak intensities of 5 x 109 accelerated 26Al ions per second were achieved at DRAGON. The reaction yield was of the order 3 x 10-13 reactions per incident ion, and only the superior primary beam suppression capabilities of the DRAGON separator were enough to enable the measurement of the 184 keV resonance strength. Using the detection signature from DRAGON’s BGO detector array, it was possible to determine the location of the narrow resonance within the extended gas cell and, comparing to stopping powers also measured at DRAGON, to derive the resonance energy with high accuracy (fig 4). The resulting resonance strength of 35±7 μeV is lower than the adopted value, while the measured resonance energy of 184±1 keV is 4 keV smaller than the adopted value. These quantities were used in a spherically symmetric, implicit hydrodynamic nova code in Lagrangian formulation to estimate an ejected 26Al yield based on as much experimental information as possible while neglecting the still uncertain 25Al(p,γ)26Si rate. It was found that the newly measured 184 keV resonance strength resulted in a 20% increase in the ejected yield with respect to the adopted value.
Fig 4 - Detected Energy vs Time-of-flight plot showing gamma-coincident 27Si recoils from the extremely weak 26Al(p,γ)27Si reaction against a random background.
The conclusions of this work are that the paradigm of novae as a small but significant source of secondary 26Al in the galaxy remains supported, and further investigations of the 25Al(p,γ)26Si reaction and to some extent the 23Mg(p,γ)24Al reaction are required to put this conclusion on a firm experimental footing with nuclear uncertainties eliminated. These results were reported in Physical Review Letters.
44Ti Formation in the Alpha-Rich Freeze-out Phase of Core-Collapse Supernovae
One of the significant observations of an astrophysical characteristic γ-ray emitting radioisotope is the detection of 44Ti (t1/2=58.9±0.3 yr) in the Cassiopeia A (Cas A) supernova remnant at ~3.4 kiloparsec distance. Enough flux of the characteristic 68, 78 and 1157 keV γ-rays resulting from the decay of the daughter 44Sc are detectable from this remnant to enable a determination of the ejected 44Ti yield, immediately giving a diagnostic for supernova models. A 44Ti yield has also been inferred from the light curve of supernova 1987A that is similar to the one observed in the Cas A remnant. Meteoric grains of pre-solar origin also show 44Ca overabundances thought to come from in situ decay of supernova-produced 44Ti.
The all-sky survey of the 44Ti characteristic γ-ray, however, shows no unambiguous identification of 44Ti sources other than Cas A, and by inference, SN1987A. This survey is at odds with the expected flux given the presently adopted Galactic supernova rate and current 1D models of supernova explosions, which show 44Ti ejected yields a factor of 2-10 smaller than observed in Cas A. The models also fail to reproduce the 44Ti/56Ni ratio inferred from the SN1987A light curve or the solar system abundance ratio of 44Ca/56Fe. The predictions of these core-collapse models are highly dependent on the location of the boundary where material falls back on to the neutron star or black hole, or becomes ejected and available for observation. Here, a detailed understanding of the formation of 44Ti can lead to constraints on the underlying physics in the model.
The production and destruction reaction rates of 44Ti are a crucial part of this understanding, and it is known that the value of certain rates can affect the ejected 44Ti mass fraction by a large amount. Of these rates, the direct production of 44Ti via alpha-capture on 40Ca has a significant influence on the final 44Ti yield and was imbued with a large uncertainty. Two experimental attempts to measure the 40Ca(α,γ)44Ti reaction, one originally motivated from nuclear structure studies using the traditional prompt-γ technique at discrete energies and the other using the technique of TRILIS to derive a total cross-section integrated over the relevant energy range to supernova temperatures were undertaken. A large discrepancy existed between the rates derived in these two studies, with the accelerator mass spectrometry measurement resulting in a total strength up to ~5 times larger than that of the prompt γ-ray measurement.
DRAGON is the ideal instrument to measure this reaction because the isotopically pure gas target, superior beam suppression, and coincidence measurement of de-excitation γ-rays and 44Ti recoils enabling a clean, energy-dependent measurement that avoids possible sources of systematic error inherent in the other techniques. The DRAGON study includes experimentally measured efficiencies and stopping powers. A 40Ca++ beam was produced using the off-line ion source and accelerated to energies between 0.605A MeV and 1.153A MeV at intensities of around 1 x 1010 ions/sec. With the beam impingent on the windowless helium target with a thickness of 1-4 x 1018 atoms/cm2, a complete excitation function of the yield of 44Ti from the 40Ca(α,γ)44Ti was obtained in the energy region Ecm=2.11-4.19 MeV.
This experiment utilized the novel technique of using thin silicon nitride foils to boost the charge state of the recoiling product nuclei after the gas target to allow maximum efficiency of transmission of the selected M/q in the separator. A 48Ti beam was also used to measure the equilibrium charge-state distributions after the gas cell and the charge state booster foil. The resulting energy-dependent cross section showed marked differences from the prompt γ-ray data; in particular, substantial resonance strength was observed in regions between narrow resonances identified in the prompt γ-ray experiment. Although the strength of the strong resonances at higher energies in the reaction was in good agreement with the prompt γ-ray data, the DRAGON result was higher by a factor two in the low-energy regime (fig 5).
Fig 5 - Excitation function of the 40Ca(α,γ)44Ti reaction in the astrophysical energy range as measured by DRAGON, shown against previous experimental data.
The reaction rate used in the supernova 44Ti yield estimations was based on the Hauser-Feshbach statistical model of the NON-SMOKER code scaled to fit available resonance data on self-conjugate nuclei. Compared to this result, which appeared to agree with the prompt γ-ray data, the DRAGON result is higher by around 40%. The results suggest that the level density in 44Ti is higher than previously thought and, in fact, the non-scaled NON-SMOKER statistical model rates agree well with the DRAGON measured rates. The predicted mass fraction of 44Ti ejected in a Cas A type event is now higher by 40% compared to the previous estimates of the scaled Hauser-Feshbach approach. The measurement uncertainties in the DRAGON work translate to an uncertainty of ±3% in ejected 44Ti yield, much lower than the uncertainty in the observed yield and a substantial improvement compared to the discrepancy seen between previous experimental works. Thus, the use of 44Ti as a supernova diagnostic is put on a firmer experimental footing where the remaining uncertainties are dominated by the model physics. The remaining nuclear physics uncertainties relevant to 44Ti yield in supernovae are the strengths of the 44Ti(α,p) and 45V(p,γ) reactions.
Stellar Evolution and Helium Burning in Massive Stars
In the quest to uncover the secrets of nucleosynthesis, reliable stellar models are required, especially for massive stars that become the progenitors of core-collapse supernovae and are responsible for a large deposition of nucleosynthetic yield into the interstellar medium. The evolution of such a supernova progenitor is sensitively dependant on the conditions in the star during core and shell helium burning. In particular, the triple-alpha reaction and the 12C(α,γ)16O reaction take place simultaneously, depending on stellar mass, and their ratio determines the subsequent C/O ratio in the ashes of nuclear burning. The composition of these ashes then determines decisive global properties of the evolving star such as the convective energy transport conditions and entropy in the core, the subsequent burning shell positions and, therefore, whether the star initiates C- and Ne-burning stages or skips straight to O burning. All these conditions have a drastic effect on the nucleosynthetic yields in the star.
The reliability of stellar evolution models hinges on the value of the 12C(α,γ)16O reaction rate at helium burning temperatures (~1-2 x 108 K). The Holy Grail of stellar astrophysics is therefore the value of the 12C(α,γ)16O cross section at the mean interaction energy of 300 keV. Because the cross section there is of the order 10-17 barns, direct measurements of sufficient accuracy are simply impossible. A rate with an uncertainty of less than 10% is required to make the stellar models reliable to the required level of accuracy.
All attempts to determine the astrophysical S-factor at 300 keV, S(300), have done so by extrapolating measurements taken at higher energies while taking into account known resonances, and most importantly including known information for the vital subthreshold resonances corresponding to the 16O 1- and 2+ states at 7.17 MeV and 6.917 MeV respectively, for the ground-state transitions. This has been achieved using a variety of reactions and experimental techniques but, where radiative-capture measurements have been performed, it has been done using both normal and inverse kinematics.
What has been ignored in the past is the radiative capture with a cascade transition through the first excited 0+ state in 16O at 6.049 MeV. This capture has been ignored because that state decays purely through e+e- production and therefore no high-energy, secondary γ ray exists following the low-energy primary γ-ray to be observed in an experimental setup. Because of the experimental difficulty in observing this transition, it has been wrongly assumed to be small.
The DRAGON facility detects reaction products from radiative-capture reactions in inverse kinematics, in coincidence with de-excitation γ-rays around the target position. Therefore, it provides a layer of background rejection and sensitivity beyond the non-inverse approach. An experiment was conducted in which a 12C beam of up to 3 x 1011 ions/sec with energies in the range Ecm=2.22-5.42 MeV was impingent on the windowless helium target at pressures of 4 to 8 Torr. The resulting BGO energy spectra for events in coincidence with detected 16O recoils were analyzed in comparison with GEANT3 simulations including the pair-decay of the 16O first excited state, showing excellent agreement for the known branching ratios in cascades from the resonance or direct capture reactions (fig 6). In this way, an excitation function was constructed, in particular for capture and decay through the 6.049 MeV state, which has been observed here for the first time.
Fig 6 - DRAGON gamma-ray spectrum for the 12C(α,γ)16O reaction, shown with GEANT simulation results which include pair-decay from the 1st excited state of 16O.
Using R-Matrix fits, including both the E1 and E2 contributions to this cascade, an extrapolation was made (fig 7) that takes into account the interference between higher lying resonances, direct capture, and the subthreshold resonances. A value of the S-factor of S6.0(300) = 25+16-15 keVb was determined from this data (C. Matei et al., Physical Review Letters 97 (2006)). Given the value of Stotal(300) = 170 keVb for static helium burning argued from stellar nucleosynthesis models, and the fact that this S-factor is required to less than 10%, the cascade through the first excited state of 16O can no longer be ignored and makes up a significant proportion (15%) of the total reaction strength. This conclusion is only made possible due to the unique capabilities of the DRAGON facility.
Fig 7 - S-factor for cascades through the 6.0 MeV state in 16O in the 12C(α,γ)16O reaction, fitted with R-Matrix and including both the E1 (short-dashed), E2 (dotted) and total (long-dashed) components.
The results of this work are an important step in the quest to know the 12C(α,γ)16O reaction strength at helium burning temperatures to high accuracy, and the reliability of stellar nucleosynthesis models depends on them.
The Nuclear Trigger for X-ray Bursts
Type I X-ray bursts are fascinating objects. They are thought to occur when H- or He-rich matter accretes onto the surface of a neutron star in a binary star system and erupts into a thermonuclear runaway. This runaway causes a massive increase in radiated power and a corresponding sequence of proton- and alpha-induced reactions up the proton-rich side of stability, known as the rp- and αp-processes. As the fuel in the runaway is used up, the burst dies down, only to have the fuel build up again before a further runaway, often leading to a regular bursting pattern. Models of X-ray bursts have been studied prolifically over the last decade, resulting in the inclusion of a large network of nuclear reaction rates using input parameters, most of which remain unmeasured and are based solely on shell model, analogue state considerations, or statistical model estimates.
Several mysteries arise in reconciling models with observed X-ray bursts. For example, bursts are observed only in systems where the accretion rate is lower than around 30% of the Eddington accretion rate, whereas some models predict that all accretion rates up to the Eddington luminosity should result in X-ray bursts. The key to solving this mystery may lie in the breakout from the hot, β-limited CNO cycle into the rp-process. This thermonuclear runaway is by a sequence of nuclear reactions that occur at a critical temperature and density. The reaction sequence 14O(α,p)17F(p,γ)18Ne(α,p)21Na is the main breakout route, carrying the most reaction flux once the runaway has been established. However, the 15O(α,γ)19Ne reaction appears to be the real trigger for the X-ray burst, causing the pre-burst temperature instability and initiating flow into the rp-process that allows the runaway to start, raising the temperature to levels where the main breakout path can occur. The 15O(α,γ)19Ne reaction has, for around two decades, played this crucial role in hot-CNO breakout, depending on the strength of a dominant resonance at ~500 keV, corresponding to the 3/2+, 4.0 MeV state in 19Ne. The strength of this resonance is known to be weak, making a direct measurement of the radiative-capture rate extremely difficult and requiring extraordinary 15O beam intensities (> 1 x 1010 ions/sec). If such a beam were available, DRAGON, which was optimized to study this reaction, would be used in a direct measurement. So far, however, an 15O beam of sufficient intensity has been difficult to develop anywhere. Therefore the focus has been on deriving the 4.0 MeV state’s resonance strength via indirect measurements to determine the alpha-decay branching ratio of the state and its lifetime so that the resonance strength can be derived via
where Bα is the alpha-branching ratio and τ is the mean lifetime of the state.
At ISAC-I, two measurements of the lifetime of the 4.0 MeV state have been carried out using the Doppler-shift attenuation method. Using an implanted 3He target and the reaction 3He(20Ne,4He)19Ne* (here, the asterisk denotes an excited nuclear state) at 34 MeV to populate the 4.0 MeV and other states, the recoiling 19Ne nucleus was decelerated in the dense Au target, leading to a angle-dependent Doppler shift of the γ-ray energy. The γ-rays were detected using a high-purity germanium detector at 0° with respect to the beam axis. Gamma rays corresponding to transitions from seven states in 19Ne lying at excitation energies from 1.536-4.602 MeV were observed, including the 4.0 MeV state. Using a line shape analysis of the γ-ray from the 4.0 MeV state to the ground state, and the γ-ray from the 4.0 MeV → 1.5 MeV state (fig 8), a joint-likelihood analysis yielded a lifetime of 6.9±1.5(stat.)±0.7(sys.) fs. With this result, and the lifetime measurements of the other six states, all astrophysically relevant states (with the exception of the 4.378 MeV state) have been measured in this work to sufficient precision to constrain the 15O(α,γ)19Ne reaction rate when combined with a precision measurement of the α-branching ratios. The TRIUMF experiment represents the most precise measurement of the lifetimes of these states in 19Ne ever, and was successful largely because of the highly efficient experimental setup, high-quality beam at the ISAC facility.
Fig 8 - Doppler-shifted line shapes due to two transitions of the 4035 keV level in 19Ne, populated in the TRIUMF 3He(20Ne,4He)19Ne experiment, from which the state lifetime was extracted.
To determine the 15O(α,γ)19Ne reaction rate at X-ray burst temperatures, it is now required to measure the alpha-decay branching ratio of the 4.0 MeV state experimentally. Recent attempts to do this have only succeeded in putting upper limits on the value, but until a direct measurement of this resonance strength can be made at some point in the future, the alpha-decay measurements remain the only viable way to determine this rate experimentally.
Additional Studies and Future Directions
In addition to the work summarized above, the ISAC-I program has also contributed both experimental and theoretical work in the region of solar nucleosynthesis and the 7Be(p,γ)8B reaction.
In 2007, the focus was on producing fluorine beams to study the 18F(p,α)15O reaction, the most significant reaction pertaining to the production of 18F in explosive hydrogen burning, thought to produce the earliest potentially observable 511 keV γ-rays in a nova explosion. An initial study using the TUDA facility at somewhat reduced beam intensities measured excitation functions for both the 18F(p,p)18F and 18F(p,α)15O reaction. Initial R-Matrix studies of this high-quality data have shown several features at odds with previous experiments and some recent theoretical estimates. This measurement may have significant implications for the production of 18F and was only possible because of the excellent performance of the FEBIAD ion source installed at ISAC-I. Further optimization of the source will allow even higher beam intensities, and more studies in the low-energy astrophysical regime concerning 18F are planned in 2008.
In 2008, a new nuclear astrophysics detector, TACTIC, will be commissioned. TACTIC is an ion tracking chamber that will be used, like TUDA, to study the nuclear reactions of astrophysical significance with charged particles in their exit channels. The chamber operates at very low beam energies, has a very large acceptance, and is tolerant of high beam fluxes. These characteristics make it an ideal detector for measuring the small cross sections associated with nuclear astrophysics. The initial experiment will be to measure the cross section of 8Li(α,n)11B, which is a seed reaction starting off the r-process.
The DRAGON facility aims to tackle two important reactions in 2008–2009 involved in explosive hydrogen burning: the 33S(p,γ)34Cl reaction, crucial in determining the yields of 33S potentially deposited in meteoric grains of nova origin; and 23Mg(p,γ)24Al, a never-before studied reaction important for 26Al production in novae and the flow of the rp-process. The latter reaction is made possible due to newly achieved intensities of 23Mg using ISAC targets and the TRIUMF Resonant Laser Ion Source (TRILIS).
Beyond these experiments, new and unique beams are planned for the areas of astrophysics described above. These challenging beams will allow experiments to be done that will advance the understanding of astrophysics of neutron-deficient nuclei. World leadership in the field of nuclear astrophysics will remain a major thrust of TRIUMF-ISAC’s physics program. The DRAGON facility will remain the best instrument in the world for these important experiments.
All of the experiment studies listed above were led by TRIUMF scientists, in close partnership with a host of international collaborators. The experiments also attracted the participation of students, who benefited from the high-quality training available at TRIUMF. The scientists of the DRAGON and TUDA facilities continue to work closely with world experts in modeling classical novae, supernovae, stellar evolution, and X-ray bursts to identify important nuclear reactions and advance results in this field.
In Canada: Deep River, McMaster University, Queen’s University, Simon Fraser University, University of Alberta, University of British Columbia, University of Guelph, l’Université de Montréal, University of Northern BC, University of Prince Edward Island, University of Toronto, and the University of Victoria.
International Partners: Austria (1), Belgium (1), China (1), France (2), Germany (5), India (1), Ireland (1), Israel (1), Italy (2), Japan (1), the Netherlands (1), Scotland (1), Spain (2), the United Kingdom (3), and the United States (7).
Nuclear astrophysics constitutes a significant part of TRIUMF’s core scientific outlook, and the laboratory continues to provide the necessary infrastructure support for this program. The facilities involved in performing astrophysics research themselves rely on dedicated annual budgets for maintenance repair and operation. TRIUMF is also responsible for ensuring targeted and groundbreaking beam development to ensure that the experiments within this program can be performed. Nuclear astrophysics experiments often require the highest intensity and most challenging exotic rare-isotope ion beams. A key part of this beam development strategy is the investment, both in financial and personnel terms, in ion-source technology and target chemistry.
TRIUMF also provides a dedicated core of staff scientists involved in, and some who specialize in, nuclear astrophysics research. Personnel includes these grant-eligible board-appointed employees: G. Ball, L. Buchmann, B. Davids, P. Delheij, J. Dilling, G. Hackman, D. Hutcheon, A. Olin, C. Ruiz, and P. Walden