IMPROVING THE RATE OF THE TRIPLE ALPHA PROCESS

1
IMPROVING THE RATE OF THE TRIPLE ALPHA PROCESS

• Clarisse Tur a, Alan Wuosmaa b , Sam Austin a ,
  J. Lighthall b, S. Marley b , N. Goodman b , J.
  J. Bosb
• International Symposium on Nuclear Astrophysics
  Nuclei In the Cosmos-IX
• June 25-30, 2006
• CERN, Geneva, Switzerland
• National Superconducting Cyclotron Laboratory
  (NSCL) / Michigan State University (MSU) a and
  Western Michigan University (WMU) b

2
What is the triple alpha experiment about?

• Goal measure more precisely than has been done
  in the past the internal pair emission branch of
  the 7.654 MeV state of .
• This will imply more precision on the knowledge
  of the rate of the fundamental triple-
  reaction that takes place in stars to form the
  element carbon. Current error on rate /- 10-12
  goal /- 5.5 to 6.
• Do this by observing the ee- decay pairs of the
  0 7.65 MeV state of
• in coincidence with inelastically
  scattered protons at from
• 
  
• 
  
• Data will be taken using tandem accelerator at
  WMU in the summer of 2006. The incoming proton
  energy will be about 10.6 MeV.

Clarisse Tur NIC9 June 29, 2006
3
More on the triple alpha reaction

• During the helium-burning stage in stars, three
  nuclei are fused into in a
  resonant two-stage process called the
  triple-alpha process
• STEP I
• STEP II
• Then

4
A closer look at the triple alpha rate

• The rate for the triple- reaction can be
  approximately written as
• 
  
• given by
• ( is the energy released in the
  decay )
• is experimentally
  determined as the product of 3 independently
  measured quantities
• ( is the radiative width of the
  7.65 MeV state of carbon 12, is the total
  width taking into account the and
  decays of the Hoyle state )

Clarisse Tur NIC9 June 29, 2006
5
Current knowledge of
ERROR ON 3-ALPHA RATE 11.6
gt
NEW H. Crannell et al. - Based on analysis of
existing electron scattering data Nucl. Phy.
A758 (2005) 399c

GOOD
ERROR ON 3-ALPHA RATE 10.0
gt
And (6 measurements of Er) -
GOOD Contribution to fractional error on 3-alpha
rate 1.2

6
Back to the triple alpha experiment

• 10.6 MeV protons impinging on
• carbon 12 target inelastic scattering
• excites nuclei to 7.654 MeV state.
• take advantage of strong resonance
• at incoming energy of 10.6 MeV and
• 135 degrees in lab scattering angle.
• to reduce gamma ray background,
• require coincidence between scintillator tube and
  outer scintillator block.
• pair branch given by

Improved version of Robertson et al, PRC 15,
1072 (77)
Clarisse Tur NIC9 June 29, 2006
7
Accuracy in which we want to achieve

• This experiment versus Robertson et al. (1977)
• we wont be statistics-limited (can run
• for a longer amount of time 2 proton
• detectors instead of 1)
• less pile-up, fewer chance coincidences (dc
• beam instead of a pulsed one, faster scintillator
  (BC404 versus NE102), faster electronics, greater
  granularity (outer block divided into 4
  quadrants), can run at lower intensity for a
  longer amount of time)
• less gamma background (thinner inner
• scintillator tube gt 7 (Robertson et al.) versus
  4.7 (us) according to simulations)
• less target contaminants (better target purity)
• The accuracy that we can get will essentially
• only be limited by systematics (how well we can
  determine the gamma background for instance).

Measure to /- 5 (down from
/- 9.2)
gt
ERROR ON 3-ALPHA RATE 5.5 to 6
Clarisse Tur NIC9 June 29, 2006
8
Why is a better accuracy needed? SNII
Nucleosynthesis A16-40
Explosion of 15, 20 and 25 M? star average. Vary
rate of 12C(a,g)16O All else same.
Production Factor Same PF for 1.2 x
standard 12C(a,g)16O rate.
9
Why is a better accuracy needed? Size of the iron
core of an SNII
CORE-COLLAPSE SUPERNOVAE (SNII) Even if
was determined with 10
accuracy, there would be an intrinsic uncertainty
of on the core mass at onset of
core-collapse (it takes an energy similar to the
energy released in supernova explosion to
dissociate that into nucleons!) gt need better
knowledge of both rates (right now, 10-12
accuracy on 3-alpha, 20 on the
). Need a better knowledge of BOTH reaction
rates. Knowing the 3-alpha rate to 6 is a first
step in that direction.
Clarisse Tur NIC9 June 29, 2006
10
Why is a better accuracy needed? Asymptotic giant
branch stars (AGB stars)
Clarisse Tur NIC9 June 29, 2006
11
AGB stars Sensitivity to Nuclear Reaction Rates
(F. Herwig, S. M. Austin)
ASYMPTOTIC GIANT BRANCH (AGB) STARS Stronger
pulses (or He flashes) are induced and hence the
carbon enrichment of the stellar surface is
increased by about a factor of two when either
the 3-alpha rate is increased by its uncertainty
or when the rate is
decreased by its uncertainty. The latter has
recently been determined to 5 (Lemut et al.,
arXivnucl-ex/0602012) gt need better knowledge
of 3-alpha rate.
Herwig et al. Astro. Journal, 613 L73-L76 (2004)
, and Herwig et al. Phys. Rev. C 73, 025802
(2006)
12
Simulations Detector geometry
An acrylic tube BC404 tube 4 BC404 quadrants
13
Simulations Stopping range of parent e and e-
11.64 of all e e- exited detector (mainly
through central hole. Those that stop in the
detector, do so close to the center.
Clarisse Tur NIC9 June 29, 2006
14
Simulations Comparing cascade gamma pairs to
ee- pairs from the Hoyle state
cut (de gt 0.) (eOut gt 0.) 3.4 of the 1M
gamma pairs thrown make it through these minimum
requirements.
Clarisse Tur NIC9 June 29, 2006
15
Summary and conclusion

• Want accuracy of 5 for the pair branch gt
  accuracy of about 6 for the rate of the 3-alpha
  process.
• Detector has been built at the NSCL and has
  already undergone extensive tests.
• The simulations give us confidence in the design
  of our detector and our measurement.
• Running in the summer of 2006, using WMU Tandem
  accelerator.
• Stay tuned!

Clarisse Tur NIC9 June 29, 2006