Fusion cross-sections for deuterium cycle fusion reactors (D-cycle): an analysis of geometric, Gamow-

Sommerfeld and astrophysical S-factors

Abstract

Fusion reactions in the deuterium cycle (D+D, D+T and D+ 3He) are the main nucleus-nucleus
interactions which occur in tokamaks and stellerators. These reactions are the limiting case between
the Woods-Saxon potential field at nuclear distances and the Coulomb electrostatic potential
(scattering) at longer distances. In this paper several fusion cross-sections, geometric, Gamow-
Sommerfeld and astrophysical S-factors have been reviewed and compared with experimental data
from the last ENDF/B-VIII.0 cross-section library. The XDC-fusion code has been developed to
calculate fusion cross-sections, geometric, Gamow-Sommerfeld and S-factors of the deuterium-cycle
(D-cycle), including resonance parameters (energy and partial width). The software estimates also
fusion reaction heat (Q) and Woods-Saxon/Coulomb proximity potentials. Although relative
differences between fusion cross-sections are lower than 5 %, S-factors present considerable
differences between the energies and partial width (FWHM) of the single-level Breit-Wigner (SLBW)
resonances. The energy at which is placed the maximum fusion cross-section is also different between
cases. In conclusion, fusion reaction models for light nuclei (deuterium, tritium and helium) should be
reviewed in order to apply fusion to energy production in safety conditions.

Keywords: Fusion reaction, fusion cross-section, astrophysical S-factor, Gamow-Sommerfeld factor

1. Introduction

Fusion reactions in the deuterium cycle (D+D, D+T and D+ 3He) are the main nucleus-nucleus
interactions which occur in the deuterium cycle (D-cycle) in tokamaks and stellerators fusion reactors
[Smith and Cowley 2010, Donné 2018, Zohm 2019]. Fusion reactions are the limiting case between
the Woods-Saxon potential field at nuclear distances and the Coulomb electrostatic potential
(scattering) at longer distances. In order to overpass the Coulomb electrostatic barrier and achieve
nuclear distances to allow fusion reactions, ionized plasma is confined by a magnetic field inside
tokamaks and stellerators [Knoepfel 1978, Woods 2006]

Fusion reactions are characterized by their fusion cross-sections and described by three energy
dependent functions: geometric, Gamow-Sommerfeld and astrophysical S-factor. These factors retain
the physical characteristics of fusion reaction: a Coulomb electrostatic potential barrier penetration,
in a form of an exponential attenuation, G(E), and a strong force interaction after tunneling effect in
the nuclear potential well (Woods-Saxon), at the boundary conditions, in a form of an astrophysical
S(E) factor or resonance function [Miley et al 1974, Li 2002, Chen 2016].

Fusion cross section is expressed in a simplified form as

 f ( E )  GS

where ϕ is the geometric factor, a term inversely proportional with incident particle energy, S is a
strong-energy correlated parameter S-factor, called also astrophysical or S-factor, which depends on
the boundary conditions between nuclear well and Coulomb electrostatic field and G(E), a tunneling
or penetration factor, called Gamow-Sommerfeld factor. In particular, the astrophysical S-factor is a
resonance function, based on the classical formula of Breit-Wigner of single-level resonances [Breit
and Wigner 1936].

Fusion cross-section of light nuclei were firstly estimated by the Naval Research Laboratory, who
published a 5 parameter formula for D+D, D+T and D+ 3He light nuclei fusion reactions, also known as
Duane 5-parameter formula, based on accelerator beam-target experiments [NRL 2019].

Corresponding author: [email protected]

Further studies, such as those carried out by Bosch and Hale in 1992, included a 9 parameter formula
for Gamow and S-factor, based on the R-matrix of thousand experiments, and an inverse energy
dependent geometric factor. However, in the case of D(d,p)T fusion reaction, no resonance peak was
included [Bosch and Hale 1992].

In the last decades, fusion cross-sections have been modeled with a complex phase-shift of the
incident wavefunction at the boundaries of Coulomb electrostatic potential and Woods-Saxon
potential for the nuclear well, including SLBW resonance peaks for all target nuclei (D, T or 3He) [Li et
al 2006,2008, Singh et al 2019].

In order to estimate fusion cross-section, geometric, Gamow-Sommerfeld, S-factors and SLBW

resonance parameters in the D-cycle, the XDC-fusion code has been developed. Fusion cross-sections
equations estimated through different expression have been compared with ENDF/B-VIII.0 cross-
section libraries: Duane 5-parameter formula, Bosch-Hale 9-parameter formula and a combined
geometric factor based on wavenumber and a complex phase-shift S-factor.
2. The D-cycle

The deuterium cycle of fusion is composed of several interdependent nuclear reactions [Atzeni and
Meyer-ter-Vehn 2001, Ball 2019]:

2
1 D + 12 D  31T (1.01 MeV) + 1
1 H ( 3.02 MeV)
2 1
1 D + 12 D  23 He (0.82 MeV) + 0 n ( 2.45 MeV)
2 1
1 D + 31T  24 He (3.52 MeV) + 0 n (14.07 MeV)
2
1 D + 23 He  24 He (3.67 MeV) + 1
1 H (14.68 MeV)

which can be summarized in

6 12 D  4
2 He (3.52 MeV) + 11 H (3.02 MeV) + 01 n (14.07 MeV)
4
2 He (3.67 MeV) + 11 H (14.68 MeV) + 01 n (2.45 MeV) + 1.83 MeV

Furthermore, protons produced by fusion reactions can also react with low probability as

1
1 H + 11 H  2
D + e  +  + 0.93 MeV
1 (99.76 %)
1 1
1 H + H +e
1
  12 D +  + 1.95 MeV (0.24 %)

Other fusion reactions of the deuterium cycle with a low branching ratio are [Robouch et al 1993,
Bystritsky et al 1998]

2
1 D + 12 D  4
2 He (0.08 MeV) + γ0(23.77 MeV)

and with tritium as target nuclei

2
1 D + 31T  4
2 He (3.52 MeV) + 01 n (14.07 MeV)
2 3 5
1 D + 1T
 2 He* (0.05 MeV) + γ0(16.65 MeV)
2 3 5
1 D + 1T
 2 He* (3.2 MeV) + γ1(13.50 MeV)

There are other fusion reactions in which the reaction heat is positive, which have not been
considered, such as

2 1
1 D + 31T  3
2 He +20n
2 3 3 1
1 D + 2 He
 2 He + 11 H + 0 n

Table 1 summarizes the main characteristics of the D-cycle fusion: branching ratios (BR), reaction heat
(Q) and emergent particle/gamma energies.
3. Geometric, Gamow-Sommerfeld and S-factor

3.1. Geometric factor (ϕ)

In most of the fusion cross-section expressions, geometric factor is expressed as  1 / E . In the

last decades, geometric factor has been expressed as the area subtended by a de Broglie
wavelength  or wavenumber k, which depends on energy [Li et al 2006, 2008]


 ( E )  2 
k2

with

k  2 E / 

where μ is the reduced mass of the system,

 m1m2 /  m1  m2 

The geometric factor ϕ(E) can also be expressed with two independent terms,

 2  1
 ( E )   
 2  E

The first term is a conversion factor which is included in some expressions in the Gamow-Sommerfeld
or in the S-factor, with units of keV.b in the SI.

3.2. Gamow-Sommerfeld factor

3.2.1 Introduction

The Gamow-Sommerfeld factor G(E) is a penetration factor of the Coulomb barrier by the incident
particle and characterize the probability of incident/target nuclei of reaching a fusion reaction. It is
expressed with an exponential attenuation function as [Brennan and Coyne 1964, Humblet et al 1987]

1
G( E )  exp  2l 
2

with ηl the Sommerfeld parameter, evaluated as

rtp
1
l 
 k (r ) dr
rn
l

with kl the wave number inside the nuclear well

1/ 2
 2 
kl ( r )  2  E  U ( r )  
 

where U(r) is the potential field, rn is the classical distance of closest approach (nuclear square-well
radius) and rtp is the classical turning point, that is, the distance in which the energy of the incident
particle energy is E U ( rtp ) [Lee and Jung 2017]
3.2.2 Case 1: Duane parametrization

Duane parametrization included an interpolation of the Gamow factor as the Mott’s form of the
Coulomb barrier penetrability [Mott and Massey 1965, NRL 2019]

1
G( E ) 

exp A1 / E  1 
3.2.3 Case 2

Let us consider that the potential field U(r) can be approximated as

 2l  l  1
U (r ) U C (r ) 
2 r 2

where UC is the Coulomb potential and the second term is the spin-orbit potential, with l the
magnetic quantum number.

The Coulomb electrostatic potential is expressed as

Z1 Z 2 e 2
U C ( r ) ke r rn
r

with k e 1 / 4 0 the Coulomb electrostatic constant and rn is estimated as


rn r0 A11/ 3  A21/ 3 
with r0 the specific radius per nucleon of fusion reaction and the classical turning point rtp as

rtp ke Z1Z 2 e 2 / E

The centrifugal term is zero at the boundaries of the nuclear potential well ( l=0), then the integral is
calculated as [Atzeni 2004]

1 Bg   E   E  E   
0  arccos   
 U   U
 1   
2 E   0  0  U 0   

 

where
rtp
U 0 U (rn ) E
rn
and

1/ 2
Bg  g 2 k e Z1Z 2e 2  / 2 2

is the Gamow constant.

The η0 can be also expressed in term of the lengths rn and rtp as

   r   r  r   
1 Bg   n    n 1  n  
0   arccos
 2
E  rtp   rtp  rtp   
      

1/ 2
The arccosine can be expanded in powers of  E /U 0  , thus


1 B    E 1/ 2 1  E  3 / 2 3  E  5 / 2   E  E  

 0  2 g              ...     1   
 E  2   U 0  6  U 0  40  U 0    U 0  U 0   

  
In the low energy limit, with E  U 0 , η0 reduces to the Sommerfeld parameter

1 Bg
0 
2 E

and the Gamow factor to


G ( E ) exp  Bg / E 
The Sommerfeld parameter can be expressed also as  0 Z1Z 2  c / v  , with  e 2 /  4 0 c 
the fine structure constant, and v the relative velocity of the incident particle, related with the kinetic
energy as, E  v 2 / 2 .

3.2.4 Case 3

In some approximations, the Gamow factor is expressed in a similar way to the Sommerfeld
expression as [Yoon and Wong 2000]

2
G( E ) 
exp 20   1

which is equivalent to [Li et al 2006,2008]

2
G( E ) 
exp 2 / k ac   1

with ac the Coulomb unit of length

2
ac 
ke Z1Z 2 e 2

Other expressions have included relativistic considerations about Gamow-Sommefeld factors, which
will be considered in future research [Arbuzov and Kopylova 2012]

3.3. S-factor

3.3.1 Introduction

The S-factor is a function which characterizes the width and energies of the resonance peaks of a
nuclear reaction. The Breit-Wigner formula was firstly applied to the non-relativistic scattering of slow
neutrons with dependence only on resonance energy position and width [Breit and Wigner 1936]
In the case of a single level Breit-Wigner resonance E0, the S-factor is

i f
S (E)  2 2
E  E0     / 2 

where E0 is the energy of the single-level resonance, Γi and Γf are the partial widths of the entrance
and exit channels of the fusion reaction and   
j
j is the total decay width, equal to the width
of the resonance peak at half height [Takano and Ishiguro 1976, Blatt and Weisskopf 1979, Belov et al
2019].

3.3.2 Case 1

The S-factor was interpolated by Duane through a [1/2] order Padé approximation with a constant
term as [Duane 1972, Miley et al 1974, NRL 2019]

A2
S (E)  A
 A4  A3 E  2  1 5
where the single-level Breit-Wigner (SLBW) resonance is at E0  A4 / A3 and the total resonance
width is  2 / A3 (see table 2) In the case of D+D fusion reactions, in which A5=0, the S-factor
equals the classical Breit-Wigner resonance formula.

3.3.3 Case 2

Further approaches increased the order of the interpolation, through a [4/4] order Padé
approximation as [Bosch and Hale 1992]

A1  E  A2  E  A3  E  A4  EA5   
S (E) 
1  E  B1  E  B2  E  B3  EB4   
Table 3 presents the coefficients for the [4/4] order Padé approximation. In the case of fusion
reactions T(d,n)4He and 3He(d,p)4He, the S-factor was interpolated in two different energy regions:
[0.5-550 keV] and [550-4700 keV] for T(d,n) 4He and [0.3-900 keV] and [900-4800 keV] for
3He(d,p)4He.

3.3.4 Case 3

Let us consider that the Woods-Saxon potential in the nuclear well is modeled as a square-well
complex potential as

U S (r ) U r  iU i

where Ur and Ui are the real and imaginary part of the potential [Mohr 1957].
The complex phase shift ω(δ0) of the incident wavefunction is

 ( 0 ) r  ii W /  2 Wr  iWi  /  2 cot  0 /  2

In this case, the dimensionless S-factor is defined through the function ω(δ0) as [Zhong-Li et al 2004, Li
et al 2008, Jing et al 2009, Singh et al 2019]
  4i
S (E)  2
  i  1 /  2 
2
r

Let us define the complex function

 r  ii  K r  iK i  rn Krn

where rn is the radius of the nuclear well, estimated as


rn r0 A11/ 3  A21/ 3 
with r0 the specific radius per nucleon of the nuclear well, which depends on target/incident nuclei.

Therefore, the wavenumber K is

2
K 2  K r  iK i  a  ib

where

2
a E  Ur 
2
2
b  2  Ui 


Equaling real and imaginary parts,

K r2  K i2 a
2 K r K i b

Hence,

1/ 2
 a  a2  b2 
K r  

 2 

b
Ki 
2K r

The boundary conditions of the wavefunction Ψ in the nuclear well as

 ( rn ) rn
 sin kr   ' Krn
cos Krn 
 cot  
sin  kr  r r n
sin  Krn 

and in the Coulomb field as [Landau and Lifshitz 1987]

rn  1 
 (rn )   2 cot 0   2 H 
ac   

Therefore, equaling both conditions

 rn
 cot        2H 
ac

then,

ac
 cot( 0 ) /  2   cot( )  2 H
rn

The real and imaginary part of the phase function ω are obtained through

 ac 
i Im   cot( ) 
 rn 
ac  cos r  ii  
 Im  r  ii  
rn  sin  r  ii  
ac  cos r cosh i  i sin r sinh i 
 Im  r  ii  
rn  sin r cosh i  i cos r sinh i 
ac   cos r cosh i  i sin r sinh i   sin r cosh i  i cos r sinh i  
 Im  r  ii 
 
  

rn   sin r cosh i  i cos r sinh i   sin r cosh i  i cos r sinh i  
ac   sin r cos r (cosh 2 i  sinh 2 i )  i sinh i cosh i (sin 2 r  cos 2 r )  
 Im  r  ii   

rn   sin 2 r cosh 2 i  cos 2 r sinh 2 i 
  r sin r cos r (cosh i  sinh i )  i sinh i cosh i (sin r  cos r )
2 2 2 2

 
a  i i sin r cos r (cosh 2 i  sinh 2 i )  r sinh i cosh i (sin 2 r  cos 2 r ) 
 c Im
rn  sin 2 r cosh 2 i  cos 2 r sinh 2 i 
 
 
ac  i sin r cos r (cosh 2 i  sinh 2 i )  r sinh i cosh i (sin 2 r  cos 2 r ) 
  
rn 
 sin 2 r cosh 2 i  cos 2 r sinh 2 i 

 ac i sin 2r  r sinh 2i 
 
 sin 2  cosh 2   cos 2  sinh 2   
 2rn r i r i 

a  i sin 2r  r sinh 2i 

 c   
2rn  sin 2 r 1  sinh 2 i   1  sin 2 r  sinh 2 i 

ac  i sin 2r  r sinh 2i 
  
2rn 
 sin 2 r  sinh 2 i 

and

a 
r Re c  cot()  2 H 
 rn 
 r sin r cos r (cosh 2 i  sinh 2 i )  i sinh i cosh i (sin 2 r  cos 2 r ) 
 
ac   ii sin r cos r (cosh 2 i  sinh 2 i )  r sinh i cosh i (sin 2 r  cos 2 r ) 
 Re  2H
rn  sin 2 r cosh 2 i  cos 2 r sinh 2 i 
 
 
ac  r sin r cos r (cosh 2 i  sinh 2 i )  i sinh i cosh i (sin 2 r  cos 2 r ) 
 
rn  sin 2 r cosh 2 i  cos 2 r sinh 2 i   2 H

ac  r sin 2r  i sinh 2i 
    2H
2rn  sin 2 r 1  sinh 2 i   1  sin 2 r  sinh 2 i 
ac  r sin 2r  i sinh 2i 
    2H
2rn  sin 2 r  sinh 2 i 

Hence,

ac  i sin 2r   r sinh  2i  

i Wi /  2   
2rn  sin 2  r   sinh 2  i  

ac  r sin  2r   i sinh  2i  

r Wr /  2     2H
2rn  sin 2 r  sinh 2 i 
The function H can be expressed as
   '   1 / kac   
H  ln 2kac   Re   2 A
    1 / kac   
or as

H  ln 2kac   y  kac   2 A

with A the Euler constant and y(kac) related to the logarithmic derivative of the Γ function given by

y ( x) 

'  x  1 1 
1
 A  ln( x )

 x 1
 2

x
 n n
n 1
2
 x 2
which can be approximated to

1 2
y ( kac )   kac 
12

Table 5 presents the nuclear potential well parameters and average atomic radius for the D+D, D+T
and D+3He fusion reactions. Figure 1 shows the nuclear Woods-Saxon and Coulomb electrostatic
potentials for fusion reactions, whereas figures 2 and 3 presents the complex real and imaginary part
of the phase function ω. From the figures, the energies at which ωr=0, corresponding with the SLBW
resonances, are 1039.58 keV, 51.97 keV and 218.84 keV for D+D, D+T and D+ 3He, respectively.
4. Fusion cross-sections

4.1 ENDF library

ENDF/B-VIII.0 (Evaluated Nuclear Data File) is the library containing the last revision of experimental
fusion cross-section reactions from Brookhaven National Laboratory, based on accelerator beam-
target experiments with incident particle (proton, deuteron, triton and helium-3) into several targets
(H, D ,T, Helium) [Brown et al 2018]

ENDF/B-VIII.0 libraries are available in the EXFOR database (Experimental Nuclear Reaction Data),
which contains an extensive compilation of experimental nuclear reaction data [Otuka 2014].

Table 5 summarizes all the fusion cross section cases that have been used, including geometric,
Gamow and S-factors. These expressions have been compared with ENDF/B-VIII.0 cross sections data
for fusion reactions of the D-cycle: D(d,p)T, D(d,n)3He, T(d,n) 4He and 3He(d,p)4He.

Plasma temperature has been estimated as

 e
T ( K ) E  MeV   
 kb 

where E is the kinetic energy of the incident ions, e the electron charge and kb=1.3807 x 10-23 m2 kg s-2
K-1 the Boltzmann constant.

4.2 Case 1

The fusion cross-section σf (E) was firstly characterized by Duane’s through a 5 parameter expression
as

1 1  A2 
 f ( E )     A 
 
E  exp A1 / E  1    A3 E  A4   1
2 5 


which assumed the geometric factor as a term dependent only of energy, Gamow factor through
Mott’s form of the Coulomb barrier penetrability and the astrophysical S-factor as a [2/2] order Padé
approximation of the classical Breit-Wigner resonance formula.

4.3 Case 2

In this case, the S-factor is estimated by a [4/4] Padé approximation and Gamow factor by its Gamow
constant,

1 1   A1  E ( A2  E ( A3  E ( A4  EA5 ))) 
 f ( E )   
 
E  exp BG / E   1  E ( B1  E ( B2  E ( B3  EB4 ))) 


4.4 Case 3

In the last case, fusion cross-section is calculated as

 1 2      2    4i 

 f ( E )  
2  2  
 
 2 2 2 
      k   exp 2 / kac   1  
  r  i  1 /   

where 1/χ2 is the Gamow factor and β is a factor which depends on the target nuclei.
5. XSDC-Fusion code

The XSDC-Fusion code is a software to compute fusion cross sections, geometric, Gamow-Sommerfeld
and S-factors in the deuterium cycle (D-cycle) and energy range from 1 keV to 5 MeV, equivalent to
plasma temperatures of 11.6 x 10 6 to 5.80 x 1010 K. The software has been developed in Python 3.8
environment and includes ENDF fusion cross section library and estimates fusion cross-sections,
energy resonances and partial widths (FWHM) of all cases above-mentioned. The XSDC-Fusion code
calculates also the reaction heat (Q) of the fusion reactions and the parameters of the Woods-Saxon
and Coulomb proximity potentials.

6. Results and discussion

Figures 4 to 12 show Gamow-Sommerfeld factors and astrophysical S-factors for different fusion
reactions. Significant differences have been observed between Gamow and S-factors, in particular
with SLBW resonance energy values and Γ (FWHM).

There are considerable differences between case 1 and the other cases considering resonance peak
energy for T(d,n)4He and 3He(d,p)4He reactions. In case 1, energy resonances are placed at 78.7 MeV
and 325.9 MeV, respectively, while for cases 2/3 are placed at 49.1/42 MeV and 215.3/207.8 MeV, as
showed in table 6.

In case 2, the S-factor of the D(d,p)T reaction does not include the Breit-Wigner resonance peak as
cases 1 and 3 in 2798.2 MeV and 931.3 keV, respectively. In addition, case 1 includes a single
resonance at 2798.16 keV for reaction D(d,p)T and 3821.43 keV for reaction D(d,n) 3He, respectively,
whereas case 2 has a resonance at 5012.7 keV for reaction D(d,n) 3He and case 3 has a single
resonance in 931.3 keV for both reactions. As a notation, this energy value is equivalent to the
conversion factor keV and uma, although no physical correlation has been observed between them.

Furthermore, D+D fusion reaction for case 3 is independent of reaction products (see fig. 11), while
other cases have different expressions from D(d,p)T and D(d,n) 3He reactions.

Figure 13 presents ENDF/B-VIII.0 fusion cross section data obtained with accelerator-based beam-
target experiments in Brookhaven National Laboratory. These fusion cross sections have been used as
reference values to compare with other fusion cross-section equations (cases 1, 2 and 3).

Figures 14, 15 and 16 show fusion cross section for D+D, D+T and D+ 3He evaluated through cases 1, 2
and 3, whereas table 6 summarizes resonance energies for all cases and fusion reactions and table 7
shows cross-section maximum values and energy Emax. As observed, resonance peaks are mostly
attenuated in all cases and shifted from the energy at which is placed the maximum cross-section
values.

Fusion cross section differences are relatively lower between all cases under study (< 5 %) in the
energy range of [0.001, 5] MeV, compared with ENDF/B-VIII.0, as observed in figures 17, 18 and 19.

In addition, the energy position of the maximum value of fusion cross-section is also shifted between
cases. In cases 2 and 3, fusion reaction 3He(d,p)4He has a saw profile at energies lower than 10 keV.
7. Conclusions

During decades, fusion cross-sections have been estimated with beam-target experiments on
accelerators and interpolated through several parametric equations to quantum mechanical models.
Fusion cross-sections (σ), S-factors and branching ratios (BR) from these experiments should be
compared with real conditions in tokamak and stellerator fusion reactors, in which target particles are
not in rest and have similar kinetic and potential energies of incident particles.

S-factors characterize fusion reactions in a more detailed way than cross sections, showing SLBW
resonance peaks at different energies. However, there are considerable differences between
resonance energies and Γ (FWHM) as observed.

Furthermore, although there are other types of Wood-Saxon proximity potentials instead of the
square-well potential (i.e. optical model), those models have been applied to fusion of heavy nuclei
(Z>8) [Chepurnov 1967, Denisov 2002, Aygun 2018]

In conclusion, fusion reaction models for light nuclei (deuterium, tritium and helium) should be
reviewed in order to apply fusion to energy production in safety conditions.
8. References

Arbuzov AB and Kopylova TV. On relativization of the Sommerfeld-Gamow-Sakharov factor. J High

Ener Phys 9 (2012)

Atzeni S and Meyer-ter-Vehn. Nuclear fusion reactions. Oxford University Press. ISBN: 978-0-19-
85624-1 (2001)

Aygun M. Comparative analysis of proximity potentials to describe scattering of 13C, 16O, 28Si and 208Pb
nuclei. Rev Mex Fis E vol 64 n.2 (2018)

Ball J. Maximizing specific energy by breeding deuterium. ArXiv:1908.00834 [physics.pop-ph]

doi: 10.1088/1741-4326/ab394c

Belov AA, Kalitkin NN and Kozlitin IA. Refinement of thermonuclear reaction rates. Fusion Engineering
and Design 141, 51-58 (2019)

Blatt JM and Weisskopf VF. Theoretical Nuclear Physics (Springer-Verlag, 1979)

Breit G and Wigner E. Capture of Slow Neutrons, Phys. Rev. 49, 519 (1936)

Brennan JG and Coyne JJ. Energy Dependence of the D-D reaction cross section at low Energies. J Res
Natl Bur Stand A Phys Chem 68A(6): 675-682 (1964)

Bosch and Hale 1992. Improved formulas for fusion cross-sections and thermal reactivities. NUCLEAR
FUSION, vol 32, N0.4 (1992)

Brown DA et al. ENDF/B-VIII.0: The 8th major release of the nuclear reaction data library with CIELO-
project cross sections, new standards and thermal scattering data. Nucl Data Sheets 148(2018)1

Bystritsky VM, Grebenyuk VM, Parzhitski SS, Pen’kov FM, Sidorov VT and Stolupin VA. Experimental
Investigation of dd reaction in range of ultralow energies using Z-pinch. D15-98-239. Nucl Inst Meth
(1998)

Chen FF. Introduction to Plasma Physics and Controlled Fusion (vol 1) Springer (2016)
ISBN 978-3-319-22309-4

Chepurnov VA and Fiz Y. 955 (1967)

Denisov VY. Interaction potential between heavy ions. Phys Letters B 526 315-321 (2002)

Donné AJH. The European roadmap towards fusion electricity. Phil. Trans. R. Soc. A 377: 20170432.
(2018) doi:10.1098/rsta.2017.0432

Duane BH. Fusion Cross Section Theory, in Rept. BNWL-1685 (BattellePacific Northwest Laboratory,
USA 1972)

Dudek J, Majhofer A, Skalski J, Werner T, Cwiokand S and Nazarewicz W. J. Phys.G 5,101359 (1979)

Dudek J, Szymanski Z, Werner T, Faessler A and Lima C. Phys. Rev.C261712 (1982)

Englefield MJ. Solution of Coulomb problem by Laplace Transform. J Math Anal Appl. 48, 270-275
(1974)

Ghasemi R and Sadeghi H. S-factor for radiative capture reactions for light nuclei at astrophysical
energies. Results in Physics 9 151-165 (2018)
Green AES. Nuclear Physics, McGraw-Hill, New York 1955.

Greiner W. Quantum Mechanics, 2nd ed. Springer-Verlag Berlin Heidelberg NY (USA) 1993.

Humblet J, Fowler WA and Zimmerman BA. Approximate penetration factors for nuclear reactions of
astrophysical interest. Astron Astrophys 177, 317-325 (1987)

Jing Z, Yuan-Yong F, Shu-Hua Z, Hai-Long X, Cheng-Bo Li and Qiu-Ying M. Measurement of the

astrophysical S factor for the low energy 2H(d,gamma)4He reaction. Chinese Phys C, 33, 5 (2009)

Kim Y, Mack JM, Herrmann HW, Young CS et al. Determination of the deuterium-tritium branching
ratio based on inertial confinement fusion implosions. Phys Rev C85, 061601(R) (2012)

Kiptily VG. On the Core Deuterium-Tritium Fuel Ratio and Temperature Measurements in DEMO.
Nuclear Fusion, 55 (2) (2015) doi: 10.1088/0029-5515/55/2/023008

Knoepfel H. Tokamak Reactors for Breakeven.A Critical Study of the Near-Term Fusion Reactor
Program. Pergamon (1978)

Landau LD and Lifshitz EM. Quantum Mechanics (Pergamon, Oxford 1987) p572

Langenbrunner JL, Weller HR and Tilley DR. Two-deuteron radiative capture: Polarization observables
at Ed ≤ 15 MeV. Phys Rev C 42, 1214 (1990)

Lee M and Jung Y. Quantum shielding effects on the Gamow penetration factor for nuclear fusion
reaction in quantum plasmas. Phys Plasmas 24, 014502 (2017)

Li XZ. Nuclear physics for nuclear fusion. Fusion Science and Technology (2002)
doi:10.13182/FST02-A201

Li XZ et al. Fusion cross sections for fusion energy. Fusion Eng and Design 81, 1517-1520 (2006)

Li XZ, Wei QM and Liu B. A new simple formula for fusion cross-sections of light nuclei. Nucl Fusion 48,
125003 (2008)

Mott NF and Massey HS. The theory of atomic collisions. Oxford University Press, London (1965)

Miley GH, Towner H and Ivich N. Fusion cross sections and reactivities. Rept COO-2218-17 (University
of Illinois, Urbana, IL, USA 1974) Np (1974) doi:10.2172/4014032

Nishitani T, Shibata Y, Tobita K and Kusama Y. Fusion gamma-ray diagnostics for D-3He experiments in
JT-60U. Review Scientific Instruments 72, 877 (2001) doi.org/10.1063/1.1321004

Nocente M, Källne J, Salewski M, Tardocchi M and Gorini G. Gamma-ray emission spectrum from
thermonuclear fusion reactions without intrinsic broadening. Nuclear Fusion 55, 12 (2015).
doi:10.1088/0029-5515/55/12/123009

NRL Plasma Formulary (2019)

https://www.nrl.navy.mil/ppd/content/nrl-plasma-formulary

Ongena J. Nuclear fusion and its large potential for the future world energy supply. Nukleonika; 61(4):
425-432. (2016) doi:10.1515/nuka-2016-0070

Otuka et al. Towards a More Complete and Accurate Experimental Nuclear Reaction Data Library
(EXFOR): International Collaboration Between Nuclear Reaction Data Centres (NRDC) Nucl Data Sheets
120, 272-276 (2014)
Rajbongshi T and Kalita K. Systematic study of deformation effects on fusion cross-sections using
various proximity potentials. Open Physics 12 (6) (2014)
doi:10.2478/s11534-014-0451-1

Ran Y, Xue L, Hu S and Su RK. On the Coulomb-type potential of the one-dimensional Schrödinger
equation. J Phys A: Math Gen 33 9265-9272 (2000)

Robouch BV et al. Gamma diagnostics of thermonuclear plasma using D(d,gamma) 4He reaction:
Feasibility study. ETDE-IT-94-23 IAEA (1993)

Singh V, Atta D, Khan MA and Basu DN. Astrophysical S-factor for deep sub-barrier fusion reactions of
light nuclei. Nuclear Physics A (986) 98-106 (2019)

Smith CL and Cowley S. The path to fusion power. Phyl Trans R Soc 368, 1091-1108 (2010)
doi:10.1098/rsta.2009.0216

Takano H and Ishiguro Y. Multi-level correction to Breit-Wigner Single-Level Formula. Jour Nucl Sci and
Tech 14(9), 627-639 (1977)

Woods LC. Theory of Tokamak Transport (Wiley-VCH, 2006)

Yoon JH and Wong CY. Relativistic modification of the Gamow factor. Phys Rev C 61, 044905 (2000)
doi:10.1103/PhysRevC.61.044905

Zhom H. On the size of tokamak fusion power plants. Phil. Trans. R. Soc. A 377: 20170437.
doi:10.1098/rsta.2017.0437

Zhong-Li X, Liu B, Chen S, Ming Wei Q and Hora H. Fusion cross-sections for inertial fusion energy.
Laser and Particle Beams 22, 469-477 (2004) doi:10.1017/S026303460404011X
Tables

Table 1 - Fusion reactions of the D-cycle between deuterium and other light nuclei
[Langenbrunner et al 1990, Jing et al 2009, Kiptily 2014, Ongena 2016]
Energy released per reaction product
Reaction BRi Q (MeV) [MeV]
(a) (b)
T+p ~ 0.45 -4.03 1.01 3.02
3
D+D He+n ~ 0.55 -3.27 0.82 2.45
4
He+γ 10-7 -10-6 -23.85 0.08 23.77
3
He+2n - +2.98 - -
4
The D-cycle D+T He+n ~1 -17.59 3.52 14.07
5
He+γ(a) 5x10-5 - 5x10-4 -16.70 0.05 / 3.2 16.65 / 13.50
3
He+p+n - +2.22 - -
D+3He 4
He+p ~1 -18.35 3.67 14.68
5
Li+γ (b) 5x10-5 - 5x10-4 -16.39 0.05 16.34

(a) D+T branching ratio by inertial confinement fusion (ICF) [Kim et al 2012]
(b) D+3He branching ratio on JT-60U tokamak [Nishitani et al 2000]

Table 2 - Fusion cross-section parameters for Duane’s 5 parameter equation

[Miley et al 1974, NRL 2019]

D(d,p)T D(d,n)3He T(d,n)4He 3

He(d,p)4He Units Comments
A1 46.097 47.88 45.95 89.27 keV1/2 Gamow constant (BG)
A2 372 482 5.02 x 104 2.59 x 104 keV.b -
A3 4.36 x 10-4 3.08 x 10-4 1.368 x 10-2 3.98 x 10-3 keV -
A4 1.22 1.177 1.076 1.297 - -
A5 0 0 409 647 keV.b -
E0 2798.16 3821.43 78.65 325.88 keV Resonance energy E0=A4dd1/A3dd1
Γa 4587.16 6493.51 146.20 502.51 keV Resonance width Γa=2/A3dd1

Table 3 - S-function (Padé expansion coefficients) [Bosch and Hale 1992]

T(d,n)4He 3
He(d,p)4He
D(d,p)T D(d,n)3He Units
[0.5-550 keV] [550-4700 keV] [0.3-900 keV] [900-4800 keV]
BG 31.3970 31.3970 34.3827 34.3827 68.7508 68.7508 keV1/2
A1 5.5576 x 104 5.3701 x 104 6.927 x 104 -1.4714 x 106 5.7501 x 106 -8.3993 x 105 keV.mb
A2 2.1054 x 102 3.3027 x 102 7.454 x 108 0.0 2.5226x 103 0.0 mb
A3 -3.2638 x 10-2 -1.2706 x 10-1 2.050 x 106 0.0 4.5566x 101 0.0 mb.keV-1
A4 1.4987 x 10-6 2.9327 x 10-5 5.2002 x 104 0.0 0.0 0.0 mb.keV-2
A5 1.8181 x 10-10 -2.5151 x 10-9 0.0 0.0 0.0 0.0 mb.keV-3
B1 0.0 0.0 6.38 x 101 -8.4127 x 10-3 -3.1995 x 10-3 -2.6830 x 10-3 keV-1
B2 0.0 0.0 -9.95 x 10-1 4.7983 x 10-6 -8.5530 x 10-6 1.1633 x 10-6 keV-2
B3 0.0 0.0 6.981 x 10-5 -1.0748 x 10-9 5.9014 x 10-8 -2.1332 x 10-10 keV-3
B4 0.0 0.0 1.728 x 10-4 8.5184 x 10-14 0.0 1.4250 x 10-14 keV-4

Table 4 - Averaged nuclear radius r12, Wood-Saxon nuclear potential well

US = Ur + iUi , Coulomb barrier UC and factor β [Li et al 2008, Singh et al 2018]

D(d,p)T / D(d,n)3He T(d,n)4He 3

He(d,p)4He Units
r0 2.778 1.887 3.331 fm
r1
7.0 5.1 9.0 fm
2

Ur -48.52 -40.69 -11.859 MeV

Ui -263.27 -109.18 -259.02 keV
U
205.71 282.42 319.97 keV
C

β 2 2  2 2 -
 Table 5 - Comparison of geometric, Gamow and S-factors of several fusion cross-section expressions

Case Φ(E) G(E) S(E) Comments Reference

1 Duane’s 5 parameter
  A1    A2  formula with a
1 1/ E  exp    1 A5   2

 [2/2] order Padé
[NRL 2019]
  E  
 A4  A3 E  1  
approximation of S(E)
1 Bosch-Hale formula
  B  A1  E ( A2  E ( A3  E ( A4  EA5 ))) with a [4/4] order [Bosch and
2 1/ E  exp  G  
  E 1  E ( B1  E ( B2  E ( B3  EB4 )))Padé approximation of Hale 1992]
S(E)
1
  2     4i  Woods-Saxon square
3  /k 2
2  exp    1
  complex potential [Li et al 2006,

  kac     2    1 /  2  2 
  with phase shift ω(δ0) 2008]
 r i  boundary conditions

Table 6 - Comparison of resonance parameters for fusion reactions

Case 1 Case 2 Case 3 Units

2798.
D(d,p)T -
2
931.3
3821. 5012.
E0 D(d,n)3He keV
4 7
4
T(d,n) He 78.7 49.10 42
3
He(d,p)4He 325.9 215.3 207.8
4587.
D(d,p)T - 846.4
2
6493. 5517.
Γ D(d,n)3He 846.4
5 9 keV
(FWHM) 4
T(d,n) He 147.4 81.2 82.1
3 287.5
He(d,p)4He 515.2 172.7
5
D(d,p)T 0.14 0.05 0.04
S(0) D(d,n)3He 0.20 0.05 0.04
MeV.b
at 1 keV T(d,n)4He 23.99 11.71 4.25
3
He(d,p)4He 10.34 5.77 1.23
D(d,p)T 0.37 - 0.21
D(d,n)3He 0.48 0.62 0.21
S(E) at E0 MeV.b
T(d,n)4He 50.61 27.29 8.47
3
He(d,p)4He 26.55 16.69 8.19

Table 7 - Comparison of fusion cross section parameters for fusion reactions

ENDF Case 1 Case 2 Case 3 Units

D(d,p)T 1700 2550 7999 912
D(d,n)3He 1200 3400 1109 912
Emax keV
T(d,n)4He 64.8 113.3 64.8 62.3
3
He(d,p)4He 247.8 430.4 262.3 230.1
D(d,p)T 8.9 x 10-2 9.6 x 10-2 1.0 x 10-1 8.9 x 10-2
D(d,n)3He 0.104 0.109 0.105 0.089
σf at Emax barn
T(d,n)4He 5.01 4.94 5.04 5.00
3
He(d,p)4He 0.81 0.72 0.81 0.80
Figures

Coulomb
barrier
Turning point
E=UC(r)
projectile
Tunneling

Fusion

Figure 1 - Woods-Saxon square-well (left) and Coulomb electrostatic proximity potential (right)
for fusion reactions

Figure 2 - Complex phase shift ωr(E) of fusion reactions (Re)

Figure 3 - Complex phase shift ωi(E) of fusion reactions (Im)

Figure 4 - Gamow factors as a function of center-of-mass energy E from Duane’s 5 parameter equation
[NRL 2019]

Figure 5 - Gamow factors as a function of center-of-mass energy E (reduced Mott’s expression)

Figure 6 - Gamow factors (1/χ2) as a function of center-of-mass energy E

 Figure 7 - S-factor of D+D fusion reactions as a function of center-of-mass energy E (Case 1)

Figure 8 - S-factor of D+T and D+3He fusion reactions as a function of center-of-mass energy E (Case 1)

Figure 9 - S-factor for D+D fusion reaction as a function of center-of-mass energy E (Case 2)
Figure 10 - S-factor for D+T and D+3He fusion reactions as a function of center-of-mass energy E (Case 2)

Figure 11 - S-function for D+D fusion reaction as a function of center-of-mass energy E (Case 3)

Figure 12 - S-function for D+T/D+3He fusion reactions as a function of center-of-mass energy E (Case 3)
 Figure 13 - Fusion cross-sections for D+D/D+T/D+ 3He reactions as a function of center-of-mass energy E from
ENDF/B-VIII.0 (Reference)

Figure 14 - Fusion cross-sections for D+D/D+T/D+3He reactions as a function of center-of-mass energy E (Case 1)

Figure 15 - Fusion cross-sections for D+D/D+T/D+3He reactions as a function of center-of-mass energy E (Case 2)
Figure 16 - Fusion cross-sections for D+D/D+T/D+3He reactions as a function of center-of-mass energy E (Case 3)

Figure 17 - Relative differences of fusion cross-sections between case 1 formula and ENDF

Figure 18 - Relative differences of fusion cross-sections between case 2 and ENDF

Figure 19 - Relative differences of fusion cross-sections between case 3 formula and ENDF