betaFeedback

WorksPhysics and Astronomy

Low energy nuclear reactions in a crystal lattice

Abstract

We extend the recently proposed mechanism for inducing low energy nuclear reactions (LENRs) to a crystal lattice. The process gets a dominant contribution at second order in quantum perturbation theory. The tunnelling barrier is evaded due to the presence of high energy intermediate states. However, the process is found to have a negligible rate in free space due to a delicate cancellation between amplitudes corresponding to different states. The delicate cancellation can be evaded in medium in special circumstances leading to an observable rate. Here we consider this process in a one dimensional crystal lattice and find that the rate depends strongly on the form of interaction. We find that for some allowed choices of interactions, the rate is substantial, while other interactions produce a very small rate.

Introduction

There exists considerable experimental evidence for nuclear fusion reactions at low energies [1, 2, 3, 4, 5, 6, 7, 8]. However, theoretically it is so far not clear how such reactions may occur, although there exist many proposals [9, 10, 11, 12, 13, 14, 15, 16, 17] In a recent series of papers [18, 19, 20] we have suggested that low energy nuclear reactions may get significant contribution at second order in perturbation theory. This is similar to the mechanism proposed in [17]. This is illustrated by the basic formula at this order [21, 22]

⟨f∣T(t0,t)∣i⟩=(iℏ)2∑m∫t0tdt′ei(Ef−Em)t′/ℏ⟨f∣HI(t′)∣m⟩∫t0t′dt′′ei(Em−Ei)t′′/ℏ⟨m∣HI(t′′)∣i⟩(1)\langle f|T(t_0,t)|i\rangle= \left(\frac{i}{\hbar}\right)^2 \sum_m \int_{t_0}^{t} dt' e^{i(E_f-E_m)t'/\hbar}\langle f|H_I(t')|m\rangle\int_{t_0}^{t'} dt'' e^{i(E_m-E_i)t''/\hbar}\langle m|H_I(t'')|i\rangle \tag*{(1)}

where ∣i⟩|i\rangle is an initial low energy state, ∣f⟩|f\rangle is the final state, HI(t)H_I(t) is the interaction Hamiltonian and ∣m⟩|m\rangle represent the intermediate states. Here EiE_i, EmE_m and EfE_f represent the initial, intermediate and final state energies and we need to sum over all the intermediate state energy eigenvalues. The process being considered is different from those generally considered at first order which are expected to be strongly suppressed at low energies [23]. At second order new processes open up which do not get any contribution at first order. In [18, 19], for example, the authors considered process with emission of two photons which necessarily requires two perturbations and hence vanishes at first order.

The process takes place in two steps. At the first interaction vertex the initial state ∣i⟩|i\rangle makes a transition to an intermediate state ∣m⟩|m\rangle. At the second interaction vertex the intermediate state ∣m⟩|m\rangle forms the final state nuclei. At both the vertices a additional particle may be emitted or some energy/momentum transfered to the lattice or an impurity particle in the medium. The important point is that the intermediate energies can take arbitrarily large values and energy need not be conserved at each interaction vertex. Hence the Coulomb barrier, may not lead to a significant suppression and the amplitude ⟨f∣HI(t′)∣m⟩\langle f|H_I(t')|m\rangle which leads to the fusion process may not be too small.

In our earlier calculations [18, 19] we assumed that the interaction vertices are electromagnetic leading to emission of a photon at each vertex. A direct calculation reveals that, although for some intermediate state energies EmE_m the amplitude becomes rather large, the sum over all EmE_m takes very small values. Hence the rate turns out to be very small. The basic problem arises at the first vertex. Here we are dealing with eigenstates in a potential which are not momentum eigenstates. We find that a large number of these states contribute and mutually cancel the contribution due to one another. This problem would not have arisen if we had momentum eigenstates and imposed momentum conservation at this vertex. However this is not applicable in the model used in [18, 19].

In [20], we studied this problem in more detail by using a toy model for the potential. We used a step model for the potential barrier since it admits simple analytic expressions for the wave functions. We found that if we assume free space wave functions at large distances, the reaction rate is very small due to a delicate cancellation among different wave functions. This is in agreement with results found in [18, 19]. However, in a medium the energy eigenvalues are expected to be discretized. This is true both for a crystalline lattice [24] and for disordered systems [25, 26]. This idea was implemented in a simple manner in [20] by imposing an infinite potential wall at a large distance from the nuclear potential well. This leads to discretization of energy eigenvalues and now the cancellation is not exact. In this case we find that the reaction rate can be substantial, much larger than rate for the first order process which is strongly suppressed by the potential barrier due to low initial state energy. This shows that although the rate is negligible in free space it can be substantial in a medium. This is also consistent with experimental claims [1, 3, 4, 27], which suggest that the process happens only in a medium.

In this paper we consider the fusion process in a crystal lattice, which provides a realistic model for solid medium. We set up the formalism in general but restrict our calculations to the one dimensional Kronig-Penney model [21]. In this case a proton is assumed to undergo fusion with a lattice ion. We assume that the proton experiences a periodic potential such that

V(r⃗+R⃗)=V(r⃗)(2)V(\vec{r}+\vec{R}) = V(\vec{r}) \tag*{(2)}

We emphasise that we are interested in the potential experienced by a proton and not an electron. Hence at the location of each ion we expect a strong repulsive potential and an attractive potential between the sites. The form of the potential is similar to that of electrons but we expect a much stronger barrier height in the present case. A sample potential in one dimension, which is taken to be a generalization of the Kronig-Penney potential [21], is illustrated in Figure 1. The deep and narrow attractive potential models the nuclear force. This leads to a modification of the wave function at distances very close to the nucleus. At large distances it is expected to lead to only a mild change. For example, in free space it leads to phase shift which shows a mild dependence on energy. The potential can be expressed as,

Figure 1. Schematic illustration of the periodic potential given in Eq. (3).

V(x)=0for a+(n−1)ξ<x<nξ−a=V0for nξ−a<x<nξ−d=−V1for nξ−d<x<nξ+d=V0for nξ+d<x<nξ+a(3)\begin{aligned} V(x) &= 0 && \text{for } a+(n-1)\xi< x < n\xi-a \\ &= V_{0} && \text{for } n\xi-a < x < n\xi-d \\ &= -V_{1} && \text{for } n\xi-d < x < n\xi+d \\ &= V_{0} && \text{for } n\xi+d < x < n\xi+a \tag*{(3)} \end{aligned}

where ξ=2a+2b\xi=2a+2b. Here dd is very small and represents the nuclear length scale. Furthermore the nuclear potential V1V_{1} is taken to be very strong.

The eigenfunctions in a periodic lattice can be expressed as

ψ(r⃗)=1NNeik⃗⋅r⃗uk(r⃗)(4)\psi(\vec{r})=\frac{1}{\sqrt{N\mathcal{N}}}e^{i\vec{k}\cdot\vec{r}}u_{k}(\vec{r}) \tag*{(4)}

such that

uk(r⃗+R⃗)=uk(r⃗)(5)u_{k}(\vec{r}+\vec{R})=u_{k}(\vec{r}) \tag*{(5)}

Here NN is the number of unit cells and N\mathcal{N} is the normalization factor in each unit cell. We shall consider the process in which the proton emits a particle, such as a photon, at the first vertex. The emitted particle may be a virtual photon which may be absorbed by another particle in the medium. In earlier papers we have considered emission of real photons [18, 19, 20]. In the present paper we perform an explicit calculation using a one dimensional model. Hence, for simplicity, we consider emission of a scalar (spin 0) particle since in one dimension only one spatial component is allowed.

General Formalism

In this section we develop the general formalism for the second order process. We shall perform an explicit calculation only in one dimension, however, here we develop the framework assuming three dimensions. The process involves two perturbations or vertices. At the first vertex, the incident particle, which we refer to as a proton, emits a spin 0 particle of high momentum. At the second vertex it undergoes fusion with a lattice site while emitting another spin 0 particle. As explained earlier, the emitted particles are taken to be spin 0 since we perform explicit calculations only in 1 dimension. The net effect is fusion of proton with a lattice site with emission of two scalar particles.

The interaction Hamiltonian can be expressed as

HI=−gϕ(r⃗,t)O(6)H_{I}=-g\phi(\vec{r},t)\mathcal{O} \tag*{(6)}

where ϕ\phi is a scalar field and can be expressed as,

ϕ(r⃗,t)=1Ω∑q⃗ℏ2ω(q)[aq⃗(t)eiq⃗⋅r⃗+aq⃗†(t)e−iq⃗⋅r⃗].(7)\phi(\vec{r},t)=\frac{1}{\sqrt{\Omega}}\sum_{\vec{q}}\sqrt{\frac{\hbar}{2\omega(q)}}\left[a_{\vec{q}}(t)e^{i\vec{q}\cdot\vec{r}}+a_{\vec{q}}^{\dagger}(t)e^{-i\vec{q}\cdot\vec{r}}\right]. \tag*{(7)}

Here ω(q)\omega(q) is the frequency of spin 0 particles, Ω\Omega the spatial volume and a(q⃗)a(\vec{q}) the annihilation operator. In Eq. (6), O\mathcal{O} is an operator which can be chosen to be an identity or some other operator. We discuss its detailed form later. The form of ϕ\phi is chosen so that it has the right dimensions in one spatial dimension since our detailed calculation will be done only for this case. The interaction Hamiltonian in Eq. (6) can have dependence on the momentum of the incident proton in analogy to the interaction with electromagnetic field [18, 19]. This is implemented by introduction of the operator O\mathcal{O}. The constant gg has dimensions which will be specified later.

We next compute the matrix element ⟨m∣HI(t′′)∣i⟩\langle m|H_I(t^{\prime\prime})|i\rangle in Eq. (1) using the interaction Hamiltonian in Eq. (6) setting the operator O\mathcal{O} equal to unity. This perturbation leads to production of a scalar particle of momentum q⃗1\vec{q}_1 and energy E1=ℏω1E_1=\hbar\omega_1. We obtain,

⟨m∣HI(t′′)∣i⟩=−gΩℏ2ω1∫d3r ψm∗(r⃗)e−iq⃗1⋅r⃗ei(Em+ℏω1−Ei)t′′/ℏψi(r⃗)(8)\langle m|H_I(t^{\prime\prime})|i\rangle=-\frac{g}{\sqrt{\Omega}}\sqrt{\frac{\hbar}{2\omega_1}}\int d^3r\,\psi_m^*(\vec{r})e^{-i\vec{q}_1\cdot\vec{r}}e^{i(E_m+\hbar\omega_1-E_i)t^{\prime\prime}/\hbar}\psi_i(\vec{r}) \tag*{(8)}

where ψi\psi_i and ψm\psi_m represent the initial and intermediate state wave function corresponding to wave vectors k⃗i\vec{k}_i and k⃗m\vec{k}_m respectively. This leads to

⟨m∣HI(t′′)∣i⟩=−gΩNiNmℏ2ω1ei(Em+ℏω1−Ei)t′′/ℏ∫d3r e−ik⃗m⋅r⃗uk⃗m∗(r⃗)e−iq⃗1⋅r⃗eik⃗i⋅r⃗uk⃗i(r⃗)(9)\langle m|H_I(t^{\prime\prime})|i\rangle=-\frac{g}{\sqrt{\Omega\mathcal{N}_i\mathcal{N}_m}}\sqrt{\frac{\hbar}{2\omega_1}}e^{i(E_m+\hbar\omega_1-E_i)t^{\prime\prime}/\hbar}\int d^3r\,e^{-i\vec{k}_m\cdot\vec{r}}u_{\vec{k}_m}^*(\vec{r})e^{-i\vec{q}_1\cdot\vec{r}}e^{i\vec{k}_i\cdot\vec{r}}u_{\vec{k}_i}(\vec{r}) \tag*{(9)}

Here we have also substituted for the proton wave functions using Eq. (4) and Ni\mathcal{N}_i and Nm\mathcal{N}_m are the normalization factors for the initial and intermediate states respectively. We now split the integral into an integral over a unit cell and sum over all cells. Hence we write r⃗\vec{r} as

r⃗=s⃗+R⃗j(10)\vec{r}=\vec{s}+\vec{R}_{j} \tag*{(10)}

This leads to,

⟨m∣HI(t′′)∣i⟩=−gΩNiNmNℏ2ω1ei(Em+ℏω1−Ei)t′′/ℏ×∑j∫d3se−ik⃗m⋅(s⃗+R⃗j)uk⃗m∗(s⃗)e−iq⃗1⋅(s⃗+R⃗j)eik⃗i⋅(s⃗+R⃗j)uk⃗i(s⃗)(11)\langle m|H_I(t^{\prime\prime})|i\rangle=-\frac{g}{\sqrt{\Omega\mathcal{N}_i\mathcal{N}_m}\mathcal{N}}\sqrt{\frac{\hbar}{2\omega_1}}e^{i(E_m+\hbar\omega_1-E_i)t^{\prime\prime}/\hbar} \times\sum_j\int d^3s e^{-i\vec{k}_m\cdot(\vec{s}+\vec{R}_j)}u_{\vec{k}_m}^{*}(\vec{s})e^{-i\vec{q}_1\cdot(\vec{s}+\vec{R}_j)}e^{i\vec{k}_i\cdot(\vec{s}+\vec{R}_j)}u_{\vec{k}_i}(\vec{s}) \tag*{(11)}

where we have used the periodicity of uku_k displayed in Eq. (5). We can now perform the sum over jj using,

∑jei(−k⃗m−q⃗1+k⃗i)⋅R⃗j=N∑G⃗δG⃗,−k⃗m−q⃗1+k⃗i(12)\sum_j e^{i(-\vec{k}_m-\vec{q}_1+\vec{k}_i)\cdot\vec{R}_j}=N\sum_{\vec{G}}\delta_{\vec{G},-\vec{k}_m-\vec{q}_1+\vec{k}_i} \tag*{(12)}

where G⃗\vec{G} are the reciprocal lattice vectors. Substituting this, we obtain,

⟨m∣HI(t′′)∣i⟩=−gΩNiNmℏ2ω1ei(Em+ℏω1−Ei)t′′/ℏ×∑G⃗δG⃗,−k⃗m−q⃗1+k⃗i∫d3se−ik⃗m⋅s⃗uk⃗m∗(s⃗)e−iq⃗1⋅s⃗eik⃗i⋅s⃗uk⃗i(s⃗)(13)\langle m|H_I(t^{\prime\prime})|i\rangle=-\frac{g}{\sqrt{\Omega\mathcal{N}_i\mathcal{N}_m}}\sqrt{\frac{\hbar}{2\omega_1}}e^{i(E_m+\hbar\omega_1-E_i)t^{\prime\prime}/\hbar} \times\sum_{\vec{G}}\delta_{\vec{G},-\vec{k}_m-\vec{q}_1+\vec{k}_i}\int d^3s e^{-i\vec{k}_m\cdot\vec{s}}u_{\vec{k}_m}^{*}(\vec{s})e^{-i\vec{q}_1\cdot\vec{s}}e^{i\vec{k}_i\cdot\vec{s}}u_{\vec{k}_i}(\vec{s}) \tag*{(13)}

Hence k⃗m\vec{k}_m takes only those values which satisfy, k⃗m=−q⃗1+k⃗i−G⃗\vec{k}_m=-\vec{q}_1+\vec{k}_i-\vec{G}.

We next consider the second matrix element ⟨f∣HI(t′)∣m⟩\langle f|H_I(t^{\prime})|m\rangle. In this case the intermediate state makes a transition to final state with emission of another scalar particle of wave vector q⃗2\vec{q}_2 and energy E2=ℏω2E_2=\hbar\omega_2. The proton undergoes fusion with a lattice ion, whose atomic number changes from ZZ to Z′=Z+1Z^{\prime}=Z+1. Hence the final state is the proton bound to a lattice ion due to the nuclear potential which is the narrow deep well shown in Figure 1. This wave function is also periodic and takes large values only very close to an ion over the range of nuclear potential and decays rapidly between two ions. Hence the final state wave function also takes the form shown in Eq. (4) and Eq. (5). Physically this implies that the proton can undergo fusion with any of the lattice ions.

We obtain

⟨f∣HI(t′)∣m⟩=−gΩNfNmNℏ2ω2ei(Ef+ℏω2−Em)t′/ℏ∫d3re−ik⃗f⋅r⃗uk⃗f∗(r⃗)e−iq⃗2⋅r⃗eik⃗m⋅r⃗uk⃗m(r⃗)(14)\langle f|H_I(t^{\prime})|m\rangle=-\frac{g}{\sqrt{\Omega\mathcal{N}_f\mathcal{N}_m}\mathcal{N}}\sqrt{\frac{\hbar}{2\omega_2}}e^{i(E_f+\hbar\omega_2-E_m)t^{\prime}/\hbar}\int d^3r e^{-i\vec{k}_f\cdot\vec{r}}u_{\vec{k}_f}^{*}(\vec{r})e^{-i\vec{q}_2\cdot\vec{r}}e^{i\vec{k}_m\cdot\vec{r}}u_{\vec{k}_m}(\vec{r}) \tag*{(14)}

Here we have replaced the wave functions using Eq. (4) and Nf\mathcal{N}_f is the normalization factor corresponding to the final state. We point out that the wave vector k⃗f\vec{k}_f depends on the value of the deep nuclear potential well. We again set r⃗=s⃗+R⃗l\vec{r}=\vec{s}+\vec{R}_l and use periodicity to replace the integral in terms of integral over a unit cell and sum over all cells. In this case the sum over cells leads to

∑lei(−k⃗f−q⃗2+k⃗m)⋅R⃗l=N∑G⃗′δG⃗′,−k⃗f−q⃗2+k⃗m(15)\sum_l e^{i(-\vec{k}_f-\vec{q}_2+\vec{k}_m)\cdot\vec{R}_l}=N\sum_{\vec{G}^{\prime}}\delta_{\vec{G}^{\prime},-\vec{k}_f-\vec{q}_2+\vec{k}_m} \tag*{(15)}

We need to compute the matrix element for a fixed value of the final state wave vector k⃗f\vec{k}_f. For the fixed value of this wave vector, the wave vector of the final state ϕ\phi particle takes the value q⃗2=−k⃗f+k⃗m−G⃗′=k⃗i−G⃗′−G⃗−q⃗1−k⃗f\vec{q}_2=-\vec{k}_f+\vec{k}_m-\vec{G}'=\vec{k}_i-\vec{G}'-\vec{G}-\vec{q}_1-\vec{k}_f where G⃗′\vec{G}' is a reciprocal lattice vector. Hence we obtain,

⟨f∣HI(t′)∣m⟩=−gΩNfNmℏ2ω2ei(Ef+ℏω2−Em)t′/ℏ×∑G⃗′δG⃗′,−k⃗f−q⃗2+k⃗m∫d3se−ik⃗f⋅s⃗uk⃗f∗(s⃗)e−iq⃗2⋅s⃗eik⃗m⋅s⃗uk⃗m(s⃗)(16)\langle f|H_I(t')|m\rangle=-\frac{g}{\sqrt{\Omega N_fN_m}}\sqrt{\frac{\hbar}{2\omega_2}}e^{i(E_f+\hbar\omega_2-E_m)t'/\hbar}\times\sum_{\vec{G}'}\delta_{\vec{G}',-\vec{k}_f-\vec{q}_2+\vec{k}_m}\int d^3s e^{-i\vec{k}_f\cdot\vec{s}}u_{\vec{k}_f}^{*}(\vec{s})e^{-i\vec{q}_2\cdot\vec{s}}e^{i\vec{k}_m\cdot\vec{s}}u_{\vec{k}_m}(\vec{s}) \tag*{(16)}

In this equation the integral is only over a unit cell.

The transition amplitude can now be expressed as

⟨f∣T(t0,t)∣i⟩=iℏg2ℏ2Ωω1ω2∫t0tdt′ei(Ef+ℏω2+ℏω1−Ei)t′/ℏI(17)\langle f|\mathcal{T}(t_0,t)|i\rangle=\frac{i}{\hbar}\frac{g^2\hbar}{2\Omega\sqrt{\omega_1\omega_2}}\int_{t_0}^{t}dt'e^{i(E_f+\hbar\omega_2+\hbar\omega_1-E_i)t'/\hbar}I \tag*{(17)}

where

I=∑m∑G⃗′∑G⃗δG⃗′,−k⃗f−q⃗2+k⃗mI2I1Em−Ei+ℏω1δG⃗,−k⃗m−q⃗1+k⃗i(18)I=\sum_m\sum_{\vec{G}'}\sum_{\vec{G}}\delta_{\vec{G}',-\vec{k}_f-\vec{q}_2+\vec{k}_m}\frac{I_2I_1}{E_m-E_i+\hbar\omega_1}\delta_{\vec{G},-\vec{k}_m-\vec{q}_1+\vec{k}_i} \tag*{(18)}
I1=∫d3sψk⃗m∗(s⃗)e−iq⃗1⋅s⃗ψk⃗i(s⃗)(19)I_1=\int d^3s\psi_{\vec{k}_m}^{*}(\vec{s})e^{-i\vec{q}_1\cdot\vec{s}}\psi_{\vec{k}_i}(\vec{s}) \tag*{(19)}

and

I2=∫d3sψk⃗f∗(s⃗)e−iq⃗2⋅s⃗ψk⃗m(s⃗)(20)I_2=\int d^3s\psi_{\vec{k}_f}^{*}(\vec{s})e^{-i\vec{q}_2\cdot\vec{s}}\psi_{\vec{k}_m}(\vec{s}) \tag*{(20)}

Here we have absorbed the normalization factors in the wave functions. We may use either the extended or the reduced band scheme. In the extended band scheme each k⃗m\vec{k}_m is uniquely associated with an energy eigenvalue EmE_m.

For some values of k⃗i\vec{k}_i, k⃗m\vec{k}_m and q⃗1\vec{q}_1, the matrix element is expected to be quite large. However, in analogy with the results in [18, 19], it is possible that as we sum the complete amplitude over all kmk_m the amplitude may cancel out. As seen in [20], such a cancellation does not happen if the energy spectrum takes discrete values. This may be applicable in the present case also. In the sum over mm, k⃗m\vec{k}_m take only those values which satisfy k⃗m=q⃗1+k⃗i−G⃗\vec{k}_m=\vec{q}_1+\vec{k}_i-\vec{G} and the energies are corresponding restricted. This leads to discretization of energy eigenvalues since in each band we get contribution from only one state. Hence we may find a substantial rate in the present case also. We examine this by using the generalized one dimensional Kronig-Penney model in the next section.

One dimensional generalized Kronig-Penney Model

We specialize to the one-dimensional potential given in Eq. (3). The eigenfunctions for E<V0E < V_0 are given by,

ψ=1N[Aneik′(x−nξ)+Bne−ik′(x−nξ)]for a+(n−1)ξ<x<nξ−a=1N[Cne−κ(x−nξ)+Dneκ(x−nξ)]for nξ−a<x<nξ−d=1N[EneiQ(x−nξ)+Fne−iQ(x−nξ)]for nξ−d<x<nξ+d=1N[Gne−κ(x−nξ)+Hneκ(x−nξ)]for nξ+d<x<nξ+a(21)\begin{aligned} \psi&= \frac{1}{\mathcal{N}}\left[A_n e^{ik'(x-n\xi)}+B_n e^{-ik'(x-n\xi)}\right] && \text{for } a+(n-1)\xi< x < n\xi-a \\ &= \frac{1}{\mathcal{N}}\left[C_n e^{-\kappa(x-n\xi)}+D_n e^{\kappa(x-n\xi)}\right] && \text{for } n\xi-a < x < n\xi-d \\ &= \frac{1}{\mathcal{N}}\left[E_n e^{iQ(x-n\xi)}+F_n e^{-iQ(x-n\xi)}\right] && \text{for } n\xi-d < x < n\xi+d \\ &= \frac{1}{\mathcal{N}}\left[G_n e^{-\kappa(x-n\xi)}+H_n e^{\kappa(x-n\xi)}\right] && \text{for } n\xi+d < x < n\xi+a \tag*{(21)} \end{aligned}

where

k′=2mEℏ2,(22)k'=\sqrt{\frac{2mE}{\hbar^2}}, \tag*{(22)}
κ=2m(V0−E)ℏ2,(23)\kappa=\sqrt{\frac{2m(V_0-E)}{\hbar^2}}, \tag*{(23)}
Q=2m(E+V1)ℏ2,(24)Q=\sqrt{\frac{2m(E+V_1)}{\hbar^2}}, \tag*{(24)}

mm is the mass of the particle, EE is the energy eigenvalue and N\mathcal{N} is the normalization factor. It is convenient to work in analogy with the standard Kronig-Penney model [21]. Here we use the same formalism and notation. The coefficients AnA_n in two adjacent cells are related by

(AnBn)=(α1+iβ1iβ2−iβ2α1−iβ1)(An+1e−ik′ξBn+1eik′ξ)(25)\begin{pmatrix} A_n \\ B_n \end{pmatrix} = \begin{pmatrix} \alpha_1+i\beta_1 & i\beta_2 \\ -i\beta_2 & \alpha_1-i\beta_1 \end{pmatrix} \begin{pmatrix} A_{n+1}e^{-ik'\xi} \\ B_{n+1}e^{ik'\xi} \tag*{(25)} \end{pmatrix}

The variables α1\alpha_1, β1\beta_1 and β2\beta_2 are given by

α1=cos⁡2k′a2[Vcosh⁡2κ(a−d)−Xsinh⁡2κ(a−d)]−sin⁡2k′a4{(k′κ−κk′)[−Vsinh⁡2κ(a−d)+Xcosh⁡2κ(a−d)]−Y(k′κ+κk′)}(26)\alpha_1 = \frac{\cos2k'a}{2}\left[V\cosh2\kappa(a-d)-X\sinh2\kappa(a-d)\right] -\frac{\sin2k'a}{4} \left\{ \left(\frac{k'}{\kappa}-\frac{\kappa}{k'}\right) \left[-V\sinh2\kappa(a-d)+X\cosh2\kappa(a-d)\right] -Y\left(\frac{k'}{\kappa}+\frac{\kappa}{k'}\right) \right\} \tag*{(26)}
β1=sin⁡2k′a2[Vcosh⁡2κ(a−d)−Xsinh⁡2κ(a−d)]+cos⁡2k′a4{(k′κ−κk′)[−Vsinh⁡2κ(a−d)+Xcosh⁡2κ(a−d)]−Y(k′κ+κk′)}(27)\beta_1 = \frac{\sin2k'a}{2}\left[V\cosh2\kappa(a-d)-X\sinh2\kappa(a-d)\right] +\frac{\cos2k'a}{4} \left\{ \left(\frac{k'}{\kappa}-\frac{\kappa}{k'}\right) \left[-V\sinh2\kappa(a-d)+X\cosh2\kappa(a-d)\right] -Y\left(\frac{k'}{\kappa}+\frac{\kappa}{k'}\right) \right\} \tag*{(27)}
β2=14{(k′κ+κk′)[Vsinh⁡2κ(a−d)−Xcosh⁡2κ(a−d)]+Y(k′κ−κk′)}(28)\beta_2 = \frac{1}{4} \left\{ \left(\frac{k'}{\kappa}+\frac{\kappa}{k'}\right) \left[V\sinh2\kappa(a-d)-X\cosh2\kappa(a-d)\right] +Y\left(\frac{k'}{\kappa}-\frac{\kappa}{k'}\right) \right\} \tag*{(28)}

where

V=2cos⁡2Qd,X=(Qκ−κQ)sin⁡2Qd,Y=(Qκ+κQ)sin⁡2Qd(29)\begin{aligned} V &= 2\cos2Qd,\\ X &= \left(\frac{Q}{\kappa}-\frac{\kappa}{Q}\right)\sin2Qd,\\ Y &= \left(\frac{Q}{\kappa}+\frac{\kappa}{Q}\right)\sin2Qd \tag*{(29)} \end{aligned}
Equation not extracted
(30)

The corresponding quantities for E>V0E > V_0 are obtained by replacing κ\kappa with −iK-iK where

K=2m(E−V0)ℏ2,(31)K = \sqrt{\frac{2m(E-V_0)}{\hbar^2}}, \tag*{(31)}

We define the transfer operator PP [21],

P=((α1−iβ1)eik′ξ−iβ2eik′ξiβ2e−ik′ξ(α1+iβ1)e−ik′ξ)(32)P = \begin{pmatrix} (\alpha_1-i\beta_1)e^{ik'\xi} & -i\beta_2e^{ik'\xi}\\ i\beta_2e^{-ik'\xi} & (\alpha_1+i\beta_1)e^{-ik'\xi} \end{pmatrix} \tag*{(32)}

which satisfy det⁡P=α12+β12−β22=1\det P=\alpha_1^2+\beta_1^2-\beta_2^2=1. The eigenfunctions take the form of Eq. (4) with

cos⁡kξ=α1cos⁡k′ξ+β1sin⁡k′ξ(33)\cos k\xi=\alpha_1\cos k'\xi+\beta_1\sin k'\xi \tag*{(33)}

For any choice of kk this leads to all the allowed energy eigenvalues. The eigenvectors are obtained by solving the matrix equation

P(A0±B0±)=p±(A0±B0±)(34)P \begin{pmatrix} A_0^\pm\\ B_0^\pm \end{pmatrix} = p_\pm \begin{pmatrix} A_0^\pm\\ B_0^\pm \end{pmatrix} \tag*{(34)}

where the eigenvalues p±=exp⁡(±ikξ)p_\pm=\exp(\pm ik\xi).

We next compute the integrals I1I_1 (Eq. (19)) and I2I_2 (Eq. (20)) using the generalized Kronig-Penney model for eigenvalue p+p_+ of the transfer operator. Here we specialize to one cell and set n=0n=0. The coefficients A0+A_0^+ and B0+B_0^+ can be written as,

A0+=β2eik′ξ/2B0+=B0a+e−ik′ξ/2(35)\begin{aligned} A_0^+ &= \beta_2e^{ik'\xi/2}\\ B_0^+ &= B_{0a}^+e^{-ik'\xi/2} \tag*{(35)} \end{aligned}

where,

B0a+=α1sin⁡k′ξ−β1cos⁡k′ξ−sin⁡kξ(36)B_{0a}^+=\alpha_1\sin k'\xi-\beta_1\cos k'\xi-\sin k\xi \tag*{(36)}

We also need to make a choice for the operator O\mathcal{O} in Eq. (6). So far we have chosen this to be just the identity operator. For this choice, I1I_1 can be written as

I1=∫dxψkm′∗(x)e−iq1xψki′(x)(37)I_1=\int dx\psi_{k_m'}^*(x)e^{-iq_1x}\psi_{k_i'}(x) \tag*{(37)}

Let ki′k_i' and km′k_m' denote the wave vectors defined in Eq. (22) corresponding to the initial and intermediate states respectively. Furthermore we denote variables, such as, β2\beta_{2}, B0+B_0^+ corresponding to the two wave functions as β2m\beta_{2m}, β2i\beta_{2i}, B0m+B_{0m}^+, B0i+B_{0i}^+ etc. The dominant contribution to the integral I1I_1 is obtained from the region −a−2b<x<−a-a-2b<x<-a where V(x)=0V(x)=0. The integral in this region, both for E<V0E<V_0 and E>V0E>V_0, is found to be equal to

I1=1NiNm2ei(a+b)q1[β2mβ2isin⁡b(km′−ki′+q1)km′−ki′+q1+β2mB0ia+sin⁡b(km′+ki′+q1)km′+ki′+q1+B0ma+β2isin⁡b(km′+ki′−q1)km′+ki′−q1+B0ma+B0ia+sin⁡b(km′−ki′−q1)km′−ki′−q1](38)I_1=\frac{1}{\mathcal{N}_i\mathcal{N}_m}2e^{i(a+b)q_1}\left[\beta_{2m}\beta_{2i}\frac{\sin b(k'_m-k'_i+q_1)}{k'_m-k'_i+q_1}+\beta_{2m}B^+_{0ia}\frac{\sin b(k'_m+k'_i+q_1)}{k'_m+k'_i+q_1}+B^+_{0ma}\beta_{2i}\frac{\sin b(k'_m+k'_i-q_1)}{k'_m+k'_i-q_1}+B^+_{0ma}B^+_{0ia}\frac{\sin b(k'_m-k'_i-q_1)}{k'_m-k'_i-q_1}\right] \tag*{(38)}

We notice that this integral can be substantial in kinematic regions where the argument of any of the sine function, such as (km′−ki′+q1)(k'_m-k'_i+q_1), vanishes. In other regions the integral is relatively small because the wave function ψi\psi_i is strongly suppressed due to the potential barrier. Furthermore the enhancement seen in the region −a−2b<x<−a-a-2b<x<-a due to vanishing of the argument of any of the sine functions, does not happen in the remaining regions and hence contributions from those regions is further suppressed. In our calculations we explicitly compute the contribution to I1I_1 in the regions adjacent to dominant region and found it to be negligible.

The integral I2I_2, which is analogous to the nuclear matrix element, can be written as

I2=∫dxψkf∗(x)e−iq2xψkm(x)(39)I_2=\int dx\psi^*_{k_f}(x)e^{-iq_2x}\psi_{k_m}(x) \tag*{(39)}

This integral gets dominant contribution from the region −d<x<d-d<x<d and the regions adjacent to this. In the adjacent regions, we get contributions only from values of xx very close to ∣x∣=d|x|=d. As xx moves further away from this region, the wave function ψkf\psi_{k_f} is very strongly suppressed. In the region −d<x<d-d<x<d we can write the wave function ψkf\psi_{k_f} as

ψkf=1Nf[E0feiQfx+F0fe−iQfx](40)\psi_{k_f}=\frac{1}{\mathcal{N}_f}\left[E_{0f}e^{iQ_fx}+F_{0f}e^{-iQ_fx}\right] \tag*{(40)}

where

Qf=2m(Ef+V1)ℏ2,(41)Q_f=\sqrt{\frac{2m(E_f+V_1)}{\hbar^2}}, \tag*{(41)}

In the adjacent regions, it is reasonable to approximate the wave function to be an exponentially falling function which is practically zero once we substantially far from the region −d<x<d-d<x<d. The contribution to the integral I2I_2 from the dominant region can be written as,

I2=2E0f∗E0msin⁡d(Qf+q2−Qm)Qf+q2−Qm+2E0f∗F0msin⁡d(Qf+q2+Qm)Qf+q2+Qm+2F0f∗E0msin⁡d(Qf−q2+Qm)Qf−q2+Qm+2F0f∗F0msin⁡d(Qf−q2−Qm)Qf−q2−Qm(42)\begin{aligned} I_2={}&2E^*_{0f}E_{0m}\frac{\sin d(Q_f+q_2-Q_m)}{Q_f+q_2-Q_m}+2E^*_{0f}F_{0m}\frac{\sin d(Q_f+q_2+Q_m)}{Q_f+q_2+Q_m}\\ &+2F^*_{0f}E_{0m}\frac{\sin d(Q_f-q_2+Q_m)}{Q_f-q_2+Q_m}+2F^*_{0f}F_{0m}\frac{\sin d(Q_f-q_2-Q_m)}{Q_f-q_2-Q_m} \tag*{(42)} \end{aligned}

where Qm=2m(Em+V1)/ℏ2Q_m=\sqrt{2m(E_m+V_1)/\hbar^2}. In our calculations we have included the contribution from adjacent regions which also give a significant contribution.

Results

We choose the following parameters for the potential:

V0=100 atomic unitsV1=50.0×106 eVa=0.1 atomic unitsb=1.0 atomic unitsd=5.0×10−5 atomic units(43)\begin{aligned} V_0 &= 100\ \text{atomic units}\\ V_1 &= 50.0 \times10^{6}\ \text{eV}\\ a &= 0.1\ \text{atomic units}\\ b &= 1.0\ \text{atomic units}\\ d &= 5.0 \times10^{-5}\ \text{atomic units} \tag*{(43)} \end{aligned}

The initial state is chosen to have a very small energy eigenvalue. For the chosen parameters we obtain an eigen state with energy equal to 2.68×10−32.68 \times10^{-3} in atomic units, which we choose for our calculations. This energy eigenvalue Ei≪V0E_i \ll V_0 and hence it shows very little dependence on kik_i within the band. The same is also true for the final state wave function whose energy eigenvalue EfE_f lies deep in the nuclear potential well. We take the final state eigenfunction corresponding to the energy eigenvalue Ef=−10.96E_f=-10.96 MeV. This is the second excited state with the ground state energy equal to −45.3-45.3 MeV. Given the initial wave number kik_i we need to sum over all energy eigenvalues EmE_m for which km=−q1+ki−Gk_m=-q_1+k_i-G where GG are the reciprocal lattice vectors. Here kik_i, and hence q1q_1 can be chosen arbitrarily. Once this choice is made, we need to sum over all kmk_m values arising due to different reciprocal lattice vectors.

The reaction rate in 1 dimension can be written as

dPdt=1ΔT∫dE1dE2ρ1ρ2∑kf∣⟨f∣T(t0,t)∣i⟩∣2(44)\frac{dP}{dt}=\frac{1}{\Delta T}\int dE_1dE_2\rho_1\rho_2\sum_{k_f}\left|\langle f|T(t_0,t)|i\rangle\right|^2 \tag*{(44)}

where E1=ℏω1E_1=\hbar\omega_1 and E2=ℏω2E_2=\hbar\omega_2 are the two emitted particle energies and the factor ρ1=ρ2=Ω/(2πℏc)\rho_1=\rho_2=\Omega/(2\pi\hbar c). The expression for the matrix element is given in Eq. (17). We obtain

dPdt=g4ℏ8πc2∫dE2E1E2∑kf∣I∣2(45)\frac{dP}{dt}=\frac{g^4\hbar}{8\pi c^2}\int\frac{dE_2}{E_1E_2}\sum_{k_f}|I|^2 \tag*{(45)}

along with the energy conservation constraint, Ef+E2+E1−Ei=0E_f+E_2+E_1-E_i=0. We next express gg as

g=g′mec5/2ℏ(46)g=g'\frac{m_ec^{5/2}}{\sqrt{\hbar}} \tag*{(46)}

where g′g' is a dimensionless coupling and mem_e is the electron mass which is equal to unity in atomic units.

We rewrite the equations for the sum/integrals II, I1I_1 and I2I_2 in form suitable for one dimension:

I=∑m∑G′∑GδG′,−kf−q2+kmI2I1Em−Ei+ℏω1δG,−km−q1+ki(47)I=\sum_m\sum_{G'}\sum_G\delta_{G',-k_f-q_2+k_m}\frac{I_2I_1}{E_m-E_i+\hbar\omega_1}\delta_{G,-k_m-q_1+k_i} \tag*{(47)}
I1=∫dxψkm∗(x)e−iq1xψki(x)(48)I_1=\int dx\psi_{k_m}^{*}(x)e^{-iq_1x}\psi_{k_i}(x) \tag*{(48)}

and

I2=∫dxψkf∗(x)e−iq2xψkm(x)(49)I_2 = \int dx\psi_{k_f}^{*}(x)e^{-iq_2x}\psi_{k_m}(x) \tag*{(49)}

With kik_i and q1q_1 fixed kmk_m takes only one value in a particular band. The sum over GG amounts to a sum over kmk_m values across different bands. We also have the sum over kfk_f. Here we will restrict ourselves to a single band and then the sum is over the kfk_f within the band. We point out that the energy EfE_f is almost independent of kfk_f within a band. With kfk_f restricted to a single band, it is clear that, for fixed q2q_2, the delta function δG′,−kf−q2+km\delta_{G',-k_f-q_2+k_m} will restrict kfk_f to a single value. The value of q2q_2 needs to consistent with the energy conservation. Since EfE_f is almost independent of kfk_f, this can be easily satisfied. This takes care of all the sums in the rate formula and we are left with just a single integral over E2E_2. As we change the value of E2E_2, the energy E1E_1 will also change and we will get contribution from different kmk_m values within a band.

Explicit calculations show that although the integral I1I_1 and I2I_2 are substantial for some values of kmk_m, after summing over all intermediate states, the amplitude II takes very small values. This is similar to what was seen earlier [18, 19] and also for the case of continuous eigenvalues in [20]. It is somewhat difficult to get a precise result for this calculation since it is found to be within the numerical error in the calculation. Furthermore the integral undergoes oscillations and to get a final result one need to go to very large values of the intermediate energy eigenvalue. This further inhibits obtaining a precise result and we do not pursue this further.

Our main aim in this paper is to determine if the proposed second order fusion mechanism is prohibited or strongly suppressed in Quantum Mechanics. In order to address this we test some other operators in the interaction Hamiltonian. One interesting choice of operator is

O=1P2−P02(50)\mathcal{O} = \frac{1}{P^2-P_0^2} \tag*{(50)}

where PP is the momentum operator and P0P_0 is a parameter such that E0=P02/(2m)E_0=P_0^2/(2m) does not belong to the eigen-spectrum of the model. We choose P0=1P_0=1 in atomic units which satisfies the above mentioned constraint for the choice of potential parameters. This is a well defined Hermitian operator and hence allowed by Quantum Mechanics. It is not clear which physical interaction may lead to this operator. It may arise if, during the initial stage, the process takes place through the Hydrogen atom rather than a free proton. In this case the incident momentum PP is shared both by the proton and the electron through their mutual Coulomb interaction. Hence the process would be suppressed by an additional factor of 1/P21/P^2, which acts as a form factor for the Hydrogen atom. This may further enhance the reaction rate since the Hydrogen atom would experience a reduced Coulomb barrier. In any case, our main aim in studying this interaction is to determine the cancellation seen earlier is a generic result or happens only for the some interactions. Hence, this operator serves our purpose to determine if mathematically Quantum Mechanics permits enhanced fusion at second order. We point out that for the first order process, the action of this operator will not lead to any enhancement of rate.

The action of the operator given in Eq. (50) on a state is facilitated by making an expansion around P0P_0. For the wave functions under consideration, the action of PP is rather simple and one can easily sum the resulting series. Hence we obtain, setting ℏ=1\hbar= 1,

1P2−P02e±ik′x=1k′2−P02e±ik′x\frac{1}{P^2-P_0^2}e^{\pm ik'x}=\frac{1}{k'^2-P_0^2}e^{\pm ik'x}
1P2−P02e±κx=1−κ2−P02e±κx(51)\frac{1}{P^2-P_0^2}e^{\pm\kappa x}=\frac{1}{-\kappa^2-P_0^2}e^{\pm\kappa x} \tag*{(51)}

Using this we can easily compute all the matrix elements. For this model we refer to the integrals corresponding to I1I_1 and I2I_2 as Iˉ1\bar{I}_1 and Iˉ2\bar{I}_2 respectively and the sum II (Eq. (47)) is denoted by Iˉ\bar{I}. We obtain

Iˉ1=∫dxψkm∗(x)e−iq1xOψki(x)\bar{I}_1=\int dx\psi_{k_m}^{*}(x)e^{-iq_1x}\mathcal{O}\psi_{k_i}(x)
Iˉ2=∫dxψkf∗(x)e−iq2xOψkm(x)(52)\bar{I}_2=\int dx\psi_{k_f}^{*}(x)e^{-iq_2x}\mathcal{O}\psi_{k_m}(x) \tag*{(52)}

and the operation of O\mathcal{O} can be computed using Eq. (51). Furthermore Iˉ\bar{I} is obtained by replacing I1I_1 and I2I_2 in Eq. (47) by Iˉ1\bar{I}_1 and Iˉ2\bar{I}_2 respectively. It is also convenient to define a dimensionless coupling g′′g'' for this interaction. In analogy with Eq. (46), we obtain,

g=g′′me3c9/2ℏ(53)g=g''\frac{m_e^3c^{9/2}}{\sqrt{\hbar}} \tag*{(53)}

We choose kmk_m at the edge of a band with cos⁡(kmξ)=1\cos(k_m\xi)=1. Since kik_i can be chosen arbitrarily we can pick any value of q1q_1. For our calculation we set q1=600q_1=600. This value is chosen since it will lead to dominant contribution from EmE_m values a little above the potential barrier V0V_0. This region dominates the reaction rate. The value of q2q_2 is now given by

q2=Ei−Efc−q1(54)q_2=\frac{E_i-E_f}{c}-q_1 \tag*{(54)}

This also fixes kfk_f by the relationship kf=−q1−q2+ki−G−G′k_f=-q_1-q_2+k_i-G-G'. We note that for each EmE_m there are two wave functions, ψ+\psi^+ and ψ−\psi^- which are complex conjugates of one another. Due to the delta function δG,−km−q1+ki\delta_{G,-k_m-q_1+k_i} only one of these would contribute for a fixed q1q_1. The values of the real part of II as a function of the upper cutoff on the energy EmE_m of the intermediate state is shown in Figure 2. Here we show the contribution from dominant region −a−2b<x<−a-a-2b<x<-a. The contribution from other regions is highly suppressed. We find that II settles to a value of −6.89×10−11-6.89 \times10^{-11} in atomic units. The imaginary part behaves in a similar manner and settles to a value approximately 2.6×10−112.6 \times10^{-11}. We also show the dependence of rate as a function of the emitted ϕ\phi particle momentum q1q_1 in Figure 3 setting the coupling parameter g′′=10−5g''=10^{-5}. The total rate after integrating over q1q_1 is found to approximately 2×10−132 \times10^{-13} in atomic units.

Figure 2. The real part of Iˉ\bar{I} (see Eq. (52)) as a function of the upper cutoff on the intermediate state energy EmE_m in atomic units. We see that the amplitude tends to a significant value as the cutoff goes to infinity.

Figure 3. The differential reaction rate dP/dtdq1dP/dtdq_{1} as a function of the momentum of the emitted particle q1q_{1}, both quantities are in atomic units. Here the value of dimensionless parameter gg has been taken to be 10−510^{-5}.

It is clear from Figure 2 that the second order process is not suppressed. This is because the final value of Iˉ\bar{I} is comparable to the peak value obtained for some intermediate value of cutoff Energy. The result would be negligible only if contributions from different energy eigenvalues show delicate cancellation, which does not happen in this case. The behaviour of the amplitude is very similar to what is seen in our earlier paper for a different 3 dimensional model [20]. Here also a cancellation was seen which results in a small but significant contribution.

We may compare the rate with that obtained from the first order process. The rate in this case in given by

dPdt=g′22ℏcω∣∫dxψf∗e−iqxψi∣2(55)\frac{dP}{dt} = \frac{g^{\prime2}}{2\hbar c\omega}\left|\int dx\psi_f^*e^{-iqx}\psi_i\right|^2 \tag*{(55)}

where ω\omega and qq are the frequency and momentum of the emitted ϕ\phi particles. The comparison depends on the choice of the dimensionless parameter g′′g^{\prime\prime}. For g′′=10−5g^{\prime\prime}=10^{-5}, chosen earlier, we find that the rate is of order 10−7610^{-76} in atomic units. Here we have chosen the same initial and final states as for the second order process. The dominant suppression arises from the exceedingly small factor corresponding to the barrier penetration probability. It is clear that the first order process is highly suppressed compared to the second order process, unless we choose exceedingly small values for the dimensionless coupling parameter g′′g^{\prime\prime}.

Conclusions

In this paper we have studied the rate of nuclear fusion at low energies using a 1+11+1 dimensional lattice model. The model is a generalization of the standard Kronig-Penney model and incorporates the deep nuclear potential. We consider fusion of a particle, which is analogous to proton, with a lattice site. The fusion rate of the standard first order process is very strongly suppressed due to tunnelling barrier. We consider the rate at second order in perturbation theory. For this case the rate can be large in some special circumstances, as described in earlier papers [18, 19, 20]. In [20] we had considered a three dimensional toy model assuming a step potential. Here we found that the rate was very small if the eigenfunctions show a continuous energy spectrum, which is achieved if the potential vanishes at large distances. However it was argued that in a medium the potential need not be zero at large distances and spectrum is not necessarily continuous. We modelled this phenomenon by assuming an infinite potential barrier beyond a certain distance. This lead to a significant rate for the fusion process [20]. Here we are interested in determining whether a lattice model leads to an observable rate.

The fusion process takes place by emission of two particles, one at each interaction. We have tried a few different models for the interaction. The simplest model in which the interaction Hamiltonian is taken to be of the form given in Eq. (6) with the operator O\mathcal{O} equal to unity we find that the rate is very small. The actual value is difficult to obtain due to a very delicate cancellation among amplitudes due to different intermediate states and we do not pursue this in detail in the present paper. A modified interaction with the operator O\mathcal{O} having a momentum dependence given in Eq. (50) is found to give a substantial rate. Although many interactions depend on momentum, it is not known whether such an interaction can arise in a real physical situation. We have suggested that it might arise as an effective interaction or a form factor if in the initial stage the process happens through the Hydrogen atom rather than a proton. However, we have not examined this possibility in detail. Nevertheless, our result demonstrates that quantum mechanics does not prohibit a relatively large fusion rate even at very low energies and high potential barrier. To be precise, there does not seem to be any general rule that the amplitude necessarily has to be small exceedingly small even for the second order process. Our results show that nuclear fusion may take place in a crystalline lattice but requires special conditions. A detailed study is required to determine whether those may be met in a real physical situation. An alternative possibility is presented by a disordered system. In such a system, all states up to a certain energy level are localized. Hence, in a given localized region, relevant for the fusion process, dominant contributions would be obtained only from discretized energy states. Hence we expect that fusion rate be a significant in this case, in analogy with results obtained in [20]. A detailed calculation in a disordered physical system is postponed to future research.

Acknowledgements: We are very grateful to Amit Agarwal for useful discussions.

References

  1. [1]S. B. Krivit, Development of Low-Energy Nuclear Reaction Research, ch. 41, pp. 479–496. John Wiley & Sons, Ltd, 2011.DOI
  2. [2]L. I. Urutskoev, Low-Energy Nuclear Reactions: A Three-Stage Historical Perspective, ch. 42, pp. 497–501. John Wiley & Sons, Ltd, 2011.DOI
  3. [3]M. Srinivasan, G. Miley, and E. Storms, Low-Energy Nuclear Reactions: Transmutations, ch. 43, pp. 503–539. John Wiley & Sons, Ltd, 2011.DOI
  4. [4]E. Storms, “Introduction to the main experimental findings of the lenr field,” Current Science, vol. 108, pp. 535–539, 02 2015.DOI
  5. [5]M. C. McKubre, “Cold fusion: comments on the state of scientific proof,” Current Science, pp. 495–498, 2015.
  6. [6]J.-P. Biberian, “Biological transmutations,” Current Science, vol. 108, pp. 633–635, 02 2015.
  7. [7]M. Srinivasan and K. Rajeev, “Chapter 13 - transmutations and isotopic shifts in lenr experiments,” in Cold Fusion (J.-P. Biberian, ed.), pp. 233 – 262, Elsevier, 2020.
  8. [8]T. Mizuno and J. Rothwell, “Excess heat from palladium deposited on nickel,” Journal of Condensed Matter Nuclear Science, vol. 29, pp. 21–33, 2019.DOI
  9. [9]S. B. Krivit and J. Marwan, “A new look at low-energy nuclear reaction research,” Journal of Environmental Monitoring, vol. 11, no. 10, pp. 1731–1746, 2009.DOI
  10. [10]K. P. Sinha, “Model of low energy nuclear reactions in a solid matrix with defects,” Current Science, vol. 108, pp. 516–518, 02 2015.DOI
  11. [11]C. L. Liang, Z. M. Dong, and X. Z. Li, “Selective resonant tunnelling–turning hydrogen-storage material into energetic material,” Current Science, vol. 108, pp. 519–523, 02 2015.DOI
  12. [12]P. Hagelstein, “Deuterium evolution reaction model and the fleischmann–pons experiment,” Journal of Condensed Matter Nuclear Science, vol. 16, pp. 46–63, 02 2015.DOI
  13. [13]F. Celani, A. Tommaso, and G. Vassallo, “The zitterbewegung interpretation of quantum mechanics as theoretical framework for ultra-dense deuterium and low energy nuclear reactions,” Journal of Condensed Matter Nuclear Science, vol. 24, pp. 32–41, 02 2017.DOI
  14. [14]P. Kálmán and T. Keszthelyi, “On low-energy nuclear reactions,” arXiv preprint arXiv:1907.05211, 2019.arxiv.org/abs/1907.05211
  15. [15]C. Spitaleri, C. Bertulani, L. Fortunato, and A. Vitturi, “The electron screening puzzle and nuclear clustering,” Physics Letters B, vol. 755, pp. 275–278, 2016.
  16. [16]J.-L. Paillet and A. Meulenberg, “On highly relativistic deep electrons,” Journal of Condensed Matter Nuclear Science, vol. 29, pp. 472–492, 02 2019.DOI
  17. [17]P. Kálmán and T. Keszthelyi, “Forbidden nuclear reactions,” Phys. Rev. C, vol. 99, p. 054620, May 2019.DOI
  18. [18]P. Jain, A. Kumar, R. Pala, and K. P. Rajeev, “Photon induced low-energy nuclear reactions,” Pramana, vol. 96, no. 96, 2022.DOI
  19. [19]P. Jain, A. Kumar, K. Ramkumar, R. Pala, and K. P. Rajeev, “Low energy nuclear fusion with two photon emission,” JCMNS, vol. 35, p. 1, 2021.DOI
  20. [20]K. Ramkumar, H. Kumar, and P. Jain, “A toy model for low energy nuclear fusion,” Pramana, vol. 97, no. 109, 2023.DOI
  21. [21]E. Merzbacher, Quantum Mechanics. Wiley, 1998.DOI
  22. [22]J. Sakurai, Advanced Quantum Mechanics. Always learning, Pearson Education, Incorporated, 1967.
  23. [23]D. D. Clayton, Principles of stellar evolution and nucleosynthesis. The University of Chicago Press, Chicago, 1968.
  24. [24]N. W. Ashcroft and N. D. Mermin, Solid State Physics. Holt-Saunders, 1976.
  25. [25]E. Abrahams, 50 Years of Anderson Localization. WORLD SCIENTIFIC, 2010.
  26. [26]P. A. Lee and T. V. Ramakrishnan, “Disordered electronic systems,” Rev. Mod. Phys., vol. 57, pp. 287–337, Apr 1985.
  27. [27]J.-P. Biberian, “Anomalous isotopic distribution of silver in a palladium cathode,” Journal of Condensed Matter Nuclear Science, vol. 29, pp. 211–218, 2019.DOI

Paper details

vlnc:W_xyx6ynz8
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents