betaFeedback

WorksPhysics and Astronomy

Nonlocal energy transfer mechanism in three-dimensional quantum turbulence

Abstract

We investigate the kinetic energy cascade in zero-temperature quantum turbulence. Using simple theoretical arguments and unprecedented numerical simulations, we unveil an universal mechanism transferring energy directly from large to very small scales, thus bypassing the Kolmogorov-like local energy cascade and resulting in nonclassical energy spectra. This mechanism rests both on the vast separation of scales typical of superfluid helium-4 flows and on the alignment between quantum vortices and large-scale velocity gradients, in direct analogy with vortex stretching in classical flows.

Introduction

Near absolute zero, helium-4 exists in a superfluid state known as He II which has remarkable hydrodynamical properties. In addition to being inviscid, all vorticity of zero-temperature He II is concentrated in atomically thin vortices with quantized circulation κ≈0.997×10−7 m2/s\kappa\approx0.997 \times10^{-7}\,\mathrm{m}^2/\mathrm{s} known as quantum vortices [1]. Despite these major differences with classical fluids, quantum turbulence (QT) in He II [2] is widely thought to present significant similarities with classical turbulence, especially at large scales where the collective behavior of quantum vortices becomes dominant [3, 4, 5, 6, 7, 8]. In typical He II experiments [3, 6, 9, 10, 11, 12, 13], QT is characterized by a very wide range of scales, ranging from the microscopic vortex core radius a0∼10−10 ma_0 \sim10^{-10}\,\mathrm{m} to the integral scale L0∼10−2L_0 \sim10^{-2}–10−1 m10^{-1}\,\mathrm{m} associated to the largest turbulent structures. Between the two, the typical distance between quantum vortices, ℓ∼10−5\ell\sim10^{-5}–10−4 m10^{-4}\,\mathrm{m}, sets the transition between the quantum scales, where the discrete nature of quantum vortices is apparent, and the inertial scales, where the collective motion and the partial polarization of quantum vortices [14, 15, 16, 17] can lead to quasi-classical behavior.

Up to now, the inertial and quantum ranges have been mostly regarded as two disconnected regions of scale space with their own very different dynamics [14, 18]. In this Letter we show that this is not truly the case, since, due to the vast separation between ℓ\ell and a0a_0 in He II turbulence, the quantum range acts as an energy sink which directly receives kinetic energy from the inertial scales, thus bypassing the locality of the energy cascade on which rests Kolmogorov’s classical turbulence phenomenology [19]. A major consequence is that, even at scales significantly larger than ℓ\ell, the kinetic energy spectrum significantly departs from the expected quasi-classical prediction E(k)∼ε2/3k−5/3E(k) \sim\varepsilon^{2/3}k^{-5/3}, where kk is the wavenumber and ε\varepsilon the (constant) energy flux. Similar spectral shortcut mechanisms have been identified in classical turbulent flows interacting with plant canopies [20] and rigid slender fibers [21, 22].

Using simple theoretical arguments, we show that in QT such nonlocal energy transfers rest on the preferential alignment of quantum vortices with large-scale straining motions induced by the vortex tangle itself. Such alignment tends to increase the quantum vortex line density L≡ℓ−2\mathcal{L} \equiv\ell^{-2} and thus the energy content Esmall∼κ2Lln⁡(ℓ/a0)E_{\mathrm{small}} \sim\kappa^2\mathcal{L}\ln(\ell/a_0) at quantum scales, at the expense of large-scale energy. Note that, unlike energy, vortex length is not a conserved quantity and therefore it is allowed to increase. This vortex line production mechanism can be directly related to the ubiquitous vortex stretching phenomenon in classical turbulent flows [23, 24, 25], which arises from similar alignments. The identified mechanism appears to be a generic property of QT as well, as it can be deduced from past numerical [26, 27, 28, 29] and experimental [9, 30, 31, 32] observations of growth of vortex length at early times in decaying settings. At later times, as a disordered vortex tangle develops, such vortex growth is likely to be arrested by the loss of vortex length associated to vortex reconnections [33, 34]. In particular, in steady-state regimes, the identified line production mechanism can be seen as a necessary condition for preserving the total vortex line density over time.

In the following, we describe quantum vortex dynamics at scales much larger than a0a_0 using the vortex filament model (VFM) [35, 36]. In the VFM, quantum vortices are described as closed or infinite lines in three-dimensional (3D) space, parametrized by the vortex positions s(ξ)\mathbf{s}(\xi) where ξ\xi is the arc length. A tangle of quantum vortices induces a velocity field through space given by the Biot–Savart law,

vs(x,t)=κ4π∮C(t)(s−x)×ds∣s−x∣3,(1)\mathbf{v}_{s}(\mathbf{x},t)=\frac{\kappa}{4\pi}\oint_{C(t)}\frac{(\mathbf{s}-\mathbf{x})\times\mathrm{d}\mathbf{s}}{|\mathbf{s}-\mathbf{x}|^3}, \tag*{(1)}

where C(t)C(t) denotes the instantaneous vortex geometry. When evaluated on a vortex location s0∈C(t)\mathbf{s}_0 \in C(t), integration must be omitted over a small distance ∣s−s0∣<a0eΔ/2|\mathbf{s}-\mathbf{s}_0|<a_0 e^{\Delta/2} (we set Δ=1/2\Delta=1/2 [36]) to avoid the singularity in Eq. (1) and to account for the finite vortex core size a0a_0 [37]. A well known result of this desingularization procedure is the localized induction approximation (LIA) [38],

vsε(s)=κ4πs′×s′′ln⁡(2εeΔa0),a0≪ε≪1∣s′′∣,(2)\mathbf{v}_{s}^{\varepsilon}(\mathbf{s})=\frac{\kappa}{4\pi}\mathbf{s}'\times\mathbf{s}''\ln\left(\frac{2\varepsilon}{e^{\Delta}a_0}\right),\qquad a_0\ll\varepsilon\ll\frac{1}{|\mathbf{s}''|}, \tag*{(2)}

which expresses the velocity induced by a small local segment ξ∈[ξ0−ε,ξ0+ε]\xi\in[\xi_0-\varepsilon,\xi_0+\varepsilon] on a vortex point s(ξ0)\mathbf{s}(\xi_0), and primes denote derivatives with respect to ξ\xi. In particular, s′\mathbf{s}' and s′′\mathbf{s}'' are respectively the local unit tangent and curvature vectors.

Spectral energy fluxes

In the following we consider a statistically isotropic and spatially homogeneous turbulent quantum vortex tangle in a cubic periodic domain of volume V=Lbox3V=L_{\mathrm{box}}^3. The kinetic energy per unit mass associated to the velocity field (1) is E=⟨∣vs∣2⟩/2=κ28πV∮C∮Cds1⋅ds2∣s1−s2∣E=\langle|\mathbf{v}_s|^2\rangle/2=\frac{\kappa^2}{8\pi V}\oint_C\oint_C\frac{\mathrm{d}\mathbf{s}_1\cdot\mathrm{d}\mathbf{s}_2}{|\mathbf{s}_1-\mathbf{s}_2|} [37], where $\langle\cdot\rangle` denotes a spatial average. The Biot–Savart law (1) is then equivalent to [39]

δEδs=κVs′×vs(s),(3)\frac{\delta E}{\delta\mathbf{s}}=\frac{\kappa}{V}\mathbf{s}'\times\mathbf{v}_{s}(\mathbf{s}), \tag*{(3)}

which is interpreted as a Magnus force (per unit length and mass) acting on each vortex location s\mathbf{s}. From the definition of the functional derivative δEδs\frac{\delta E}{\delta\mathbf{s}}, the energy variation rate due to an arbitrary vortex velocity vL(s)\mathbf{v}_{L}(\mathbf{s}) is dEdt∣vL=∮CδEδs⋅vL dξ\left.\frac{\mathrm{d}E}{\mathrm{d}t}\right|_{\mathbf{v}_{L}}=\oint_{C}\frac{\delta E}{\delta\mathbf{s}}\cdot\mathbf{v}_{L}\,\mathrm{d}\xi. By decomposing vs(x)\mathbf{v}_{s}(\mathbf{x}) into Fourier modes, one can readily generalize Eq. (3) to δEkδs=κVs′×vs<k(s)\frac{\delta E_{k}}{\delta\mathbf{s}}=\frac{\kappa}{V}\mathbf{s}'\times\mathbf{v}_{s}^{<k}(\mathbf{s}), where vs<k(x)=∑∣k∣≤kv^s(k)eik⋅x\mathbf{v}_{s}^{<k}(\mathbf{x})=\sum_{|\mathbf{k}|\leq k}\hat{\mathbf{v}}_{s}(\mathbf{k})e^{i\mathbf{k}\cdot\mathbf{x}} is a low-pass filtered velocity field and Ek=⟨∣vs<k∣2⟩/2E_{k}=\langle|\mathbf{v}_{s}^{<k}|^{2}\rangle/2 is the cumulative energy up to wavenumber kk [19]. Ultimately, this allows us to define the energy flux through wavenumber kk as the energy removed from the large scales by the total induced velocity vs\mathbf{v}_{s} (1),

Πk=−dEkdt∣vs=−∮CδEkδs⋅vs dξ=κV∮C(s′×vs)⋅vs<k dξ.(4)\Pi_{k}=-\left.\frac{\mathrm{d}E_{k}}{\mathrm{d}t}\right|_{\mathbf{v}_{s}}=-\oint_{C}\frac{\delta E_{k}}{\delta\mathbf{s}}\cdot\mathbf{v}_{s}\,\mathrm{d}\xi=\frac{\kappa}{V}\oint_{C}(\mathbf{s}'\times\mathbf{v}_{s})\cdot\mathbf{v}_{s}^{<k}\,\mathrm{d}\xi. \tag*{(4)}

Since the superfluid vorticity can be formally expressed as ωs(x)=κ∮Cδ(x−s) ds\boldsymbol{\omega}_{s}(\mathbf{x})=\kappa\oint_{C}\delta(\mathbf{x}-\mathbf{s})\,\mathrm{d}\mathbf{s}, this is in fact equal to the usual definition Πk=⟨vs<k⋅(ωs×vs)⟩=⟨vs<k⋅(vs⋅∇vs)⟩\Pi_{k}=\langle\mathbf{v}_{s}^{<k}\cdot(\boldsymbol{\omega}_{s}\times\mathbf{v}_{s})\rangle=\langle\mathbf{v}_{s}^{<k}\cdot(\mathbf{v}_{s}\cdot\nabla\mathbf{v}_{s})\rangle in incompressible Navier–Stokes turbulence [19]. Note that, as in classical fluids, the flux vanishes as k→∞k\to\infty.

Nonlocal energy transfer—In 3D classical isotropic turbulence, for kk in the inertial range, the energy flux Πk\Pi_{k} is constant and equal to the energy cascade rate ε\varepsilon. Moreover, the energy cascade is local in the sense that vs<k\mathbf{v}_{s}^{<k} mostly transfers energy to wavenumbers k′≲2kk'\lesssim2k [40, 41, 42, 43]. In the following we show that this is generally not the case in zero-temperature QT, as energy is directly transferred from inertial scales (k≪kℓ≡2π/ℓk\ll k_{\ell}\equiv2\pi/\ell) to quantum scales k′≫kℓk'\gg k_{\ell}. Here the mean inter-vortex distance is ℓ=L−1/2\ell=\mathcal{L}^{-1/2} where L=1V∮Cdξ\mathcal{L}=\frac{1}{V}\oint_{C}\mathrm{d}\xi is the vortex line density. We start by generalizing Eq. (4) to write the nonlocal energy transfer from large (<k<k) to small (>q>q) scales as Πk>q=κV∮C(s′×vs>q)⋅vs<k dξ\Pi_{k}^{>q}=\frac{\kappa}{V}\oint_{C}(\mathbf{s}'\times\mathbf{v}_{s}^{>q})\cdot\mathbf{v}_{s}^{<k}\,\mathrm{d}\xi, where vs>q≡vs−vs<q\mathbf{v}_{s}^{>q}\equiv\mathbf{v}_{s}-\mathbf{v}_{s}^{<q} is a high-pass filtered (i.e. small-scale) velocity. Since nonlocal vortex interactions are negligible at scales below ℓ\ell, one can safely assume that vs>kℓ\mathbf{v}_{s}^{>k_{\ell}} is well approximated by the LIA (2) with ϵ∼ℓ\epsilon\sim\ell. This ultimately leads to

Πk>kℓ≈k24πln⁡(2ℓeΔa0)Sk,Sk≡−1V∮Cs′′⋅vs<k dξ,(5)\Pi_{k}^{>k_{\ell}}\approx\frac{k^{2}}{4\pi}\ln\left(\frac{2\ell}{e^{\Delta}a_{0}}\right)S_{k},\qquad S_{k}\equiv-\frac{1}{V}\oint_{C}\mathbf{s}''\cdot\mathbf{v}_{s}^{<k}\,\mathrm{d}\xi, \tag*{(5)}

where we have used the geometrical relation s′×(s′×s′′)=−s′′\mathbf{s}'\times(\mathbf{s}'\times\mathbf{s}'')=-\mathbf{s}''. Equation (5) states that nonlocal energy transfers from inertial to quantum scales are enabled by a possible antialignment between the large-scale velocity vs<k\mathbf{v}_{s}^{<k} and the local vortex curvature s′′\mathbf{s}'', and are amplified by the scale separation between ℓ\ell and a0a_{0}. Importantly, one can interpret SkS_{k} as the production of vortex line density L\mathcal{L} by vs<k\mathbf{v}_{s}^{<k}. Indeed, using integration by parts,

Sk=1V∮Cs′⋅∂vs<k∂ξ dξ=1V∮Cs′⋅∇vs<k⋅s′ dξ.(6)S_{k}=\frac{1}{V}\oint_{C}\mathbf{s}'\cdot\frac{\partial\mathbf{v}_{s}^{<k}}{\partial\xi}\,\mathrm{d}\xi=\frac{1}{V}\oint_{C}\mathbf{s}'\cdot\nabla\mathbf{v}_{s}^{<k}\cdot\mathbf{s}'\,\mathrm{d}\xi. \tag*{(6)}

Here s′⋅∂vs<k∂ξ\mathbf{s}'\cdot\frac{\partial\mathbf{v}_{s}^{<k}}{\partial\xi} expresses the local vortex line growth rate (per unit length) due to vs<k\mathbf{v}_{s}^{<k}. Remarkably, the last integral is strongly reminiscent of the enstrophy (Ω≡⟨∣ω∣2⟩/2\Omega\equiv\langle|\boldsymbol{\omega}|^{2}\rangle/2) production term from the Navier–Stokes equations, which writes PΩ=⟨ω⋅∇v⋅ω⟩=⟨ωiSijωj⟩P_{\Omega}=\langle\boldsymbol{\omega}\cdot\nabla\mathbf{v}\cdot\boldsymbol{\omega}\rangle=\langle\omega_{i}S_{ij}\omega_{j}\rangle where Sij=(∂ivj+∂jvi)/2S_{ij}=(\partial_{i}v_{j}+\partial_{j}v_{i})/2 is the strain-rate tensor [23]. This term is generally positive in 3D classical turbulence due to the vortex stretching mechanism [23, 25, 44], intimately linked to energy cascade dynamics. In the case of Eq. (6), SkS_{k} corresponds to vortex stretching by the nonlocal (or large-scale) strain rate tensor, whose most extensional eigenvector has been shown to preferentially align with vorticity in classical turbulence [45, 46]. Therefore, it is natural to expect SkS_{k} to be positive for k≪kℓk\ll k_{\ell} if the large scales of QT have any similarities with classical fluid turbulence. Paradoxically, such similarities would result in positive nonlocal energy transfers Πk>kℓ\Pi_{k}^{>k_{\ell}} [Eq. (5)], ultimately leading to nonclassical scaling laws at scales larger than ℓ\ell. This scenario is confirmed using numerical simulations below.

Numerical simulations—We perform VFM numerical simulations of statistically stationary zero-temperature QT at high vortex densities with an unprecedented separation of scales between L0L_{0} and ℓ\ell. To circumvent the O(N2)O(N^{2}) cost of directly computing far-ranged interactions between NN vortex points, we adopt a novel fast Fourier transform (FFT)-based approach [47] inspired by Ewald summation methods [48, 49], commonly used in molecular dynamics to evaluate electrostatic interactions in periodic systems. Our method, summarized in the End Matter, enables the efficient evaluation of Eq. (1) in O(Nlog⁡N)O(N\log N) time. Unlike other fast approaches [50, 51], this method naturally accounts for the infinity of periodic vortex images (instead of effectively introducing a cut-off distance rcut∼Lboxr_{\mathrm{cut}}\sim L_{\mathrm{box}} [36]), while also providing highly accurate Fourier-space velocity fields at no extra cost, facilitating the evaluation of energy spectra and fluxes (4). The methods are implemented in the open-source GPU-accelerated VortexPasta.jl solver [52].

To eliminate limitations associated to usual time-decaying configurations – such as statistical convergence issues and the influence of initial conditions – we devise a novel energy injection scheme to keep the system in a statistically stationary state. Our approach mirrors standard practice in numerical classical turbulence studies. Concretely, we introduce an external forcing velocity vf\mathbf{v}_{f} such that vortices move according to

dsdt=vs(s)+vf(s).(7)\frac{\mathrm{d}\mathbf{s}}{\mathrm{d}t}=\mathbf{v}_{s}(\mathbf{s})+\mathbf{v}_{f}(\mathbf{s}). \tag*{(7)}

As detailed in the End Matter [Eq. (9)], vf\mathbf{v}_{f} can be defined such that it injects kinetic energy at the largest scales of the system (wavenumbers k<kfk<k_{f}) at rate εinj\varepsilon_{\mathrm{inj}}, while leaving smaller scales unmodified (as verified in Figure 2, open circles). We also ensure that vf⋅s′′=0\mathbf{v}_{f}\cdot\mathbf{s}''=0 so that vf\mathbf{v}_{f} does not directly modify the total vortex length (and thus the energy at quantum scales). Note that determining vf\mathbf{v}_{f} requires knowing the Fourier-filtered velocity v<kf\mathbf{v}^{<k_{f}} at all times, which is greatly facilitated by our FFT-based numerical method.

Figure 2. Normalized spectral energy fluxes (run εinj=3200\varepsilon_{\mathrm{inj}} = 3200, a0=10−7a_0 = 10^{-7}). Accumulated energy injection rate Fk\mathcal{F}_k (open circles, Eq. (8)), total energy flux Πk\Pi_k (filled circles, Eq. (4)), local energy flux ΠkΠ<2k\Pi_k^{\Pi<2k} (crosses), nonlocal energy flux ΠkΠ>kℓ\Pi_k^{\Pi>k_\ell} (squares), and estimated nonlocal flux based on vortex line production SkS_k (dashed line, Eq. (5)). The energy injection scales (k<kfk < k_{\mathrm{f}}) are highlighted in light red.

Vortex lines are discretized as quintic splines, ensuring an accurate estimation and interpolation of geometrical quantities such as ss, s′s' and s′′s''. All line integrals are evaluated using 3-point Gauss–Legendre quadratures between neighboring vortex points, and the relative accuracy of Biot–Savart computations is kept below 10−610^{-6}[47]. As in previous works [27, 53, 54, 55], the distance Δξ\Delta\xi between vortex points is kept within [δres,2δres)[\delta_{\mathrm{res}}, 2\delta_{\mathrm{res}}) using adaptive line refinement. The resolution is set to δres≈ℓ/6\delta_{\mathrm{res}} \approx\ell/6, adjusted according to the mean inter-vortex distance ℓ\ell in each simulation. Following standard practice, filaments are reconnected when their distance is below δres\delta_{\mathrm{res}}, but only if this leads to loss of vortex length (effectively dissipating energy at quantum scales) and if the reconnecting segments are not locally parallel [56]. Equation (7) is advanced in time using an adaptive 4th-order Runge–Kutta scheme.

We set the nondimensional vortex circulation to κ=1\kappa= 1, the periodic domain size to Lbox=2πL_{\mathrm{box}} = 2\pi and the vortex core radius to a0=10−7a_0 = 10^{-7}. Since a0≈10−10 ma_0 \approx10^{-10}\,\mathrm{m} in He II, this gives Lbox≈2π mmL_{\mathrm{box}} \approx2\pi\,\mathrm{mm} in physical units, which is representative of a typical measurement section in current experiments. We also perform simulations with a0=10−4a_0 = 10^{-4} to illustrate the effect of reducing the ℓ/a0\ell/a_0 ratio. Forcing is applied up to wavenumber kf=2.5k_{\mathrm{f}} = 2.5. A total of 7 runs are performed with different values of a0a_0 and of the energy injection rate εinj\varepsilon_{\mathrm{inj}} as detailed in the End Matter (Table I). An instantaneous visualization of a typical run is shown in Figure 1, highlighting the formation of highly polarized vortex bundles, a key feature of quasi-classical QT [14, 15, 17, 54, 57].

Visualization of QT simulation in a statistically steady state

Figure 1. Visualization of QT simulation in a statistically steady state (run εinj=3200\varepsilon_{\mathrm{inj}} = 3200, a0=10−7a_0 = 10^{-7}). Quantum vortices are colored by the Gaussian-filtered superfluid vorticity at scale ℓ\ell, highlighting highly polarized vortex bundles (yellow to red colors).

Numerical results—We start by plotting in Figure 2 the spectral

energy fluxes obtained from one of our runs. The total energy flux from large to small scales (Πk\Pi_k, filled circles) displays a plateau at scales larger than ℓ\ell with a value close to the energy injection rate εinj\varepsilon_{\mathrm{inj}}. At first sight, this is consistent with the energy cascade picture in classical turbulence [58]. At smaller scales, the flux starts decreasing very slowly (logarithmically) with kk, and is expected to reach zero near the microscopic scale ka∼1/a0k_a \sim1/a_0 (not visible in the figure). This drastically differs from the Navier–Stokes case where the energy flux rapidly decays to zero when the viscous scales are reached.

The flux Πk\Pi_k provides no indication on the locality of the energy cascade. With this aim, we consider the local flux ΠkΠ<2k\Pi_k^{\Pi<2k} to neighboring scales k′∈[k,2k]k' \in[k, 2k] (crosses in Figure 2), defined by replacing vsv_s with vs<2kv_s^{<2k} in Eq. (4). The local flux clearly decreases beyond the injection scales, thus breaking the hypothesis of a constant energy flux associated to Kolmogorov’s (K41) scaling laws. This is also consistent with the behavior of the nonlocal flux ΠkΠ>kℓ\Pi_k^{\Pi>k_\ell} (squares), which is positive and increases throughout the inertial range. The nonlocal flux is in fairly good agreement with our estimation (5) based on the vortex length production term SkS_k (dashed line), providing strong support to the proposed picture of a nonlocal energy transfer driven by the alignment between quantum vortices and large-scale velocity gradients. Our results confirm that a preferential alignment exists at all scales (up to the largest simulated scale) leading to a continuous depletion of the local cascade. More profoundly, they also suggest that this is the main mechanism enabling a transfer of energy from large to quantum scales, before it being converted to acoustic energy (phonons) at microscopic scales via vortex reconnections [33, 59, 60] and the Kelvin wave cascade [61, 62, 63, 64, 65, 66]. The existence of nonlocal energy transfers is expected to lead to the breakdown of K41 scaling laws in QT. This is confirmed in Figure 3, which shows the kinetic energy spectra obtained from all numerical runs. The normalized spectra collapse at quantum scales, meaning that they all follow the analytical prediction E(k)=(κ2L/4π)k−1E(k) = (\kappa^2\mathcal{L}/4\pi)k^{-1} associated to an isolated straight vortex for ka0≪1ka_0 \ll1 [27, 47]. Note that a production of vortex lines directly increases the energy content throughout the quantum scales, shifting the spectrum upwards. At inertial scales (kℓ≪1k\ell\ll1) the spectra become much steeper, indicating the emergence of collective dynamics and energy-containing vortex bundles generally associated to a partial polarization of the tangle. While it is possible to fit power law scalings within this region, these are substantially steeper than Kolmogorov’s k−5/3k^{-5/3} law, suggesting a higher degree of polarization than that traditionally associated to quasi-classical QT [14, 17]. As expected from Eq. (5), the deviations are further amplified as the ratio ℓ/a0\ell/a_0 is increased from ∼103\sim10^3 (blue curves) to ∼106\sim10^6 (red curves).

Figure 3. Kinetic energy spectrum from different runs. Two sets of curves are shown with ℓ/a0∼106\ell/a_0 \sim10^6 (red) and 10310^3 (blue), see the End Matter (Table I) for precise values. Different markers correspond to different energy injection rates εinj\varepsilon_{\mathrm{inj}}. The spectra are normalized according the analytical prediction E(k)=(κ2/4πℓ2)k−1E(k) = (\kappa^2/4\pi\ell^2)k^{-1} in the quantum scales. Inset: Local energy flux Πk2κ\Pi_k^{2\kappa} for the different runs.

Despite these major deviations, one can attempt to reconcile the observed energy spectra with K41 theory by considering a nonconstant local energy flux which, for simplicity, behaves as εlocal(k)∼k−α\varepsilon_{\mathrm{local}}(k) \sim k^{-\alpha}. Then, dimensional analysis leads to E(k)∼εlocal(k)2/3k−5/3∼k−(5+2α)/3E(k) \sim\varepsilon_{\mathrm{local}}(k)^{2/3}k^{-5/3} \sim k^{-(5+2\alpha)/3}. Taking Πk2κ\Pi_k^{2\kappa} (inset of Figure 3) as a proxy for εlocal(k)\varepsilon_{\mathrm{local}}(k), we find α≈1.25\alpha\approx1.25 and 0.80.8 for the two considered values of ℓ/a0\ell/a_0 over about half a decade in kk. These values exactly correspond to the E(k)∼k−2.5E(k) \sim k^{-2.5} and k−2.2k^{-2.2} scalings fitted in the main panel of Figure 3. This gives strong support to the idea that, similarly to classical turbulence, inertial range scalings are independent of the specific energy injection mechanisms and of quantum scale dynamics, despite the lack of a constant local energy cascade.

**Discussion—**We have identified a novel mechanism transferring energy from inertial to quantum scales in zero-temperature QT. Besides significantly altering QT scaling laws at scales above ℓ\ell, this mechanism seems to predominantly explain how energy arrives to quantum scales before being ultimately dissipated by phonon emission. The crucial realization enabling these transfers is that, while individual quantum vortices cannot be stretched in the classical sense [2, 67], nothing disallows them from being elongated by large-scale velocity gradients, as vortex length is not a conserved quantity. Importantly, this mechanism transfers energy from large scales to all quantum scales at once, which highlights the relevance of the scale separation between ℓ\ell and a0a_0 in this process.

One major question is whether nonlocal transfers vanish at larger scales as the extent of the inertial range (or the Reynolds number) is increased. Our simulations suggest that this is indeed the case, as the local energy flux (inset of Figure 3) appears to saturate to a constant value of the order of the energy injection rate εinj\varepsilon_{\mathrm{inj}} – as expected based on classical turbulence phenomenology. Physically, this may be explained by an increasing decorrelation between the orientations of individual quantum vortices and of the resulting coarse-grained vorticity at large scales. The alignment between velocity gradients and vorticity at different scales will be the subject of a future investigation.

Being based on an alignment between large-scale velocity gradients and quantum vortices, we expect the identified mechanism to stay valid in finite-temperature QT driven by classical means (e.g. mechanical forcing), where vortices coexist and interact with a viscous normal fluid via a mutual friction force. These two components are expected to be tightly coupled at large scales, roughly exhibiting one-fluid behavior. The identified nonlocal transfers can be expected to promote their decoupling and thus the mutual friction between the two components, potentially leading to temperature-dependent energy spectra as the quantum scale is approached.

While our work is motivated by isotropic turbulence in He II, the identified mechanism may also be relevant in rotating settings or in other superfluids where the ratio ℓ/a0\ell/a_0 also spans several orders of magnitude in realistic settings. This can be the case in 3{}^{3}He-B, where a0∼10−8 ma_0 \sim10^{-8}\,\mathrm{m} while values ℓ∼10−4 m\ell\sim10^{-4}\,\mathrm{m} can be experimentally achieved [68]. The separation of scales can be even more dramatic in the core of neutron stars [69], where a0∼10−14 ma_0 \sim10^{-14}\,\mathrm{m} [70, 71] while the inter-vortex distance associated to the star rotation alone can be estimated as ℓ∼10−5 m\ell\sim10^{-5}\,\mathrm{m} [72].

Acknowledgments

This work was supported by the French ANR through the project ANR-23-CE30-0024-04 (QuantumVIW). This project was granted access to computational resources of CINES and IDRIS under the allocation 2025-A0192A00611 made by GENCI. Computations were also performed using the GRICAD infrastructure which is supported by the Grenoble research communities.

References

End Matter

Definition of forcing velocity—In analogy with Eq. (4), one can express the cumulative energy injection rate up to scale kk due to an external velocity vf\mathbf{v}_{\mathrm{f}} as

Fk≡dEkdt∣vf=κV∮C(s′×vs<kf)⋅vf dξ.(8)\mathcal{F}_{k} \equiv\left.\frac{\mathrm{d}\mathcal{E}_{k}}{\mathrm{d}t}\right|_{v_{\mathrm{f}}} = \frac{\kappa}{V}\oint_{C}\left(\mathbf{s}' \times\mathbf{v}_{\mathrm{s}}^{<k_{\mathrm{f}}}\right)\cdot\mathbf{v}_{\mathrm{f}}\,\mathrm{d}\xi. \tag*{(8)}

Clearly, choosing vf0=αs′×vs<kf\mathbf{v}_{\mathrm{f}}^{0}=\alpha\mathbf{s}'\times\mathbf{v}_{\mathrm{s}}^{<k_{\mathrm{f}}} (with α>0\alpha>0 a nondimensional forcing coefficient) ensures energy injection up to scale k=kfk=k_{\mathrm{f}}. However, to avoid the unwanted injection of energy at quantum scales see Eq. (5), we also require vf\mathbf{v}_{\mathrm{f}} to be orthogonal to s′′\mathbf{s}''. This leads to

vf(s)=vf0−vf0⋅s′′∣s′′∣2s′′=αs′′⋅vs<kf∣s′′∣2s′×s′′.(9)\mathbf{v}_{\mathrm{f}}(\mathbf{s})=\mathbf{v}_{\mathrm{f}}^{0}-\frac{\mathbf{v}_{\mathrm{f}}^{0}\cdot\mathbf{s}''}{\lvert\mathbf{s}''\rvert^{2}}\mathbf{s}''=\alpha\frac{\mathbf{s}''\cdot\mathbf{v}_{\mathrm{s}}^{<k_{\mathrm{f}}}}{\lvert\mathbf{s}''\rvert^{2}}\mathbf{s}'\times\mathbf{s}'' . \tag*{(9)}

From Eq. (8), the large-scale energy injection rate associated to this choice is then

εinj≡Fkf=ακV∮C∣s′′⋅vs<kf∣2∣s′′∣2 dξ.(10)\varepsilon_{\mathrm{inj}}\equiv\mathcal{F}_{k_{\mathrm{f}}}=\alpha\frac{\kappa}{V}\oint_{C}\frac{\left\lvert\mathbf{s}''\cdot\mathbf{v}_{\mathrm{s}}^{<k_{\mathrm{f}}}\right\rvert^{2}}{\lvert\mathbf{s}''\rvert^{2}}\,\mathrm{d}\xi. \tag*{(10)}

In practice, we dynamically tune the value of α\alpha to achieve a constant energy injection rate εinj\varepsilon_{\mathrm{inj}}, similarly in spirit to commonly used forcing schemes in classical turbulence [73].

Fast evaluation of Biot–Savart integrals—In the following, we summarize the FFT-based approach used to efficiently evaluate Biot–Savart integrals in periodic domains. For details the reader is referred to Ref. [47]. Similarly to standard Ewald summation methods [48], the main idea is to split the self-induced velocity field (1) as vs(x)=vs(n)(x)+vs(f)(x)v_s(\boldsymbol{x}) = v_s^{(\mathrm{n})}(\boldsymbol{x}) + v_s^{(\mathrm{f})}(\boldsymbol{x}), where vs(n)v_s^{(\mathrm{n})} and vs(f)v_s^{(\mathrm{f})} respectively account for the influence of vortex elements located near and far from the evaluation point x\boldsymbol{x}. To obtain expressions for these two components, we first introduce the vector potential ψs(x)\psi_s(\boldsymbol{x}) such that vs=∇×ψsv_s = \nabla\times\psi_s. It is related to the vorticity by Poisson’s equation ωs=∇×vs=−∇2ψs\omega_s = \nabla\times v_s = -\nabla^2\psi_s, whose solution can be written as the convolution ψs=G∗ωs\psi_s = G * \omega_s where G(r)=1/(4πr)G(r) = 1/(4\pi r) is the Green’s function associated to the 3D Poisson’s equation. In the context of the VFM, this becomes

ψs(x)=κ4π∮Cds∣s−x∣.(11)\psi_s(\boldsymbol{x}) = \frac{\kappa}{4\pi}\oint_C \frac{\mathrm{d}\boldsymbol{s}}{|\boldsymbol{s}-\boldsymbol{x}|}. \tag*{(11)}

It can be readily shown that ∇×ψs\nabla\times\psi_s leads to the Biot–Savart law (1). As a side note, knowing ψs\psi_s at vortex locations s∈C\boldsymbol{s} \in C conveniently allows to evaluate the total kinetic energy per unit mass as E=κ2V∮Cψs(s)⋅dsE = \frac{\kappa}{2V}\oint_C \psi_s(\boldsymbol{s})\cdot\mathrm{d}\boldsymbol{s}.

The main idea of standard Ewald methods is to split the Green’s function G(r)G(r) as

G(r)=erfc⁡(αr)4πr+erf⁡(αr)4πr=G(n)(r)+G(f)(r),(12)G(r) = \frac{\operatorname{erfc}(\alpha r)}{4\pi r} + \frac{\operatorname{erf}(\alpha r)}{4\pi r} = G^{(\mathrm{n})}(r) + G^{(\mathrm{f})}(r), \tag*{(12)}

where erf⁡(x)\operatorname{erf}(x) is the error function and erfc⁡(x)=1−erf⁡(x)\operatorname{erfc}(x) = 1-\operatorname{erf}(x). Here α\alpha is a tunable splitting parameter (an inverse length scale) setting the transition between near- and far-ranged interactions. Taking the gradient of Eq. (12), a similar splitting can then be obtained for the Biot–Savart kernel, ∇G(r)=−r/(4πr3)=∇G(n)(r)+∇G(f)(r)\nabla G(r) = -\boldsymbol{r}/(4\pi r^3) = \nabla G^{(\mathrm{n})}(r) + \nabla G^{(\mathrm{f})}(r), with

∇G(n)(r)=−[erfc⁡(αr)+2αrπe−(αr)2]r4πr3,(13)\nabla G^{(\mathrm{n})}(r) = -\left[\operatorname{erfc}(\alpha r) + \frac{2\alpha r}{\sqrt{\pi}}e^{-(\alpha r)^2}\right]\frac{\boldsymbol{r}}{4\pi r^3}, \tag*{(13)}
∇G(f)(r)=−[erf⁡(αr)−2αrπe−(αr)2]r4πr3.(14)\nabla G^{(\mathrm{f})}(r) = -\left[\operatorname{erf}(\alpha r) - \frac{2\alpha r}{\sqrt{\pi}}e^{-(\alpha r)^2}\right]\frac{\boldsymbol{r}}{4\pi r^3}. \tag*{(14)}

allowing to define vs(n)=∇G(n)∗ωsv_s^{(\mathrm{n})} = \nabla G^{(\mathrm{n})} * \omega_s and vs(f)=∇G(f)∗ωsv_s^{(\mathrm{f})} = \nabla G^{(\mathrm{f})} * \omega_s. With the above choice of splitting, it can be shown that vs(f)v_s^{(\mathrm{f})} is nothing else than the velocity field induced by a Gaussian-filtered vorticity field ωs(f)(x)=(φα∗ωs)(x)\omega_s^{(\mathrm{f})}(\boldsymbol{x}) = (\varphi_\alpha* \omega_s)(\boldsymbol{x}) with φα(r)=(α/π)3e−(αr)2\varphi_\alpha(r) = (\alpha/\sqrt{\pi})^3 e^{-(\alpha r)^2}.

The first crucial point here is that, unlike the full Biot–Savart kernel, ∇G(n)(r)\nabla G^{(\mathrm{n})}(r) decays very quickly with rr, which means that one only needs to consider nearby pairs within a radius rcut=β/αr_{\mathrm{cut}} = \beta/\alpha to obtain an arbitrarily accurate estimation of vs(n)v_s^{(\mathrm{n})}. Here β\beta is a nondimensional parameter controlling truncation errors [47], which we set to β=3.5\beta= 3.5 for 10−610^{-6} relative accuracy.

Here β\beta is a nondimensional parameter controlling truncation errors [47], which we set to β=3.5\beta= 3.5 for 10−610^{-6} relative accuracy. Secondly, ∇G(f)(r)\nabla G^{(\mathrm{f})}(r) is nonsingular and well-behaved near r=0r = 0, unlike ∇G(r)\nabla G(r). In periodic domains, this allows to represent vs(f)v_s^{(\mathrm{f})} as a truncated Fourier series up to wavenumber kmax⁡=2βαk_{\max} = 2\beta\alpha [47]. Then, the Fourier coefficients of vs(f)v_s^{(\mathrm{f})} are simply

v^s(f)(k)=ik×ω^s(f)(k)∣k∣2=ik×ω^s(k)∣k∣2e−k2/4α2,(15)\widehat{v}_s^{(\mathrm{f})}(\boldsymbol{k}) = \frac{i\boldsymbol{k}\times\widehat{\omega}_s^{(\mathrm{f})}(\boldsymbol{k})}{|\boldsymbol{k}|^2} = \frac{i\boldsymbol{k}\times\widehat{\omega}_s(\boldsymbol{k})}{|\boldsymbol{k}|^2}e^{-k^2/4\alpha^2}, \tag*{(15)}

where ω^s(k)=κV∮Ce−ik⋅s ds\widehat{\omega}_s(\boldsymbol{k}) = \frac{\kappa}{V}\oint_C e^{-i\boldsymbol{k}\cdot\boldsymbol{s}}\,\mathrm{d}\boldsymbol{s} is a Fourier coefficient of the superfluid vorticity field.

In practice, to obtain vs(f)v_s^{(\mathrm{f})}, we start by approximating the ω^s(k)\widehat{\omega}_s(\boldsymbol{k}) line integral using 3-point Gauss–Legendre quadratures on each discrete vortex segment, requiring 3N3N spline interpolations of s(ξ)s(\xi) and s′(ξ)s'(\xi). This results in a Fourier sum over 3N3N nonequispaced points, which is then efficiently evaluated using the type-1 nonuniform FFT (NUFFT) algorithm [74, 75]. This step yields ω^s(k)\widehat{\omega}_s(\boldsymbol{k}) coefficients on a uniform 3D Fourier grid truncated at ±kmax⁡\pm k_{\max} in each direction. Then, once Eq. (15) has been evaluated, we interpolate vs(f)v_s^{(\mathrm{f})} at vortex locations in physical space using the type-2 NUFFT. Overall, by properly choosing the splitting parameter α\alpha, this method leads to O(Nlog⁡N)O(N\log N) complexity, enabling the fast and arbitrarily accurate evaluation of Biot–Savart interactions in dense vortex systems.

Numerical runs—In Table I we detail relevant parameters associated to the VFM simulations discussed in the main text.

εinj\varepsilon_{\mathrm{inj}}a0a_0ReκRe_\kappaL0L_0L0/ℓL_0/\ellℓ/a0\ell/a_0NNTsim/T0T_{\mathrm{sim}}/T_0
5010−410^{-4}5.91.147.61.5×1031.5\times10^32.8×1052.8\times10^511.1
20010−410^{-4}8.71.1010.51.1×1031.1\times10^38.8×1058.8\times10^56.6
80010−410^{-4}14.81.1615.93.3×1023.3\times10^22.5×1062.5\times10^66.0
5010−710^{-7}5.71.236.31.9×1061.9\times10^61.3×1051.3\times10^515.8
20010−710^{-7}9.41.269.41.3×1061.3\times10^64.3×1054.3\times10^56.4
80010−710^{-7}14.91.2413.19.5×1059.5\times10^51.2×1061.2\times10^616.7
320010−710^{-7}23.31.2318.46.7×1056.7\times10^53.4×1063.4\times10^611.4

Table 1. Physical and numerical parameters of the simulations. The superfluid Reynolds number is Reκ=vrmsL0/κRe_\kappa= v_{\mathrm{rms}}L_0/\kappa [6, 7, 76, 77] where vrms2=(2/3)Ekv_{\mathrm{rms}}^2 = (2/3)\mathcal{E}_{k} is the large-scale velocity variance (associated to turbulent fluctuations). The integral scale is estimated as L0=π2vrms2∫0kℓk−1E(k) dkL_0 = \frac{\pi}{2v_{\mathrm{rms}}^2}\int_0^{k_\ell} k^{-1}E(k)\,\mathrm{d}k where E(k)E(k) is the energy spectrum. NN is the number of discrete vortex points. Tsim/T0T_{\mathrm{sim}}/T_0 is the simulation time in units of the eddy turnover time T0=L0/vrmsT_0 = L_0/v_{\mathrm{rms}}. See main text for other definitions.

References

  1. [1]R. J. Donnelly, Quantized Vortices in Helium II (Cambridge University Press, 1991).
  2. [2]C. F. Barenghi, L. Skrbek, and K. R. Sreenivasan, Quantum Turbulence (Cambridge University Press, Cambridge, 2023).
  3. [3]J. Maurer and P. Tabeling, Local investigation of superfluid turbulence, Europhys. Lett. 43, 29 (1998).
  4. [4]A. W. Baggaley, C. F. Barenghi, A. Shukurov, and Y. A. Sergeev, Coherent vortex structures in quantum turbulence, EPL 98, 26002 (2012).
  5. [5]N. P. Müller, J. I. Polanco, and G. Krstulovic, Intermittency of Velocity Circulation in Quantum Turbulence, Phys. Rev. X 11, 011053 (2021).DOI
  6. [6]C. Bret, P. Diribarne, J. Duplat, and B. Rousset, Kolmogorov cascade as the governing mechanism for intervortex spacing in quantum turbulence, Phys. Rev. Fluids 10, 094701 (2025).DOI
  7. [7]J. I. Polanco, P.-E. Roche, L. Danaila, and E. Lévêque, Disentangling temperature and Reynolds number effects in quantum turbulence, Proc. Natl. Acad. Sci. U.S.A. 122, e2426598122 (2025).DOI
  8. [8]L. Galantucci, Quantum turbulent flows: a model for classical turbulence?, J. Fluid Mech. 1022, F1 (2025).DOI
  9. [9]P. M. Walmsley, A. I. Golov, H. E. Hall, A. A. Levchenko, and W. F. Vinen, Dissipation of Quantum Turbulence in the Zero Temperature Limit, Phys. Rev. Lett. 99, 265302 (2007).DOI
  10. [10]M. S. Paoletti, M. E. Fisher, K. R. Sreenivasan, and D. P. Lathrop, Velocity Statistics Distinguish Quantum Turbulence from Classical Turbulence, Phys. Rev. Lett. 101, 154501 (2008).
  11. [11]P. Švancara and M. La Mantia, Flows of liquid 4He due to oscillating grids, J. Fluid Mech. 832, 578 (2017).
  12. [12]Y. Tang, S. Bao, T. Kanai, and W. Guo, Statistical properties of homogeneous and isotropic turbulence in He II measured via particle tracking velocimetry, Phys. Rev. Fluids 5, 084602 (2020).DOI
  13. [13]J. Salort, F. Chillà, E. Rusaouën, P.-E. Roche, M. Gibert, I. Moukharski, A. Braslau, F. Daviaud, B. Gallet, E.-W. Saw, B. Dubrulle, P. Diribarne, B. Rousset, M. Bardin, J.-P. Moro, A. Girard, C. Baudet, V. L’vov, A. Golov, and S. Nazarenko, Experimental signature of quantum turbulence in velocity spectra?, New J. Phys. 23, 063005 (2021).DOI
  14. [14]V. S. L’vov, S. V. Nazarenko, and O. Rudenko, Bottleneck crossover between classical and quantum superfluid turbulence, Phys. Rev. B 76, 024520 (2007).DOI
  15. [15]P.-E. Roche and C. F. Barenghi, Vortex spectrum in superfluid turbulence: Interpretation of a recent experiment, EPL 81, 36002 (2008).
  16. [16]A. W. Baggaley and C. F. Barenghi, Vortex-density fluctuations in quantum turbulence, Phys. Rev. B 84, 020504 (2011).DOI
  17. [17]J. I. Polanco, N. P. Müller, and G. Krstulovic, Vortex clustering, polarisation and circulation intermittency in classical and quantum turbulence, Nat. Commun. 12, 7090 (2021).
  18. [18]W. F. Vinen and J. J. Niemela, Quantum turbulence, J. Low Temp. Phys. 128, 167 (2002).
  19. [19]U. Frisch, Turbulence: The Legacy of A.N. Kolmogorov, 1st ed. (Cambridge University Press, Cambridge, UK, 1995).
  20. [20]J. Finnigan, Turbulence in Plant Canopies, Annu. Rev. Fluid Mech. 32, 519 (2000).
  21. [21]S. Olivieri, L. Brandt, M. E. Rosti, and A. Mazzino, Dispersed Fibers Change the Classical Energy Budget of Turbulence via Nonlocal Transfer, Phys. Rev. Lett. 125, 114501 (2020).DOI
  22. [22]I. Cannon, S. Olivieri, and M. E. Rosti, Spheres and fibers in turbulent flows at various Reynolds numbers, Phys. Rev. Fluids 9, 064301 (2024).DOI
  23. [23]H. Tennekes and J. L. Lumley, A First Course in Turbulence (MIT Press, 1972).DOI
  24. [24]Wm. T. Ashurst, A. R. Kerstein, R. M. Kerr, and C. H. Gibson, Alignment of vorticity and scalar gradient with strain rate in simulated Navier–Stokes turbulence, Phys. Fluids 30, 2343 (1987).DOI
  25. [25]D. Buaria, E. Bodenschatz, and A. Pumir, Vortex stretching and enstrophy production in high Reynolds number turbulence, Phys. Rev. Fluids 5, 104602 (2020).DOI
  26. [26]C. Nore, M. Abid, and M. E. Brachet, Kolmogorov Turbulence in Low-Temperature Superflows, Phys. Rev. Lett. 78, 3896 (1997).DOI
  27. [27]T. Araki, M. Tsubota, and S. K. Nemirovskii, Energy Spectrum of Superfluid Turbulence with No Normal-Fluid Component, Phys. Rev. Lett. 89, 145301 (2002).pubmed.ncbi.nlm.nih.gov/12366052
  28. [28]A. Villois, D. Proment, and G. Krstulovic, Evolution of a superfluid vortex filament tangle driven by the Gross-Pitaevskii equation, Phys. Rev. E 93, 061103 (2016).DOI
  29. [29]N. P. Müller and G. Krstulovic, Kolmogorov and Kelvin wave cascades in a generalized model for quantum turbulence, Phys. Rev. B 102, 134513 (2020).DOI
  30. [30]S. R. Stalp and R. J. Donnelly, Decay of Grid Turbulence in Superfluid 4He, in Flow at Ultra-High Reynolds and Rayleigh Numbers, edited by R. J. Donnelly and K. R. Sreenivasan (Springer New York, New York, NY, 1998) pp. 419–435.DOI
  31. [31]C. F. Barenghi, A. V. Godeev, and L. Skrbek, Depolarization of decaying counterflow turbulence in He II, Phys. Rev. E 74, 026309 (2006).pubmed.ncbi.nlm.nih.gov/17025541
  32. [32]P. Walmsley, D. Zmeëv, F. Pakpour, and A. Golov, Dynamics of quantum turbulence of different spectra, Proc. Natl. Acad. Sci. U.S.A. 111, 4691 (2014).DOI
  33. [33]M. Leadbeater, T. Winiecki, D. C. Samuels, C. F. Barenghi, and C. S. Adams, Sound Emission due to Superfluid Vortex Reconnections, Phys. Rev. Lett. 86, 1410 (2001).
  34. [34]A. Villois, D. Proment, and G. Krstulovic, Irreversible Dynamics of Vortex Reconnections in Quantum Fluids, Phys. Rev. Lett. 125, 164501 (2020).DOI
  35. [35]K. W. Schwartz, Three-dimensional vortex dynamics in superfluid 4He: Line-line and line-boundary interactions, Phys. Rev. B 31, 5782 (1985).
  36. [36]R. Hänninen and A. W. Baggaley, Vortex filament method as a tool for computational visualization of quantum turbulence, Proc. Natl. Acad. Sci. USA 111, 4667 (2014).DOI
  37. [37]P. G. Saffman, Vortex Dynamics (Cambridge University Press, 1993).
  38. [38]R. J. Arms and F. R. Hama, Localized-Induction Concept on a Curved Vortex and Motion of an Elliptic Vortex Ring, Phys. Fluids 8, 553 (1965).DOI
  39. [39]E. B. Sonin, Vortex oscillations and hydrodynamics of rotating superfluids, Rev. Mod. Phys. 59, 87 (1987).DOI
  40. [40]R. H. Kraichnan, Isotropic Turbulence and Inertial-Range Structure, Phys. Fluids 9, 1728 (1966).DOI
  41. [41]Y. Zhou, Degrees of locality of energy transfer in the inertial range, Phys. Fluids 5, 1092 (1993).DOI
  42. [42]G. L. Eyink and H. Aluie, Localness of energy cascade in hydrodynamic turbulence. I. Smooth coarse graining, Phys. Fluids 21, 115107 (2009).
  43. [43]J. I. Cardesa, A. Vela–Martín, and J. Jiménez, The turbulent cascade in five dimensions, Science 357, 782 (2017).
  44. [44]G. I. Taylor, Production and dissipation of vorticity in a turbulent fluid, Proc. R. Soc. Lond. A 164, 15 (1938).DOI
  45. [45]P. E. Hamlington, J. Schumacher, and W. J. A. Dahm, Direct assessment of vorticity alignment with local and nonlocal strain rates in turbulent flows, Phys. Fluids 20, 111703 (2008).
  46. [46]D. Buaria and A. Pumir, Nonlocal amplification of intense vorticity in turbulent flows, Phys. Rev. Research 3, L042020 (2021).DOI
  47. [47]J. I. Polanco, Fast and Accurate Evaluation of Biot–Savart Integrals over Spatial Curves in Periodic Domains, SIAM J. Sci. Comput. 47, B822 (2025).DOI
  48. [48]A. Arnold and C. Holm, Efficient Methods to Compute Long-Range Interactions for Soft Matter Systems, in Advanced Computer Simulation Approaches for Soft Matter Sciences II, Advances in Polymer Science, edited by C. Holm and K. Kremer (Springer, Berlin, Heidelberg, 2005) pp. 59–109.DOI
  49. [49]A. Arnold, F. Fahrenberger, C. Holm, O. Lenz, M. Bolten, H. Dachsel, R. Halver, I. Kabashow, F. Gäßler, F. Heber, J. Is- eringhausen, M. Hofmann, M. Pippig, D. Potts, and G. Sutmann, Comparison of scalable fast methods for long-range interactions, Phys. Rev. E 88, 063308 (2013).pubmed.ncbi.nlm.nih.gov/24483585
  50. [50]A. W. Baggaley and C. F. Barenghi, Tree Method for Quantum Vortex Dynamics, J. Low Temp. Phys. 166, 3 (2012).
  51. [51]S. Yui, H. Kobayashi, M. Tsubota, and R. Yokota, Quantum Turbulence Coupled with Externally Driven Normal-Fluid Turbulence in Counterflow of Superfluid 4He, J. Phys. Soc. Jpn. 94, 043601 (2025).DOI
  52. [52]J. I. Polanco, jipolanco/VortexPasta.jl: v0.32.12, Zenodo (2025).
  53. [53]H. Adachi, S. Fujiyama, and M. Tsubota, Steady-state counterflow quantum turbulence: Simulation of vortex filaments using the full Biot-Savart law, Phys. Rev. B 81, 104511 (2010).DOI
  54. [54]A. W. Baggaley, J. Laurie, and C. F. Barenghi, Vortex-Density Fluctuations, Energy Spectra, and Vortical Regions in Superfluid Turbulence, Phys. Rev. Lett. 109, 205304 (2012).
  55. [55]L. Galantucci, A. W. Baggaley, N. G. Parker, and C. F. Barenghi, Crossover from interaction to driven regimes in quantum vortex reconnections, Proc. Natl. Acad. Sci. 116, 12204 (2019).
  56. [56]A. W. Baggaley, The Sensitivity of the Vortex Filament Method to Different Reconnection Models, J. Low Temp. Phys. 168, 18 (2012).DOI
  57. [57]A. W. Baggaley, The importance of vortex bundles in quantum turbulence at absolute zero, Phys. Fluids 24, 055109 (2012).DOI
  58. [58]A. Alexakis and L. Biferale, Cascades and transitions in turbulent flows, Phys. Rep. 767, 1 (2018).DOI
  59. [59]C. Nore, M. Abid, and M. E. Brachet, Decaying Kolmogorov turbulence in a model of superflow, Phys. Fluids 9, 2644 (1997).DOI
  60. [60]M. Kobayashi and M. Tsubota, Kolmogorov Spectrum of Superfluid Turbulence: Numerical Analysis of the Gross-Pitaevskii Equation with a Small-Scale Dissipation, Phys. Rev. Lett. 94, 065302 (2005).pubmed.ncbi.nlm.nih.gov/15783741
  61. [61]D. Kivotides, J. C. Vassilicos, D. C. Samuels, and C. F. Barenghi, Kelvin Waves Cascade in Superfluid Turbulence, Phys. Rev. Lett. 86, 3080 (2001).pubmed.ncbi.nlm.nih.gov/11290112
  62. [62]W. F. Vinen, M. Tsubota, and A. Mitani, Kelvin-Wave Cascade on a Vortex in Superfluid 4He at a Very Low Temperature, Phys. Rev. Lett. 91, 135301 (2003).pubmed.ncbi.nlm.nih.gov/14525312
  63. [63]E. Kozik and B. Svistunov, Kelvin-Wave Cascade and Decay of Superfluid Turbulence, Phys. Rev. Lett. 92, 035301 (2004).pubmed.ncbi.nlm.nih.gov/14753882
  64. [64]V. S. L’vov and S. Nazarenko, Spectrum of Kelvin-wave turbulence in superfluids, JETP Lett. 91, 428 (2010).DOI
  65. [65]G. Krstulovic, Kelvin-wave cascade and dissipation in low-temperature superfluid vortices, Phys. Rev. E 86, 10.1103/Phys- RevE.86.055301 (2012).pubmed.ncbi.nlm.nih.gov/23214835
  66. [66]A. W. Baggaley and J. Laurie, Kelvin-wave cascade in the vortex filament model, Phys. Rev. B 89, 014504 (2014).
  67. [67]S. Z. Alamri, A. J. Youd, and C. F. Barenghi, Reconnection of Superfluid Vortex Bundles, Phys. Rev. Lett. 101, 215302 (2008).
  68. [68]D. I. Bradley, D. O. Clubb, S. N. Fisher, A. M. Guénault, R. P. Haley, C. J. Matthews, G. R. Pickett, V. Tsepelin, and K. Zaki, Decay of Pure Quantum Turbulence in Superfluid 3He-B, Phys. Rev. Lett. 96, 035301 (2006).pubmed.ncbi.nlm.nih.gov/16486721
  69. [69]J. M. Lattimer and M. Prakash, The Physics of Neutron Stars, Science 304, 536 (2004).DOI
  70. [70]S. Seveso, P. M. Pizzochero, F. Grill, and B. Haskell, Mesoscopic pinning forces in neutron star crusts, Mon. Not. R. Astron. Soc. 455, 3952 (2016).DOI
  71. [71]J.-K. Liu, A. W. Baggaley, C. F. Barenghi, and T. S. Wood, Vortex Avalanches and Collective Motion in Neutron Stars, Astrophys. J. 984, 83 (2025).
  72. [72]V. Graber, N. Andersson, and M. Hogg, Neutron stars in the laboratory, Int. J. Mod. Phys. D 26, 1730015 (2017).DOI
  73. [73]A. G. Lamorgese, D. A. Caughey, and S. B. Pope, Direct numerical simulation of homogeneous turbulence with hyperviscosity, Phys. Fluids 17, 015106 (2005).DOI
  74. [74]A. Dutt and V. Rokhlin, Fast Fourier Transforms for Nonequispaced Data, SIAM J. Sci. Comput. 14, 1368 (1993).DOI
  75. [75]L. Greengard and J.-Y. Lee, Accelerating the Nonuniform Fast Fourier Transform, SIAM Rev. 46, 443 (2004).DOI
  76. [76]A. P. Finne, T. Araki, R. Blaauwgeers, V. B. Eltsov, N. B. Kopnin, M. Krusius, L. Skrbek, M. Tsubota, and G. E. Volovik, An intrinsic velocity-independent criterion for superfluid turbulence, Nature 424, 1022 (2003).pubmed.ncbi.nlm.nih.gov/12944960
  77. [77]J. Salort, P.-E. Roche, and E. Lévêque, Mesoscale equipartition of kinetic energy in quantum turbulence, EPL 94, 24001 (2011).DOI

Paper details

vlnc:W_9xasmch8
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