betaFeedback

WorksPhysics and Astronomy

Toward interpreting the IBEX ribbon with mirror diffusion in interstellar turbulent magnetic fields

Abstract

We investigate the role of the magnetohydrodynamic (MHD) turbulence measured by Voyager in the very local interstellar medium (VLISM) in modeling the Interstellar Boundary Explorer (IBEX) ribbon. We demonstrate that the mirroring by compressible modes of MHD turbulence dominates over that by the mean magnetic field. Based on the new mirror diffusion mechanism identified by Lazarian and Xu for particles with large pitch angles in MHD turbulence, we find that the mirror diffusion can both confine pickup ions and preserve their initial pitch angles, and thus accounts for the enhanced intensity of energetic neutral atoms that return to the heliosphere. The ribbon width is determined by both the range of pitch angles for effective turbulent mirroring and the field line wandering induced by Alfvénic modes. It in turn provides a constraint on the amplitude of magnetic fluctuations of fast modes. The field line wandering also affects the coherence of the ribbon structure across the sky. By extrapolating the magnetic energy spectrum measured by Voyager, we find that the injection scale of the turbulence in the VLISM is less than ∼500\sim 500 au for the ribbon structure to be coherent.

Introduction

The “ribbon” of enhanced energetic neutral atom (ENA) emissions discovered by the Interstellar Boundary Explorer (IBEX) [29] provides valuable information about the local interstellar magnetic field [35, 31] and the interaction of pickup ions with the pre-existing turbulent magnetic fluctuations outside the heliopause [53] (Giacalone & Jokipii 2015). Turbulent magnetic fluctuations following a Kolmogorov spectrum are observed by Voyager 1 and 2 in the outer heliosheath [4, 52, 5, 25, 13]. With a higher amplitude than that of the interstellar turbulent spectrum [4, 25, 30] and ion-neutral collisional damping of the interstellar turbulence in the partially ionized very local interstellar medium (VLISM) [44], the turbulence detected by Voyager is more likely to be of heliospheric origin [50]. Among the numerous models for explaining the origin of the IBEX ribbon, the magnetic mirroring of pickup ions in the turbulent magnetic fields in the VLISM is found to be important in determining the ribbon structure [53] (Giacalone & Jokipii 2015).

The magnetic mirrors created by summing over a large number of randomly directed plane waves with random polarizations and phases [14] can trap particles as particles are reflected back and forth between two mirror points [6]. Unlike compressible magnetohydrodynamic (MHD) waves, MHD turbulence contains fast and slow (or pseudo-Alfvénic) modes that generate magnetic mirrors and Alfvénic modes that cause perpendicular superdiffusion of magnetic field lines [1] (Cho & Lazarian 2003). Following turbulent magnetic field lines, particles also undergo superdiffusion in the direction perpendicular to the magnetic field [24] (Xu & Yan 2013; Hu et al. 2022) while interacting stochastically with different magnetic mirrors along the magnetic field. As a result, rather than being trapped, particles exhibit the mirror diffusion [22] parallel to the magnetic field. The mirror diffusion is usually a much slower diffusion process compared to that associated with pitch-angle scattering [22, 41].

The MHD turbulence measured by Voyager contains both incompressible and compressible modes [50, 25, 13]. In this work, we will investigate the importance of the recently identified mirror diffusion of particles in the MHD turbulence in the VLISM in explaining the origin of the IBEX ribbon and affecting its structure. In Section 2, we first examine the mirroring effect of the compressed interstellar mean magnetic field. In Section 3, we move to the mirroring effect of turbulent magnetic fields. The mirror diffusion resulting from the perpendicular superdiffusion of turbulent magnetic fields and its effect on the ribbon width is discussed in Section 4. The effect of field line wandering on ribbon width and structure is studied in Section 5. Further discussion is presented in Section 6. A summary of our main results follows in Section 7.

Mirroring effect of the interstellar mean magnetic field

The compression of the large-scale mean field in the outer heliosheath gives rise to the magnetic mirroring effect. It can cause the concentration of pickup protons in a narrow region where the radial component of the compressed interstellar mean magnetic field is close to zero [29, 7].

For a particle with rg<Lmr_g < L_m, where rgr_g is the particle gyroradius, and LmL_m is the size of the magnetic mirror, i.e., the variation scale of the mean magnetic field, the particle preserves its first adiabatic invariant, with a constant magnetic moment

mv⊥22B=constant.(1)\frac{m v_\perp^2}{2B}=\text{constant}. \tag*{(1)}

Here v⊥v_\perp is the particle velocity perpendicular to the magnetic field, mm is the particle mass, and BB is the magnetic field strength. When a particle with constant energy and magnetic moment moves into a region with converging magnetic fields, its velocity along the magnetic field v∥v_\parallel decreases, and the angle between the particle velocity vv and the magnetic field BB, i.e., the pitch angle, increases. The deceleration toward the mirror point where the mean magnetic field has the maximum strength and is perpendicular to the heliocentric radial direction can potentially lead to the accumulation of the pickup ions with large pitch angles [7].

The magnetic mirror force is [6]

d(pμ)dt=−v⊥p⊥2B0δBmLm,(2)\frac{d(p\mu)}{dt}=-\frac{v_\perp p_\perp}{2B_0}\frac{\delta B_m}{L_m}, \tag*{(2)}

where μ\mu is cosine of the pitch angle, pp is the particle momentum, tt is time, B0B_0 is the mean magnetic field strength, and δBm\delta B_m is the parallel magnetic fluctuation over LmL_m. It follows that the rate of change in μ\mu due to mirroring, i.e., the mirroring rate, is

Γm=∣1μdμdt∣=v2B0δBmLm1−μ2μ,(3)\Gamma_m=\left|\frac{1}{\mu}\frac{d\mu}{dt}\right|=\frac{v}{2B_0}\frac{\delta B_m}{L_m}\frac{1-\mu^2}{\mu}, \tag*{(3)}

where vv is the particle speed. Under the consideration of keV protons (v≈438 km s−1v\approx438\ {\rm km\ s}^{-1}), δBm/B0≈1/4\delta B_m/B_0\approx1/4 and Lm≈100L_m\approx100 au [7], Γm\Gamma_m as a function of μ\mu is presented in Figure 1. It depends on the longitudinal gradient in the mean magnetic field. The maximum μ\mu for mirroring is determined by

Rate of mirroring and scattering for keV protons

Figure 1. The rate of mirroring Γm\Gamma_{\rm m} by the interstellar mean magnetic field and the rate of scattering Γsc\Gamma_{\rm sc} by the turbulent magnetic fluctuations in the outer heliosheath for keV protons. The turbulence parameters are adopted (see text) to reach an approximate balance between Γm\Gamma_{\rm m} and Γsc\Gamma_{\rm sc}. For the magnetic mirroring to be effective, the turbulence parameters should satisfy Γm>Γsc\Gamma_{\rm m}>\Gamma_{\rm sc}.

μmax⁡=δBmB0+δBm,(4)\mu_{\max}=\sqrt{\frac{\delta B_m}{B_0+\delta B_m}}, \tag*{(4)}

which is derived from the first adiabatic invariant

v⊥2B0=v2B0+δBm.(5)\frac{v_\perp^2}{B_0}=\frac{v^2}{B_0+\delta B_m}. \tag*{(5)}

The turbulent magnetic fluctuations detected by Voyager 1 and 2 in the outer heliosheath (e.g., [4]) affects the transport of pickup protons via the field line wandering (see § 5), turbulent mirroring (see § 3), and pitch-angle scattering. As efficient scattering causes breaking of the first adiabatic invariant of particles, an effective magnetic mirroring imposes constraints on the amplitude of turbulent fluctuations. Here we focus on the gyroresonant scattering [19, 20, 32] of pickup protons by fast modes of MHD turbulence. Compared with Alfvenic modes, fast modes usually have a smaller energy fraction [9] (Hu et al. 2022). However, their isotropic energy scaling [9] leads to a more efficient gyroresonant scattering (Yan & Lazarian 2004; Xu & Yan 2013; Xu & Lazarian 2020) compared to the anisotropic Alfvenic modes [8, 47, 3].

The rate of scattering is defined as [19, 6]

Γsc=1μ2⟨δμ2t⟩=2Dμμμ2,(6)\Gamma_{\rm sc}=\frac{1}{\mu^2}\left\langle\frac{\delta\mu^2}{t}\right\rangle=\frac{2D_{\mu\mu}}{\mu^2}, \tag*{(6)}

where δμ\delta\mu is the change of μ\mu due to scattering, and DμμD_{\mu\mu} is the pitch-angle diffusion coefficient of fast modes [40]

Dμμ=Cμ∫d3kk∥2k2[J1′(x)]2I(k)R(k).(7)D_{\mu\mu}=C_\mu\int d^3k\frac{k_\parallel^2}{k^2}[J_1'(x)]^2I(k)R(k). \tag*{(7)}

In the above expression, J1′(x)J_1'(x) is the derivative of the Bessel function, and I(k)I(k) is the magnetic energy spectrum of fast modes [9],

I(k)=Cfk−7/2,Cf=116πδBf2L−1/2.(8)I(k)=C_f k^{-7/2},\quad C_f=\frac{1}{16\pi}\delta B_f^2L^{-1/2}. \tag*{(8)}

kk is the wavenumber, LL is the injection scale of turbulence, δBf\delta B_f is the rms strength of magnetic fluctuations of fast modes at LL, R(k)R(k) is the resonance function for gyroresonance,

R(k)=πδ(ωk−v∥k∥+Ω),Ω=qB0mc.(9)R(k)=\pi\delta(\omega_k-v_\parallel k_\parallel+\Omega),\quad\Omega=\frac{qB_0}{mc}. \tag*{(9)}

Ω\Omega is the particle gyrofrequency, ωk=kVA\omega_k=kV_A is the wave frequency, VA=B0/4πmHniV_A=B_0/\sqrt{4\pi m_H n_i} is the Alfvén speed in ions, mHm_H is the hydrogen atomic mass, nin_i is the ion number density, k∥k_\parallel is the parallel component of kk, qq is the electric charge, mm is the proton mass, cc is the light speed,

Cμ=(1−μ2)Ω2B02,(10)C_\mu=(1-\mu^2)\frac{\Omega^2}{B_0^2}, \tag*{(10)}
x=k⊥v⊥Ω=k⊥rg−1,rg=v⊥Ω,(11)x=\frac{k_\perp v_\perp}{\Omega}=\frac{k_\perp}{r_g^{-1}},\quad r_g=\frac{v_\perp}{\Omega}, \tag*{(11)}

and k⊥k_\perp is the perpendicular component of kk. The approximate expression for Γsc\Gamma_{\rm sc} in the limit of a small xx is (Xu & Lazarian 2020),

Γsc≈π2δBf2B02(vLΩ)1/2Ω(1−μ2)μ−3/2,(12)\Gamma_{\rm sc}\approx\frac{\pi}{2}\frac{\delta B_f^2}{B_0^2}\left(\frac{v}{L\Omega}\right)^{1/2}\Omega(1-\mu^2)\mu^{-3/2}, \tag*{(12)}

which depends on δBf\delta B_f.

As discussed above, for the mirroring of the mean magnetic field to be effective, there should be Γsc<Γm\Gamma_{\rm sc}<\Gamma_m. In Figure 1, we present Γsc\Gamma_{\rm sc} for keV protons calculated using Eqs. (6) and (7), with B0=5 μGB_0=5\ \mu{\rm G}, L=100L=100 au, δBf/B0=0.04\delta B_f/B_0=0.04, and ni=0.07 cm−3n_i=0.07\ {\rm cm}^{-3} [38, 39]. We adopt 5×1075\times10^7 cm as the cutoff scale of turbulent spectrum, which is comparable to the ion inertial length ([13]) and smaller than rg≈9×108r_g \approx9 \times10^{8} cm at μ=0\mu= 0. Due to the lack of turbulent magnetic fluctuations at small scales, the scattering is absent at small μ\mu. Given the above parameters, the mirroring by the mean field and the scattering by turbulence are approximately in balance over the μ\mu values where they are both present. The condition Γsc<Γm\Gamma_{\rm sc}<\Gamma_{\rm m} can be used to constrain δBf\delta B_f and LL for the MHD turbulence in the outer heliosheath.

Mirroring effect of turbulent magnetic fields

The mirroring of the interstellar mean magnetic field alone may not be able to account for the ribbon. Moreover, Zirnstein et al. [54] argued that the mean field mirror acts to push away pickup ions from the mirror point and reduces the ribbon flux. In addition to the mirroring effect of the interstellar mean magnetic field, the compressible MHD turbulence in the outer heliosheath can also contribute to the magnetic mirroring. Both fast and slow modes create multi-scale magnetic mirrors within the inertial range of turbulence. Isotropic fast modes are more efficient in mirroring than anisotropic slow modes (Xu & Lazarian 2020), as the latter have the magnetic fluctuations decrease more rapidly with decreasing scales in the direction parallel to the local magnetic field. Similar to the analysis in § 2, one easily finds that the mirroring rate of fast modes is

Γtm=v2B0bfkk1−μ2μ,(13)\Gamma_{\rm tm}=\frac{v}{2B_0}b_{fk}k\frac{1-\mu^2}{\mu}, \tag*{(13)}

where bfkb_{fk} is the magnetic fluctuation of fast modes at kk. Among the mirrors at different wavenumbers, the ones that are most effective in reflecting the particles at μ\mu have ([6])

μ2≈bfkB0.(14)\mu^2 \approx\frac{b_{fk}}{B_0}. \tag*{(14)}

Smaller mirrors would have insufficient magnetic fluctuation to reflect the particles, and larger mirrors would have a lower mirroring rate. Given the scaling relation of fast modes ([9]),

bfk=δBf(kL)−14,(15)b_{fk}=\delta B_f(kL)^{-\frac{1}{4}}, \tag*{(15)}

there is (Xu & Lazarian 2020),

Γtm≈v2L(δBfB0)41−μ2μ7.(16)\Gamma_{\rm tm}\approx\frac{v}{2L}\left(\frac{\delta B_f}{B_0}\right)^4\frac{1-\mu^2}{\mu^7}. \tag*{(16)}

We note that unlike the mirroring of the mean field, for the mirroring of turbulent magnetic fields, particles with a given μ\mu predominantly interact with the mirrors at the corresponding kk. Γtm\Gamma_{\rm tm} has a much stronger dependence on μ\mu compared to Γm\Gamma_{\rm m}. Despite the much smaller fluctuation of turbulent magnetic fields, Γtm\Gamma_{\rm tm} is significantly larger than Γm\Gamma_{\rm m} due to the large magnetic field gradients at small scales (see Figure 1 and Figure 2). It suggests that the mirroring effect of the turbulent magnetic fields dominates over that of the mean field.

Rates of mirroring and scattering by turbulent magnetic fluctuations

Figure 2. The rate of mirroring Γtm\Gamma_{\mathrm{tm}} (Eqs. (16) and (20)) and the rate of scattering Γsc\Gamma_{\mathrm{sc}} (Eqs. (6) and (7)) by the turbulent magnetic fluctuations in the outer heliosheath for keV protons. Same parameters as in Figure 1 are used except for δBf/B0\delta B_f/B_0 values. The vertical dashed lines denote μc\mu_c.

Eq. (16) stands for the case with rg<1/k<Lr_g<1/k<L, that is (Eqs. (14) and (15)) (Xu & Lazarian 2020),

μ>μrg≈bfk(rg)B0=(δBfB0)12(rgL)18,(17)\mu>\mu_{r_g}\approx\sqrt{\frac{b_{fk}(r_g)}{B_0}}=\left(\frac{\delta B_f}{B_0}\right)^{\frac{1}{2}}\left(\frac{r_g}{L}\right)^{\frac{1}{8}}, \tag*{(17)}

and

μ<μmax⁡=δBfB0+δBf,(18)\mu<\mu_{\max}=\sqrt{\frac{\delta B_f}{B_0+\delta B_f}}, \tag*{(18)}

which is derived from the first adiabatic invariant,

v⊥2B0=v2B0+δBf.(19)\frac{v_\perp^2}{B_0}=\frac{v^2}{B_0+\delta B_f}. \tag*{(19)}

For 1/k<rg1/k<r_g, i.e. μ<μrg\mu<\mu_{r_g}, the mirrors at 1/k≈rg1/k\approx r_g dominate the mirroring. Therefore, there is (Xu & Lazarian 2020)

Γtm=v2B0bfk(rg)rg1−μ2μ=v2rgδBfB0(rgL)141−μ2μ.(20)\Gamma_{\rm tm}=\frac{v}{2B_0}\frac{b_{fk}(r_g)}{r_g}\frac{1-\mu^2}{\mu}=\frac{v}{2r_g}\frac{\delta B_f}{B_0}\left(\frac{r_g}{L}\right)^{\frac{1}{4}}\frac{1-\mu^2}{\mu}. \tag*{(20)}

In Figure 2, we present Γtm\Gamma_{\rm tm} and Γsc\Gamma_{\rm sc} for the mirroring and scattering of keV protons by fast modes. The same parameters as in Figure 1 are used but for different δBf/B0\delta B_f/B_0 values. If Γtm\Gamma_{\rm tm} and Γsc\Gamma_{\rm sc} can reach balance, the critical μc\mu_c is defined at their balance. With Γtm>Γsc\Gamma_{\rm tm}>\Gamma_{\rm sc} at μ<μc\mu<\mu_c, the diffusion of protons (parallel to the local magnetic field) is dominated by mirroring (see § 4). At larger μ\mu, scattering becomes more important than mirroring. As an approximation, by using Eqs. (12) and (16), we find Xu & Lazarian (2020)

μc≈[14πδBf2B02(vLΩ)12]211\mu_c \approx\left[\frac{14}{\pi}\frac{\delta B_f^2}{B_0^2}\left(\frac{v}{L\Omega}\right)^{\frac{1}{2}}\right]^{\frac{2}{11}}
≈0.11(δBf/B00.04)411(L100 au)−111(B05 μG)−111(v438 km s−1)111,(21)\approx0.11\left(\frac{\delta B_f/B_0}{0.04}\right)^{\frac{4}{11}}\left(\frac{L}{100\ \mathrm{au}}\right)^{-\frac{1}{11}}\left(\frac{B_0}{5\ \mu\mathrm{G}}\right)^{-\frac{1}{11}}\left(\frac{v}{438\ \mathrm{km\ s^{-1}}}\right)^{\frac{1}{11}}, \tag*{(21)}

which is close to the exact value of μc\mu_c at the intersection between Γtm\Gamma_{\mathrm{tm}} and Γsc\Gamma_{\mathrm{sc}} (see Figure 2). The actual μc\mu_c is slightly larger than the above estimate due to the cutoff of turbulent spectrum that was not taken into account in Xu & Lazarian (2020). If the balance between Γtm\Gamma_{\mathrm{tm}} and Γsc\Gamma_{\mathrm{sc}} cannot be reached (see the case with δBf/B0=0.01\delta B_f/B_0 = 0.01 in Figure 2), μc\mu_c is determined by μmax⁡\mu_{\max} in Eq. (18). By using Eqs. (14) and (15), we find that the corresponding length scale is

kc−1=L(δBfB0)−4μc8k_c^{-1}=L\left(\frac{\delta B_f}{B_0}\right)^{-4}\mu_c^8
≈43 au(δBf/B00.04)−4(L100 au)(μc0.18)8,(22)\approx43\ \mathrm{au}\left(\frac{\delta B_f/B_0}{0.04}\right)^{-4}\left(\frac{L}{100\ \mathrm{au}}\right)\left(\frac{\mu_c}{0.18}\right)^8, \tag*{(22)}

as the maximum size of the mirrors in turbulent magnetic fields for mirroring keV protons, which is much larger than rgr_g.

Mirror diffusion and ribbon width

Based on the mirroring by fast modes and perpendicular superdiffusion and wandering of magnetic fields induced by Alfvén modes, our model for interpreting the IBEX ribbon is

illustrated in Figure 3. Unlike the trapping of particles between the mirror points within static magnetic bottles, in MHD turbulence with the perpendicular superdiffusion of magnetic fields (Lazarian & Vishniac [21]; Eyink et al. [11]), particles that follow turbulent magnetic field lines also undergo the perpendicular superdiffusion and thus cannot be spatially trapped (Xu & Yan 2013; Lazarian & Yan [24]; Hu et al. 2022). During the perpendicular superdiffusion, particles always encounter different mirrors, rather than bouncing back and forth between two mirror points. The corresponding diffusion in the direction parallel to the magnetic field is termed mirror diffusion in Lazarian & Xu [22].

Illustration for the ribbon width

Figure 3. Illustration for the ribbon width. The shaded regions, after projection along the line of sight, correspond to the ENA flux distribution transverse to the ribbon structure in the sky. The width WW corresponds to the region in the outer heliosheath where pickup ions undergo the slow mirror diffusion in interstellar turbulent magnetic fields. The misalignment between red and green shaded regions shows the additional broadening of the ribbon ΔW\Delta W due to the field line wandering caused by Alfvénic modes over a radial distance ∼λ\sim\lambda.

It is known that the diffusion arising from pitch-angle scattering faces the so-called 90° problem, that is, a particle cannot be scattered through 90° (i.e., μ=0\mu= 0) according to the quasi-linear theory (Jokipii [19]) (see Figure 2), leading to an infinitely large parallel mean free path (however, see, e.g., Xu et al. (2016); Xu & Lazarian (2018) for gyroresonant scattering with turbulence-broadened resonance). The mirror diffusion does not depend on the pitch angle scattering. For a particle with a given μ\mu, the step size of its random walk along the magnetic field is the size of the mirrors that are most effective in reflecting the particle (see § 3). Therefore, its spatial diffusion coefficient along the turbulent field line is Eqs. (14) and (15) (Lazarian & Xu [22])

D∥(μ)≈vμk−1=vL(δBfB0)−4μ9,μrg<μ<μc,(23)D_{\parallel}(\mu) \approx v\mu k^{-1} = vL\left(\frac{\delta B_f}{B_0}\right)^{-4}\mu^9,\quad\mu_{rg}<\mu<\mu_c, \tag*{(23)}

which decreases drastically with decreasing μ\mu. For μ<μrg\mu<\mu_{rg}, there is

D∥(μ)≈vμrg,μ<μrg.(24)D_{\parallel}(\mu) \approx v\mu r_g,\quad\mu<\mu_{rg}. \tag*{(24)}

Near the region with B⋅r=0B\cdot r=0, where rr is the radial line of sight direction, the pickup ions from neutral solar wind (Zirnstein et al. [56]) are subject to the mirroring effect of the mean field, provided that the turbulent magnetic fluctuations are sufficiently low for the mirroring to dominate over the turbulent scattering (see § 2). The pitch-angle distribution of the pickup ions is expected to become more anisotropic toward the mirror point, with an excess at large pitch angles. The pickup ions accumulated near the mirror point of the mean field with μ<μc\mu<\mu_c and the pickup ions with initial μ\mu’s less than μc\mu_c can interact with the mirrors in turbulent magnetic fields (see § 3) and undergo the mirror diffusion. The largest diffusion distance over their ∼1\sim1 year lifetime (Giacalone & Jokipii 2015) can be estimated as

SD(μ)=2D∥(μ)tS_D(\mu)=\sqrt{2D_{\parallel}(\mu)t}
≈38 au(δBf/B00.04)−2(L100 au)12\approx38\ \mathrm{au}\left(\frac{\delta B_f/B_0}{0.04}\right)^{-2}\left(\frac{L}{100\ \mathrm{au}}\right)^{\frac{1}{2}}
(v438 km s−1)12(μ0.18)32(t1 year)12.(25)\left(\frac{v}{438\ \mathrm{km\ s}^{-1}}\right)^{\frac{1}{2}}\left(\frac{\mu}{0.18}\right)^{\frac{3}{2}}\left(\frac{t}{1\ \mathrm{year}}\right)^{\frac{1}{2}}. \tag*{(25)}

With the strong dependence of SD(μ)S_D(\mu) on μ\mu, the pickup ions with smaller μ\mu’s have much more suppressed diffusion and are accumulated near the point with B⋅r=0B\cdot r=0, causing the more enhanced ENA flux toward the center of the ribbon. With Γm≪Γtm\Gamma_m\ll\Gamma_{\rm tm} (see § 3), the effect of the mean field mirror on reflecting particles away from the mirror point (Zirnstein et al. [54]) is insignificant.

The width of the ribbon is determined by μc\mu_c,

W≈2[90∘−cos⁡−1(μc)].(26)W\approx2[90^\circ-\cos^{-1}(\mu_c)]. \tag*{(26)}

With μc\mu_c approximately proportional to (δBf/B0)4/11(\delta B_f/B_0)^{4/11} (Eq. (21)), WW increases with δBf\delta B_f. For μc=0.18\mu_c=0.18 in the case with δBf/B0=0.04\delta B_f/B_0=0.04 (see Figure 2), the corresponding W≈21∘W\approx21^\circ is consistent with the observed ribbon width of ∼20∘\sim20^\circ at ∼1\sim1 keV (Schwadron et al. [36]). The observed ribbon width provides a constraint on the maximum δBf\delta B_f.

As shown in Figure 3, the shaded region indicates the ribbon width WW. The pickup ions with initial μ\mu’s smaller than μc\mu_c are subject to the mirroring of the interstellar turbulent magnetic fields and undergo the μ\mu-dependent mirror diffusion, resulting in the enhanced pickup-ion intensity. The pickup ions with larger initial μ\mu’s are scattered by interstellar turbulent magnetic fields. The diffusion associated with scattering is much faster than the mirror diffusion (Lazarian & Xu [22]), and the particles lose their memory of the initial pitch angles. Therefore, the pickup ions with initial μ\mu’s larger than μc\mu_c, after being neutralized, are unlikely to return to the inner heliosphere to be observed by IBEX. We note that as δBf\delta B_f of fast modes may change with the distance from the heliopause, WW corresponding to different distances beyond the heliopause can be different. Future observations by Interstellar Mapping and Acceleration Probe (IMAP, Schwadron et al. [34]) that will resolve the substructure of the ribbon might provide more information on the variation of WW along the line of sight to test our theory.

The effect of field line wandering on the ribbon

The confinement of pickup ions near the region with B⋅r=0B\cdot r=0 is attributed to their mirroring interaction with the compressible modes of MHD turbulence on length scales less than kc−1k_c^{-1} (Eq. (22)). The incompressible Alfvénic modes do not contribute to mirroring and have a negligible contribution to scattering due to their anisotropy, but they dominate the wandering of turbulent magnetic field lines (Lazarian & Vishniac [21]). Next we will discuss the effect of field line wandering on broadening of the ribbon and distortion of the ribbon structure.

The ribbon source region stretches out from the heliopause to a radial distance ∼100\sim100 au that is determined by the mean free path λ\lambda of neutral solar wind atoms beyond the heliopause (Heerikhuisen et al. [17]). We extrapolate the observed magnetic fluctuation δBobs≈0.3 μG\delta B_{\rm obs}\approx0.3\ \mu\mathrm{G} at length scale lobs≈20l_{\rm obs}\approx20 au by Voyager 1 in the VLISM (Lee & Lee [25]) by following the Kolmogorov scaling (Burlaga et al. [4]; Lee & Lee [25]) and find the magnetic fluctuation at length scale λ\lambda,

δBλ=δBobs(lobsλ)−13\delta B_{\lambda}=\delta B_{\rm obs}\left(\frac{l_{\rm obs}}{\lambda}\right)^{-\frac{1}{3}}
≈0.5 μG(δBobs0.3 μG)(lobs20 au)−13(λ100 au)13.(27)\approx0.5\ \mu\mathrm{G}\left(\frac{\delta B_{\rm obs}}{0.3\ \mu\mathrm{G}}\right)\left(\frac{l_{\rm obs}}{20\ \mathrm{au}}\right)^{-\frac{1}{3}}\left(\frac{\lambda}{100\ \mathrm{au}}\right)^{\frac{1}{3}}. \tag*{(27)}

We assume that the measured magnetic fluctuation is mainly induced by Alfvénic modes based on the observed Kolmogorov magnetic energy spectrum. Different from the mirroring effect of compressible fast and slow modes, incompressible Alfvénic modes cause wandering of magnetic field lines. The degree of wandering over a perpendicular length scale λ\lambda is determined by the corresponding magnetic fluctuation, with

tan⁡θλ=δBλB0.(28)\tan\theta_{\lambda}=\frac{\delta B_{\lambda}}{B_0}. \tag*{(28)}

The wandering of field lines within the ribbon source causes shift of the points with B⋅r=0B\cdot r=0 and thus shift of the ribbon center (see Figure 3). Given the above estimate of δBλ\delta B_{\lambda} and B0=5 μGB_0=5\ \mu\mathrm{G}, we find that the additional broadening of the ribbon due to the field line wandering along the line of sight

ΔW=tan⁡−1(δBλB0)≈5.7∘,(29)\Delta W=\tan^{-1}\left(\frac{\delta B_{\lambda}}{B_0}\right)\approx5.7^\circ, \tag*{(29)}

which provides the upper limit for the broadening caused by Alfvénic modes. Combining the broadening effects of both Alfvénic and fast modes, we have the ribbon width (Eqs. (26) and (29))

W+ΔW≈2[90∘−cos⁡−1(μc)]+tan⁡−1(δBλB0).(30)W+\Delta W \approx2[90^\circ-\cos^{-1}(\mu_c)]+\tan^{-1}\left(\frac{\delta B_\lambda}{B_0}\right). \tag*{(30)}

Both broadening effects are illustrated in Figure 3.

In addition to the above broadening effect, the field line wandering induced by Alfvénic modes can also cause distortion of the ribbon structure across the sky. The turbulent magnetic fluctuation at LL in the VLISM is

δBL=δBobs(lobsL)−13.(31)\delta B_L=\delta B_{\mathrm{obs}}\left(\frac{l_{\mathrm{obs}}}{L}\right)^{-\frac{1}{3}}. \tag*{(31)}

Under the assumption that δBL\delta B_L is mainly associated with Alfvénic modes, the field line wandering over LL in the direction perpendicular to the radial line of sight and perpendicular to the interstellar mean magnetic field can give rise to the misalignment of ribbon centers across the sky by θL\theta_L, with

tan⁡θL=δBLB0.(32)\tan\theta_L=\frac{\delta B_L}{B_0}. \tag*{(32)}

For the ribbon structure to remain spatially coherent across the sky, θL\theta_L should be smaller than half of the average ribbon width, i.e., θL<10∘\theta_L<10^\circ. This constraint leads to

δBL<0.88 μG(tan⁡θLtan⁡(10∘))(B05 μG),(33)\delta B_L<0.88\ \mu\mathrm{G}\left(\frac{\tan\theta_L}{\tan(10^\circ)}\right)\left(\frac{B_0}{5\ \mu\mathrm{G}}\right), \tag*{(33)}

and (Eq. (31))

L<508 au(tan⁡θLtan⁡(10∘))3(δBobs0.3 μG)−3(B05 μG)3(lobs20 au).(34)L<508\ \mathrm{au}\left(\frac{\tan\theta_L}{\tan(10^\circ)}\right)^3\left(\frac{\delta B_{\mathrm{obs}}}{0.3\ \mu\mathrm{G}}\right)^{-3}\left(\frac{B_0}{5\ \mu\mathrm{G}}\right)^3\left(\frac{l_{\mathrm{obs}}}{20\ \mathrm{au}}\right). \tag*{(34)}

The above estimate is consistent with the simulation in Zirnstein et al. [53], where they found that given L=500 auL=500\ \mathrm{au}, the ribbon at large scales changes drastically, with the peak of the ribbon meandering around the sky.

In Xu & Li [44], an upper limit of LL was found to be ∼194 au\sim194\ \mathrm{au} under the ion-neutral decoupling condition in the partially ionized VLISM. It is more stringent than that imposed by the spatial coherence of the ribbon (Eq. (34)).

Discussion

Earlier studies on magnetic mirroring (Cesarsky & Kulsrud [6]) and its application to trapping of pickup ions (Giacalone & Jokipii 2015; Zirnstein et al. [53]) adopt a model of MHD turbulence as a superposition of MHD waves. In the MHD turbulence model with strong nonlinear interactions between oppositely directed wave packets, the turbulent energy cascade mainly happens in the direction perpendicular to the local magnetic field (Goldreich & Sridhar [16]; Lazarian & Vishniac [21]; Cho & Lazarian [9]). The anisotropic nature of MHD turbulence was not taken into account in Giacalone & Jokipii (2015); Zirnstein et al. [53], so it is unclear what types of magnetic fluctuations are most efficient in interacting with the ∼keV\sim\mathrm{keV} ions in their model. In nonlinear

MHD turbulence, due to the stochasticity and perpendicular superdiffusion of turbulent magnetic fields (Eyink et al. [12]; Beresnyak & Lazarian [2]), particles simultaneously undergo the perpendicular superdiffusion (Xu & Yan 2013; Lazarian & Yan [24]; Hu et al. 2022; Lazarian et al. [23]) by following the turbulent field lines and mirror reflection by magnetic compressions (Zhang & Xu [51]). The former process enables encounters of a particle with multiple magnetic mirrors and its diffusive motion along the magnetic field. Therefore, in both parallel and perpendicular directions, we expect that particles are not trapped. More detailed comparison with earlier studies will be carried out by using test particle simulations in MHD turbulence in our future work.

In addition, the relation between the characteristic MHD turbulence parameters, e.g., the energy fraction and scaling of compressible modes, the injection scale of turbulence, and the ribbon width was not clearly established in earlier studies. In this work, we quantified the range of pitch angles for effective turbulent mirroring with μ<μc\mu<\mu_c, where μc\mu_c is determined by the balance between mirroring and scattering (Eq. (21)) or by μmax⁡\mu_{\max} (Eq. (18)), the corresponding range of scales rg<k−1<kc−1r_g<k^{-1}<k_c^{-1} (Eq. (22)), and the diffusion coefficient and distance of mirror diffusion (Eqs. (23), (24), and (25)). The ribbon width depends on μc\mu_c and thus can provide a constraint on the magnetic fluctuation of compressible modes, whose energy fraction may change with the radial distance away from the heliopause. Analysis of Voyager 1 data reveals that the magnetic fluctuations in the VLISM are of mixed compressible and transverse nature (Burlaga et al. [4]; Fraternale & Pogorelov [13]). Studies by, e.g., Zank et al. [49]; Matsukiyo et al. [28] suggest that the turbulent fluctuations in the solar wind are transmitted into the VLISM as fast modes. A significant level of compressible magnetic fluctuations is still seen in the VLISM in late 2018 (Fraternale & Pogorelov [13]). Based on these studies, we assume that magnetic compressions are present in an extended region spanning over ∼100 au\sim100\ \mathrm{au} beyond the heliopause and account for the turbulent mirroring. The mirror diffusion is sufficiently slow to account for the confinement of pickup ions near the point with B⋅r=0B\cdot r=0. In addition, it preserves the initial pitch angles of particles for the ENAs to travel back to 1 au1\ \mathrm{au} and generate the ribbon.

With λ\lambda of neutral atoms increasing with increasing energies (Lindsay & Stebbings [26]), the ribbon source region extends farther into the VLISM at higher energies (Zirnstein et al. [55]). With more significant field line wandering within the ribbon source, a broader ribbon is expected at higher energies. This is consistent with the observed ribbon width that increases with increasing ENA energies (Schwadron et al. [36]). The multi-scale turbulent motions of magnetic field lines, including both turbulent compression of magnetic fields and field line wandering, can also give rise to fine structure seen in the ribbon (McComas et al. [29]) and temporal variations.

We note that the turbulent spectrum in the VLISM on scales less than ∼6×1011 cm\sim6\times10^{11}\ \mathrm{cm} becomes shallower than the Kolmogorov scaling (Lee & Lee [25]). Despite the unclear physical origin and large data uncertainty [13], the enhanced small-scale magnetic fluctuations can affect both mirroring and scattering.

We did not consider pickup-ion generated waves (e.g., [33]). Although the enhanced scattering by self-generated waves can suppress the spatial diffusion of pickup ions, strong scattering also causes significant changes in pitch angles. Pitch-angle isotropization by strong scattering near B⋅r=0B \cdot r = 0 would reduce the ribbon flux.

We did not consider the draping of the large-scale interstellar magnetic field around the heliosphere, which can also act to broaden the ribbon [53].

Summary

To investigate the origin of the IBEX ribbon, we studied the interaction of pickup ions with MHD turbulence in the VLISM. The magnetic compressions in the MHD turbulence serve as magnetic mirrors. The mirroring effect of turbulent magnetic fields dominates over that of compressed mean magnetic field due to the large turbulent magnetic field gradient at small scales. Due to the perpendicular superdiffusion of turbulent magnetic fields induced by Alfvénic modes, particles interacting with turbulent magnetic mirrors undergo mirror diffusion parallel to the magnetic field. The mirror diffusion of pickup ions is sufficiently slow, causing their accumulation near B⋅r=0B \cdot r = 0. With the initial pitch angles preserved during the mirror diffusion, after the pickup ions are neutralized, the resulting secondary ENAs can return to the heliosphere to be observed by IBEX.

The smallest pitch angle for effective mirroring is determined by the balance between turbulent mirroring and pitch-angle scattering and the amplitude of magnetic compression. The ribbon width depends on the range of pitch angles for effective turbulent mirroring (see Eq. (26)). It provides a constraint on the magnetic fluctuation of fast modes that dominate both mirroring and scattering.

The field line wandering caused by Alfvénic modes can further broaden the ribbon by shifting the locations with B⋅r=0B \cdot r = 0 within the ribbon source region (see Eq. (30)). For the ribbon structure to be coherent across the sky, under the assumption that the MHD turbulence measured by Voyager 1 in the VLISM is dominated by Alfvénic modes, we find that the injection scale of turbulence is limited up to ∼500\sim500 au (Eq. (34)). This is larger than the upper limit (∼194\sim194 au) imposed by ion-neutral decoupling condition in the partially ionized VLISM [44].

Acknowledgments

S.X. acknowledges the support for this work provided by NASA through the NASA Hubble Fellowship grant # HST-HF2-51473.001-A awarded by the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Incorporated, under NASA contract NAS5-26555. H.L. acknowledges the support by LANL LDRD program and DOE OFES program. H.L. gratefully acknowledges discussions with Fan Guo.

Software: MATLAB [27]

References

  1. [1]Beresnyak, A. 2013, ApJL, 767, L39, doi: 10.1088/2041-8205/767/2/L39DOI
  2. [2]Beresnyak, A., & Lazarian, A. 2019, Turbulence in Magnetohydrodynamics
  3. [3]Beresnyak, A., Yan, H., & Lazarian, A. 2011, ApJ, 728, 60, doi: 10.1088/0004-637X/728/1/60DOI
  4. [4]Burlaga, L. F., Florinski, V., & Ness, N. F. 2018, ApJ, 854, 20, doi: 10.3847/1538-4357/aaa45aDOI
  5. [5]Burlaga, L. F., Ness, N. F., Berdichevsky, D. B., et al. 2022, ApJ, 932, 59, doi: 10.3847/1538-4357/ac658eDOI
  6. [6]Cesarsky, C. J., & Kulsrud, R. M. 1973, ApJ, 185, 153
  7. [7]Chalov, S. V., Alexashov, D. B., McComas, D., et al. 2010, ApJL, 716, L99, doi: 10.1088/2041-8205/716/2/L99DOI
  8. [8]Chandran, B. D. G. 2000, Physical Review Letters, 85, 4656, doi: 10.1103/PhysRevLett.85.4656DOI
  9. [9]Cho, J., & Lazarian, A. 2002, Physical Review Letters, 88, 245001, doi: 10.1103/PhysRevLett.88.245001DOI
  10. [11]Eyink, G., Vishniac, E., Lalescu, C., et al. 2013, Nature, 497, 466, doi: 10.1038/nature12128DOI
  11. [12]Eyink, G. L., Lazarian, A., & Vishniac, E. T. 2011, ApJ, 743, 51, doi: 10.1088/0004-637X/743/1/51DOI
  12. [13]Fraternale, F., & Pogorelov, N. V. 2021, ApJ, 906, 75, doi: 10.3847/1538-4357/abc88aDOI
  13. [14]Giacalone, J., & Jokipii, J. R. 1999, ApJ, 520, 204, doi: 10.1086/307452DOI
  14. [16]Goldreich, P., & Sridhar, S. 1995, ApJ, 438, 763, doi: 10.1086/175121DOI
  15. [17]Heerikhuisen, J., Gamayunov, K. V., Zirnstein, E. J., & Pogorelov, N. V. 2016, ApJ, 831, 137, doi: 10.3847/0004-637X/831/2/137DOI
  16. [19]Jokipii, J. R. 1966, ApJ, 146, 480, doi: 10.1086/148912DOI
  17. [20]Kulsrud, R., & Pearce, W. P. 1969, ApJ, 156, 445, doi: 10.1086/149981DOI
  18. [21]Lazarian, A., & Vishniac, E. T. 1999, ApJ, 517, 700, doi: 10.1086/307233DOI
  19. [22]Lazarian, A., & Xu, S. 2021, ApJ, 923, 53, doi: 10.3847/1538-4357/ac2de9DOI
  20. [23]Lazarian, A., Xu, S., & Hu, Y. 2023, Frontiers in Astronomy and Space Sciences, 10, 1154760, doi: 10.3389/fspas.2023.1154760DOI
  21. [24]Lazarian, A., & Yan, H. 2014, ApJ, 784, 38, doi: 10.1088/0004-637X/784/1/38DOI
  22. [25]Lee, K. H., & Lee, L. C. 2020, ApJ, 904, 66, doi: 10.3847/1538-4357/abba20DOI
  23. [26]Lindsay, B. G., & Stebbings, R. F. 2005, Journal of Geophysical Research (Space Physics), 110, A12213, doi: 10.1029/2005JA011298DOI
  24. [27]MATLAB. 2021, MATLAB and Statistics Toolbox Release 2021b (Natick, Massachusetts: The MathWorks Inc.)
  25. [28]Matsukiyo, S., Noumi, T., Zank, G. P., Washimi, H., & Hada, T. 2020, ApJ, 888, 11, doi: 10.3847/1538-4357/ab54c9DOI
  26. [29]McComas, D. J., Allegrini, F., Bochsler, P., et al. 2009, Science, 326, 959, doi: 10.1126/science.1180906DOI
  27. [30]Ocker, S. K., Cordes, J. M., Chatterjee, S., et al. 2021, Nature Astronomy, 5, 761, doi: 10.1038/s41550-021-01363-7DOI
  28. [31]Pogorelov, N. V., Heerikhuisen, J., Zank, G. P., et al. 2011, ApJ, 742, 104, doi: 10.1088/0004-637X/742/2/104DOI
  29. [32]Schlickeiser, R. 2002, Cosmic Ray AstrophysicsDOI
  30. [33]Schwadron, N. A., & McComas, D. J. 2013, ApJ, 764, 92, doi: 10.1088/0004-637X/764/1/92DOI
  31. [34]Schwadron, N. A., Opher, M., Kasper, J., et al. 2016, in Journal of Physics Conference Series, Vol. 767, Journal of Physics Conference Series, 012025, doi: 10.1088/1742-6596/767/1/012025DOI
  32. [35]Schwadron, N. A., Bzowski, M., Crew, G. B., et al. 2009, Science, 326, 966, doi: 10.1126/science.1180986DOI
  33. [36]Schwadron, N. A., Allegrini, F., Bzowski, M., et al. 2011a, ApJ, 731, 56, doi: 10.1088/0004-637X/731/1/56DOI
  34. [38]Slavin, J. D., & Frisch, P. C. 2008, A&A, 491, 53, doi: 10.1051/0004-6361:20078101DOI
  35. [39]Swaczyna, P., McComas, D. J., Zirnstein, E. J., et al. 2020, ApJ, 903, 48, doi: 10.3847/1538-4357/abb80aDOI
  36. [40]Voelk, H. J. 1975, Reviews of Geophysics and Space Physics, 13, 547, doi: 10.1029/RG013i004p00547
  37. [41]Xu, S. 2021, ApJ, 922, 264, doi: 10.3847/1538-4357/ac2d8fDOI
  38. [44]Xu, S., & Li, H. 2022, ApJL, 941, L19, doi: 10.3847/2041-8213/acal43
  39. [47]Yan, H., & Lazarian, A. 2002, Physical Review Letters, 89, B1102+, doi: 10.1103/PhysRevLett.89.281102DOI
  40. [49]Zank, G. P., Du, S., & Hunana, P. 2017, ApJ, 842, 114, doi: 10.3847/1538-4357/aa7685DOI
  41. [50]Zank, G. P., Nakanotani, M., & Webb, G. M. 2019, ApJ, 887, 116, doi: 10.3847/1538-4357/ab528cDOI
  42. [51]Zhang, C., & Xu, S. 2023, submitted
  43. [52]Zhao, L. L., Zank, G. P., & Burlaga, L. F. 2020, ApJ, 900, 166, doi: 10.3847/1538-4357/ababa2DOI
  44. [53]Zirnstein, E. J., Giacalone, J., Kumar, R., et al. 2020, ApJ, 888, 29, doi: 10.3847/1538-4357/ab594dDOI
  45. [54]Zirnstein, E. J., Heerikhuisen, J., & Dayeh, M. A. 2018, ApJ, 855, 30, doi: 10.3847/1538-4357/aaaf6dDOI
  46. [55]Zirnstein, E. J., Heerikhuisen, J., Funsten, H. O., et al. 2016, ApJL, 818, L18, doi: 10.3847/2041-8205/818/1/L18DOI
  47. [56]Zirnstein, E. J., McComas, D. J., Schwadron, N. A., et al. 2019, ApJ, 876, 92, doi: 10.3847/1538-4357/ab15d6DOI

Paper details

vlnc:W_ecx8q93m
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 8

downloads · 1

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