An Aligned Very-Low-Mass Star Orbiting an M dwarf and Obliquity Patterns Across Giant Planets, Brown Dwarfs, and Binary Stars

Abstract: Stellar obliquity serves as a key diagnostic for tracing the dynamical evolution of bound systems-from giant planets and brown dwarfs to stellar binaries-revealing whether these diverse populations share analogous histories. Here, we report the first obliquity measurement for a double M dwarf system, determined via the Rossiter-McLaughlin effect. The spin axis of the primary star, TOI-5375 ($M_\ast=0.62\pm0.02\,M_\odot$), is well aligned with the orbit of its low-mass stellar companion ($M_c=84.8\pm1.5\, M_J$, $\rm P=1.72\,days$) with a projected obliquity of $λ=-13.5_{-13.8}<sup>{+12.4}\,<sup>{\circ}$ and a true 3D obliquity of $ψ=37.5_{-13.4}<sup>{+10.6}\,<sup>{\circ}$. The result indicates that the system either formed with a primordially aligned configuration or has undergone tidal realignment. We further investigate obliquity patterns across giant planets, brown dwarfs and binary stars. It turns out that a few obliquity trends observed in giant planets also tentatively exhibit in the latter two higher-mass populations: 1) well-aligned orbits are preferentially found around cooler host stars ($T_{\rm eff}\leq 6250\,K$); 2) wide-orbit ($a/R_\ast\geq 10$) companions are predominantly aligned; 3) no significant correlation shows up between obliquity and orbital eccentricity in any of the companion classes. By modeling $|λ|$ with a two-component Gaussian distribution, we find that the low-$|λ|$ components of binary stars and brown dwarfs are more concentrated near zero than giant planets while the high-$|λ|$ components of brown dwarfs and binaries remain unclear due to the small sample size.

• The paper presents the first Rossiter–McLaughlin (RM) measurement of a double M dwarf system, TOI-5375, revealing a projected obliquity of -13.5+12.4−13.8∘ and true obliquity of 37.5+10.6−13.4∘. This makes the companion just above the hydrogen-burning limit as a very-low-mass M dwarf.
• The study extends the analysis to 205 systems, finding tentative evidence that alignment trends and patterns similar to hot Jupiters extend to higher-mass companions, suggesting a correlation with efficient tidal realignment for brown dwarfs and binaries, and finding that alignment peaks at larger intervals for binaries and brown dwarfs than for planets.
• The authors note the system can be explained with either a primordial quenched disk or a realigned spindle-disk on a 46 Myr timescale.

Overview

This paper reports the first Rossiter–McLaughlin (RM) measurement of a double M dwarf system, TOI-5375, and uses the result as the anchor for a comparative analysis of spin–orbit alignment across three companion populations: giant planets, brown dwarfs, and stellar binaries. The host star ($M_\ast = 0.62 \pm 0.02\,M_\odot$, $T_{\rm eff} = 3897$ K) is orbited every 1.72 days by a companion of $M_c = 84.8 \pm 1.5\,M_J$ — just above the hydrogen-burning limit — making it a very-low-mass M dwarf rather than a brown dwarf. The joint fit yields a projected obliquity $\lambda = -13.5^{+12.4}_{-13.8}\,^{\circ}$ and, after deprojection using the stellar inclination, a true obliquity $\psi = 37.5^{+10.6}_{-13.4}\,^{\circ}$. The authors conservatively classify the system as aligned, though the large uncertainties on both quantities are dominated by the low signal-to-noise ratio (SNR) of the RM data on this faint, rapidly rotating target.

Observations and template construction

Two transits were observed on UT 2025 February 10 and 17 with SPIRou at CFHT ($R \approx 75{,}000$, 0.98–2.5 μm), each covering the full 1.7 hr transit plus roughly 1.5 hr of out-of-transit baseline, achieving SNRs of 42 and 49 near 1.75 μm. Radial velocities were extracted with the line-by-line (LBL) method after masking atmospheric OH lines and the oxygen fluorescence band.

A methodological contribution of the paper is the construction of a "vetted" LBL template for TOI-5375. Because the star's high rotational broadening ($v\sin i = 16.7 \pm 0.9$ km s$^{-1}$) makes it a poor match to standard SPIRou templates built from slowly rotating stars, the authors model each observed spectrum as a non-negative linear combination of 15 rotationally broadened reference templates spanning K2V–M7V, together with a two-component telluric absorption model fitted to pre-telluric-correction spectra. The resulting template yields $v_{\rm sys} = -63.5 \pm 4.0$ km s$^{-1}$ and $T_{\rm eff} = 3897$0 km s$T_{\rm eff} = 3897$1, consistent with prior measurements. The final RVs have median per-point uncertainties of 63 and 57 m s$T_{\rm eff} = 3897$2. No significant correlations were found between the RVs and activity indicators (CRX, dTemp), arguing against spot-induced rotational RV contamination of the RM signal.

Joint analysis and system parameters

The photometry (TESS Sectors 40, 47, 50, 63, 73, 74 at 2-minute cadence; ground-based Bessell-I data from Red Buttes Observatory), literature RVs from HPF and HARPS-N, and the two nights of SPIRou RM data were jointly modeled in Allesfitter, with the RM computed via flux-weighted velocities in ellc. A GP with a Matérn-3/2 kernel detrended the TESS simple aperture photometry, and a contaminating flux ratio was fitted to account for dilution. The companion's own flux was assumed negligible in photometry and spectral lines based on its mass ratio and luminosity–mass relation.

Key derived parameters include $T_{\rm eff} = 3897$3, mass ratio $T_{\rm eff} = 3897$4, orbital eccentricity $T_{\rm eff} = 3897$5, scaled semi-major axis $T_{\rm eff} = 3897$6, and $T_{\rm eff} = 3897$7 km s$T_{\rm eff} = 3897$8 from the joint fit. The companion's stellar nature is supported independently by a secondary eclipse depth of $T_{\rm eff} = 3897$9 ppm in TESS, consistent under blackbody assumptions with an effective temperature of $M_c = 84.8 \pm 1.5\,M_J$0 K, corresponding to an M9V star.

Combining the refined photometric rotation period ($M_c = 84.8 \pm 1.5\,M_J$1 d) with the stellar radius gives $M_c = 84.8 \pm 1.5\,M_J$2 km s$M_c = 84.8 \pm 1.5\,M_J$3, which via Bayesian inference of $M_c = 84.8 \pm 1.5\,M_J$4 yields $M_c = 84.8 \pm 1.5\,M_J$5 and hence the deprojected obliquity quoted above. The authors note that the first night's SPIRou data show higher jitter and a tentative residual slope consistent with zero only at about $M_c = 84.8 \pm 1.5\,M_J$6, likely attributable to detector persistence and systematic biases in template-matching RV extraction; they explicitly caution that these data should be interpreted with care.

Evolutionary interpretation

The measured alignment admits two scenarios: primordial formation in a quiescent disk-aligned configuration, or subsequent tidal realignment. Using the Zahn (1977) formalism, the estimated realignment timescale is $M_c = 84.8 \pm 1.5\,M_J$7 Myr. Here the paper confronts an unresolved tension: the short rotation period suggests a young host (~400 Myr), while metallicity and kinematics indicate an old field star. Under either age estimate, however, the damping timescale is shorter than the system age, so sufficient time existed for realignment. This ambiguity means the observation cannot by itself distinguish between the two histories.

Comparative obliquity patterns

Assembling samples of 173 giant planets, 11 brown dwarfs, and 21 binary stars (primary-star obliquities only, from SOCat/TEPCat and recent binary compilations), the paper tests whether trends established for hot Jupiters extend to higher-mass companions:

Trend	Giant planets	Brown dwarfs	Binaries
Alignment below Kraft break ($M_c = 84.8 \pm 1.5\,M_J$8 K)	Established	Tentative	Tentative
Wide orbits ($M_c = 84.8 \pm 1.5\,M_J$9) aligned	Yes	Yes (5 systems)	Yes (12 systems)
Obliquity–eccentricity correlation	None found	None found	None found

Fitting two-component Gaussian mixture models to $\lambda = -13.5^{+12.4}_{-13.8}\,^{\circ}$0 distributions (with Monte Carlo resampling of measurement uncertainties over 1000 iterations), the low-$\lambda = -13.5^{+12.4}_{-13.8}\,^{\circ}$1 component peaks at $\lambda = -13.5^{+12.4}_{-13.8}\,^{\circ}$2 for binaries, $\lambda = -13.5^{+12.4}_{-13.8}\,^{\circ}$3 for brown dwarfs, and $\lambda = -13.5^{+12.4}_{-13.8}\,^{\circ}$4 for giant planets. The authors interpret this ordering as evidence that binaries and brown dwarfs were either born more aligned than giant planets or experienced more efficient tidal realignment. All brown dwarf and binary systems lie above the empirical mass-ratio boundary $\lambda = -13.5^{+12.4}_{-13.8}\,^{\circ}$5 proposed for aligned high-mass-ratio systems, consistent with the scaling $\lambda = -13.5^{+12.4}_{-13.8}\,^{\circ}$6, though a primordial interpretation (isolated formation versus dynamical instability in compact configurations) remains viable.

One qualitative distinction is emphasized: retrograde orbits ($\lambda = -13.5^{+12.4}_{-13.8}\,^{\circ}$7) appear among giant planets but are essentially absent in sub-stellar companions and stellar binaries. The misaligned systems that do exist among brown dwarfs (CoRoT-3, XO-3) and binaries (DI Herculis, CV Velorum) orbit hot stars above the Kraft break, mirroring the hot-Jupiter pattern — but with only four such systems total, the authors state plainly that no robust conclusion is possible. DI Herculis, notably eccentric and misaligned despite $\lambda = -13.5^{+12.4}_{-13.8}\,^{\circ}$8, is attributed to perturbations from a tertiary companion.

Limitations and open questions

Several caveats bear directly on the strength of the results. The obliquity constraints for TOI-5375 itself are limited by the faintness of the host and the modest SNR of two transit visits; the first night's data carry elevated jitter and possible template-matching systematics, and additional SPIRou visits or larger telescopes would be needed to tighten $\lambda = -13.5^{+12.4}_{-13.8}\,^{\circ}$9 and $\psi = 37.5^{+10.6}_{-13.4}\,^{\circ}$0. The Gaussian mixture decomposition depends on the choice of priors and models, and the high-$\psi = 37.5^{+10.6}_{-13.4}\,^{\circ}$1 components for brown dwarfs and binaries are not statistically robust given sample sizes of 11 and 21. The absence of an obliquity–eccentricity correlation may itself be an observational bias, since obliquity measurements favor short-period systems subject to tidal circularization. Finally, the hierarchical Bayesian treatment applied to deprojected planetary obliquities elsewhere has not yet been extended to brown dwarf and binary distributions — a specific gap this paper identifies but does not fill.

Conclusion

TOI-5375 provides the first obliquity constraint for an M-dwarf binary, showing a projected alignment of $\psi = 37.5^{+10.6}_{-13.4}\,^{\circ}$2 and a true obliquity of $\psi = 37.5^{+10.6}_{-13.4}\,^{\circ}$3, consistent either with primordial alignment or tidal realignment on a ~46 Myr timescale. Extending the comparison to 205 systems across three companion classes, the paper finds tentative evidence that the cool-star alignment trend, the alignment of wide-orbit companions, and the lack of an eccentricity correlation hold across masses up to the stellar regime, while the concentration of binary and brown dwarf low-obliquity components nearer zero than giant planets hints at mass-dependent formation or tidal physics. Whether these patterns survive as the brown dwarf and binary samples grow beyond a few dozen systems remains the central open question raised by this work.