{"id":"45269a3d-f0df-4b2b-b738-633a466a627e","arxiv_id":"2412.13763","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"The authors derive from a microscopic t-J model a mesoscale transport theory in which charge currents exert transverse spin-transfer torques, enabling current-driven domain-wall motion in d-wave altermagnets.","lead":"This paper builds a mathematical model of how electrons flow through a d-wave altermagnet, a magnetic material with no net magnetization but spin-split bands. It predicts that an electric current can push magnetic domain walls in a direction forbidden in ordinary antiferromagnets, due to new transverse couplings between charge currents and spin texture gradients.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (19) appears to contain a wrong coupling constant: the surviving torque from Eq. (18) is proportional to g_4^xy, not g_4^AM, which changes the microscopic definition and spin-splitting scaling of η_FL while preserving the qualitative transverse form.","rationale":"The reader's weakest assumption was the slave-boson adiabatic pinning in Appendix A1. While that is a reasonable foundational concern, the most load-bearing problem I found is more concrete and closer to the central claim: Eq. (19), which directly yields the flagship torque (20), does not follow algebraically from Eq. (18) as written. My term-by-term simplification shows that the g_4^AM contributions cancel and the surviving torque is proportional to g_4^xy, the conventional spin-exchange coupling, rather than g_4^AM, the altermagnetic anisotropy coupling. This conclusion is robust: the g_4^xy term is linear in the spin splitting (through 1/m_s_xy), matching the abstract's statement that η_FL is proportional to the spin splitting, whereas the printed Eq. (19) is quadratic. The error is therefore likely a typo or an index slip, but it changes the microscopic prediction for η_FL and the associated domain-wall force. Because the transverse derivative form (j_x^e ∂_y + j_y^e ∂_x)n survives with either g_4^xy or g_4^AM, and because the torque still vanishes in conventional antiferromagnets where 1/m_s_xy → 0, the central conceptual claim is not destroyed. The dissipative part η_DL is also still an assumption rather than a derivation. These are addressable but not cosmetic issues, so the reader's CONDITIONAL verdict remains appropriate. I therefore recommend no change to the verdict, while flagging the specific algebraic correction needed in Eq. (19).","tokens_in":25172,"tokens_out":19738,"duration_ms":154992,"concrete_test":"Recompute Eq. (19) by direct substitution: take Eq. (18), insert J_i^κ = (m_parallel/m_s_xy)(ℏ/2e) σ_x|κβ n_i j_β^e, impose n·m = 0 and uniform j, and collect terms. Verify that the surviving g_4^xy term has coefficient (4 m_parallel^2/m_s_xy)(g_4^xy s^2/ℏ^2)(ℏ/2e) and that all g_4^AM terms cancel after dropping ∂m contributions. If the paper's g_4^AM instead appears, recheck the index contractions in the sixth and seventh terms of Eq. (18).","verdict_should_be":"UNCHANGED","load_bearing_attack":"The derivation of the central spin-transfer torque, Eq. (19), does not survive a term-by-term simplification of the preceding Eq. (18). Substituting the spin-splitter relation J_i^κ = C σ_x|κβ n_i j_β^e, with C = (m_parallel/m_s_xy)(ℏ/2e), and using n·m = 0 with uniform j, the g_4^AM terms cancel in pairs: the first term cancels against the m-derivative part of the sixth term, and the seventh term is a ∂m contribution that the authors explicitly discard. The g_4^xy terms give, after the fourth term cancels the third, a single surviving contribution (4 m_parallel^2/m_s_xy)(g_4^xy s^2/ℏ^2)(ℏ/2e)(j_x^e ∂_y + j_y^e ∂_x)n_i. The paper instead reports g_4^AM in Eq. (19). Since g_4^AM ∝ (t−t') and 1/m_s_xy ∝ (t'−t), the printed Eq. (19) is quadratic in the spin splitting, directly contradicting the sentence in the same paragraph that quadratic terms 'via g_4^AM and 1/m_s_xy' are disregarded, and also contradicting the abstract's statement that η_FL is proportional to the d-wave spin splitting. With the correct g_4^xy coefficient, the torque is linear in (t−t') and the qualitative form (and its vanishing as t'→t) is unchanged, so the conceptual claim survives but the specific microscopic expression for η_FL must be corrected.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript develops an effective long-wavelength theory for itinerant carriers in d-wave altermagnets, starting from a slave-boson representation of a t-J model on a rutile lattice. The central results are a spin-polarized mass tensor with off-diagonal elements proportional to the Néel order, which yields the spin-splitter effect, and spin-transfer torques and spin-pumping currents exhibiting a transverse derivative structure, τ ∝ (j_x^e ∂_y + j_y^e ∂_x) n. The paper also discusses consequences for current-driven domain-wall motion and the effect of elastic strain on the transport coefficients.","tokens_in":25582,"tokens_out":17413,"duration_ms":140438,"significance":"If the central results hold, the paper identifies a new, symmetry-required spin-transfer torque in d-wave altermagnets that couples charge currents to magnetization gradients in a transverse, anisotropic combination not present in conventional antiferromagnets. The derivation is self-contained within a microscopic model and produces falsifiable predictions, such as transverse-current-driven domain-wall motion in the same geometry where the torque vanishes in bipartite antiferromagnets. The main limitations are the load-bearing mean-field slave-boson ansatz, the partly asserted passage from Eq. (18) to Eq. (19), and the phenomenological status of the dissipative torque. Nevertheless, the qualitative form of Eq. (20) is a valuable contribution to altermagnetic spintronics.","major_comments":[{"comment":"The reduction from Eq. (18) to Eq. (19) is not shown, and the cancellations involved are nontrivial. After substituting the spin-splitter relation J_κ = (m_∥/m_{sxy})(ℏ/2e) n σ_x j^e and using n·m=0 with uniform j^e, the g_4^AM terms in Eq. (18) cancel pairwise, and the surviving g_4^xy contribution emerges only after a partial cancellation among the third, fourth, and fifth terms of that equation. The authors should present this simplification explicitly, because the microscopic expression for η_FL is a central result. In addition, the factor s^2 in Eq. (19) is unexplained: it does not appear in the spin-splitter relation (15) or in the energy functional (16), and its presence changes the spin-density scaling of η_FL. Note that, as printed, Eq. (19) contains g_4^xy (not g_4^AM), so it is linear in the spin splitting once 1/m_{sxy} is accounted for, consistent with the abstract; however, the derivation must be made transparent and the s^2 factor either derived or corrected.","section":"IV.A, Eqs. (18)–(19)"},{"comment":"The dissipative (antidamping-like) torque is introduced by appealing to \"the usual phenomenological arguments\" rather than by deriving it from the effective Lagrangian. Since the central physical prediction in Sec. VI — transverse-current-driven domain-wall motion — relies specifically on the dissipative component (see the discussion of Fig. 2), the paper should either derive η_DL within the same microscopic framework or explicitly state that Eq. (20) is a phenomenological ansatz with η_DL as an independent parameter. As written, the abstract's claim to \"elucidate the spin-transfer response\" overstates the status of the dissipative term.","section":"IV.A, Eq. (20)"},{"comment":"The derivation of the spin-pumping currents from Onsager reciprocity is under-specified. In particular, the thermodynamic force f_m ≈ −n ∂_t n/s^2 stated after Eq. (22) does not follow directly from the LLG equations (22); the standard relation in the exchange-dominated limit involves f_m ∝ n × ∂_t n up to factors of the spin density and susceptibility. Please provide the derivation and clarify the definitions of s and f_m, since the pumped currents in Eq. (24) depend sensitively on these factors. The qualitative reciprocal relation between Eqs. (20) and (24) is plausible, but the quantitative coefficients need support.","section":"IV.B, Eq. (24)"}],"minor_comments":[{"comment":"The symbol s is used both for the sublattice spin magnitude (Appendix A) and for the saturation spin density in the dissipative torque (Eq. (20)); please use distinct notation to avoid ambiguity.","section":"Eqs. (19)–(20) and Appendix A"},{"comment":"The sentence stating that terms \"quadratic in the spin splitting ∝ t′−t (via g_4^AM and 1/m_{sxy})\" are disregarded should be rephrased: it is the combination g_4^AM / m_{sxy} that is quadratic, not the factors individually.","section":"After Eq. (19)"},{"comment":"The neglect of valley-off-diagonal terms is justified by an energy-cost argument; a quantitative estimate, even an order-of-magnitude one, would make the long-wavelength expansion more convincing.","section":"Sec. II, footnote 31"},{"comment":"The strain dependence of t−t′ is a useful and clear result; however, the discussion after Eq. (27) could be shortened and the conditions for shear-strain enhancement stated more precisely (e.g., relative magnitude of |t0−t′0| vs. |t0+t′0|).","section":"Sec. V, Eq. (27)"},{"comment":"There are several typographical errors and formatting inconsistencies in the reference list, including \"Pys.\" in Ref. 36 and inconsistent spelling of \"Jaeschke-Ubiergo\" in Refs. 25 and 34; please harmonize the bibliography.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The manuscript overlaps with works acknowledged by the authors (Refs. 43 and 44); its differential contribution is the microscopic t-J derivation and the explicit reciprocal spin-pumping result. The route to acceptance should focus on making the derivation of Eq. (19) fully transparent and on clearly labeling the dissipative torque as phenomenological or deriving it microscopically. The specific stress-test concern that Eq. (19) contains g_4^AM is not supported by the text as provided — Eq. (19) contains g_4^xy — but the unexplained factor s^2 remains a genuine point that needs clarification."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a serious analytic paper and worth a referee. The central result — a spin-transfer torque and reciprocal spin pumping with transverse derivative combinations (j_x^e ∂_y + j_y^e ∂_x) — is physically interesting and derived from a genuine microscopic starting point, not just symmetry phenomenology. The slave-boson derivation in the appendices is extensive, and the off-diagonal spin-polarized mass tensor in Eq. (5) is a genuinely new piece of structure. The strain analysis is a nice extra; the note added is honest about overlap with Vakili et al. and Kokkeler et al.\n\nI checked the stress-test note about Eq. (19) against the text. It does not land. As printed, Eq. (19) uses g4^xy, not g4^AM. The coefficient is (m∥²/m_s^xy) g4^xy, and since 1/m_s^xy ∝ (t'−t), this is linear in the d-wave spin splitting. The sentence in the same paragraph says quadratic terms via g4^AM and 1/m_s^xy are disregarded, which is consistent: g4^AM ∝ (t−t'), so g4^AM/m_s^xy would be quadratic. The concern's own conclusion — that the qualitative form survives — is right, but the claimed correction to the microscopic expression is unnecessary.\n\nSoft spots, in proportion. The dissipative (antidamping) torque is introduced by the usual phenomenological argument, not derived; that is a limitation, but a standard one. Valley-off-diagonal terms are dropped at subleading order with a short justification. There is no numerical check of the algebra, which matters because the appendix is long, but I did not find an obvious error in the main chain. The load-bearing assumption is the mean-field pinning of the vector slave-boson fields to the sublattice spin densities (adiabatic J_sd → ∞). If that fails, Eq. (3) and everything after inherits the problem. That is the main thing I would ask the authors to discuss more carefully.\n\nWho is this for? Condensed-matter theorists working on altermagnet transport and spin-texture dynamics. They will use the effective Lagrangian and the transverse torque form; the lack of numerics means experimentalists won't rely on the specific coefficients yet. It deserves a serious referee, and with a revision that strengthens the dissipative-torque derivation and the adiabaticity discussion, it could be a solid contribution.","headline":"A serious microscopic derivation of transverse spin-transfer torques in d-wave altermagnets; the novelty is real but narrower than claimed, and the advertised stress-test red flag on Eq. (19) does not hold up.","tokens_in":26090,"tokens_out":3769,"would_cite":true,"duration_ms":30995,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper derives the altermagnetic spin-transfer torque $\\tau = \\eta_{\\rm FL}(j^e_x\\partial_y + j^e_y\\partial_x)n + \\eta_{\\rm DL} s\\, n\\times(j^e_x\\partial_y + j^e_y\\partial_x)n$ and its reciprocal spin pumping from a microscopic model.","keywords":["altermagnetism","d-wave altermagnet","spin-transfer torque","spin pumping","spin-splitter effect","slave-boson t-J model","domain-wall motion","mesoscale transport theory"],"falsifier":"Measure the current-direction dependence of domain-wall motion in a rutile d-wave altermagnet such as RuO$_2$. For a wall whose N\\'eel order varies only along $x$, Eq. (20) predicts a torque proportional to $j^e_y\\,\\partial_x n$, so the wall should move when the current flows along $y$ (transverse to the wall normal) and stay put for current along $x$; a conventional antiferromagnet shows the opposite pattern. Observing this 90-degree switching rule, or finding its absence, would settle whether the central claim is correct.","tokens_in":24969,"feed_emoji":"🧲","tokens_out":13224,"duration_ms":112052,"temperature":0.7,"pith_summary":"Using a slave-boson formulation of the $t$--$J$ model, this paper derives a mesoscale transport theory for itinerant carriers in a d-wave altermagnet, a compensated magnet whose opposite-spin sublattices are linked by a fourfold rotation and whose band structure shows momentum-dependent d-wave spin splitting. Its central result is the altermagnetic spin-transfer torque $\\tau^{\\rm AM}_{m,\\mathrm{ST}} = \\eta_{\\rm FL}(j^e_x\\partial_y+j^e_y\\partial_x)n + \\eta_{\\rm DL} s\\, n\\times(j^e_x\\partial_y+j^e_y\\partial_x)n$, whose crossed combination of current and spatial derivative exists only for d-wave altermagnets. The same crossed structure appears in the reciprocal spin-pumping current. If this is right, charge currents flowing transverse to a domain wall's normal can move the wall, and the spin-splitter effect follows from an off-diagonal spin-polarized mass tensor. The paper also shows that strain renormalizes the spin splitting and hence tunes these effects.","feed_headline":"Altermagnets gain a transverse spin-transfer torque","feed_subtitle":"Microscopic t-J derivation gives a torque ∝ j_x∂_y + j_y∂_x: transverse currents move domain walls.","key_machinery":"The load-bearing object is the effective Euclidean Lagrangian (3), obtained from a spin-rotation-invariant slave-boson mean-field expansion of the single-band $t$--$J$ model on a rutile lattice. Its decisive ingredient is the spin-polarized mass tensor $(M^s)^{-1}$, whose off-diagonal elements $\\propto n/m^s_{xy}$ parametrize the transverse intertwining of charge and spin currents. Together with the emergent gauge fields $A_x = g^{xy}_4\\, n\\,\\partial_x n + g^{\\rm AM}_4\\, m\\,\\partial_y n$ and $A_y = g^{xy}_4\\, n\\,\\partial_y n + g^{\\rm AM}_4\\, m\\,\\partial_x n$, the energy term $E_{\\rm ST} = -\\frac{4m_\\parallel}{\\hbar^2}(J_x\\cdot A_x + J_y\\cdot A_y)$ generates the spin-transfer torque through the Poisson-bracket dynamics of the N\\'eel order and the macroscopic magnetization.","core_discovery":"The paper's central claim is that the long-wavelength theory of itinerant carriers in a d-wave altermagnet contains a spin-polarized diffusive term with off-diagonal mass-tensor elements proportional to the N\\'eel order, and that this term couples charge currents to spatial gradients of the order parameter in a crossed way. Combining the spin-splitter relation $J_\\kappa = \\frac{m_\\parallel}{m^s_{xy}} \\frac{\\hbar}{2e}\\, n\\, \\sigma^x_{\\kappa\\sigma} j^e_\\sigma$ with the gauge-field coupling in the energy functional yields the spin-transfer torque of Eq. (20), $\\tau^{\\rm AM}_{m,\\mathrm{ST}} = \\eta_{\\rm FL}(j^e_x\\partial_y + j^e_y\\partial_x)n + \\eta_{\\rm DL} s\\, n\\times(j^e_x\\partial_y + j^e_y\\partial_x)n$, with $\\eta_{\\rm FL}$ and $\\eta_{\\rm DL}$ proportional to the d-wave spin splitting. The Onsager-reciprocal spin-pumping current has the same crossed structure. The same framework produces the spin-splitter effect as a real-space off-diagonal spin conductivity and predicts that elastic strain modifies these effects by renormalizing the spin-splitting parameter $t-t'$.","pith_inferences":["If Eq. (20) is confirmed, the 90-degree switching rule for current-driven domain-wall motion becomes a practical diagnostic for d-wave altermagnetism, and devices could steer textures by rotating the current direction instead of changing its magnitude.","Because the crossed derivative combination is the real-space fingerprint of d-wave symmetry, g- and i-wave altermagnets would plausibly show higher-order derivative combinations of the same type; deriving them would give a testable symmetry hierarchy, though this goes beyond the paper.","Strain tuning suggests that placing an altermagnet on a piezoelectric substrate could modulate the spin-splitter efficiency and the spin-transfer torque in situ; this is an extrapolation from the paper's strain analysis, not a claim it makes.","Since the derivation is presented at the $\\Gamma$-valley and the $Z$-valley flips the sign of $z$-derivative terms, quasi-two-dimensional devices that average over valleys could partially cancel or enhance the predicted torque; the paper notes the valley-independence only for in-plane physics."],"forward_implications":["The spin-splitter effect follows from the off-diagonal spin-polarized mass tensor: an electric field generates a spin current polarized along the N\\'eel order and flowing transversely, with no spin-orbit coupling needed.","Altermagnetic spin-transfer torques contain a fieldlike term linear in the N\\'eel order, allowed because the sublattice symmetry is broken, plus a dissipative term transverse to it; both vanish when altermagnetism is switched off.","Spin pumping is Onsager reciprocal to the torque and carries the same crossed $j^e_x\\partial_y + j^e_y\\partial_x$ structure, so a moving texture pumps spin currents that have no counterpart in bipartite antiferromagnets.","Domain walls and skyrmions experience a spin-transfer force with a distinct angular dependence; in particular, the reactive force on a skyrmion is isotropic in the current direction, unlike the transverse dependence in ferrimagnets and magnetoelectric antiferromagnets.","Elastic strain renormalizes the spin-splitting parameter $t-t'$: shear strain can enhance the altermagnetic transport effects when the unstrained hoppings have the same sign, and merely renormalizes them when the signs are opposite."],"supporting_citations":[{"why":"Supplies the spin-rotation-invariant slave-boson representation of the t–J model on which the effective Lagrangian (3) is built.","marker":"27–30"},{"why":"Predicts the spin-splitter effect that the paper reproduces as a real-space off-diagonal spin conductivity.","marker":"21"},{"why":"Provides the RuO2 experimental platform for the spin-splitter effect that the transport theory targets.","marker":"22"},{"why":"Gives the continuum d-wave exchange model and stability arguments for domain walls and skyrmions used in the spin-texture discussion.","marker":"25"},{"why":"Establishes the slave-boson mean-field expansion for itinerant carriers in a magnetic background and the adiabatic J_sd → ∞ treatment.","marker":"35"},{"why":"Supplies the effective Lagrangian for itinerant carriers in antiferromagnets that Eq. (3) reduces to in the limit of vanishing spin splitting.","marker":"36"},{"why":"Defines the conventional antiferromagnetic spin-transfer torque baseline against which the crossed altermagnetic torque is contrasted.","marker":"38"},{"why":"Independently derives an equivalent altermagnetic spin-transfer torque on phenomenological grounds, noted by the authors as complementary.","marker":"44"}],"fun_headline_variants":["Transverse spin torque emerges in d-wave altermagnets","d-wave altermagnets get a cross-gradient spin torque","Crossed currents and gradients yield altermagnet torque","Spin-transfer goes sideways in d-wave altermagnets"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the adiabatic slave-boson mean-field assumption that the vector boson fields stay rigidly pinned to the localized sublattice spin densities, i.e. the $J_{sd}\\to\\infty$ limit; if that pinning fails, the effective Lagrangian (3) and every torque and pumping formula derived from it inherit the error.","fun_headline_variants_meta":{"raw":{"variants":["Transverse spin torque emerges in d-wave altermagnets","d-wave altermagnets get a cross-gradient spin torque","Crossed currents and gradients yield altermagnet torque","Spin-transfer goes sideways in d-wave altermagnets"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00027,"raw_usage":{"total_tokens":1663,"prompt_tokens":1019,"completion_tokens":644,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":635,"completion_tokens_details":{"reasoning_tokens":588}},"tokens_in":635,"tokens_out":644,"duration_ms":6063,"temperature":1.0,"reasoning_tokens":588,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T12:49:33.484064+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the current-direction dependence of domain-wall motion in a rutile d-wave altermagnet such as RuO$_2$. For a wall whose N\\'eel order varies only along $x$, Eq. (20) predicts a torque proportional to $j^e_y\\,\\partial_x n$, so the wall should move when the current flows along $y$ (transverse to the wall normal) and stay put for current along $x$; a conventional antiferromagnet shows the opposite pattern. Observing this 90-degree switching rule, or finding its absence, would settle whether the central claim is correct.","supporting_citations":[{"cited_title":"Gonz\\' a lez-Hern\\' a ndez, L","cited_arxiv_id":null,"evidence_quote":"Predicts the spin-splitter effect that the paper reproduces as a real-space off-diagonal spin conductivity."},{"cited_title":"Bose, N.J","cited_arxiv_id":null,"evidence_quote":"Provides the RuO2 experimental platform for the spin-splitter effect that the transport theory targets."},{"cited_title":"S mejkal, A","cited_arxiv_id":null,"evidence_quote":"Gives the continuum d-wave exchange model and stability arguments for domain walls and skyrmions used in the spin-texture discussion."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the slave-boson mean-field expansion for itinerant carriers in a magnetic background and the adiabatic J_sd → ∞ treatment."},{"cited_title":"Grimmer-ACSA1993","cited_arxiv_id":null,"evidence_quote":"Supplies the effective Lagrangian for itinerant carriers in antiferromagnets that Eq. (3) reduces to in the limit of vanishing spin splitting."},{"cited_title":"Gomonay, V.P","cited_arxiv_id":null,"evidence_quote":"Defines the conventional antiferromagnetic spin-transfer torque baseline against which the crossed altermagnetic torque is contrasted."},{"cited_title":"Kim, K.-J","cited_arxiv_id":null,"evidence_quote":"Independently derives an equivalent altermagnetic spin-transfer torque on phenomenological grounds, noted by the authors as complementary."}],"review_version":1}