{"id":"d0a5ae38-f5cf-4fc8-a2b8-2282c5e4a28e","arxiv_id":"1908.08664","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The global minima of a periodic acoustic radiation potential are controlled by the smallest-eigenvalue eigenspace of a small matrix, giving particle arrangements that are points, lines, or planes within restricted Bravais lattice classes.","lead":"This paper works out which periodic particle patterns can be made by trapping small particles with standing sound waves in a liquid. It shows that the pattern type, such as dots, lines, or sheets, follows from the smallest eigenvalue and eigenvectors of a small matrix, and it lists which crystal lattice shapes are reachable.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Bravais classification is for the wave lattice A, while the actual particle arrangement can be a finer lattice; a simple eigenvalue with hexagonal A yields a rectangular arrangement, contradicting the abstract.","rationale":"The reader's weakest assumption is exactly right. I checked the mechanics: Q(0) is correctly diagonalized, and Lemmas 2.2-2.6 correctly characterize level sets for real eigenvectors. The mathematical core of Theorem 2.1 is sound. The problem is the bridge from the level-set structure to the Bravais-class enumeration. When T_{λ,u} contains more than just s=0, the minimizer set is a union of cosets that may form a lattice finer than A. For a simple eigenvalue, T={0,1} and the union is the lattice generated by A and A1/2. This lattice can fall in a Bravais class different from A. The concrete 2D hexagonal case gives A hexagonal but minimizers rectangular primitive, directly contradicting Table 1. The paper's remark that Def. 3.1 is 'irrespective of the particular particle arrangement inside a primitive cell' shows the authors are aware of the basis but miss that the basis can change the translation lattice. This is fixable by redefining achievability for the generated lattice of minima or by explicitly distinguishing 'wave-lattice class' from 'particle-arrangement class'. Since the core linear algebra is correct and the issue is a classification gap, the conditional verdict stands.","tokens_in":13828,"tokens_out":20458,"duration_ms":186217,"concrete_test":"In 2D, take hexagonal wavevectors K = k[[1,1/2],[0,√3/2]] and choose a,b (e.g., a=1,b=1) so that λ_min(Q(0)) is simple. Construct the unit eigenvector u in the H- eigenspace with no zero entries. Using Lemma 2.4 with T={0,1}, compute the minima set S = A Z^2 ∪ A(1/2+Z^2). Show S is the lattice generated by b1=(a1+a2)/2 and b2=(a1-a2)/2, that b1·b2=0 and |b1|≠|b2|, so S is primitive rectangular. Then note that this contradicts Table 1, which says the orthorhombic class is not achievable; this uses only the paper's own theorem.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 3 (Definition 3.1, Tables 1-2) classifies achievable Bravais classes under the constraint |k1|=...=|kd|, but the objects classified are the wave lattices A=2πK^{-T}. Lemma 2.4 shows the set of minima is generally a union of cosets A(n+s/2). For a simple eigenvalue the cosets are A Z^d and A(1/2+Z^d), whose union is the index-2 superlattice generated by A and A1/2, not A itself. This finer lattice need not belong to the same Bravais class as A: if A is hexagonal in 2D, the minimizer lattice has basis (a1+a2)/2 and (a1-a2)/2, which are perpendicular with unequal lengths, i.e., a primitive rectangular (orthorhombic) lattice. Table 1 declares orthorhombic unachievable, so the abstract's statement that 'the arrangement's periodicity is limited to certain Bravais lattice classes' is unsupported. The remark after Def. 3.1 ('irrespective of the particular particle arrangement inside a primitive cell') does not address that the union of cosets can itself be a lattice of a different class; it only dismisses the arrangement inside the cell.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies periodic acoustic radiation potentials generated by superpositions of d plane waves with equal wavenumbers in a non-viscous fluid, and uses them to predict where small spherical particles are trapped. The potential is written as a quadratic form u*Q(x)u, and a unitary similarity argument reduces the position dependence to the matrix Q(0). Lemma 2.1 gives the explicit eigendecomposition of Q(0) in terms of the eigenvectors of KK^T, and the paper shows that level sets corresponding to eigenvalues are unions of lattices of the form A(n+s/2) when the eigenspace lies in one of the H± subspaces, or contain lines or planes when the eigenspace straddles the two subspaces. Section 3 then enumerates, in two and three dimensions, which Bravais lattice classes are compatible with the constraint that all reciprocal vectors have equal length, concluding that only 3 of 5 two-dimensional and 6 of 14 three-dimensional classes are achievable.","tokens_in":102,"tokens_out":14002,"duration_ms":246670,"significance":"The Section 2 analysis is elegant, self-contained, and largely correct. It gives a parameter-free recipe: for physical constants a and b and a chosen matrix K, the optimal transducer amplitudes are eigenvectors of Q(0), and the geometry of the resulting minima is read off from the corresponding eigenspace without further numerical optimization. The lemmas are stated with proofs, and Examples 2.1-2.4 verify the level-set predictions numerically. If the Section 3 classification were corrected to describe the translation lattice of the particle positions rather than the wave lattice, the paper would provide a useful design rule for ultrasound-directed self-assembly. As it stands, however, the central application claim, namely that the achievable particle arrangements are limited to the enumerated Bravais classes, is not established.","major_comments":[{"comment":"The classification in Section 3 is for the wave lattice A = 2πK^{-T}, not for the lattice of particle positions. Lemma 2.4 shows that for a simple eigenvalue the set of global minima is L = A Z^d ∪ A(1/2+Z^d), which is the index-2 superlattice generated by A and A(1/2,...,1/2), not A itself. This lattice can belong to a different Bravais class from A. For example, in 2D take A hexagonal; then L has basis (a1+a2)/2 and (a1-a2)/2, which are orthogonal with unequal lengths, i.e. a primitive rectangular (orthorhombic) lattice, a class that Table 1 declares unachievable. Consequently, the abstract's statement that the arrangement's periodicity is limited to the enumerated Bravais classes is unsupported. The remark after Definition 3.1, which says the definition is irrespective of the particular particle arrangement inside a primitive cell, does not address this point, because the issue is not the configuration inside the cell but the translation lattice of the set of minima itself. The tables need to be recomputed for the minimizer lattice L (and for the other cases in Lemma 2.4), or the claims need to be explicitly restricted to the wave lattice.","section":"Section 3, Definition 3.1, Tables 1-2"},{"comment":"The discussion of the 'two points per primitive cell' case in Lemma 2.4 conceals the distinction between the primitive cell of A and the period of the set of minima. For a simple eigenvalue, the two points 0 and 1/2 in atomic coordinates generate the superlattice L described above, and the period of the arrangement is L, not A. For an eigenvalue of multiplicity d, T = {0,1}^d and the minima form the refinement A/2 Z^d, which has the same Bravais class as A; for intermediate multiplicities the union of cosets may or may not be a lattice. The paper does not analyze these cases before stating the enumerations in Tables 1 and 2. A complete treatment must either classify the Bravais class of L for each possible T_{λ,u} or explicitly state that only the wave lattice is being classified.","section":"Section 2.4, Lemma 2.4 and Section 3"}],"minor_comments":[{"comment":"The word 'Tegragonal' in the 'Implied symmetry' column should be 'Tetragonal'.","section":"Table 2, Hexagonal primitive row"},{"comment":"The sentence beginning 'We remark that definition 3.1 is irrespective of the particular particle arrangement inside a primitive cell' is ambiguous and should be replaced by an explicit statement of whether the classification refers to the wave lattice A or to the period lattice of the minima.","section":"Section 3, after Definition 3.1"},{"comment":"The captions describe the Bravais class of the wave lattice, but the plotted minima may lie on a different lattice; it would be helpful to state which eigenvalue multiplicity and which set T_{λ,u} is used in each panel, since the same wave lattice can lead to different minimizer sets.","section":"Figures 7 and 8"},{"comment":"In the definition of T_{λ,u}, the sign '±' is not quantified within the set-builder notation; from context it means the sign chosen in u = [v;±v], but this should be stated explicitly.","section":"Proof of Lemma 2.4"}],"recommendation":"major_revision","confidential_remarks":"The Section 2 results appear sound and the paper is well written. The main problem is in Section 3, where the classification of achievable Bravais classes is performed for the wave lattice A rather than for the actual lattice of particle positions, which can be an index-2 superlattice of A by Lemma 2.4. This is fixable by recomputing the Bravais classes of the minimizer lattices for each case in Lemma 2.4, or by restricting the claims to the wave lattice. If the authors can do that, the paper would be publishable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"I read the paper. The two sections that matter are very different. Section 2 is a genuinely clean piece of applied linear algebra: writing the radiation potential as u* Q(x) u, observing that shifts are unitary similarities, and then characterizing level-sets by sign changes in the minimal eigenspace. The results—points, lines, planes, and the explicit lattice coset formulas—are new relative to the earlier eigendecomposition papers [6,11] and are verified numerically in the examples. I trust that part.\n\nThe problem is Section 3. The abstract says the particle arrangement's periodicity is limited to certain Bravais classes, but Definition 3.1 classifies the wave lattice A = 2πK^{-T} under the constraint |k1|=...=|kd|. The actual minima are not necessarily on A. Lemma 2.4 shows that for a simple eigenvalue the minima are the union of two cosets A Z^d and A(1/2 + Z^d). That union is itself a lattice, generated by A and A(1/2,...,1/2), and it can be finer than A and have a different Bravais class. Concretely, if A is hexagonal in 2D, the lattice generated by (a1+a2)/2 and (a1-a2)/2 is primitive orthorhombic—exactly the class Table 1 says cannot be achieved. The remark after Definition 3.1 only says the arrangement inside a primitive cell doesn't matter, which sidesteps this issue. So the classification is about the pressure field's periodicity, not the particle arrangement's, and the abstract overclaims.\n\nThere are smaller caveats: the Gor'kov potential assumes small spherical particles and ignores size and shape (disclosed in Remark 1.1), and the theory only guarantees global minima of the ideal potential, so local minima could trap particles. Those are acceptable limitations.\n\nWho this is for: people designing ultrasound-directed assembly, and mathematicians interested in symmetric quadratic forms and lattices. The Section 3 fix is not hard—either restate the claim as applying to A, or compute the actual lattice of minima—but it is load-bearing for the title.\n\nMy recommendation: send it to peer review. The Section 2 contribution is solid and worth refereeing; the referee should ask for a correction or clarification in Section 3 before publication.","headline":"Solid level-set mathematics; the Bravais classification section classifies the wave lattice, not the particle lattice, and the abstract overstates the result.","tokens_in":14598,"tokens_out":5041,"would_cite":true,"duration_ms":50944,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35J05","74J05","82D25"],"pacs":[],"model":"deepseek-v4-flash","headline":"A small matrix predicts where acoustic waves trap particles","keywords":["acoustic radiation potential","Bravais lattices","crystallographic symmetries","ultrasound directed self-assembly","standing acoustic waves","level sets","eigenvalue method"],"falsifier":"Take a two-dimensional wave set with two equal-length wavevectors at 60 degrees to each other, pick a unit vector $u$ in a simple smallest-eigenvalue eigenspace of $Q(0)$, and compute the full set of global minima of the radiation potential. Lemma 2.4 predicts minima on the two lattices $A n$ and $A(n+1/2)$; determine the Bravais class of the resulting pattern. If that class is not tetragonal, hexagonal, or orthorhombic centred, the paper's claim that only three Bravais classes are achievable is false.","tokens_in":13603,"feed_emoji":"🔊","tokens_out":10416,"duration_ms":97794,"temperature":0.7,"pith_summary":"This paper determines which periodic crystal-like materials can be fabricated by letting a standing acoustic wave push small particles in a liquid resin and then curing the resin to freeze them in place. It proves that the positions where particles collect are governed by a 2d×2d real symmetric matrix built from the wave directions, and specifically by the eigenspace belonging to its smallest eigenvalue. Depending on the symmetries of that eigenspace, the minima of the acoustic radiation potential form periodic sets of isolated points, lines, or planes. Because the interfering wavevectors must all have the same length, the particle lattice can only belong to certain Bravais lattice classes, three in two dimensions and six in three, which the paper enumerates together with the transducer settings that realize them. The value is a concrete recipe for designing such materials and a statement of the theoretical limits of ultrasound-directed self-assembly.","feed_headline":"A small matrix predicts where acoustic waves trap particles","feed_subtitle":"Minima become points, lines, or planes; only six Bravais classes are achievable in 3D.","key_machinery":"The central object is the $2d\\times 2d$ Hermitian matrix $Q(x)=M(x)^*\\operatorname{diag}(a,-bI_d)M(x)$ that expresses the acoustic radiation potential as a quadratic form $\\psi(x;u)=u^*Q(x)u$ in the transducer amplitudes. The key structural fact is that a translation by $\\varepsilon$ is realized by a unitary similarity, so the eigenvalues of $Q(x)$ are position-independent; the whole spatial dependence is carried by the phase factors $\\exp(i[K,-K]^T x)$ acting on $u$. At $x=0$, the matrix decomposes as $Q(0)=a11^T-b[K;-K][K;-K]^T$, and its eigenspaces are built from the constant vector and the singular vectors of $K$. The level-set criteria in Theorem 2.1 then reduce to checking whether the eigenvector $u$ stays in the $\\lambda$-eigenspace after sign flips of its entries; the sets $T_{\\lambda,u}$, $T^{\\pm}_{\\lambda,u}$, and $R^{\\pm}_{\\lambda,u}$ record exactly which sign flips are allowed, and each allowed flip contributes a shifted lattice, a line, or a plane.","core_discovery":"Writing the acoustic radiation potential at position $x$ as $\\psi(x;u)=u^*Q(x)u$, where $u$ are the complex amplitudes of the $d$ plane waves, the paper shows that a spatial shift is a unitary similarity of $Q$, so all eigenvalue information is contained in $Q(0)=a11^T-b[K;-K][K;-K]^T$. For a unit-power parameter vector $u$ in the eigenspace of the smallest eigenvalue of $Q(0)$, the origin is a global minimum of $\\psi(\\cdot;u)$. Theorem 2.1 then characterizes the level sets $L_{\\lambda,u}$ for any eigenpair: if $u$ has a zero entry the minima contain lines or planes; if $u$ has no zero entries and the $\\lambda$-eigenspace lies within one of the symmetric subspaces $H_+$ or $H_-$, the level set is a union of up to $2d$ lattices $\\{A(n+s/2): n\\in\\mathbb{Z}^d\\}$ with $s\\in\\{0,1\\}^d$; and if the eigenspace straddles $H_+$ and $H_-$, lines or planes may appear. Applying the equal-length constraint $|k_1|=\\cdots=|k_d|$ to the enumeration of Bravais classes, the paper concludes that only three classes are achievable in two dimensions (tetragonal, hexagonal, orthorhombic centred) and six in three dimensions (triclinic primitive, orthorhombic face-centred, trigonal primitive, cubic primitive, cubic face-centred, cubic body-centred), with explicit reciprocal vectors for each.","pith_inferences":["The paper's recipe can be read backwards as a design algorithm: choose a desired Bravais class from the tables, set the wavevectors accordingly, compute $Q(0)$, and select $u$ from the minimal eigenspace; whether the desired within-cell particle motif is actually produced then hinges on the sign-flip sets $T_{\\lambda,u}$, which the paper gives but does not fully explore as a design tool.","Because the radiation potential itself rests on the small-sphere, inviscid-fluid approximation, real experiments with finite-size or non-spherical particles may see the trap positions shift or additional minima appear; quantifying that shift is a natural experimental follow-up the paper does not attempt.","The same quadratic-form-plus-unitary-shift structure appears in other wave-based manipulation settings, such as optical tweezers or structured light, so the Bravais-class limitations may be a general property of wave-directed assembly rather than an acoustic-specific feature.","One could test the enumeration by fabricating a two-dimensional hexagonal wavefield with a simple smallest eigenvalue and checking whether the predicted two-point-per-cell pattern indeed belongs to one of the three listed classes; the paper does not report such an explicit check."],"forward_implications":["Choosing transducer parameters from the smallest-eigenvalue eigenspace of $Q(0)$ makes the origin a global minimum of the radiation potential at fixed power, and by periodicity this minimum repeats on the lattice $A$.","Real particle patterns need not be the lattice $A$ alone: the minima can be a union of up to $2d$ shifted lattices $A(n+s/2)$, so a single wave set can produce a finer two-point or $2d$-point basis within each primitive cell.","The equal-length constraint on wavevectors rules out most Bravais classes; in two dimensions only tetragonal, hexagonal, and orthorhombic-centred lattices are achievable, and in three dimensions only six of the fourteen classes.","Turning off one or more transducers (zero entries in $u$) turns the minima from isolated points into continuous lines or planes, which could be used to make fibrous or lamellar particle arrangements.","The level-set description holds for every eigenvalue of $Q(0)$, not just the minimum, so the same machinery can describe other preferential surfaces inside the potential, although the paper leaves the study of non-global local minima open."],"supporting_citations":[{"why":"Supplies the acoustic radiation potential formula for small spherical particles in an ideal fluid, the quantity whose minima the paper classifies.","marker":"[5]"},{"why":"Introduces the eigendecomposition approach to minimize the acoustic radiation potential at user-specified points, which this paper extends to an explicit global characterization.","marker":"[6]"},{"why":"Extends the eigendecomposition minimization to three dimensions, motivating the 3D analysis here.","marker":"[11]"},{"why":"Source for the Bravais lattice classes and reciprocal lattice conventions used in the enumeration tables.","marker":"[9]"},{"why":"Reference tables for the reciprocal vectors of the three-dimensional Bravais classes, used to impose equal-length constraints in the enumeration.","marker":"[2]"},{"why":"Prior demonstration of acoustically trapped, reconfigurable colloidal crystals, the fabrication context the paper's limits apply to.","marker":"[3]"},{"why":"Prior experimental and theoretical study of particle patterning by ultrasonic standing waves in a rectangular cavity, an alternative arrangement method the analysis complements.","marker":"[13]"}],"fun_headline_variants":["Acoustic crystal types dictated by a small matrix eigenspace","Matrix eigenspace determines acoustic particle arrangements","Six Bravais lattices from acoustic standing waves","Acoustic traps arranged as points, lines, planes by matrix","Smallest eigenvalue of matrix fixes acoustic trap locations"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The enumeration of achievable Bravais classes assumes the particle pattern's periodicity is the wave lattice $A$, even though Theorem 2.1 shows minima can also form finer shifted lattices $A(n+s/2)$ that may belong to classes outside the enumeration.","fun_headline_variants_meta":{"raw":{"variants":["Acoustic crystal types dictated by a small matrix eigenspace","Matrix eigenspace determines acoustic particle arrangements","Six Bravais lattices from acoustic standing waves","Acoustic traps arranged as points, lines, planes by matrix","Smallest eigenvalue of matrix fixes acoustic trap locations"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000823,"raw_usage":{"total_tokens":3642,"prompt_tokens":1028,"completion_tokens":2614,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":644,"completion_tokens_details":{"reasoning_tokens":2538}},"tokens_in":644,"tokens_out":2614,"duration_ms":20751,"temperature":1.0,"reasoning_tokens":2538,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:35:16.179402+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a two-dimensional wave set with two equal-length wavevectors at 60 degrees to each other, pick a unit vector $u$ in a simple smallest-eigenvalue eigenspace of $Q(0)$, and compute the full set of global minima of the radiation potential. Lemma 2.4 predicts minima on the two lattices $A n$ and $A(n+1/2)$; determine the Bravais class of the resulting pattern. If that class is not tetragonal, hexagonal, or orthorhombic centred, the paper's claim that only three Bravais classes are achievable is false.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the acoustic radiation potential formula for small spherical particles in an ideal fluid, the quantity whose minima the paper classifies."},{"cited_title":"Greenhall, F","cited_arxiv_id":null,"evidence_quote":"Introduces the eigendecomposition approach to minimize the acoustic radiation potential at user-specified points, which this paper extends to an explicit global characterization."},{"cited_title":"Prisbrey, J","cited_arxiv_id":null,"evidence_quote":"Extends the eigendecomposition minimization to three dimensions, motivating the 3D analysis here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Source for the Bravais lattice classes and reciprocal lattice conventions used in the enumeration tables."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reference tables for the reciprocal vectors of the three-dimensional Bravais classes, used to impose equal-length constraints in the enumeration."},{"cited_title":"Caleap and B","cited_arxiv_id":null,"evidence_quote":"Prior demonstration of acoustically trapped, reconfigurable colloidal crystals, the fabrication context the paper's limits apply to."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Prior experimental and theoretical study of particle patterning by ultrasonic standing waves in a rectangular cavity, an alternative arrangement method the analysis complements."}],"review_version":1}