2025-11-30T10:28:18.745802

Brauer groups of certain affine cubic surfaces

Alfaraj
We study the Brauer groups of affine surfaces that are complements of singular hyperplane sections of smooth cubic surfaces over a field $k$ of characteristic $0$. We determine the Brauer group over the algebraic closure as a Galois module for all the possible singular hyperplane sections. For the case when the hyperplane section is geometrically the union of three lines, we give explicit examples where transcendental elements of order $2$ and $3$ exist over $\mathbb{Q}$. We end with an application on the integral Brauer-Manin obstruction to the integral Hasse principle.
academic

Brauer groups of certain affine cubic surfaces

Basic Information

  • Paper ID: 2509.16042
  • Title: Brauer groups of certain affine cubic surfaces
  • Author: Abdulmuhsin Alfaraj (University of Bath)
  • Classification: math.AG (Algebraic Geometry), math.NT (Number Theory)
  • Publication Date: October 2025 (arXiv v2)
  • Paper Link: https://arxiv.org/abs/2509.16042v2

Abstract

This paper investigates the Brauer group of affine surfaces obtained as complements of singular hyperplane sections in smooth cubic surfaces over fields of characteristic 0. The author determines the structure of the Brauer group on the algebraic closure as a Galois module for all possible singular hyperplane sections. For the case where the hyperplane section is geometrically a union of three lines, explicit examples are provided showing the existence of transcendental elements of order 2 and 3 over Q\mathbb{Q}. The paper concludes with applications of integral Brauer-Manin obstructions to the integral Hasse principle.

Research Background and Motivation

Problem Background

  1. Arithmetic Significance of Brauer Groups: The Brauer group is a geometric object containing rich arithmetic information and is a birational invariant for smooth proper varieties. In 1970, Manin successfully used the Brauer group to explain failures of the Hasse principle, initiating the study of Brauer-Manin obstructions.
  2. Computational Challenges: Determining the Brauer group, particularly its transcendental part, is typically a difficult problem. For non-proper varieties, the situation becomes more complex—the Brauer group is expected to be larger than that of any smooth compactification.
  3. Importance of log K3 Surfaces: Log K3 surfaces over number fields have special arithmetic significance. It is conjectured that the positivity of the log anticanonical class of an affine variety controls the abundance of integral points. Log K3 surfaces (with trivial anticanonical class) represent the boundary case and are of fundamental importance.
  4. Existing Research:
    • Colliot-Thélène and Wittenberg studied families of diagonal cubic surfaces (complements of smooth genus 1 curves)
    • Bright and Lyczak provided uniform bounds on the size of Brauer groups of complements of smooth anticanonical divisors in del Pezzo surfaces
    • Markoff surfaces (unions of three rational lines) were studied by Colliot-Thélène et al. and by Loughran-Mitankin

Research Motivation

This paper studies surfaces that can be embedded as complements of singular anticanonical divisors in smooth cubic surfaces for the following reasons:

  • These surfaces are abundant log K3 surfaces with interesting arithmetic properties
  • From a geometric perspective, they may possess infinite discrete automorphism groups
  • In the Markoff surface case, the automorphism group combined with Brauer-Manin obstructions yields applications

Core Contributions

  1. Complete Classification Theorem: Determines the complete structure of the Brauer group Br(U)\text{Br}(U) as a Galois module Γk=Gal(kˉ/k)\Gamma_k = \text{Gal}(\bar{k}/k) for all possible singular hyperplane sections (Theorem 1.1)
  2. Existence of Transcendental Brauer Groups: For the case of three geometric lines, provides explicit constructions showing the existence of transcendental elements of order 2 and 3 over Q\mathbb{Q} (Theorem 1.2)
  3. Integral Brauer-Manin Obstruction Application: For the equation 9x3+y3=z2+39x^3 + y^3 = z^2 + 3, proves the existence of an integral Brauer-Manin obstruction from a 3-torsion element (Theorem 1.3), answering a question posed by Grechuk on MathOverflow
  4. Computational Methods: Develops systematic methods for computing Brauer groups using Gysin sequences, purity theorems, and Galois cohomology

Detailed Methodology

Problem Setup

Given a smooth cubic surface X⊂P3X \subset \mathbb{P}^3 over a characteristic 0 field kk and a singular hyperplane section HH, compute the structure of the Brauer group Br(U)\text{Br}(U) as a Galois module for the affine surface U:=X∖HU := X \setminus H.

Theoretical Framework

1. Stratification of the Brauer Group

For a kk-variety XX, define:

  • Br0(X)=im[Br(k)→Br(X)]\text{Br}_0(X) = \text{im}[\text{Br}(k) \to \text{Br}(X)] (constant part)
  • Br1(X)=ker⁡[Br(X)→Br(Xˉ)]\text{Br}_1(X) = \ker[\text{Br}(X) \to \text{Br}(\bar{X})] (algebraic Brauer group)
  • Bra(X):=Br1(X)/Br0(X)\text{Br}_a(X) := \text{Br}_1(X)/\text{Br}_0(X)
  • Br(X)/Br1(X)\text{Br}(X)/\text{Br}_1(X) (transcendental Brauer group)

2. Key Technical Tools

Gysin Sequence (Lemma 3.3): For a smooth kk-variety XX and a smooth closed subscheme ZZ of codimension cc, there is a long exact sequence: 0→H2c−1(X,μn)→H2c−1(X∖Z,μn)→H0(Z,Z/n(−c))→H2c(X,μn)→⋯0 \to H^{2c-1}(X,\mu_n) \to H^{2c-1}(X \setminus Z, \mu_n) \to H^0(Z,\mathbb{Z}/n(-c)) \to H^{2c}(X,\mu_n) \to \cdots

Semi-purity (Lemma 3.4): For reduced closed subschemes ZZ of codimension 1 everywhere (with singularities of codimension ≥2\geq 2), an analogous sequence holds.

Poincaré Duality and Absolute Purity Theorem (Gabber): Connect different cohomology groups.

Computational Strategy

Step One: Algebraic Brauer Group

Use the short exact sequence (Proposition 2.5): 0→⨁i=13Z[ℓi]→Pic(X)→Pic(U)→00 \to \bigoplus_{i=1}^3 \mathbb{Z}[\ell_i] \to \text{Pic}(X) \to \text{Pic}(U) \to 0

Compute Bra(U)≅H1(k,Pic(U))\text{Br}_a(U) \cong H^1(k, \text{Pic}(U)) (Lemma 2.4).

Through Magma computations of all possible Galois actions, obtain finitely many possibilities (Proposition 2.6).

Step Two: Global Brauer Group

From the Kummer sequence obtain the commutative diagram (3.7):

0 @>>> (\text{Pic}X)/n @>>> H^2(X,\mu_n) @>>> (\text{Br}X)[n] @>>> 0\\ @. @VVV @VVV @VVV @.\\ 0 @>>> (\text{Pic}U)/n @>>> H^2(U,\mu_n) @>>> (\text{Br}U)[n] @>>> 0 \end{CD}$$ Key result: $\text{Br}(U)[n] \cong \text{coker}(\Psi_n)$ (Lemma 3.8). #### Step Three: Case-by-Case Computation **Case 1: Line + Smooth Conic** (Proposition 3.9) - If tangent: $\text{Br}U = 0$ - If intersecting at two rational points: $\text{Br}U \cong \mathbb{Q}/\mathbb{Z}(-1)$ - If intersecting at a degree 2 point $k(\sqrt{d})$: $(\text{Br}U)[n] \cong M_d/nM_d(-1)$, where $M_d = (\text{Ind}_{k(\sqrt{d})/k}\mathbb{Z})/\mathbb{Z}$ **Case 2: Geometrically Irreducible Singular Cubic Curve** (Proposition 3.10) - Cusp: $\text{Br}U = 0$ - Split multiplicative node: $\text{Br}U \cong \mathbb{Q}/\mathbb{Z}(-1)$ - Non-split multiplicative node: $(\text{Br}U)[n] \cong M_d/nM_d(-1)$ **Case 3: Three Geometric Lines** (Proposition 3.11) Let $L$ be the minimal splitting field. Key computation explicitly describes the map $\phi: H^1(C', \mathbb{Z}/n\mathbb{Z}) \to H^3(X', \mu_n)$. Through Lemma 3.7 establish the commutative diagram (3.14), using the residue formula (3.15): $$\partial(a,b,c) = (c-b, a-c, b-a)$$ Obtain $\ker(\phi) \cong \mathbb{Z}/n\mathbb{Z}(-1)$ as a $\Gamma_L$-module. By analyzing Galois actions (explicit calculations of 3-cycles and 2-cycles), determine the structure as a $\Gamma_k$-module. ### Technical Innovations 1. **Unified Framework**: Incorporates all singular hyperplane section cases into a unified computational framework 2. **Explicit Galois Actions**: Through explicit formulas for residue maps (3.15), precisely determine Galois module structures 3. **Use of Induced Modules**: Introduces $M_d = (\text{Ind}_{k(\sqrt{d})/k}\mathbb{Z})/\mathbb{Z}$ to concisely describe non-split cases 4. **Constructive Methods**: Provides completely explicit constructions of transcendental elements (Proposition 4.4) ## Main Results ### Theorem 1.1 (Complete Classification of Brauer Groups) Let $k$ be a characteristic 0 field, $\Gamma_k := \text{Gal}(\bar{k}/k)$, $X$ a smooth cubic surface, $H$ a hyperplane section, and $U := X \setminus H$. **(i) Line + Smooth Conic Case**: $$\text{Br}U \cong \begin{cases} 0, & \text{if tangent}\\ \mathbb{Q}/\mathbb{Z}(-1), & \text{if } |\ell(k) \cap C(k)| = 2\\ \varinjlim(M_d/nM_d(-1)), & \text{if } \ell \cap C = \text{Spec}(k(\sqrt{d})) \end{cases}$$ **(ii) Geometrically Irreducible Singular Cubic Curve**: $$\text{Br}U \cong \begin{cases} 0, & \text{if cusp}\\ \mathbb{Q}/\mathbb{Z}(-1), & \text{if split multiplicative node}\\ \varinjlim(M_d/nM_d(-1)), & \text{if non-split node} \end{cases}$$ **(iii) Three Lines**: Let $L$ be the minimal splitting field, $d \in k$ the discriminant of $L/k$, and $\tilde{L} := L(\sqrt{d})$. $$\text{Br}U \cong \begin{cases} 0, & \text{if Eckard point}\\ \mathbb{Q}/\mathbb{Z}(-1), & \text{if } L=k \text{ and } \cap\ell_i = \emptyset \text{, or } \text{Gal}(L/k) \cong \mathbb{Z}/3\mathbb{Z}\\ \varinjlim(M_d/nM_d(-1)), & \text{if } \text{Gal}(\tilde{L}/k) \cong S_3 \text{ or } \text{Gal}(L/k) \cong \mathbb{Z}/2\mathbb{Z} \end{cases}$$ ### Theorem 1.2 (Existence of Transcendental Elements) There exist affine cubic surfaces $U$ over $\mathbb{Q}$ (complements of three geometric lines) such that: - $\mathbb{Z}/2\mathbb{Z} \subset \text{Br}(U)$ and $\mathbb{Z}/2\mathbb{Z} \not\subset \text{Br}_1(U)$ - $\mathbb{Z}/3\mathbb{Z} \subset \text{Br}(U)$ and $\mathbb{Z}/3\mathbb{Z} \not\subset \text{Br}_1(U)$ **Explicit Construction** (Section 4.1): **Order 2 Case**: Take $f_1 = X, f_2 = Y, f_3 = Z$ with points chosen as: - $P_1 = [0:1:1], P_4 = [0:4:1]$ (on $\ell_1$) - $P_2 = [9:0:1], P_5 = [16:0:1]$ (on $\ell_2$) - $P_3 = [-25:1:0], P_6 = [-36:1:0]$ (on $\ell_3$) Generator: $A_2 = (f_1/f_3, f_2/f_3)$. **Order 3 Case**: Over $L = \mathbb{Q}(\omega)$, take: $$f_1 = X + \omega Y + Z, \quad f_2 = X + \omega^2 Y + Z, \quad f_3 = Z$$ Carefully chosen six points satisfy the residue conditions, with generator: $$A_3 := \text{cor}_{L(T_L)/\mathbb{Q}(T)}\left(\frac{f_1}{f_3}, \frac{f_2}{f_3}\right)_\omega$$ ### Theorem 1.3 (Integral Brauer-Manin Obstruction) Consider the $\mathbb{Z}$-scheme: $$U: 9x^3 + y^3 = z^2 + 3 \subset \mathbb{A}^3_{\mathbb{Z}}$$ Let $\omega$ be a primitive cube root of unity. Then: 1. $\text{Br}(U_{\mathbb{Q}})/\text{Br}(\mathbb{Q}) \cong \mathbb{Z}/3\mathbb{Z}$ 2. The generator is the cyclic algebra: $$B := \text{cor}_{\mathbb{Q}(U_{\mathbb{Q}})(\omega)/\mathbb{Q}(U_{\mathbb{Q}})}\left(-9, z + \sqrt{-3}\right)_\omega$$ 3. $U(\mathbb{A}_{\mathbb{Z}})^B = \emptyset$ **Proof Outline** (Section 5): - Prove $U(\mathbb{A}_{\mathbb{Z}}) \neq \emptyset$ (verify local points exist for all primes) - For $x_3 = (x,y,z) \in U(\mathbb{Z}_3)$, analysis modulo 9 yields $z \equiv \pm 4 \pmod{9}$ - Use explicit formulas for cyclic algebras (Colliot-Thélène et al., [3]): - $(ω, 1+aλ+bλ^2+\cdots)_{ω,\mathfrak{p}} = a+a^2+b \in \mathbb{Z}/3\mathbb{Z}$ - $(λ, 1+aλ+bλ^2+cλ^3+\cdots)_{ω,\mathfrak{p}} = (a-a^3)/3+ab-c \in \mathbb{Z}/3\mathbb{Z}$ - Compute: $$\text{inv}_{\mathfrak{p}}A(x_3) = \begin{cases} 2 \in \mathbb{Z}/3\mathbb{Z}, & z \equiv -4 \pmod{9}\\ 1 \in \mathbb{Z}/3\mathbb{Z}, & z \equiv 4 \pmod{9} \end{cases}$$ - Therefore $\sum_p \text{inv}_p B(x_p) \neq 0$, so $U(\mathbb{A}_{\mathbb{Z}})^B = \emptyset$ ## Technical Details ### Galois Cohomology Computations For $M_d = (\text{Ind}_{k(\sqrt{d})/k}\mathbb{Z})/\mathbb{Z}$, Lemma 4.1 gives: If $\sqrt{d} \notin k(\zeta_n)$: $$(M_d/nM_d(-1))^{\Gamma_k} \cong \begin{cases} \mathbb{Z}/2\mathbb{Z}, & n = 2^i\\ 0, & \text{otherwise} \end{cases}$$ If $\sqrt{d} \in k(\zeta_n)$ and $[k(\zeta_n):k] = [\mathbb{Q}(\zeta_n):\mathbb{Q}]$: $$(M_d/nM_d(-1))^{\Gamma_k} \cong \begin{cases} \mathbb{Z}/4\mathbb{Z}, & n = 2^i\\ \mathbb{Z}/3\mathbb{Z}, & n = 3^i\\ 0, & \text{otherwise} \end{cases}$$ **Proof Strategy**: Transform the problem into computing Galois invariants of Hom modules $H_n$, using explicit actions of $G_n = \text{Gal}(k(\zeta_n, \sqrt{d})/k)$. ### Bounds on Transcendental Brauer Groups Theorem 4.2 summarizes: Assuming $k$ contains no nontrivial roots of unity, $\text{Br}U/\text{Br}_1 U$ is a subgroup of: | Case | Upper Bound | |------|-------------| | Tangent/Cusp/Eckard point | 0 | | Two rational points/Split node/$L=k$ or $\mathbb{Z}/3\mathbb{Z}$ | $\mathbb{Z}/2\mathbb{Z}$ | | $k(\sqrt{d}) \subset k(\zeta_4)$ | $\mathbb{Z}/4\mathbb{Z}$ | | $k(\sqrt{d}) \subset k(\zeta_3)$ | $\mathbb{Z}/3\mathbb{Z} \times \mathbb{Z}/2\mathbb{Z}$ | | Other | $\mathbb{Z}/2\mathbb{Z}$ | ## Related Work ### Historical Background 1. **Manin (1970)**: First used Brauer groups to explain failures of the Hasse principle, initiating Brauer-Manin obstruction theory 2. **Colliot-Thélène & Sansuc (1981)**: Developed Brauer group theory for algebraic groups 3. **Colliot-Thélène & Wittenberg (2012)**: Studied integral Brauer-Manin obstructions for diagonal cubic surfaces ### Related Research 1. **Bright & Lyczak (2019)**: Uniform bounds on Brauer groups of complements of smooth anticanonical divisors in del Pezzo surfaces 2. **Colliot-Thélène, Wei & Xu (2020)**: Brauer-Manin obstructions for Markoff surfaces 3. **Loughran & Mitankin (2021)**: Integral Hasse principle and strong approximation for Markoff surfaces 4. **Harpaz (2017)**: Geometry and arithmetic of log K3 surfaces ### Innovations in This Paper - **Completeness**: First systematic study of all singular hyperplane section cases - **Explicitness**: Complete explicit constructions of transcendental elements - **Applicability**: Provides new instances of integral Brauer-Manin obstructions ## Conclusions and Discussion ### Main Conclusions 1. Completely determines the structure of Brauer groups of complements of singular hyperplane sections as Galois modules 2. Proves existence of transcendental elements of order 2 and 3 3. Provides new applications of integral Brauer-Manin obstructions ### Limitations 1. **Characteristic Restriction**: Methods rely on characteristic 0 assumption (absolute purity theorem) 2. **Computational Complexity**: Algebraic Brauer group computations require computer algebra systems like Magma 3. **Generality**: Focuses on cubic surfaces; higher degree cases not addressed ### Future Directions 1. Extend to other types of log K3 surfaces 2. Study interactions between infinite automorphism groups and Brauer-Manin obstructions 3. Explore applications of transcendental Brauer groups to other arithmetic problems 4. Develop more efficient computational methods ## In-Depth Evaluation ### Strengths 1. **Theoretical Completeness**: - Provides complete classification for all possible cases (Theorem 1.1) - Systematically develops methodology for computing Brauer groups - Clear theoretical framework with rigorous logic 2. **Technical Depth**: - Skillfully combines Gysin sequences, purity theorems, and Galois cohomology - Explicit calculations of Galois actions (particularly 3-cycles and 2-cycles) demonstrate sophisticated technique - Precise computations with cyclic algebra residues reflect deep expertise 3. **Constructive Nature**: - Proof of Theorem 1.2 provides completely explicit examples - All point coordinates explicitly given, highly verifiable - Proof of Theorem 1.3 includes all computational details 4. **Practical Value**: - Answers an open question posed on MathOverflow - Provides new tools for integral Hasse principle research - Methods generalizable to other log K3 surfaces ### Weaknesses 1. **Computational Dependence**: - Algebraic Brauer group computations depend on Magma without fully independent proofs - Some Galois cohomology calculations are highly technical with limited readability 2. **Generalization Limitations**: - Main results restricted to cubic surfaces - Applicability of methods to more general log K3 surfaces unclear 3. **Limited Examples**: - While proving transcendental element existence, provides only few explicit examples - Lacks statistical understanding of "typical" transcendental Brauer group size 4. **Application Scope**: - Integral Brauer-Manin obstruction examples somewhat specialized - Potential applications to other Diophantine problems not discussed ### Impact 1. **Theoretical Contribution**: - Fills gap in Brauer group research for singular boundary cases - Provides new perspective for arithmetic of log K3 surfaces - Developed techniques applicable to other geometric objects 2. **Practical Value**: - Provides new tool for determining integral point existence - Explicit construction methods have direct application value - Computational framework adoptable by other researchers 3. **Reproducibility**: - All constructions explicit and easily verifiable - Main theorem proofs logically clear and traceable - Sufficient technical detail for others to learn ### Applicable Scenarios 1. **Theoretical Research**: - Brauer group computations for log K3 surfaces - Explicit Galois cohomology calculations - Invariant studies in birational geometry 2. **Arithmetic Applications**: - Hasse principle and strong approximation problems - Solvability determination for Diophantine equations - Obstruction theory for integral point distribution 3. **Computational Aspects**: - Algorithm foundations for computer algebra systems - Brauer group computations for specific cubic surfaces - Explicit cyclic algebra constructions ## Selected References [3] J.-L. Colliot-Thélène, D. Kanevsky, and J.-J. Sansuc, *Arithmétique des surfaces cubiques diagonales*, Lecture Notes in Math., vol. 1290, Springer, 1987. [4] J.-L. Colliot-Thélène, A. Skorobogatov, *The Brauer–Grothendieck Group*, Springer, 2021. [5] J.-L. Colliot-Thélène, D. Wei, and F. Xu, *Brauer-Manin obstruction for Markoff surfaces*, Ann. Sc. Norm. Super. Pisa, 2020. [14] D. Loughran, V. Mitankin, *Integral Hasse principle and strong approximation for Markoff surfaces*, IMRN, 2021. [18] J. S. Milne, *Étale cohomology*, Princeton University Press, 1980. --- **Overall Assessment**: This is an excellent algebraic geometry paper of high technical depth with complete results. The author systematically studies an important but previously incompletely understood class of geometric objects, develops effective computational methods, and provides meaningful arithmetic applications. The paper's principal value lies in its completeness, constructive nature, and methodological contributions, establishing a solid foundation for subsequent research.