On Hermitian manifolds with constant mixed curvature

Basic Information

• Paper ID: 2503.12432
• Title: On Hermitian manifolds with constant mixed curvature
• Authors: Shuwen Chen, Fangyang Zheng
• Classification: math.DG (Differential Geometry)
• Publication Date: August 28, 2025 (arXiv v2)
• Paper Link: https://arxiv.org/abs/2503.12432

Abstract

Recently, Kai Tang conjectured that any compact Hermitian manifold with nonzero constant mixed curvature must be a Kähler manifold. He verified this conjecture in complex dimension 2 and for Chern Kähler-like manifolds in general dimensions. This paper verifies the conjecture for several classes of special Hermitian manifolds, including complex nilmanifolds, solvmanifolds with complex duals, almost abelian Lie groups, and Lie algebras containing J-invariant abelian ideals of codimension 2. The paper also verifies the conjecture for all compact balanced BTP threefolds when the Bismut connection has parallel torsion. These results provide partial evidence for the validity of Tang's conjecture.

Research Background and Motivation

Problem Background

1. Generalization of space forms: In Riemannian geometry, the simplest manifolds are space forms with constant sectional curvature. In complex geometry, since the sectional curvature of a Hermitian metric generally cannot be constant (unless the metric is flat), one considers the case of constant holomorphic sectional curvature.
2. Constant holomorphic sectional curvature conjecture: There has long existed a conjecture that a compact Hermitian manifold with nonzero constant holomorphic sectional curvature must be a Kähler manifold. This conjecture has been proven in complex dimension 2, but remains open in higher dimensions.
3. Introduction of mixed curvature: Chu-Lee-Tam introduced the concept of mixed curvature, defined as a convex combination of the first Ricci curvature and holomorphic sectional curvature: $C_{\alpha,\beta}(X) = \alpha\text{Ric}(X,X) + \beta H(X)$.

Research Motivation

Tang proposed an analogous conjecture regarding constant mixed curvature. This paper aims to provide evidence for this conjecture by studying specific types of Hermitian manifolds.

Core Contributions

1. Verification of Tang's conjecture for four classes of Lie-Hermitian manifolds:
   • Complex nilmanifolds
   • Solvmanifolds with complex duals
   • Almost abelian Lie groups
   • Lie algebras containing J-invariant abelian ideals of codimension 2
2. Results on manifolds with parallel Bismut torsion:
   • Verification of the conjecture for compact balanced BTP threefolds
   • Proof that mixed curvature cannot be nonzero constant for non-balanced BTP manifolds
3. Establishment of a systematic technical framework: Development of a unified method for analyzing mixed curvature on Lie-Hermitian manifolds
4. Generalization of known results: Extension of results on constant holomorphic sectional curvature to the mixed curvature case

Detailed Methodology

Core Conjecture

Tang's Conjecture: Let $(M^n, g)$ be a compact Hermitian manifold with constant mixed curvature $C_{\alpha,\beta} = c$. If $c \neq 0$, then $g$ is a Kähler metric.

Technical Framework

1. Structure Constants Method for Lie-Hermitian Manifolds

For a left-invariant complex structure $J$ and metric $g$ on a Lie group $G$, structure constants are defined as: $C^j_{ik} = \phi^j([e_i, e_k]), \quad D^j_{ik} = \phi^i([e_j, e_k])$

where $\{e_1, \ldots, e_n\}$ is a basis of $g^{1,0}$ and $\{\phi^1, \ldots, \phi^n\}$ is the dual coframe.

2. Characterization of Mixed Curvature

Under a unitary orthonormal frame, the constant mixed curvature condition is equivalent to: $4\beta\hat{R}_{i\bar{j}k\bar{\ell}} + \alpha(R_{i\bar{j}}g_{k\bar{\ell}} + R_{k\bar{j}}g_{i\bar{\ell}} + R_{i\bar{\ell}}g_{k\bar{j}} + R_{k\bar{\ell}}g_{i\bar{j}}) = 2c(g_{i\bar{j}}g_{k\bar{\ell}} + g_{i\bar{\ell}}g_{k\bar{j}})$

3. Key Technical Tools

• Salamon frames: Special frame choices for nilpotent Lie algebras
• Admissible frames: Applicable to Lie algebras with abelian ideals
• Properties of Bismut connection: Exploitation of special geometric structures of BTP manifolds

Proof Strategy

For Lie-Hermitian Manifolds

1. Frame establishment: Selection of appropriate unitary orthonormal frames with special structure constants
2. Utilization of unimodularity condition: $\sum_r(C^r_{ri} + D^r_{ri}) = 0$
3. Analysis of curvature components: Computation of Chern curvature components via structure constants
4. Application of constant mixed curvature condition: Derivation of $c = 0$ and proof of Chern flatness

For BTP Manifolds

1. Classification of known results: Utilization of the classification of balanced BTP threefolds
2. Case-by-case analysis: Discussion according to the rank of the Streets-Tian tensor
3. Proof by contradiction: For non-balanced cases, derivation of contradictions to prove $c = 0$

Main Theorems

Theorem 1 (Lie-Hermitian Manifolds)

Let $(M^n, g)$ be a Lie-Hermitian manifold whose universal cover is $(G, J, g)$. Assume the Lie algebra $\mathfrak{g}$ satisfies one of the following conditions:

• (i) $\mathfrak{g}$ is nilpotent
• (ii) $\mathfrak{g}$ is solvable and $J\mathfrak{g}' = \mathfrak{g}'$
• (iii) $\mathfrak{g}$ is unimodular and almost abelian
• (iv) $\mathfrak{g}$ is unimodular and contains a J-invariant abelian ideal of codimension 2

If $C_{\alpha,\beta} = c$, then $c = 0$. If $C_{\alpha,\beta} = 0$ and $\beta \neq 0$, then $\mathfrak{g}$ must be Chern flat.

Theorem 2 (Generalized Result)

Let $(M^n, g)$ be a Lie-Hermitian manifold. Assume $\mathfrak{g}' + J\mathfrak{g}'$ is nilpotent. If $C_{\alpha,\beta} = c$, then $c = 0$. If $C_{\alpha,\beta} = 0$ and $\beta \neq 0$, then the manifold must be Chern flat.

Theorem 3 (Balanced BTP Threefolds)

Let $(M^3, g)$ be a compact balanced BTP threefold that is non-Kähler. If $C_{\alpha,\beta} = c$, then $c = 0$. If $C_{\alpha,\beta} = 0$ and $(\alpha,\beta) \neq (0,0)$, then either the manifold is Chern flat or $\beta = 0$ and it is intermediate type Chern Ricci flat.

Theorem 4 (Non-balanced BTP Manifolds)

Let $(M^n, g)$ be a compact non-balanced BTP manifold. If $C_{\alpha,\beta} = c$, then $c = 0$.

Technical Innovations

1. Unified Framework

Establishment of a unified method for handling different types of Lie-Hermitian manifolds through systematic analysis of mixed curvature via special properties of structure constants.

2. Refined Case Analysis

For cases containing J-invariant abelian ideals, detailed classification according to different ranges of the parameter $\alpha/\beta$.

3. Geometric Analysis of BTP Manifolds

Full utilization of classification results for BTP manifolds, particularly the complete classification of balanced BTP threefolds.

4. Proof by Contradiction Techniques

For non-balanced BTP manifolds, clever exploitation of admissible frame properties to derive contradictions.

Experimental Results and Applications

Verification of Main Results

1. Complex nilmanifolds: Verification of Tang's conjecture, consistent with known results of Li-Zheng
2. Almost abelian case: Provision of more general proofs than known results on constant holomorphic sectional curvature
3. BTP manifolds: First systematic study of mixed curvature problems on these manifolds

Specific Computational Examples

• Wallach threefolds: Detailed calculations showing that mixed curvature cannot be nonzero constant
• Intermediate type BTP threefolds: Specific proof that $C_{1,-2} \equiv 0$ is impossible

Related Work

Historical Background

1. Constant holomorphic sectional curvature conjecture: Balas-Gauduchon (1985), Apostolov-Davidov-Muskarov (1996)
2. Mixed curvature theory: Introduction by Chu-Lee-Tam, important advances by Yang, Ni, and others
3. Lie-Hermitian manifolds: Foundational work by Salamon, Lauret, Valencia, and others

Positioning of This Paper

This paper applies mixed curvature theory to Lie-Hermitian manifolds and BTP manifolds, filling a gap in this interdisciplinary area.

Conclusions and Discussion

Main Conclusions

1. Tang's conjecture holds for multiple important classes of Hermitian manifolds
2. The constancy of mixed curvature imposes strong constraints on geometric structures
3. BTP manifolds provide an important platform for studying non-Kähler geometry

Limitations

1. Unable to prove Tang's conjecture in complete generality
2. Dependence on certain technical conditions (such as $\beta \neq 0$)
3. The case of higher-dimensional balanced BTP manifolds requires further investigation

Future Directions

1. Investigation of more general classes of Hermitian manifolds
2. Exploration of geometric significance of mixed curvature
3. Development of new technical tools for handling non-homogeneous cases

In-Depth Evaluation

Strengths

1. Technical rigor: Detailed proofs, accurate computations, clear logic
2. Systematicity: Establishment of a unified theoretical framework
3. Completeness: Coverage of multiple important classes of manifolds
4. Originality: First systematic study of mixed curvature on BTP manifolds

Weaknesses

1. Limitations: Restriction to specific types of manifolds
2. Technical complexity: Some proofs are overly technical, affecting readability
3. Geometric intuition: Lack of in-depth discussion of geometric significance of results

Impact

1. Theoretical contribution: Provides important evidence for Tang's conjecture
2. Methodological value: The established technical framework is applicable to related problems
3. Foundation for future research: Lays groundwork for further investigation of mixed curvature theory

Applicable Scenarios

The methods and results of this paper are applicable to:

1. Research on curvature problems in complex geometry
2. Hermitian geometry on Lie groups
3. Classification problems for non-Kähler manifolds
4. Verification of conjectures in differential geometry

References

The paper cites 62 related references, covering important works in complex geometry, Lie group theory, differential geometry, and other fields, demonstrating the depth and breadth of the research.