Functional Bi‐MOFs and Their Derivatives: From Structural Engineering to Multifunctional Applications in Energy, Catalysis, and Biomedicine

ABSTRACT

Bismuth‐based metal–organic frameworks (Bi‐MOFs) are emerging as a distinctive class of functional porous materials that combine structural tunability, biocompatibility, and unique physicochemical characteristics of Bi(III). In contrast to conventional transition‐metal‐based MOFs, however, their assembly is strongly governed by the stereochemically active 6s² lone pair, variable coordination environments, and the pronounced hydrolytic tendency of Bi(III), which together complicate the realization of predictable structures, permanent porosity, and operational stability. This review provides a systematic analysis of the structure–property relationships of Bi‐MOFs, focusing on how Bi(III) coordination chemistry governs framework design, topology evolution, and stability. The influence of major ligand families, including carboxylates, phosphonates, and phenolates, is critically discussed, and the representative strategies for transforming Bi‐MOFs into functional derivatives, such as porous carbons, oxides, sulfides, and phosphides, are summarized. Their applications in photocatalysis, electrocatalysis, energy storage, biomedicine, and environmental remediation are also highlighted, with particular attention to the relationship between precursor structure and functional performance. Key challenges, including green synthesis, rigorous evaluation of porosity and stability, and the development of multivariate and application‐oriented systems, are further discussed. Rather than viewing Bi‐MOFs as universal substitutes for established MOFs, this review identifies them as specialized functional platforms and provides a critical perspective for the rational development of next‐generation Bi‐MOF‐based materials.

Bi(III) 6s² lone pair shapes a unique Bi‐MOF assembly distinct from transition‐metal MOFs. This review dissects coordination‐driven framework structures, ligand effects, and derivative fabrication routes. It maps structure‐performance links across catalysis, energy, biomedicine and environmental uses, pinpointing unresolved obstacles and framing Bi‐MOFs as tailored advanced functional platforms.

1. Introduction

Metal–organic frameworks (MOFs) are a highly versatile class of crystalline porous materials assembled from metal nodes and organic linkers, and they are distinguished by the exceptional tunability of their structures and functions [1, 2]. Over the past two decades, extensive studies have established relatively mature design principles for conventional MOFs, particularly those based on transition metals such as Zr, Cu, Fe, and Zn [3, 4, 5, 6, 7]. These systems often benefit from comparatively well‐defined coordination geometries, predictable secondary building units, and, in representative cases, high permanent porosity, robust chemical stability, and increasing technological maturity [8]. By contrast, MOFs derived from main‐group metal centers remain far less systematically understood, leaving an important gap in the broader landscape of framework chemistry.

Among these less explored systems, bismuth‐based MOFs (Bi‐MOFs) are of particular interest because they differ from classical transition‐metal‐based MOFs in several key structure‐directing aspects. Bi(III) possesses a stereochemically active 6s² lone pair, which frequently gives rise to asymmetric coordination environments, flexible coordination numbers, and distorted coordination polyhedra. As a result, the construction of Bi‐MOF is often characterized by lower geometric predictability and greater local structural irregularity than that observed in many established MOF families [9, 10]. In addition, the strong hydrolysis tendency of Bi³⁺ imposes a relatively narrow synthetic and stability window, such that framework formation requires careful control over precursor speciation, ligand deprotonation, solvent environment, and pH‐related conditions. These features indicate that Bi‐MOFs should not be regarded simply as another extension of conventional MOF chemistry. Rather, they represent a distinct subclass of framework in which lone‐pair effects, hydrolysis behavior, and coordination flexibility play unusually important roles.

At the same time, the significance of Bi‐MOFs does not lie in directly outperforming benchmark MOF families in terms of ultrahigh surface area, topological regularity, or hydrolytic robustness. Their real promise lies instead in combining the structural tunability of porous frameworks with the distinctive physicochemical characteristics of bismuth. As a heavy element, bismuth can introduce strong spin–orbit coupling and influence charge separation, excited‐state behavior, and redox‐related processes, which is potentially advantageous for photocatalysis, sensing, and related functional applications [11, 12, 13, 14, 15]. In parallel, the relatively low toxicity and favorable bio‐related profile of many bismuth compounds make Bi‐containing frameworks attractive for biomedical exploration when compared with a number of transition‐metal‐based counterparts [16, 17]. More broadly, Bi‐MOFs occupy a distinctive position at the intersection of porous‐material design, main‐group coordination chemistry, and Bi‐centered functional materials [18].

This position becomes even clearer when Bi‐MOFs are benchmarked against other established porous materials. Compared with mature MOF platforms such as UiO‐ [5], MIL‐ [6], and HKUST‐type systems [7], Bi‐MOFs generally remain less predictable in framework construction, less rigorously validated in terms of permanent porosity, and less comprehensively assessed under realistic stability conditions [19]. Compared with traditional porous solids such as zeolites, porous oxides, or related inorganic hosts, Bi‐MOFs do not currently offer superior practical maturity, durability, or process readiness [20, 21, 22, 23]. Their present level of material development is therefore better described as one of exploratory synthesis, proof‐of‐concept functionality, and early‐stage application demonstration, rather than standardized deployment. Yet this relative immaturity is precisely why a dedicated review is timely. The field has now expanded sufficiently to warrant a critical framework capable of distinguishing genuinely Bi‐specific opportunities from claims that simply transfer the general language of MOF research to a much less mature system.

At present, the Bi‐MOF literature remains fragmented [24, 25, 26]. Many studies focus on the successful synthesis of new structures or on preliminary demonstrations of catalytic [27, 28], adsorption [29], electrochemical [30, 31], or biomedical performance [17, 32], whereas comparatively few attempt to extract general design rules concerning coordination constraints, framework stabilization strategies, porosity verification, or structure–property trade‐offs. In particular, several fundamental questions remain unresolved: which ligand environments are most compatible with hydrolytically vulnerable Bi centers; how cluster nuclearity and connectivity affect framework robustness; under what circumstances structural distortion is functionally beneficial rather than detrimental to framework integrity; and which characterization criteria are sufficient to support claims of permanent porosity or application‐relevant stability. Without clearer design heuristics and more consistent evaluation standards, the field risks remaining largely descriptive and case‐specific.

Against this background, the aim of this review is not simply to catalogue reported Bi‐MOFs, but to assess them as a distinct yet still developing class of materials. We first examine the coordination chemistry of Bi(III), with particular emphasis on the structural consequences of the 6s² lone pair and the synthetic constraints associated with hydrolysis‐prone Bi species. We then discuss how ligand type, connectivity, cluster nuclearity, and topology evolution govern framework formation, stability, and functional behavior, thereby extracting design‐relevant structure–property relationships. In addition, we emphasize the need for more rigorous evaluation of porosity, stability, reproducibility, and application claims in the current Bi‐MOF literature. By comparing Bi‐MOFs with classical MOFs and other porous materials, we further clarify where their most meaningful opportunities may lie and what is required for the field to progress from exploratory synthesis toward more mature functional platforms.

2. Structure Features of Bi‐MOFs

Bi‐MOFs are intricate structures formed through the linkage of polyatomic inorganic Bi‐containing cores or clusters and organic ligands. The diversity of these components, along with the various ways of connection, results in a wide array of structural configurations. Recent studies, encompassing analyses up to the year 2020, have systematically summarized the structures of most reported Bi‐MOFs, organizing them based on the principal frameworks of the ligands employed [25]. As an extension, this section delves into the chemical behavior of Bi cations in solution, elucidating the challenges researchers face in deciphering the coordination “habits” of Bi ions. Furthermore, it shifts focus toward the coordination structures and topological networks of Bi‐MOFs, newly developed in the last five years, with a special emphasis on those exhibiting permanent pores. Meanwhile, the publications related to Bi‐MOFs over the past decade have been systematically reviewed, and the major research directions have been categorized and summarized based on their specific content, to reveal the developmental trends and research hotspots in this field (Figure 1a,b).

FIGURE 1.

Development of Bi‐MOFs in the last decade. (a) Number of scholarly publications published in the last ten years related to Bi‐MOFs. (b) Categorization of research areas analyzed. (Data is collected from the Web of Science.).

2.1. The Chemical Properties of the Bismuth Cation

Prior to examining the intricacies of Bi‐MOF structures, it is crucial to understand the chemical characteristics of Bi cations and their coordination chemistry in liquid‐phase synthesis. Bismuth, recognized since antiquity as the heaviest stable metal element, is distinguished by its remarkably low toxicity—lower than that of sodium chloride for most bismuth compounds, attributed often to their low solubility. Therefore, metallic bismuth and its compounds are deemed biologically safe and non‐toxic. Positioned in the fifth group of the sixth period in the periodic table, the unique ground‐state electronic configuration (4f¹⁴5d¹⁰6s²6p³) of bismuth predisposes bismuth ions toward a predominantly stable +3 oxidation state in most complexes, denoted as Bi(III). In certain instances, bismuth can share two 6s electrons with a ligand to form bonds, showing a +5 valence, Bi(V), upon the loss of both outermost and sub‐outer electrons. Although both Bi(II) states (indicating an incomplete loss of outermost electrons) and Bi(V) states (representing a complete loss of outermost electrons) are observed in various compounds, they appear less frequently in Bi‐MOF structures compared to Bi(III), primarily due to their relative instability [33].

According to Pearson's hard‐soft acid‐base theory, Bi(III), characterized as a borderline acid, displays a marked preference for forming bonds with negatively charged, oxygen‐based ligands. Despite this affinity, leveraging Bi(III) for the development of porous MOFs is still significantly challenging [34]. Bi(III) ion, being the most stable oxidation state due to its low electronegativity and strong polarization capability, is nevertheless susceptible to rapid hydrolysis in an aqueous solution (with a pKa of 1.51) [35]. Common bismuth salts, such as bismuth nitrate (Bi(NO₃)₃·5H₂O) and bismuth trichloride (BiCl₃), undergo hydrolysis in neutral aqueous environments, leading to the formation of basic salt precipitation (BiONO₃) and bismuth oxychloride (BiOCl), respectively [36]. These unavoidable hydrolysis reactions foster the uncontrolled transformation of bismuth precursor into various bismuth‐oxygen nuclei or clusters, resulting in erratic nucleation and growth phases during the Bi‐MOFs synthesis [37]. In addition, bismuth coordination compounds may show an inert lone‐electron‐pair effect, akin to that observed in other high‐coordination main‐group elements. Specifically, the three outermost 6p electrons are engaged in bonding during the coordination process, giving rise to a stereochemicallyactive lone pair of 6s² electrons. This phenomenon is particularly pronounced in mononuclear coordination compounds, where the voids in bismuth's coordination sphere are filled by stereochemically active lone pairs. This results in the distortion of the coordination polyhedron and contributes to the asymmetry of the overall structure. Consequently, the synthesis process is more likely to yield dense, non‐porous bismuth coordination compounds or layered materials, rather than the sought‐after porous Bi‐MOF structures [33, 38].

2.2. The Coordination “Habit” of Bi(III)

When delving into the design and customization of MOFs, a critical factor to consider is the coordination “habits” of metal ions—specifically, how these metal ions interact with oxygen atoms from the organic ligands or solvents to form cores/clusters. Understanding these coordination behaviors is not only essential for deciphering the coordination mechanism but also pivotal in predicting the topological structure of MOFs.

Certain metal ions are known for their distinctive coordination tendencies. For example, aluminum ions, Al(III), tend to assemble into [Al₃(µ₃‐O)(COO)₆] clusters with six linkers, which can incorporate various ligands in different configurations, often resulting in cage‐like structures with high porosity and substantial specific surface area [39]. Zirconium ions are predisposed to form Zr₆O₈ clusters, serving as 6, 8, or 12‐connect nodes [40]. However, the coordination geometries of Bi(III) exhibit significant variability due to differences in the reaction system. Bi(III) ions can exhibit coordination numbers ranging from 3 to 10, leading to diverse and irregular Bi(III)‐based MOF structures, including pyramidal (coordination number (CN) = 5, [Bi(BPDC)₂H₂O]·xGuest, with H₂BPDC = 4,4′‐biphenyldicarboxylic acid) [41], octahedral (CN = 6, Bi‐mna, where H₂mna = 2‐mercaptonicotinic acid) [42], monocapped trigonal prism (CN = 7, [Bi(L)(HL)], where H₂L = camphoric acid) [43], distorted bi‐capped octahedral (CN = 8, Bi(pydc)₂·(H₃O⁺)(H₂O)_0.83, where H₂PyDC = 2,5‐pyridinedicarboxylic acid) [44] distorted square antiprism (CN = 8, (H₃O)[Bi₂(PTC)(HPTC)₂(H₂O)₂, where H₃PTC = 2,4,6‐pyridinetricarboxylic acid) [45], and triangular prism (CN = 9, Bi(BTB), where H₃BTB = 1,3,5‐benzenetrisbenzoate, denoted as CAU‐7) [46]. The variability of coordination cores in Bi(III)‐MOFs, influenced by the lone pair effect and diverse coordination environments, poses a challenge in predicting the structure from primary building units (PBUs) or secondary building units (SBUs). The connectivity and symmetry of SBUs are crucial factors that determine the topology of the MOFs network. While grasping the formation mechanism of SBUs in Bi‐MOFs may be challenging, it remains a topic of significant interest and is worthy of further investigation by researchers.

Given the remarkable diversity of Bi(III) coordination environments, inorganic building units, and framework architectures, establishing clear structure–topology relationships in Bi‐MOFs remains challenging. To facilitate a more systematic discussion and provide consistent structural guidance throughout the review, representative Bi‐MOFs are summarized in Table 1 according to their organic linker types, Bi(III) coordination environments, and framework topologies. Where topology information was not reported in the original literature, additional topology analyses were conducted using the available crystallographic data whenever possible. This summary enables readers to follow the structure–topology narrative more intuitively and highlights recurring design principles that emerge from the interplay between linker geometry and Bi(III) coordination chemistry.

TABLE 1.

Representative structure–topology relationships in Bi‐MOFs categorized by ligand type and Bi(III) coordination environment.

CN	4	5	6	7	8	9
Linker
			dia [47]
					sql [48] pcu [48]
			sra [49] hcb [49]
					‐ [50]
					‐ [51]
		‐ [52]
					‐ [45]
						htb [38]
				‐ [43]
						sql1 [53]
					‐ [54]
				esg [55]
					tfi [12]
					‐ [56]
	sql1 [57]				pcu [57]
					dia [58]
					btt [59]
			‐ [60]
						hcb [46]
			svd [17]
		‐ [61]
					dia [62]
					bex [63]
						rtl [64]
					sra [65] scu [66]
				‐ [67]

1.‘¹’ Topology assignments marked with superscript 1 were determined by the authors through independent ToposPro analyses based on the crystallographic information files (CIFs) reported in the original literature.

2.‘‐’ Topology not assigned. The original publication did not provide a topological classification, and subsequent analysis either yielded a topology that could not be matched to any net currently included in the RCSR database or a reliable topological simplification could not be established from the available crystallographic data.

2.3. Diverse Structures With Different Ligands

Researchers have adeptly imbued MOF materials with unique functionalities by utilizing various organic ligands, each bearing distinct characteristics. So far, approximately 70 different organic ligands have been employed in constructing Bi‐MOFs, as detailed in Figure 1. Notably, carboxylate ligands — encompassing both pure carboxylate and multifunctional organic molecules that contain carboxyl groups alongside other functional groups like hydroxyl, amino, and sulfhydryl —constitute a significant portion of these organic ligands, mirroring the trend observed in other transition metal‐based MOFs [68]. The prevalence of carboxylate ligands can be easily understood. These ligands boast a strong coordination ability with metals, and their length is optimally suited for creating cavities of appropriate sizes, conducive to forming a porous structure [69]. Additionally, the soft acid nature of Bi(III) ions manifests a pronounced affinity for organic linkers with multiple negatively charged functional groups based on oxygen atoms, promoting the chelation of carboxylate anions [34]. However, Bi‐MOFs incorporating phosphate and sulfonic acid ligands are notably less common. The low solubility of phosphate ligands and their tendency to yield dense and non‐porous layered structures with metal ions pose significant challenges to the controllable and rational design of Bi‐MOFs based on these ligands [70]. The pure organic sulfonate is a weak Brønsted base with weak coordination ability to many metal ions, so it is difficult to use for the synthesis of Bi‐MOFs. Yet, it's noteworthy that Bi‐MOFs featuring free sulfonic acid groups can be more readily synthesized using multifunctional organic ligands, benefiting from the synergistic effect of other functional groups (carboxyl and phosphate groups) [71]. Besides, aryl oxide (phenolic) ligands have been explored in recent years for Bi‐MOFs design, offering potent chelation with metal ions and endowing the resulting Bi‐MOFs products with enhanced chemical stability compared to those based on carboxylates [17]. Recently, the discovery of Bi‐MOFs based on various ring ligands marks a significant advancement in the field [72]. In summary, the functionality of ligands determines the diversity of products. Figure 2 summarizes the array of organic ligands utilized in constructing Bi‐MOFs to date. Our focus will be on Bi‐MOFs reported in the past five years, categorized according to their organic ligands with distinct functional groups, and we will delve into the diversification of their coordination structures and topological networks in detail (Table 2). The topology network will be presented in bold (e.g., sra, sql, pcu, hcb, esg) and space group symbols in italic (e.g., Pnma, P $\overline{1}$, P2₁/n, Pbcn, Cc, I2/a) throughout the following text.

FIGURE 2.

The array of organic ligands used to construct Bi‐MOFs to date.

TABLE 2.

Structure‐directing factors and design implications in Bi‐MOFs.

Structure‐directing factor	Typical consequence	Opportunity	Risk	Design implication	Structure‐directing factor
Bi(III) hydrolysis	oxo/hydroxo species formation	cluster‐based nodes	precipitation, dense phases	regulate pH, water content, and precursor speciation	Bi(III) hydrolysis
6s2 lone pair	distorted coordination geometry	unusual local functionality	lower topological predictability	avoid over‐reliance on rigid reticular assumptions	6s2 lone pair
Variable coordination number	broad structural diversity	flexible framework assembly	poor reproducibility	use constrained donor environments	variable coordination number
multitopic carboxylate ligands	higher framework dimensionality	possible 3D porosity	interpenetration or collapse	pair rigid ligands with controlled node chemistry	multitopic carboxylate ligands
multifunctional O‐donor ligands	stronger Bi coordination	better stability/functionality	added complexity	useful for stability‐oriented design	multifunctional O‐donor ligands

2.3.1. Carboxylate Ligands

2.3.1.1. Monocarboxylic Acid Ligand

Constructing topological networks using pure monotopic carboxylate ligands alongside metal ions is challenging due to their poor connectivity. Practically, achieving a porous MOF structure necessitates incorporating monotopic carboxylate ligands that contain additional functional groups, such as sulfonic acid groups and sulfhydryl groups. This strategy not only facilitates the construction of more functional Bi‐MOF structures but has also led to the development of only a handful of Bi‐MOFs using six reported monotopic carboxylate ligands to date.

A notable example is 2‐mercaptonicotinic acid (H₂MNA), a representative monotopic carboxylate ligand that coordinates with the Bi(III) atom through S, N chelation and O atoms bridge. This coordination results in the formation of {[Bi(MNA)₂]⁻·(HDMA)⁺}_n (DMA = dimethylamine), denoted as Bi‐mna, the first reported Bi‐MOF to incorporate mixed monotopic carboxylate, pyridine, and thiolate donors from a single ligand. In its asymmetrical unit, the Bi distorted octahedra are coordinated by two S atoms from sulfhydryl groups, two N atoms from pyridine, and two O atoms from carboxylate groups. The BiO₂N₂S₂ cores serve as 4‐connected nodes, linked by MNA²⁻ ligands as a bridge to form a diamond topology with a Schläfli symbol of (66) (Figure 3a) [47]. Particularly, the structure features two distinct homo‐chiral helices along the b‐axis and c‐axis, respectively, creating an unprecedented 3D network with helical channels of opposite chirality and different sizes along the a‐axis, i.e., large right‐handed helical channels and small left‐handed helical channels. Further explorations by Iram and co‐workers in 2018 introduced two other mercapto‐containing monocarboxylic acid ligands, 2‐mercapto‐3‐methyl‐4‐thiazoleacetic acid (H₂MMTA) and 4‐mercaptobenzoic acid (H₂MBA), leading to the synthesis of [Bi(HMMTA)₃]_n and [Bi(MBA)₃]_n, respectively [73]. Unfortunately, due to limited solubility, it is impossible to obtain high‐quality single crystals and perform a comprehensive crystal structure analysis. The only thing that can be identified is that both crystals are orthorhombic. In addition, the N₂ adsorption and desorption curves of these structures at 77 K present type II isotherms, indicative of mesoporous materials, and exhibited significantly large specific surface areas of 5316 and 2184 m² g⁻¹, respectively. Lei et al. further expanded the Bi‐MOF landscape by introducing Bi‐MMTAA, synthesized from H₂MMTA and triphenyl bismuth (c) [28]. Bi‐MMTAA crystallizes in the P2₁/m space group and features 1D chains composed of Bi and MMTAA via Bi‐O and Bi‐S bonds, showcasing a pyramid coordination core, where one benzene ring from BiPh₃, two S atoms from two MMTAA molecules, and two O atoms in ‐COOH from another two MMTAA molecules coordinate with each Bi(III). This example proves how the diversity of metal salts, in addition to organic ligands, can further enrich the variety of Bi‐MOF structures (Figure 3b).

FIGURE 3.

Representative helical structure and coordination unit of a Bi‐MOF constructed using a monocarboxylic acid ligand. (a) a₁) View of the right‐handed helix constructed from mna‐I ligands and Bi(III) atoms along the c axis; a₂) View of the right‐handed helix made of mna‐II ligands and Bi(III) atoms along the b axis; a₃) The small left‐handed helical channel along the a axis; a₄) The large right‐handed helical channel along the a axis; a₅) View of the supramolecular left‐handed helical chain; a₆) The supramolecular left‐handed helical chain in the large right‐handed helical channel. Color code: Bi, purple; C, white; N, blue; S, yellow; O, red; H, green. L/R: left/right. Reprinted with permission [47]. Copyright 2013, Royal Society of Chemistry. (b) (b₁ and b₂) Forward polarizing microscope images of Bi‐MMTAA. Projection of the single‐crystal structure of Bi‐MMTAA along the crystallographic b₃ a‐axis and b₄ c‐axis. b₅) Illustration of the pyramidal structure consisting of atoms that coordinate with Bi. (c) Ball‐and‐stick view of the asymmetric unit of BiPF‐4, including five DBS organic linkers and two Bi7 oxo‐hydroxo clusters. Bi is green, C is gray, O is red, and S is yellow. Reprinted with permission [28, 48]. Licensed under CC‐BY 4.0. Copyright 2022, American Chemical Society.

For the first time, Gómez‐Oliveira et al. utilized 3,5‐disulfobenzoic acid (3,5‐DSB), a monocarboxylic acid linker integrating carboxylic and sulfonic acid moieties, to synthesize the bismuth framework [Bi₁₄(µ₃‐O)₉(µ₄‐O)₂(µ₃‐OH)₅(3,5‐DSB)₅(H₂O)₃]·7H₂O, denoted as BiPF‐4 [48]. Polynuclear inorganic SBUs constructed from seven Bi atoms with different coordination environments are not common in Bi‐MOFs. Within its crystal structure, the asymmetric unit consists of 14 Bi(III) ions of different coordination numbers. Nine of these ions are octacoordinated (CN = 8), three have a CN = 7, one is connected to six oxygen atoms (CN = 6), and the remaining one coordinates with nine oxygen atoms (CN = 9), respectively. This complicated assembly is mainly attributed to the versatile binding mode facilitated by the 3,5‐DSB linker (Figure 3c). The Bi₇ clusters as 4‐connected nodes were interlinked by DSB³⁻ ligands, culminating in forming 2D networks aligned perpendicular to the [100] direction and showing a sql topological configuration. The extension of these layers along the [100] direction, using organic linkers, results in a 3D structure with a pcu topology. Such topological networks, while previously observed in other materials featuring Bi and sulfonate‐carboxylate linkers, were based on different inorganic clusters [57]. Evidently, the inorganic connectors played an essential role in regulating the connectivity of Bi‐based clusters within the synthesis system, ultimately leading to the emergence of diversified network architectures of Bi‐MOFs. This innovative work not only expands the structural diversity of Bi‐MOFs but also underscores the significance of carefully selecting linkers to tailor the connectivity and architecture of the resulting frameworks.

Leveraging the active coordination sites in the ligands is pivotal for the construction of high‐dimensional structures. A notable example of this principle in action is bismuth subgallate (BSG), traditionally known for its use as an active pharmaceutical ingredient, which has been proven to be a microporous 1D coordination polymer [74]. In its structure, Bi(III) ions are linked solely through the phenolic groups of gallic acid ligands, leaving the carboxylic acid group uncoordinated. Building on this concept, A. Ken Inge and co‐workers explored the extended coordination capabilities of gallate linkers(GA), through both phenolate and carboxylate groups, to develop 2 or 3‐periodic framework structures [49]. This exploration led to the synthesis of two novel bismuth gallate phases in methanol (MeOH) solution by extending the reaction duration, namely Bi(GA)(MeOH) and Bi(GA), respectively. Bi(GA)(MeOH) crystallizes in the polar orthorhombic space group P2₁2₁2₁. Its structure is characterized by a skewed trapezoidal bipyramidal BiO₆ core, coordinated by four bridging phenolate groups and a carboxylate group from the surrounding gallate, along with a methanol molecule. This coordination forms a zigzag rod‐shaped SBU, resulting in a 3‐periodic framework structure with a sra net topology. Upon further extending the reaction time, the coordination involving alcohol and carboxylate disappears, leading to the deprotonation and subsequent coordination of the non‐coordinating phenol oxygen. This process results in the crystallization of a new phase, Bi(GA), within the monoclinic space group P2₁ /c. In its structure, each Bi(III) ion is coordinated by six bridging phenolates to form a semi‐oriented pentagonal pyramid BiO₆ core. The emergence of Bi(GA) is attributed to the structural rearrangement from Bi(GA)(MeOH). Contrary to Bi(GA)(MeOH), where zigzag rod‐shaped SBUs are interconnected, in Bi(GA), these SBUs do not join through the third phenol oxygen of the gallate anions. Instead, they formed a layered structure, simplifying it into a 2‐periodic honeycomb hcb net. This discovery underscores the significance of reaction conditions and ligand coordination in dictating the structural outcomes of Bi‐MOF synthesis, offering new avenues for designing complex MOF architectures.

2.3.1.2. Ditopic Carboxylate Ligands

Among the two‐connected ligands, the terephthalic acid (H₂BDC) ligand is the simplest one. Adopting different solvothermal conditions, several Bi‐BDC materials with distinct structures have been synthesized [75, 76, 77]. Using N, N‐dimethylformamide (DMF) as the reaction solvent, a 3D supramolecular layered Bi‐MOF, {[Bi(BDC)_1.5(DMF)(H₂O)]·DMF}_n, was reported by Bai's group, which exhibits a triclinic crystal system and crystallizes in the P‐1 space group [50]. Within the structure, an irregular BiO_8B dodecahedron is observed, with Bi(III) ions coordinated through six carboxylic O atoms, one O atom from a water molecule, and one O atom from a DMF molecule. These BiO₈ dodecahedron cores are bridged through a BDC²⁻ ligand to another Bi(III), forming zigzag chains. Subsequently, these chains are connected through another BDC²⁻ ligand to form 2D sheets along the ac‐plane with hexagonal channels measuring 18.86 Å × 23.60 Å apertures. The formation of this 3D supramolecular structure is facilitated by the p‐π interaction between the N atom on the coordinated DMF molecule and the benzene ring in the adjacent layers (Figure 4a). Further investigations have explored the use of pre‐functionalized dicarboxylate linkers, incorporating nitro, amino, and hydroxyl groups, for the construction of functional Bi‐MOF [75, 78]. Recently, Lulich and co‐workers developed three Bi‐MOFs first through fluorinated ligand, tetrafluoroterephtalate (F₄BDC), including [Bi₂(OH)₂(H₂O)₄(F₄BDC)₂·2(H₂O)] (Bi₂‐F₄BDC), [Bi₄O₂(H₂O)₂(F₄BDC)₄(C₂H₆SO)·0.5(H₂O)] (Bi₄‐F₄BDC), and [Bi(H₂O)(F₄BDC)_1.5(DMSO)] (Bi_chain‐F₄BDC), respectively [51]. All three crystallize in the P $\overline{1}$ space group. Bi₂‐F₄BDC shows the SBUs based on a Bi₂ dimer, essentially two nine‐coordinated BiO₉ units bridged by dihydroxide. These Bi₂ dimers are bidirectionally connected into chains via bridging F₄BDC ligands. In Bi₄‐F₄BDC, a Bi₄O₂ tetramer cluster features seven F₄BDC ligands that straddle two Bi centers and one F₄BDC ligand that chelates one Bi and bridges to another. One additional dimethyl sulfoxide (DMSO) and water molecule bind to two Bi centers. The Bi₄O₂ tetramer clusters are linked in the bc‐plane via the benzene ring from the F₄BDC linker. In Bi_chain‐F₄BDC, eight‐coordinate BiO₈ cores create chains along the a‐axis, and the chains are bridged by F₄BDC ligands, resulting in a 3D network, via the Bi chains in the a‐axis and through the F₄BDC linkers in the b‐axis and c‐axis (Figure 4b). It is worth noting that there is a temperature‐mediated phase transition between Bi₂‐F₄BDC and Bi_chain‐F₄BDC, that is, the nucleation degree of bismuth nodes is increased by hydrolysis reactions, which highlights the flexibility of the system, with labile combination between linkers and nodes.

FIGURE 4.

Representative Bi‐MOF structures using ditopic carboxylate ligands, including coordination environments, 1D/2D/3D frameworks, ligand connection modes, interlayer interactions, and RE³⁺ substitution at Bi³⁺ sites. (a) a₁) Coordination environment and the space configuration of metal Bi(III); a₂) A one‐dimensional chain; a₃) 3D structure; a₄) The interaction of p‐π between two adjacent layers; a₅) 2D layered structure; a₆) The connection fashion of the ligand. Reprinted with permission [50]. Copyright 2022, Elsevier. (b) b₁) Bichain‐F₄BDC view emphasizing the pi‐pi stacking of the F₄BDC ligands and the linking between the Bi chains in the b‐direction. b₂) View of Bichain‐F₄BDC down the a‐axis in space‐filling representation to highlight the fluorine‐lined tunnels. The DMSO molecules that reside in these tunnels are not shown for ease of viewing. Reprinted with permission [51]. Licensed under CC‐BY‐NC‐ND 4.0. Copyright 2023, American Chemical Society. (c) Crystallographic a‐projection (Bi: green; O: red; N: blue; C: gray; H atoms omitted for clarity), showing 1‐D [Bi₂O₂]²⁺ chains. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) Reprinted with permission [52]. Copyright 2009, Elsevier. (d) d₁) Asymmetric unit with atom labeling of Bi‐PDC (all hydrogen atoms are omitted for clarity) and the thermal ellipsoids are drawn at the 50% probability level, d₂) The coordination environments of Bi³⁺ ions, (symmetry codes: #1 = ‐x, 2‐y, 2‐z; #2 = 1‐x, 1‐y, 1‐z; #3 = ‐1+x, +y, 1+z; #4 = +x, +y, 1+z) d₃) Monocapped square antiprism coordination atmosphere of Bi1, d₄) Connecting modes of PDC²⁻ ligand, d₅) 2D plane being parallel to ab‐plane, d₆) Hofmann‐type networks constructed by 2D Bi‐coordinated polyhedral layers with PDC²⁻ ligands viewed along b‐axis in Bi‐PDC. Reprinted with permission [85]. Copyright 2022, Elsevier Inc. (e) e₁) The coordination environments of the ligand 1,4‐ndc²⁻ and Bi³⁺ metal ion, e₂) The 2D layer parallel to the ac plane with hydrogen atoms, guest DMF molecules, and Hdma cations omitted for clarity. In situ RE⁺³⁺ ion replacement happens at the Bi³⁺ sites. Symmetry codes: #1 x, ‐y + 1/2, z + 1/2; #2 ‐1 + x, y, z; #3 x, ‐y + 1/2, z – 1/2; #4 x + 1, y, z. Reprinted with permission [53]. Copyright 2019, Royal Society of Chemistry.

Additionally, heterocyclic dicarboxylate ligands have been widely used in the development of Bi‐MOF, with 3,5‐pyridine dicarboxylic acid standing out as one of the most impactful. It was utilized by Tran et al. in 2009 to construct Bi₂O₂[NC₅H₃(CO₂)₂] (denoted as ARL‐3) [52], making the inception of Bi‐MOFs based on heterocyclic dicarboxylate ligands. ARL‐3 features an asymmetric unit containing two crystallographically distinct Bi centers, Bi₍₁₎O₅ and Bi₍₂₎NO₄, both showing the umbrella‐shaped geometry typical of lower p‐block metals. The structure is characterized by 1D cationic [Bi₂O₂]²⁺ chains extending parallel to the c‐axis, with pyridine dicarboxylate moieties bridging these chains via Bi‐carboxylate oxygen bonds to form a 3D network (Figure 4c). Initially, research on Bi‐MOFs often focused on dipicolinic acid ligands due to the variability of carboxylic acid substituent positions on the pyrrole ring. However, these structures generally demonstrated limited porosity [79, 80, 81, 82, 83, 84]. Subject to the unsatisfactory performance of the resulting products, the use of pyridine dicarboxylic acid as ligands in designing novel Bi‐MOF structures is no longer the focus among researchers. Over the past five years, only one new Bi‐MOF based on pyridine dicarboxylic acid has been explicitly reported. Utilizing 2,4‐pyridine dicarboxylic acid (2,4‐H₂PDC), Guo et al. synthesized [Bi(PDC)(OH)] with a 3D pillared‐layer structure [85]. Each Bi(III) ion in [Bi(PDC)(OH)] is connected to one nitrogen atom and six carboxylate oxygen atoms from five PDC²⁻ ligands, alongside two oxygen atoms from two bridging hydroxides, adopting a distorted monocapped square antiprism sharing edges to form a 2D coordination polyhedral monolayer parallel to the ab‐plane. The pillared‐layer framework arises from the connection of adjacent coordinated polyhedral monolayers along the c‐axis via PDC²⁻ ligands (Figure 4d). Beyond pyridine dicarboxylic acid, ligands such as imidazole dicarboxylic acid and thiophene dicarboxylic acid have been employed in the construction of Bi‐MOF, further illustrating the versatility and potential of heterocyclic dicarboxylates in MOF synthesis [83, 86].

Apart from phenyl and heterocyclic ligands, naphthalene and biphenyl‐based dicarboxylate ligands have been explored in the synthesis of Bi‐MOFs, showcasing the versatility of aromatic dicarboxylates in constructing complex MOF architectures. In 2019, Lu et al. reported the synthesis of [(HDMA)Bi(1,4‐NDC)₂(DMF)]_n·nDMF using π‐conjugated 1,4‐naphthalenedicarboxylic acid (1,4‐H₂NDC) as the ligand. This compound crystallizes in the space group P2₁/c [53], where each Bi center is nine‐coordinated by nine oxygen atoms. Eight of these oxygen atoms come from four carboxylate groups of four distinct 1,4‐H₂NDC ligands, and the remaining one is from a coordinated DMF molecule, resulting in a deformed tricapped trigonal prism geometry. The resulting structure features infinite plicate 2D layers stacked along the b‐axis, constructing a 3D supramolecular network through hydrogen bonds (Figure 4e). Norbert Stock et al. in 2018 successfully synthesized a series of Bi‐MOFs using 4,8‐disulfonyl‐2,6‐naphthalenedicarboxylic acid (H₄DSNDC) through high‐throughput methods [54]. These MOFs include [Bi₆O₆(OH)₂(H₂O)₄(DSNDC)], [Bi₂(OH)₂(DSNDC)], [Bi₈O₇(OH)₂(H₂O)₂(DSNDC)₂], [Bi₇O₅(OH)₃(H₂O)₄(DSNDC)₂]·4H₂O, [Bi₂(OH)₂(H₂O)₂(DSNDC)(H₂DSNDC)] and [Bi₆O₄(OH)₄(H₂O)₁₂(H₂DSNDC)₃]·xH₂O). Notably, three previously unreported IBUs were identified within these products, including {Bi₃O₃(OH)}_∞, Bi₁₆O₁₄(OH)₄ clusters, and Bi₇O₅(OH)₃ clusters, highlighting the innovative potential of using complex dicarboxylate ligands for MOF synthesis. Recently, Iram and co‐workers reported a new MOF based on biphenyl‐4,4′,‐dicarboxylic acid (H₂BDA) ligand [87]. Unfortunately, detailed information on the crystal structure remains undisclosed in the original text due to unsuccessful attempts at growing single crystals. This challenge underscores the complexities involved in the crystallization of Bi‐MOFs and the ongoing quest for suitable conditions that facilitate the formation of well‐defined crystalline structures.

2.3.1.3. Tritopic Carboxylate Ligands

Benzene‐1,3,5‐tricarboxylic acid (H₃BTC), a fundamental three‐connected ligand, has been extensively utilized in the synthesis of Zr, Cu, and Al‐MOFs due to its ability to construct complex structures [39, 40, 88]. This ligand has also found its way into the realm of Bi‐MOFs. Among the H₃BTC‐based Bi‐MOFs reported to date, CAU‐17 remains the most comprehensively characterized porous framework. Its experimentally validated porosity, together with its unusually complex topology, makes it a particularly informative system for understanding structure–porosity relationships in Bi‐MOFs.

CAU‐17 represents a significant focus within the larger Bi‐MOFs family due to its unique structural features. Recently, several porous structures based on this ligand have also been reported, but their permanent porosity remains to be verified. For instance, [Bi(BTC)(H₂O)]·H₂O, prepared by Sun and co‐workers, crystallizes in the orthorhombic space group Pnma. In this structure, the Bi(III) ion exhibits a coordination number of nine, coordinating with eight carboxyl oxygen atoms from five H₃BTC ligands and an additional oxygen atom from a coordinated water molecule. Notably, a carboxyl oxygen atom coordinates simultaneously with two Bi(III), forming a symmetrical dimer Bi₂O₁₈ unit with a shared edge. These dimer units are connected by BTC³⁻ into a 3D network possessing two types of micropore channels with sizes of 12.449 × 8.553 Å and 7.970 × 8.838 Å, respectively, indicating the versatility of H₃BTC in generating porous architectures [11]. In 2021, Åhlén et al. reported UU‐200, a trimester‐based structure that also crystallizes in the orthorhombic Pnma space group but has a different coordination mode compared to [Bi(BTC)(H₂O)]·H₂O. UU‐200 exhibits a 3D structure containing three Bi(III) ions and four BTC³⁻ anions within its asymmetric unit, with ten oxygen atoms from six distinct BTC³⁻ anions coordinating with Bi₁ ³⁺ and Bi₃ ³⁺ cations, and nine oxygen atoms from five different BTC³⁻ anions coordinating to Bi₂ ³⁺ cations. Each of the four symmetry‐independent BTC³⁻ anions occupies a different chemical environment, involving a bidentate chelate mode and additional individual bonds (Figure 5a). This complex arrangement of ligands and metal ions culminates in a 3D pore system with two types of channels, about 3.49 and 3.42 Å, respectively [89].

FIGURE 5.

(a) Coordination environment of the benzene‑1,3,5‑tricarboxylate (BTC³⁻) moieties in UU‑200 synthesized from tritopic carboxylate ligands. a₁) Two carboxylate groups of the BTC³⁻‐anions coordinate with the Bi₂ ³⁺ and Bi₁ ³⁺ in a bidentate chelate mode, and the third carboxylate group forms two bonds to Bi₃ ³⁺ and one bond to Bi₂ ³⁺. a₂) All three carboxylate groups of the BTC³⁻‐anions coordinate with the Bi₂ ³⁺ and Bi₃ ³⁺ in a bidentate chelate mode. a₃) One carboxylate group of the BTC³⁻‐anions coordinates with the Bi₃ ³⁺ in a bidentate chelate mode, and the remaining two form chelates to one Bi₁ ³⁺ and coordinate with one additional bond to a second symmetry‐dependent Bi₁ ³⁺. a₄) One carboxylate group of the BTC³⁻‐anions coordinates with the Bi₁ ³⁺ in a bidentate chelate mode, and two carboxylate groups chelate to a Bi₂ ³⁺ and form an additional bond to a Bi₃ ³⁺. Reprinted with permission [89]. Licensed under CC‐BY. Copyright 2021, Elsevier. (b) View along the a‐axis for 4 (top left) and the corresponding augmented net (top right) along the same viewing direction, as well as similar representations along the b‐axis. For the augmented nets, two‐coordinated linkers are represented as solid lines and three‐coordinated linkers as yellow triangles. The bismuth dimers are shown as distorted purple dodecahedra. Reprinted with permission [55]. Copyright 2020, American Chemical Society.

The influence of solvent systems on MOF synthesis is well‐documented, with different solvents capable of inducing distinct coordination environments. For the first time, Sanchez‐Sala et al. used hot acetic acid (HAc) as a solvent to synthesize [Bi(HBTC)(Ac)], with an inorganic node based on a BiO₇ core. In the structure, four partially deprotonated HBTC²⁻ ions and one acetate ligand coordinate in a monodentate manner to Bi(III), while another acetate ligand engages in chelation. Bismuth atoms are arranged into chains parallel to the b‐axis, then assembled into sheets parallel to the (−1 0 1) plane by a triple bridge. The establishment of a 3D supramolecular network through hydrogen bonding that links the protonated carboxylic acid of the HBTC²⁻ ligand from one layer to the next [90].

Generally, symmetric expansion, partial substitution, and backbone desymmetrization are frequently utilized in designing tritopic carboxylate ligands to enhance ligand functionality and structural diversity in MOFs [91]. These methods are critical for developing MOFs with desired properties, including permanent porosity. The prime example of symmetric expansion in a ligand for the construction of Bi‐MOFs with permanent pore size is CAU‐7, using the 1,3,5‐benzenetrisbenzoic acid (H₃BTB). This ligand is an extension of the simplest tricarboxylic acid ligand, H₃BTC, demonstrating how expanding the symmetry of a ligand can lead to the formation of complex Bi‐MOF structures [46]. Researchers have also explored ligand functionalization and desymmetrization by altering the type of benzene rings or the relative position of tethering carboxylate groups, for instance, substituting H₃BTC with 1,2,4‐benzenetricarboxylic acid (H₃Tri) and 2,4,6‐pyridinetricarboxylic acid (H₃PTC), or H₃BTB with triazine‐2,4,6‐triyltribenzoic acid (H₃TATB) and 2‐(4‐carboxy‐3‐aminophenyl)‐4,6‐bis(4‐carboxyphenyl)‐1,3,5‐triazine (H₃TATB‐NH₂), to achieve diverse structural motifs [26, 92]. In 2018, Norbert Stock's group reported four new Bi‐based coordination polymers using H₃PTC [45]. [Bi(PTC)(H₂O)₂] crystallizes in the triclinic space group P $\overline{1}$, featuring each Bi(III) ion coordinated by one N atom from a PTC³⁻ molecule and eight O atoms from three H₂O molecules and three PTC³⁻ molecules, resulting in distorted monocapped square antiprism BiNO₈ polyhedra. The structure consists of infinite zigzag chains of edge‐sharing BiNO₈ polyhedrons through µ‐O atoms interconnected by fully deprotonated PTC³⁻ linkers. Similarly, (H₃O)[Bi₂(PTC)(HPTC)₂(H₂O)₂], also crystallizing in the triclinic space group P $\overline{1}$, contains chains of alternating edge‐ and corner‐sharing BiNO₇ and BiN₂O₆ polyhedra linked through µ‐O atoms. These chains are connected by hydrogen bonds to form a 2D layered framework. The inorganic nodes in α‐[Bi(PTC)] are based on Bi₂N₂O₁₂ dimers of edge‐sharing BiNO₇ polyhedra. Each dimeric SBU is coordinated by ten PTC³⁻ linkers to 24 other SBUs to construct a 3D network. β‐[Bi(PTC)], a polymorph of α‐[Bi(PTC)], crystallized in the orthorhombic space group Pbcn, contains dimers of edge‐sharing BiNO₈ polyhedra that form 1D chains. These chains are further interconnected into a 3D network through ligands.

Linker desymmetrization, achieved by incorporating additional benzene rings into the ligand structure, has paved the way for the synthesis of novel and complex Bi‐MOFs. A notable advancement in this area is the creation of the first breathable, flexible Bi‐MOF by A. Ken Inge and co‐workers, based on biphenyl‐3,4′,5‐tricarboxylic acid (H₃BPT) [55]. The MOF, SU‐100 (Bi(BPT)·2MeOH), indicates a plate‐shaped crystal and is distinguished by its inorganic building units based on a Bi₂O₁₂ dimer. This dimer is regarded as a composition of two seven‐coordinated BiO₇ units bridged by two oxygen atoms in an edge‐sharing manner. SU‐100 crystallized in the space group of I2/a. The unique structure of SU‐100, where Bi₂O₁₂ dimers act as 6‐connected nodes connected by BPT³⁻ ligands as the 3‐connected linkers, forms a (3,6)‐connected 3D network with the esg topology. The SU‐100 possessed a BET surface area of 385 m²·g⁻¹ and demonstrated solvent‐related flexibility, including changes in unit cell volume and solvent‐accessible void volume. This flexibility is attributed to the relatively fragile rhombic part of the Bi₂O₁₂ dimers and only an even number of cycles in the esg topological network. In the same work, a series of Bi‐based dense coordination polymers, Bi₂O₂(HBPT), Bi(OH)(H₂BPT)₂(H₂O)₂·H₂O, and Bi₂(HBPT)₃(H₂O)₃·H₃BPT, were synthesized using the BPT³⁻ ligand in an aqueous solution through hydrothermal reactions. Among these, Bi₂O₂(HBPT) features two crystallographically independent Bi(III) centers with coordination numbers of six and seven, forming distorted octahedra and distorted pentagonal bipyramids, respectively. The BPT³⁻ ligands, acting as the 3‐connected linkers, contribute to a 3,5,5‐c network with the point symbol of {44.64.72}{44.65.7}₂{63}. Another dense phase structure, Bi(OH)(H₂BPT)₂(H₂O)₂·H₂O, is a bismuth‐oxo rod, in which each Bi(III) ion is eight‐coordinated by O atoms from two carboxylate groups of H₂BPT molecules, two terminal water molecules, and two µ ₂‐hydroxide ions. By adjusting the ratio of the reactants, large prismatic crystals of a dense phase, Bi₂(HBPT)₃(H₂O)₃·H₃BPT, were obtained, crystallizing in the P1 space group. This structure presents two distinct crystallographically independent Bi(III) ions with coordination numbers of six and seven. In its structure, the three‐coordinated linkers can be regarded as a 3‐connected node, and the two‐coordinated molecules can be viewed as an edge between eight‐coordination nodes of a dimer, leading to a (3,3,8)‐connected tfz‐d topological network (Figure 5b).

The diverse configurations of tritopic carboxylate ligands facilitate the synthesis of Bi‐MOFs with complex structures. Advancing the complexity of tricarboxylate ligands is deemed beneficial for crafting novel Bi‐MOF structures featuring nondefault topologies.

2.3.1.4. Tetratopic Carboxylic Ligands

Tetratopic carboxylate ligands, with their multiple coordination sites and complex coordination patterns, surpass both ditopic and tritopic carboxylate ligands in terms of structural diversity. Their extensively dynamic configurations, including square, rectangular, and tetrahedral geometries, have led to their frequent employment in tailoring porous Bi‐MOFs in recent years. These ligands, derived from the basic tetratopic carboxylic ligand pyromellitic acid (H₄Pyr), are generally divided into two categories based on the phenyl backbone. One category features an inner functional core tethering outer tetraphenyl rings (such as 1,3,6,8‐tetrakis(p‐benzoic acid)pyrene (TBAPy), 1,2,4,5‐tetrakis‐(4‐carboxyphenyl)benzene (TCPB), 4′,4″,4‴,4‴′‐(ethene‐1,1,2,2‐tetrayl)tetrabiphenyl‐4‐carboxylic acid (TCBPE), 4,4′,4″,4‴‐methanetetrayltetrabenzoic acid (MTBA), and meso‐tetrakis(4‐carboxyphenyl)porphyrin (TCPP)), while the other involves a biphenyl backbone tethering tetracarboyxlate groups (such as biphenyl‐3,3′5,5′‐tetracarboxylic acid (BPTC), [1,1′:4′,1″]terphenyl‐3,3″,5,5″‐tetracarboxylic acid (TPTC), and 3,3′,5,5′‐azobenzenetetracarboxylic acid (ABTC)). The first category is particularly suited for constructing high‐dimensional Bi‐MOFs due to its central functionality or the moderate flexibility of pendant benzoate groups. A prime example is H₄TBAPy, which has led to three unique structures with permanent porosity, to be discussed in detail in the following chapter. Another tetracarboxylic acid ligand, H₄TCPB, an expanded version of H₄Pyr, was used by Köppen et al. to construct three different Bi‐MOFs in 2018 [63]. The synthesis underscored the critical role of solvent effects, demonstrating how the physical and chemical properties of the solvent affect the reaction equilibrium and metal center coordination mode and, ultimately, determine the structure of the reaction product. For instance, [Bi₂(TCPB)(H₂TCPB)(H₂O)]·xH₂O (CAU‐31) was obtained in pure methanol, (NH₂(CH₃)₂)[Bi(TCPB)(H₂O)]·xH₂O (CAU‐32) in a methanol‐DMF mix, and [Bi₄(O)₂(OH)₂(H₂TCPB)(TCPB)(H₂O)₂]·xH₂O (CAU‐33) in pure DMF, with only CAU‐33 exhibiting permanent porosity. CAU‐33 contains two crystallographically unique BiO₅ irregular polyhedron coordination cores in the crystal structure, forming a bismuth oxide rod along the c‐axis in an angle‐sharing manner. This structure exhibits two well‐defined 1D lozenge‐shaped channels of 9.5 × 4.6 Ų and 4.4 × 4.1 Ų, respectively, constructed by IBU of bismuth oxygen rods and TCPB4⁻ ligands in two different connection modes (two‐connected and four‐connected). The BET surface area of CAU‐33 is measured to be 450 m²·g^−1, and the total pore volume is 0.21 cm³·g⁻¹. Meanwhile, CAU‐31 and CAU‐32, both dense‐phase structures, were also obtained. CAU‐31 contains only one type of BiO₈ coordination sphere, in which six oxygen atoms belong to three carboxylic acid groups and the other two oxygen atoms are from water molecules. In the underlying 2D net, each BiO₈ polyhedron forms a bex topology network by chelating with three completely deprotonated carboxyl groups from three different TCPB⁴⁻ ligands. This results in a 3D interlocked structure with uncommon inclined interpenetration. CAU‐32 contains a Bi(III) coordinated by eight oxygen atoms of four carboxylic acids and one oxygen atom of water, forming an anionic 3D interpenetrating framework.

Large sterically hindered tetracarboxylic ligands can effectively mitigate framework interpenetration, as demonstrated by the synthesis of the non‐interpenetrating structure [Bi₂(C₁₆H₆O₈)1.5(H₂O)₂]·(DMF)_3.5·(H₂O)₃, denoted as NOTT‐220, by Mathew Savage et al. using 3,3′,5,5′‐tetracarboxylic acid [12]. In a more recent study, Zhu et al. used 5′,‐(4‐carboxyphenyl)‐[1,1″,:3″,1″‐terphenyl]‐3,4″,5‐tricarboxylic acid (H₄CPTTA) to synthesize a robust Bi‐MOF featuring a two‐fold interpenetrated structure with saturated coordinated mononuclear SBUs ([Bi(COO)₄]), known as JLU‐MOF120. This MOF exhibits excellent water and pH stability [58]. JLU‐MOF120 crystallizes in the monoclinic system within the Cc space group. In its structure, each [Bi(COO)₄] SBU consists of one eight‐coordinated Bi(III) linked to four carboxylates from four distinct CPTTA4⁻ ligands. These [Bi(COO)₄] SBUs, as 4‐connected nodes, are joined by CPTTA4⁻ ligands, acting as 4‐connected linkers, to give rise to a 4‐connected 3D interpenetrated framework with a dia topology (Figure 6a). Notably, JLU‐MOF120 presents three different twisted channels, with pore sizes measuring 2.9 × 6.4 Ų, 4.7 × 7.3 Ų, and 0.8 × 4.6 Ų along different directions, respectively (Figure 6b).

FIGURE 6.

Representative structure of Bi‐MOF based on tetratopic carboxylic ligands. (a) Structural description of JLU‐MOF121. a₁) Simplification of the [In(COO)₄] SBU and CPTTA⁴⁻ ligand; a₂) Connolly surface areas of a single net; a₃) Ball and stick view of JLU‐MOF121; a₄). Polyhedral view of JLU‐MOF121; a₅) Dia topology network. Color scheme: In = orange, C = gray, and O = red. (b) Three different pore sizes along the different directions of JLU‐MOF120. Reprinted with permission [58]. Copyright 2022, American Chemical Society. (c) Representation of the asymmetric unit of IEF‐2. Reprinted with permission [56]. Copyright 2019, Royal Society of Chemistry.

Another kind of tetracarboxylic acid ligands with biphenyl or terphenyl backbones generally demonstrate a high degree of structural flexibility, endowing Bi‐MOFs with diverse structural dimensions. Li and co‐workers employed the H₄BPTC ligand to develop two distinct Bi‐MOF structures, diverging from their previously synthesized [Bi₂(BPTC)_1.5(H₂O)₂]·(DMF)_3.5·(H₂O)₃ (NOTT‐220‐solv), by varying the temperature and duration of the synthesis reaction. The resulting structures, [Bi(BPTC)]·Me₂NH₂·(DMF)_1.5 (MFM‐221), and [Bi₂(HBPTC)·(H₂BPTC)·(DMF)·(OH)] (MFM‐222), exhibit unique structure configurations [93]. MFM‐221 crystallizes in the C2/c space group of the monoclinic crystal system, where each Bi(III) ion is coordinated by nine oxygen atoms from five ligands, with eight in chelating mode and one in monodentate mode. Adjacent Bi(III) ions are bridged by two carboxylates to form a binuclear Bi₂O₁₆ center, connected by ligands into an anionic, non‐interpenetrating framework. The structure of MFM‐222 has two central Bi(III) ions (Bi₁, Bi₂). Bi₁ is coordinated with eight oxygen atoms from six ligands and a hydroxyl group (µ₂‐OH), four in bidentate and four in monodentate mode, while Bi₂ is connected with nine O atoms of five BPTC ligands, one µ₂‐OH, and one DMF molecule. Bi₁ and Bi₂ are joined by one carboxylate oxygen from a BPTC ligand and µ₂‐OH to form binuclear centers, which are further connected through BPTC linkers, yielding a non‐interpenetrated and non‐porous framework. Attempting an isoreticular expansion of NOTT‐220‐solv, A. Ken Inge's group constructed a coordination polymer, Bi₆O₄(H₂TPTC)_5, using the symmetrically expanded ligand H₄TPTC. However, the goal of isoreticular expansion was not realized, resulting in a structure with diverse connectivity and a unique (3,3,16)‐c network, distinct from NOTT‐220's 3,6‐connected network [55].

Additionally, employing H₄AzoBTC as the ligand, another symmetric rigid organic spacer bearing four carboxylic acid groups, Vilela et al. prepared [Bi₄(HAzoBTC)₂(AzoBTC)(OH)₂(H₂O)₄]·7H₂O (IEF‐2), which crystallizes in the triclinic space group P $\overline{1}$ [56]. In its structure, the asymmetric unit contains two Bi(III) ions, both of which are coordinated with eight oxygen atoms, six from the carboxyl group, one from the terminal H₂O molecule, and one from the µ₂‐OH. The edge‐sharing BiO₈ polyhedra exhibit 1D chain structures along the a‐axis. These chains, linked through the ligands, compile into a 3D network with small channels of 3–4 Å (Figure 6c).

2.3.1.5. Hexatopic Carboxylate Ligands

Utilizing hexatopic carboxylic acid linkers for the construction of Bi‐MOFs has been relatively rare. Only such a linker, 2,4,6‐tris(3,5‐dicarboxylphenylamino)‐1,3,5‐triazine (H₆TDPAT), has six carboxyl functional groups and three flexible arms, enabling its role as a µ‐bridge to assemble bismuth PBUs or SBUs into 3D frameworks. Employing the flexible triazine‐polycarboxylic linker H₆TDPAT as the ligand, You et al. prepared H₃O·[Bi(TDPAT)_2/3]·3H₂O·4DMA (Bi‐TDPAT), which crystallizes in the cubic space group I432 [59]. In its structure, each Bi(III) atom is coordinated with eight carboxylate O atoms, forming a BiO₈ dodecahedral configuration. The coordination mode is µ₆‐η_O ²η_O ²η_O ²η_O ²η_O ²η_O ², and the BiO₈ cores as 4‐connected nodes are connected by TDPAT6⁻ ligands as the 6‐connected linkers to give rise to a (4,6)‐connected 3D extended network with a novel topological type btt (Figure 7a). In addition, the structure features a bilayer cage with a distorted snub cube shape, incorporating an inner cavity of 20.8 Å. Expanding on this theme, Francoise M. Amombo Noa et al. developed a series of MOFs using ubiquitous metals and the 1,2,3,4,5,6‐hexakis(4‐carboxyphenyl)‐benzene (H₆CPB) ligand, including [Bi₂(CPB)(HOAc)₂(DMF)₂]·2DMF (CTH‐6) using Bi(III) as a metal source [64]. In its structure, each Bi atom is coordinated to nine oxygen atoms—six from the carboxyl group of three cpb6⁻ ligands by chelation, two from DMF molecules, and one from an acid molecule. Each CPB6⁻ linker connects with six BiO₉ cores in a 6‐connected mode, and each BiO₉ core is linked by three CPB6⁻ linkers in a 3‐connected node, resulting in a (3,6)‐connected rtl topological network, displaying a unique arrangement composed of hexagons and triangular pyramids (Figure 7b). The inaccurate S_BET (67 m² g⁻¹) of CTH‐6 was calculated from the N₂ adsorption isotherms of the mixed phase because of the failure in obtaining the pure phase.

FIGURE 7.

(a) Structure of Bi‑TDPAT synthesized using hexatopic carboxylate ligands. a₁) Spatial configuration of the Bi‐O polyhedron. a₂) Coordination environment of Bi³⁺ (#1 y + 1/2, x – 1/2, ‐z + 1/2; #2 ‐z + 1, y, x; #3 y + 1/2, ‐z + 1/2, ‐x + 1/2). a₃) Linking pattern of the TDPAT ligand. a₄) Three‐dimensional framework structure. Hydrogen atoms are not shown for clarity. (b) CTH‐6 [Bi₂(cpb)(HOAc)₂(dmf)₄]: Bi‐O bonds are on the order of 2.22 Å to 3.06 Å. Ellipsoids are drawn at 50% probability. Reprinted with permission [59, 64]. Copyright 2022, 2020, American Chemical Society.

The exploration of mesoporosity or cage‐like structures in Bi‐MOF is limited, highlighting the need for further research into the use of multitopic carboxylate ligands, particularly hexa‐, octa‐, and dodeca‐carboxylates, to diversify the topologies and porosities of Bi‐MOF. This area remains ripe for investigation, promising exciting developments in the synthesis and application of Bi‐MOFs.

2.3.2. Phosphate Ligands

Phosphate‐based MOFs (P‐MOFs) are drastically less prevalent in the research literature compared to their carboxylate‐based equivalents, mainly due to the poor solubility of the phosphate ligands. Compounds incorporating phosphate ligands tend to form layered, dense‐phase products with low crystallinity, complicating their structure analysis through single‐crystal PXRD and seriously impeding the development of P‐MOFs [70]. Although the advent of three‐dimensional electron diffraction (3D‐ED) has transcended the limitations of single‐crystal structure analysis, allowing for the acquisition of accurate atomic information from nanocrystals, submicron crystals, and samples with multiple phases, the number of reported Bi‐MOFs constructed solely from pure phosphate ligands remains exceedingly sparse. To date, only one Bi‐MOF constructed with pure phosphate ligands has been reported: Bi₂(Py(PO₃)₂(PO₃H)₂ (IEF‐7), synthesized using a lab‐made pyrenyl derivative, pyrene tetraphosphonic acid (Py(PO₃H)₂), as the ligand by Salcedo‐Abraira and co‐workers [60]. IEF‐7 crystallizes in the triclinic crystal system, with the $\overline{P}$ 1 space group. Structurally, each Bi atom in IEF‐7 is coordinated by six phosphate O atoms, forming a distorted pentagonal pyramidal geometry (Figure 8). The pyramids share an edge and exhibit inversion symmetry, resulting in Bi₂O₁₀ clusters. Within each phosphate ligand, two phosphate groups coordinate with two Bi cations through two monodentate O atoms, while the remaining two phosphonate groups coordinate with three Bi cations through three completely deprotonated O atoms. Each cluster is connected to five different phosphonate linkers to yield an unprecedented trinodal 3,3,5‐connected network. IEF‐7 displayed a BET surface area of 34 m² g⁻¹ and a pore size of 3.64 Å, close matching the kinetic diameter of nitrogen molecules.

FIGURE 8.

SBU of IEF‑7 was synthesized using phosphate ligands. Bi, P, C, and O are represented in purple, orange, brown, and red, respectively. Hydrogen atoms are omitted for clarity. Reprinted with permission [60]. Copyright 2023, American Chemical Society.

Theoretically, three oxygen atoms of a phosphate group can coordinate with a metal in any of its protonated states, providing a variety of coordination modes and the potential for more functionally diverse products. However, the intentional design of such a system is rare. Furthermore, considering the stereochemically active lone pair effect characteristic of bismuth ions, designing Bi‐MOFs with controllable structures based on phosphate ligands remains a significant challenge.

2.3.3. Phenolate Ligands

The phenolic hydroxyl group (‐OH) present in phenolic ligands acts as an exemplary electron donor due to its lone pair of electrons. This distinctive character enables effective coordination with metal cations, particularly with phenolates exhibiting stronger basicity compared to carboxylates, resulting in a strong bond between the O atom and the Lewis acidic Bi(III) center. Consequently, compounds formed from such interactions tend to exhibit enhanced chemical stability and thermal durability. Empirical evidence suggests that combining metal cations with high charge density with ligands featuring high pKa values is a strategic approach for creating highly chemically stable MOFs [94]. In addition, bioactive phenolic hydroxyl ligands, being predominantly derived from environmentally benign sources, promote the development of sustainable Bi‐MOFs. A significant milestone in this area was achieved with the successful utilization of Ellagic acid, a natural plant extract, for the construction of high‐quality Bi₂O(H₂O)₂(C₁₄H₂O₈)·nH₂O crystals by A. Ken Inge et al., and named SU‐101 [17]. This breakthrough marked the first instance of Bi‐MOF construction based on phenolic phytochemicals, which will be mentioned separately in the next chapter. Following this, Gao et al. obtained a conductive catecholate‐based Bi‐MOF, denoted as Bi‐HHTP, using 2,3,6,7,10,11‐hexahydroxytriphenylene (HHTP) as the ligand. Bi‐HHTP, crystallizing in the monoclinic P2₁/n space group, displays a 1D zigzagging corrugated chain structure of unsaturated BiO₅ units, which are coupled by the Bi‐O bonds to form a 3D framework [61]. In its structure, each HHTP ligand coordinates with seven Bi(III) ions at various angles, with Bi(III) ions engaging in coordination with distorted catechol groups, leading to an unusual, unsaturated coordination mode akin to a distorted tetragonal pyramid.

2.4. Representative Bi‐MOFs

A recurring issue in the Bi‐MOF literature is that crystallographic voids or channel‐like motifs do not necessarily indicate permanent porosity. In many cases, frameworks that appear open in the crystal structure may partially collapse, rearrange, or lose accessibility after guest removal because of labile coordination environments, hydrolysis‐derived inorganic motifs, or solvent‐dependent structural features. Therefore, structural description alone is not sufficient to assess the practical significance of a reported Bi‐MOF.

In this review, permanent porosity is considered on the basis of available experimental evidence rather than crystallographic analysis alone. Particular attention is given to framework activation procedures, low‐pressure gas sorption measurements, and post‐activation structural characterization when such information is available [95]. Framework stability under relevant operating conditions is also considered when evaluating the significance of reported porosity.

As summarized in Table 3, a number of representative Bi‐MOFs have been reported with varying levels of porosity‐related evidence. Some frameworks are supported by activation procedures and gas sorption measurements, whereas others are primarily characterized by crystallographic voids or surface‐area data without complete porosity validation. Distinguishing between these cases provides a more transparent basis for discussing structure–porosity relationships in Bi‐MOFs.

TABLE 3.

Representative Bi‐MOFs classified according to the experimental evidence supporting porosity‐related claims.

Bi‐MOF	Linker type	Node/structural feature	Activation & low‐pressure sorption	BET/Pore volume/Stability	Refs.
Experimentally Validated Permanent Porosity
CAU‐7	BTB, tritopic carboxylate	BiO9 chains, 3D honeycomb	160°C, vacuum, 12 h	1150 m2 g−1; 0.43 cm3 g−1; stable to 380°C	[46]
CAU‐7‐TATB	H3TATB, expanded tritopic	CAU‐7‐type BiO9 chains	High‐temperature activation	813 m2 g−1; 0.31 cm3 g−1 guest‐free phase discussed	[26]
CAU‐17	BTC, tritopic	BiO9 node, complex 3D	120°C, vacuum, 3 h	530 m2 g−1; PXRD retained	[96]
CAU‐33	TCPB, tetratopic	BiO5 polyhedron, 1D lozenge channel	High temperature and low pressure	450 m2 g−1; framework stable	[63]
NOTT‐220	Tetracarboxylate	BiO rods, double‐crossed	Acetone exchange; 100°C vacuum 24 h	1014 m2 g−1; 0.39 cm3 g−1; framework stable	[12]
SU‐101	Ellagate/phenolate	Bi2O rods	150°C vacuum 10 h	412/350 m2 g−1; stable over pH 2–14, PBS	[17]
Bi‐NU‐901	TBAPy, tetratopic	Bi6 node, scu topology	EtOH exchange; scCO2 drying	320 m2 g−1; PXRD/SEM retained; PBS stable	[66]
Potentially Porous Bi‐MOFs Requiring Further Validation
Bi‐TBAPy‐A	TBAPy, tetratopic	BiO8, octahedral	—	594 m2 g−1; stable	[67]
Bi‐PMOF	Porphyrin	Bi node	—	742 m2 g−1; functional stability reported	[97]
IEF‐5	S‐rich dicarboxylate	Bi3 cluster	N2 sorption (77 K)	30 m2 g−1; limited pore accessibility	[60]
IEF‐7	Phosphonate	Bi node	N2 sorption (77 K)	34 m2 g−1; high chemical stability	[55]
SU‐100	BPT, tritopic	BiO rods	Gas sorption isotherms	385 m2 g−1, breathing/flexible	[55]
BiPF‐5	MTBA, tetratopic	Bi node	—	135 m2 g−1; less robust	[62]
BiPF‐7	MTBA, tetratopic	Bi node	—	425 m2 g−1; framework adaptability limited	[62]
Non‐permanent/potential porosity
Bi‐HHTP	6‐hydroxytriphenylene	BiO5 chain, 1D zigzag	—	—	[61]
Bi‐TBAPy‐B	TBAPy	BiO8 capped triangular prism	—	‐; activation unknown	[65]
UU‐200	BTC	Bi node	—	‐; post‐activation unknown	[89]
Bi‐PDC	PDC	BiO8 node	—	—	[85]

Notes:

1. Experimentally validated permanent porosity refers to Bi‐MOFs for which framework activation procedures and low‐pressure gas sorption measurements have been explicitly reported, with structural integrity after activation confirmed whenever such information is available.

2. Potentially porous Bi‐MOFs refer to materials for which crystallographic voids, gas sorption data, or surface‐area values have been reported, but complete porosity validation (e.g., activation procedures and/or post‐activation structural characterization) remains unavailable.

3.‘‐’ = Not reported or not clearly specified in the original source.

From this perspective, the most informative Bi‐MOFs are not necessarily those with the largest apparent pores or highest reported surface areas, but those for which accessible porosity and framework stability have been experimentally demonstrated. Such distinctions are important for identifying Bi‐MOF architectures that are most promising for practical applications and for guiding future efforts toward more rigorous porosity characterization.

2.4.1. CAU‐7/CAU‐7‐TATB/CAU‐7‐TATB‐NH₂

Marked a significant milestone as the first highly crystalline Bi‐MOF with confirmed permanent porosity, Bi‐BTB (BTB = 1,3,5‐benzenetrisbenzoate), denoted as CAU‐7, was first reported by Norbert Stock and co‐workers in 2012 through conventional solvothermal synthesis or microwave‐assisted synthesis in methanol solvent [46]. On a microscopic scale, CAU‐7 exhibits a typical rod‐like morphology, making it easy to identify the structure based on the electron diffraction data collected by single‐crystal automated diffraction tomography (ADT). Each rod is a triangular prism‐like particle assembled along the [001] axis and crystallizes in the space group Pb21a. The 3D honeycomb network in CAU‐7 is composed of infinite 1D chains of face‐sharing BiO₉ polyhedra and BTB³⁻ linkers. The Bi (III) ions are coordinated with nine oxygen atoms from the BTB³⁻ linkers, forming a 1D hexagonal channel with an aperture of 10 Å (Figure 9a). CAU‐7 demonstrates remarkable thermal stability up to 380°C, as evidenced by temperature‐dependent X‐ray powder diffraction (TD‐XPRD). Compared to other BTB⁻based MOFs with honeycomb networks, such as CAU‐4 (Al(BTB)) and MIL‐103 (Tb(H₂O)(BTB), CAU‐7 exhibited the highest porosity, with a pore volume of 0.43 cm³ g⁻¹ and a BET surface area of 1150 m² g⁻¹.

FIGURE 9.

Reconstructed three‐dimensional diffraction space, crystal structure, and post‐synthetic modification of CAU‐7. (a) Reconstructed three‐dimensional diffraction space of CAU‐7 projected along the main directions (top), a rod tilted to expose the triangular base (bottom left), and a sketch of the trilling arrangement (bottom right). Reprinted with permission [46]. Copyright 2012, WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim (b) Crystal structure of CAU‐7‐TATB (view along [001]). The unit cell dimension is represented by the box. (c) Post‐synthetic modification of CAU‐7‐TATB‐NH₂with the respective anhydrides and 1,3‐propane sultone. Degree of conversion in %. Reprinted with permission [26]. Copyright 2017, Royal Society of Chemistry.

Based on this foundational work, Norbert Stock et al. synthesized CAU‐7‐TATB in 2017, using triazine‐2,4,6‐triyltribenzoic acid (H₃TATB) as a nitrogen‐containing ligand [26]. The powder X‐ray diffraction (PXRD) pattern of CAU‐7‐TATB is highly similar to that of CAU‐7. In the structure of the guest‐free CAU‐7‐TATB obtained after high‐temperature activation, the benzoate groups have less rotation relative to the central aromatic ring, resulting in a more co‐planar arrangement of the rings attributed to the minimal steric hindrance from hydrogen atoms (Figure 9b). The BET surface area of CAU‐7‐TATB was measured to be 813 m² g⁻¹, with a total micropore volume of 0.31 cm³ g⁻¹. Furthermore, the first reported functionalized Bi‐MOF, isoreticular CAU‐7‐TATB‐NH₂, using an amino‐functionalized H₃TATB‐NH₂ ligand, was synthesized under almost the same synthesis conditions as CAU‐7‐TATB. This variant opened avenues for exploring post‐synthetic modification (PSM) reactions, although only two modified products, CAU‐7‐TATB‐NH₂‐AA and CAU‐7‐TATB‐NH₂‐VA, formed by acetic anhydride and valeric anhydride, maintained a porous crystal structure after PSM reactions of CAU‐7‐TATB‐NH₂ with eight reagents (Figure 9c). These results indicate the potential for CAU‐7‐TATB‐NH₂ to host acidic groups within its pores, highlighting the diverse interaction possibilities, from weak coordination bonds to strong covalent interactions. The exploration of PSM in bismuth‐based MOFs represents a nascent yet promising field [95]. A deeper understanding of the SBUs’ characteristics and the transformations within the topological network could pave the way for developing more functional Bi‐MOFs, expanding their applicability across various fields.

2.4.2. NOTT‐220

Savage and co‐workers developed a 3D porous framework, [Bi₂(BPTC)1.5(H₂O)₂]·(DMF)3.5·(H₂O)₃ (NOTT‐220‐solv), using H₄BPTC (biphenyl‐3,3′,5,5′‐tetracarboxylic acid) as the organic ligand by a solvothermal method. Incorporating piperazine as a critical regulator in this synthesis effectively impeded the coordination interaction between organic ligands and bismuth cations. This strategic intervention regulated the nucleation rate and crystal growth, showing sophisticated control over the material's structural assembly [12]. NOTT‐220‐solv crystallizes in the space group P2/c and has a neutral, non‐interpenetrated structure anchored by irregular, distorted pseudo‐tetrahedral Bi₂O₁₄ dimer inorganic nodes bridged by BPTC4⁻ ligands. The unique aspect of NOTT‐220‐solv lies in its ability to manifest in two distinct crystal structures (α phase and β phase), differentiated by the coordination modes of water molecules. In both phases, each Bi (III) ion in the Bi₂O₁₄ dimer is coordinated with three carboxylate groups from three different BPTC4 ⁻ ligands and shares three oxygen atoms from three bridging carboxylate groups alongside two coordinated water molecules. The subtle distinction between the phases emerges in the coordination pattern of oxygen atoms from water molecules to the Bi (III) ions, marking a nuanced structural variation that underscores the framework's complexity. Each Bi (III) ion is coordinated with an oxygen atom from a water molecule in the α phase, while the same Bi (III) is connected to two water molecules in the β phase (Figure 10a,b). Remarkably, upon the removal of coordinated water molecules and free solvent from the pores, these two phases converge structurally, indicating their intrinsic similarity. The extended structure NOTT‐220‐solv, characterized by the Bi₂O₁₄ dimer (as a singular 6‐c vertex) linked with six BPTC4 ⁻ ligands (as two 3‐c vertices), unfolds into a new 3,6‐connected framework topology. This intricate network presents three unique interlinked pore channels, offering a diverse size range with inner diameters of 3.2, 5.3, and 8.3 Å. The BET‐specific surface area of NOTT‐220a is 1014 m² g⁻¹, with a total pore volume of 0.39 cm³ g⁻¹.

FIGURE 10.

Coordination geometry of NOTT‐220. (a) α‐NOTT‐220‐solv showing a water molecule bound to each Bi^III centre; (b) β‐NOTT‐220‐solv showing two water molecules coordinated to the same Bi^III centre. Reprinted with permission [12]. Copyright 2014, WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim.

2.4.3. CAU‐17

A. Ken Inge and Norbert Stock's group made a significant contribution to the field of Bi‐MOFs by reporting a porous [Bi(BTC)(H₂O)]·2H₂O·MeOH, denoted as CAU‐17, using the H₃BTC ligand [38]. This MOF stands out due to its unprecedented topological complexity, meticulously constructed from simple building units. CAU‐17 is characterized by nine kinds of BiO₉ nodes, each displaying slight variations in bond lengths and angles yet maintaining a uniform coordination mode. The structure is distinguished by each Bi (III) ion being coordinated by nine oxygen atoms, with eight originating from the carboxylate groups of five different BTC³⁻ ligands and one from a terminal water molecule. These Bi (III) ions share four oxygen atoms with two adjacent Bi (III), forming helical rods of twisted face‐sharing octahedra, which are composed of three symmetry‐independent Bi (III) ions. The interlinking of these helical rods through bridging BTC³⁻ ligands builds a complex htb network. Remarkably, CAU‐17 exhibits five crystallographically independent channels, including two hexagonal channels (double‐linked), two rectangular channels, and one triangular channel (single‐linked), based on different rod connection modes (Figure 11a‐e). The 3‐periodic net of CAU‐17, with its extensive array of carboxylate nodes, shows the framework's topological intricacy, emphasizing the potential of simple units to create complex networks. However, designing such controllably complex topological networks remains a formidable challenge, especially in Bi‐based framework materials.

FIGURE 11.

Structure of CAU‐17. (a) Crystal structure of CAU‐17. Bismuth‐oxygen polyhedra are colored purple. (b) The rods fall on the 2‐periodic htb net, which has one type of node (purple), one type of triangle (green), one type of square (light blue), and one type of hexagon (yellow). (c) The htb net is overlaid with symmetry elements found in CAU‐17 (space group P3̅) and the unit cell edges. Unlike the standard htb net, CAU‐17 has three symmetry‐independent rods (orange, green, and blue nodes), two hexagonal channels (orange and yellow hexagons), two independent rectangular channels (light and dark blue rectangles), and one type of triangular channel (green triangle). (d) The inner wall of one of the hexagonal channels of CAU‐17 viewed perpendicular to the c‐axis. (e) The connectivity within the inner walls of the hexagonal channels resembles achiral carbon nanotubes of the armchair type. Reprinted with permission [38]. Copyright 2016, American Chemical Society.

Further exploration by Ying's group into the synthesis of CAU‐17 revealed a rapid crystal structural evolution from the initially formed dense phase structure [Bi(BTC)(H₂O)]·H₂O to CAU‐17, facilitated by the cationic surfactant cetyltrimethylammonium bromide (CTAB) in an aqueous solution [98]. This discovery highlights the significant impact of micelles and solvent molecules on the local environment surrounding the Bi (III) coordination center, influencing processes such as the removal of non‐coordinated water molecules (dehydration) and the formation of ‐O‐ bridges. While surfactants are commonly used as capping agents to regulate crystal growth and morphology, their role in affecting structure and coordination is less explored. The insights from this in‐depth study open new avenues for synthesizing and designing Bi‐MOFs with controllable structures, suggesting that the strategic use of surfactants could be key to unlocking further complexities in MOF architecture.

2.4.4. Bi‐NU‐901/Bi‐TBAPy

Exploring the isoreticular analogs of MOFs presents a strategic and effective approach to developing materials with predictable structures and functionalities. In 2019, Farha's group ventured into this realm by employing Bi(III) and the tetratopic 1,3,5,8‐(p‐benzoate)pyrene (TBAPy) to construct Bi₆(µ₃‐OH)₈(HCO₂)₂(TBAPy)₂ (Bi‐NU‐901) with the same scu topology network as its Zr counterpart, Zr‐NU‐901 (Figure 12a) [66]. This process illustrates the adaptability of the isoreticular principle, effectively transitioning from one metal center to another within the SBUs. Bi‐NU‐901 is constructed from 8‐connected hexanuclear Bi₆ nodes and tetratopic TBAPy ligands, forming an scu framework with diamond‐shaped microporous one‐dimensional channels of approximately 11 Å and a BET surface area of 320 m² g⁻¹. The TBAPy ligand's moderate flexibility and the perpendicular orientation of its benzoic carboxylates relative to the pyrene core significantly influence the assembly and conformation of metal nodes, facilitating the construction of complex high‐dimensional Bi‐MOFs. Subsequently, two additional H₄TBAPy‐based Bi‐MOFs were reported, featuring a linker coordination mode distinct from that of Bi‐NU‐901. Xiao et al. successfully prepared a highly stable porous framework, denoted as Bi‐TBAPy‐A [99], where each Bi cation is octacoordinated by eight oxygen atoms from four linkers, and each ligand is chelated to four Bi (III) ions, forming two kinds of channels with different shapes (Figure 12b). The as‐constructed rhomboid and quadrilateral channels have the same size of 7.37 × 11.65 Ų. Bi‐TBAPy‐A was determined to have permanent porosity with a BET surface area of 594 m² g⁻¹. Guan and co‐workers introduced another 3D microporous fluorescent Bi‐MOF, denoted as Bi‐TBAPy‐B [65], featuring a BiO₈ double‐cap triangular prism coordination core. This structure produces a single network with square‐like pores of 16.3 × 11.6 Ų, assembled through BiO₈ cores and ligands into a layered structure with the sra topology (Figure 12c). Although the Bi‐TBAPy‐B is reported to have permanent microporosity, the lack of supporting data within the article casts uncertainty on the claim.

FIGURE 12.

Structure of Bi‐NU‐901 and Bi‐TBAPy‐A/B. (a) Bi‐NU‐901 is comprised of a₁ Bis₆ metal nodes and a₂ 1,3,6,8‐tetrakis (p‐benzoate) pyrene linkers and exhibits the a₃ scu topology. Reprinted with permission [66]. Copyright 2019, American Chemical Society. (b) Structure and analysis of Bi‐TBAPy‐A. b₁) Coordination mode of the Bi atom in Bi‐TBAPy; b₂) Coordination mode of the TBAPy moiety in Bi‐TBAPy; b₃) Crystal structure of Bi‐TBAPy viewed along the a axis. Color representations: red, O; yellow, N; gray, C; green, Bi. H atoms are removed for clarity. Reprinted with permission [99]. Copyright 2019, Elsevier. (c) Structure of the Bi‐TBAPy‐B: c₁) Linking pattern of the H₄TBAPy ligand; c₂) Rotationally symmetric BiO₈ double cap triangular prism; c₃) Binding of four different TBAPy ligands to a Bi center; c₄) 3D structure of Bi‐TBAPy; c₅) Singlet network of a complex viewed along the b axis; and c₆) Simplified 4‐connected topological net analyzed using TOPOS software. Color code: C, gray; O, red; and Bi, green. Reprinted with permission [65]. Copyright 2019, American Chemical Society.

2.4.5. SU‐101

In 2021, Inge's group developed the first bioinspired microporous Bi‐MOF with excellent biocompatibility based on the phenolic ligand ellagic acid, i.e., Bi₂O(H₂O)₂(C₁₄H₂O₈)·nH₂O, denoted as SU‐101 [17]. SU‐101 consisted of rod‐shaped IBUs that aligned in parallel layers, perpendicular to the c‐axis, yielding an interrod spacing of about 6 Å. This architecture culminates in a uninodal 7‐coordinated network (svd topology). Within its asymmetric unit, each Bi(III) ion exhibited a hexa‐coordination environment, involving three hydroxyl oxygen atoms from phenolate ligands, two µ₄ ‐oxygens, and a coordinated water molecule. SU‐101 exhibits a permanent porosity with an apparent BET surface area of 412 m² g⁻¹ and a pore size predominantly at 6.8 Å, in accordance with the crystallographically determined pore size of 6–7 Å (Figure 13a–h). In addition, SU‐101 exhibits excellent chemical stability, and the framework remains intact across a variety of organic solvents and aqueous solutions ranging from pH 2 to 14, primarily due to the robust chelation between two deprotonated hydroxyl oxygen atoms and Bi(III) ions. The exploration and successful synthesis of SU‐101 spotlights the potential for large‐scale, green synthesis and development of robust and bio‐friendly Bi‐MOF structures, including but not limited to the synthesis of crystal structures based on other previously unexplored phenolic phytochemicals.

FIGURE 13.

Structure of SU‐101. (a) The structure of SU‐101 as viewed down the c‐axis. Hydrogen atoms and water molecules in the pores are omitted for clarity. (b) The coordination environment around Bi³⁺. The bond to a coordinated water molecule is represented as a dashed line. (c, d) Chelation of ellagate toward the bismuth oxo rods. (e) The tiling of the svd net. (f) The underlying svd net. (g, h) Choice of nodes for the deconstruction of the infinite IBU. Reprinted with permission [17]. Licensed under CC‐BY. Copyright 2020, American Chemical Society.

2.4.6. IEF‐5

Alba García‐Sanchez et al. constructed a new Bi(III)‐based framework, [Bi₁₂(DTTDC)₂₄]·12[DMA]·4DMF·2H₂O, denoted as IEF‐5, using an electron‐rich linker, dithieno[3,2‐b:2′,3′‐d]thiophene‐2,6‐dicarboxylic acid (DTTDC), known for its hole transport capability [57]. IEF‐5 crystallized in the C2 space group within the monoclinic system, showing a unique structural composition where three linearly arranged Bi atoms are coordinated to 12 carboxylic groups from the DTTDC linkers to form [Bi₃(‐CO₂)₁₂]³⁻ clusters. These clusters are connected with the nearest six analogous clusters via DTTDC linkers to build a 3D pcu network structure (Figure 14a–c). Within this assembly, the framework accommodates water, and DMF solvent molecules interpenetrate between the two interpenetrated networks, along with dimethylammonium cations—byproducts of DMF decomposition—anchored within its pores. Nevertheless, the resulting BET surface area of IEF‐5 is notably low, recorded at only 30 m² g⁻¹. In the pursuit of refining the synthesis process, an additional phase, IEF‐6, was discovered, albeit inseparable from IEF‐5. IEF‐6's structural integrity is sustained by each Bi(III) ion being tetra‐chelated by eight oxygen atoms of four carboxyl groups, while each DTTDC ligand serves as a chelate to two Bi atoms, forming a stacked square layered structure.

FIGURE 14.

Structure of IEF‐5. (a) The inorganic SBU in IEF‐5 is formed by three bismuth atoms. (b) Each SBU is connected to the other six ones, forming a pcu‐type network. (c) Representation of the simulated interaction of the DMA cations with the framework atoms. Bismuth is purple, oxygen is red, carbon is black, sulfur is yellow, and nitrogen is blue. Reprinted with permission [57]. Copyright 2019, American Chemical Society.

2.4.7. Porphyrin‐Based Bi‐MOF (Bi‐PMOF)

The TCPP, also corresponding to 5,10,15,20‐tetra(p‐benzoato)porphyrin (TBP), with its tetradentate symmetry, is considered to be a captivating ligand for constructing MOFs because of its versatile biological and chemical functionalities, including light‐harvesting capabilities, additional central coordination sites for metal ions, and a robust structure. The adaptability in the connection number and symmetry of H₂TCPP opens vast topological possibilities, enabling the creation of diverse frameworks with varying pore sizes and geometries, thereby enhancing the potential applications of these MOFs in various fields.

The first Bi‐PMOF, named Bi‐TBP, was constructed by Lin's group in 2019 [97], presenting a neutral 3D framework with a formula of Bi₂(TBP)(O) formed through coordination between infinite Bi‐oxo chains and the TBP ligands (Figure 15a). In a subsequent development, Xie et al. in 2022 prepared a novel, robust 2D porphyrin‐based Bi‐MOF, named Bi‐TCPP, composed of BiO₆ cores and H₂TCPP ligands [67]. Bi‐TCPP, crystallizing in the orthorhombic space group P2₁2₁2₁ , features each Bi atom coordinated by five oxygen atoms from carboxyl groups across four porphyrin molecules and an additional oxygen atom from a water molecule (Figure 15b). The microscopic morphology of Bi‐TCPP resembles a decahedral crystal with a remarkable permanent porosity toward N₂, boasting a theoretical S_BET of 742 m² g⁻¹.

FIGURE 15.

Structure of Bi‐PMOF. (a) Coordination environments of Bi³⁺ ions in Bi‐oxo chain SBUs. Reprinted with permission [97]. Copyright 2020, Wiley‐VCH Verlag GmbH & Co. KGaA, Weinheim. (b)Crystal structure of the Bi‐PMOF. Bi, green spheres; C, gray; O, red; N, blue; H, white. Reprinted with permission [67]. Copyright 2022, Royal Society of Chemistry. (c) Chemical structure diagram of Bi‐PMOFs, C (gray), N (blue), O (red), H (white), Bi (purple). Reprinted with permission [13]. Copyright 2023, Wiley‐VCH GmbH.

Further expanding the scope of porphyrin‐based Bi‐MOFs, Cheng et al. in 2023 constructed another variant with hexagonal prism geometry by incorporating formic acid as a modifier in the reaction. This MOF crystallizes in the orthogonal Cmmm space group (Figure 15c) [13]. Notably, the central coordination site of the porphyrin ligand remains underutilized in these structures, highlighting a potential area for further exploration in the design of Bi‐PMOFs. As of recent developments, two additional Bi‐PMOFs have been introduced. Although these latest entries lack a detailed crystal structure analysis as per the referenced studies, their PXRD patterns suggest distinct structural characteristics from each other and the previously reported Bi‐PMOF [100, 101].

2.4.8. BiPF‐5/BiPF‐7

The innovative work of Eloy P. Gómez‐Oliveira and co‐workers has led to the introduction of two novel Bi‐MOFs, [Bi(MTBA)][(CH₃)₂NH₂] (BiPF‐5) and [Bi₈MTBA₆(H₂O)₉] (BiPF‐7), both using 4,4′,4″,4‴‐methanetetrayltetrabenzoic acid (H₄MTBA) as the organic ligand [62]. BiPF‐5, crystallizing in the orthorhombic space group Pnna, features a BiO₉ core where each Bi atom is coordinated by eight oxygen atoms from four different phosphonate groups and one from a water molecule, resulting in a dia topology network (Figure 16a). However, its N₂ sorption isotherm at 77K indicated limited micropore accessibility, with a BET surface area of 135 m² g⁻¹, likely attributed to the presence of cations obstructing gas adsorption within the pores, similar to the observation in IEF‐5.

FIGURE 16.

(a) Structure of InPF‐60. a₁) The inorganic SBU in InPF‐60 is composed of one metal atom coordinated to four carboxylic groups; a₂) Structural representation of InPF‐60; a₃) The inorganic SBU in BiPF‐5, composed of one metal atom coordinated to four carboxylic groups and a water ligand; a₄) Structural representation of BiPF‐5. The central C atom of the organic linker is represented as a black tetrahedron in a₂ and a₄, while blue and magenta polyhedra represent indium and bismuth atoms, respectively. In both cases, hydrogen atoms and counter‐cations are omitted for clarity. (b) Structure of BiPF‐7. b₁) The structure of BiPF‐7 is composed of vertex‐sharing cages that extend along the [001] direction. In the simplified view, two sets of catenated chains are represented in orange and blue colors. b₂) Pairs of catenated chains are connected at the positions highlighted in red, resulting in the formation of a single, self‐catenated three‐dimensional framework b₃). Reprinted with permission [62]. Copyright 2022, Wiley‐VCH GmbH.

In contrast, BiPF‐7 shows a distinctive structure with two types of SBUs: [Bi₂(‐CO₂)₆(H₂O)₂] and [Bi₂(‐CO₂)₆(H₂O)₃], both comprising two Bi atoms and six carboxyl groups. In the structure of [Bi₂(‐CO₂)₆(H₂O)₂], each Bi atom was 8‐coordinated to six oxygen atoms from three different MTBA4⁻ linkers and two oxygen atoms from two terminal water molecules, with an edge‐sharing connection between two Bi atoms. In the structure of [Bi₂(‐CO₂)₆(H₂O)₃], each Bi atom is 9‐coordinated, with three water molecules bridging two Bi atoms in a face‐sharing manner. These SBUs, acting as 6‐connected nodes, are connected by MTBA⁴⁻ ligands as 4‐connected linkers to generate a self‐catenated 4,6‐connected three‐dimensional framework. The resulting network contains characteristic mtq (Edshammer) cages as its fundamental cavity units (Figure 16b). This sophisticated structure endows BiPF‐7 with a larger BET‐specific surface area of 425 m² g⁻¹ and a type I isotherm indicative of a microporous structure.

3. Bi‐MOF Synthesis Method

Up to now, the synthesis of Bi‐MOFs has paralleled methodologies common to the broader field of MOF synthesis, with solvothermal synthesis being the most widely utilized technique. The prolonged high‐temperature and high‐pressure reaction promotes the solubility of the reactants in an organic solvent, thereby facilitating the reaction and crystallization process and yielding high‐quality crystalline MOFs. Single‐crystal products are indispensable for structural determination, particularly for the novel MOFs whose structures have not yet been elucidated [102]. The choice of synthesis conditions and methods, tailored to specific precursor systems (including metal salts and organic ligands), can significantly influence the resultant MOF's coordination structures and its physicochemical properties [103]. Alongside solvothermal and hydrothermal methods, alternative approaches, including microwave‐assisted, sonochemical, and mechanochemical syntheses, have been utilized for manufacturing Bi‐MOFs. This chapter reviews all the reported synthesis methods for Bi‐MOFs in detail, and the advantages and disadvantages of different methods are discussed.

3.1. Solvothermal/Hydrothermal Synthesis

Solvothermal synthesis is a crucial method in the preparation of Bi‐MOFs, offering a versatile approach through which the synthesis conditions can be finely tuned to affect crystal formation significantly. This method involves dissolving metal salt and organic linkers in liquid organic solvents under elevated temperatures and pressures within a closed system, such as an autoclave. The choice of solvent, its polarity, and the solubility of both metal salts and organic linkers are crucial factors that directly influence the morphology, crystallinity, and porosity of the resultant Bi‐MOFs(Figure 19a) [104].

FIGURE 19.

Schematic illustration of the current approach in the preparation of Bi‐MOFs. (a) Overview of synthesis methods, possible reaction temperatures, and final reaction products in MOF synthesis. (b) Mechanochemical methods: Neat Grinding (NG), Liquid‐Assisted Grinding (LAG), Ion‐ and Liquid‐Assisted Grinding (ILAG). Reprinted with permission [104, 110]. Copyright 2012, American Chemical Society. Copyright 2021, Elsevier Ltd.

Feyand and co‐workers presented the first highly crystalline, porous, and catalytically active Bi‐MOF, CAU‐7, employing Bi(NO₃)₃·5H₂O and H₃BTB, via solvothermal synthesis in methanol (Figure 17a) [46]. The addition of a minor amount of DMF to the reaction solvent not only improved crystal size but also prevented the agglomeration observed with methanol alone, highlighting the significant role solvent choice plays in determining the product's morphology and facilitating structure determination through electron diffraction. Furthermore, variables such as the raw materials ratio, pre‐stirring time, and reaction duration are pivotal in determining crystal growth. These parameters must be carefully optimized to achieve the desired structural and functional properties of Bi‐MOFs.

FIGURE 17.

(a) Crystal structure of CAU‐7. The connection of 1,3,5‐benzenetrisbenzoate (BTB³⁻) and Bi³⁺ ions (top left) leads to face‐sharing BiO₉ polyhedra, which form chains along the [001] direction (top right). These chains are interconnected into a honeycomb network (bottom). (b) Fragment of the crystal structure of IEF‐1 projected on the (bc) plane. Hydrogen atoms, not participating in hydrogen bonding, are omitted for the sake of clarity. The local coordination of the central atom is schematically shown at the top left. (Color codes: Bi orange, O red, and C gray balls). Reprinted with permission [43, 46]. Copyright 2012, 2018, WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim.

In comparison, hydrothermal synthesis employs water as the sole reaction solvent and typically favors the formation of low‐dimensional bismuth oxides over high‐dimensional Bi‐MOFs, attributed to the unique ground‐state electronic configuration of the bismuth metal (4f¹⁴5d¹⁰6s²6p³). The outermost 6p electrons are responsible for bonding during the coordination process, leading to the formation of sterically active lone pairs of 6s² electrons. The solvent water molecules tend to occupy one or more of the coordination sites, leading to the formation of dense, non‐porous, or layered materials easily [54, 105]. Vilela's group [43] in 2018 actualized a large‐scale preparation of 2D bismuth‐camphorate IEF‐1 (formulated as [Bi(L)(HL)]) with camphoric acid (H₂ L) as a linker by hydrothermal synthesis. The central Bi ion was coordinated with three doubly deprotonated carboxylic acid groups in a bidentate manner (L ²⁻) and one singly deprotonated carboxylic acid group in a monodentate manner (HL ⁻) to close the four‐membered and six‐membered rings, respectively (Figure 17b). In the following year, this group also prepared a 3D carboxylic acid‐based MOF that can act as a proton carrier by hydrothermal synthesis using 3,3′,5,5′‐azobenzene tetracarboxylic acid (H₄AzoBTC) as an organic ligand, formulated as [Bi₄(HAzoBTC)₂(AzoBTC)(OH)₂(H₂O)₄]·7H₂O (IEF‐2). Its asymmetric unit consists of one‐half ligand, one hydroxyl group, and five water molecules, showing a micrometer‐scale lamellar morphology [56].

Furthermore, although ionothermal synthesis, which employs ionic liquids as solvents, offers another environmentally friendly alternative for MOF synthesis, its application in Bi‐MOF synthesis remains less explored.

3.2. Microwave‐Assisted Synthesis

Microwave‐assisted synthesis represents a significant advancement in the fabrication of Bi‐MOFs, offering a rapid, efficient, and uniform heating method that contrasts sharply with the more time‐intensive solvothermal/hydrothermal techniques. This method capitalizes on the ability of microwaves to penetrate the reaction mixture, providing instantaneous and homogeneous heating throughout the solution, thus facilitating faster reaction rates and potentially yielding purer products due to the reduced likelihood of impurity formation over shorter reaction times (Figure 19a) [106]. The utilization of microwave‐assisted synthesis for Bi‐MOFs has been demonstrated by Feyand et al. in the fast synthesis of CAU‐7, where a pure‐phase product was achieved in merely 20 min—a stark contrast to the prolonged durations often required in conventional solvothermal methods [46]. While the product retained the same structural integrity as those obtained through solvothermal synthesis, the microwave‐assisted method produced distinct morphologies, namely agglomerates composed of nanocrystals, illustrating the method's influence on the physical attributes of the synthesized MOF.

More recently, Oliveira et al. synthesized a heterogeneous catalyst of BiPF‐4 with excellent chemical stability and thermal stability by microwave hydrothermal synthesis, formula [Bi₁₄(µ ₃‐O)₉(µ ₄‐O)₂(µ ₃‐OH)₅(3,5‐DSB)₅(H₂O)₃]·7H₂O. This MOF, composed of special inorganic oxy‐hydroxy‐bismuthate layers interconnected by the 3,5‐DSB ligand [48], exemplifies the potential of microwave‐assisted synthesis not only in reproducing known MOFs but also in exploring novel structures with desirable properties (Figure 18a–c). The efficiency and rapidity of microwave‐assisted synthesis make it an attractive alternative for MOF production, especially for applications requiring high‐throughput or rapid prototyping of materials.

FIGURE 18.

Synthesis of BiPF‐4. (a) Ball‐and‐stick view of the asymmetric unit of BiPF‐4, including five DBS organic linkers and two Bi7 oxo‐hydroxo clusters. Bi is green, C is gray, O is red, and S is yellow. (b)Different binding modes of the 3,5‐DSB linker were observed in the BiPF‐4 material. (c) Material BiPF‐4 is obtained via the hydrothermal reaction of bismuth nitrate with the sodium salt of the 3,5‐DSB3‐Linker under Microwave Heating. Reprinted with permission [48]. Licensed under CC‐BY 4.0. Copyright 2022, American Chemical Society.

3.3. Sonochemical Synthesis

The advent of sonochemistry in MOF synthesis presents a novel approach that leverages the intense energy and unique conditions produced by ultrasonic irradiation. Ultrasonic irradiation could generate local hotspots in the liquid medium with extremely high temperatures (ca. 5000 °C) and pressures (ca.1000 atm), opening up unprecedented opportunities for MOF fabrication [107]. The acoustic cavitation effect, central to the sonochemical reactions, creates a highly reactive microenvironment conducive to the disintegration of bulk materials into finer nanoparticles, thus promoting the effective coordination of metal ions with organic ligands. One of the remarkable benefits of sonochemistry is the homogenization of the reaction mixture through vigorous shock waves and microjets, ensuring uniformity in the reaction system and enhancing the likelihood of successful coordination reactions (Figure 19a) [108].

In 2021, Ying et al. showcased the potential of this technique by effectively producing the hexagonal prismatic pure‐phase CAU‐17 at room temperature within just 1 h, utilizing surfactant‐assisted sonochemical synthesis [109].

Furthermore, the environmental friendliness of surfactant‐mediated sonochemical synthesis was highlighted in a subsequent study by Ying's group, which achieved the synthesis of CAU‐17 nanocrystals in H₂O for the first time [98]. The employment of surfactants, particularly cetyltrimethylammonium bromide (CTAB), circumvented the common issue of incomplete coordination between ligands and metal centers due to the poor solubility of organic ligands in aqueous solution. This innovation underscores the importance of choosing suitable surfactants to simplify the synthesis conditions and attain MOFs with precisely controlled structures.

3.4. Other Synthesis Methods

The landscape of Bi‐MOF synthesis is continually evolving, with a distinct pivot toward greener, more sustainable methods that not only streamline the synthesis process but also minimize environmental impact. The exploration of water‐based room‐temperature synthesis, mechanochemical synthesis, and heating reflux synthesis adds to the repertoire of techniques employed in fabricating Bi‐MOFs, reflecting the field's ongoing commitment to eco‐friendly practices.

In addition to the aforementioned surfactant‐assisted sonochemical synthesis method, Ying's group also developed a secondary growth method, which was mediated by solvent (volume ratio of ethanol to H₂O is 1:1) and ligand (H₃BTC), to achieve the phase transformation from a mixed phase to a pure phase, resulting in CAU‐17 nanofibers with a higher specific surface area [109]. The synthesis of the biocompatible bismuth ellagic acid framework SU‐101 by the Inge group exemplifies the potential of water‐based synthesis at room temperature in 2021 [17]. Using H₂O as the only solvent and ordinary stirring under ambient conditions yields rod‐like crystals with remarkable crystallinity and uniform dispersion. This method is notable not only for its simplicity and efficiency but also for its scalability (space‐time yield is about 5 kg m⁻³ day⁻¹), offering a compelling strategy for green synthesis that could be applied broadly across MOF research.

Mechanochemical synthesis, including extrusion, ball milling, and grinding, represents another avenue toward green and efficient synthesis. By enabling MOF fabrication under minimal or no solvent conditions, this method addresses both the environmental concerns associated with solvent use and the challenges posed by the solubility limitations of certain reactants (Figure 19b) [110]. In 2014, Tröbs et al. [82]. Mechanochemically synthesized two metal‐organic structures (denoted (H₂Im)[Bi(1,4‐BDC)₂] and [Bi(2,5‐PDC)(NO₃)₂(H₂O)₂]·H₂O) composed of ligand 1,4‐benzene‐dicarboxylate (1,4‐H₂BDC), 2,5‐pyridine‐dicarboxylate (2,5‐H₂PDC), and bismuth cations, respectively, highlighting the versatility and efficiency of this approach in producing MOF materials.

In addition to these methods, the electrothermal plate heating approach recently reported by Gómez‐Oliveira et al. has proven effective for obtaining single crystals for PXRD analysis, further expanding the toolkit available for MOF synthesis [62].

The diversity of synthesis methods reflects a broader trend toward sustainable MOF production. Each approach has its own advantages and disadvantages, and they might play out differently in different situations. Microwave‐assisted and sonochemical syntheses offer an advantage for small‐scale exploratory experiments, facilitating high‐throughput screening. In particular, water‐based room‐temperature and mechanochemical syntheses stand out for their potential to achieve large‐scale, industrially viable MOF synthesis. As the field progresses, the emphasis on green synthesis underscores the research community's commitment to environmental stewardship and innovation in material science. This shift toward greener synthesis methods is expected to continue shaping the development of Bi‐MOFs, with the dual goals of advancing scientific understanding and promoting sustainability at the forefront of researchers' efforts.

3.5. Sustainability Considerations in Bi‐MOF Synthesis

Although terms such as “green”, “eco‐friendly”, and “sustainable” are frequently used throughout the Bi‐MOF literature, the criteria supporting these descriptions are often not explicitly defined. To improve consistency and provide a more objective basis for comparison, this review adopts several criteria when discussing the sustainability of Bi‐MOF synthetic strategies. A synthesis route is considered relatively more sustainable when it exhibits one or more of the following characteristics: (i) reduced use of toxic organic solvents, (ii) lower reaction temperatures and shorter reaction times, (iii) reduced energy consumption, (iv) utilization of benign or renewable precursors, (v) minimized waste generation, and (vi) potential scalability for practical production.

Based on these criteria, sustainability‐related descriptions in this review are linked to specific synthetic features rather than being used as general qualitative labels. For example, aqueous synthesis, room‐temperature synthesis, mechanochemical approaches, and the use of bio‐derived ligands may offer sustainability advantages owing to their reduced solvent consumption, lower energy demand, or improved environmental compatibility compared with conventional solvothermal methods that rely heavily on high‐boiling organic solvents. Nevertheless, the sustainability performance of different synthetic routes should be evaluated within the context of these specific criteria rather than solely through descriptive terminology.

Future studies should further establish standardized and quantitative sustainability metrics for Bi‐MOF synthesis, which would facilitate more objective comparisons among different preparation routes and accelerate the development of genuinely sustainable Bi‐MOF materials.

4. Properties and Applications

Extensive efforts have been devoted to exploring the application potential of Bi‐MOFs and their derivatives, driven not only by the structural tunability inherent to MOF chemistry but also by the distinctive physicochemical characteristics imparted by bismuth centers. In contrast to conventional MOFs, the functional advantages of Bi‐MOFs do not simply stem from ultrahigh surface areas or rigid permanent porosity. Rather, they are more closely associated with Bi‐specific features, including flexible coordination environments, accessible Lewis acidic sites, heavy‐atom‐induced photophysical effects, and favorable biocompatibility. Accordingly, this section critically discusses recent advances in the applications of Bi‐MOFs in catalysis, adsorption and separation, energy conversion, and bioactive platforms, with particular emphasis on the underlying structure–property–application relationships.

4.1. A Structure, Property, Application Perspective

Before discussing specific application domains, it is important to establish a unifying perspective that links the intrinsic structural features of Bi‐MOFs to their physicochemical properties and functional behavior. Although the development of Bi‐MOFs has often proceeded in a phenomenological manner, a more rational understanding of this field requires a structure–property–application framework.

As discussed in Section 2, the coordination chemistry of Bi³⁺ is dominated by its stereochemically active 6s² lone pair, hydrolysis‐sensitive speciation, and flexible coordination environment. These characteristics often give rise to distorted coordination polyhedra, anisotropic local charge distribution, variable cluster nuclearity, and, in some cases, coordinatively unsaturated Bi sites. Such structural features can translate into several functionally relevant properties, including strong host–guest interactions, accessible Lewis acidic sites, tunable photophysical behavior, and framework‐dependent chemical reactivity. At the same time, the relatively limited structural predictability and often moderate stability of Bi‐MOFs also impose important constraints on their practical use, particularly in applications that require highly robust permanent porosity.

These structure‐derived properties help define the application landscape of Bi‐MOFs. In photocatalysis and redox‐related processes, the heavy‐atom character of bismuth and the associated electronic structure can favor charge separation and interfacial reaction pathways. In electrochemical systems, Bi‐MOFs are often more effective as precursors or hybrid components than as pristine framework materials, because compositing or derivatization can improve conductivity, structural robustness, and active‐site accessibility. In biomedical and antibacterial contexts, the low toxicity and favorable biocompatibility of bismuth‐containing species provide an additional advantage that is less common among transition‐metal‐based MOFs.

Therefore, the functional value of Bi‐MOFs should not be assessed solely by conventional MOF metrics such as surface area or crystallographic porosity. Instead, their performance is more appropriately understood as the result of a balance between Bi‐specific chemical functionality, framework accessibility, and structural stability. From this perspective, the following sections evaluate the applications of Bi‐MOFs not as isolated examples, but as manifestations of their underlying coordination chemistry and structure‐derived properties.

4.2. Catalysis

Catalysis is a prominent application field for MOFs [111, 112, 113]. Regarding this, a number of thorough reviews that emphasized and examined the state‐of‐the‐art in MOF catalysis have been published [114, 115, 116]. The catalytic capabilities of Bi‐MOFs and their functionalized derivatives are explored in the section that follows. These characteristics are based on several catalytic reaction types, such as Lewis acid catalysis [96], photocatalysis [117, 118], and biomimetic catalysis. Electrocatalysis will be detailed in the electrochemical part of the next chapter.

4.2.1. Lewis Acid Catalysis

Based on the inherent properties of bismuth cations, Bi‐MOFs exhibit Lewis acidity and can provide active sites for catalyzing specific processes. Compared with the transition‐metal or rare‐earth‐metal MOFs‐based catalysts, Bi‐MOFs as catalysts are relatively less studied. CAU‐17, a representative Bi‐MOF with Lewis acid sites, was first reported by Köppen et al. as a heterogeneous catalyst in the ring‐opening of styrene oxide to 2‐methoxy‐2‐phenylethanol [96]. The structural characteristics of CAU‐17 are as follows: the flexible removal of coordinated water molecules in BiO₉ nodes can produce free coordination unsaturated sites. These open metal sites act as electron‐pair acceptors that can accelerate the reaction process, which in turn act as catalytic Lewis acid sites. The experimental results confirmed the stability of CAU‐17 as a heterogeneous catalyst. Although its catalytic activity is inferior to that reported for other MOFs, this study confirmed the presence of unsaturated sites in the crystal structure of CAU‐17 for the first time, which is of great significance.

4.2.2. Photocatalysis

Due to their generally poor light absorption ability and usually unstable structure, Bi‐MOFs suffer from problems such as insufficient exposure of active sites, limited generation of photogenerated carriers, and a high recombination rate of photogenerated electron‐hole pairs in photocatalytic reactions. Therefore, most Bi‐MOFs are not used directly as photocatalysts. Instead, they are typically employed as composite catalysts, utilizing bismuth‐based compounds or stable single atoms derived from Bi‐MOFs to achieve photocatalytic degradation of organic pollutants [119] and photocatalytic reduction [120, 121] reactions. In addition, Bi‐MOFs constructed based on certain ligands with unique properties, such as porphyrin [120] and MMTAA [28], also have photocatalytic properties.

From a structure, property, and application perspective, the photocatalytic performance of Bi‐MOF‐based systems generally arises from several synergistic factors, including the distinctive electronic structure of bismuth, the ability of the MOF scaffold to disperse active sites and create porous or hierarchical interfaces, and the facilitation of charge separation through ligand‐to‐metal charge transfer (LMCT), built‐in electric fields, or rational heterojunction design. Accordingly, recent studies can be broadly classified into pollutant degradation, energy‐related small‐molecule conversion, and photoelectrocatalytic systems with tailored charge‐transport pathways.

In the realm of pollutant degradation, Bi‐MMTAA as a photocatalyst was applied for Rhodamine B (RhB) decomposition under visible light (λ > 420 nm), and exhibited high photocatalytic activity with 96% degradation efficiency in 1 h. The experimental results suggest that O²⁻ is the major reactive oxygen species (ROS) during the photocatalytic process over Bi‐MMTAA. Furthermore, certain bimetallic MOF derivatives have been employed as photocatalysts for the degradation of organic pollutants under visible light irradiation. Specifically, Dai and colleagues fabricated Ag/AgCl/BiOCl composites by utilizing Ag/Bi‐CAU‐17 as a self‐sacrificial template [122]. This composite exhibited enhanced charge transfer efficiency, intensified light absorption, and superior photocatalytic activity toward tetracycline hydrochloride (TC) in comparison to the pristine CAU‐17 material (Figure 20a). Recently, Zou et al. fabricated a micro‐nano hierarchical flower‐like spherical Bi‐MOF microsphere/Bi₂MoO₆ (denoted as Bi‐MOF‐M/BMO) heterojunction structure in 2022 through self‐assembly and in situ growth techniques [123]. They employed TC as a model pollutant to assess the photocatalytic efficacy of the synthesized material. The Bi‐MOF‐M/BMO heterojunction demonstrated superior photocatalytic activity, achieving a degradation efficiency of 93.6% within a 60‐min illumination period (Figure 20b). It also exhibited excellent mineralization capabilities and structural stability. The homogeneous integration of the Bi‐MOF‐M/BMO chemically bonded heterojunction not only extends the spectral absorption range and intensifies light absorption but also facilitates the separation of photogenerated charge carriers. This is attributed to the formation of a favorable interface electric field and the alignment of energy bands that are well‐matched, thereby significantly enhancing the photocatalytic performance (Figure 20c). Subsequently, Ding et al. achieved the in situ growth of lead‐free halide perovskite Cs₃Bi₂Br₉ quantum dots (QDs) on Bi‐MOF [124], realizing atomic‐level interfacial contact between Cs₃Bi₂Br₉ QDs and Bi‐MOF through shared bismuth atoms. This approach overcomes the charge transfer barriers typically found at conventional heterojunction interfaces (Figure 20d). The Bi atoms act as electron transport channels, accelerating interfacial charge transfer while also inducing oxygen vacancies that enhance CO₂ adsorption, thereby improving photocatalytic efficiency. Similarly, Shi et al. constructed an S‐scheme heterojunction of BiOCl/Bi‐MOF via an in situ growth strategy [125], which combined the advantages of both components to enhance photocatalytic performance (Figure 20e). Recently, Lee et al. also employed an in situ growth method to design an S‐scheme heterojunction composed of the lead‐free halide perovskite CSBX (Cs₃Sb_0.5Bi_1.5Cl₄Br₅) and Bi‐MOF [126]. This strategy enabled intimate contact between CSBX and Bi‐MOF, thus improving their electronic interaction and charge transfer efficiency. In the CSBX/Bi‐MOF heterojunction, the conduction band (CB) and valence band (VB) levels differ between CSBX and Bi‐MOF. This energy band alignment allows photogenerated electrons to transfer from CSBX to Bi‐MOF, while photogenerated holes migrate in the opposite direction—from Bi‐MOF to CSBX—realizing effective charge separation and transfer. The higher CB level of CSBX facilitates electron acceptance from Bi‐MOF, and the lower VB level of Bi‐MOF favors hole accumulation. The built‐in electric field between the two components promotes the migration of photogenerated charge carriers (electrons and holes), significantly enhancing the efficiency of the photocatalytic reaction (Figure 20f). Moreover, Zhang et al. developed a CdS/Bi₂O₂CO₃ (BOC) heterojunction photocatalyst via a Bi‐MOF‐derived strategy for the synergistic degradation of antibiotics and simultaneous H₂O₂ generation under visible light [127]. Compared to Bi₂O₂CO₃ synthesized via conventional chemical methods (C‐BOC), the Bi‐MOF‐derived BOC exhibited a higher specific surface area (24.05 m²/g) and larger pore volume, which enhanced adsorption capacity toward reactants such as O₂ and antibiotic molecules. The MOF‐derived material retained a porous framework and structural integrity, allowing uniform growth of CdS nanoparticles on its surface and the formation of a tightly contacted CdS/BOC heterojunction. This type‐II heterostructure facilitated directional electron transfer from BOC to CdS, significantly improving charge separation efficiency and thus enhancing photocatalytic activity.

FIGURE 20.

Current photocatalysis applications of Bi‐MOFs. (a) EIS of BiOCl and Ag/AgCl/BiOCl. Reprinted with permission [122]. Copyright 2021, Elsevier B.V. (b) Photocatalytic degradation of TC with the as‐prepared catalysts. (c) Left c₁: The energy band structures of Bi‐MOF‐M and BMO, and right c₂: the transfer mechanism of photogenerated carriers in Bi‐MOF‐M/BMO under visible light irradiation. Reprinted with permission [123]. Copyright 2021, Elsevier. Inc. (d) Photocatalytic mechanism for CO₂ conversion with the Cs₃Bi₂Br₉/Bi‐MOF composite. Reprinted with permission [124]. Copyright 2023, American Chemical Society. (e) The efficiency of RhB degradation of different samples (triple experiment). Reprinted with permission [125]. Copyright 2024, Elsevier Inc. (f) Schematic illustration of the band structures of CSBX and BOF (left), interfacial charge transfer and the formation of an IEF upon hybridisation (middle), and CSBX/BOF S‐scheme heterojunction under light irradiation (right). Reprinted with permission [126]. Copyright 2024, Elsevier. B.V. (g) NH₃ production rate of BMO, BMO/Bi‐MOF, BMO/Bi‐MOF‐R1, BMO/Bi‐MOF‐R2, and BMO/Bi‐MOF‐R3. Reprinted with permission [128]. Copyright 2024, Elsevier. Inc. (h) Photocatalytic activity of all samples for the removal of NO. Reprinted with permission [129]. Copyright 2025, Elsevier. B.V. (i) NH₃ evolution rates over prepared catalysts. Reprinted with permission [130]. Copyright 2023, American Chemical Society. (j) UV–vis absorption spectra of RhB in the presence of Bi‐MMTAA under light irradiation. Reprinted with permission [28]. Copyright 2022, American Chemical Society. (k) Bi 4f XPS spectra of Bi‐TBAPy and Bi‐TBAPy after H₂ evolution. Reprinted with permission [99]. Copyright 2019, Elsevier. (l) Scheme of photo(electro)catalytic H₂ production mechanism, where S₀ is the fundamental state, S₁ the singlet excited state, T₁ the triplet excited state, ^•D⁻ the radical anion, and ISC the intersystem crossing. Reprinted with permission [57]. Copyright 2019, American Chemical Society.

Bi‐MOF as a photocatalyst or photoelectrocatalyst for reactions such as N₂ fixation, water splitting (hydrogen/oxygen release), and CO₂ reduction is of vital importance in promoting the sustainable development of green energy. Wang et al. in 2024 anchored Bi‐MOF onto the surface of Bi₂MoO₆(BMO) to serve as a functional platform and precisely regulated the concentration of oxygen vacancies (OVs) within the Bi‐MOF component of the BMO/Bi‐MOF photocatalyst by modulating the reducing agent dosage for light‐induced applications [128]. This system was utilized for the photocatalytic synthesis of ammonia. The BMO/Bi‐MOF photocatalyst, modified with surface OVs, exhibited a markedly enhanced photocatalytic NH₃ production rate of 125.78 µmol g⁻¹ h⁻¹, which was 21.4 and 3.7‐fold higher than that of the pristine BMO and the unmodified BMO/Bi‐MOF, respectively (Figure 20g). The porous structure and high surface area characteristics of Bi‐MOF components enhance the adsorption of N₂ at the OVs sites within the Bi‐MOF framework. Furthermore, the introduction of surface OVs facilitates the migration of photoexcited electrons from the BMO to the Bi‐MOF, thereby promoting the activation of N₂ molecules at the OVs sites.

Wang et al. constructed a TiO₂/Bi‐MOF Z‐scheme heterojunction, effectively addressing the issues of charge carrier recombination and limited light absorption (Figure 20h). This work provides a new approach for developing efficient and stable NO_x removal technologies [129]. Bakhtian et al. proposed the construction of a ternary Z‐scheme heterojunction structure, MOF‐808/Bi‐BDC/BiOBr, forming an MOF‐on‐MOF heterostructure [130]. Through interfacial engineering, they achieved effective electron‐hole separation and, by utilizing a Z‐scheme band alignment mechanism, enabled efficient transfer of photogenerated charge carriers. The MOF‐808/Bi‐BDC BiOBr system achieved an ammonia production rate of 189.5 µmol·g⁻¹·h⁻¹ under sacrificial‐agent‐free conditions (Figure 20i). Xie et al. reported a two‐dimensional bismuth‐based porphyrin metal‐organic skeleton photocatalyst for white‐driven selective oxidation of sulfides in 2022 [67]. Porphyrin ligands were excited to produce a triplet excited state (³TCPP^*) under visible light, and ¹O₂ was produced by a triplet energy transfer from ³TCPP^* to O₂. The O₂ molecule could also accept an electron from the Bi³⁺ center to generate ˙O₂ ⁻. Finally, ³TCPP^* could transfer one electron to bismuth metal to reform Bi³⁺ based on the ligand‐to‐metal charge transfer (LMCT) mechanism, and the generated ˙O₂ ⁻ and ¹O₂ oxidized the thioether to sulfoxide. Lei et al. reported a Bi‐MMTAA for visible‐light‐driven photocatalytic oxidation of phenylboronic acids(Figure 20j) [28]. The major ROS produced over Bi‐MMTAA under photocatalytic conditions was O²⁻•. Under light irradiation, the yield of phenol reaches 90.61% within 10 h under an O₂ atmosphere. When air is used instead of O₂, the reaction slows, and the yield drops to 72.60%. Xiao et al. synthesized a novel 3D Bi‐MOF (Bi‐TBAPy) with good stability and suitable band edge positions for promising photocatalytic H₂ evolution [99]. Based on our optimization, Bi‐TBAPy exhibits the optimal H₂ evolution rate of 140 µmol h^‒1 g^‒1 when triethanolamine (TEOA) is used as the sacrificial reagent. The LMCT mechanism that occurred in Bi‐TBAPy during the photocatalytic process was confirmed by XPS spectra, and this intrinsic charge separation process was favorable to the photocatalytic reactions because charge recombination was suppressed (Figure 20k). García‐Sanchez reported the synthesis and characterization of a new MOF prepared with the use of bismuth and DTTDC [57], an electron‐rich linker with hole transport ability. By combining theoretical studies and time‐resolved spectroscopies, it has been possible to determine the intermediate states and to understand the charge dynamics behavior that governs the kinetically controlled PEC reaction mechanism. First, charges are photogenerated, separated (fs), and transferred from the excited state to the triplet state (ns‐ms), where the holes react with electron donors (SO₃ ²⁻) through a radical anion intermediate. The electrons are transferred to the Pt cathode, where H₂ is produced (Figure 20l). These results give new insights into the photo(electro)catalytic behavior of MOFs in light‐induced reactions for environmental remediation.

4.3. Electrochemistry

MOFs, as a class of porous materials with exceptional properties, have garnered significant attention in the fields of electrochemical carbon dioxide reduction reaction (ECO₂RR) and electrochemical energy storage and conversion devices (EESCDs). The d⁰ electronic configuration of Bi³⁺ ions provides a suitable environment for the adsorption and activation of CO, making Bi‐MOFs particularly suitable for the targeted synthesis of formate without the significant production of by‐products like carbon monoxide (CO) or hydrogen (H₂) gas, which act as electron donors or acceptors during the CO₂RR process. In the context of EESCDs, the inherent properties of Bi‐MOFs contribute to improved charge storage and transfer capabilities, potentially leading to better performance in terms of energy density and power density. The research on Bi‐MOFs in these areas is rapidly evolving, with numerous studies exploring their synthesis methods, structural modifications, and performance optimizations.

4.3.1. Electrocatalysis

Bismuth is an exceptional electrocatalyst for the CO₂ reduction reaction (CO₂RR), featuring a unique electronic structure and surface chemistry, endowing it with highly selective adsorption capabilities for key reaction intermediates. Unlike surfaces that actively catalyze the hydrogen evolution reaction (HER), Bi displays a weak adsorption energy for the ^*H intermediate, effectively suppressing unwanted H₂ generation. Simultaneously, its moderate yet stable binding energy for the ^*OCHO intermediate favors a preferential pathway toward formic acid (HCOOH) formation. However, this binding energy remains insufficient to facilitate further reduction of the ^*CO intermediate, significantly reducing the production rates of CO and multi‐carbon (C₂ ⁺) products. Under carefully optimized potentials, Bi‐based materials frequently demonstrate Faradaic efficiencies for formate exceeding 90%, coupled with low overpotentials and remarkable operational stability. These findings substantiate the mechanism wherein Bi, through precisely modulating surface electronic density and orbital energy levels, selectively promotes the efficient conversion of CO₂ to HCOO⁻/HCOOH [131].

Recent studies have focused on tuning the intrinsic catalytic activity and selectivity of a catalyst by changing composition, morphology, and atomic arrangement. Ying et al. developed a CTAB‐mediated morphology engineering method to reconstruct a CAU‐17‐fiber with a larger accessible surface area [109]. The morphology transformation processes were systematically studied and further employed to build a CAU‐17‐fiber hierarchical structure by calcination for formate production through the ECO₂RR process. The CAU‐17‐fiber_400 electrode exhibited superior electrocatalytic performance in a broad range of potentials and demonstrated a formic acid Faradaic efficiency (FE_HCOO‐) of 96.4% with a current density of about 20.4 mA cm⁻² at a potential of ‐0.9 V_RHE in the H‐type cell, as well as outstanding durability (Figure 21a). Li et al. describe three Bi‐MOFs, MFM‐220, MFM‐221, and MFM‐222 [93]. Three Bi‐MOFs with the same ligand but distinct porosity (ranging from 49.6% to 0%) have been synthesized by tuning the reaction conditions (Figure 21b). The faradaic efficiency for formate (FE formate) can reach 90.4% at 1.1 V vs. RHE using evolved MFM‐220‐p/carbon paper (CP). The high catalytic ability of MFM‐220‐p/CP is due to the structural evolution promoted by the highly porous structure of MFM‐220 (void of 49.6%) compared to the cation‐blocked MFM‐221 and non‐porous MFM‐222 (voids of 33.6% and 0%, respectively). Li et al. synthesized a highly stable SU‐101 catalyst using a wet chemical method, which showed high activity and selectivity for HCOO⁻ for CO₂ electrochemical reduction under ambient conditions (Figure 20c) [132]. The FE generated by HCOO was as high as 93.66% at a potential of 1.06 V vs. RHE, and it had excellent stability for 10 h.

FIGURE 21.

Current electrocatalysis application of Bi‐MOFs. (a) Left: Schematic representation of the preparation of CAU‐17‐hp by an ultrasound‐assisted wet chemical reaction and the construction of CAU‐17‐fiber. Right: Schematic representation of the experimental setup (H‐cell configuration) for ERCO₂ employing CAU‐17‐derived electrocatalysts. Reprinted with permission [109]. Copyright 2021, Wiley‐VCH GmbH. (b) Views of the coordination environments of Bi (III) sites and pore structures of (b₁ and b₂) MFM‐220, (b₃ and b₄) MFM‐221, and (b₅ and b₆) MFM‐222 (Bi, purple; O, red; C, gray; H, light gray; N, blue). The Me₂NH₂ ⁺ cations and coordinated DMF molecules in MFM‐221 and MFM‐222, respectively, are highlighted in ball‐and‐stick mode. Reprinted with permission [93]. Licensed under CC BY 3.0 Unported. Copyright 2022, Royal Society of Chemistry. (c) Schematic illustration of the preparation procedure of the SU‐101 ‐NRs. Reprinted with permission [132]. Copyright 2022, Royal Society of Chemistry. (d) d₁. The general process for the synthesis of Bi‐HHTP. (d₂–d₄) The views are in a different dimension. d₅. Chelation of HHTP toward Bi³⁺ ions. d₆. The coordination environment of Bi³⁺ ions (distorted tetragonal pyramid). Reprinted with permission [61]. Licensed under CC BY 3.0 Unported. Copyright 2023, Royal Society of Chemistry. (e) Schematic illustration of the ion transfer in the paired electrolysis with BPM. Reprinted with permission [133]. Copyright 2022, Wiley‐VCH GmbH. (f) The schematic diagram of the solar‐driven ECR||MOR electrolytic system. Reprinted with permission [134]. Copyright 2025, American Chemical Society. (g) In situ surface reconstruction of the Bi‐MOF to Bi NS for the CRR. Schematic of the two‐step reconstruction process, including 1) the electrolyte‐mediated chemical and structural transformation from Bi‐MOF to Bi₂O₂CO₃ NS, and 2) electrochemical reduction of Bi₂O₂CO₃ NS to Bi NS. Reprinted with permission [135]. Copyright 2021 Wiley‐VCH GmbH.

The unsaturated coordination Bi sites in some Bi‐MOFs with stable structures can also serve as active sites, which could promote charge transfer and lower the energy barrier. Gao et al. constructed the Bi‐HHTP of zigzagging corrugated topology with Bi‐enriched sites (Figure 20d), which exhibited an outstanding performance for selective formate production of 95% with a maximum turnover frequency of 576 h⁻¹ in a flow cell. Significantly, the structure of Bi‐HHTP could be well maintained after catalysis [61].

Yang and co‐workers developed an interesting strategy to construct paired electrolytes with enhanced internal relationships (Figure 21e) [133]. A novel Bi‐MOF, Bi‐FDCA, was constructed by converting biomass‐derived 5‐hydroxymethylfurfural (HMF) to 2,5‐furandicarboxylic acid (FDCA), which is capable of catalyzing the reduction of CO₂ to HCOOH with over 95% FE and an average current density of 19.6 mA cm⁻² at 1.2 V vs. RHE. With this inner relationship, paired electrolysis was conducted with cathodic potential fixed at 1.2 V vs. RHE, producing HCOOH with 95.6% FE and FDCA with 75% FE in one single electrolyzer.

In a recent study, Cao et al. synthesized a series of structurally distinct Bi‐MOFs (Bi‐BDC, Bi‐BTC, Bi‐PMA) by tuning the structures of the organic ligands, aiming to optimize their catalytic performance for CO₂ reduction to formate (HCOO⁻) [134]. In addition, they innovatively introduced methanol oxidation (MOR) as the anodic reaction and constructed a solar‐driven ECR||MOR (electrochemical CO₂ reduction || methanol oxidation reaction) tandem system (Figure 21f). This setup enabled the simultaneous production of formate at both the cathode and anode, significantly improving electrolysis efficiency and reducing energy consumption. Among the Bi‐MOFs, Bi‐BDC exhibited the highest conductivity, the largest specific surface area, and the optimal electron density, resulting in the best ECO₂RR performance. In the constructed Bi‐BDC||CuO/CF ECR||MOR system, replacing the conventional oxygen evolution reaction (OER) with MOR at the anode reduced the system voltage by 310 mV. The system achieved synchronized HCOO⁻ production at both electrodes and maintained stable operation over 20 h.

Though most reports announced that Bi‐MOFs show good stability during the ECO₂RR process, the catalyst materials may undergo dynamic changes, such as phase transformation, nanoclustering, degradation, and reduction, under operating conditions. This reconstruction will result in uncontrollable and detrimental deviation from the target active sites and expected catalytic performance. Yao et al. used Bi‐MOF for the electrochemical reduction of CO₂ as the research object for the first time [135]. They reported that monitoring and controlling the reconstruction of materials under working conditions is crucial for accurately identifying active sites, clarifying reaction mechanisms, and rationally designing advanced catalysts, confirming that the complex reconstruction of Bi‐MOFs can be controlled in two steps: 1) electrolyte‐mediated dissociation and conversion of Bi‐MOF to Bi₂O₂CO₃ (BOC), and 2) potential‐mediated reduction of BOC to Bi. The intentionally reconstructed Bi catalyst exhibits excellent activity, selectivity, and durability for formate production, and the unsaturated surface Bi atoms formed during reconstruction become the active sites (Figure 21g). Abundant unsaturated Bi atoms were formed adjacent to the surface vacancies during the two‐step reconstruction, contributing to more favorable adsorption of ^*OCHO intermediates and ultimately benefiting the reaction process.

Overall, current studies suggest that Bi‐MOFs can contribute to ECO₂RR through multiple roles, including as intact catalysts with accessible Bi sites, as porous and structurally tunable precursors for reconstructed active phases, and as platforms for integrated paired electrolysis systems. At the same time, the field also reveals a recurring challenge: high catalytic performance does not necessarily imply that the pristine MOF framework remains the true active phase under operating conditions. Therefore, future work should place greater emphasis on operando characterization, reconstruction tracking, and rigorous benchmarking against non‐MOF Bi‐based catalysts, so as to distinguish intrinsic framework‐mediated catalysis from activity arising from dynamically evolved Bi species.

4.3.2. Electrochemical Energy Storage and Conversion Devices

MOFs are widely studied as innovative materials for energy storage and conversion devices, such as a supercapacitor and a battery. However, owing to their insufficient conductivity and limited mechanical and chemical stability, the direct implementation of MOFs generally does not offer satisfactory performance. Therefore, more interest and investment were attached to MOF composites and MOF derivatives for practical applications. Like other MOFs, Bi‐MOFs also have the characteristics of high surface area and adjustable porosity. Particularly, bismuth's unique electronic structure and chemical properties endow Bi‐MOFs with distinct electrochemical properties. Recently, the study of Bi‐MOFs in the energy storage and conversion field is still in its infancy.

A recent breakthrough by Ma et al. in 2024 developed a new type of solid‐state supercapacitor utilizing Bi‐MOF materials combined with CP electrodes to enhance humidity harvesting and energy storage capabilities [136]. Two new hygroscopic Bi‐MOFs, [Bi₂(HABTC)(ABTC)_0.5·4H₂O] and [Bi₄(ABTC)₃(DMF)₂]·DMF, were grown in situ on carbon paper and modified with conductive polyaniline to enhance ion transport, charge transfer, and hygroscopicity. Under conditions of 90% relative humidity and 70°C, solid‐state capacitors constructed with this material achieved a high energy density of 40.40 Wh kg⁻¹ and exhibited excellent performance at a power density of 499.82 W kg⁻¹. After 1000 charge‐discharge cycles, the supercapacitor retained 92.17% of its initial capacitance and successfully powered a micro‐LED, demonstrating its practicality (Figure 22a).

FIGURE 22.

Current applications of Bi‐MOFs in energy storage and conversion. (a) Capacitance retention after 1000 cycles. Reprinted with permission [136]. Copyright 2024, Wiley‐VCH GmbH. (b) Variations of the specific capacity of Bi‐MOF‐based electrodes at different current densities. (c) Cycling performances of Bi‐MOF‐based electrodes at 2.0 A g⁻¹. (d) Specific capacitance calculated by GCD curves at different bending angles of the flexible device. Reprinted with permission [137, 138]. Copyright 2024, 2023, Elsevier Ltd. (e) Schematic illustration of the synthesis of Bi@SA Cu‐C. Reprinted with permission [140]. Copyright 2024, Wiley‐VCH GmbH. (f) Schematic representation for the synthesis of BiCo‐MOF@VS₂ nanocomposite on NF. Reprinted with permission [141]. Copyright 2024, Elsevier B.V. (g) Scheme illustration of BiMOF‐CP cathode. (h) Schematic diagrams comparing the MnO₂ deposition mechanisms on CP (top) and Bi‐PYDC (bottom), illustrating the three functions (modulation of the deposition rate, stabilization of the structure of the deposited MnO₂, and H storage ability) that Bi‐PYDC performed. Reprinted with permission [142]. Copyright 2023, Wiley‐VCH GmbH.

Wang et al. precisely controlled the synthesis parameters—particularly the Bi³⁺ to H₃BTC molar ratio—to successfully prepare Bi‐MOF materials with a three‐dimensional hierarchical structure and systematically investigated the relationship between structure and performance [137]. When the Bi³⁺/H₃BTC molar ratio was 2:1 (denoted BM‐2), the resulting material exhibited a three‐dimensional spherical hierarchical architecture composed of interconnected ultrathin nanosheets. Electrochemical evaluation of BM‐2 revealed a specific capacitance of 896.1C g⁻¹ at a current density of 0.5 A g⁻¹, significantly outperforming control samples BM‐1 (812.9C g⁻¹) and BM‐3 (459.1C g⁻¹); even at 5 A g⁻¹, BM‐2 retained 80.2% of its capacitance (Figure 22b). Cyclic voltammetry (CV) profiles exhibited characteristic dual redox peaks (Bi ↔ Bi(I) ↔ Bi(III)), indicating that Faradaic pseudocapacitance predominates the energy storage mechanism. Moreover, BM‐2 demonstrated excellent cycling stability, retaining 71.2% of its initial capacitance after 2000 charge‐discharge cycles at 2.0 A g⁻¹ (Figure 22c). When assembled into a symmetric supercapacitor (BM‐2 as the positive electrode and activated carbon as the negative electrode), the device achieved an energy density of 33.7 Wh kg⁻¹ and a power density of 1699.2 W kg⁻¹; after 1000 cycles at 5 A g⁻¹, the capacity retention remained at 79.5%. Compared with reported Bi‐based materials such as Bi₂S₃/g‐C₃N₃, Bi₂O₃@C, and BiFeO₃, BM‐2 exhibits superior specific capacitance and device performance, demonstrating strong potential for practical applications.

Beyond precursor ratio regulation, recent studies further show that morphology engineering under mild and green conditions can also effectively enhance the electrochemical performance of pristine Bi‐MOFs. Qin et al. reported an ultrafast room‐temperature synthesis of CAU‐17 in a water/methanol binary solvent system using Pluronic P123 as a soft template, affording a hierarchical sea‐urchin‐like architecture within only 5 min. Benefiting from its improved ion accessibility and facilitated charge‐transfer kinetics, the optimized CAU‐17 delivered a high specific capacity of 731 C g⁻¹ at 0.2 A g⁻¹ and retained 81.25% of its capacitance after 6100 cycles [31]. When assembled into a hybrid supercapacitor with MnO₂ as the positive electrode, the device achieved an energy density of 42.2 Wh kg⁻¹ at a power density of 400 W kg⁻¹. These results indicate that, in addition to composition and precursor control, rational morphology modulation provides an effective route for improving the energy‐storage performance of Bi‐MOFs.

In contrast to these studies based on pristine Bi‐MOF electrodes, further performance enhancement can also be achieved by converting Bi‐MOFs into conductive derivative architectures. Huang et al. prepared Bi‐MOF‐containing fibers by mixing Bi‐MOF with a PAN (polyacrylonitrile) solution and using electrospinning technology to deposit the mixture onto a carbon fiber substrate [138]. The resulting fibers were then subjected to pyrolysis to convert Bi‐MOF into Bi₂O_{3‐ x} . Further oxidation at an appropriate temperature yielded Bi₂O_{3‐ x} @carbon fiber (CF) composites. To further enhance conductivity, poly(3,4‐ethylenedioxythiophene) (PEDOT) was used as a shell material to coat the Bi₂O_{3‐ x} @CF, forming a typical core‐shell structure. The Bi₂O_3‐0.5@CF@PEDOT electrode exhibited a high specific capacitance of up to 460 F g⁻¹ at a current density of 1 A g⁻¹, and maintained a good energy density (16.4 Wh kg⁻¹) even at a high power density of 500.34 W kg⁻¹. The electrode demonstrated excellent cycling stability, retaining 99% of its capacitance after 8500 charge‐discharge cycles. A symmetric supercapacitor (SSC) device assembled using Bi₂O_3‐0.5@CF@PEDOT electrodes and a PVA‐KOH gel electrolyte delivered an energy density of 16.4 Wh kg⁻¹ and a power density of 500.34 W kg⁻¹ at 1 A g⁻¹. The device showed minimal capacitance loss after bending (Figure 22d), indicating strong durability and flexibility for practical applications. More recently, MOF‐derived Bi‐based composites have emerged as promising battery‐type electrodes for hybrid supercapacitors. Qin et al. reported a CAU‐17‐derived Bi₂O₃/N‐doped carbon composite (Bi₂O₃@NC‐700) prepared through an air‐assisted phase stabilization strategy followed by nitrogen‐doping treatment [139]. By introducing a pre‐oxidation step prior to carbonization, the reduction and volatilization of Bi species during thermal conversion were effectively suppressed, enabling the retention of electrochemically active Bi₂O₃ within a conductive N‐doped carbon framework. The resulting electrode delivered a high specific capacity of 1329C g⁻¹ at 0.5 A g⁻¹ and maintained 89.02% capacity retention after 10 000 cycles. When assembled into a hybrid supercapacitor with NiCoMn‐LDH, the device achieved an energy density of 58.13 Wh kg⁻¹, demonstrating the considerable potential of Bi‐MOF‐derived materials for advanced electrochemical energy‐storage applications.

Unlike single‐metal MOFs, bi‐metallic MOFs have stable framework structures. The coexistence and synergy of different transition metals, such as nickel, cobalt, and iron, provide higher conductivity, better redox properties, more active sites, multiple valence states, enhanced charge compensation, and improved charge storage, thereby enhancing the performance of energy storage applications. To improve the performance of Bi‐MOFs in batteries, researchers have employed various strategies to optimize their electrochemical properties. These strategies include metal doping, organic ligand regulation, and composite material design. For example, by introducing conductive polymers, carbon nanotubes, and other conductive materials to form composites with Bi‐MOFs, the conductivity of Bi‐MOFs can be significantly improved, accelerating ion diffusion and enhancing the cycling stability of the batteries. Li et al. proposed the single atom Cu regulated Bi@C with Cu‐C₃ site material (Bi@SA Cu‐C) using a simple MOF‐derived precursor method [140]. A Bi‐MOF precursor doped with copper (Cu) was first synthesized and then subjected to pyrolysis. During pyrolysis, Cu atoms were incorporated into the carbon matrix, forming Cu‐C₃ sites, which are coordination structures where a Cu atom is directly bonded to three carbon atoms (Figure 22e). These Cu‐C₃ sites regulate the decomposition of the electrolyte and promote the formation of the solid electrolyte interphase (SEI) layer. They significantly reduce the dissociation energy of the P‐F bond in the NaPF₆ electrolyte and the C‐O bond in 1,2‐dimethoxyethane (DME), facilitating SEI formation. Moreover, the local electric field generated by the Cu‐C₃ sites enhances Na⁺ adsorption and migration, lowering the diffusion energy barrier and improving sodium‐ion storage kinetics. The Bi@SA Cu‐C material exhibited excellent electrochemical performance, achieving a high reversible capacity of 404 mAh g⁻¹ at a low current density (0.2 A g⁻¹), and maintaining 315 mAh g⁻¹even at a high current density (70 A g⁻¹), indicating outstanding rate performance. Additionally, it demonstrated excellent cycling stability, retaining 95.8% of its capacity after 6,000 cycles, significantly outperforming Bi@C (without Cu single atoms), highlighting the critical role of Cu single atoms in SEI regulation and Na⁺ diffusion enhancement. Sugasri Chinnasamy et al. proposed a high‐performance electrode composed of BiCo‐MOF@VS₂ using a hydrothermal technique (Figure 22f) [141]. The BiCo‐MOF@VS₂ heterostructure outperforms BiCo‐MOF in electronic transport, ion diffusion rates, and spatial resistance. Compared with other MOF@VS₂ composites, BiCo‐MOF@VS₂ showed the largest specific capacitance and retained 85% of its capacitance after 3000 cycles at a current density of 9 A g⁻¹. Gou et al. first proposed the “MOF as an electrodeposition surface” strategy to regulate the Mn²⁺ deposition reaction rate [142], using a 3D MOF structure, Bi‐PYDC. The experiments showed that a cathode‐free ZIB with Bi‐PYDC surface modification had an excellent cycle life of 10 000 cycles at a current density of 1 mAh cm^{−2 (}Figure 22g,h). Su et al. utilized a Bi‐MOF as a precursor for direct carbonization treatment [143]. During pyrolysis at 700 °C, Bi³⁺ ions were in situ reduced to metallic Bi nanoparticles, which were uniformly dispersed in the carbon framework. The carbonized MOF not only constructed a continuous conductive network, but also formed a hierarchical architecture consisting of carbon nanorods and an ultrathin outer carbon layer. This structure effectively buffers the volume expansion of Bi nanoparticles during cycling, thus preventing material pulverization and electrode failure. The Bi‐MOF‐derived composite UCF@CNs@BiN effectively addresses the key challenges of traditional Bi‐based anodes in potassium‐ion batteries (PIBs), such as significant volume variation, unstable solid electrolyte interphase (SEI), and short cycle life. The design notably enhances specific capacity, rate capability, and cycling stability, making it a promising anode candidate for high‐performance PIBs.

In summary, current studies suggest that Bi‐MOFs can play three principal roles in electrochemical energy storage and conversion: (i) as direct electroactive frameworks, particularly in pseudocapacitive systems; (ii) as precursors or sacrificial templates for constructing conductive and structurally robust derivative electrodes; and (iii) as interfacial modifiers or heterostructure components that regulate charge transfer, ion transport, and deposition behavior. Among these, derivative engineering and composite design currently appear to be the most effective strategies for overcoming the intrinsic conductivity and stability limitations of pristine Bi‐MOFs. Future progress will depend on more rigorous benchmarking against advanced non‐MOF Bi‐based materials, clearer operando identification of the true electrochemically active phases, and deeper structure–property correlations between the parent framework and the performance of the derived functional architectures.

4.4. Molecular Adsorption, Detection, and Separation

The high porosity and tunable chemical functionalities of Bi‐MOFs make them ideal candidates for molecular adsorption and separation applications. They have been the subject of extensive research for applications such as gas storage, separation, and the removal of pollutants from water and air.

4.4.1. Gas Adsorption and Detection

Köppen et al. demonstrated that CAU‐17 [96], with its microporous structure and a BET surface area of up to 530 m²/g, exhibits excellent greenhouse gas adsorption properties, especially in the selective adsorption of SF₆/N₂ and CO₂/N₂, with selectivities exceeding 31 and 29, respectively. This makes CAU‐17 potentially valuable for greenhouse gas separation and capture (Figure 23a). Mathew Savage et al. reported a highly porous structure and high‐density Bi‐MOF, NOTT‐220‐solv [12], for high‐capacity selective adsorption of CH₄ and greenhouse gas CO₂. Using grand canonical Monte Carlo (GCMC) simulations and density functional theory (DFT) calculations, the adsorption process was modeled, showing that CH₄ and CO₂ adsorb first at the {Bi₂} node's Bi‐O bonds (Figure 23b). The stable adsorption of CH₄ is primarily driven by weak hydrogen bonds between the oxygen centers of the carboxyl group on CH₄, while CO₂ is adsorbed through dipole interactions between the positive carbon center and the oxygen center of the carboxyl group. The design of bismuth‐based nodes and multi‐level pore channels achieves a balance between high volumetric adsorption performance and structural stability. Future work could enhance selectivity and capacity through ligand functionalization or by introducing open metal sites. This research provides a new approach for the application of heavy‐metal‐based MOFs in gas storage. Meanwhile, Bi‐MOFs, as precursors, have their derivatives applied in gas removal. Li et al. reported a defect‐engineered Bi@BiOBr/C hierarchical micro‐rod photocatalyst derived from CAU‐17 [144], which was applied for efficient visible‐light‐driven NO gas removal. CAU‐17 served not only as a precursor but also provided a morphology‐controlling template favorable for constructing hierarchical structures. Through a one‐step calcination‐reduction strategy, carbon coating, Bi nanoparticle loading, and oxygen vacancy introduction were achieved simultaneously, resulting in improved charge transport pathways and the creation of abundant active sites.

FIGURE 23.

Current applications of Bi‐MOFs and their derivatives in gas adsorption and detection. (a) Adsorption isotherms of CAU‐17 with CO₂, SF_6, and N₂ recorded at 0°C. (b) View of the preferred binding mode of b₁) and b₂) CH₄; and b₃) and b₄) CO₂ on the binuclear {Bi₂} node of NOTT‐220a. b₁) and b₃) show the side‐on view while b₂) and b₄) show the top‐down view. Reprinted with permission [12, 96]. Copyright 2018, 2014, WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim. (c) Comparison of the iodine sorption capacity of other Bi‐based sorbents. Reprinted with permission [145]. Copyright 2022 Elsevier B.V. (d) Breakthrough curve of H₂S adsorption by SU‐101 at 25°C and 1 bar. The inset shows the comparative H₂S adsorption capacities for each cycle. Reprinted with permission [17]. Licensed under CC‐BY. Copyright 2020, American Chemical Society. (e) Solid‐state emission spectra of activated SU‐101 at different excitation wavelengths. (f) Solid‐state emission spectra of the ellagic acid linker (brown) and as‐synthesized SU‐101 (yellow). Reprinted with permission [146]. Copyright 2025, Royal Society of Chemistry.

Qin et al. synthesized NOTT‐220 [145], which exhibited excellent iodine adsorption performance at 75°C, with an adsorption capacity of 955 mg·g⁻¹ (Figure 23c). The experiments showed that the captured iodine existed not only in the form of iodine molecules but also converted into BiOI, BiI₃, and polyiodide anions. Grape et al. found that Bi₂O(H₂O)₂(C₁₄H₂O₈)·nH₂O (SU‐101) exhibited excellent chemical stability and was used for the adsorption of acidic gases such as SO₂ and H₂S (Figure 23d) [17]. In SO₂ adsorption experiments, SU‐101 performed excellently, reaching a high adsorption capacity of 15.95 mmol/g, making it one of the most efficient MOFs for hydrogen sulfide adsorption. In addition to its excellent H₂S adsorption properties, Valeria B. López‐Cervantes' team recently investigated its detection capabilities for H₂S [146]. SU‐101 itself generated an emission peak around 461 nm when excited at 370 nm (Figure 23e). When compared with its organic ligand‐metal coordination, it was found that after metal coordination, the emission originally dominated by the ligand was partially quenched due to ligand‐to‐metal charge transfer (LMCT). When SU‐101 was exposed to an H₂S atmosphere, polysulfide generation caused structural rigidity and blocked the LMCT process, resulting in the fluorescence “turn‐on” phenomenon, significantly enhancing the emission intensity (Figure 23f). This resulted in a marked increase in fluorescence signal, leading to high selectivity and sensitivity in H₂S detection. The detection limit was as low as 22 ppm, with almost no response to other gases such as water vapor and CO₂, demonstrating excellent selectivity and reproducibility.

4.4.2. Water Purification

Bi‐MOFs have been widely studied in AOP (advanced oxidation process) for pollutant removal due to their low toxicity, environmental friendliness, and high stability. In a recent study, Gao et al. precisely designed and synthesized Bi‐BDC‐Fc _x using a microwave‐assisted synthesis method, in which BDC was partially substituted with different ratios of ferrocene (Fc) ligands (Figure 24a) [147]. The introduction of Fc ligands effectively provided additional electrons, facilitating electron transfer upon light excitation. After photoexcitation, electrons could migrate from both H₂BDC and Fc to the Bi centers, enhancing the separation and transfer efficiency of photogenerated electrons. The prepared Bi‐BDC‐Fc _x materials were applied in the photocatalytic activation of PMS (peroxymonosulfate), PDS (peroxydisulfate), and H₂O₂ to degrade pollutants such as bisphenol A (BPA), 4,4′‐Propane‐2,2‐diylbis(2‐methylphenol) (BPC), and 4,4′‐Sulfonyldiphenol (BPS). Quenching experiments and electron spin resonance (ESR) analysis were used to elucidate the roles of reactive species in different oxidation systems. In the PMS and PDS systems, SO₄ ⁻·, ·OH, and ¹O₂ were identified as the main active species, whereas in the H₂O₂ system, ·OH and ¹O₂ played the dominant roles. Additionally, Ouyang et al. found that CAU‐17 showed high adsorption capacity for selenate ions (SeO₃ ²⁻) [148], reaching 255.3 mg/g, which is much higher than that of traditional adsorbents. Even in the presence of high concentrations of competitive anions like CO₃ ²⁻, Cl⁻, and PO₄ ³⁻, CAU‐17 maintained high selectivity and showed a strong affinity for selenate ions. The structure of CAU‐17 contains hexagonal, rectangular, and triangular channels, providing a large number of pathways for SeO₃ ²⁻ anions to access the active Bi‐O sites (Figure 24b). Experiments using Density Functional Theory (DFT) calculations, X‐ray absorption near‐edge structure (XANES) modeling, and extended X‐ray absorption fine structure (EXAFS) fitting to explore the adsorption mechanism revealed that CAU‐17 can fix one SeO₃ ²⁻ ion via two Bi‐O‐Se bonds in the triangular channels (Figure 24c). The tetravalent adsorption behavior primarily involves chemical adsorption, while the hexavalent SeO₄ ²⁻ is physically adsorbed. Lu et al. also conducted experiments on CAU‐17 for phosphate removal and compared it with a La‐doped Bi‐MOF (La‐CAU‐17) in a parallel experiment [149]. They found that La‐CAU‐17 exhibited the best phosphate adsorption capacity, reaching 216.07 mg/g, surpassing CAU‐17's 185.57 mg/g. Fourier Transform Infrared Spectroscopy (FTIR) showed that the organic ligands H₃BTC and La³⁺ can modulate the coordination of the central Bi metal during the doping process, exposing more active adsorption sites (Figure 24d). He et al. used solvothermal methods to synthesize Bi‐BDC and Bi‐NH‐BDC for the adsorption of diclofenac sodium (DCF) from water [150]. To address potential secondary pollution caused by adsorption, they combined adsorption with photocatalytic degradation. The results showed that DCF molecules were effectively degraded, and the degradation pathway was verified by intermediate products. Ultimately, organic materials were degraded into CO₂, water, and Cl⁻, among other harmless substances (Figure 24e). After five cycles, Bi‐NH‐BDC still maintained a high DCF removal rate.

FIGURE 24.

Current water purification application of Bi‐MOFs (a) The schematic illustration for the synthesis of Bi‐BDC‐Fc_x. Reprinted with permission [147]. Copyright 2024, Elsevier B.V. (b) Structure of CAU‐17 with 3 kinds of channels (H for hexagonal channels, T for triangular channels, and R for rectangular channels), and the structure of the minimum repeat unit. (c) Illustration of the formation mechanism of the multilayered structure of CAU‐17‐SeO₃ ²⁻. Reprinted with permission [148]. Copyright 2018, Wiley‐VCH Verlag GmbH & Co. KGaA, Weinheim (d) FTIR spectra of Bi‐MOFs. (e) Possible degradation pathways of DCF. Reprinted with permission [149, 150]. Copyright 2021, 2024, Elsevier B.V.

4.5. Biological and Medical Applications

Bismuth, as a heavy metal, stands out due to its recognized low toxicity, making it unique among many metal elements. Traditional bismuth‐based compounds, such as bismuth subsalicylate, have long been used in the treatment of gastrointestinal disorders, demonstrating good safety and clinical utility [151, 152, 153]. This favorable biological profile of bismuth provides a solid foundation for its further development in modern biomedical materials.

MOFs, known for their exceptionally high surface area, tunable pore structures, and versatile functionalities, have emerged as promising platforms in drug delivery and other therapeutic applications. Bi‐MOFs, which integrate bismuth ions into the MOF structure, not only retain these advantageous properties but also inherit bismuth's low toxicity. By optimizing pore size and surface chemistry, Bi‐MOFs can improve the affinity and loading capacity for specific drug molecules. Moreover, studies in both cell and animal models have demonstrated excellent biocompatibility, reinforcing their potential for safe in vivo applications.

4.5.1. Drug Delivery

The porous structure and chemical nature of Bi‐MOFs make them ideal candidates for drug delivery. Grape et al. synthesized SU‐101 from ellagic acid [17], which showed good biocompatibility and colloidal stability in vitro. SU‐101 demonstrated no significant toxicity, indicating its high safety for in vivo applications (Figure 25a–c). Additionally, the porous structure of SU‐101 allows it to effectively load drug molecules and release them as needed, suggesting its potential for drug delivery. Subsequently, Zhang et al. investigated the drug‐loading and light‐controlled release capabilities of SU‐101 [16]. In their work, SU‐101 exhibited high loading capacities for various drugs such as ciprofloxacin (CIP), norfloxacin (NOR), and amoxicillin (AMO), with a CIP loading rate reaching up to 85.8%. The primary mechanism involves the formation of hydrogen bonds between the hydroxyl groups (─OH) of drug molecules and the carbonyl groups (C═O) in SU‐101. Upon light irradiation, the electron density around the C═O groups decreases, causing the hydrogen bonds to break and triggering drug release (Figure 25d). To verify the general applicability of this mechanism, SU‐101 was also tested with other hydroxyl‐containing drugs (e.g., NOR, AMO, tetracycline (TET), and doxorubicin (DOX)), all of which showed similarly high loading capacities and light‐controlled release behaviors.

FIGURE 25.

Current drug delivery application of Bi‐MOFs. (a) Colloidal stability of SU‐101 (reagent grade) in water (red) and RPMI (green), representing a) the NP size and b) ζ‐potential evolution vs. time. (c) Cell viability of the HL‐60 cell line after 24 h incubation with SU‐101 (blue), ellagic acid (green), and bismuth acetate (red). Note that the shown data correspond to the average of triplicates for each concentration, obtained in two independent experiments (a total of n = 6). The vertical error bars drawn in the diagram indicate the range of fluctuations (0.06–0.8) from which the standard deviations were calculated. Reprinted with permission [17]. Licensed under CC‐BY. Copyright 2020, American Chemical Society. (d) Proposed light‐triggered drug release mechanism for SU‐101. (O₁, O₂, O₃, O₄, and O₅ number the different atoms of oxygen in SU‐101). Reprinted with permission [16]. Copyright 2022, Elsevier Inc. (e) Snapshots of DCA (e₁–e₃) and α‐CHC (e₄–e₆) in CAU‐7 at different loadings: e₁) 121 mg g⁻¹ or 10.8 wt.%, e₂) 307 mg g⁻¹ or 23.5 wt.%, e₃) Saturation with 500 mg g⁻¹ or 33.3 wt.%, e₄) 42 mg g⁻¹ or 4.0 wt.%, e₅) 126 mg g⁻¹ or 11.2 wt.%, and e₆) Saturation with 248 mg g⁻¹ or 19.9 wt.%. The drug molecules are represented in the green‐stick mode. Reprinted with permission [32]. Copyright 2020, American Chemical Society. (f) Drug release profile of CAU‐17‐CTAB@dye in PBS solution at pH = 7.4. (g) Drug release profile of CAU‐17‐CTAB@dye in HCl aqueous solution at pH = 2.0 for 4 h after in PBS solution at pH = 7.4 for 8 h. (h) Schematic illustration of structural changes of CAU‐17‐CTAB in the gastrointestinal tract and drug release. Reprinted with permission [98]. Licensed under CC‐BY. Copyright 2023, Wiley‐VCH GmbH.

Besides SU‐101, more Bi‐MOFs, such as CAU‐7 and CAU‐17, have been applied to drug delivery. Orellana‐Tavra et al. utilized CAU‐7 with the organic ligand BTB³⁻ (1,3,5‐benzenetrisbenzoate) for co‐delivering dichloroacetate (DCA) and α‐cyano‐4‐hydroxycinnamic acid (α‐CHC), achieving high loadings of 33 and 9 wt.%, respectively (Figure 25e). Release durations reached 17 days (DCA) and 31 days (α‐CHC). Thermal amorphization of CAU‐7 extended α‐CHC release by 32%. CAU‐7 enhanced drug internalization and cytotoxicity against HeLa cells, highlighting its potential in biomedical applications. Our team has also explored the application of CAU‐17 in the field of drug delivery, focusing on its potential for intestine‐targeted drug release [98]. We initially selected three model dyes—methylene blue (MB), RhB, and methyl orange (MO)—to represent different types of drug molecules, to evaluate the loading capacity and drug release behavior of CAU‐17‐CTAB. The experimental results showed that the material exhibited high loading capacities for the cationic dyes MB and RhB, reaching 155 and 137 mg/g, respectively, while the loading capacity for the anionic dye MO was relatively low, at only 43 mg/g. Notably, the crystal structure of CAU‐17‐CTAB remained intact throughout the drug loading process, demonstrating excellent structural stability. In the drug release experiments, simulated gastric fluid (pH = 2) and simulated intestinal fluid (PBS buffer, pH = 7.4, containing phosphate) were used to mimic the human gastrointestinal environment and assess their responsive release behavior. The results indicated that the material maintained good chemical stability in gastric fluid, with a drug release amount of less than 10%, effectively preventing premature release in the stomach. In contrast, under intestinal conditions, CAU‐17‐CTAB underwent rapid structural degradation, and the drug release rate significantly increased, exceeding 80% within 10 min (Figure 25f). In further experiments simulating the gastrointestinal pathway, the drug‐loaded material was pretreated in gastric fluid for 4 h and then transferred to intestinal fluid. It still achieved efficient drug release, confirming its responsive and controllable release capability under complex physiological conditions (Figure 25g,h). Additionally, we conducted real drug loading experiments using procainamide (PA), a clinically used antiarrhythmic drug. The results showed that PA had a high loading capacity in CAU‐17‐CTAB, reaching up to 400 mg/g, which is significantly higher than that of the model dyes. FTIR spectroscopy further confirmed the successful encapsulation of PA molecules within the MOF structure, with the crystal structure remaining stable and undamaged. These findings once again demonstrate the excellent encapsulation capacity and structural compatibility of CAU‐17‐CTAB for cationic drugs.

4.5.2. Antibacterial Applications

As widely reported, Bi‐containing nanomaterials, Bi‐MOFs, also exhibit effective antibacterial activity in the form of individual MOF, or hybridized and modified MOFs, mainly relying on physical contact, metal ions and ligands, oxidative stress, photothermal effect, and synergistic effect [154, 155].

To date, various Bi‐MOFs have been demonstrated to be effective in the antibacterial field. Iram et al. investigated the antibacterial activity of four Bi‐MOFs with different morphologies [73], [Bi(C₆H₆O₂S₂N)₃]_n, [Bi(C₁₂H₁₆O₄S₄)₃]_n, [Bi(C₄H₈O₂SN)₃]_n, [Bi(C₇H₄O₂S)₃]_n, against five types of bacteria (Figure 26a). When compared to the reference drug erythromycin, enhanced antibacterial activity was observed. The increase in antibacterial performance was attributed to the chelation with bismuth ions. Through this coordination, Bi‐MOFs exhibited excellent inhibition against pathogenic bacteria, making them potential new antibacterial materials. Huang et al. synthesized bismuth‐based materials with ellipsoid and rod shapes [156], and the experimental results showed that rod‐shaped materials, i.e., Bi‐MOFs (CAU‐17), exhibited strong antibacterial effects against Gram‐negative bacteria (e.g., periodontal pathogens). At a concentration of 50 µg/mL, both ellipsoids and rods effectively disrupted the bacterial membrane, especially showing significant inhibition against the biofilm of Porphyromonas gingivalis (P. gingivalis) (Figure 26b,c). Minimum inhibitory concentration (MIC) tests showed that these Bi‐MOFs were highly effective in inhibiting common oral pathogens, especially Aggregatibacter actinomycetemcomitans (Aa), Fusobacterium nucleatum (Fn), and P. gingivalis.

FIGURE 26.

Current antibacterial application of Bi‐MOFs. (a) Synthesis of bismuth organic frameworks 1–4. Reprinted with permission [73]. Copyright 2019 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim. (b) Anti‐adhesion and (c) Eradication effects on biofilms (scale bar: 50 µm). Reprinted with permission [156]. Copyright 2022, Elsevier Ltd. (d) The antibacterial effects after treatment of CAU‐17. (e) Two possible photodegradation mechanisms: e₁ type II heterojunction and e₂ Z‐scheme heterojunction. (f) Photodegradation of complex bacterial communities in duck farm wastewater. Relative abundance of the top 20 complex bacterial communities in duck farm wastewater. Reprinted with permission [157, 158]. Copyright 2024, Elsevier B.V.

To enhance the antibacterial effects, researchers are currently attempting to construct heterojunction materials, including hybridized and modified Bi‐MOFs. For example, Li et al. constructed a type II p‐n heterojunction CAU‐17/MoS₂ composite photocatalyst by loading MoS₂ nanosheets onto the rod‐shaped CAU‐17 [157]. Compared to single MoS₂ or CAU‐17 materials, the composite exhibited significantly improved antibacterial activity under light exposure (Figure 26d). Recently, Yu et al. designed and synthesized a novel Z‐scheme heterojunction composite: In³⁺‐doped Bi/BiOCl/Bi(In³⁺)‐MOF [158]. In this material, BiOCl serves as the main photocatalytic component, while In³⁺ doping increases the positive surface charge. Simultaneously, in situ reduction generates metallic Bi to form a Z‐scheme heterojunction. The introduction of oxygen vacancies enhances the separation efficiency of photogenerated charge carriers, and the MOF component increases the specific surface area and provides a confinement effect. ESR and quenching experiments confirmed that the main reactive oxygen species (ROS) involved are hydroxyl radicals (·OH), superoxide radicals (∙O₂ ⁻), photogenerated holes (h⁺), and singlet oxygen (¹O₂). This material achieves efficient electron‐hole separation and ROS generation through the Z‐scheme charge migration pathway, thereby enhancing bactericidal activity and genetic damage (Figure 26e). It was capable of eliminating 107.34 CFU/mL of drug‐resistant E. coli within 80 min and completely inactivating complex microbial communities (e.g., from duck farm wastewater) within 100 min (Figure 26f).

4.5.3. Other Bio‐Medical Applications

Ni et al. found that Bi‐TBP offers a promising approach for local tumor treatment owing to its low toxicity and photodynamic activity [97]. The photosensitive Bi‐TBP can serve as a photodynamic therapy (PDT)‐based combination therapy carrier to effectively induce immunogenic cell death (ICD) of tumors by loading with CpG oligodeoxynucleotides, particularly suitable for local tumors with low systemic immune activation requirements. Li et al. designed a novel nano‐immunomodulator (Bi‐MOF‐L‐Cys@PEG@HA, abbreviated as BMCPH) [159], which enables tumor‐site‐specific release of H₂S through enzyme‐triggered reactions, addressing the challenge of delivering unstable gas molecules. By combining microwave thermal therapy with H₂S‐mediated immune modulation, the system achieved prolonged tumor suppression and reduced risks of recurrence and metastasis (Figure 27). Zhou et al. developed a bismuth‐based metal‐organic framework (Bi‐MOF@CS‐Se) nanodrug modified with selenized chitosan (CS‐Se). This is the first integration of Bi‐MOFs with selenized chitosan to realize multifunctional synergistic effects, including antibacterial, anti‐inflammatory, gastric retention, and gut microbiota protection. Bi‐MOF@CS‐Se responds intelligently to gastric acid, targets mucosal adhesion, and provides multi‐mechanism antibacterial and anti‐inflammatory actions, offering an efficient and safe treatment platform for Helicobacter pylori infections. Lee Robison et al. synthesized Bi‐NU‐901 [66], which has a similar topological structure to Zr‐NU‐901, and selected it as a CT contrast agent. In CT performance tests, Bi‐NU‐901 showed 14 times and 7 times higher CT contrast intensity than Iodixanol at 35 and 70 kV, respectively, outperforming the other two materials. Preliminary stability tests also showed that Bi‐NU‐901 remained stable in phosphate‐buffered saline (PBS) buffer for at least 24 h. Bi‐NU‐901 demonstrated excellent X‐ray absorption ability and is a promising material for CT contrast agents. This study provides a new direction for the application of MOFs in biological imaging, especially for the potential value of heavy‐element MOFs in multi‐energy CT and targeted imaging.

FIGURE 27.

Synthesis of Bi‐MOF@l‐Cys@PEG@HA (BMCPH) nanoparticles and a schematic diagram of BMCPH for mediating tumor microwave thermal therapy (MWTT) and reactivating the antitumor immune effect. Reprinted with permission [159]. Copyright 2023, American Chemical Society.

Taken together, these studies indicate that the biomedical relevance of Bi‐MOFs lies not only in their function as passive carriers, but also in their ability to integrate therapeutic delivery, imaging contrast, photodynamic or photothermal activity, and stimulus‐responsive behavior within a single platform. At the same time, the practical advancement of Bi‐MOFs in biomedicine will require more systematic evaluation of biodegradation, biosafety, long‐term stability, and in vivo clearance. Therefore, future work should place greater emphasis on clarifying how framework composition, Bi release behavior, and structural integrity collectively determine therapeutic efficacy and translational potential.

5. Challenges and Perspective

A balanced perspective on Bi‐MOFs should not be built simply around what they have not yet achieved relative to conventional MOFs. Their long‐term significance depends instead on identifying application scenarios in which they offer capabilities that are not readily accessible to classical MOFs, covalent organic frameworks (COFs), porous carbons, or zeolites. In this broader context, the value of Bi‐MOFs does not lie in competing directly with established porous materials in every area, but in exploiting the distinctive chemistry of bismuth.

In particular, the flexible and often asymmetric coordination environments of Bi³⁺, the photophysical consequences associated with a heavy p‐block element, its selective redox behavior, and its generally favorable biocompatibility together give Bi‐MOFs a profile that differs substantially from more established porous solids. Their future relevance is therefore unlikely to come from matching the highest surface areas, the most rigid pore architectures, or the best hydrolytic robustness already achieved elsewhere. Rather, it will depend on whether Bi‐specific functionality can compensate for, or even outweigh, these shortcomings in carefully defined applications. The key issue is thus not whether Bi‐MOFs can become universal replacements for existing porous materials, but where they can emerge as genuinely differentiated functional platforms (Figure 28).

FIGURE 28.

Challenges and prospects of Bi‐MOFs. The figure highlights the major challenges and future directions of Bi‐MOFs in structural design, performance optimization, application development, and industrial translation.

5.1. Where Bi‐MOFs are Unlikely to Dominate

Bi‐MOFs are unlikely to become leading porous materials in applications that mainly require ultrahigh surface area, rigid permanent porosity, and outstanding chemical stability. In such areas, mature porous platforms, including UiO‐, MIL‐, and HKUST‐type MOFs, as well as zeolites and porous carbons, remain more competitive. Their structural predictability, scalable synthesis, and long‐term operational stability are advantages that Bi‐MOFs do not yet provide in a consistent manner.

This limitation is closely related to the intrinsic coordination chemistry of bismuth. As discussed in Section 2, the stereochemically active 6s² lone pair and the strong hydrolysis tendency of Bi³⁺ often give rise to irregular coordination environments, variable cluster nuclearity, and a relatively narrow synthesis window [33, 34, 35]. As a consequence, Bi‐MOFs are generally less likely than many transition‐metal‐based MOFs to form highly robust and topologically predictable frameworks with permanently accessible pore systems. Although such structural flexibility may be advantageous in some cases, it is clearly disadvantageous where rigid and reproducible pore architectures are required.

From a practical point of view, these features are reflected in several recurring challenges highlighted throughout this review, including the difficulty of achieving green and reproducible synthesis, insufficient framework stability under realistic operating conditions, and the absence of unified criteria for evaluating true porosity [98, 160, 161, 162, 163, 164, 165, 166]. These issues make Bi‐MOFs less competitive in traditional porosity‐driven applications, such as bulk gas storage and highly standardized molecular separation, where reproducibility, hydrolytic stability, and processability are essential.

More broadly, in applications where bismuth‐specific chemistry offers no clear advantage—for example, where the key requirements are simply high conductivity, fast mass transport, exceptional water stability, or low‐cost large‐scale production—Bi‐MOFs should not be framed as a delayed attempt to reproduce what mature porous materials already do well.

Recognizing these limitations does not diminish the significance of Bi‐MOFs. On the contrary, it helps clarify their real value: not as universally competitive porous materials, but as candidates for specific application spaces where the distinctive coordination chemistry, electronic structure, and bio‐related properties of bismuth may provide advantages that other porous systems cannot easily offer.

5.2. Where Bi‐MOFs Could Become Uniquely Competitive

A meaningful position for Bi‐MOFs within the broader porous‐material landscape will most likely emerge in applications where Bi‐specific functionality is central to performance, rather than in areas dominated by conventional porosity metrics. In other words, Bi‐MOFs are most promising when they are evaluated not only by surface area, pore volume, or structural rigidity, but by whether the chemistry of bismuth can be translated into clear functional advantages.

One particularly promising direction lies in light‐driven and redox‐related systems. As a heavy main‐group element, bismuth exhibits strong spin–orbit coupling and distinctive electronic behavior, which makes Bi‐MOFs attractive for photocatalysis, photoelectrocatalysis, and selective redox transformations [28, 99, 128]. In these systems, the framework serves not merely as a porous scaffold, but as a structural platform that regulates the spatial distribution of Bi sites, local coordination environments, and interfacial electronic interactions. This provides a basis for exploiting the intrinsic photophysical and redox characteristics of bismuth in a controlled manner.

A second important direction is biomedicine. Compared with many transition‐metal‐based MOFs, Bi‐MOFs combine the relatively low toxicity and favorable biocompatibility of bismuth compounds with the modularity of MOF chemistry [66, 97, 159]. This combination is well suited to multifunctional biomedical platforms integrating drug delivery, photothermal or photodynamic therapy, imaging, and stimulus‐responsive release. In particular, the high atomic number of bismuth (Z = 83) is advantageous for X‐ray attenuation and CT imaging, while the framework structure can facilitate controlled cargo loading, degradation behavior, and targeting functions [66]. If in vivo degradation pathways and long‐term biosafety can be evaluated more systematically, Bi‐MOFs may become especially valuable in biomedical scenarios where conventional MOFs and simple inorganic Bi compounds each have inherent limitations.

Bi‐MOFs may also become particularly competitive as precursors for functional derivatives. In many catalytic and electrochemical systems, their main value may lie not in the direct use of the intact framework, but in their role as chemically programmable precursors. Bi‐MOFs provide a versatile route to porous Bi‐containing oxides, sulfides, carbons, and single‐atom‐regulated composites [109, 138, 140, 143]. Through precursor conversion, it becomes possible to control morphology, elemental distribution, and interfacial architecture in ways that are difficult to achieve through conventional solid‐state synthesis. This precursor‐oriented strategy is likely to remain one of the most practical routes for translating Bi‐MOF chemistry into functional materials and devices.

Another area of potential competitiveness lies in applications governed more by selective interfacial chemistry than by maximum pore volume. Representative examples include electrochemical CO₂ reduction to formate, localized catalytic transformations, and aqueous ion separation [109, 133]. In such systems, the distorted coordination environments of Bi centers and the presence of accessible Lewis acidic sites may create local chemical environments with enhanced affinity toward specific ions or reaction intermediates. Further tuning of charge distribution through bimetallic design or ligand regulation may also enable the development of Bi‐MOF for selective capacitive deionization and related separation processes. In this context, they may show potential for the selective electrosorption of challenging species such as fluoride (F⁻) [167] and chloride (Cl⁻) [168, 169], thereby addressing selectivity limitations commonly encountered with conventional carbon‐based electrodes. These advantages are not adequately captured by conventional descriptors such as BET surface area alone.

Taken together, these considerations suggest that the most realistic future of Bi‐MOFs lies in function‐oriented niches in which the chemistry of bismuth is directly linked to performance. Their long‐term value is unlikely to come from replacing established porous materials across the board, but from enabling combinations of structural tunability and Bi‐specific functionality that are difficult to realize with other material classes.

5.3. What Must be Solved to Reach this Positioning

If Bi‐MOFs are to occupy such a differentiated position, the next stage of research should move beyond the continued reporting of new structures and instead address a set of enabling problems that currently limit the credibility of this positioning. These are not merely generic challenges; each one directly determines whether Bi‐MOFs can be meaningfully distinguished from alternative materials.

The first requirement is improved synthetic controllability, including greener and more reproducible synthesis routes. At present, most Bi‐MOFs are still synthesized by solvothermal methods using organic solvents such as DMF [25], which raises concerns regarding sustainability, energy consumption, and scale‐up. Although microwave‐assisted [164], mechanochemical [165], ultrasound‐assisted [98, 170], and green‐solvent‐based approaches have been explored [171, 172], their implementation in Bi‐MOFs remains difficult because Bi³⁺ is highly susceptible to hydrolysis and competing side reactions. This challenge is illustrated by attempts to synthesize CAU‐17 in choline chloride‐based deep eutectic solvents (DES), which yielded BiOCl rather than the target framework, while analogous bromide‐containing systems produced BiOBr [173, 174, 175]. These examples indicate that green synthesis in Bi‐MOF chemistry is not simply a matter of process optimization, but a deeper chemical problem involving Bi precursor speciation, competing oxyhalide formation, and nucleation control. Unless these issues are addressed, Bi‐MOFs will remain difficult to prepare reproducibly and therefore difficult to position as reliable functional materials.

The second requirement is improved control over multimetal incorporation and framework stability. Bimetallic or heterometallic design is often proposed as a means of tuning charge distribution, active sites, and interfacial properties. In Bi‐MOFs, however, this strategy is particularly challenging because the large ionic radius of Bi³⁺, its high and often irregular coordination number, and its strong hydrolysis tendency can readily introduce lattice strain, coordination mismatch, or phase separation. More importantly, Bi‐MOFs may undergo rapid transmetalation under conditions relevant to synthesis or operation. For example, CAU‐17 can convert to HKUST‐1 in the presence of Cu²⁺ and to a La‐BTC phase in the presence of La³⁺, indicating that the parent Bi‐based framework may be readily lost through ion exchange. This means that, unless metal retention and genuine incorporation are rigorously demonstrated, many nominally bimetallic Bi‐MOFs remain chemically ambiguous. Future studies, therefore, need to combine more careful synthetic design with advanced characterization techniques such as HRTEM [176], XPS [177] and EXAFS to determine whether secondary metal ions are truly integrated into the Bi‐MOF structure rather than present as separate phases or transformation products [178, 179].

The third requirement is stricter validation of porosity, framework integrity, and working‐state structure. In Bi‐MOFs, crystallographic openness alone is not a guarantee of practical porosity, because apparently open frameworks may partially collapse or reconstruct during activation. Likewise, structures that appear intact under idealized characterization conditions may behave differently in water, under electrochemical bias, or during catalytic turnover. More consistent standards are therefore needed for activation protocols, gas sorption measurements, aqueous and electrochemical stability tests, and post‐use structural characterization. Without such standards, it remains difficult to determine whether a given Bi‐MOF should be regarded as a stable porous material, a metastable coordination solid, or a precursor that transforms into the actual active functional phase during operation.

The fourth requirement is clarification of the true functional phase. In catalysis, photocatalysis, and electrochemical systems, Bi‐MOFs often undergo reduction, reconstruction, or partial decomposition. In some cases, the observed activity may arise primarily from defect‐rich surfaces, in situ‐generated Bi nanoparticles, or reconstructed intermediate species rather than from the pristine framework itself [135]. This issue is central to the strategic positioning of Bi‐MOFs, because it determines whether their main value lies in serving as stable functional frameworks or as intentionally transformation‐prone precursors. Resolving this question will require broader use of operando and post‐operando methods to establish direct relationships between parent structure, the evolving active species, and the measured performance.

The fifth requirement is realistic translational assessment. Although bismuth is generally considered less toxic than many heavy metals, the practical safety of Bi‐MOFs also depends on ligand toxicity, residual solvent content, particle size effects, degradation products, and long‐term exposure behavior. At the same time, bismuth is more expensive than abundant metals such as Al or Zn, and many Bi‐MOFs still show limited conductivity, restricted porosity, or only moderate structural robustness under operating conditions. Accordingly, the practical relevance of Bi‐MOFs cannot be established through proof‐of‐concept performance alone. It must also be assessed through meaningful benchmarking against state‐of‐the‐art alternatives, including other MOFs, inorganic Bi‐based materials, porous carbons, and hybrid functional systems. Only through such comparison can the claimed differentiation of Bi‐MOFs be convincingly validated.

5.4. Long‐Term Significance in the Broader Landscape of Porous Materials

Even if the challenges are addressed systematically, Bi‐MOFs are unlikely to redefine the porous‐materials field through record surface area, maximum framework rigidity, or universal applicability. Their long‐term significance lies elsewhere. More realistically, Bi‐MOFs should be viewed as a specialized class of function‐oriented porous materials that connect several domains usually treated separately: porous framework chemistry, Bi‐centered electronic and redox functionality, biomedically relevant heavy‐element behavior, and derivative‐enabled nanostructured material design.

This perspective is important because it positions Bi‐MOFs not as delayed competitors to mature MOFs, but as materials whose strongest opportunities arise precisely in areas where conventional porous frameworks are not intrinsically optimized. Classical MOFs are typically valued for structural regularity, permanent porosity, and framework robustness. By contrast, simple inorganic Bi compounds are more often used for composition‐driven optical, catalytic, or biomedical functions, but they lack the structural modularity of MOF chemistry. Bi‐MOFs occupy an intermediate yet potentially powerful position between these two material spaces. In principle, they can combine programmable framework architectures with Bi‐specific functionality, while also serving as chemically designable precursors to derivative materials with controlled morphology, dispersion, and interface structure. This dual role is not fully reproduced by either conventional MOFs or simple Bi‐based solids alone.

From this perspective, the future of Bi‐MOFs should be judged less by whether they can catch up with other porous materials on conventional metrics, and more by whether they can enable material functions that are otherwise difficult to realize. Their real strategic value lies in the possibility of integrating Bi‐centered photophysical behavior, redox activity, interfacial selectivity, and biomedical utility within structurally tunable framework systems. If this integration can be demonstrated convincingly and reproducibly, Bi‐MOFs will not need to outperform all other porous materials in general terms to justify their importance. Their significance will instead lie in establishing a differentiated functional territory within the broader porous‐material landscape.

Accordingly, the next stage of Bi‐MOF research should become more function‐driven, benchmark‐driven, and mechanism‐driven. The central question is no longer how many new Bi‐MOF structures can be reported, but which among them can occupy application spaces where their combination of porous framework design and Bi‐specific chemistry provides genuine and nontrivial advantages. Only under this logic can the field move beyond structural novelty and develop into a conceptually mature and sustainable branch of porous materials research.

6. Summary and Outlook

This review systematically summarizes recent advances in Bi‐MOFs, with a particular focus on the chemical behavior of Bi³⁺ ions in solution and the complexities associated with their coordination behavior. The diversity in coordination geometries and topological networks of Bi‐MOFs has been comprehensively analyzed, highlighting their distinct structural features and promising prospects for future development. In terms of synthesis strategies, various reported methodologies for Bi‐MOFs preparation have been critically reviewed, comparing their respective advantages and limitations. Additionally, the review presents a detailed overview of the applications of Bi‐MOFs across diverse fields, recently including catalysis, energy storage, biomedical imaging, drug delivery, fluorescence sensing, adsorption, and separation, emphasizing their multifunctional capabilities and broad utility as advanced functional materials.

Despite the significant progress and clear advantages demonstrated by Bi‐MOFs, considerable challenges remain in structural design, precise synthetic control, stability optimization, and the expansion of their application scope. Ultimately, this critical review aims to provide new insights and inspire researchers by establishing a theoretical foundation and suggesting promising directions for the ongoing development of Bi‐MOFs. It is anticipated that these perspectives will encourage further investigation and facilitate the practical deployment of Bi‐MOFs within the broader MOF research community.

Conflicts of Interest

The authors declare no conflicts of interest.

Biographies

Jingjing Dai is currently a second‐year Master's student at Zhejiang Sci‐Tech University. Her research interests mainly focus on the green synthesis of metalorganic frameworks (MOFs) and their photoelectric performance applications.

Haowei Yang received his Master's degree from the School of Materials Science and Engineering at Zhejiang Sci‐Tech University. His research interests focus on the synthesis of Bi‐based metalorganic frameworks (Bi‐MOFs) and their applications in environmental remediation.

Liang Huang, male, graduated from the Department of Chemical and Biological Engineering, Zhejiang University with a Master's degree. He is a Senior Engineer and currently serves as Vice President of the Chemical Engineering Design Institute of Zhejiang Tianzheng Engineering Co., Ltd., as well as General Manager of the Second Chemical Design Department.

Yi Guo received his Ph.D from Zhejiang University under the supervision of Prof. Xinsheng Peng, and worked as a postdoctoral researcher at Nanyang Technological University with the supervision of Prof. Yanli Zhao. He is now a faculty in Jiangsu university since 2021. His research interests focused on porous functional membrane, CO2 electroreduction and fuel cells.

Bing Wu is currently a postdoctoral researcher at the University of Chemistry and Technology Prague, where he received his Ph.D. in December 2023. His research primarily focuses on the synthesis and exfoliation of 2D materials, as well as their applications in energy storage and conversion.

Sheng Wang is a professor and doctoral supervisor at Zhejiang Sci‐Tech University. He received his PhD in Engineering in Japan in 2003. From 2003 to 2005, he worked as a part‐time lecturer at the Catalysis Research Center of Hokkaido University, Japan, and joined Zhejiang Sci‐Tech University in December 2005. His research interests include fiber composites, antibacterial and antiviral composite materials, photocatalytic materials, material surfaces/interfaces, and textile printing and dyeing finishing.

Zdenek Sofer completed his PhD at the University of Chemistry and Technology Prague (Czech Republic) in 2008 with Prof. Stejskal. He worked at Forschungszentrum Jülich (Germany; 20042005) and at Universität Duisburg‐Essen (Germany; 2009). Since 2008 he has been a faculty member at University of Chemistry and Technology Prague (Czech Republic). In 2013 and 2016 he received the rector prize for young talented scientists and in 2016 a Neuron Impuls grant. His interests are in the area of twodimensional nanomaterials and electrocatalysis.

Yulong Ying is an associate professor in the School of Materials Science and Engineering, Zhejiang Sci‐Tech University. He received his Ph.D. degree in the School of Materials Science and Engineering, Zhejiang University (China, 2017). He has been a senior scientist in the Center for Advanced Functional Nanorobots at the University of Chemistry and Technology, Prague (Czech Republic). His research focuses on micro/nanomotor design with MOFs and stimuli‐responsive materials for environmental remediation, green MOF synthesis, and 2D/nanostructured materials for water treatment and energy storage.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.