High-throughput synthesis of zinc-functionalized bio-ceramics propels the rational design of bone implants

Abstract

Endowing basic ceramics with functional elements effectively improves their physicochemical and biological properties. Nevertheless, these properties are strongly influenced by particle size, thermodynamic and kinetic conditions, and lattice structure. Such complexity requires extensive trial-and-error optimization, greatly limiting materials development efficiency. To overcome this challenge, we established a high-throughput synthesis platform that enables precise control over the morphologies and crystalline phases of zinc-doped calcium phosphate (Zn-CaPs) by independently manipulating 18 reaction times and 48 reaction temperatures, facilitating the execution of up to 144 experiments concurrently. The potential mechanisms by which system concentration, zinc/calcium ratio, pH, reaction time, temperature, and binary solvent concentration influence the crystallization of Zn-CaPs were investigated. The Zn-CaPs exhibited remarkable cytocompatibility, osteogenic bioactivity, and antibacterial properties. A porous scaffold (Zn-CaPS) fabricated from Zn-CaPs demonstrated a strength of over twice that of a commercial CaPs scaffold. Notably, after 6 and 12 weeks of implantation in a rabbit femoral defect model, the Zn-CaP scaffold achieved over 4-fold and 2-fold higher BV/TV than the control and commercial CaPs scaffolds, respectively. This work develops an innovative material genome framework for high-performance ion-doped bioceramics. It integrates high-throughput design, synthesis, and performance evaluation to accelerate R&D and yield socioeconomic benefits.

Graphical abstract

Highlights

• •A high-throughput synthesis platform (18 reaction times, 48 temperatures, 144 parallel experiments) is established for Zn-CaPs.
• •Key parameters' regulatory mechanisms on Zn-CaPs’ structure, morphology, and physicochemical properties are clarified.
• •High-throughput screened Zn-CaPs exhibit excellent cytocompatibility, osteogenic bioactivity, and antibacterial properties.
• •Zn-CaPs scaffold has superior mechanical strength and in vivo osteogenic capacity vs commercial CaPs scaffolds.

1. Introduction

The traditional materials development process takes decades to identify novel materials suitable for technological applications and even longer to refine them for industrial implementation [1]. During biomaterial synthesis, researchers frequently modulate relevant parameters based on experimental experience and a review of the existing literature [2]. The traditional trial-and-error approach to synthesizing biomaterials is inefficient, costly, and time-consuming, making it challenging to link experimental parameters to experimental outcomes [3,4]. Additionally, traditional synthesis techniques face challenges due to variations in reagent batches, durations, and ambient temperature, affecting the materials prepared [2]. Inspired by the Human Genome Project, the “Material Genome Initiative” (MGI) integrates high-throughput computation, experiments, and dedicated materials databases to revolutionize conventional materials development. It targets the inefficiencies of traditional R&D to accelerate materials discovery, commercialization, and deployment at lower cost [5]. “Materials Genome Technology” aims to accelerate material development by utilizing high-throughput experimental devices to analyze material gene chips and create a comprehensive material database [6]. High-throughput technology relies on the capability to conduct multiple parallel experiments quickly. Regrettably, high-throughput technology is primarily applied to semiconductor, ceramic, and nanomaterials [[7], [8], [9], [10]]. High-throughput methods have only recently been applied to the creation of new biomaterials and their structures or substrates for potential use in regenerative medicine [11,12]. Fang et al. developed an innovative biomaterial design approach by combining high-throughput gradient surface generation with statistical learning, leading to a unique ternary functional surface for orthopedic implants that optimally supports osteogenesis, angiogenesis, and neurogenesis [13]. Bioceramic materials are typically produced via the hydrothermal method, which enables the formation of crystalline powders without high-temperature sintering and offers advantages such as low energy consumption, high yield, and a straightforward preparation process [[14], [15], [16]]. Furthermore, the biological properties and application range of bioceramics prepared utilizing the hydrothermal technique can be greatly enhanced [17,18]. However, the synthesis process of this method is affected by various factors such as reaction time, temperature, pH, element species, and ratios, and employing high-throughput techniques can significantly improve efficiency in hydrothermal synthesis. To the best of our knowledge, there is currently no research on the application of hydrothermal technology in high-throughput bioceramic synthesis. Despite researchers' extensive understanding of the structure-activity relationships in multi-element materials, similar genomics studies in ceramic materials remain relatively limited.

Calcium phosphate (CaPs) is essential for body support, organ protection, and blood formation as the primary inorganic mineral in human hard tissues [19]. In 1920, CaPs were proposed for fracture treatment; their efficacy was even superior to autografts in certain instances [20,21]. Owing to their outstanding properties, CaPs could also be used for gene delivery, cancer therapy, catalysis, environmental governance, and other fields [[22], [23], [24]]. Certainly, CaPs have notable drawbacks, including brittleness, low strength, poor toughness, and limited biological function, which hinder their ability to meet the demands of various clinical therapies [[25], [26], [27], [28], [29], [30]]. Research indicates that combining CaPs with polymers and growth factors can eliminate their limitations [25,31]. Low-concentration functional ions show advantages in cost and stability over protein-based growth factors for bone tissue repair [32].

Zinc (Zn²⁺) is referred to as the “21st-century calcium” for its essential role in the metabolism, growth, and development of organisms, and Zn-based biodegradable materials are gaining traction for their biocompatibility and regenerative properties [33,34]. The bactericidal, osteogenic, and cardiac function-preserving actions of Zn²⁺ have been well documented [33,35]. Therefore, Zn²⁺ was added to CaPs to enhance their biological functions, endowing them with outstanding osteogenic, angiogenic, and antibacterial properties. Zn²⁺ doping into β-tricalcium phosphate (β-TCP) could change the microstructure of β-TCP and thus enhance its mechanical strength [36]. In addition, Zn²⁺ enhanced the osteogenic effect of biphasic calcium phosphate by regulating the phasor ratio of HA and β-TCP, influencing ectopic new bone formation in vivo [37]. Zn²⁺ modified the morphology and lattice of CaPs, enhancing their biological properties, stabilizing β-TCP, and facilitating the transformation of HA into the β-TCP phase [36,38]. Despite this, there is still insufficient understanding of Zn²⁺‘s role in converting HA to β-TCP, which limits insights into its biological effects.

Herein, we constructed a high-throughput synthesis device featuring 144 reaction sites, 48 temperatures, and 18 times, as detailed in Figs. S1–13 and Tables S1–7. Taking the Zn²⁺-modified CaPs as the research topic, combining high-throughput technology with the hydrothermal method to systematically investigate the influence of Zn²⁺ on the physicochemical properties and the biological functions of CaPs, as shown in Scheme 1 and Video S1. The research focused on the cytocompatibility, osteogenic differentiation, and antibacterial effects of Zn-CaPs. Additionally, the physicochemical properties and in vivo bone-regeneration effects of Zn-CaPS, constructed using 3D printing technology, were thoroughly analyzed.

Scheme 1.

Schematic Diagram of the Application and Research Process of Hydrothermal Technology in High-throughput Bio-ceramic Synthesis.

Supplementary data related to this article can be found online at https://doi.org/10.1016/j.bioactmat.2026.06.001

The following are the Supplementary data related to this article:

2. Results and discussion

2.1. High-throughput screening of ammonia content and reactant concentration for Zn-CaPs synthesis

The influence of ammonia content and reactant concentration on the formation of Zn-CaPs crystals was first analyzed simultaneously. The morphological results of CaPs samples synthesized with varying ammonia content and concentration gradients of Ca²⁺+Zn²⁺ on the equipment platform are shown in Fig. 1A and Fig. S14, where the molar ratio of Zn²⁺/(Ca²⁺+Zn²⁺) is 10 mol%. The volume of ammonia water added was successively increased from 0.1 mL to 3 mL from column 1 to column 9; the reaction concentration of Ca²⁺ with Zn²⁺ was raised from 0.1 M to 2.0 M from row 1 to row 6 (namely 10Zn-M_XA_Y, X = 0.1, 0.2, 0.5, 1.0, 1.5, and 2.0 M, Y = 0, 0.1, 0.2, 0.5, 1.0, 1.5, 2.0, 2.5, and 3 mL). Ammonia content significantly alters the micromorphology of the synthesized CaPs. Increasing ammonia content transformed Zn-CaPs micromorphology from agglomerates to cubic particles, which then gradually disintegrated into irregular nanoparticles. By contrast, increasing system concentration (The total concentration of (Zn²⁺+Ca²⁺) in the reaction system) induced the opposite morphological evolution to that of rising ammonia content. Phase analysis of Zn-CaPs powders revealed that the diffraction peaks of the main product components matched those of CaHPO₄ (Monetite) in ammonia-free or low-ammonia reaction systems (e.g., 10Zn-M_0.1A₀ and 10Zn-M_1.5A_0.5). Increasing ammonia content (e.g., 10Zn-M_0.1A_0.2) led the main diffraction peaks of samples to match those of β-Ca₃(PO₄)₂ (β-TCP); further elevation (e.g., 10Zn-M_0.1A_1.0) ultimately yielded low-crystallinity hydroxyapatite (Ca₁₀(PO₄)₆•(OH)₂, HA), as shown in Fig. 1B. Conversely, Zn-CaPs phase composition exhibited the opposite trend with increasing system concentration (e.g., 10Zn-M_0.1A_1.0 vs. 10Zn-M_0.5A_1.0, 10Zn-M_0.5A_2.0 vs. 10Zn-M_1.5A_2.0; Fig. 1C).

Fig. 1.

The effects of ammonia amount and system concentration on the morphology and phase of Zn-CaPs. (A) SEM images of Zn-CaPs prepared with different ammonia amounts and different system concentrations, where the molar ratio of Zn²⁺/(Ca²⁺+Zn²⁺) is 10 mol%, namely 10Zn-M_XA_Y, the samples marked by the red dashed box were selected to further investigate the influences of other synthetic parameters; (B) XRD patterns of 10Zn-M_0.1A_Y and 10Zn-M_1.5A_Y prepared with different ammonia amounts, where Y represents the volume of ammonia (Y = 0, 0.2, 0.5, 1.0, 2.0, and 4.0 mL, respectively); (C) XRD patterns of 10Zn-M_XA_0.2, 10Zn-M_XA_1.0, and 10Zn-M_XA_2.0 prepared with different system concentrations, where X represents the concentration of (Ca²⁺+Zn²⁺) (X = 0.1, 0.5, 1.0, and 1.5, respectively). The reaction temperature and time were 150 °C and 8 h, respectively.

Crystallization nucleation is driven by supersaturation to overcome the energy barrier, with higher supersaturation accelerating crystallization and promoting crystal growth. However, excessive supersaturation that exceeds the limiting value of the metastable zone induces the formation of an excessive amount of crystal nucleus. Rapid explosive nucleation in the solution depletes calcium and phosphorus sources, exhausting free Ca²⁺, phosphate, and other ions required for CaP particle growth and thus forming small, irregular nanoparticles [39]. The formation enthalpy, the energy released or absorbed as reactants convert from ionic to product crystalline states, correlates with crystal structural stability [40]. In the absence of ammonia, OH⁻ forms via H₂O ionization; the system is then severely deficient in OH⁻, leading to simultaneous H₂O ionization and hydrolysis. At 90 °C, the water ionization constant is approximately 55 × 10⁻¹⁴ mol²/L², giving a system OH⁻ concentration of 7.4 × 10⁻⁷ mol/L. When the dimensionless numerical ratio of ammonia-to-system concentration is insufficient, abundant Ca²⁺ and HPO₄⁻ combine with trace OH⁻; low-enthalpy CaHPO₄ thus precipitates first and subsequently crystallizes into the CaHPO₄ phase. However, in confined systems with low c (OH⁻)/c (Ca²⁺+Zn²⁺), low supersaturation and enthalpy of the products hindered the conversion of clustered CaHPO₄ to high-enthalpy, crystalline β-Ca₃(PO₄)₂. Zn²⁺ doping induced preferential crystal growth along the (0210), (128), (214), (220), (1010), and (4010) planes, yielding cubic β-Ca₃(PO₄)₂ (Fig. 1C). Elevated ammonia content increased the c (OH⁻)/c (Ca²⁺+Zn²⁺) ratio, boosting product supersaturation and formation enthalpy and thus promoting the crystalline growth of clustered CaHPO₄. Ultimately, this transformation of the reactants into the high-enthalpy β-Ca₃(PO₄)₂ was achieved. Further increases in system ammonia content raised the c (OH⁻)/c (Ca²⁺+Zn²⁺) ratio to excess levels, leading to HPO₄²⁻ and Ca²⁺ in relatively low concentrations, which markedly elevated product supersaturation, triggering rapid crystal nucleation and precipitation of high-enthalpy spherical HA nanoparticles (Fig. 2A). By contrast, system concentration exerted the opposite effect on product supersaturation and formation enthalpy to ammonia content, leading to the reverse regulatory trend in Zn-CaP phase composition and micromorphology (Fig. 2B).

Fig. 2.

The effects of process parameters on the supersaturation of reaction products. (A) Ammonia content; (B) System concentration; (C) Ammonia content and reaction temperature; (D) Zinc content.

Moreover, in a neutral environment, free phosphate ions in the system exist as H₂PO₄⁻ (monovalent) and, at low concentrations, HPO₄²⁻ (divalent), rather than PO₄³⁻ (trivalent) [41]. Elevated OH⁻ in the reaction system reacts with H₂PO₄⁻ and HPO₄²⁻ to form H₂PO₄⁻ and PO₄³⁻. Increasing OH⁻ levels lead to the accumulation of excess OH⁻ around Ca²⁺ in solution, ultimately forming a stable HA structure [42]. Thus, increasing ammonia content caused the dominant CaP phase to shift progressively from CaHPO₄ to β-Ca₃(PO₄)₂, and ultimately to Ca₁₀(PO₄)₆(OH)₂.

FTIR analysis confirmed characteristic CaP absorption peaks at 1120, 1020, 605, and 549 cm⁻¹ for all prepared samples, as highlighted by the grey regions in the spectra (Fig. 3A and B) [43]. The peaks at 1120 and 1020 cm⁻¹ arose from the antisymmetric stretching vibration of PO₄³⁻, while those at 605 and 549 cm⁻¹ stemmed from its asymmetric bending vibration. At lower pH, characteristic peaks from P–O(H) stretching vibrations of HPO₄²⁻ and PO₄³⁻ were observed at 874 and 1120 cm⁻¹. The peak at 1380 cm⁻¹ was assigned to surface-adsorbed CO₃²⁻, while the broad bands at 3447, 3283, and 1650 cm⁻¹ corresponded to O–H stretching and H₂O bending vibrations, respectively [44,45]. The band at 650 cm⁻¹ corresponds to the crystalline water absorption of scholzite [46]. Moreover, the antisymmetric stretching peak of PO₄³⁻ shifted to higher wavenumbers with increasing pH, corresponding to a blue shift. Infrared spectra reflect molecular vibrations, which involve low-energy transitions and oscillations. A decrease in the characteristic peak wavenumber implies lower energy required for molecular vibration and greater group instability. Thus, we hypothesize that a higher pH (i.e., increased ammonia) favors the stabilization of CaPs, yielding more stable products.

Fig. 3.

The effects of ammonia content and system concentration on the chemical structure, mass percentages of Zn²⁺, Ca²⁺ in Zn-CaPs, Zn²⁺ doping rate, crystal structure, and element composition of Zn-CaPs. (A) FTIR spectra of 10Zn-M_0.1A_Y and 10Zn-M_1.5A_Y prepared with different ammonia amounts, where Y represents the volume of ammonia (Y = 0, 0.2, 0.5, 1.0, 2.0, and 4.0 mL, respectively); (B) FTIR spectra of 10Zn-M_XA_0.2, 10Zn-M_XA_1.0, and 10Zn-M_XA_2.0 prepared with different system concentrations, where X represents the concentration of (Ca²⁺+Zn²⁺) (X = 0.1, 0.5, and 1.5, respectively); (C1-C3) Atomic absorption spectroscopy (AAS) of 10Zn-M_0.1A_Y, 10Zn-M_0.5A_Y and 10Zn-M_1.5A_Y prepared with different ammonia amounts, where Y represents the volume of ammonia (Y = 0, 0.2, 0.5, 1.0, and 2.0 mL, respectively); (D1-D3) AAS of 10Zn-M_XA_0.2, 10Zn-M_XA_1.0, and 10Zn-M_XA_2.0 prepared with different system concentrations, where X represents the concentration of (Ca²⁺+Zn²⁺) (X = 0.1, 0.5, and 1.5, respectively); (E) TEM images of different samples; (F) Element mapping of different samples. Scale bar: 2 μm. The reaction temperature and time of 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 were set at 150 °C and 8 h, respectively. The grey areas in FTIR spectra represent the absorption peak of PO₄³⁻. The reaction temperature and time were set at 150 °C and 8 h, respectively.

At a fixed system concentration, atomic absorption spectroscopy showed zinc doping efficiency decreased with increasing ammonia water addition. At constant ammonia levels, doping efficiency increased with higher system concentration (Fig. 3C1-C3, D1-D3). Wang et al. confirmed that Zn²⁺ reacts with ammonia to form the zinc-ammonia complex [Zn(NH₃)₄]²⁺, which retains partial Zn²⁺ in solution and thus reduces the zinc doping efficiency of the samples [47]. 10Zn-M_1.5A_2.0 exhibited a higher actual zinc doping efficiency than 10Zn-M_1.5A_0.5, as its dominant phase contained monetite that was not fully converted to β-TCP. Ca²⁺ and Zn²⁺ concentrations of the 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 after 1, 3, and 7 days of PBS immersion are shown in Fig. S15. Ca²⁺ concentrations decrease sequentially due to increased phase stability. 10Zn-M_0.5A_1.0 has the highest Zn²⁺ concentration (higher Zn²⁺ doping rate, Fig. 3C2).

2.2. High-throughput screening of reaction temperature and reaction time for Zn-CaPs synthesis

10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 were selected to investigate the effects of reaction temperature and time on the morphology, phase composition, and Zn²⁺ doping efficiency of Zn-CaP products. The results demonstrated that reaction time and temperature modulated the Zn-CaP morphology consistently (Fig. 4A–S16, S17). Unreacted 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 featured irregular spherical/striped nanoparticles and lamellar structures (Fig. S16). With increasing reaction time and temperature, the lamellar structure of 10Zn-M_0.5A_0.2 shortened and eventually disappeared, and its Zn-CaP product transformed from large-particle agglomerates to small-particle agglomerates; 10Zn-M_0.5A_1.0 evolved from spherical particles and long lamellae into elongated striped structures with distinct edges, and finally cubic particles; for 10Zn-M_0.5A_2.0, the as-synthesized Zn-CaPs first changed from striped nanoparticles and lamellae to spherical nanoparticles, with no further morphological variation thereafter (Fig. S16). XRD patterns confirmed that unreacted 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 exhibited monetite, brushite, and HA phases, respectively. However, with increasing reaction temperature and time, the

Fig. 4.

The effects of reaction temperature, reaction time, and zinc content on the morphology and phase of Zn-CaPs. (A) SEM images of Zn-CaPs prepared at different reaction temperatures, reaction times, and zinc contents; (B-D) XRD patterns of Zn-CaPs prepared at different reaction temperatures and reaction times, where Zn²⁺/(Ca²⁺+Zn²⁺) = 10 mol%; (E) XRD patterns of different XZn-M_0.5A_1.0 (X = 0, 1, 3, 5, 10, and 20 mol%). The reaction temperature and time were set at 150 °C and 8 h, respectively.

phase of 10Zn-M_0.5A_0.2 gradually evolved from monetite to brushite and back to monetite; 10Zn-M_0.5A_1.0 transformed from brushite to β-TCP; and 10Zn-M_0.5A_2.0 retained HA as its dominant phase with no phase variation (Fig. 4B–D, S18-S19). Frost et al. demonstrated that Ca²⁺ reacts with HPO₄²⁻ and H₂O to form brushite [48]. Previous studies have shown that increased OH⁻ promotes the hydration of monetite to form brushite, while excess OH⁻ induces the transformation of the product to HA [49,50]. Brushite dehydrates upon heating to form monetite [51]. As the ammonia content increases, HPO₄²⁻ reacts with OH⁻ to produce PO₄³⁻, thus causing Monetite to transform into the β-TCP phase. Increasing ammonia content drives HPO₄²⁻ to react with OH⁻ to form PO₄³⁻, thereby converting monetite to the β-TCP phase [42]. Excess ammonia yielded HA as the product's dominant phase, and a higher initial pH in the CaPs synthesis system favored the formation of more crystalline products with high Ca/P molar ratios [52]. The addition of a fixed amount of ammonia meant that reaction temperature and time did not affect the dominant β-TCP and HA phases in the products. While the Gibbs free energy of the reaction system is temperature-dependent, elevated reaction temperatures provide a phase transition driving force, facilitating the transformation of the parent phase into a more stable new phase [53]. HA is the most thermodynamically stable crystalline CaPs salt, so increasing reaction temperature or extending reaction time does not induce phase changes [54]. Moreover, reaction temperature and concentration elevation increase the enthalpy of the reaction system. The effects of reaction temperature and ammonia content on product morphology were elucidated by estimating the supersaturation changes of the reaction products. Low ammonia and temperature levels led to low supersaturation in the system, thus causing the low-enthalpy CaHPO₄ to precipitate first in the solution. Low-enthalpy CaHPO₄ exhibited inhibited crystal directional growth and formed agglomerated particles instead. With increasing ammonia content or reaction temperature, elevated system supersaturation and enthalpy facilitated the directional crystal growth of reaction products, yielding β-Ca₃(PO₄)₂ with a crystalline lamellar structure. With further increases in system ammonia content or reaction temperature, the supersaturation and enthalpy of reaction products rose sharply, inhibiting directional crystal growth and leading to the formation of low-solubility, high-enthalpy HA nanoparticles (Fig. 2C). With increasing reaction temperature and time, the dominant phases of Zn-CaPs products for 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 under different synthesis conditions were monetite, brushite, and β-TCP, respectively. Their functional groups and chemical compositions were further verified via FTIR analysis (Fig. S20). TEM was used to investigate the internal structure of Zn-CaPs samples synthesized at 150 °C for 8 h 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_1.0 exhibited distinct crystalline surface structures, corresponding to monetite, β-TCP, and HA, respectively, consistent with previous physicochemical analyses (Fig. 3E) [55]. EDS mapping revealed homogeneous distribution of Ca, P, O, and Zn across the surfaces of 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 (Fig. 3F). EDS spectra (Fig. S21) and elemental analysis (Table S8) show that the Zn²⁺/(Zn²⁺+Ca²⁺) ratios of 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 decline progressively, matching Fig. 3 (C2). Moreover, reaction time and temperature effects on Zn²⁺ doping efficiency in Zn-CaPs correlated with variations in dominant phase, crystallinity, and secondary phase, leading to fluctuating actual Zn²⁺ doping efficiencies in high-throughput synthesized CaPs samples (Fig. 5A and B).

Fig. 5.

The effects of reaction temperature and reaction time on the mass percentage of Zn²⁺ and Ca²⁺ in Zn-CaPs, as well as the Zn²⁺ doping rate. AAS of 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 were prepared at different reaction temperatures (A) and reaction times (B), respectively.

2.3. High-throughput screening of Zn content for Zn-CaPs synthesis

We further investigated the effect of Zn addition on the morphology, phase, and Zn²⁺ doping efficiency of CaPs. For samples prepared under the M_0.5A_1.0 condition, increasing Zn doping transformed the Zn-CaPs structure from rod-like nanoparticles to rod-like nanoparticle agglomerates; the rod-like nanoparticles within these agglomerates then gradually fused to form crystalline surfaces, which further developed into complete cubic particles. In this process, rod-like nanoparticles gradually disappeared, and cubic particles increased in abundance, consistent with the trends shown in Fig. 4A and S22. With increasing Zn doping, the samples maintained a cubic morphology. XRD analysis shows that 0Zn-M_0.5A_1.0 had monetite as its primary phase; introducing Zn²⁺ into the system shifted the Zn-CaPs phase to β-TCP, likely due to Zn²⁺‘s stabilizing effect on β-TCP [56]. With increasing Zn addition, the Zn-CaP phase ultimately transformed to scholzite (Fig. 4E). Gomes et al. confirmed that Zn²⁺ converts β-TCP to scholzite by fully occupying the Ca4 and Ca5 sites [38]. XRD Rietveld (Table S9) refinement results revealed that both the lattice parameters (a, c) and unit cell volume of the as-synthesized CaPs gradually decreased with increasing Zn doping content [38,57]. This phenomenon is consistent with the lattice shrinkage induced by the substitution of Ca²⁺ (ionic radius: 1.00 Å, CN = 6) with smaller Zn²⁺ (0.74 Å, CN = 6) [38]. XPS full spectra (Fig. S23 and Table S10) confirmed the characteristic binding energy shifts of Ca 2p and Zn 2p, indicating that Zn²⁺ was chemically bonded within the lattice rather than merely physically adsorbed on the surface [58]. Combined with the variations in Ca/P ratio and Zn²⁺/(Ca²⁺+Zn²⁺) atomic ratio, these results collectively demonstrate that Zn²⁺ ions have been successfully incorporated into the CaPs lattice by replacing Ca²⁺ sites.

FTIR analysis confirmed the characteristic functional groups of monetite, β-TCP, and scholzite in the synthesized CaP samples (Fig. S24A). Fig. S24B shows that the actual Zn²⁺ doping efficiency in high-throughput synthesized Zn-CaP analogs increased with increasing Zn addition, indicating that Zn²⁺ gradually replaced Ca²⁺ in the β-TCP structure, consistent with the theoretical formula Ca_3-xZn_x (PO₄)₂ [59]. At low Zn²⁺ concentrations, Zn²⁺ did not participate in the reaction. Rapid reaction of Ca²⁺, HPO₄²⁻, and OH⁻ yielded low-supersaturation, low-enthalpy rod-like CaHPO₄ nanoparticles, which precipitated first in the system. Increasing Zn²⁺ concentration raised the c (Zn²⁺)/c (Ca²⁺+Zn²⁺) ratio, elevating the supersaturation and enthalpy of reaction products; this promoted the fusion of rod-like CaHPO₄ nanoparticles into a lamellar crystalline structure, converting the products to high-enthalpy β-Ca₃(PO₄)₂. At this stage, Zn²⁺ in the β-Ca₃(PO₄)₂ lattice interstices gradually participated in the reaction, leading to the gradual formation of CaZn₂(PO₄)₂·2H₂O in the product. At high Zn²⁺ concentrations, Zn²⁺ participated fully in the reaction; the product reached high supersaturation, Zn²⁺ substituted partial Ca²⁺ lattice sites, and a cubic CaZn₂(PO₄)₂·2H₂O single-crystal structure was formed. This process is schematically illustrated in Fig. 2D. Calculations based on supersaturation theory yielded the supersaturation values of reaction products as a function of system concentration, ammonia dosage, reaction temperature, reaction time, and Zn addition, presented in Table S11–13.

2.4. High-throughput screening of binary solvent system for Zn-CaPs synthesis

Adding organic co-solvents to aqueous crystallization solutions modulates the incorporation of soluble additives and induces marked changes in crystal morphology, enabling tailored design of material morphology, structure, and composition—an outcome unattainable in pure aqueous systems [60]. Herein, this platform was used to establish a binary solvent concentration gradient for investigating the effects of varying ethanol/water volume ratios on CaPs synthesis. A series of Zn-CaPs samples were synthesized under the 10Zn-M_0.5A_1.0 condition. SEM images revealed that the morphology of unreacted Zn-CaPs evolved from spherical nanoparticles to irregular lamellar and blocky structures, eventually forming porous microsphere agglomerates assembled from nanoparticles (Fig. 6A). Post-reaction Zn-CaPs exhibited a distinct lamellar structure (Fig. 6B). With a decreasing anhydrous ethanol/deionized water volume ratio, the lamellar structure gradually diminished and increased in size; a further decline in this ratio transformed the lamellar structure into nanoparticle-assembled agglomerates, which grew with increasing water content. This is because alcohol molecules bind more strongly to crystal surfaces than water molecules; alcohol solvents thus drastically modify the structured water layer on crystal surfaces, replacing water to form a highly ordered alcohol molecular layer [61]. This alcohol molecular layer

Fig. 6.

The effects of the binary solvent system on the morphology, phase, and mass percentages of Zn²⁺ and Ca²⁺ in Zn-CaPs, as well as the Zn²⁺ doping rate. (A) SEM images: reacted; (B) SEM images: unreacted, enlarge area scale: 1 μm; (C) XRD patterns: reacted; (D) XRD patterns: unreacted; (E) AAS: reacted. (Zn²⁺/(Ca²⁺+Zn²⁺) = 10 mol%, the reaction temperature and time were 150 °C and 8 h, respectively.).

Modified the crystal surface charge and wettability, strongly inhibiting the deposition of Ca²⁺ and PO₄³⁻ on the mineral surface and retarding crystal growth and dissolution; thus, higher ethanol content promoted the formation of more lamellar structures [62]. With increasing water ratio, the order of the alcohol molecular layer on crystal step edges was disrupted, and water molecules gradually occupied these sites—promoting directional crystal growth along step edges and larger crystal sizes [63]. XRD analysis showed the predominant phase of as-prepared Zn-CaPs shifted from monetite to β-TCP with decreasing anhydrous ethanol/water ratio. Post-reaction, the crystalline phase of Zn-CaPs converted entirely from β-TCP to monetite as the major phase, with the scholzite secondary phase also decreasing steadily (Fig. 6C and D). In ethanol-free reaction systems, post-reaction samples converted from brushite to β-TCPs, as β-TCP is a high-temperature CaPs, while monetite and brushite are low-temperature CaPs [64]. Organic solvent addition to the system promotes the formation of metastable phases, attributed to hydrogen bonding between alcoholic hydroxyl groups and acid-group oxygen atoms, as well as electrostatic interactions between hydroxyl oxygen atoms and Ca²⁺—both of which delay crystalline phase precipitation [60,65]. Thus, post-reaction Zn-CaPs samples in the binary solvent were dominated by monetite as the major phase. FTIR analysis further confirmed the characteristic functional groups of monetite and scholzite (Fig. S25A and B). The dominant phase of binary solvent-synthesized Zn-CaPs products remained unchanged, indicating a dynamic equilibrium of Zn²⁺ doping efficiency (Fig. 6E).

In brief, the effects of key reaction parameters—including system concentration, ammonia dosage, reaction time/temperature, Zn doping amount, and binary solvent—on the microstructure, phase, and Zn²⁺ doping efficiency of Zn-CaPs materials are illustrated in Fig. 7A and B. Regulating target gradient parameters enables rapid screening of the conditions required to fabricate distinct Zn-CaPs morphologies and crystalline phases, accelerating the rational design of CaP materials.

Fig. 7.

The effects of process parameters on the morphology, phase, and Zn²⁺ doping rate of reaction products. (A) I: Morphology, II: Phase (the red line represents regulation and the black line represents non-regulation), and III: Zn²⁺ doping rate; (B) The synergistic effects of the parameters on the morphology, phase, and Zn²⁺ doping rate of the samples. The waveform fluctuation represents the change of Zn²⁺ doping rate, and the color alternation represents the change of morphology.

2.5. Biological performance of Zn-CaPs

10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 were selected to verify the biological functions of Zn-CaPs. The cytocompatibility of Zn-CaPs was assessed via co-culturing ion extracts with MC3T3-E1 cells. Fig. 8A shows that the green fluorescence intensity of 10Zn-M_0.5A_0.2 and 10Zn-M_0.5A_1.0 was higher than that of the control group, whereas that of 10Zn-M_0.5A_2.0 exhibits similarity to the control group. With increasing culture time, the MC3T3-E1 cell number increased in the 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 groups (Fig. 8B). Additionally, no obvious dead cells were observed in any group. These results demonstrate that 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 all exhibit excellent cytocompatibility, consistent with related studies indicating good cytocompatibility of Zn-CaPs at an appropriate Zn-doping concentration [66,67]. The superior cytocompatibility effect of 10Zn-M_0.5A_0.2 and 10Zn-M_0.5A_1.0 compared with 10Zn-M_0.5A_2.0 may be attributed to their higher release levels of Zn²⁺ and Ca²⁺ (Fig. S15). Within an appropriate concentration range, higher concentrations of Zn²⁺ and Ca²⁺ can promote cell proliferation and differentiation [68]. Cytocompatibility shows the same trend as the release quantity of Zn²⁺ and Ca²⁺.

Fig. 8.

In vitro cytocompatibility and osteogenic bioactivities of the Zn-CaPs extracts after culturing with BMSCs for 7 days. (A) The cell live/death staining of MC3T3-E1 after culturing with the sample extract for 5 days; (B) The cytocompatibility was evaluated by co-culture of the sample extract with MC3T3-E1; (C) RT-qPCR analysis of osteogenic-related genes; (D) Western blot analysis of osteogenic-related proteins; (E) Immunofluorescence staining of Runx2 and OCN; (F) Alkaline phosphatase staining. (All data were performed using one-way or two-way ANOVA for multiple comparisons and presented as mean ± standard deviation (SD), n = 3; the error bars represented the SD, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.).

To evaluate the in vitro osteogenic differentiation potential of Zn-CaPs, osteogenic-associated protein and gene expression were detected via RT-PCR and Western blot. RT-PCR results showed that osteogenic-associated gene levels in the 10Zn-M_0.5A_1.0 group were higher than those in other groups (except BMP-2). All experimental groups exceeded the control (except Runx2) (Fig. 8C). Western blot results showed osteogenic-associated protein expression in the 10Zn-M_0.5A_1.0 group was higher than in other groups (except BMP-2) (Fig. 8D). Consistent with related studies, Zn²⁺ exerts a robust pro-osteogenic differentiation effect [66,69]. Additionally, immunofluorescent staining of Runx2 and OCN (Fig. 8E) and ALP staining (Fig. 8F) yielded consistent results, demonstrating that Zn²⁺ enhances CaPs’ osteogenic differentiation-promoting capacity in BMSCs. Yuan et al. demonstrated that Zn²⁺ significantly upregulates Runx2 and OCN expression in BMSCs cultured with calcium silicate/CaPs cement [70]. However, Runx2 and OCN expression in 10Zn-M_0.5A_2.0 was lower than that of the control group, likely due to its higher Zn²⁺ release. As reported by Xiong et al., high Zn²⁺ concentrations inhibit the expression of osteogenesis-related proteins and impair the osteogenic activity of CaPs [68].

To assess the in vitro antibacterial activity of Zn-CaPs, their ion extracts were co-cultured with E. coli and S. aureus for 24 h (Fig. S26A and B). The results showed that OD values in the 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 groups were lower than the control, for both bacterial strains. This confirms that Zn²⁺ incorporation enhances the antibacterial activity of CaPs, consistent with previous studies [71,72]. However, OD values for 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 decreased progressively, indicating Zn²⁺ inhibits bacterial proliferation, with Ca²⁺ also exerting a regulatory effect on bacterial growth. Adawy et al. confirmed that CaPs promote E. coli proliferation [73]. Beyond Zn²⁺, 10Zn-M_0.5A_0.2 released substantially more Ca²⁺ than the other groups, which counteracted part of Zn²⁺‘s antibacterial activity. Thus, its antibacterial performance was weaker than that of 10Zn-M_0.5A_2.0 and 10Zn-M_0.5A_1.0. For further intuitive validation of Zn-CaPs antibacterial activity, ion extracts were mixed with bacterial suspension and spread on agar plates. After 24 h of incubation, the results (Fig. S26C) showed fewer E. coli and S. aureus colonies in the 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 groups than in the control, consistent with the in vitro co-culture data. Biological function assessments confirmed that Zn-CaPs exhibit favorable cytocompatibility, osteogenic differentiation potential, and antibacterial activity in vitro.

2.6. Construction and mechanical properties of 3D printed Zn-CaPs scaffold

5Zn-M_0.5A_1.0, 10Zn-M_0.5A_1.0, and 20Zn-M_0.5A_1.0 were fabricated into porous scaffolds (5Zn-β-TCPS, 10Zn-β-TCPS, 20Zn-β-TCPS) via 3D printing. The scaffolds appeared yellow pre-sintering and turned white post-sintering, while retaining a cylindrical shape both before and after sintering (Fig. 9A, S27A). After sintering, all scaffolds maintained a porous structure with pore diameters of approximately 400–500 μm (Fig. 9A–S27B), which aligns with Karageorgiou et al.‘s finding that scaffold pores >300 μm benefit capillary growth and new bone formation [75]. XRD results showed that the monetite and scholzite phases in the scaffolds disappeared post-sintering, with β-TCP as the main phase and a small amount of calcium pyrophosphate (Ca₂P₂O₇) formed (Fig. 9B), consistent with Tas et al.‘s finding that monetite can transform into Ca₂P₂O₇ upon calcination [76]. In our study, a small amount of zinc pyrophosphate (Zn₂P₂O₇) was detected in 10Zn-β-TCPS and 20Zn-β-TCPS, with the latter having a higher Zn₂P₂O₇ content—this phase forms from high-temperature sintering of scholzite. Both Ca₂P₂O₇ and Zn₂P₂O₇ exhibit good cytocompatibility [77,78]. XRD patterns of sintered 10Zn-M_0.5A_0.2, 10Zn-M_0.5A_1.0, and 10Zn-M_0.5A_2.0 confirm the conversion of CaHPO₄ to β-TCP, and 10Zn-M_0.5A_1.0 and 10Zn-β-TCPS possess identical phase compositions after sintering (Fig. S28). FTIR analysis of the scaffolds confirmed characteristic functional groups of β-TCP at corresponding wavenumbers, consistent with the above findings, and an absorption peak at 727 cm⁻¹ assigned to pyrophosphate (Fig. S27C) [79].

Fig. 9.

Characterization and efficacy of the Zn-CaPs scaffolds for femoral defect repair. (A) Macroscopic pictures and SEM images of scaffolds post-sintering. Scale bar: 2 mm; (B) XRD patterns; (C) Compression strength; (D) A comparison of compressive strength and porosity between Zn-β-TCPS and other CaPs bioceramic scaffolds was conducted in Ashby diagram [74]; (E) Comprehensive performance evaluation for 10Zn-β-TCPS and other CaPs bioceramic scaffolds in a radar map [74]; (F) A schematic illustration of the procedure for creating a femur bone defect in rabbits with or without implantation of the scaffolds; (G) Micro-CT reconstruction images of the defects treated with PLAS, β-TCPS, and 10Zn-β-TCPS at 6 and 12 weeks; (H) Statistical analysis of bone volume per total volume (BV/TV), bone mineral density (BMD), and trabecular number (Tb. N) of repaired tissue in different groups. Data are shown as means ± SD. (n = 3, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, NS: no significant difference.)

EDS results showed uniform distribution of Ca, P, O, and Zn on the surface of 10Zn-β-TCPS (Fig. S27E). Mechanical tests revealed that 10Zn-β-TCPS exhibited the highest compressive strength; as shown in Fig. 9C, β-TCPS had comparable compressive strength to 5Zn-β-TCPS but significantly higher than 20Zn-β-TCPS. This difference was attributed to the broadest grain size distribution of β-TCPS, followed by 5Zn-β-TCPS and 20Zn-β-TCPS. In contrast, 10Zn-β-TCPS exhibited the most uniform particle size distribution. The lowest compressive strength of 20Zn-β-TCPS was attributed to phase changes induced by its high Zn²⁺ doping content (Fig. 4E–S24B). Pre-sintering SEM images of 5Zn-β-TCPS and 10Zn-β-TCPS further confirmed the relatively uniform particle distribution of 10Zn-β-TCPS (Fig. S27D). Our prior work demonstrated that ceramic scaffolds with uniform particle distribution exhibit higher compressive strength, as this morphology effectively mitigates excessive stress concentration [80]. Moreover, the porosities of β-TCPS, 5Zn-β-TCPS, 10Zn-β-TCPS, and 20Zn-β-TCPS all exceeded 80% (Fig. S27B). Wei et al. demonstrated that BCP scaffolds with 84% porosity facilitate vascular ingrowth and effectively repair large femoral defects in goats [81]. Notably, the compressive strength of Zn-β-TCPS exceeded 0.5 MPa, meeting the mechanical strength requirements for load-bearing bone repair scaffolds [82]. Scaffold degradation rate is vital to its in vivo mechanical stability and bone regeneration capacity; in vitro results show β-TCPS degrades faster with higher Ca²⁺ release (Fig. S29). High-concentration Zn²⁺ doping (≥2 mol% Zn) induces lattice distortion and forms a phosphorus-rich surface passivation layer, lowering material solubility and Zn²⁺ release [83]. 10Zn-β-TCPS was fabricated from 10Zn-M_0.5A_1.0, which exhibits excellent cytocompatibility, osteogenic differentiation, and antibacterial activity. To evaluate the osteogenic performance of β-TCPS, 5Zn-β-TCPS, 10Zn-β-TCPS, and 20Zn-β-TCPS, ALP staining of BMSCs seeded on the scaffolds was conducted (Fig. S30). At 7 and 14 days, BMSCs on 5Zn-β-TCPS, 10Zn-β-TCPS, and 20Zn-β-TCPS showed higher ALP activity than those on β-TCPS, with 10Zn-β-TCPS yielding the highest activity. ALP staining results confirmed that an optimal Zn²⁺ concentration enhances the osteogenic activity of β-TCPS [68]. Finally, the compressive strength and porosity of 10Zn-β-TCPS were compared with those of other CaP scaffolds (Fig. 9D) [84]. Notably, conventional-structured 10Zn-β-TCPS (vs. TPMS structure) achieved a high porosity of 86% while retaining a relatively high compressive strength of 2.68 MPa. A radar chart was further used to systematically summarize the mechanical and biological properties of 10Zn-β-TCPS and other CaPs scaffolds (β-TCP, HA, BCP; Fig. 9E) [74]. 10Zn-β-TCPS outperformed other CaPs scaffolds in compressive strength, porosity, cytocompatibility, osteogenic activity, and processability, and its compressive strength meets the mechanical requirements for load-bearing bone repair materials. Collectively, the 10Zn-β-TCPS developed in this study exhibits superior mechanical and biological properties, making it a promising candidate material for the clinical repair of large load-bearing bone defects.

2.7. In vivo osteogenesis of the Zn-β-TCPS

A rabbit femoral defect model was established to evaluate the performance of commercial polylactic acid scaffolds (PLAS), β-TCPS, and 10Zn-β-TCPS. Bone repair efficacy was assessed at 6 and 12 weeks post-implantation via micro-CT, H&E, Masson, and immunohistochemical (IHC) staining (Fig. 9F). Coronal plane images showed more new bone formation in the 10Zn-β-TCPS group than in the control, PLAS, and β-TCPS groups at 6 weeks. Abundant new bone formed in the 10Zn-β-TCPS group at 12 weeks, with the β-TCPS group exhibiting inferior bone formation (Fig. 9G). In contrast, only a small amount of new bone formed in the control and PLAS groups. Micro-CT 3D reconstructions revealed a gradual increase in new bone formation across the control, PLAS, β-TCPS, and 10Zn-β-TCPS groups at 6 weeks, with the same trend observed at 12 weeks. Micro-CT quantitative analysis showed that the 10Zn-β-TCPS group exhibited the highest bone volume/total volume (BV/TV), bone mineral density (BMD), and trabecular number (Tb.N) of the four groups at 6 and 12 weeks (Fig. 9H). Notably, BV/TV and Tb. N in the 10Zn-β-TCPS group differed significantly from the β-TCPS group at both 6 and 12 weeks, confirming that Zn²⁺ exerts a positive effect on bone formation [68,84]. Furthermore, BV/TV, BMD, and Tb. N in the 10Zn-β-TCPS group also displayed significant differences compared to the control group at 6 and 12 weeks. BV/TV, BMD, and Tb. N in the control group, PLAS group, β-TCPS group, and 10Zn-β-TCPS group increased successively, whether at 6 or 12 weeks.

Histological staining further confirmed that 10Zn-β-TCPS improved the integrated repair of femoral defects relative to the PLAS and control groups. H&E (Fig. 10A) and Masson's trichrome staining (Fig. 10B) showed significantly more new bone formation in the β-TCPS group than the PLAS group at 6 weeks; the 10Zn-β-TCPS group had the highest new bone formation, while new bone formation in the control group was only minimal. At 12 weeks, all groups exhibited more new bone formation than at 6 weeks, with levels increasing in the order of the control, PLAS, β-TCPS, and 10Zn-β-TCPS groups. Additionally, bone maturity was higher in the 10Zn-β-TCPS and β-TCPS groups than in the control and PLAS groups, with the 10Zn-β-TCPS group showing the highest new bone maturity (Fig. 10B). Histological staining results confirmed that the 10Zn-β-TCPS group had the optimal osteogenic performance, and Zn²⁺ can simultaneously induce new bone formation and maturation [77,85]. IHC staining detected the expression of bone-specific proteins (OCN, Runx2) in the defect area (Fig. 10C and D). OCN and Runx2 expression levels were higher in the β-TCPS group than in the PLAS group, with the 10Zn-β-TCPS group showing the highest levels; expression was extremely low in the control group. This confirms that Zn²⁺ enhances the osteogenic differentiation of β-TCPS. From 6 to 12 weeks, OCN and Runx2 expression increased across all groups, with the 10Zn-β-TCPS group showing the highest levels. Quantitative analysis of magnified images quantified the H&E, Masson, OCN, and Runx2 stained areas (Fig. 10E–S31). The 10Zn-β-TCPS group differed significantly from the β-TCPS group, which in turn was markedly higher than the control group. Histological and IHC staining results confirmed that 10Zn-β-TCPS promotes new bone formation and maturation by inducing the expression of bone-specific proteins.

Fig. 10.

Histological analysis of the newly regenerated tissues in the defect area with the scaffold implantation. (A) H&E staining images; (B) Masson staining images; (C) OCN staining images; (D) Runx2 staining images; (E) Quantitative analysis of the percentage of H&E stained area, Masson stained area, and OCN stained area to the magnified images area. “S” represents scaffold, and “N” represents new bone. Scale bars, 2500 μm; scale bars of magnified images, 500 μm.

Material research and development commences with a methodology for material design. Materials genomics represents a research mode for materials development, for which reliable methods for efficient materials construction are essential [6]. Advanced material synthesis and screening technology are crucial for material design as they can speed up the design process, shorten the research and development time, and identify the best conditions and parameters [3,4]. This study achieved high-throughput fabrication of Zn-CaPs via a self-built high-throughput synthesis system combined with the hydrothermal method, with this strategy exhibiting far higher preparation efficiency than conventional ovens or multi-station reactors (Table S15). Furthermore, the samples synthesized via this apparatus possess high reproducibility (Fig. S32). It enables simultaneous investigation of the effects of multiple process parameters on the physicochemical properties of target materials; moreover, the impacts of these parameters on Zn-CaP crystallization were elucidated based on supersaturation theory. Zn²⁺ doping significantly enhanced the biological functions of Zn-CaPs, and 10Zn-β-TCPS possessed markedly higher mechanical strength than pure β-TCPS. Zn-CaPs thus exhibit great potential for biomedical applications. Our findings confirm that high-throughput technology markedly boosts material preparation efficiency, facilitates the construction of a material gene database, and provides a data reference for subsequent material fabrication. However, this work has limitations: only the effects of a single trace element (Zn) on the physicochemical properties and biological functions of CaPs were investigated, and the research nodes for reaction time, temperature, and binary solvent conditions were insufficient; the specific crystallization mechanism regulated by reaction parameters (including Zn²⁺) remains unclear; the link between high-throughput material synthesis and optimal biological performance lacks sufficient experimental support. Future work will explore the synergistic effects of multiple ions and the enhancement of CaP physicochemical and biological properties via multi-trace element co-doping. We will also identify the optimal doping contents of multi-element modified CaPs for high-efficiency biological functions (drug delivery, antibacterial activity, antitumor efficacy, and osteogenesis), integrate XRD/Raman analysis with computational simulations to further elucidate the underlying crystallization mechanisms, which will enrich the material database and accelerate the development of biofunctional materials. However, the challenges in extending high-throughput hydrothermal synthesis to multi-element doped CaPs mainly involve lattice incompatibility and phase disorder, parametric dimensional explosion, inhomogeneous reaction environment, ambiguous structure-performance relationships, and restricted high-throughput screening and scale-up application. We conclude that integrating high-throughput technology with material fabrication offers broad prospects for the future R&D of biomaterials.

3. Conclusions

In this study, a high-throughput synthesis equipment was adopted to evaluate the effects of system concentration, ammonia concentration, reaction duration, temperature, zinc concentration, and binary solvent on the morphology, phase, chemical structure, and zinc doping rate of Zn-CaPs produced via hydrothermal methods. The synthesized Zn-CaPs demonstrated excellent cytocompatibility, osteogenic differentiation, and antibacterial properties in vitro. Furthermore, the 3D-printed scaffolds (10Zn-β-TCPS) exhibited enhanced mechanical strength and superior femur defect repair compared to conventional CaPs scaffolds (Zinc-free CaPs), highlighting their potential for load-bearing bone regeneration. This study establishes a versatile and efficient platform for the synthesis of high-performance ion-doped CaPs with tailored morphology and structure. Future integration of AI-driven material design with high-throughput techniques is expected to enable prediction of material composition, structure, and properties, thereby advancing controllable synthesis and the rapid development of novel materials.

Ethics approval and consent to participate

In this study, all animal experiments were conducted in accordance with the guidelines of the Chinese Committee for Animal Protection and were approved by the Medical Ethics Committee of Southwest Jiaotong University (SWJTU-2403-NSFC (053)).

CRediT authorship contribution statement

Hao Huang: Investigation, Methodology, Writing – original draft. Bingjun Zhang: Funding acquisition, Writing – review & editing. Shaojie Zhu: Investigation. Dongbiao Chang: Methodology. Liwen Yang: Methodology. Jinsheng Li: Investigation. Lei Qiang: Investigation. Bobo Jiang: Investigation. Ruiquan Li: Writing – review & editing. Xiong Lu: Methodology. Tailin Guo: Supervision. Xin Chen: Funding acquisition, Supervision, Writing – review & editing. Pengfei Zheng: Conceptualization, Supervision, Writing – review & editing. Jie Weng: Funding acquisition, Methodology, Supervision, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no conflict of interest.

Appendix A. Supplementary data

The following are the Supplementary data to this article: