Effect of calcination temperature on the bioactivity of a sol-gel-derived 40CaO-60B₂O₃ glass composition

Abstract

Sol-gel-derived borate glasses are highly bioactive where they demonstrate rapid production of hydroxycarbonated apatite (HCA), in vitro; making them promising candidates for biomedical applications. Calcination, a critical step in sol-gel processing, significantly influences product properties. This study investigated the effect of calcination temperature on the texture, structure, as well as the reactivity and bioactive properties of sol-gel-derived 40CaO-60B₂O₃ (mol%) samples synthesized from calcium lactate pentahydrate. X-ray diffraction (XRD) confirmed that samples calcined between 400 and 600 °C were amorphous, whereas crystallization occurred when samples were calcined at 700 °C. Attenuated total reflectance-Fourier transform infrared (ATR-FTIR) spectroscopy revealed a decrease in BO₄ units with increasing calcination temperature. There was a decrease in specific surface area, attributable to reduced network connectivity and densification, which correlated with a decrease in vapor reactivity, as indicated by dynamic vapor sorption analysis. Bioactivity was confirmed by HCA formation within 7 days in simulated body fluid in all calcined samples, as characterized by ATR-FTIR, XRD, and scanning electron microscopy. In particular, samples calcined at 400 and 500 °C exhibited HCA formation within 1 day. Thermogravimetric analysis revealed that the sample calcined at 400 °C contained the highest amount of organic residues from sol-gel processing. In contrast, the sample calcined at 500 °C combined high bioactivity with optimal thermal decomposition, indicating that this calcination temperature may be suitable for this glass composition. In summary, the successful use of calcium lactate pentahydrate as a low-cost precursor highlights its scalability and potential for producing high-performance sol-gel-derived bioactive borate glasses.

Highlights

• Effect of calcination temperature on sol-gel-derived calcium borate samples was investigated.

• Sol-gel-derived 40CaO-60B₂O₃ (mol%) samples were synthesized from calcium lactate pentahydrate.

• Calcination temperature significantly impacted the textural and structural properties.

• Specific surface area showed correlation with vapor reactivity.

• Samples calcined at 400 and 500 °C exhibited superior bioactivity in simulated body fluid.

Introduction

Bioactive glasses are extensively studied in simulated body fluid (SBF) to test their ability in forming hydroxycarbonated apatite (HCA), a key indicator of acellular bioactivity, in vitro. This property ultimately enables bioactive glasses to bond both physically and chemically with mineralized tissues, in vivo, making them valuable for bone regeneration and in the treatment of dentine hypersensitivity. The first commercial bioactive glass, Bioglass^® 45S5 (46.1SiO₂-26.9CaO-24.4Na₂O-2.6P₂O₅ mol%), was pioneered by Dr. Larry Hench in 1969 [1]. Recently, bioactive borate-based glasses have gained attention due to their rapid conversion rate to HCA, which can be attributed to their inherently lower chemical durability [2–7].

Glass products, such as window glass and smartphone cover glass, are traditionally fabricated using melt-quenching, where precursor oxide powders are melted at high temperatures (>1200 °C) followed by rapid cooling to prevent crystallization and maintain an amorphous structure. In contrast, sol-gel offers numerous advantages, including high purity, homogeneity and low temperature processing [8]. For biomedical applications, sol-gel-derived glasses provide additional benefits, such as increased surface area and porosity, which enhance degradation and ion release rates, thereby accelerating HCA formation. For example, the sol-gel-derived 40CaO-60B₂O₃ (mol%) glass has been shown to form HCA in SBF within 2 h [9, 10]. Moreover, the synergy between borate glass composition and the sol-gel method offers promising potential for wider medical applications [11]. For example, hemostatic, antibacterial and pro-angiogenic activities of a series of sol-gel derived binary CaO-B₂O₃ bioactive glasses (CaO ranging from 50 to 90 mol%) were investigated by Zheng et al. [12]. It was reported that the biological properties of the sol-gel derived CaO-B₂O₃ bioactive glasses are composition-dependent and dose-dependent. These previous studies [11, 12] on sol-gel-derived binary CaO-B₂O₃ bioactive glasses have primarily focused on the composition dependence of their bioactivity and biological properties.

The sol-gel process consists of four key steps: sol formation, gelation, drying, and calcination. Calcination, as the final synthesis step, is particularly influential, as it can amplify or mitigate the effects of the earlier stages. Calcination significantly influences the morphology, phase composition and chemical composition of the glasses [8]. During the calcination step, many phenomena such as polycondensation, structural reorganization and sintering can take place [8]. For example, calcination can lead to a collapsing of the porous structure, in which the temperature range can be critical in impacting various ranges of pore sizes, i.e., between 400 and 500 °C affect smaller pores, whereas between 700 and 900 °C affect larger pores. Additionally, absorbed species are desorbed or decomposed between 100 and 500 °C, while organic-based solvent residuals and reaction products can break down between 300 and 500 °C [8]. To date, the effects of calcination temperature on sol-gel-derived silicate-, phosphate- and borate-based glasses have been investigated. For example, Siqueira et al. [13] examined the sol-gel derived 46.1SiO₂-26.9CaO-24.4Na₂O-2.6P₂O₅ (mol%) glass system calcined at 700, 800, 900 and 1000 °C. They reported that all calcined samples were crystalline, where different crystal phases, such as Na₂Ca₂Si₃O₉ and Na₂Ca₃Si₆O₁₆, were formed depending on the calcination temperature. At lower calcination temperature, a higher fraction of Na₂Ca₂Si₃O₉ was obtained, which exhibits higher bioactivity than Na₂Ca₃Si₆O₁₆. Consequently, the bioactivity decreased with increasing calcination temperature. In the case of phosphate-based glasses, Carta et al. [14] evaluated the sol-gel derived 40P₂O₅-xB₂O₃-(60-x)Na₂O (x = 10, 15, 20 and 25 mol%) glass system. According to the results, changes in crystallinity and in the structural connectivity of the glass network with increasing calcination temperature were observed. They concluded that these structural changes lead to variations in the dissolution rate in SBF. In addition, in terms of borate glasses, the sol-gel-derived 46.1B₂O₃-26.9CaO-24.4Na₂O-2.6P₂O₅ (mol%) glass system [15] was investigated by Lepry et al. and demonstrated that crystallization occurred with increasing calcination temperature, which caused a reduction in the specific surface area (SSA). Furthermore, it was stated that this decrease in SSA affects the dissolution rate in SBF, ultimately influencing the bioactivity. Nevertheless, despite the advantages of the sol-gel method, it also faces challenges, including high precursor costs, residual pores, and extended processing times [16]. For example, in order to synthesize 40CaO-60B₂O₃ (mol%) glass [9], a previous study utilized calcium methoxyethoxide, which is a relatively expensive precursor, whereas a follow-up study demonstrated that calcium lactate pentahydrate may be a more cost-effective alternative precursor, yielding glasses with comparable HCA conversion times and the potential for scalability [17]. However, these glasses lack optimization for calcination conditions. Therefore, this study aimed at investigating the influence of calcination temperature on 40CaO-60B₂O₃ (mol%) glasses generated from calcium lactate pentahydrate and to propose an optimal calcination temperature for enhanced bioactivity and scalability.

Experimental section

Sample preparation

Sol-gel-derived 40CaO-60B₂O₃ (mol%) calcined samples were fabricated as previously reported [17]. In order to generate a batch size of 10 g, 11.6 g of boric acid (≥99%, Sigma Aldrich, Canada) and 58 g of methanol (≥99%, Sigma Aldrich, Canada) were initially mixed under magnetic stirring in a covered glass beaker for ~1 h until the solution was clear. Then 19.4 g of calcium lactate pentahydrate (≥98%, Fisher Scientific, Canada) was added and mixed for 1 h before the sol was poured as a thin layer into a crystallization dish and covered with a porous parafilm and stored for 7 days for gelation and ageing. The gel was then taken through a step-wise drying process, which consisted of 60 °C for 1 day, followed by 120 °C for further 2 days, after which the non-calcined samples were collected. These samples were then calcined in covered porcelain crucibles at either 400, 500, 600 or 700 °C. All calcination temperatures were reached using a heating ramp rate of 0.5 °C/min, with a 2 h dwell time, and followed by furnace cooling (Table 1). All calcined and non-calcined samples were ground and sieved to a particle size range of 25 to 75 μm and stored in a desiccator for characterization.

Table 1.

Glass composition and calcination temperatures investigated in this study

ID	Glass composition (mol%)	Firing rate (°C/min)	Calcination temperature (°C)	Dwell time (h)
LAC400	40CaO-60B2O3	0.5	400	2
LAC500			500
LAC600			600
LAC700			700

Particle characterization

Calcined powder particle size (D₅₀) was determined in 2-propanol suspension using a Microtrac Flow Sync (Microtrac Inc., USA). Powder specific surface areas (SSA) was also measured with nitrogen gas adsorption and desorption isotherms using a Micromeritics TriStar II Plus (Micromeritics Instrument Corporation, USA) gas sorption system. Samples were degassed at 120 °C overnight under nitrogen gas to remove contaminants before measurement of SSA values as determined from collected isotherms using the Brunauer–Emmett–Teller (BET) method.

Thermogravimetric analysis (TGA)

TGA was carried out on both non-calcined and calcined powders using a Simultaneous Thermal Analyzer STA 449 F3 Jupiter (NETZSCH, Germany). Measurements were performed under flowing argon purge (20 mL/min) on 30 mg specimens in Pt/Rh crucibles using a temperature range of 25 and 900 °C and a heating rate of 10 °C/min. The output was used to determine the weight loss due to the synthesis residues and hydroxyl functional groups derived from the sol-gel process.

X-ray diffraction (XRD)

XRD of calcined samples before and after dynamic vapor sorption analysis and immersion in SBF was carried out using a Bruker D2 Phaser X-ray Diffractometer (Bruker Corporation, USA) equipped with Cu Kα target set to a power level of 30 kV and 10 mA. Diffractograms were collected from 10 to 80° 2θ angles with an approximate step size of 0.055°, and step time of 0.5 s. Phase identification was carried out using a DIFFRAC. EVA (Bruker Corporation, USA).

Attenuated total reflectance-Fourier transform infrared (ATR-FTIR) spectroscopy

ATR-FTIR spectroscopy was carried out on calcined samples before and after vapor sorption and desorption analysis and immersion in SBF. Spectra were collected using a Nicolet™ iS50 FTIR Spectrometer (Thermo Fisher Scientific, USA) between 4000 and 400 cm⁻¹ with a resolution of 2 cm⁻¹ using 64 scans per sample. All collected spectra were baseline corrected. For all calcined samples, the resultant maximum intensities of peaks between 400 and 4000 cm⁻¹ were normalized to 1 using OMNIC Spectra software (Thermo Fisher Scientific, USA).

Scanning electron microscopy (SEM) and electron-dispersive X-ray spectroscopy (EDS)

SEM imaging and EDS analysis of the calcined particles, as-prepared and post immersion in SBF, were carried out on Pt sputter coated samples (using a Leica Microsystems EM ACE600 sputter coater). SEM imaging was conducted by using an FlexSEM 1000II (Hitachi High-Tech Canada, Inc., Canada) at 10.0 kV. SEM images were captured at magnifications of 500× and 15k×. To determine HCA conversion, EDS using an EDAX and a TEAM EDS Analysis System was performed at 20 kV on sample surface areas to determine the Ca/P ratio at the 7 day-time point.

Dynamic vapor sorption (DVS)

DVS was used to characterize the vapour sorption and desorption of the calcined particles under controlled relative humidity and temperature. A DVS Resolution Instrument (Surface Measurements Systems Ltd., UK) was used to measure sample mass changes ( ± 0.1 μg) at 37 ± 0.1 °C with nitrogen (N₂) flow as purge (200 sccm). Samples of 10 mg were placed in an aluminium pan and first equilibrated at 0% RH to determine their dry weight, before immediately being exposed to 90% RH for 24 h followed by 0% RH for a further 24 h. After each measurement, the samples were collected for further structural analyses (XRD and ATR-FTIR as described above).

Bioactivity in SBF

The ability of the calcined samples to convert to HCA was tested using Kokubo’s SBF at pH 7.4 [18]. Sterile 50 mL falcon tubes were used to mix calcined particles and SBF at a ratio of 1.5 mg/mL and stored at 37 ± 1 °C in an incubator shaker (120 rpm; KS 4000 i control, IKA, USA). The powders were removed from the SBF solution at 6 h, and 1 and 7 day-time points, where they were gently rinsed with deionized water then twice with anhydrous ethanol, dried overnight at room temperature, and then dried in an oven at 60 °C for 1 day.

Results and discussion

Characterization of calcined samples

TGA curves provide insights into the mass changes of the non-calcined and calcined samples as a function of temperature, thereby determining weight loss due to the removal of synthesis residues and hydroxyl functional groups derived from the sol-gel process. Figure 1a shows the TGA curves of non-calcined samples and those calcined at 400, 500, 600 and 700 °C, and Table 2 summarizes the total weight loss in the samples when heated from room temperature to 900 °C. The TGA curves presented an initial weight loss at around 120 °C, which is associated with the removal of physically bound water. This was followed by a second main region of weight loss in the range of 350–550 °C, which was due to the decomposition of organic residues. The non-calcined sample, which can be assumed to contain the highest amount of the synthesis residues and hydroxyl functional groups, demonstrated the largest weight loss, at 51.5%. Furthermore, among the calcined samples, the weight loss decreased as the calcination temperature increased, with those calcined at 500 and 600 °C exhibiting a similar extent of weight loss. On the other hand, LAC400 and LAC700 contained the highest and lowest amounts of organic residues, respectively, as determined by their extent of weight loss. From these results, it can be expected that lowering the calcination temperature below 400 °C is not preferable, as it leads to a much higher amount of organic residues in the products compared with the other calcined samples in this study.

Fig. 1.

Thermal analysis and structural characteristics of non-calcined samples and those calcined at 400, 500, 600 and 700 °C. a TGA, b XRD, c ATR-FTIR, and d expansion of the 1800–400 cm⁻¹ region

Table 2.

Summary of total weight change of calcined samples after TGA, fraction of boron tetrahedra (N₄) as estimated by FTIR spectroscopy, calcined particle median diameter (D₅₀), and SSA

ID	Total weight change (%)	N4	D50 (μm)	SSA (m2/g)
Non-calcined	−51.5	−	−	−
LAC400	−24.6	0.55	51.6	52.3
LAC500	−14.9	0.54	42.5	43.5
LAC600	−14.2	0.54	60.4	39.9
LAC700	−0.8	0.44	49.6	2.0

XRD and ATR-FTIR were carried out to morphologically and structurally analyze the samples calcined at the various temperatures. XRD diffractograms suggested that the particles calcined at 400, 500 and 600 °C were amorphous, as they displayed several broad humps (Fig. 1b). In contrast, LAC700 displayed strong sharp peaks, which was identified as a calcium metaborate CaB₂O₄ crystal phase (COD 1010468) [19, 20]. Figure 1c shows the ATR-FTIR spectra of the calcined samples, with the wavenumber region covering the absorption bands at lower wavenumbers expanded in Fig. 1d. All calcined samples indicated hydroxyl-related signals in their spectra at ~3400 and 1630 cm⁻¹. The spectra of LAC400, LAC500 and LAC600 indicated three main regions: the stretching vibrations of B-Ø and B–O^- bonds in borate triangle containing groups, such as BØ₃ at 1150–1500 cm⁻¹, the stretching vibrations of B-Ø in BØ^4- tetrahedra at 800–1150 cm⁻¹, and B-Ø bending in borate groups at 600–750 cm⁻¹ (Ø and O denote bridging and non-bridging oxygen atoms, respectively) [17, 21–24]. LAC700 also showed the three main regions at 630–750 cm⁻¹, 1000–1350 cm⁻¹ and 1350–1550 cm⁻¹. The bands between 630–750 cm⁻¹ can be attributed to the borate chain deformation, the 1000–1350 cm⁻¹ band is the stretching vibration of B-Ø, and the bands in the 1350–1550 cm⁻¹ region are the stretching vibration of B-O^- [25, 26]. From Fig. 1d, it can also be observed that the absorbance of the peak related BO₄ in the spectrum of LAC700 was smallest among all calcined samples. In addition, LAC700 exhibited the highest peak intensity of 630–750 cm⁻¹ band (borate chain deformation), indicating a transformation toward a less three-dimensional network structure. Simultaneously, the peak intensity of B-O^- (1350–1550 cm⁻¹) was also the highest, suggesting that the bonds within BO₄ units were broken. Interestingly, the peak widths of LAC700 were narrower than those of the other samples, indicating that LAC700 possesses a less amorphous structure compared to the others. Indeed, FTIR may be used to investigate the presence of different structural groups, thus providing insights into possible changes in their rearrangement with calcination temperature. Here, the fraction of four-coordinated boron atoms (N₄) was calculated using Gaussian functions [27–31] and Eq. (1) below to assess the effect of calcination temperature on the structural network:

$$N_4=\frac{A_4}{A_4+A_3}$$ 1

where A₄ and A₃ are the peak areas of BO₄ units (800–1150 cm⁻¹) and BO₃ units (1150–1500 cm⁻¹) in the FTIR spectra (Fig. 1d), respectively. Table 2 presents the N₄ fraction as a function of calcination temperature, along with SSA and average median diameter (D₅₀) of the calcined samples. The estimated N₄ values for LAC400, LAC500 and LAC600 were of similar range and consistent with that previously reported [9] on the same sol-gel-derived glass composition, which utilized calcium methoxyethoxide as a precursor. Moreover, the N₄ value in LAC700 was lowest, which is also in agreement with a previous finding [31]. Based on these values, it can be concluded that the network connectivity of the calcined samples decreased with increasing calcination temperature due to a reduction in BO₄ units. Additionally, Fig. 1d shows an increase in the relative intensity of bands at around 700 and 1220 cm⁻¹, attributable to the B-O-B bending vibrations in symmetric BO₃ triangles [20] and B-O stretching vibrations of BO₃ units [19], respectively. These changes indicate an increase in BO₃ units, which supports the trend observed in the estimated N₄ values.

Table 2 also shows a decrease in SSA with increasing calcination temperature, consistent with findings from a previous study on sol-gel-derived borate glasses [15], where SSA was found to be influenced not only by N₄, but also by increased densification at higher calcination temperatures. Similarly, in this study, the decrease in SSA with increasing calcination temperature can be attributed to both reduced N₄ and increased densification.

Reactivity under vapor

DVS was used to investigate the reactivity of the calcined powders through their immediate aqueous interactions [17, 21]. Their sorption and desorption of water vapor was gravimetrically measured by exposure to 90% RH for 24 h followed by exposure to 0% RH for a further 24 h, respectively (Fig. 2a). During the sorption step, LAC400 exhibited the most rapid weight increase within the first 2 h, followed by LAC500 and LAC600, which showed a more gradual weight gain. In contrast, LAC700 demonstrated almost no weight change throughout the test. Notably, aside from LAC700, other calcined samples did not reach vapor sorption saturation, as their weight continued to increase up to 24 h. Table 3 summarizes the mass changes after 24 h sorption and desorption. The data from Tables 2 and 3 suggest a strong correlation between SSA and the amount of vapor sorption, with the permanent vapor sorption confirming that the calcined samples react with water vapor. Typically, higher N₄ values are associated with lower reactivity under vapor [17, 21, 31]. Despite the reduction in N₄ with increasing calcination temperature, the permanent vapor sorption also decreased. This counterintuitive trend can be explained by the reduction in SSA, which limits the surface area available for water molecules to react with the calcined samples. Thus, the influence of decreased SSA outweighs the impact of reduced network connectivity on vapor reactivity.

Fig. 2.

Calcined sample-reactivity with vapor and structural changes post exposure to vapor. a DVS measurements of vapor sorption and desorption when calcined samples were immediately exposed to 90% RH for 24 h followed by 0% RH for another 24 h. b ATR-FTIR spectra, and c XRD diffractograms of each calcined sample before and after exposure to vapor

Table 3.

Vapor sorption (90% RH) and desorption (0% RH) mass change of calcined samples

ID	Vapor sorption mass change
	After 24 h sorption [%]	After 24 h desorption [%]
LAC400	13.9	5.8
LAC500	7.5	3.9
LAC600	6.7	3.2
LAC700	0.5	0.1

Figure 2b, c presents the ATR-FTIR spectra and XRD diffractograms of the calcined samples before and after exposure to vapor, respectively. ATR-FTIR spectroscopy (Fig. 2b) revealed an increase in the relative intensity of hydroxyl groups at 3400 cm⁻¹ and a decrease in the relative intensities of the BO₃ bands (1220 cm⁻¹ and 1378 cm⁻¹) after vapor sorption. Additionally, XRD analysis indicated that LAC400, LAC500, and LAC600 maintained their amorphous nature post-exposure to vapor.

Bioactivity in SBF

The bioactivity of calcined samples, as assessed by their ability to convert a biologically relevant HCA, was analyzed over 7 days of immersion in SBF. HCA formation was characterized using ATR-FTIR, XRD, SEM and EDS. Figure 3 shows the ATR-FTIR spectra of all calcined samples, which suggested the initiation of HCA formation within 6 h in SBF, as indicated by the appearance of a strong band at ~1020 cm⁻¹ and its shoulder region at ~960 cm⁻¹, which are characteristic of the bending mode ν1 of PO₄^3- [9, 32]. Peaks at ~1470 and ~1421 cm⁻¹ represent the stretching modes ν1 and ν3 of CO₃²⁻ respectively, and a sharp peak at around 870 cm⁻¹ indicates the bending mode ν2 of CO₃²⁻ [33, 34]. This latter band may also be a combination of the B–O stretching of boroxol rings found in the calcined glasses as well as the bending mode ν2 of CO₃²⁻. The ν2 bending mode of water can also be seen by the band at ~1640 cm⁻¹ [35, 36]. In particular, in LAC400 and LAC500, the peaks related to phosphate and carbonate peaks appeared at earlier times, and at longer immersion times in SBF, these peaks became more defined among all calcined samples, indicating the maturation of HCA. In addition, the strong peaks attributed to HCA formation were also observed in the ATR-FTIR spectrum of the crystalline LAC700 sample at 7 days. In contrast to these findings, an equivalent melt-quenched-derived 40CaO-60B₂O₃ (mol%) glass [37] demonstrated HCA formation after 15 days of SBF immersion. This indicates that the HCA conversion rates of all samples calcined at the various temperatures in this study are more rapid than that of melt-quenched glass. Nevertheless, a direct comparison is difficult, as information on the SBF immersion test conditions, such as particle size and the sample-to-SBF ratio, was not provided in that previous work [37].

Fig. 3.

ATR-FTIR spectra of calcined samples, as-prepared, and after 6 h, 1 day and 7 days in SBF

Mineralization was confirmed by XRD diffractograms (Fig. 4), which by 7 days, indicated strong HCA peaks [38] at ~ 25 and ~ 32° 2θ (“■”, JCPDS 19-272) in all calcined samples. In line with the FTIR spectra, LAC400 and LAC500 displayed these HCA derived peaks at relatively earlier time points (1 day) compared to LAC600 and LAC700. In particular for LAC700, no distinct peaks attributable to HCA formation were observed until 7 days, which was due to its lower reactivity, and although HCA-related peaks were present, some original crystalline peaks of CaB₂O₄ still remained, indicating the coexistence of low crystalline HCA and CaB₂O₄.

Fig. 4.

XRD diffractograms of calcined samples, as-prepared, and after 6 h, 1 day and 7 days in SBF. All calcined particles converted to HCA (“■”, JCPDS 19-272) within 7 days

Figure 5 shows the SEM micrographs of all calcined particle surfaces before and after SBF immersion. Before SBF immersion, all calcined samples exhibited relatively smooth morphologies across the surfaces of the particles. However, after 1 day of SBF immersion, defined spherulitic-like HCA crystals [9, 15, 39] were observed on the surfaces of LAC400 and LAC500 particles, and these morphologies became more prominent by 7 days. In contrast, for LAC600 and LAC700, relatively defined spherulitic-like HCA crystals were not observed until 7 days of SBF immersion. EDS was used to estimate the Ca/P ratio of all samples after 7 days of immersion in SBF (Fig. 5), supporting the findings from ATR-FTIR and XRD.

Fig. 5.

SEM images of calcined samples before and after immersion in SBF, showing changes over time (as-prepared: top, 1 day: middle, 7 days: bottom). Ca/P ratios of samples at 7 days immersion in SBF as provided by EDS

Typically, in sol-gel-derived glasses, higher SSA allows for more rapid ion release or higher reactivity under water vapor [15, 40]. Based on this, as the calcination temperature increases and the SSA decreases, reactivity, ion release and bioactivity are expected to follow the same decreasing trend, as was observed with time in SBF. Given these findings, it can be predicted that as the calcination temperature increases, the bonds within BO₄ units in the glass structure are broken, leading to the formation of BO₃ units. This results in an increased fraction of BO₃ units in the glass network structure, causing a reduction in network connectivity. Simultaneously, pore collapse occurs, followed by densification. As a result, SSA decreases with increasing calcination temperature, lowering the reactivity in SBF. Therefore, it can be concluded that the expected decrease in ion release in SBF leads to a reduction in the HCA conversion rate as the calcination temperature increases. Among the calcined samples, the FTIR and XRD results for LAC400 and LAC500 are remarkably similar, and within the observation period of this study, these two samples exhibited the highest bioactivity with negligible differences in their trends for HCA formation. FTIR spectra showed that the detection of phosphate peaks, indicative of HCA formation, was delayed with increasing calcination temperature. On the other hand, the XRD results revealed that at 7 days of SBF immersion, LAC700 exhibited more distinct HCA-related peaks compared to LAC600. In a previous study [15], Lepry et al. examined the effect of varying calcination temperatures on the bioactivity of a sol-gel-derived B₂O₃-CaO-Na₂O-P₂O₅ glass system. Consistent with the findings in this study, increasing calcination temperatures led to crystallization and a decrease in SSA beyond a certain threshold. However, despite the reduced SSA, the crystallized sample retained their ability to convert to HCA. Similarly, in this study, and despite its crystalline structure and diminished SSA, the demonstrated bioactivity of LAC700 suggests that factors other than SSA have a role in HCA formation. Therefore, further investigations, particularly on the ion release behavior of LAC700 in SBF, are essential to elucidate the mechanisms underlying its bioactivity.

Conclusions

The influence of calcination temperature on the texture, structure, reactivity, and bioactivity of sol-gel-derived 40CaO-60B₂O₃ (mol%) glasses synthesized from calcium lactate pentahydrate was investigated. XRD confirmed the amorphous nature of samples calcined at 400, 500 and 600 °C, while calcination at 700 °C induced crystallization. ATR-FTIR revealed a decrease in N₄ with increasing calcination temperature. Moreover, SSA decreased due to a reduction in N₄ and densification, correlating with lower reactivity under vapor as indicated by DVS. This suggests that SSA significantly influences the reactivity under vapor. Bioactivity was confirmed by HCA formation within 7 days in SBF for all calcined samples by ATR-FTIR, XRD and SEM. In particular, LAC400 and LAC500 exhibited higher bioactivity. Therefore, in terms of bioactivity, calcination at 400 or 500 °C can be most effective. TGA showed that LAC400 contained the highest amount of organic residues, reducing its effective bioactive glass component per weight compared to LAC500. Given these findings, calcination at 500 °C is optimal for maximizing bioactivity. Additionally, the high bioactivity of 40CaO-60B₂O₃ (mol%) glasses synthesized from calcium lactate pentahydrate highlights that this low-cost precursor can be scalable and a promising choice for these bioactive glasses.

Author contributions

Kenta Katsumi: Conceptualization, Methodology, Investigation, Visualization, Writing – original draft, Writing – review & editing. Zhiqiu Ye: Methodology, Investigation. Pierre Hudon: Investigation, Writing – review & editing. Showan N Nazhat: Funding acquisition, Resources, Supervision, Writing – review & editing.

Funding

Funding was provided by Canada Natural Sciences and Engineering Research Council, Canada Foundation for Innovation and McGill University.

Data availability

No datasets were generated or analysed during the current study.

Compliance with ethical standards

Competing interests

The authors declare no competing interests.