Raman Spectroscopy Analysis of Wollastonite/Tricalcium Phosphate Glass-Ceramics after Implantation in Critical Bone Defect in Rats ()
1. Introduction
The treatment of pathologic or traumatic extensive bone losses requires the application of bone substitutes since the natural physiological bone healing mechanism is insufficient to reach total bone tissue regeneration in a short time [1]. Currently, all the known materials used as bone substitutes still have some limitations and the search for the ideal bone substitute continues [2]. Several synthetic materials have been proposed for bone regeneration, among them, those consisting of several components that contribute with properties to render enhanced osteoinduction and osseointegration are attracting considerable attention [3].
Glass-ceramics belonging to the binary system wollastonite-tricalcium phosphate (W/TCP) has been extensively studied in the last two decades as they exhibit greater bioactivity and better mechanical resistance than the well-known calcium phosphate bioceramics [3] - [8]. The in vitro and in vivo bioactivity of W and its potential in bone regeneration have been well established in many published and explained on the basis of its incongruent dissolution and the calcium and silicon ions released when in contact with physiological fluids [9]. In particular, the silicon ion has a well-recognized capacity to induce osteogenesis and angiogenesis [9] [10].
Furthermore, the addition of W to calcium phosphates to manufacture sintered composites contributes to reinforcing and densifying the resulting glass-ceramic and improving its bioactivity [10].
Calcium phosphates are similar to the mineral phase of bone matrix, teeth, and calcified tissues. TCP is one of the most utilized calcium phosphates as bone substitute, due to its partial biodegradability, osteoconductivity and bone adhesion. Such properties result in its wide clinical use, as well as its application in the development of new composites with the purpose of bone regeneration [11] [12] [13].
Thus, combining the properties of W and TCP in the form of sintered glass-ceramic with different W/TCP ratios for the sake of increased osteoinduction and resorption may result in a bone substitute with better properties than its components. Moreover, the use of different W/TCP weight ratios may be a good strategy to tune the in vivo resorption rate of W/TCP glass-ceramics [14] [15]. The ideal amount of each component in the glass-ceramic becomes a challenge in the quest for the better performance of the material in the host organism, which is necessary to evaluate by in vivo implantation models.
Among the various techniques available, Raman spectroscopy has been widely used in the characterization of glasses and ceramics as well as in the assessment of in vivo responses to bone regeneration materials [16] [17] [18] [19]. It is a non-destructive technique requiring no or little preparation of samples and is widely used in molecular and structural analysis of complex compounds. This technique is based on the inelastic scattering of monochromatic light by the sample and the analysis of the frequency shift of the scattered light, which provides information about vibrational, rotational and other low-frequency transitions in molecules or crystals [20] [21] [22] [23].
In this scenario, this work studied the compositional modifications of W/TCP glass-ceramics with different W/TCP weight ratios after implantation in a critical bone defect model in rats for several periods, using Raman spectroscopy, and describes the ability of the studied glass-ceramics to induce the formation of organic matrix and mineralization of newly formed tissue, aiming to identify the best performance during the bone regeneration.
2. Materials and Methods
2.1. Glass-Ceramics
Experimental W/TCP glass-ceramics (W/TCP weight ratios: 80/20, 60/40 and 20/80) were kindly supplied by the Institute of Ceramics and Glass (ICV-CSIC, Madrid, Spain).
They were manufactured from powders of natural wollastonite, CaSiO3 (NYCO, NYAD M1250, Batch Number M1404227J01) and tricalcium phosphate, Ca3(PO4)2 (Sigma-Aldrich, Ref. 21218, Batch Number BCBM7330V). Briefly, the required amounts of both raw materials were mixed with isopropanol and the mixture was ball milled in an attrition mill. The resulting slurry was dried at 60˚C overnight and the dry cake was sintered at the selected temperature and dwelling time (20 - 80: 3 hr. at 1250˚C; 60 - 40: 1 hr. at 1250˚C; and 80 - 20 = 5 min. at 1250˚C). The sintered cakes were crushed in a jaw mill and sieved to collect particles with sizes in the range of 0.6 - 0.4 mm.
Granulate 60/40 was received sterile (25 kGy, e-beam) whereas 20/80 and 80/20 granulates were received non-sterile and sterilized by dry heat (180˚C, 30 min) prior to being used.
The 20 - 80 granules consisted mainly of crystalline β and αTCP and βW, in this order of abundance, and minor amounts of crystalline αW and amorphous glassy phase (AGP). On the other hand, 60/40 and 80/20 granules were composed of crystalline αW and βTCP and AGP. The chemical and phase compositions of the experimental materials are resumed in Table 1.
2.2. Animal Model and Implantation Procedure
The study was approved by the Ethics Committee on Animal Use of the Federal
![]()
Table 1. Chemical and phase compositions of the experimental glass-ceramics.
a. Semi-quantitative scale: +++ Most abundant phase; ++ Abundant; + Minor; - Scarce. b. Trace impurities as MgO, Al2O3, K2O, Na2O, Fe2O3, TiO2. c. nd: non detected; AGP.: amorphous glassy phase.
University of Bahia, Brazil, (No. 128/2017), and followed the Guide for the Care and Use of Laboratory Animals prepared by the National Academy of Sciences [24], as well as the norms of the National Council for the Control of Animal Experimentation (CONCEA). Seventy-two adult male rats (Rattus norvegicus) weighing 350 - 400 g and 4 - 5-month age were used. Animals were kipped in individual cages at a temperature of 22˚C (±2) exposed to alternating periods of 12 hr. light and 12 hr. dark and fed and watered ad libitum. Animals were randomly distributed into experimental 4 groups (n = 20 each): three groups underwent surgery to create a circular bone defect of 8 mm in diameter in the central portion of the calvaria which was filled with one of the experimental glass-ceramics 20/80, 60/40, and 80/20. The fourth group (CG) did not undergo surgery and was used as a control to establish the spectroscopic parameters of the normal cortical bone of the calvaria.
Inside each group, subgroups (n = 5) were sacrificed after implantation periods of 7, 15, 45 and 120 days. The calvarias were removed, dissected, and fixed in 10% formaldehyde (v/v) at 4˚C - 8˚C. Fixed samples were then prepared according to Meade et al. [25], this is, were removed from the fixing solution, washed in filtered water, and immersed in 0.9% saline solution for ten minutes, then removed and dried with a paper towel. The experimental design is shown in Figure 1.
2.3. Raman Spectroscopic Analysis
An FT-Raman Spectrophotometer RFS 100/S (Bruker, Germany) with YAG: Nd laser (1064 nm) excitation source and Ge-diode, cooled at liquid N2 temperature was used. The sample was directly positioned in the sample holder with the laser beam adjusted to a power of 10 mW and focused on the central portion.
Spectra in the region 300 - 3500 cm−1 were obtained with a resolution of 4 cm−1 and an accumulation of 512 scans. These parameters were obtained from previous experiments to obtain the maximum signal-to-noise ratio without altering the physical and chemical integrity of samples.
OPUS 6.0 software (Bruker, Germany) was used to acquire and process Raman data, and all spectra were obtained at least three times, after focusing the
![]()
Figure 1. Experimental design: (a) Site were deffect surgery was practiced; (b) Aspect of the critical size bone defect (8 mm); (c) Deffect filled with experimental glass-ceramic; (d) Calvaria specimen with the implant site at the central portion (green arrow); and (e) Position of sample in the FT-Raman instrument.
laser beam at a different point on the sample, to verify the reproducibility of the system and possible uncertainties in intensity or Raman shift of the vibrational bands observed.
GRAMS/AI 7.02 software package (Thermo Galactic, USA) was used for import/export of Bruker OPUS File Format data and conversion to ASCII-XY. The spectra were plotted in OriginLab® version 2020b program, and analyzed for the presence, position, width, and intensities of the Raman bands. The spectra of the glass-ceramics, as received, and after the different implantation periods were analyzed to identify the modifications associated with resorption or permanence of the constituting crystalline phases, or the deposition of new organic or inorganic substances related to new bone tissue formation at the implant site. The assignation of the bands observed in the experimental spectra was carried out on the basis of the phase composition of the experimental glass-ceramics in Table 1 and the data published in the literature [26] - [31].
3. Results
The spectra of the group 20/80 after the different periods of implantation is displayed in Figure 2. After 7, 15 and 45 days the characteristic Raman bands of αW, β- and αTCP phases were still distinguishable. Additionally, new bands arose at 1668, 1452 and 877 cm−1 with very weak intensities during implantation. At 120-day period, the characteristic bands of the βTCP phase remain but those corresponding to αW (580 and 374 cm−1) are no longer observed [19] [27].
The results for the 60/40 and 80/20 experimental groups are shown in Figure 3 and Figure 4, respectively. In these experimental groups bands at 1668 cm−1 (amide I); 1452 cm−1 (CH2 and CH3 lipids/collagen); 1271 and 1248 cm−1 (amide III); 1005 cm−1 (phenylalanine); 877, 856 and 818 cm−1 (proline, hydroxyproline and tyrosine), all consistent with the deposition of collagen on the implant site, were observed [32].
These groups also showed spectral modifications around 1071, 592 and 432 cm−1, consisting of the appearance of bands with significant intensity differences between groups. From the 7-day period, the main bands attributed to the αW mineral phase (1076, 985, 580 and 374 cm−1) [19] [26] [28] are no longer detected, suggesting that it is quickly resorbed in the physiological environment.
![]()
Figure 2. Raman spectra of 20/80 glass-ceramic: before (0 d) and after implantation for 7-, 15-, 45-, and 120-day periods, and the spectrum of non-implanted rat calvaria (CG). Wavenumber range (a) 3500 - 1000 cm−1; (b) 1000 - 900 cm−1; (c) 900 - 300 cm−1.
On the other hand, the most prominent bands attributed to βTCP and αTCP phases (974 and 954 cm−1, respectively) are observed up to the 45-day period. At 120-day period, 60/40 and 80/20 groups presented a unique band centered at 962 cm−1. Only in the 80/20 group was visible a low intensity shoulder at 972 cm−1 due to a small amount of βTCP remaining [19] [29] [30] [31].
![]()
Figure 3. Raman spectra of 60/40 glass-ceramic: before (0 d) and after implantation for 7-, 15-, 45-, and 120 - day periods, and the spectrum of non-implanted rat calvaria (CG). Wavenumber range (a) 3500 - 1000 cm−1; (b) 1000 - 900 cm−1; (c) 900 - 300 cm−1.
In the 20/80 group, after implantation for 120 days, only a few bands associated with the deposition of organic components at the implant site were observed, whereas the bands attributed to the inorganic constituents of the glass-ceramic remained.
![]()
Figure 4. Raman spectra of 80/20 glass-ceramic: before (0 d) and after implantation for 7-, 15-, 45-, and 120-day periods, and the spectrum of non-implanted rat calvaria (CG). Wavenumber range (a) 3500 - 1000 cm−1; (b) 1000 - 900 cm−1; (c) 900 - 300 cm−1.
Groups 60/40 and 80/20 presented bands very similar to Raman shifts, but with significant differences in their intensity ratios.
4. Discussion
The results of this study suggest that the experimental materials studied, i.e., W/TCP glass-ceramics with nominal weight ratios of 20/80, 60/40 and 80/20, exhibit quite different behavior with respect to resorption and new bone tissue deposition when implanted in rat calvaria.
The Raman vibrational profiles of the studied glass-ceramics experienced drastic modifications with the implantation time evidencing partial resorption of the inorganic constituents of the implanted material and the simultaneous deposition of organic and inorganic substances consistent with the formation of new bone tissue at the implant site.
The modifications in Raman shift, width and intensity of bands were dependent on the implantation period, and permit monitoring of the resorption of the different crystalline phases constituting the glass-ceramics as well as the deposition and formation of new substances both organic and inorganic.
Results indicated that the composition of 20/80 glass-ceramic remained almost unvaried up to the 45-day period; only after 120 days, the bands assigned to the αW disappeared. This suggests that the resorption rate of the 20/80 glass-ceramic is low and probably insufficient for bone regeneration purposes, keeping in mind that an ideal bone substitute should be resorbed and replaced by new bone tissue in a simultaneous and coordinated process [33]. However, results indicate that it can be used as an osteoconductive filling material for repairing bone defects.
Results indicated that the αW phase at the 7-day period of implantation of glass-ceramics was completely resorbed in groups 60/40 and 80/20, as evidenced by the absence of the characteristic bands at 1075, 985, 933, 580, 559 and 511 cm−1 [19] [26] [27] [28]. These results corroborate the results of previous studies that evaluated the behavior of ceramics containing αW and TCP, both in vitro and in vivo and reported its fast and incongruent dissolution and the pseudomorphic transformation of the TCP phase into biological apatite (Ap) [9].
Another study used wollastonite as bioactive, resorbable, and reinforcing filler in a W/polyvinyl alcohol composite for bone regeneration and concluded that it was possible to obtain controlled biodegradation and induce the formation of apatite on the material surface after 7 days of immersion in simulated body fluid [34]. Others also demonstrated that the presence of W in composites enhanced bioresorption rates and favored the initial bone repair dynamics [34] [35] [36] [37].
Considering the high in vivo resorption rate of αW it is reasonable to consider that different amounts of Ca and Si ions are released from the experimental materials used in this study which should be proportional to the αW content. The amount of surface amorphous silica and Ap nucleation on the implanted glass-ceramic should be dependent on its αW content as well [9] [38]. It is well known that calcium and silicon ions play an important in angiogenesis and osteogenesis in vivo [39] [40] [41].
Glass-ceramics with osteogenic behavior should primarily influence the local cell metabolism to promote cell differentiation into osteoblasts and stimulate their proliferation, with subsequent production of sufficient new bone tissue to fill the critical bone defect. Maia Filho et al. [42] used Raman spectroscopy to study in vivo ceramic implants and found that those that induced the higher amounts of organic matrix deposition also achieved higher mineralization and new bone formation.
The results of the present study showed different spectral behavior for the formation of Raman bands around 1668 cm−1 (amide I), 1452 cm−1 (methyl group), 1271 and 1248 cm−1 (amide II), 1005 cm−1 (phenylalanine), 877, 856 and 818 cm−1 (proline, hydroxyproline and tyrosine) characteristic of collagen, as reported in studies that evaluated bone repair through Raman spectroscopy [43] [44] [45].
It is understood that the different percentage proportions of αW in glass-ceramics were decisive for the formation of organic components with extremes (20/80 and 80/20) exhibiting lower performances, and the 60/40 showing better performance, as resumed in Figure 5.
Although the literature does not define the ideal amount for silicon ion saturation and osteogenesis promotion, it is reasonable to infer that the amount of silicon ions released through αW resorption influences the osteogenesis potential of glass-ceramics under study.
Uribe et al. [46] investigated the in vitro osteogenic effects of dental follicle stem cells in simulated body fluid containing different Si concentrations and concluded that 25 µg/ml of Si had a significant positive effect on osteogenesis. On the other hand, solutions with Si < 25 µg/mL and >75 µg/mL did not show significant effects, whereas concentrations ≥ 100 µg/ml had an inhibitory effect on osteogenesis. Such findings corroborate the results of the present study, and permit to explain the lower deposition of organic matter on glass-ceramics with extreme contents of αW (20/80 and 80/20) and expect a poorer performance for bone regeneration as well.
![]()
Figure 5. Raman band intensities reached by the experimental groups at 120 days after glass-ceramic implantation and comparison with CG.
However, dos Santos et al. [47] found better results for bone regeneration using a 3D scaffold made of glass-ceramic of W/TCP with a weight ratio of 20/80 evidencing that, besides chemical composition, biomaterial porosity (proportional to the material/body contact area) exerts a decisive influence on the interaction with the host organism.
The group 60/40 showed the formation of characteristic bands of organic components and achieved great spectral similarity with CG. In the present study, the positions, and intensities of bands such as those observed in CG indicate the formation and organization of new tissue with characteristics similar to the normal cortical bone [18].
Regarding the compositional modifications associated with mineral polymorphs, the spectral results of the group 20/80 group demonstrated the permanence of bands characteristic of the β- and α-TCP phases throughout all analyzed periods. This indicates that the 20/80 glass-ceramic performed as an osteoconductive filling material in the critical bone defect model. Groups 60/40 and 80/20 also presented bands attributed to β- and α-TCP phases (971 and 954 cm−1, respectively), but only up to the 45-day period. After 120 days of implantation, only the 972 cm−1 band was visible in the 80/20 group as a shoulder of weak intensity, indicating a slower biodegradation rate of β- with regard to α-TCP. It is known that β-TCP and α-TCP mineral phases are isomorphs with the same chemical composition; however, their differences in crystal structures promote different in vivo behaviors [9]. Thus, the results found corroborate other studies that verified that the solubility index of the β-phase was lower than that of the αTCP phase resulting in longer permanence of βTCP at the implant site [48] [49] [50].
The results obtained also showed that in 60/40 and 80/20 groups, there were spectral changes consisting of the occurrence of new bands characteristics of vibrational modes of the PO4 group. The Raman shifts observed (962, 592 and 432 cm−1) are characteristic of the Ap composing bone mineral [32]; however, these bands presented important differences in intensities between the experimental groups.
The intensities of these bands can be directly related to the osteogenic potentials of the glass-ceramics under study. It is understood that, among several factors, the intensities of Raman bands are also related to the amount of active molecular species present in the sample [51]. Therefore, the intensity of bands exhibited at the implant site can be related to components formed and/or deposited over the evaluated periods.
The mineralization quality achieved in 60/40 and 80/20 groups was evaluated by comparing the values of Full Width at Half Maximum (FWHM) in cm−1 for the Raman band centered at 962 cm−1, which is the more intense band in the Raman spectrum bone mineral phase, Ap. According to the literature, the FWHM value of this band is inversely related to the crystallinity degree of Ap [52]. In our study, the comparison of FWHM values of the 962 cm−1 band indicated different crystallinity degrees between experimental groups when compared to CG (see Figure 6).
![]()
Figure 6. Comparison of FWHM of the 962 cm−1 band at 120 days after the implantation of glass-ceramics. (a) Control group; (b) 60/40 group; (c) 80/20 group.
The group 80/20 showed higher FWHM than the group 60/40 and CG. This indicates that 80/20 glass-ceramic induced the formation of poor crystalline Ap, probably because the fast dissolution of αW obtained lower performance for the formation of organic components and mineralization of the newly formed tissue, and 60/40 resulted in a greater amount and better organization of mineral components deposited and similarity to the spectrum of biological apatite in the CG.
Results of this study showed that the 60/40 glass-ceramic presented the greatest potential for future clinical applications as bone regeneration material.
5. Conclusion
The spectral changes observed for 60/40 and 80/20 glass-ceramics demonstrated that they exhibited gradual biodegradation and compositional changes during the in vivo implantation periods studied, indicating they have osteogenic potential, mainly the 60/40, which achieved the greater spectral similarity with normal bone. During the analysis periods, 20/80 glass-ceramic showed scarce spectral changes which evidence lower in vivo reactivity and resorption, and behavior typical of osteoconductive filling materials.
Acknowledgements
The authors would like to thank the Foundation for Research Support of the State of Bahia (FAPESB), the Department of Health Technology and Biology and the Laboratories of Materials Characterization, and of Radiological Physics of the Federal Institute of Education, Science and Technology of Bahia (IFBA) and express special thanks to Professor Fabiana Paim Rosa.
Raul Garcia Carrodeguas thanks grant PTQ-16-08573 funded by MICINN/AEI, Spain.