and P2O5-Containing Silicate and Borosilicate Bioactive Glasses

Aug 12, 2019 - The present work elucidates about the structure of bioactive glasses having chemical compositions expressed as (mol %) (50.0 ...
0 downloads 0 Views 4MB Size
Subscriber access provided by Nottingham Trent University

B: Biomaterials and Membranes 2

5

Structure and Stability of High CaO and PO Containing Silicate and Borosilicate Bioactive Glasses Sakthi Prasad, Anuraag Gaddam, Anuradha Jana, Shashi Kant, Prasanta Kumar Sinha, Sucheta Tripathy, Kalyandurg Annapurna, José M. F. Ferreira, Amarnath Reddy Allu, and Kaushik Biswas J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.9b02455 • Publication Date (Web): 12 Aug 2019 Downloaded from pubs.acs.org on August 15, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Structure and Stability of High CaO and P2O5 Containing Silicate and Borosilicate Bioactive Glasses

Sakthi Prasad1,2, Anuraag Gaddam3, Anuradha Jana1,4, Shashi Kant5,6, P. K. Sinha7, Sucheta Tripathy5,6, K. Annapurna1,2, José M. F. Ferreira3, Amarnath R. Allu1,2 and Kaushik Biswas*1,2

1Academy

of Scientific and Innovative Research, Campus: CSIR – Central Glass & Ceramic

Research Institute, Kolkata, India - 700032 2Glass

Division, CSIR – Central Glass & Ceramic Research Institute, Kolkata, India –

700032 3Department

of Materials and Ceramic Engineering, CICECO, University of Aveiro, 3810-

193, Aveiro, Portugal 4Bioceramics

and Coating Division, CSIR – Central Glass & Ceramic Research Institute,

Kolkata, India - 700032 5Academy

of Scientific and Innovative Research, Campus: CSIR-Indian Institute of Chemical

Biology, Kolkata, India - 700032 6Structural

Biology and Bioinformatics Division, CSIR – Indian Institute of Chemical

Biology, Kolkata, India– 700032 7Analytical

Chemistry Group, CSIR – Central Glass & Ceramic Research Institute, Kolkata,

India - 700032

*Corresponding

Author: [email protected] 1 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 48

Abstract The present work elucidates about the structure of bioactive glasses having chemical compositions expressed as (mol %) (50.0-x) SiO2  x B2O3  9.3 Na2O  37 CaO  3.7 P2O5, where x = 0.0, 12.5, 25 and 37.5, and establishes a correlation between the structure and thermal stability. The structural modifications in the parent boron free glass (B0) with the gradual substitutions of B2O3 for SiO2 are assessed by Raman and

29Si, 31P, 11B

and

23Na

magic angle spinning (MAS)-nuclear magnetic resonance (NMR) spectroscopies. The structural studies reveal the presence of Q2Si and Q3Si structural units in both silicate and borosilicate glasses. However, Q4Si(3B) units additionally form upon incorporating B2O3 in B0 glass. B-containing silicate glasses exhibit both 3 coordinated boron (BIII) and 4 coordinated boron (BIV) units. The

31P

MAS-NMR studies reveal that the majority of

phosphate species exist as isolated orthophosphate (Q0P) units. The incorporation of B2O3 in B0 glass increases the crosslinking between the SiO4 and BO4 structural units. However, incorporation of B2O3 lowers the glass thermal stability (ΔT) as shown by differential scanning calorimetry (DSC). Although both silicate and borosilicate glasses exhibit good in vitro apatite forming ability and cell compatibility, the bactericidal action against E.coli bacteria is more evident in borosilicate glass in comparison to silicate base glass. The controlled release of (BO3)3 ions from boron modified bioactive glasses improves both the cell proliferation and the antibacterial properties, making them promising for hard tissue engineering applications.

Keywords: Bioactive glasses, SiO2 by B2O3 substitution, Thermal Stability, MTT assay, Broth-microdilution, Simulated body fluid (SBF) and Apatite formation.

2 ACS Paragon Plus Environment

Page 3 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

1.

Introduction The importance of bioactive glasses in hard tissue engineering applications has been

continuously increasing after their ability to bond to living tissues was first demonstrated by Hench.1 Fabricating bioactive glasses in the form of scaffolds and fibers can increase the ease of usage in patient specific and bone defect specific applications.2 In this aspect, good thermal stability of glasses is essential for accomplishing the sintering of the scaffolds and the fiber drawing without devitrification.2 Low thermal stability and uncontrolled crystallization are known to negatively affect the sintering behavior of bioactive glasses.3,4 Early crystallization occurs in commercial bioactive glasses with low thermal stability values such as 45S5 and S53P4 before achieving maximum densification.2 Therefore, achieving a good thermal stability as given by ΔT = Tx – Tg, where Tx is the crystallization onset temperature and Tg is the glass transition temperature, is very important for promoting viscous flow sintering and preventing crystallization.3 High ΔT values signifies a delayed crystallization process above Tg, enabling to obtain well-densified scaffolds upon sintering. The thermal stability (ΔT) can be tuned through suitable changes in glass compositions. One approach to increase ΔT is by using a high MO/M2O ratio where MO and M2O stand for alkaline earth oxide and alkali oxide, respectively.3,5,6 Studies of cation mixing and ordering in sodium calcium silicate glasses revealed non-random distributions of Na and Ca modifier cations and a preference for dissimilar pairs in silicate network structure, thereby contributing to an increase in the viscosity of calcium sodium silicate glasses.7,8 Another strategy for improving the thermal stability of the glasses is fixing a given MO/M2O ratio and explore the mixed alkali effect or mixed ion effect,6,9–13 i.e., using different oxides of each MO or M2O type.14,15 The enhanced thermal stability of 13-93 glass in comparison to those of 45S5 and S53P4 bioactive glasses has been achieved exploring the mixed alkali effect among Na2O, K2O, MgO and CaO.2,9,13 This behavior is explained based on a diffusion 3 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 48

phenomenon. According to which, the presence of an appropriate secondary ion concentration limits the energetically favorable hopping sites or ionic diffusion paths, thereby delaying the glass crystallization process.13,14,16,17 Certain additions of CaF2 have also been reported to improve ΔT due to the association of F anions with modifier cations with or without increasing silicate network connectivity.18–22 Apart from modifier cation variations, changes in the network forming cations, like the incorporation of B2O3 in 45S5, 13-93 and S53P4 glasses, also improved ΔT and the scaffold processing.23–28 These effects are ascribed to the ability of the borate structural units (BO3 and BO4) to be associated with silicate structural units.29,30 Increasing the P2O5 content in bioactive glass compositions is another way to enhance ΔT as the orthophosphate units attract modifier cations, thereby reducing the concentration of non-bridging oxygens (NBO) associated with silicate structural units.31 The compositional variation approaches often resulted either in limited improvements of ΔT or affect the bioactivity of the glasses and delay surface apatite formation. Nevertheless, proper concomitant compositional variations are likely to improve ΔT and faster apatite forming ability, like in the case of ICIE-16 glass in comparison to that of 45S5,6 while exploring the mixed alkali effect (Na2O+K2O) and a high MO/M2O ratio (3:1). Several multiple compositional modifications have induced favorable changes in ΔT as well as in apatite formation rate11,12,32–34 crediting the suitability of this strategy to enhance the overall performance of bioactive glasses. The same approach is also likely to enhance the apatite formation, the cell attachment behavior and the antibacterial effects, additionally preventing nosocomial infections in bioactive glass implants.35 Ionic releases from bioactive glasses containing K2O, MgO, SrO, B2O3, and CaF2 have been reported to exert therapeutic effects.10–12,23,32 B2O3containing glasses have exhibited enhanced thermal stability, faster apatite formation, cell compatibility,23 and antibacterial properties.23,25–28,35 These attractive features make the B-

4 ACS Paragon Plus Environment

Page 5 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

containing glasses reliable materials for hard and soft tissue engineering applications.36 The present work aims at further exploring the beneficial effects of B2O3 incorporation on thermal stability against devitrification, in vitro apatite formation, in vitro cell viability and antibacterial properties of a glass composition with high CaO:Na2O ratio and high P2O5 content. The parent boron free glass composition (B0) is given by 50 SiO2  9.3 Na2O  37.0 CaO  3.7 P2O5 (mol%) in which SiO2 was gradually replaced by B2O3 in 12.5, 25.0 and 37.5 mol % and the resulting glasses were labelled as B1, B2 and B3, respectively. The structural features of glasses have been assessed by Raman and MAS-NMR (for 31P, 29Si, 23Na and 11B nuclei) spectroscopic techniques. Apatite forming ability in simulated body fluid (SBF), in vitro cell viability, and antibacterial activity against E. coli bacteria have been evaluated to infer about the applicability of the biomaterials for hard tissue engineering applications. The effects of structural changes in glasses on their thermal, in vitro behavior and antibacterial properties are disclosed.

2.

Experimental Procedure

2.1. Materials and Reagents The glass compositions investigated in this work are presented in Table 1 along with that of ICIE-16 glass reported elsewhere.6 The parent glass composition (B0) was designed to have a CaO:Na2O ratio of 4:1 and a P2O5 content of 3.7 mol% (Ca/P = 5). B2O3 was gradually added to B0 glass, replacing SiO2 (12.5, 25 and 37.5 mol%) to obtain the glasses B1, B2 and B3, respectively. The glasses were prepared by the conventional melt quenching technique using Pt crucibles. The starting raw materials included SiO2 (99.8%, Sipur A1 Bremtheler Quartz-itwerk), Na2CO3 (99.5%, Merck), CaCO3 (99%, Sigma Aldrich), CaHPO4.2H2O (98% Sigma Aldrich) and H3BO3.2H2O (99.9999%, Sigma Aldrich). The melting temperature of B0 glass (1450 ºC) was gradually decreased to 1350 ºC, 1250 ºC and 5 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 48

1200 ºC, for glasses B1, B2 and B3, respectively. The melts, which were poured in a metallic mold and annealed, revealed an increasing devitrification trend for the B2 and B3 (discussed in the Section 3.1.1-XRD results). Accordingly, only the annealed B0 and B1 glasses were used for further characterization after being crushed in an agate mortar and pestle to obtain particle sizes < 45 m. Additionally, polished bulk samples of dimensions 10×10×2 mm3 were also prepared. 2.2. Characterization techniques The glass powders before and after in vitro testing were characterized by different techniques: X-ray diffraction (XRD, Rigaku, Ultima IV) within the 10  90º 2 with 0.02º step size and acquisition speed of 6 steps per second using Cu Kα X-rays (λ = 1.5418 Å); Fourier transform infrared spectroscopy (FTIR, Perkin Elmer, Frontier IRL 1280119) was performed using KBr transmission method within the mid-IR region (4000  400 cm1). The KBr (Uvasol® (particle size between 100 and 500 μm with mean size around 200 μm) was used to prepare the samples for Infrared spectroscopy (FTIR, Merck Millipore, Germany)). The powder samples were mixed with KBr in 100:1 weight ratio. To prevent the possibility of scattering loss, this mixture was further manually ground before the experiments using an agate mortar and pestle. This ground powder was then sieved through a set of sieves with a mesh size of 125, 106, 75, 63, 45, 20 and 10 μm and the obtained particle sizes were between 10  106 μm with a mean size around 45 μm. This sample-KBr mixture was subjected to load of 10 ton for 5 minutes holding time using hydraulic press to cylindrical pellets with 13 mm diameter. Throughout the study, the humidity of the room was maintained below 30% using dehumidifier to avoid possible hydration of samples. Polished bulk samples were used for density measurements using the Archimedes principle. Raman spectroscopy (Horiba Scientific, LabRAM HR Evolution) was performed

6 ACS Paragon Plus Environment

Page 7 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

on polished bulk samples within the range of 200 – 1600 cm1, using a 488 nm laser as incident beam source. The thermal properties of all investigated samples, including the ICIE16 glass, were assessed through differential scanning calorimetry (DSC, Netzsch, STA 44953) using 30 mg of each powder in Pt crucibles under a heating rate of 10 ºC min1. The network connectivity of silica (the average number of bridging oxygens per Si tetrahedral) in B0 glass (NCSi = 2.596) was calculated using the formula given by equation (1).37

(1)

NCSi =

where SiO2, P2O5, Na2O, CaO are obtained from the glass composition in mol%. 23Na, 29Si, 31P

MAS-NMR analysis was performed for B0 and B1 powders while 11B

MAS-NMR spectroscopy was performed only for sample B1. Bruker Avance III HD 700 MHz narrow-bore, and Bruker Avance III HD 400 MHz were used for the nuclei, and for the

31P

and

29Si,

23Na

and

11B

respectively. 4 mm triple-resonance MAS probe was

employed at 185.20 MHz (23Na), 224.63 MHz (11B) and 161.9 MHz (31P) Larmor frequencies. 7 mm MAS probe was used for 29Si nuclei at 79.5 MHz Larmor frequency. All samples were spun in ZrO2 rotors. A spinning rate of 15 kHz was used for nuclei; while nuclei

31P

and

29Si

23Na

were spun at 12 kHz and 5 kHz, respectively.

and

23Na

11B

MAS

spectra were recorded with a 15º excitation pulse with recycle delay of 2 s. 11B MAS NMR spectra were recorded using a Hahn-echo experiment with a 90º excitation soft pulse with recycle delay of 2 s. 31P MAS spectra were recorded with a 10º excitation pulse with recycle delay of 60 s. 29Si MAS spectra were recorded with a 90º excitation pulse with recycle delay of 60 s. Chemical shifts are quoted in ppm using the following references: saturated aqueous solution of NaCl (0 ppm), 0.1 M aqueous solution of H3BO3 (19.6 ppm), 85% H3PO3 (0 ppm) and tetramethylsilane (0 ppm), respectively, for

23Na, 11B, 31P

and

29Si.

To identify the 7

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 48

relative proportions of various structural units, the NMR spectra obtained for 11B, 31P and 29Si nuclei were deconvoluted using DMfit software.38

2.3. Bioactivity Study The apatite forming ability of glasses was tested using SBF (pH 7.4) prepared according to the Kokubo’s method.39 The TC-04 method was adapted for the powder sample based bioactivity study, in which the powders were immersed in SBF for 1, 3, 7 and 14 days in an incubator at 37 ± 0.5 ºC, using a proportion of 1.5 mg ml1.40 At each time point, the powders were filtered, washed dried and used to assess the surface apatite formation by XRD and FTIR. The microstructural changes were observed using FESEM instrument (Carl Zeiss, FESEM Supra 35 VP). The SBF supernatants collected at each time point were analyzed for pH and concentrations of (BO3)3, Na+ and Ca2+ ions using inductively coupled plasma – atomic emission spectroscopy (ICP-AES) at the emission wavelengths at 252 nm, 589 nm and 316 nm, respectively. The pH analysis was carried out on SBF supernatants at 37 ºC using pH meter fitted with a temperature electrode (Labard, L1-1122pH).

2.4.

Cell Proliferation Study The cell proliferation studies were carried out on polished bulk samples using

MC3T3 (Mouse osteoblast) cells in triplicates adapting MTT assay protocol given elsewhere.41,42 The S53P4 glass sample was synthesized as reported elsewhere43 and used as control. All the samples, including the control, were seeded with 5000 cells. The optical density (OD) readings were measured at 595 nm using enzyme-linked immunosorbent assay (ELISA) plate reader (Bio rad, iMark™) after 3, 5 and 7 days. The mean cell proliferation (%) calculated with respect to control and the standard deviation values of triplicate OD 8 ACS Paragon Plus Environment

Page 9 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

measurements at 3, 5 and 7 days were plotted for comparison. Student’s t-test was performed to check the significance level (P value) compared to the S53P4 glass considering two independent sets with equal variance and two-tailed hypothesis.

2.5.

Antibacterial Properties The samples were tested for their antibacterial activity against E.coli bacteria by

dispersing glass powder in inoculum made from Luria-Bertani (LB) broth and E.coli bacteria (initial concentration: 1.1±0.1 × 107 cells ml1). The glass powders were weighed and added to inoculum to obtain two concentrations, 5 mg ml1 and 10 mg ml1. S53P4 glass powder ( phosphates > borates.45 The band observed at 950 cm1 in B0 can be jointly attributed to stretching vibrations, ν(PO) in orthophosphate (Q0P)46 and ν(SiO) in Q2Si units.47–49 The broad shoulder with the peak maximum at 1076 cm1 can be primarily attributed to stretching vibrations, ν(SiO) of Q3Si units.47–49 The vibration band observed at around 602 cm1 which can be attributed to mixed stretching-bending vibrational bands of SiOSi units in Q3Si and Q2Si units are present in the B0 glass sample.47–49 A weak band observed at around 777 cm1 corresponds to mixed stretching-bending vibrations of SiOSi bridging bonds.47–49 The weak band near 430 cm1

10 ACS Paragon Plus Environment

Page 11 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

corresponds to the bending vibrations in OPO bonds within orthophosphate units.46 Therefore, the Raman spectra of B0 glass primarily indicates the presence of silicates in Q3Si as well as Q2Si form and phosphates in the Q0P form. Most of the band features observed in B0 glass were still found to be preserved in B1 glass, confirming that B2O3 incorporation did not induce significant structural modifications. The decrease in intensity of the band observed at 1076 cm1 could be primarily due to the lower SiO2 content in the B1 glass composition compared to that of B0 glass. The intensity of the broad band with a peak at 602 cm1, decreased due to the substitution of SiO2 by B2O3 in B1 glass. The broadening of the band may be attributed to the presence of various possible bending vibrational bands corresponding to BIIIOBIV, BIIIOSi, BIVOSi,50–55 POB56 bonds along with SiOSi47–49 bonds present in between 500 to 800 cm1. The weak band in between 14001500 cm1 in B1 glass can be attributed to stretching vibrations of BO and BO bonds in BIII units.50–55 The presence of metaborate, danburite and boroxol ring structures were ruled out due to absence of sharp bands in the respective band positions described elsewhere.50–55 3.1.3. MAS-NMR Spectra Analysis The MAS-NMR distributions of 31P, 29Si, 11B and 23Na units were analyzed to gain further understandings about the influence of chemical composition on the glass structure. The specific features of 31P, 29Si, 11B and 23Na MAS-NMR spectra for B0 and B1 samples are displayed in Fig. 3 and detailed below. 11B

MAS-NMR: 11B

MAS-NMR spectrum for B1 glass (Fig. 3a) shows two peaks centered at 14.8

ppm and 0.12 ppm, corresponding to BIII and BIV units, respectively. According, the spectrum was deconvoluted using two peaks: one peak with second order quadrupolar effects for BIII

11 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

units; and one Gaussian/Lorentzian peak for BIV units. The deconvoluted

Page 12 of 48

11B

MAS-NMR

spectrum for B1 glass (Fig. 3a) and the resulting NMR parameters are presented in Table 2. According to the NMR parameters, BIII peak would corresponds to asymmetric BIII units with one NBO.57,58 However, due to the broad nature of this peak, there might be other BIII units in smaller concentrations that could not be reliably identified. 31P

MAS-NMR: 31P

MAS-NMR spectra for B0 and B1 samples (Fig. 3b and 3c) show broad

resonance bands with peak maxima at around 4 ppm, indicating the predominant presence of orthophosphate (Q0P) structural units.16 Moreover, the chemical shifts of phosphorus resonance bands for B0 and B1 glasses are close to that of Ca3(PO4)2, indicating that Q0P units are mainly coordinated with Ca2+ cations. The deconvolution of 31P MAS-NMR spectra (Fig. 3b and 3c) revealed small fractions of pyrophosphate structural units in B0 and B1 glasses.59 These results together with peak assignments reported in Table 2 are in good agreement with those reported by Yu et al.30 Their glasses also showed that around 90% of P exists as orthophosphate units and the rest as pyrophosphate units. The chemical shift of Q0P units marginally moved towards the shielding side in B1 glass compared to B0. The

31P

MAS-NMR results of B0 glass are in agreement with the results reported by Lockyer et al.60 where the chemical shift of Q0P units was only influenced by the Na2O/CaO ratio in the composition. 29Si

MAS-NMR: The

29Si

MAS-NMR spectra of B0 and B1 glasses shown in Fig. 3d and 3e,

respectively, show broad resonance peaks centered at around 85 ppm, depicting the dominance of Q2Si and Q3Si structural units in both glasses.16,60 Adding B2O3 caused an enrichment in Q2Si at expenses of Q3Si units and a consequent overall move of the peak center towards higher chemical shift side. The chemical shift of a 29Si NMR spectrum is known to

12 ACS Paragon Plus Environment

Page 13 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

strongly depend on the type of atoms distributed in the next nearest neighborhood. The deconvolutions of the

29Si

NMR spectra with two Gaussian/Lorentzian peaks, Fig. 3d, Fig.

3e, together with the results reported in Table 2, show that about 40 and 48% of Si exists as in B0 and B1, respectively, and the rest exists as Q3Si. The NMR deconvolution analysis for 11B, 31P and 29Si spectra for B0 agrees well with the glass composition, but for B1 there is a discrepancy, suggesting an excess amount of modifiers than expected from the chemical composition. MAS-NMR and Raman data on alkali-containing borosilicate glasses revealed the coexistence of both BIII and BIV units mix with silicate units.50,53,54 The influence of BIV units on

29Si

MAS-NMR chemical shift is apparently similar to the effect of AlO4 on

29Si

MAS-NMR chemical shift of Q4Si to 90 ppm as Q3Si when surrounded by 3 AlO4 in the next nearest neighbourhood.61 The discrepancy in the B1 analysis might be due to the overlapping of Q3Si and Q4Si(3B or 4B) units. Based on the 11B NMR spectra and the amount of BIV units (33%), the overall estimated fraction of Q4Si(3B or 4B) is between 7 (for Q4Si(4B)) and 9% (for Q4Si(3B)), which would overlap with the Q3Si peak. Therefore, the relative amounts of Q3Si units will be between 45 to 42%. A better match between NMR composition and actual composition is achieved when purely (Q4Si(3B)) units are considered. 23Na

MAS-NMR: 23Na

MAS-NMR spectra of B0 and B1 glasses displayed in Fig. 3f show broad

resonance peaks centered at 6.11 for B0 and at 6.61 ppm for B1. Such broad features denote that the Na+ ions are distributed among different silicate, phosphate and borate structural units. There is a clear chemical shifting trend of the 23Na MAS-NMR peak towards lower values with the addition of B2O3, which was attributed to an increase in the NaO bond length.62 The length of Na‒O bonds tends to increase when sodium acts as a charge compensator than as a modifier cation. Therefore, the decrease in 23Na MAS‒NMR chemical shift in the present study is due to the role of Na changing from network modifier to charge 13 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 48

compensator. Stone-Weise et al.63 have also reported that sodium cations originally present as NBOs, tend to convert from modifier cations into charge compensators for the

units upon

substituting B2O3 for SiO2 in a SiO2B2O3Na2O glass system. 3.2. Thermal and Physical Properties The DSC thermographs of the glass powders recorded at 10 ºC min−1 are shown in Fig. 4. The intersection points of the two tangents drawn at first endothermic dip and at the first exothermic peak correspond to Tg and Tx, respectively. The values of thermal stability (ΔT = TxTg) could be calculated from these parameters and data are reported in Table 3. The maximum ΔT = 175 ºC is found for B0 glass, and that all the thermal parameters decrease with the incorporation of B2O3 at the expenses of SiO2. A similar reduction in Tg value with addition of successive amounts of B2O3 in the Si–BNaPCa/Sr glass system was reported elsewhere.29 The decrease in ΔT is mainly due to more significant reductions of Tx in comparison to Tg values. The density values measured for the B0 and B1 glasses were 2.77±0.01 and 2.75±0.01 g cm–3, respectively, from which Network Volumes (NV) of the glasses were calculated using equation (2).58 The NV is the volume occupied per mole of network forming structural units in a glass.

NV =

(2)

where SiO2, P2O5, B2O3, Na2O and CaO being the molar oxide concentrations in the glass composition, while MW is the molecular weight of each oxide denoted in the subscript and ρ is the measured density value.

14 ACS Paragon Plus Environment

Page 15 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

The resulting NV values of 42.17±0.11 and 35.50±0.11 cm3 mol–1 found for B0 and B1, respectively, are in accordance with the decreasing trend reported when the overall polymerization of the glass network increases.

3.3. Crystallization Behavior of the Glasses Heat treating the glasses at Tx followed by identification of crystalline phases (Figures not shown), revealed the formation of pseudowollastonite (Ca3Si3O9) (JCPDS No.: 74-0874) and devitrite (Na2Ca3Si6O16) (JCPDS No.: 23-0671) in B0; while crystalline dicalcium silicate (Ca2SiO4) (JCPDS No.: 86-0399) and oxyapatite (Ca10(PO4)6O) (JCPDS No.: 89-6495) phases were found for B1. Thus, the incorporation of B2O3 in B0 glass exerted significant effects on the resulting crystalline phase assemblage. 3.4. Bio-Mineralization Ability The XRD patterns of B0 and B1 glasses recorded before and after immersion in SBF for different periods are shown in Fig. 5a and 5b, respectively. After 1 day, only amorphous halos at 2θ ~22.5º and 2θ ~30º, corresponding to silica gel layer,23 and to amorphous calcium phosphate (ACP) layer64 can be seen. After 3 days, sharp peaks with positions matching the standard XRD profiles of HAp (JCPDS No.: 09-0432) and HCA (JCPDS No.: 04-0697) are clearly observed, the intensity of which increase with immersion time, and coexisting with the amorphous halo at 2θ ~22.5º. The FTIR spectra of B0 and B1 samples before and after immersion in SBF for different periods are shown in Fig. 5c and 5d, respectively. The FTIR spectrum of B0 glass before SBF immersion shows broad vibrational bands corresponding to silicates, while B1 glass before SBF immersion exhibits vibrational bands of both borates and silicates. The distinct absorption band at 1423 cm1 corresponds to BO bond stretching in

units. The

15 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

asymmetric stretching vibrational bands of BO bonds in

Page 16 of 48

units occur in the range 800–

1050 cm1 and could not be clearly distinguished in B1 glass65,66 due to overlapping of SiOSi asymmetric stretching and SiNBO stretching vibrational bands at 1075 cm1 and 913 cm1, respectively. Similarly, the variations in the silicate structural units accompanied with B2O3 incorporation are difficult to distinguish in the present FTIR spectra. The FTIR spectra of B0 and B1 glasses after SBF immersion showed a broad band around 3400 cm1 (not shown), corresponding to the vibrational modes of adsorbed water. The bands denoting stretching vibrations of OH groups in HAp lattice should also be present between 3600 cm1 and 3700 cm1, but were not clearly distinguishable from the broad band. Apart from that, B0 and B1 samples immersed in SBF for 1, 3, 7 and 14 days primarily show bending of OPO bond and stretching bands of PO bond in PO4 units, SiOSi bonds and CO bonds in carbonate units. A weak band at 588 cm1 in the as-prepared B0 and B1 samples corresponds to OPO bending vibrations in amorphous form.67,68 However, the split bands corresponding to bending in OPO bond appear around 560 and 600 cm1 after 3 days of immersion in SBF, denoting crystalline apatite formation.69 The additional presence of carbonate bending vibrational bands at around 872 cm1 and the carbonate stretching vibrations at around 1473 cm1 and 1415 cm1 confirm the formation of B-type carbonated apatite67,70 in both B0 and B1 glasses. Additionally, the bands signifying stretching modes in BIII units around 1423 cm1 disappeared after immersion in SBF for 1 day, and split bands corresponding to stretching modes in carbonates appear at around 1415 cm1 and 1473 cm1. This corroborates with the release of BIII units into the SBF while the carbonates from SBF precipitated on the glass surface. The FESEM micrographs obtained for the glass samples immersed in SBF for 1 and 3 days are shown in Fig. 6. After 1 day, both B0 (Fig. 6a) and B1 (Fig. 6b) samples show

16 ACS Paragon Plus Environment

Page 17 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

droplet type structures deposited on surface. This type of structure evolved to a flake like morphology with increasing immersion time to 3 days as shown in Fig. 6c and Fig. 6d, for B0 and B1 glasses, respectively. Both XRD and FTIR results confirm the formation of polymerized silica gel layer upon immersing the silicate (B0) or borosilicate (B1) glasses in SBF for 1 day denoting the ion release from glasses.23 Additionally, XRD and FTIR results also confirm the formation of ACP layer in B0 and B1 glasses after 1 day SBF immersion which could be related to the droplet structures in the FESEM micrographs of samples. There is no doubt about the formation of crystalline apatite, but the HAp or HCA phases could not be clearly distinguished from XRD results. However, the presence of carbonate vibrational bands in FTIR spectra confirmed the formation of B-type HCA phase. The flake like morphology in the FESEM micrographs of samples after 3 days of immersion in SBF is consistent with the formation of the apatite layer as confirmed from XRD and FTIR results. These results confirm the excellent bio-mineralization capability of the investigated glasses. 3.5.

pH and Ion Concentration Analysis The average pH values and concentrations of (BO3)3, Ca2+ and Na+ ions of SBF

supernatants after 1, 3, 7 and 14 days are shown in Fig. 7a and 7b, respectively along with standard deviations as error bars. The initial pH = 7.4 of SBF solution increased upon immersion of both B0 and B1 glasses due to the ionic exchange reactions, with maximum values being registered at 3 days, followed by a gradual decrease at longer immersion periods. Since, HAp and B-type HCA formation requires OH ions from the surrounding medium; pH is likely to decrease with apatite formation. According to the XRD results, the apatite formation becomes predominant after immersion of 3 days, explaining this pH evolution trend. Ionic concentrations of Na+ and Ca2+ ions were slightly higher for B1 samples in comparison to the B0 ones, indicating a marginal increase in the release of modifier cations induced by the 17 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 48

incorporation of B2O3 in silicate glass. The concentration of (BO3)3 ions within one day of immersing the B1 glass in SBF was ~46 ppm and gradually increased to ~64 ppm within 14 days. 3.6.

In vitro Cell Proliferation Studies The percentages of MC3T3 cells proliferation onto the surface of glass samples with

respect to control at 3, 5 and 7 days (Fig. 8a) reveal an increase up to 5 days with the incorporation of B2O3. At 7 days, B0 shows the highest cell proliferation followed by B1. Neither of the investigated glass samples did show any evidence of toxicity, exhibiting better cell compatibility than the control, as confirmed from the increase in the cell proliferation along with the testing period. The significance levels for B0 and B1 glasses were statistically highly significant (P < 0.001) compared to S53P4 sample. 3.7.

Antibacterial Studies Figure 8b plotting the CFU counts shows that the number in the negative control

doubled after 24 h, while drastic reductions were observed for all glasses at both the concentrations tested. The higher inoculum concentration (10 mg ml1) exerted a more accentuated reduction in CFU in comparison to 5 mg ml1, reflecting the extents of bactericidal action. The commercial S53P4 glass developed for osteomyelitis treatments71 shows highest bactericidal action among all the samples studied. Nearest bactericidal effect compared to S53P4 glass was observed for B0 at 5 mg ml1, and for B1 glass at 10 mg ml1. A statistically significant level (P