-
ace
Su
100325
Boron doping;Low melting temperature;Degradation rate;Bioactive
glass ceramics;Solgel
Cs)tionC
properties of BGCs via low-temperature co-red process. The
B2O3-free BG-B0
e glass
Journal of Non-Crystalline Solids 358 (2012) 11711179
Contents lists available at SciVerse ScienceDirect
Journal of Non-Cr
evtem (e.g., Bioglass 45S5) was able to bond to bone mineral
[1]. Thebioactivity of BGs and BGCs is attributed to the formation
of a bone-like hydroxyl-carbonated apatite (HCA) layer on their
surface in the(simulated) biological environments, thus a strong
bond can formwith the bone tissue. The typical 60SiO236CaO4P2O5 (in
mol%;termed 58S) system has been studied extensively and this
systemproduced by a solgel technique was more bioactive than the
melt-quenching BG of the same composition [2]. SiO2CaOP2O5
systemincorporated with ZnO, MgO, and SrO as network modiers
demon-strated the effect of slowing down the degradation of the
material[36]. Additional drawbacks found in several families of
porous
P2O5MgO system composed of crystalline wollastonite, apatite
andthe residual glassy phase (Cerabone-AW) has been developedwith
improved mechanical strength, whereas the over-high
sinteringtemperature (>1100 C) leads to apatite production and
undesiredlow degradability [13]. The liquidus temperatures of
mostphosphorus-, and silica-rich phosphosilicate BGs are higher
than1000 C, and it therefore suggests that the crystalline apatite
in sin-tered BGCs decreases the material bioactivity, mechanical
strengthand dissolution rate [10,1315]. To fabricate load-bearing
BGCs it isnecessary to sintering the BGs at high temperature to
full densitywhile maintaining porosity which is required for
nutrient diffusionsilicate-based BG materials such as high
liqbrittleness and low biodegradation ratewhere load bearing is
required [710].
Corresponding author. Tel.: +86 571 8697 1782; faE-mail address:
[email protected] (Z. Gou).
0022-3093/$ see front matter 2012 Elsevier B.V.
Alldoi:10.1016/j.jnoncrysol.2012.02.005-ceramics (BGCs) of spe-ur
decades since HenchOSiO2Na2OP2O5 sys-
[11]. It is accepted that the bioactive behavior was related
with theinitial composition of the BG, the phase composition after
thermaltreatment and sintering process [12,13]. A well known
SiO2CaOcic compositions have been studied for foet al. when they
found the melt-quenching Ca1. Introduction
Bioactive glasses (BGs) and bioactivof the 20 mol% B2O3-doped
BG-B20 was lowered to ~648 C and ~952 C, respectively. The BG-B20
thermallytreated at 850950 C was transformed into wollastonite and
calcium borate, and crystallization decreasedthe kinetics but did
not inhibit the development of hydroxyapatite on their powder and
disc surface whenimmersed in simulated body uid. The in vitro
degradation in Tris buffer conrmed that the degradationrate
markedly increased with increasing boron content in BG-Bx. The
compressive strength and exuralstrength of the 10%
BG-B20-reinforced 45S5 porous BGC sintered at 850 C was nearly four
times than thatof 45S5 porous constructs. These studies suggest
that the boron-rich, phosphorus-low CaOSiO2P2O5B2O3system is a
promising biomaterial and potential low temperature co-red aid for
improving the mechanicaland biological properties of porous
BGCs.
2012 Elsevier B.V. All rights reserved.
Previous studies have shown that crystallization decreases
thelevel of bioactivity and can even turn a BG into an inert
materialKeywords:
shrunk well at ~726 C and melted at over 1050 C, while the onset
shrinking and melting temperaturesAvailable online 2 March 2012
improving the mechanicalIncorporation of B2O3 in CaO-SiO2-P2O5
bioof low-temperature co-red porous glass
Xianyan Yang a, Lei Zhang b, Xiaoyi Chen a, XiaoliangChangyou
Gao a, Zhongru Gou a,a Zhejiang-California International
NanoSystems Institute, Zhejiang University, Hangzhou 3b Rui'an
People's Hospital & the 3rd Hospital Afliated to Wenzhou
Medical College, Ruian
a b s t r a c ta r t i c l e i n f o
Article history:Received 8 November 2011Received in revised form
5 January 2012
Bioactive glass ceramics (BGon their chemical composiboron-rich,
phosphorus-low
j ourna l homepage: www.e lsuidus temperature, highlimit their
applications
x: +86 571 8697 1539.
rights reserved.ctive glass system for improving
strengthramics
n b, Guojing Yang b, Xingzhong Guo a, Hui Yang a,
29, China200, China
have different rates of biodegradation and mechanical properties
dependings and sintering temperatures. The present study was aimed
to develop theaOSiO2P2O5B2O3 bioactive glasses (BG-Bx, X=0, 10, 20)
potentially for
ystalline Solids
i e r .com/ locate / jnoncryso land tissue ingrowth. However,
high sintering temperatures result incoarse-grained microstructure
having poor bioresorption rate faraway from the growth rate of new
bone at the site of implantation.In this respect, the BGs with low
liquidus temperature and highdegradation rate are preferably
low-temperature co-red aids toimprove the mechanical and biological
properties of the BGCs.
It is generally accepted that the crystallization,
microstructure, andeven dissolution rates can be governed by
adjusting the compositions
-
of the BGs and controlling crystallization to suit with their
end appli-cations [1618]. There are several ways to reduce the
sinteringtemperature of advanced ceramics, such as addition of the
lowmeltingglass [19,20]. The melt-quenching calcium borosilicate
glass-ceramics(CaOB2O3SiO2; CBS) with a wollastonite-type main
crystallinephase have to date achieved industrial applications in
wireless com-munication due to its low-temperature co-ring property
[21]. Saranti
1172 X. Yang et al. / Journal of Non-Crystalline Solids 358
(2012) 11711179et al. reported that the boron in the glass network
of CaOB2O3P2O5systemhas a catalytic effect at favoring bioactivity
of the other calciumphosphate glasses [22]. Recently, Lee et al.
found that this ternary CBSsystem was bioactive, and may be
favorable as orthopaedic implantsdue to its low chemical durability
in physiological uids [23,24].More recently, they have developed a
high-phosphorus, low-boronCaOSiO2P2O5B2O3 BGC melted at 1550 C
(with composition(wt.%): CaO 41.8, SiO2 35.8, P2O5 13.9, B2O3 0.5,
CaF2 2.0 and MgO6.0). They reported that the cancellous screws
coated with suchnominal quaternary BGC could bond directly to
cancellous bone toimprove the bone-implant osseointegration [25],
and there was notoxic effect by a subchronic intravenous
administration [26]. However,the melt-quenching involves a high
temperature process whichled to the volatilization of the oxides
with low-melting points, suchas B2O3, it thus becomes difcult to
get the desired composition ofthose oxides.
In this study, we developed a new B2O3-doping SiO2CaOP2O5system
with low melting temperature via solgel process. Accordingto the
literatures about the effect of incorporating boric oxide to
bio-materials, boron has well-dened biological effects such as
stimula-tion of wound healing in vivo, increase of the
extracellular matrixturnover [27], and plays a role in bone
physiology [28]. The purposeof this study is to examine the effect
of B2O3 addition on the thermaland chemical properties of the
SiO2CaOP2O5 system, and gain infor-mation in choosing glass
compositions suitable for low-temperatureco-red BGC
preparation.
2. Experimental procedure
2.1. Materials
Tetraethylorthosilicate (TEOS), triethylphosphate (TEP),
boracicacid (H3BO3), ammonia (NH3H2O, ~28 wt.%), calcium
nitrate(Ca(NO3)24H2O) and absolute ethanol (99.8 wt.%) were
purchasedfrom Sinopharm Chemical Reagent Co. China, as starting
materialsfor preparation of BG-Bx. The high-purity grade NaCl,
NaHCO3, KCl,K2HPO43H2O, MgCl26H2O, CaCl2, and trishydroxymethyl
amino-methane (Tris) as required materials (BBI, Canada) for
preparingsimulated body uid (SBF) and Tris buffer. The
melt-quenching45S5 Bioglass (with similar composition to Biolgass
45S5) has thefollowing chemical composition (wt.%): 45.0 SiO2, 24.5
CaO, 24.5Na2O, and 6 P2O5. Comminution by planetary milling until a
nalparticle size of 210 m was performed.
2.2. Preparation of BG-Bx
The SiO2CaOP2O5B2O3 systems were prepared using the solgel
process. Samples were labeled using the following convention:BGs
named BG-Bx were composed of SiO2, CaO, P2O5, and
B2O3,respectively, with x=0, 10, and 20 mol%. The compositions of
thequaternary systems were listed in Table 1. The content of B2O3
was
Table 1Compositions (mol%) of the as-prepared BG-Bx in the
experiments.
Series B2O3 SiO2 CaO P2O5
BG-B0 0 60.68 35.83 3.49BG-B10 9.98 49.02 39.46 1.54BG-B20 20.00
29.87 48.43 1.70limited at a maximum value of 20 mol%, since it is
detrimental tothe biocompatibility of the highly biodegradable BG.
In a typical pro-cedure for preparing BG-B10, the 0.041 g H3BO3,
2.07 mL TEOS, and0.22 mL TEP were added to 20 mL of ethanol and
stirred for 20 min.Then, 1.404 g Ca(NO3)24H2O and 1.0 mL NH3H2O
were added2.0 mL of deionized water and mixed with 23.0 mL absolute
ethanolunder continuous magnetic stirring for 20 min. The two kinds
ofsolutions were mixed and the suspension was aged at 80 C for 12
hand nally calcined at 600 C for 90 min. To investigate the effect
ofB2O3 content on thermal and chemical properties of BG, the
BG-B20was prepared in the presence of B2O3 while the other
conditionsremained the same. Similarly, as a control, the
boron-free CaOSiO2P2O5 (BG-B0) was prepared in the absence of H3BO3
by thesame method.
2.3. Sintering characteristic evaluation of BG-Bx
In order to understand the effects of glass compositions on
thesintering characteristic, dilatometric analyses were performed
using adilatometer (DIL 402PC, NETZSCH) with a heating rate of 5
Cmin1
in air to characterize the shrinkage of the disc-shaped BG-Bx
compactswith respect to temperature. The discs were prepared using
a uniaxialpressure of 12 MPa to yield the nal dimension of 6 mm in
diameterand 4 mm in thickness, and then pre-heated at 560 C for 2
h. Thethermogravimetric and differential thermal analysis (TG/DTA)
wascarried out on TG/DTA6200 of TA Instruments with a 10 Cmin1
heating rate under an air atmosphere. According to the thermal
analy-sis, powders were subsequently sintered by electric furnace
at 850 Cand 950 C for 2 h, respectively.
2.4. Biological evaluation of BG-Bx in vitro
In vitro apatite formation test was carried out by soaking
BG-Bxpowders and discs in SBF at nal concentration of 1 mgmL1
andmonitoring the formation of HCA on the sample surface at 37 Cfor
different time intervals (3 h14 d). The SBF was prepared
bydissolving inorganic salts reagents in deionized water and
bufferedat pH=7.40 with Tris and 1.0 molL1 HCl at 37 C, according
toKokubo's recipe [29]. The circular BG-Bx discs ( 6 mm2 mm)were
prepared using a uniaxial pressure of 4 MPa and sintered at850 C
and 950 C for 2 h, respectively. After soaking, the discs
wereultrasonically washed in the fresh SBF solution for 3 min, and
thepowder samples were ltered, rinsed with deionized water,
anddried in an oven at 60 C for 24 h before analysis. All the
solutionsafter soaking powder samples were saved for inductively
coupledplasma (ICP; Varian Co., USA) analysis of B, Si, Ca and P to
measureionic concentrations.
2.5. In vitro degradation test
Degradation tests were conducted in two different
simulatedphysiological conditions following the ISO 10993
Biological evalua-tion of biomedical devices Part 14: Identication
and quanticationof degradation products from ceramics using the
as-prepared BG-Bxpowders. The tests were performed at 37 C in SBF,
and in0.05 molL1 Tris buffer at pH 7.4, simulating the body's pH as
afunction of immersion time (up to 28 d). A ratio of 0.50 g
powdersto 50.0 mL solution was used, and the aqueous media were
replacedevery three days. Weight changes were measured by
separating thepowders from the solutions, washing with deionized
water, anddrying at 95 C.
2.6. Fabrication of 45S5/BG-B20 composite BGCs
The 45S5 Bioglass powder was mixed with BG-B20 powder in
two weight proportions (95:5, 90:10) in liquid media (ethanol)
by
-
of thermal decomposition of the nitrate used in the
experiment.Then little weight loss took place up to 800 C. In
general, the BG-Bxmanifested a Tg and at heating crystallized
within the temperatureregion of 800900 C and mostly in two stages,
corresponding to cal-cium borate and calcium silicate
crystallization (Fig. 2c, d). The princi-pal characteristic
difference was the presence of a Tm at ~956 C forBG-B10, and ~952 C
for BG-B20. The liquidus peaks (Tl) were fullyshifted toward lower
temperature in comparison with that of BG-B0(>1050 C), which did
not appear in the curve (Fig. 2b). Accordingto the phase diagram,
the Ca-B-O system demonstrated a full liquidphase below 980 C [31].
It is noted that the temperature of endother-mic peaks (~650750 C)
of BG-B10 and BG-B20 coincided with thetemperature range for the
rapid densication of BG-B10 and BG-B20in Fig. 1b. This coincidence
indicated that endothermic peaks wereattributed to the densication
by liquid phase. Meanwhile, withincreasing B2O3 content, both Tg
and the onset of crystallization peakTc shifted toward lower
temperatures. This clearly corresponds tothe theory that the
network addition is charge balanced resulting inpolymerization of
silicate network and also decrease of the Tg [19,32].
3.3. Phase transformation analysis of the BG-Bx
XRD analysis results for the powders before and after sintering
at850 and 950 C were shown in Fig. 3. The amorphous phases at 600
Cgave an indication of the formation of homogeneous silica
networkin the BG-B0, but the gels with different B2O3 addition
after calciningat 600 C contained trivial crystallites, represented
by a set of low
Fig. 1. Size distribution of the as-prepared BG-Bx particles (a)
and sintering shrinkingcurves for the as-prepared BG-Bx discs
(b).
1173X. Yang et al. / Journal of Non-Crystalline Solids 358
(2012) 11711179planetary milling for 2 h in an agate jar with agate
balls. After dryingthe 45S5/BG-B20 mixture was then mixed
homogeneously withparafn spheres of ~500 m in diameter. The
cylindrical and cuboidBGC green bodieswere prepared using two kinds
of stainless steel dies( 6 mm20 mm or 8 mm10 mm40 mm) and densied
with4 MPa pressure. The green samples were sintered at 700950 C
for2 h, respectively. To understand the mechanical strengths of
the45S5 and BG-B20 individuals, their porous materials were
preparedas controls at the same conditions.
2.7. Sample characterization
The particle size distributionwas determined by laser
granulometryon a netasizer nano (Malvern, S90). The sintered BGCs
and powdersamples before and after high heat treatment, and after
soaking inSBF and Tris buffers were observed using scanning
electric microscopy(SEM; JEM-6700F, Japan), with energy dispersive
X-ray (EDX) analysis.Prior to examination, the samples were coated
with a thin layerof gold. Additionally, the phase of samples was
examined in a RigakuD/max-rA (Geigerex) X-ray diffractometer (XRD)
using Cu Karadiation with a scanning rate of 0.02min1. The
compressivestrengths of the cylindrical porous BGC samples sintered
at 850 Cwere measured using a universal testing machine (Instron,
Canton,MA). The compressive strength and three-point bending
strengthof the long rod-like porous BGCs were determined using a
universaltesting machine (Instron, Canton, MA) with a crosshead
speed of0.5 mmmin1.
3. Results
3.1. Primary characterizations of the BG-Bx powders
In the solgel process, using ethanol/water as phase
separationsolvent, the BG-Bx with nanoscale feature was
synthesized. Accordingto the ICP analysis (Table 1), BG-B0 was a
boron-free, silicon-high BGwith compositions similar to 58S, BG-B20
represented a new boron-rich, silicon-low BG, but BG-B10 was a
boron-containing, moderate-silicon BG. The size distribution of the
BG-Bx was shown in Fig. 1a.The BG-B0 showed a particle size ranging
between 600 and 1250 nm,and an average equivalent diameter of ~930
nm. The BG-B10 showeda particle size distribution ranging from 230
and 1000 nm, the averageequivalent diameter being equal to ~580 nm.
The BG-B20 showed anarrow particle size distribution from 160 to
580 nm, and an averageequivalent diameter of ~330 nm. It suggests
that increasing B2O3/SiO2 ratio result in particle size decrease.
Dilatometeric resultsfor various BG-Bx discs, which indicated the
shrinkage proles versustemperature, were shown in Fig. 1b. The
onset of softening tempera-ture (Ts) was dependent on the addition
of B2O3. The silica-richphosphosilicate BG-B0 caused a rapid
shrinkage of over 10% after726 C. This rapid densication of BG-B0
powder compacts mightresult from viscous sintering which was
observed in silicate glasses[30]. As B2O3 content increased, the
onset of shrinkage moved towardthe lower temperature. Both BG-B10
and BG-B20 had low Ts at 648664 C, which produced a rapid
densication accompanied with79% shrinkage. This change undoubtedly
corresponded to thechange in the nature of bonding in the
structural network.
3.2. Thermal analysis of the BG-Bx
Fig. 2 illustrated the TG/DTA curves for the dried gels. The
glasstransition temperature (Tg), onset of exothermic
crystallization peak(Tc), and onset of endothermic melting peak
(Tm) were determined.On TG curves (Fig. 2a), different stages were
found with the changeof oxide contents, ascribing to the residual
water and ethanol elimina-tion in the systems below 200 C. As the
heating process proceeded,
a steep weight loss stage occurred at 200600 C, which was
because intensity XRD peaks distributed in the glass matrix (Fig.
3a). The
-
1174 X. Yang et al. / Journal of Non-Crystalline Solids 358
(2012) 11711179sample pattern was in agreement with the standard
XRD pattern forwollastonite (-CaSiO3, PDF# 42-0550) in the BG-B0
when sinteringtemperature risen to 850950 C (Fig. 3b, c). However,
BG-B20 was
Fig. 2. TG and DTA curves for the as-prepared BG-Bx precursors.
(a) TG f
Fig. 3. XRD patterns of the BG-Bx before (atransformed into BGC
with two crystalline phases after nucleationand crystallization at
850 or 950 C (Fig. 3d, e). The main phase was-CaSiO3 and the
characteristic diffraction peaks of calcium borate
or BG-Bx, (b) DTA for BG-B0, (c) DTA for BG-B10, (d) DTA for
BG-B20.
) and after (b-e) sintering treatment.
-
(CaB2O4, PDF# 32-0155) began to appear. It must be noted that
nocrystalline phase containing phosphorus in the BGC composition
canbe detected by XRD, mainly due to the limited amount of
phosphateexisting in the glass matrix. We compared the as-sintered
powdersto the documented SiO2CaOB2O3 system prepared by the
solid-state reaction method [33] and detected that there was no
cristobaliteor quartz (SiO2) phase in the sintered BG-B20 (Fig. 3d,
e). It is probablethat powder prepared by solgel route is more
uniform in the micro-scopic eld, resulting in a smaller space
amount molecule adjustedduring crystallization. In addition, the
crystallization peaks ofCaB2O4 became more obvious with the
increase of thermal treatmenttemperature. That is the reason why
there are two exothermic peaksat 800900 C in the DTA curves.
3.4. Formation of HCA on the particle surface
The bioactivity study was performed with the samples calcined
at600 C and those sintered at 850 and 950 C, respectively,
becausethese samples were representative of the different phase
composi-tions of the BGs and BGCs. The unsoaked BG-B0 and BG-B20
powderswere composed of aggregates of several hundreds of
nanometers(Fig. 4a, g). In contrast, the BG-B0 particles softened
and fused afterthermal treatment at 850950 C (Fig. 4c, e). The
surface morphologyof the as-sintered BG-B20 possessed more ceramics
nature withsmoother surface, which contained CaSiO3/CaB2O4 binary
crystallites(Fig. 4i, l). After 168 h the surfaces of BG-B0 and
BG-B20 exhibitedsimilar morphology to those of unsoaked samples,
though the prima-
after sintering at 950 C. Since the crystallization (XRD
patterns) andsurface microstructures (SEM images) of BG-B10 before
and aftersintering and soaking treatment were similar to those of
BG-B20,data of BG-B10 were not shown.
To take into account the possibility of spontaneous HCA
precipita-tion in SBF at physiological temperature, the BGC discs
were alsosuspended and immersed in SBF for 37 days. It can be seen
thatthe as-sintered discs exhibited tough surface with the primary
parti-cle cements due to incomplete sintering (Fig. 5ad). After
soakingfor 3 days, the nanoscale heterogeneous plates or noduses
were de-posited on the surface of the ultrasonically washed discs
(Fig. 5eh).With the increase of soaking time up to 7 days, a
continuous coatinglayer thought to be HCA completely covered the
discs (Fig. 5il).The face-scanning EDX analyses (Fig. 5 il, insets)
conrmed thebiomimetic Ca-decient apatitic characteristic as
observed on theother BG and BGC surface reported previously
[35,36].
3.5. Ionic concentration changes of SBF solutions
Examining the compositional change of SBF with time may pro-vide
some insight into the mechanisms of enhanced bioactivity
andbiodegradability of the BGs and BGCs. Fig. 5 showed the changes
inSi, Ca, B, and P concentrations of the SBF as a function of
soakingtime. As for the calcined BG-B0, the Si and Ca
concentrations in SBFincreased rapidly during the rst 6 h and then
Ca concentration de-creased (Fig. 6a, b). Phosphorus (P)
concentration decreased abruptlyrstly and kept stable after 72 h
(Fig. 6c). In contrast, as for the
1175X. Yang et al. / Journal of Non-Crystalline Solids 358
(2012) 11711179ry irregular particle individuals in the aggregates
converted to ovoidnodules. In the case of the sintered BG-B0 at 950
C, no signs of forma-tion of biomimetic HCA could be found. Only
the microstructures con-sisted of small particles on the crystal
surface (Fig. 4f). The surfacesof the sintered BG-B20 particles
were covered by nanoparticles. Ithas been reported that excessive
crystallinity in BG might defer theformation of HCA when exposed to
SBF [34,35], and thus lead tolower bioactivity. However, the EDX
analysis (Fig. 4, insets) indicatedthat the surface layers were
essentially composed of calcium, phos-phorus, carbon, and oxygen,
with Ca/P molar ratio of 1.521.56 afterimmersion in SBF, indicating
that the material maintained bioactivityFig. 4. SEM images of the
as-prepared BG-B0 (a-f) and BG-B22sintered BG-B0 samples at 850 and
950 C, the Si and Ca concentra-tions of the SBFs soaking gradually
increased during the initial 72 hand P concentration gradually
decreased. Appreciable differencesfor Si and Ca concentrations were
observed at 72 h. For the BG-B20samples, the concentrations of B
and Si had similar trends showinga high increase during the initial
2 h, followed by a slowly decreaseto 168 h (Fig. 6d, e). Calcium
concentration also increased rstlyand then decreased (Fig. 6f).
Obviously, the increase in B, Si and Caconcentrations in the
initial stage and the decrease of P concentrationwere attributed to
the dissolution of particles and the formationof HCA on the surface
of the particles. The concentration of P in the(g-m) particles
before and after soaking in SBF for 7 days.
-
BG-B20 glasses soaking in SBF experienced a decreased and
stabilizedat a certain value after 72 h, similar to that of BG-B0
(data not shown).
particles. As a result, the powders gradually degraded
themselvesand HCA precipitated on the surfaces of the particles,
which may dis-
Fig. 5. SEM images of the as-sintered BG-B0 and BG-B22 discs
before (ad) and after (el) soaking in SBF for 3 and 7 days,
respectively.
1176 X. Yang et al. / Journal of Non-Crystalline Solids 358
(2012) 11711179It must be mentioned that the dissolution in SBF of
the BG-B20 with-out and with sintering treatment caused the
supersaturation of Caions for HCA formation, and the dissolution of
particles hydrated thesurface to provide the HCA nucleation sites.
Through this dissolutionand enhanced HCA formation processes, the
BGC in the CaOSiO2P2O5B2O3 system become highly bioactive as well
as biodegradable.
3.6. Degradation in vitro
During the immersion in SBF, the powder particles were
attackedby the surround environment, and the ions were leached from
theFig. 6. Changes in ionic concentrations in SBF during soaking
BG-B0 (ac) and BG-B20 (turb the weight loss measurement. Thus, the
weight variation in theTris buffer would be more sensitive to
detect the degradation. Fig. 7showed the weight changes of the BG
samples, normalized to thestarting weight of the powders. Then, the
weight loss increased sig-nicantly with the increase of the B
content in the BG-Bx systems inthe Tris buffer (Fig. 7a). In
contrast, the weight of the BG-Bx increasedwhen immersion in SBF,
and BG-B0 showed the highest increase by~45%, due to HCA
precipitation from SBF (Fig. 7b). These results sug-gested that the
B-doping in the CaOSiO2P2O5 system was favorableto improve its
degradation rate. More importantly, it was found fromSEM images and
EDX analysis that the soaked samples were richdf) with and without
sintering treatment. (a) Si, (b) Ca, (c) P, (d) B, (e) Si, (f)
Ca.
-
Fig. 7. Weight variation and SEM images of the soaked BG-Bx
samples in 0.05 molL1 Tris buffer (a, c, d, e) and SBF (b, f, g,
h), respectively. Insets represent face-scanning EDXspectra of the
samples after soaked in Tris buffer and SBF. (c, f) BG-B0, (d, g)
BG-B10, (e, h) BG-B20.
1177X. Yang et al. / Journal of Non-Crystalline Solids 358
(2012) 11711179of silicon (>20 at.%) after soaking in Tris
buffer (Fig. 7ce) and ofCaP (>4.5 at.%) in SBF (Fig. 7fh), which
was the essential featurefor BG-Bx dissolution in Tris buffer and
apatite deposition in SBF.
3.7. Microstructure and strength of BGCs
Fig. 8 showed the SEM images of the cross-section of the
BG-B20-reinforced 45S5 BGCs porous materials; for comparison, SEM
imagesof constructs prepared by 45S5 Bioglass were also shown.
Generally,there was some similarity on the macropores with
spherical mor-phology, resulting from the thermolysis of parafn
microspheres.The interconnective pores and the small pores in the
strut wallscould be observed by SEM. Thus, after low temperature
co-ring pro-cess, the microstructure consisted of solid grains with
solidied liquidnetwork, and possibly residual pores. To regenerate
bone tissue in situFig. 8. SEM images and high-magnication of the
cross-section of BG-B20-reinforced 45S5 BB20=95:5; (c)
45S5/BG-B20=90:10.a partially or fully open porous network is more
desirable, for vascu-lar inltration of the articial scaffold [37].
Moreover, these highlyporous constructs would be advantageous for
degradation in thebody.
As shown in Fig. 9a, the compressive strengths of 5% and 10%
BG-B20-reinforced 45S5 BGC were nearly 1.43-fold and 4.15-fold
thanthat of unreinforced porous 45S5 BGC sintered at 850 C. The
primaryreason for this is that the high temperature softens the
BG-B20 parti-cles, further assisting densication. It can be seen
from the fractureface of the porous samples sintered at 700 C that
the nanoscale BG-B20 particles (~250 nm in size) were embedded by
the 45S5 particles(28 m in size), but those were confused with the
large 45S5 parti-cles after sintering at 850 C (Fig. 9a, insert).
It suggests the BG-B20nanoparticles turn into a viscous liquid to
wet and coalesce the45S5 particles, while there is progressive
microstructure coarseningGC porous materials sintered at 850 C for
2 h. (a) 45S5/BG-B20=100:0; (b) 45S5/BG-
-
ced(b)
1178 X. Yang et al. / Journal of Non-Crystalline Solids 358
(2012) 11711179and bonding to increase rigidity. This densication
can be also con-rmed from the decreased porosity, from ~69.62.2% to
64.22.9%for the porous constructs with increasing BG-B20 content.
Thedecrease of porosity occurred in 10% BG-B20-reinforced 45S5
BGC,in part due to annihilation of small pores, as shown in the
high-magnication SEM image of Fig. 8. Similarly, the exural
strength ofthe BGCs composed of 45S5 and BG-B20 showed an increase
comparedwith 45S5 without doping BG-B20 (Fig. 9b). The long,
rod-like 10%BG-B20-reinforced 45S5 BGC exhibited considerable
bending strength(~3.21 MPa) which was more than four-fold that of
single 45S5 BGC(0.67 MPa). It could be deduced that the addition of
a small amountof low-melting point BG caused a signicant
improvement of themechanical properties of 45S5 BGC.
4. Discussion
Sintering is usually used to form materials which have a
highmelting point. However, this method requires high temperature
andlong duration for heating in order to bond the junctions
betweenground material particles by congelation or diffusion.
Boron-dopedlow-melting co-ring aid, employed in BGCs, makes it
possible toprocess the sintering at lower temperature and short
duration.Some complex systems such as SiO2CaOP2O5Na2OMgOK2OB O and
SiO CaOP O Na OB O Al O have also been investi-
Fig. 9. Compressive strength, porosity (a) and exural strength
(b) of the BG-B20-reinfor(a) Insets represent SEM images of the
porous materials sintered at 700 C and 850 C.2 3 2 2 5 2 2 3 2
3
gated [38,39] for the possibility of manufacturing machinable
high-strength BGC into dental implants. However, high magnesium
oraluminum composition in such systems is still inferior to those
biode-gradable quaternary bioglasses in bone repare already in
existence.Manupriya et al. investigated that P2O5 played an
important rolein controlling chemical durability and bioactivity of
a silica-freeborate glass system (B2O3CaOP2O5Na2O) [40]. O'Donnell
et al.also found that the crystalline phase of biomimetic HCA
occurredrapidly in SBF as the phosphate content increased in the
soda-lime-phosphosilicate glass system. [41] However, it is known
that increas-ing P2O5 contents in BGs increased the apatite phase
in the BGCs aftersintering treatment, and this would reduce the
resorption rate in theBGCs [4244]. Recently, some investigations
has demonstrated thatthe bioactivity and resorption rate of -CaSiO3
ceramics and -CaSiO3/CaB2O4 BGCs were much higher than that of
-tricalciumphosphate and apatite, and these porous materials may
signicantlyenhance the spinal fusion or calvarial defect repair in
rabbit model[23,45]. It suggests that, the boron-rich,
phosphorus-low CaOSiO2P2O5B2O3 system (e.g. BG-B20) may not only be
appropriate biode-gradable materials for bone repair, but also has
signicant implica-tion as low-temperature co-ring aid with respect
to the other BGCs.5. Conclusions
In this study, the boron-rich, phosphorus-low CaOSiO2P2O5B2O3
was prepared and characterized by the solgel method at
con-siderably lower temperatures than required for conventional
meltingmethod. The BG-Bx-derived BGC materials can be sintered at
notmore than 950 C, and at sintering temperature above 800 C
crystal-lization occurred and glass-ceramics with wollastonite and
calciumborate were formed. The melting temperatures of the BG-Bx
werelowered to ~950 C by doping B, and the size of the BG-Bx
particlescould be controlled in the range of 200900 nm.
Crystallizationdecreased the kinetics but did not inhibit the
development of HCAlayer, even in fully crystallized BG-Bx, and the
biodegradation rateof the BG-Bx was based on the boron content.
Such improved thermaland biological properties of boron-rich,
phosphorus-low quaternarysystem should be helpful for lowering the
sintering temperature ofBGC porous constructs potentially for
improving bone regenerationand defect repair in situ.
Acknowledgments
The authors would like to acknowledge the Zhejiang
ProvincialNatural Science Foundation of China (No. Y2090009),
theHealth Bureau
45S5 BG porous materials with different 45S5/BG-B20 ratio
thermally treated at 850 C.Insets represent the cuboid samples and
three-point bending test images.of Zhejiang Province Foundation
(2010SSA005 and 2011C33049) andthe Science and Technology Bureau of
Wenzhou City (H20080039 andH20100076).
References
[1] L.L. Hench, Bioceramics: from concept to clinic, J. Am.
Ceram. Soc. 74 (1991)14871510.
[2] P. Saravanapavan, J.R. Jones, R.S. Pryce, L.L. Hench,
Bioactivity of gel-glass pow-ders in the CaO-SiO2 system: a
comparison with ternary (CaOP2O5SiO2) andquaternary glasses
(SiO2CaOP2O5Na2O), J. Biomed. Mater. Res. A 66 (2003)110119.
[3] M. Vallet-Reg, A.J. Salinas, J. Romn, M. Gil, Effect of
magnesium content on thein vitro bioactivity of CaOMgOSiO2P2O5
solgel glasses, J. Mater. Chem. 9(1999) 515518.
[4] J. Ma, C.Z. Chen, D.G. Wang, Y. Jiao, J.Z. Shi, Effect of
magnesia on the degradabilityand bioactivity of solgel derived
SiO2CaOMgOP2O5 system glasses, ColloidsSurf. B Biointerfaces 81
(2010) 8795.
[5] S. Cai, W. Zhang, G. Xu, W. Zhou, W. Jiang, D. Wang,
Dissolution behavior and bio-activity study of glass ceramic
scaffolds in the system of CaOP2O5Na2OZnOprepared by solgel
technique, Mater. Sci. Eng. C. 30 (2010) 105111.
[6] M.D. O'Donnell, R.G. Hill, Inuence of strontium and the
importance of glasschemistry and structure when designing bioactive
glasses for bone regeneration,Acta Biomater. 6 (2010) 23822385.
[7] P. Sepulveda, J.R. Jones, L.L. Hench, Bioactive solgel foams
for tissue repair,J. Biomed. Mater. Res. 59 (2001) 40348.
-
[8] D.C. Clupper, L.L. Hench, J.J. Mecholsky, Strength and
toughness of tape castbioactive glass 45S5 following heat
treatment, J. Eur. Ceram. Soc. 24 (2004)29292934.
[9] M. Hamadouche, A. Meunier, D.C. Greenspan, C. Blanchat, J.P.
Zhong, G.P.L. Torre,L. Sedel, Long-term in vivo bioactivity and
degradability of bulk solgel bioactiveglasses, J. Biomed. Mater.
Res. 54 (2001) 560566.
[10] H. Arstila, E. Vedel, L. Hupa, M. Hup, Factors affecting
crystallization of bioactiveglasses, J. Eur. Ceram. Soc. 27 (2007)
15431546.
[11] O. Peitl, G. La. Torre, L.L. Hench, Effect of
crystallization on apatite-layer formationof bioactive glass 45S5,
J. Biomed. Mater. Res. 30 (1996) 509516.
[12] A. Hoppe, N.S. Gldal, A.R. Boccaccini, A review of the
biological response to ionicdissolution products from bioactive
glasses and glass-ceramics, Biomaterials 32(2011) 27572774.
[13] V. Cannillo, F. Pierli, S. Sampath, C. Siligardi, Thermal
and physical characterizationof apatite/wollastonite bioactive
glass-ceramics, J. Eur. Ceram. Soc. 29 (2009)611619.
[14] J. Ma, C.Z. Chen, D.G. Wang, X.G. Meng, J.Z. Shi, Inuence
of the sintering temper-ature on the structural feature and
bioactivity of solgel derived CaOP2O5SiO2bioglass, Ceram. Int. 36
(2010) 19111916.
[15] Q.Z. Chen, G.A. Thouas, Fabrication and characterization of
solgel derived 45S5Bioglass ceramic scaffolds, Acta Biomater. 7
(2011) 36163626.
[16] A. Lucas-Girot, F.Z. Mezahi, M. Mami, H. Oudadesse, A.
Harabi, M. Le Floch, Solgelsynthesis of a new composition of
bioactive glass in the quaternary system SiO2CaONa2OP2O5:
comparison with melting method, J. Non-Crsyt. Solids 357(2011)
33223327.
[17] E. Vedel, D. Zhang, H. Arstila, L. Hupa, M. Hupa,
Predicting physical and chemicalproperties of bioactive glasses
from chemical composition. Part 4: Tailoringcompositions with
desired properties, Glass Technol.: Eur. J. Glass Sci.
Technol.,Part A 50 (2008) 916.
[18] V.-V. Vlimki, H.T. Aro, Molecular basis for action of
bioactive glass, Scand. J. Surg.
[26] J.H. Lee, H.S. Ryu, J.H. Seo, B.S. Chang, C.K. Lee, A
90-day intravenous administra-tion toxicity study of
CaOSiO2P2O5B2O3 glass-ceramics (BGS-7) in rat, DrugChem. Toxicol.
33 (2010) 3847.
[27] M. Dzondo-Gadet, R. Mayap-Nzietchueng, K. Hess, P. Nabet,
F. Belleville, B.Dousset, Action of boron at the molecular level
effects on transcription and trans-lation in an acellular system,
Biol. Trace Res. 85 (2002) 2333.
[28] S.S. Hakki, B.S. Bozkurt, E.E. Hakki, Boron regulates
mineralized tissue-associatedproteins in osteoblasts (MC3T3-E1), J.
Trace Elem. Med. Biol. 24 (2010) 243250.
[29] T. Kokubo, H. Kushitani, S. Sakka, T. Kitsugi, T. Yamamuro,
Solutions able to repro-duce in vivo surface-structure changes in
bioactive glass-ceramic A-W, J. Biomed.Mater. Res. 24 (1990)
721734.
[30] J.H. Jean, T.K. Gupta, Liquid-phase sintering in the
glass-cordierite system, J. Mater.Sci. 27 (1992) 15751584.
[31] E.T. Carlson, The system: calcium oxideboric oxide, J. Res.
Natl Bur. Stand. 9 (1932)825832.
[32] M.D. O'Donnell, S.J. Watts, R.V. Law, R.G. Hill, Effect of
P2O5 content in two seriesof soda lime phosphosilicate glasses on
structure and properties Part I: NMR,J. Non-Cryst. Solids 354
(2008) 35543560.
[33] C.C. Chiang, S.F. Wang, Y.R. Wang, Y.F. Hsu,
Characterization of glass-ceramics:thermal and electric properties,
J. Alloy. Comp. 461 (2008) 612616.
[34] P. Li, Q. Yang, F. Zhang, T. Kokubo, The effect of residual
glassy phase in a bioactiveglassceramic on the formation of its
surface apatite layer in vitro, J. Mater. Sci.Mater. Med. 3 (1992)
452456.
[35] O. Peitl, G.P. LaTorre, L.L. Hench, Effect of
crystallization on apatite-layer forma-tion of bioactive glass
45S5, J. Biomed. Mater. Res. 30 (1996) 509514.
[36] M. Mozafari, F. Moztarzadeh, M. Tahriri, Investigation of
the physico-chemicalreactivity of a mesoporous bioactive
SiO2CaOP2O5 glass in simulated bodyuid, J. Non-Cryst. Solids 356
(2010) 14701478.
[37] V. Karageorgious, D. Kaplan, Porosity of 3D biomaterial
scaffolds and osteogenesis,Biomaterials 26 (2005) 54755486.
[38] D. Zhang, E. Vedel, L. Hupa, H.T. Aro, M. Hupa, Predicting
physical and chemical
1179X. Yang et al. / Journal of Non-Crystalline Solids 358
(2012) 11711179mechanisms of the conversion of silicate (45S5),
borate, and borosilicate glassesto hydroxyapatite in dilute
phosphate solutions, J. Mater. Sci. Mater. Med. 17
(2006)583596.
[20] M. Brink, The inuence of alkali and alkaline earths on the
working range forbioactive glasses, J. Biomed. Mater. Res. 36
(1997) 107117.
[21] G.H. Chen, L.J. Tanbg, J. Chen, M.H. Jiang, Synthesis and
characterization ofCBS glass/ceramic composited for LTCC
application, J. Alloy. Comp. 478 (2009)858862.
[22] A. Saranti, I. Koutselas, M.A. Karakassides, Bioactive
glasses in the system CaOB2O3P2O5: preparation, structural study
and in vitro evaluation, J. Non-Cryst.Solids 352 (2006) 390398.
[23] J.H. Lee, C.K. Lee, B.S. Chang, H.S. Ryu, J.H. Seo, K.S.
Hong, H. Kim, In vivo studyof novel biodegradable and
osteoconductive CaOSiO2B2O3 glass-ceramics,J. Biomed. Mater. Res.
77A (2006) 362369.
[24] H.S. Ryu, J.K. Lee, J.H. Seo, H. Kim, K.S. Hong, D.J. Kim,
J.H. Lee, D.H. Lee, B.S. Chang,C.K. Lee, S.S. Chung, Novel
bioactive and biodegradable glass ceramics with highmechanical
strength in the CaOSiO2B2O3 system, J. Biomed. Mater. Res.
68A(2004) 7989.
[25] J.H. Lee, H. Nam, H.S. Ryu, J.H. Seo, B.S. Chang, C.K. Lee,
Bioactive ceramic coatingof cancellous screws improves the
osseointegration in the cancellous bone,J. Orthop. Sci. 16 (2011)
291297.properties of bioactive glasses from chemical composition.
Part 3: In vitro reactivity.Glass, Techn. Euro. J. Glass Sci.
Techn. Part. A. 50 (2008) 18.
[39] O.H. Andersson, G. Liu, K.H. Karlsson, L. Niemi, J.
Miettinen, J. Juhanoja, In vivobehavior of glasses in the
SiO2CaOAl2O3Na2OB2O3P2O5 system, J. Mater.Sci. Mater. Med. 1 (1990)
219227.
[40] K.S. Manupriya, G. Thind, K. Sharma, V. Singh, S. Aravindan
Rajendran, Solubleborate glasses: in vitro analysis, J. Am. Ceram.
Soc. 90 (2006) 467471.
[41] M.D. O'Donnell, S.J. Watts, R.G. Hill, R.V. Law, The effect
of phosphate content onthe bioactivity of soda-lime-phosphosilicate
glasses, J. Mater. Sci. Mater. Med. 20(2009) 16111618.
[42] T. Kitsugi, T. Yamamuro, T. Nakamuro, T. Nakamura, S.
Higashi, Y. Kakutani, K.Hyakuna, S. Ito, T. Kokubo, M. Takagi, T.
Shibuya, Bone bonding behavior of threekinds of apatite containing
glass ceramics, J. Biomed.Mater. Res. 20 (1986) 12951307.
[43] T. Kitsugi, T. Nakamura, T. Yamamura, T. Kokubu, T.
Shibuya, M. Takagi, SEM-EPMAobservation of three types of
apatite-containing glass-ceramics implanted in bone:the variance of
a Ca-P-rich layer, J. Biomed. Mater. Res. 21 (1986) 12551271.
[44] P. Sponer, K. Urban, E. Urbanova, K. Karpas, P.G. Mathew,
Behaviour of nonresob-able bioactive glass-ceramic implanted into
bone defects: comparison withcancellous allografts, Arch. Orthop.
Trauma. Surg. 129 (2009) 13531360.
[45] S. Xu, K. Lin, Z. Wang, J. Chang, L. Wang, J. Lu, C. Ning,
Reconstruction of calvarialdefect of rabbits using porous calcium
silicate bioactive ceramics, Biomaterials 29(2008) 25882596.95
(2006) 95102.[19] W. Huang, D.E. Day, K. Kittiratanapiboon, M.N.
Rahaman, Kinetics and
Incorporation of B2O3 in CaO-SiO2-P2O5 bioactive glass system
for improving strength of low-temperature co-fired porous glass
ceramics1. Introduction2. Experimental procedure2.1. Materials2.2.
Preparation of BG-Bx2.3. Sintering characteristic evaluation of
BG-Bx2.4. Biological evaluation of BG-Bx in vitro2.5. In vitro
degradation test2.6. Fabrication of 45S5/BG-B20 composite BGCs2.7.
Sample characterization
3. Results3.1. Primary characterizations of the BG-Bx
powders3.2. Thermal analysis of the BG-Bx3.3. Phase transformation
analysis of the BG-Bx3.4. Formation of HCA on the particle
surface3.5. Ionic concentration changes of SBF solutions3.6.
Degradation in vitro3.7. Microstructure and strength of BGCs
4. Discussion5. ConclusionsAcknowledgmentsReferences