biomaterials and bioactivity - fzusestak/yyx/biomaterials.pdfpotentially bioactive titanium surface...

Contents of the section:

Biomaterials and bioactivity

BIO-COMPATIBLE CERAMICS AS MIMETIC MATERIAL FOR BONE TISSUE SUBSTITUTION Zdeněk Strnad, Jaroslav Šesták Invited plenary lecture at the conference on inteligent materials, Hawai 2003, from proceedings THERMODYNAMICS OF NON-BRIDGING OXYGEN IN SILICA BIO-COMPATIBLE GLASS-CERAMICS Mimetic material for the bone tissue substitution, N. Koga, Z. Strnad, J. Šesták and J. Strnad Journal of Thermal Analysis and Calorimetry, Vol. 71 (2003). BIO-ACTIVATED TITANIUM SURFACE UTILIZABLE FOR MIMETIC BONE IMPLANTATION IN DENTISTRY— Part III: Surface characteristics and bone–implant contact formation Jakub Strnad, Zdeněk Strnad, Jaroslav Šesták, Karel Urbanec, Journal of Physics and Chemistry of Solids 68 (2007) 841–845 PHYSICO-CHEMICAL PROPERTIES AND HEALING CAPACITY OF POTENTIALLY BIOACTIVE TITANIUM SURFACE J. Strnad, Z. Strnad and J. Šesták Journal of Thermal Analysis and Calorimetry, Vol. 88 (2007) 3,

Bio-Compatible Ceramics as Mimetic Materialfor Bone Tissue Substitution

Zdenek Strnad*, Jaroslav Šesták**

*Laboratory for Glass and Ceramics (LASAK), Papírenská 25, CZ-16000Prague 6, Czech Republic

**Division of Solid-State Physics, Institute of Physics of the Czech Academy ofSciences, Cukrovarnická 10, CZ-16253 Prague 6, Czech Republic

ABSTRACTBone-like apatite formation on the surface of implant is of key importance during the physical andchemical processes leading to the formation of bonds between the implanted material and the newly formedbone tissue. The smartness of such a mimetic process is likely the action of silanole groups (Si-OH) whichserve as the nucleation sites for the biocompatible interface formation capable to coexist between theoriginal tissue and the implants which can be made from ceramics, glass-ceramics, composites as well ascertain metals (titanium) respecting the condition of suitable surface reactivity. Lasak Co.Ltd., is theleading manufacturer of these materials in the Czech Republic and provides various kinds of bioactiveimplants, based on calcium phosphate ceramics, apatite wollastonite glass-ceramics and implants withhydroxyapatite surface coatings, permitting differentiated applications in clinical practice. The bioactivematerials used as bone substitutes are all the subject of continuing research to attain biological, mechanicaland chemical properties as similar as possible to those of the tissue to be replaced – mimetic materials.Clinical applications in orthopaedics, neurosurgery, maxillofacial surgery, auricular surgery, dental surgeryand in other fields are demonstrated.

Keywords: biaoctive implants, bone substitutes, glass-ceramics, hydroxyapatite

INTRODUCTIONDegeneration of the skeletal system in time results in dysfunction of bones, teeth and joints. Extensive bonedefects left after the removal of tumours, infections or as a result of injuries are ideally replaced byautogenic bone tissue. As the amount of this material for the patient is limited and the use of allogenic boneis accompanied by biological, mechanical and also sociological difficulties, there is a great need foralternate material.

Since discovery of Bioglass in 1971 (1), various kinds of bioactive materials have been found and clinicallyused. The uniqueness of surface bioactive materials is their high bioactivity, opening qualitatively newapplication fields, especially for anchoring of the implant in the host tissue, with practical use inorthopedics, stomatology, neurosurgery, oncology, craniofacial surgery and possibly other fields.

LASAK developed and provides three basic kinds of bioactive materials, BAS-O, BAS-HA and BAS-R,permitting differentiated applications in clinical practice. More than 7000 people received these implants astheir bone substitutes during last eight years.

BAS-O GLASS-CERAMICS - BIOACTIVE LONG-TERM STABLE IMPLANTMATERIAL WITH HIGH MECHANICAL STRENGTHBAS-O is an inorganic, polycrystalline material prepared by controlled crystallization of glass, whose maincomponents are CaO, P205, Si02 MgO, and Al2O3 During the crystallization process, the amorphousmaterial is converted to a glass-ceramic material whose main crystalline phases are apatite and wollastonite(2,3). Controlled crystallization permits not only controlled phase conversion during the process, but also

control of the chemical composition and structure of the residual glass phase, which are decisive factors indetermining the bioactivity of the final material (4).

BAS-O exhibits extraordinary biocompatibility, which has been demonstrated in many experiments andclinical tests. The basic condition for the formation of a bond between the BAS-O implant and the livingbone tissue is the formation of a thin layer enriched in Ca and P on the glass-ceramic surface as a result of areaction between the implant and body fluids (Fig,1).

a. b.

Fig. 1. Surface layer formed on a BAS-O implant after exposure in a simulated body fluid for 28 days.a. Cross-section through the surface layer (SEM 1000x),b. Content of elements (P, Ca and SiO2) in the surface layer.

This layer, which is initially amorphous, changes in time to form a polycrystalline layer of apatiteagglomerates, into which are incorporated organic components in the interface zone, produced byosteoblasts, such as collagen fibers, with the formation of a tight bond (Fig. 2).

Fig.2: Detail of the interface of the bone tissue with aBAS-O implant 6 months after implantation(SEM).

Fig.3: Photomicrograph of a section of BAS-0 granuleimplant in bone, 2 years after implantation(toluidine blue stain).

Thus, the implant is not considered to be a foreign body; on the contrary, a strong bond is formed directlybetween the implant and the bone tissue, without an intermediate layer of soft tissue, in a time period of 4-8weeks after implantation (Fig. 3).

BAS-O exhibits intense osteoconductive properties and also the ability to form bonds between theindividual BAS-O particles in a body fluid medium. BAS-O is a white material with an apparent density of3000-3100 kg/m3. The material has a similar bending strength to the cortical bone, 170 MPA, andapproximately double the compressive strength, > 400 MPa. The strength of the junction of the BAS-Oimplant with bone tissue, measured by the push-out test (with shear stress) after implantation for 2 monthsis 15-20 MPa.

Clinical applicationBAS-0 granules and ground material are used to fill cysts, defects left by injuries, defects left byexcochleation of benign tumors, and to reconstruct extensive acetabular defects (Fig. 4). Compact, wedge-shaped blocks (with various heights and surfaces) can be used, e.g., for condyl elevation. Individuallyshaped implants can be used in neurosurgery to cover defects left from cranial trepanation and as onlays inplastic surgery.

a. b.Fig. 4: Reconstruction of extensive acetabular defects by bioactive glass-ceramics BAS-0 in reoperation of

total endoprotheses. a. Schematic drawing of implantation. b. Radiogram taken 8 months post-operatively. [5]

The special shape of intervertebral prostheses have been developed and successfully used in surgery of spin(Fig. 5). In cranio-facial surgery, the material can be used as plates, blocks, or individually shaped implantsto replace bone defects, rebuilding of orbit, for reconstruction of partial mandibular defects. It can be usedto enlarge the mandible or for plastic chin and nose profiles.

BAS-O granule

a. b.

Fig.5: a. BAS-0 intervertebral prostheses. b. Radiogram taken post-operatively [6].

BAS-HA - BIOACTIVE NONRESORBABLE IMPLANT MATERIAL BASED ONHYDROXYAPATITE CA10(P04)6(OH)2Hydroxyapatite is synthesized from aqueous solutions under precisely defined pH, temperature and otherphysical parameters, which ensure reproducible preparation of a highly pure, crystallographically definedproduct, which does not contain any unwanted calcium phosphates. This product is further processed toyield the final BAS-HA product with defined biophysical properties., Its structure and composition aresimilar to bio-apatite, which is the main inorganic component of living bone tissue. Implants form a strong:bond between the bone tissue and the implant material without an intermediate fibrous layer (Fig. 6).

Fig.6: Direct contact between the BAS-HA implant and bone tissue, 2 months after implantation(thin-section, toluidine blue stain)

BAS-HA exhibits high biocompatibility, which has been demonstrated in many preclinical and clinicaltests, including tests of cytotoxicity, carcinogenesis and mutagenic effects. The material exhibitsosteoconductive properties. After implantation in the defective part of the bone, bone tissue is newly

formed in the space between the granules of this substance. A complex of artificial substances and livingbone tissue is formed.

BAS-HA is a very dense ceramic with apparent porosity of 1.7 %. The Ca/P molar ratio is 1.66. Thematerial exhibits a bending strength of 60 MPa and compression strength of 200 MPa. The strength of thejunction with the bone tissue measured by the push-out test (shear stress) is equal to 19 MPa two monthsafter implantation and 29 MPa 4 months after implantation.

Clinical applicationBAS-HA material is designed for bone grafting especially at sites where only compressive forces areexpected to act on the implant. It can be used in dentoalveolar surgery to fill bone defects left afterextirpation of cysts, surgical extractions, or to remodel the alveolar ridge. In paradontology, it can be usedto treat bone paradontological defects. Bioactive BAS-HA material is also used for production of middleear implants (Fig. 7) and dental implants with hydroxyapatite coating (Fig. 8) [7, 8].

Fig.7: Middle ear implants made of BAS-HA. Fig.8: Dental implant (Impladent)

with hydroxyapatite coating.

BAS-R - BIOACTIVE RESORBABLE IMPLANT MATERIAL BASED ONTRICALCIUM PHOSPHATE CA3(P04)2BAS-R is a surface bioactive, resorbable, inorganic, crystalline material based on tricalcium phosphate(β-TCP) .The material is prepared by a special procedure at high temperature by melting and controlledcooling of the melt. BAS-R forms direct bonds with living bone tissue without forming a fibrous interlayer.

Fig. 9. Gradual resorption of granules ( - - - ) of BAS-R and simultaneous formation of new bone tissueat the edges of the granules; 8 months after implantation (thin-section, toluidine blue stain)

The material greatly stimulates formation of new bone tissue and has osteoconductive properties. Thematerial gradually disintegrates in the body as .a consequence of hydrolytic corrosion and activephagocytosis, accompanied by resorption and replacement with newly formed bone tissue (Fig. 9) BAS-Ris white in color and has an apparent density of 2900-3100 kg/m3

Clinical applicationIt is designed for bone replacement where resorption is required, with gradual replacement by living bonetissue. It is used in paradentology and in dentoalveolar surgery to treat bone defects [10] (see Fig.10).

Fig.10: Filling of bone defects left by tooth extraction and extirpation of a radicular cyst a) prior tothe operation b) after application of BAS-R

OUTLOOKToday bioceramics are used in a broad field of devices inside the human body. This is mainly due to theirgood biocompatibility. Among the ceramic materials used for bone replacement, bioactive ceramics appearparticularly promising because of their ability to form stable interface with living host tissue. The majorproblem of these materials, which inhibits their application on several types of implants, is their poor matchof mechanical behavior of the implant with the tissue to be replaced. Principally, the coating, composites,porous structured materials and/or new resorbable materials are promising ways for the next development.

However, no one has succeeded todate, in finding a material, which fully corresponds to bone or otherliving parts of the body. It is the task of a growing number of researchers and institutions working in thefield of biomaterials to further improve performance of these materials. Nature is still the better engineer.

REFERENCES1. L.L.Hench, R.S.Splinter, W.S.Allen, 1971. Bonding mechanisms at the interface of ceramic prosthetic

materials, J.Biomed,Res.Symp. 2, 117.2. Z.Strnad, 1986. Glass-Ceramic Materials/Liquid Phase Separation, Nucleation and Crystallization in

Glasses, Elsevier, Amsterdam.3. Z.Strnad, K.Urban, 1989. Surface Bioactive Glass-Ceramic Materials, Sklár a keramik, 39, 292.4. Z.Strnad, 1992. Role of the Glass Phase in Bioactive glass-ceramics, Biomaterials, 13(5), 317.5. K.Urban, P.Šponer, 1998. Reconstruction of Extensive Acetabular Defects by Bioactive Glass Ceramics

in Re-operations of Total Endoprostheses, Acta Chir.Orthop.Traum.Cech., 65, 17.6. M.Filip, P.Veselský, Z.Strnad, P.Laník, 1995. The Replacement of the Intervertebral Disc by Ceramic

Prosthesis in Treatment of Degenerative Diseases of the Spine, Acta Chir.Ortho.Traum. Cech., 62, 226.7. A. Šimunek, A. Štepánek, V.Zábrodský, Z.Nathanský, Z.Strnad, 1997. A 3-year Multicenter Study on

Osseointegrated Implants- Impladent, Quintessenz 6(3).8. Z.Strnad, J. Strnad, M.Psotová, C.Povýšil, K. Urban, 1998. The Osseoconductive Ability of

Plasmatically Deposited Hydroxyapatite and Pure Ti in Vitro and in Vivo, Quintessenz, 7, 5.9. K.Urban, Z.Strnad, C.Povýšil, P.Šponer, 1996. Tricalcium phosphate as a bone tissue substitute, Acta

Chir.Orthop.Traum.Cech, 63, 16.10. V.Pávek, Z.Novák, Z.Strnad, D.Kudrnová, 1994. Clinical Applications of Bioactive Glass-ceramics

BAS-O for Filling Cyst Cavities in Stomatology, Biomaterials, 15(5), 353.

Journal of Thermal Analysis and Calorimetry, Vol. 71 (2003)

THERMODYNAMICS OF NON-BRIDGING OXYGEN INSILICA BIO-COMPATIBLE GLASS-CERAMICSMimetic material for the bone tissue substitution

N. Koga1, Z. Strnad2, J. Šesták3* and J. Strnad 2

1Chemistry Laboratory, Department of Science Education, Graduate School of Education,Hiroshima University, 1-1-1 Kagamiyama, Higashi-Hiroshima 739-8524, Japan2Laboratory for Glass and Ceramics (LASAK), Papírenská 25, CZ-16000 Praha 6, Czech Republic3Institute of Physics of the Academy of Sciences, Cukrovarnicka 10, CZ-16253 Praha and Instituteof Interdisciplinary Studies, te West Bohemian University, Husova 17, CZ-30114 Pilsen,Czech Republic

Abstract

Correlations between the structural properties of Na2O–CaO–SiO2 glasses characterized by the ac-tivity of oxygen ions and the bioactivity were examined by comparing the compositional de-pendence of the structural parameters calculated on the basis of a thermodynamic consideration withthat of the bioactivity. A simple model of characterizing the glass structure by considering the bridg-ing and non-bridging oxygen ions was employed as the first step for this purpose. Further detailedthermodynamic analysis on the anionic constitution in the glass was performed and thecompositional dependences of the relative proportions of bridging, non-bridging and free oxygenions were calculated. The bioactive region corresponded to the compositional region characterizedby the higher relative proportion of non-bridging oxygen ions with co-existing an appreciable con-centration of bridging oxygen ions, suggesting a possible important role of the non-bridging oxygenions on the surface chemical process of bone-like apatite layer formation.

Kevwords: bio-compatible, bone-like apatite, glass-ceramics, mimetic material, thermodynamics

Introduction

Degeneration of a human skeletal system in time results in dysfunction of bones,teeth and joints. Extensive bone defects, left after the removal of tumors, infections oras a result of injuries, are ideally replaced by autogenous bone tissue. As the amountof this material for the patient is limited and the use of allogenic bone is accompaniedby biological, mechanical and also sociological difficulties, there is a great need foralternate non-human synthetic sources.

Merely four decades ago it was considered inconceivable that a man-made materialcould bond to living tissues in view of the deep-rooted experience that it would result in a

1418–2874/2003/ $ 20.00

© 2003 Akadémiai Kiadó, Budapest

Akadémiai Kiadó, Budapest

Kluwer Academic Publishers, Dordrecht

Yariv04

* Author for correspondence: E-mail: [email protected]

foreign body reaction and the formation of non-adherent scar tissue at the interface withinserted material. This understanding was irreversibly altered when a special composi-tion of soda-lime-phosphate-silica glass was synthesized by Hench [1] and successfullyimplanted in the femurs of rats. About 6% of P2O5 was added to simulate the Ca/P con-stituents of hydroxyapatite, Ca10(PO4)6(OH)2, that is the inorganic mineral phase naturallyexisting in bones. Therefore the bone-like apatite formation on the surface of implant isof a key importance during the physical and chemical processes leading to the formationof an enough firm connection between the implanted material and the newly formed bonetissue [2–6]. The bioactivity leads to both the osteoconduction and osteoproduction as aconsequence of rapid reaction on the bioactive glass surface. The surface reac-tions [7–16] involve ionic dissolution of calcium and sodium ions, phosphates andhydrated silica that give rise to both the intercellular and extracellular responses at the in-terfaces of the glass with its physiological environment. The smartness of such a mimeticprocess is likely hidden in the activity of oxygen characterized by the action of silanolegroups (Si–OH). They likely serve as the nucleation sites for the bio-compatible interfaceformation capable to coexist between the original tissue and the implants which can beexpediently made from glass, glass-ceramics, ceramics, cements and other composites aswell as from certainly treated metals (etched titanium) respecting the set-in condition ofits suitable surface reactivity.

We have taken part in the research progress of bioactive materials since early eight-ies [17–22]. This matured in their actual appliance in practical implantology under thetrademark IMPLADENT- (a system for oral implantology produced by LASAK, Co.

J. Therm. Anal. Cal., 71, 2003

2 KOGA et al.: BIO-COMPATIBLE GLASS-CERAMICS

Fig. 1 Surface layer formed on a BAS-O implant after exposure in a simulated bodyfluid for 28 days (a) and the cross section of the actual body implant (c):a – View of the surface layer formed on the BAS-O material after exposure inSBF (SEM 1000×, fracture), b – Content of elements (P, Ca and SiO2) in thesurface layer, and c – Detail of the interface of the bone tissue with a BAS-Oimplant 6 months after implantation (SEM 1200×)

Ltd) and BAS-O, BAS-HA, BAS-R-bioactive bone tissue substitutes [23]. Figure 1 illus-trates the bone-bonding ability of BAS-O which is based on inorganic, polycrystallinematerial prepared by controlled crystallization of glass, whose main components areCaO, P2O5, SiO2 and MgO. During the crystallization process, the glassy material is con-verted to a glass-ceramic material whose main crystalline phases are apatite andwollastonite. BAS-O granules and ground material are used to fill cysts, defects left byinjuries, defects left by excochleation of benign tumors, and can be of help to reconstructextensive acetabular defects. Compact, wedge-shaped blocks (with various heights andsurfaces) became useful for, e.g., condyl elevation. Individually shaped implants can beused in neurosurgery to cover defects left from cranial trepanation and as onlays in plasticsurgery. In Fig. 2, there is revealed another case of biomaterial BAS-HA (hydroxy-apatite) which is synthesized from aqueous solutions under precisely defined pH, temper-ature and other physical parameters, which ensure reproducible preparation of a highlypure, crystallographically defined product, which does not contain any unwanted calciumphosphates. Its structure and composition are similar to bio-apatite, which is the main in-organic component of living bone tissue. Implants form a strong bond between the bonetissue and the implant material without any intermediate fibrous layer. Final product isthe BAS-R, which is the surface bioactive, resorbable, inorganic, crystalline materialbased on tricalcium phosphate.

Bioactivity has since attracted increased attention being aimed to further molecularmanipulation (doping surfactants, micro-additives of various organic molecules such asproteins, glyco-proteins and polysaccharides, useful in easier mine-realization) which in-telligent response by host organism is evaluated in order to achieve well-tailored im-plants. Bio-glass-ceramics that activate genes offer the possibilities of repairing, or per-haps even preventing, many disease states, such as osteoporosis, in which a large fractionof women lose a substantial amount of bone mass as they age. They can be also used as asecond phase in a composite that mimics the structure and properties of bone. In future


KOGA et al.: BIO-COMPATIBLE GLASS-CERAMICS 3

Fig. 2 a – Direct contact between the BAS-HA implant and bone tissue, 2 months afterimplantation. BAS-HA material is a very dense ceramics with apparent porosity of1.7%. The Ca/P molar ratio is 1.66. The material exhibits a bending strength of60 MPa and compression strength of 200 MPa. The strength of the junction withthe bone tissue measured by the push-out test (shear stress) is equal to 19 MPa(two months after implantation) and 29 MPa (4 months after implantation).b – Dental implant (‘Impladent’) with hydroxyapatite coating

the implication of glass activation of genes it may be possible to design therapeutic treat-ments or food additives that will inhibit the deterioration of connective tissues with age.Further understanding bioactivity may even help in better perception of the creation oflife [24–26]. It shows a great variability in the application of different glasses and amor-phous materials in many fields of human activities [23, 27–31].

Structural parameters of the soda-lime-silica glasses andbioactivity

It is generally accepted that the bioactivity of glass and glass-ceramics is closely con-nected with the surface chemical reactions of formation of a bone-like apatite layerthat bonds to the bone [2–6]. Very important is systematic understanding on mecha-nisms and kinetics of the reactions that occur at the surface of glassy implants of sil-ica-specialized composition. The early stages of the bone-like apatite formation on asurface of bioactive glass have been studied systematically [7–16] and summarizedas follows [16]. The alkali ions in the bioactive glass are rapidly exchanged by hydro-gen ions in the surrounding solution, e.g., body or simulated body fluids (SBF), andthe network dissolution rapidly reduce the amount of the Si–O–Si and Si–O–Camodes and replace them with Si–OH bonds at the glass-solution interface. Singlenon-bridging oxygen modes of Si–OH are then gradually replaced by more sphericalOH–Si–OH and the condensation and repolymerization of a SiO2-rich layer takeplace on the surface. An amorphous CaO–P2O5-rich film is produced on the top of theSiO2-rich layer by incorporating soluble calcium and phosphate from solution. Thecharacteristic double layer composed of a SiO2-rich layer and a mixed crystallinelayer of hydroxy carbonate apatite and/or hydroxyl fluorapatite results from the crys-tallization of the amorphous CaO–P2O5-rich film by incorporating OH–, CO3

2– , or F–

anions from the surrounding solution.The rates of consecutive and/or concurrent chemical processes of bone-like apatite

formation on the glass surface depend largely on the composition of glass or glass-ceramics and on the associated physico-chemical properties of these materials such assolubility, volumetric ratios of glass phase and so on [7–16]. An evaluation of the correla-tion of the bioactivity with the structure of glasses is thus of interest, as have been thecases of the compositional dependence of all physico-chemical properties of glass, suchas viscosity, electrical properties, immiscibility, glass formation region, nucleation, crys-tallization, phase separation, and so on [23, 27, 28, 32–40]. The constitution of bridgingand non-bridging oxygen ions in glasses [23] is one of the most frequently used conceptfor characterizing the glass structure. Various physico-chemical properties have beencorrelated to the chemical composition in view of the thermodynamic state of oxygenions. Although there are various different theories on evaluating the constitution of bridg-ing and non-bridging oxygen ions in glasses [41–45], earlier we evaluated the state of ox-ygen ions in an iron-rich borate glasses [46] according to the model proposed by Toopand Samis [42]. In the present case of bioactivity, the state of oxygen ions is also likelyimportant in relation to the chemical processes of the early stage of the bone-like apatite



formation, especially in the dissolution process of Si–O–Si mode with bridging oxygenand intermediate formation of Si–OH with non-bridging oxygen. Previously, one of thepresent authors was evaluated the correlation of the bioactivity of Na2O–CaO–SiO2–P2O5

and CaO–MgO–SiO2–P2O5 glass-ceramic systems with the structural parameter relatedto the bridging oxygen [17] using the Stevels model [41].

The Stevels’s parameters [17, 41], i.e., X and Y, can be correlated to the meannumber of non-bridging (O–) and bridging (Oo) oxygen ions per polyhedron in theglass lattice, respectively, and calculated from the molar composition of glass accord-ing to the following equations.

X=2R–Z

Y=2Z–2R

where Z is the mean number of all types of oxygen ions per polyhedron, i.e., the mean co-ordination number of the glass-forming cations and R is the ratio of the total number ofoxygen ions to the total number of glass-forming cations in glass. The parameters (X, Y)vary from (0, 4) for, e.g., the pure silica glass, to (1, 3) for, e.g., Na2O�2SiO2 glass, and to(2, 2) for, e.g., CaO�Na 2O�2SiO2 glass. When X>2 and Y<2, the glasses are called as in-vert glasses. Having the same values of (X, Y), these glasses are characterized as structur-ally similar.

In the previous work on the Na2O–CaO–SiO2–P2O5 and CaO–MgO–SiO2–P2O5

glass-ceramic systems [17], the bioactivities of these systems evaluated by in vitrotest of mutual bonding after soaking in SBF and by in vivo test of implantation in dogtibia were correlated to one of the Stevels’s parameter Y, i.e., mean number of bridg-ing oxygen ions. It was found that the Y value of the residual glass phase in theglass-ceramic system to be close to 2 is the suitable condition for the higher bio-activity. When Y>3, the glass loses its bioactivity.

In the present study, the attention was turned to the non-bridging oxygen ions in theglass as expressed by the Stevels’s parameter X. In order to correlate the structural pa-rameter X of a glass with its bioactivity, we focused on an empirical index of bioactivity,



Fig. 3 Bioactive index IB as a function of mean number of non-bridging oxygen ions X,calculated using the data of compositional dependence of IB in the soda-lime--silica system containing 6 mass% P2O5 reported by Hench [2, 16, 47, 48]

IB, introduced by Hench [47] as IB=100/t0.5bb, where t0.5bb is the time for more than 50% ofthe implant interface to be bonded to bone. The compositional dependence of IB in theNa2O–CaO–SiO2 system containing 6 mass% P2O5 has been clearly represented byHench [2, 16, 47, 48] and employed usefully to discuss the bioactivity of the glass with acertain composition. By reading the values of IB at various typical points of glass compo-sition from the reported Iso IB plots in the Na2O–CaO–SiO2 ternary system [2, 16, 47, 48],we calculated the structural parameters (X, Y) from the molar composition and relationsof these structural parameters and the value of IB were examined. Figure 3 shows the cor-relation between the calculated structural parameter X and the value of IB for theNa2O–CaO–SiO2 ternary system containing 6 mass% P2O5. When X<1.5 (Y>2.5), theglass looses its bioactivity and bioactivity index, IB~0. The value of X>1.5 (and Y<2.5) in-dicates the range of bioactive glasses, IB>0. Essentially a linear dependence with negativeintercept has been found between the mean number of non-bridging oxygen ions andbioactivity index.

Anionic constitution in the soda-lime-silica glass-forming meltsand bioactivity

Since the Stevels model considers only non-bridging and bridging oxygen ions, itcannot hold in the more basic regime (X>>2) where free oxygen ions reach an appre-ciable concentration. Therefore, after Fincham and Richardson [49], one can proposethe dominant reaction in the formation of silicate solution (glass): Oo+O2–=2O– whereO2– refers to the free oxygen ions. The compositional dependence of relative propor-tion of non-bridging oxygen ions [O–] in the soda-lime-silica system has been alsocalculated applying methods currently used in the chemistry of organic polymers.

According to Masson [43–45], the following assumptions were applied to derivethe anionic distribution in M2O–SiO2 or MO–SiO2 glass forming melts:

1) The silicate ions are presented exclusively as linear and branched chains of generalformula: Si Ox 3x+1

(2x+2)– . These species may arise by the poly-condensation reactions ex-

pressed generally according to the following equation:

Si O Si O Si Ox 3x+1

(2x+2)–

y 3y+1

(2y+2)–

x+y 3(x+y)+1

(2x+� � 2y+2)– 2–O� (1)

2) The equilibrium constant kxy of Eq. (1) can be approximated using that for the low-est k-members k11, i.e., x=1 and y=1.3) The Eq. (1) may be written in a more general form:

2O–= Oo+ O2– (2)

where O–, Oo and O2– are the non-bridging oxygen, bridging oxygen and free oxygenions, respectively.4) According to the Temkin’s equation [50], the activity of M2O or MO oxides in theM2O–SiO2 or MO–SiO2 binary melt aMO is equated to the ion fraction of free oxideion N

O2 – .



a NMO O2 –� (3)

5) For linear and branched chains, the aMO has the following relation with respect tothe mole fraction of SiO2, i.e., for X SiO2

it follows

12

1

1

3

13

111

X aa

kSiO MO

MO2

� �

��

��

�

�

––

–

(4)

6) Knowing the value of k11 for the binary system M2O–SiO2 or MO–SiO2, the ionfraction Nx of any silicate ion Si Ox 3x+1

(2x+2)– is given by the following equation:

Nx

x x a

k a

x

MO

MO

x–1

��

�

�

��

�

��

( )!

( )! !

( – )

3

2 1

1

13

1

1

11

11

3

111�

�

��

�

��

k a

a

a( – )

( – )MO

MO

2x+1

MO (5)

7) The ion fraction of non-bridging oxide NO– in Eq. (2) is approximated by the sum-

mation of Nx using x up to 50.

N NO x

x=1

– ��50

(6)

8) The ion fraction of bridging oxide NOo is then obtained by

N N NO O Oo 2 – –�1– – (7)

Taking the literature values for k11=1.6�10–3 and =8�10–8 for CaO–SiO2 andNa2O–SiO2 binary melts [51], respectively, the X SiO2

dependence of aMO can be calcu-lated according to Eq. (4). Figure 4 shows the compositional dependence of aMO forCaO–SiO2 and Na2O–SiO2 systems. The ion fraction Nx is obtained according to Eq. (5).The calculated ion fractions Nx for CaO–SiO2 and Na2O–SiO2 binary melts are repre-sented in Fig. 5 as a function of mass fraction of SiO2. Using the compositionaldependences of the activity of oxides, i.e., N

O2 – , shown in Fig. 4 and the relative propor-



Fig. 4 The dependence of activity of sodium and calcium oxides, aMO, on the massfraction of SiO2 in the binary CaO-SiO2 and Na2O-SiO2 systems

tions of O– obtained by summing up Nx from x=1 to 50, the relative fraction of bridgingoxygen ions N

Oo at a mass fraction of SiO2 was calculated according to Eq. (7). Figure 6shows the compositional dependences of the relative fractions of Oo, O– and O2– in the bi-nary CaO–SiO2 and Na2O–SiO2 systems. According to the conventional pseudo-binaryassumption [52], the compositional dependence of the relative proportion of O– inNa2O–CaO–SiO2 ternary melts was obtained using the values for the binary melts shownin Fig. 6. The compositional dependence (in mass fraction) of the calculated relative pro-portion of O– in the Na2O–CaO–SiO2 ternary melts is shown in Fig. 7. These findingswere compared with the compositional dependence of IB for the system containing



Fig. 5 The dependence of ion fraction Nx of silicate ions Si Ox 3x+1

(2x+2)– on the mass of SiO2

in the binary a – CaO–SiO2 and b – Na2O–SiO2 systems

Fig. 6 The relative proportions of non-bridging oxygen ions O–, bridging oxygen ions Oo

and free oxygen ions O2– in the binary a – CaO–SiO2 and b – Na2O–SiO2 systems,calculated vs. the mass fraction of SiO2 using the relations in Eqs (3), (6) and (7)

6 mass% P2O5 [2, 16, 47, 48]. The region of the remarkable high bioactivity IB>8 can beobserved in the compositional region characterized by the higher relative proportion ofnon-bridging oxygen ions >0.8 in the side of SiO2-rich composition where an appreciableconcentration of bridging oxygen ions exist, Fig. 7.

The ability to form the bone-like apatite layer after soaking P2O5-freeNa2O–CaO–SiO2 glasses in SBF has been investigated experimentally by Kim et al. [53].It was shown that the compositional region of the apatite layer formation on the surfaceof the P2O5-free Na2O–CaO–SiO2 glasses corresponds closely to the region of IB>0 forthe Na2O–CaO–SiO2 glasses containing 6 mass% P2O5. It was also pointed out that therates of apatite formation of P2O5-containing Bioglass 45S5-type and a correspondingP2O5-free Na2O–CaO–SiO2 glass are comparable. Those findings indicate that thebioactivity of the glass characterized by the bone-like apatite formation depends largelyon the properties of the basic P2O5-free Na2O–CaO–SiO2 ternary glass. From the presentresults, it is predicted that a close connection exists between the structural property of theglass characterized by the activity of oxygen ions and its bioactivity. The special notice isconcerning the proportion of non-bridging oxygen ions in the glasses, where the region ofremarkable high bioactivity is observed for the compositional region with the higher rela-tive proportion of non-bridging oxygen ions >0.8, indicating a possible important role ofthe non-bridging oxygen ions during the course of the surface chemical processes ofbone-like apatite layer formation.



Fig. 7 A comparison of the relative proportion of non-bridging oxygen ions O- in theNa2O–CaO–SiO2 ternary system calculated in the present study with the Iso IB

plots for the system with 6 mass% P2O5 reported by Hench [2, 16, 47, 48]

Conclusions

A close correlation between the structural properties of the Na2O–CaO–SiO2 glassesand its bioactivity was found. Using the structural parameters X and Y related, respec-tively, to the mean number of non-bridging and bridging oxygen ions, the bioactiveregion can be correlated to the structural parameters X>1.5 and Y<2.5. Further de-tailed analysis of the anionic constitution in the glass system indicated that the re-markable high bioactive region is seen in the compositional region of higher relativeproportion of non-bridging oxygen ions with co-existing an appreciable concentra-tion of bridging oxygen ions and no significant free oxygen ions.

The structural correlations of the bioactivity shown in the present study can beused as a useful tool for designing the bioactive glass and for tailoring bioactiveglass-ceramics and composites with a glass-ceramic matrix. At the same time, a pos-sible important role of the non-bridging oxygen ions on the surface chemical pro-cesses of the bone-like apatite layer formation is expected from the close correlationof the relative proportion with the bioactivity.

* * *

The authors express thanks and gratitude to the Grant Agency of the Academy of Sciences of theCzech Republic for the project financing. The grant A 4010101 is especially acknowledged as wellas the project support by the dept. of industry of the Czech Republic under the number: FB-CV/64.

References

1 L. L. Hench, R. J. Splitner, W. C. Allen and T. K. Greenlee, J. Biomed. Mater. Res., 2 (1971) 117.2 L. L. Hench, J. Am. Ceram. Soc., 74 (1991) 1487.3 T. Kokubo, Biomater., 12 (1991) 155.4 L. L. Hench and J. Wilson, Introduction to Bioceramics, World Sci. Publ., London 1993.5 L. L. Hench and J. K. West, Life chemistry reports, 13 (1996) 187.6 L. L. Hench, J. Am. Ceram. Soc., 81 (1998) 1705.7 L. L. Hench and H. A. Paschal, J. Biomed. Mater. Res., 4 (1973) 25; 5 (1974) 49.8 T. Kokubo, S. Ito, S. Sakka and T. Yamamuro, J. Mater. Sci., 20 (1985) 2001; 21 (1986) 536.9 W. Holand, W. Vogel, K. Neumann and J. Gummel, J. Biomed. Mater. Res., 19 (1985) 303.

10 A. E. Clark and L. L. Hench, J. Non-Cryst. Solids, 113 (1989) 195.11 L. L. Hench and J. K. West, Chem. Rev., 90 (1990) 33.12 T. Kokubo, J. Non-Cryst. Solids, 120 (1990) 138.13 O. H. Andersson, K. H. Karlsson and K. Kangasniemi, J. Non-Cryst. Solids, 119 (1990) 290.14 T. Kokubo, Thermochim. Acta, 280/281 (1996) 479.15 W. Cao and L. L. Hench, Ceramics Int., 22 (1996) 493.16 O. Peitl, E. D. Zanotto and L. L. Hench, J. Non-Cryst. Solids, 292 (2001) 115.17 Z. Strnad, Biomaterials, 13 (1992) 317.18 Z. Strnad and J. Šesták, ‘Bio-compatible ceramics’ in the Proceedings of the 3rd Inter. Conf.

Intelligent Processing and Manufacturing of Materials, (J. Meel, Ed.) Vancouver Univ., p. 123.19 A. Šimunek, A. Štepánek, V. Zábrodský, Z. Nathanský and Z. Strnad, Quintessenz, 6 (1997) 3.20 Z. Strnad, J. Strnad, M. Psotová, C. Povýšil and K. Urban, Quintessenz, 7 (1998) 5.



21 K. Urban, Z. Strnad, C. Povýšil and P. Sponer, Acta Chir. Orthop. Traum. Czech.63 (1998) 16.

22 J. Strnad, A. Helebrant and J. Hamáèková, Glastech. Ber. Glass Sci. Tech., 73C (2000) 262.23 Z. Strnad, Glass-ceramic materials; Liquid Phase Separation, Nucleation and Crystallization in

Glasses, Elsevier, Amsterdam 1986.24 L. L. Hench, Glastech. Ber. Glass Sci. Tech., 70C (1997) 439.25 L. L. Hench, J. Biomed Mater. Res., 23 (1989) 685.26 L. L. Hench and J. K. West, J. Mater. Sci., 29 (1994) 3601.27 J. Šesták (Ed.), Vitrification, Transformation and Crystallization of Glasses, special issue of

Thermochim. Acta, Vol. 280/281, Elsevier, Amsterdam 1996.28 J. Šesták, in Z. Chvoj, J. Šesták and A. Triska (Eds), Kinetic Phase Diagrams; Non-equilibrium

Phase Transformations, Elsevier, Amsterdam 1991.29 J. Šesták, Glastech. Ber. Glass. Sci. Tech., 70C (1997) 439.30 J. Šesták, J. Therm. Anal. Cal., 61 (2000) 305.31 B. Hlaváèek, J. Šesták and J. J. Mareš, J. Therm. Anal. Cal., 67 (2002) 239.32 J. Šesták and Z. Strnad, in Proc. 8th Int. Conf. Reactivity of Solids, Gothenburg, Elsevier,

Amsterdam 1976, p. 410.33 J. Šesták and Z. Strnad, in Proc. Int. Congress Glass �77, CVTS Publ. House, Prague 1977,

Vol. 2, p. 249.34 N. Koga, Z. Strnad and J. Šesták, Thermochim. Acta, 203 (1992) 361.35 N. Koga and J. Šesták, Biol. Soc. Esp. Ceram. Vidrio, 31 (1992) 185.36 M. C. Weinberg, Thermochim. Acta, 280/281 (1996) 63.37 N. Koga, K. Yamaguchi and J. Šesták, J. Therm. Anal. Cal., 56 (1999) 755.38 M. C. Weinberg, J. Mining Metal. (Bor, Yugoslavia), 35 (1999) 197.39 N. Koga and J. Šesták, J. Therm. Anal. Cal., 60 (2000) 667.40 N. Koga and J. Šesták, J. Am. Ceram. Soc., 83 (2000) 1753.41 I. M. Steevels, Philips Tech. Rundschau, 9/10 (1960) 337.42 G. W. Toop and C. S. Samis, Trans. Metal. Soc. AIME, 224 (1962) 878.43 C. R. Masson, Proc. Roy. Soc. A, 287 (1965) 201.44 C. R. Masson, J. Am. Ceram. Soc., 51 (1968) 34.45 C. R. Masson, in Proc. Int. Congress Glass ’77, CVTS Publ. House, Prague 1977, Vol. 1, p. 1.46 K. Závìta and J. Šesták, in Proc. Int. Congress Glass ’77, CVTS Publ. House, Prague 1977,

Vol. 1, p. 399.47 L. L. Hench, in P. Ducheyne and J. Lemons (Eds), Bioceramics, Vol. 523, Annals of New

York Academy of Sciences, New York 1988, p. 54.48 L. L. Hench, Handbook of Bioactive Ceramics, Vol.1, CRT Press, 1990, p. 7.49 C. J. B. Fincham and F. D. Richardson, Proc. Roy. Soc., A223 (1954) 40.50 M. Temkin, Zh. Fiz. Khim, 20 (1946) 105, (in Russian).51 P. Balta and E. Balta, Introduction to Physical Chemistry of the Vitreous State, Abacus Press,

Kent 1976, Chapter 3.52 Y. Kawamoto and T. Tomozawa, Phys. Chem. Glasses, 22 (1981) 11.53 H. M. Kim, F. Miyaji, T. Kokubo, C. Ohtsuki and T. Nakamura, J. Am. Ceram. Soc.,

78 (1995) 2405.



Jour nal of Ther mal Anal y sis and Cal o rim e try, Vol. 76 (2004)

INTERACTION OF ACID AND ALKALI TREATEDTITANIUM WITH DYNAMIC SIMULATED BODYENVIRONMENTMimetic material for the bone tissue substitution Part. II

J. Strnad 1*, J. Protivínský2, D. Mazur3, K. Veltruská3, Z. Strnad 1, A. Helebrant2 and J. Šesták4

1Laboratory for Glass and Ceramics, Lasak Ltd., Papírenská 25, CZ-16000 Prague 6, Czech Republic2Department of Glass and Ceramics, Institute of Chemical Technology, Technická 5, CZ-16628 Praha 6, Czech Republic3Department of Electronics and Vacuum Physics, Faculty of Mathematics and Physics of CharlesUniversity, V. Holeëovi¹kách, CZ-18000 Prague 8, Czech Republic4Institute of Physics, Academy of Sciences, Cukrovarnická 10, CZ-16253 Prague 6, Czech Republic

Abstract

Interaction of acid and acid+alkali treated titanium samples with simulated body fluid was studied. Incase of alkali treated titanium, the dynamic arrangement of the test enabled the detection of primarycalcium and phosphate ion adsorption from the solution and later apatite crystal growth (XRD). Theinduction time for crystal growth was 24.2±0.3 h. On acid-only treated titanium no crystal growth wasdetected. The calcium phosphate adsorption layer formed on the acid treated samples was detectableby XPS only, however it differed from that one formed on the acid+alkali treated samples. Theadsorption layer formed on the acid+alkali treated samples contained larger amount of calcium,especially in the shortest exposure times. Charging of the apatite crystallites during the XPSmeasurement enabled the determination their Ca/P ratio separately from Ca/P ratio of the adsorptionlayers. XPS and EDS analyses indicated that the spherulitic crystallites consisted of carbonatedhydroxyapatite with the Ca/P ratio close to that one of the stoichiometric hydroxy apatite. It isproposed that the adsorption layer formed spontaneously and immediately on the acid+alkali treatedtitanium can provide an ideal interface between the metal implant and the apatite cement line, the firststructure formed by osteoblast cells during the formation of the new bone on foreign surfaces.

Keywords: bioactivity, body liquid, calcium phosphate, dental implant, EPS, osseointegration,osteoblast, surface treatment, titanium, XPS

Introduction

Titanium has been the most widely used material for dental implants since the lateseventies, when its osseointegrative ability was first discovered and documented [1].At that time the phenomenon of osseointegration was defined as a formation of an

1388–6150/2004/ $ 20.00

© 2004 Akadémiai Kiadó, Bu da pest



6237

Please,shortenthe title

* Author for the correspondence: E-mail: [email protected]

intimate contact of the implant with the bone tissue without intermediate layers on theoptical microscopy level. Machined titanium implants were used in the early days;later, implants with surface modifications took over offering more stable and long- term functional interface between the implant and the bone bed. Sand-blasted or titaniumplasma sprayed implants showed higher bone-implant contact or removal torque values compared to machined surfaces [2, 3]. Currently, chemical modification of titanium byacid etching is introduced to support and speed-up the healing and the bone formationprocesses around the implant further [4–8]. The acid etching procedure creates amicro-rough texture that is able to retain the fibrin network of the blood clot, throughwhich cells migrate to the surface of the implant where the new bone is formed [8].

Non-metal bioactive materials such as bioactive glasses, glass-ceramic, silicaand titania gels or hydroxyapatite have the ability to form a very stable interface withbone tissue by a formation of a calcium phosphate layer on their surfaces as aconsequence of chemical interaction with body fluids [9]. The mechanical strength of this interface usually exceeds the strength of the bone tissue to which bioactivematerial is bonded [10, 11]. Due to their poor mechanical properties, which disablethem from application under load-bearing conditions, bioactive materials (especiallyhydroxy apatite) are applied onto the surface of titanium implants to achieve fasterand more reliable bonding with bone tissue. In spite of the success in accelerating thebone healing process and in the increase of the bone-implant contact in the earlyphases of healing [12–14], hydroxyapatite plasma sprayed coatings were subject tomany controversies regarding their long-term stability.

The ability of bioactive materials to form a bone-like apatite on their surfaces inthe body can be reproduced in vitro using a simulated body fluid (SBF) [15]. Allbioactive materials, e.g. bioactive glasses, glass-ceramics, SiO2 and TiO2 gels, usetheir hydrated surfaces rich in hydroxyl group to adsorb calcium and phosphate ionsand induce spontaneous apatite crystal growth within hours or days [9, 16, 17]. It isthe kinetic of precipitation that differs from one material to another [18] and it hasbeen shown that the rate of apatite formation is related to the bone bonding ability ofmaterial. The most bioactive of contemporary materials – the 45S5 Bioglass® – forms an interfacial apatite layer 0.8 mm thick within 1 h of implantation in a rat bone [19].

In contrast to bioactive materials, machined titanium forms calcium phosphatelayers in SBF’s at much lower rate and the thickness of the precipitated layers is muchsmaller. Hanawa [20] used XPS to detect apatite-like calcium phosphate layers afterimmersing c.p. titanium in Hanks balanced salt solution and after 30 days of soaking thethickness of the adsorbed layer was 7.9 nm. Ducheyne [21] detected a calcium phosphate precipitated on the surface of c.p. titanium by XPS only after 30 days of soaking in SBF.It is assumed that c.p. titanium does not provide such suitable surface for apatite forma -tion as bioactive materials and that this can be the cause of its poorer bone-bondingability (e.g. compared to hydroxyapatite) especially in shorter healing periods and undernon ideal healing conditions [14, 22, 23].

There fore, var i ous at tempts have been made to mod ify the sur face of ti ta nium in or -der to make it bioactive, but with out the use of a thick coat ing of other bioactive ma te rial. The most suc cess ful meth ods of ti ta nium bioactivation are e.g. al kali or flu o ride treat -


2 STRNAD et al.: OSSEOINTEGRATION

ment [24]. Es pe cially the abil ity of al kali-treated ti ta nium to in duce ap a tite pre cip i ta tionaf ter such short times as 24 h of soak ing in SBF was well documented [25, 26].

Sur face treat ments mod i fy ing mi cro-rough ness (acid etch ing) or bioactivity (al kalietch ing) showed prom is ing re sults in tests run in-vivo, as well as su pe rior clin i cal per for -mance com pared to ma chined or grit blasted sur faces [5–7, 27, 28]. It was found thatthe com bi na tion of treat ments, en sur ing both op ti mal rough ness and bio active prop er ties,could re sult in an im plant sur face with out stand ing abil ity of quick and re li ableosseointegration [29]. It was con firmed that acid+al kali treated ti ta nium sur faces in duce re -pro duc ible ap a tite for ma tion in vi tro [30] and ex hibit prom is ing clin i cal per for mance [31].

The ion ex change be tween the biomaterial’s sur face and the body fluid is thefast est in ter ac tion mode; it pre cedes and af fects the ad sorp tion of larger mol e culessuch as amino ac ids and pro teins. There fore, a hy poth e sis can be made that the ionex change has a pro found ef fect on the later stages of the heal ing pro cess.

The aim of this study was to com pare the ini tial ion in ter ac tions of acid andacid+al kali treated ti ta nium with sim u lated body fluid (SBF) and to dis cuss the pos si -ble ef fect of the sur face treat ments on im plant heal ing.

Materials and methods

Preparation of samples

Titanium samples were used in the form of turnings or titanium discs (diameter 14 mmc.p. Ti grade 4), Fig 1. The machined samples (Ti-M) were washed in isopropanol in anultrasonic cleaner and dried at 110°C. The first group of samples was acid-etched (Ti- AE) in a solution of hydrochloric acid (40°C, 90 min), washed in deionized (DI) waterand ethanol in an ultrasonic cleaner. The second group of samples, designated Ti-AAE in


STRNAD et al.: OSSEOINTEGRATION 3

Fig. 1 Samples - titanium discs and turnings used for exposure in SBF

Fig. 2 Surface topography of machined- Ti-M a – acid etched-Ti-AE b – acid andalkali treated – titanium Ti-AAE and c – (SEM, magnification 4000)

this text, was acid-etched under the same conditions and subsequently etched in sodiumhydroxide solution (60°C, 4 h), washed in DI water and ethanol in ultrasonic cleaner. Allsamples were dried at 110°C after washing. The surface topography is showed in Fig. 2.

Experimental arrangement

The SBF so lu tion was pre pared us ing fol low ing re agents: KCl, NaCl, NaHCO3,MgSO4×7H2O, CaCl2, Tris, NaN3 and KH2PO4. Tris buffer was used to ad just the pH to7.55–7.60 at 25°C. So dium azide was added to in hibit bac te rial growth. Sam ples wereex posed in a flow-through re ac tion cell with SBF pre heated at 37°C (Fig. 3). The flowrate of SBF so lu tion through the cell was 0.042 mL min–1. Sam ples of the out put so lu tion were col lected for cal cium and phos phate ion con cen tra tion mea sure ment (Fig. 4). Thecom po si tion of SBF in com par i son to the in or ganic part of the blood plasma is given inTa ble 1. The prep a ra tion of the so lu tion was car ried out af ter Jonasova et al. [32].

The ex po sure in SBF is usu ally car ried out as a static ex per i ment, where a sam ple isplaced in the SBF so lu tion of a con stant vol ume, or the so lu tion is pe ri od i cally re newed[14, 21, 33]. If the so lu tion vol ume is suf fi ciently high to keep ap prox i mately con stant so -lu tion com po si tion it is dif fi cult to de tect compositional changes in the so lu tion caused bythe ma te rial-SBF in ter ac tion. If the vol ume is small, con cen tra tion changes are mea sur ablebut in case of ap a tite for ma tion the driv ing force for nu cle ation and crys tal li za tion de -creases as cal cium and phos phate ions are con sumed from the so lu tion. Static ex po sure



Fig. 3 Flow-through cell used for exposure of samples to SBF

Fig. 4 Experimental arrangement of the exposure in SBF

also does not re sem ble the con di tions dur ing im plan ta tion, where blood cir cu lates. There -fore, a dy namic ex po sure is sug gested to keep con stant com po si tion dur ing the ex per i mentand to de tect compositional changes in SBF dur ing the ini tial stages of in ter ac tion.

Surface and solution analysis

Composition of the Ti samples’ surfaces was analyzed using X-ray photoelectronspectroscopy. UHV facility with Microtech X-ray source and Omicron multi-channelhemispherical analyzer EA 125 was used. The source has Mg/Al dual anode and the Mganode was used for the measurements. The X-rays were not monochromated. The photo -electron spectra were measured with the analyzer pass energy of 20 eV. The spectra were fitted using a standard fitting procedure with Gaussian–Lorentzian curves.

Af ter ex po sure in SBF the sur face of sam ples was an a lyzed by Scan ning elec tron mi -cros copy and en ergy dispersive spec tros copy (SEM-EDS, Jeol XA-733- super probe, JeolUSA, Inc.). The EDS anal y sis re sults are based on semi-quan ti ta tive anal y sis with outthe use of stan dards. X-ray dif frac tion (XRD) mea sure ments were per formed us ing 3000Pdiffractometer, Cu an ode, mea sured at 40 kV, 30 mA (Seifert Co., Ahrensburg, Ger many).

The con cen tra tion of cal cium in the out put so lu tion was de ter mined us ingatomic ab sorp tion spec tros copy (Varian-Spectr AA300) and phos phate con cen tra -tion was de ter mined spec tro pho to met ri cally (UV-1201, Shimadzu Eu rope, Ltd.).

The rate of hydration of the prepared titanium samples was measured using FTIR(Fourier transformed infra-red) spectroscopy (Nicolet 740, Nicolet Madison, USA). Awavenumber interval of 400–4000 cm–1 was used. Incident radiation angle was 45°.

Results and discussion

Concentration changes in the output solution

The time de pend ence of Ca2+ and PO 43– ion con cen tra tion in the out put so lu tion, mea -

sured af ter ex po sure of Ti-AAE sam ples in SBF, in di cated three main phases of in ter -ac tion (Fig.)??. First, a rapid but tem po rary de crease of cal cium and phos phate con -



Ta ble 1 The com po si tion of SBF used in ex per i ments com pared to that of hu man blood plasma

SBF/mmol L–1 Blood plasma/mmol L–1

Na+ 142.0 137–147.0

K+ 5.0 3.8–5.1

Ca2+ 2.5 2.25–2.75

Mg2+ 1.0 0.75–1.25

Cl– 131.0 98–106

HCO3– 5.0 24–35

HPO42– 1.0 0.65–1.62

SO42– 0.5 0.5

Which fig ure do you mean?Please, re fer tofig ures in or der

cen tra tion in the ef flu ent was de tected. This de crease in di cated a sig nif i cantad sorp tion of Ca2+ and PO 4

3– ions by the ma te rial. Dur ing the sec ond phase, ef flu entcon cen tra tion al most re turned to its ini tial value (100 mg L–1 Ca2+, 96 mg L–1 PO 4

3– ).Ap prox i mately 25 h from the be gin ning of the ex per i ment, an other con cen tra tion de -crease was de tected. This con cen tra tion de crease sta bi lized at the levelof 60–63 mg L–1 of PO 4

3– , in di cat ing the crys tal growth at a con stant rate. The for ma -tion of ap a tite crys tals on the sur face of Ti-AAE sam ples was also con firmed by SEM and XRD mea sure ment (Fig. 8)??.

The dynamic experimental arrangement enabled simple determination of theinduction times using a plot of total mass of PO4

3– consumed by samples’ surfaces vs.time and by extrapolation of its late-time linear part to the zero mass (Fig. 6). After theinduction period, a spontaneous consumption of calcium and phosphate ions from thesolution occurred. The induction time value, which is 24.2±0.3 h, is in good agreementwith the results obtained using static exposure and reported earlier [26, 34]. In case ofTi-AAE samples the initial calcium and phosphate adsorption was remarkable in thetime dependence of the output ion concentrations. The amount of calcium andphosphate ions adsorbed was significantly higher than in case of Ti-AE samples, where no changes in the effluent concentration were detected (Fig. 5).

The find ing that only Ti-AAE sam ples in duce ap a tite for ma tion is in agree mentwith re sults of Nancollas et al. [35], who de tected ap a tite growth only on the KOHtreated ti ta nium sur face, whereas on acid (HNO3) treated ti ta nium the growth did notoc cur within 3 days of ex po sure.



Fig. 5 Time dependence of the PO43– and Ca2+ concentration in the output SBF solution

during exposure of Ti-AE (upper) and Ti-AAE (below) samples, indicatingthree phases of interaction

FTIR analysis

It was found by Li [16] that not only the negative charge of hydroxyl groups (oftitanium oxides) but also their high density on the surface is necessary prerequisitefor apatite formation. The FTIR analysis was used to determine the level of hydration of the titanium samples. The peak intensity of hydroxyl groups in the wavenumberrange of 3300–3600 was evaluated. The Ti-AAE samples showed significantlyhigher absorption at 3384.50 cm–1 (1.36 Kubelka–Munk units, Fig. 7) compared toTi-AE samples. The absorption of the Ti-M samples was negligible compared toTi-AAE. The position of the peak is characteristic to OH groups present e.g. insol–gel prepared TiO2 films [36].

These re sults in di cate that the abil ity of al kali-treated ti ta nium to in duce ap a titefor ma tion can be caused by higher den sity of hydroxyl groups pres ent in its po rousgel- like struc ture com pared to the lower den sity on the acid-treatedtitanium.

The results of the dynamic flow-through test (Fig. 5) and XPS analysis (Fig. 9)indicate that although ion adsorption occurs on both acid and alkali treated titanium



Fig. 6 Time dependence of the precipitated PO43– mass plotted for Ti-AAE and Ti-AE samples

Fig. 7 The FTIR spectrum of Ti-M, Ti-AE and Ti-AAE samples

surfaces, the adsorbed amount of phosphate and especially calcium ions is larger in caseof more hydrated alkali-treated titanium in the first stages of interaction. This could becaused by the detected difference in the density of hydroxyl groups present on the surface of Ti-AAE and Ti-AE samples. It was shown that during the process of peri-implanthealing, calcium phosphate cement line containing non-collagenous proteins is formedon the implant’s surface before collagenous matrix formation occurs [37, 38].

Intensive calcium adsorption mediated by hydroxyl groups can induce apatitecrystal growth but also attract non-collagenous proteins as good calcium binders [39].By this mechanism an apatite-protein cement layer can be formed on the surface of animplant. Chemical and structural similarity between the preformed cement layer on the



Fig. 8 XRD spectrum of the Ti-AAE sample after 48 h of exposure in SBF solution

Fig. 9 Photoelectron spectra of Ca 2p and P 2p of a – Ti-AE (HCl-corroded)samples b – Ti-AAE (HCl+NaOH-treated) samples and c – immersed in SBFfor various periods of time

implant’s surface and that one produced later by cells can result in a faster formation ofa more mechanically stable interface with newly formed bone tissue.

XRD analysis

Although apatite crystallites did not cover the surface of Ti-AAE samples completely,it was possible to detect weak diffraction peaks confirming the presence of hydroxy -apatite crystalline phase (Fig. 8).

XPS analysis

On the sam ple Ti-AE (cor roded in HCl with out sub se quent SBF treat ment) car bon C, ni tro gen N, ox y gen O and ti ta nium Ti were de tected, so dium Na was pres ent inbarely de tect able amount. In the spec trum of Ti-AAE sam ple (be fore SBF treat ment)so dium Na 1s sig nal was more in tense (by ~3 or ders of mag ni tude) than for Ti-AEsam ple, as a re sult of the NaOH treat ment. The en er getic in ter vals mea sured in thepho to elec tron spec tra were those of C 1s, N 1s, O 1s, Na 1s, Ti 2p, Ca 2p and P 2p. Inthis work, the evo lu tion of cal cium and phos pho rus spec tra was cru cial and it will bedis cussed in de tail.

The spectra of carbon, oxygen, calcium and phosphorus exhibited substantiallydifferent behavior in case of Ti-AAE compared to Ti-AE samples. In this article, theCa 2p and P 2p spectra (Fig. 9) are used for illustration. Spectra of O 1s and C 1sevolved likely and they are not shown here. The spectra of Ti 2p showed no difference(between Ti-AE and Ti-AAE samples) and no evolution in peak positions, as well asthe N 1s spectra. The intensity of Na 1s peak in Ti-AAE measurements decreased withtime in SBF in agreement with models of Kim et al. [40] and de Andrade et al. [41], for Ti-AE it remained at the just detectable intensity mentioned above. All the analyzedspectra, including those in Fig. 9, were corrected with respect to position of C 1s peakat 285.0 eV to eliminate the shifts in peak positions due to surface charging underX-ray illumination. As shown in Fig. 9, the calcium peak position in case of Ti-AEsamples is 347.5 eV, whereas in case of Ti-AAE samples it is 347.3 eV. Additionally,the spectra of Ca 2p of Ti-AAE samples divide in two overlapping peak doublets forthe 48 and 74 h periods of SBF treatment. The non-shifted Ca 2p position at 347.3 eV(Ti-AAE samples) corresponds to that of hydroxyapatite [42].

Phos pho rus spec tra ex hibit a sim i lar be hav ior. For Ti-AE sam ples, the P 2p peak po si tion is 133.6 eV and it does not change with time of SBF treat ment (peak at138.7 eV shall be dis cussed later). In case of Ti-AAE sam ples, the P 2p peak is at133.3 eV with ad di tional peaks at 136.5 and 137.9 eV, ap pear ing at 48 and 74 h inSBF, re spec tively. The non-shifted P 2p po si tion at 133.3 eV (Ti-AAE sam ples) cor -re sponds to that of hydroxyapatite [43].

As shown in Fig. 9, ad di tional in tense peaks ap peared in cal cium and phos pho -rus spec tra of Ti-AAE sam ples treated in SBF for 48 and 74 h pe ri ods. These newpeaks are shifted by 3.2 and 4.5 eV, re spec tively. Sim i lar new peaks at the same shifts from ba sic po si tions were ob served in the C 1s and O 1s spec tra (not shown here),



too. In co her ence with SEM ob ser va tion of crys tal growth on the Ti-AAE sam ples(48 and 74 h of SBF treat ment, see SEM-EDS re sults be low) we con clude that theshifted sig nal co mes from the crys tal lites. The con sti tu tion of cal cium, phos pho rus,ox y gen and car bon in di cates a car bon ated cal cium-phosphate na ture of the crys tal -lites. Be ing of in su lat ing na ture, the crys tal lites take on elec tri cal charge. Since theycharged in de pend ently from the sam ple sur face, i.e. much more, it was pos si ble todis tin guish them from the ad sorbed cal cium- and phos phate-rich layer. The largeshift may also in di cate their weak bond to the sam ple sur face. The peak at 138.7 eVin P 2p (74 h of SBF treat ment) spec trum stands out of all other data, be cause nospec trum of other el e ments has shown such a fea ture. Hav ing elim i nated the pos si bil -ity of pres ence of for eign el e ment as an im pu rity on the sam ple, it was sug gested thatsome phos pho rus-based clus ters may have formed on the sur face. How ever, an otherex per i ment will have to be done to ver ify this sug ges tion.

The Ca/P ratios measured for Ti-AE samples and Ti-AAE samples are shown inFig. 10. Supposing that all ‘charged’ signal comes from HA crystallites, the Ca/Pratios for the crystallites and for the adsorbed calcium phosphate were calculatedseparately. For the Ti-AAE samples the Ca/P ratio of the adsorbed layer varies from2.5 up to 6.2. The Ca/P ratio of the crystallites is in the range from 1.0 to 1.7, that is,significantly lower than the Ca/P ratio of the adsorbed layer. This relation betweenCa/P ratios of the adsorbed layer and the growing crystallites corresponds to themodel of Takadama et al. [43] saying that the calcium deposition precedes thedeposition of phosphate groups. The values of Ca/P ratio for the set of Ti-AE samples were between 0.7 and 1.4, which is lower than any of the previous.

From the cal cium and phos pho rus pho to elec tron peak po si tions, as well as fromthe Ca/P ra tio mea sure ments, it can be con cluded, that crys tal lites ap pear ing atTi-AAE (HCl+NaOH-treated) sur faces con sist of car bon ated hydroxyapatite. Thead sorbed pre cur sor layer cre ated on the sur face within the first 24 h could be dis tin -guished from the hydroxyapatite crys tal lites on ba sis of the Ca/P ra tio, un der the as -



Fig. 10 The Ca/P ratios measured for the Ti-AE (HCl-corroded) samples and Ti-AAE(HCl+NaOH-treated) samples, immersed in SBF for various periods of time

sump tion of crys tal lites charg ing in de pend ently on the ad sorbed pre cur sor layer.Be tween these two no sig nif i cant dif fer ence in bind ing en er gies of el e ments in -volved has been ob served. On the sam ples with sole HCl treat ment (Ti-AE sam -ples), a thin layer of a cal cium phos phate (not de tected by other meth ods used),which dif fers from HA, is cre ated. The Ca/P ra tio here is re mark ably lower thanthe Ca/P ra tio of the layer ad sorbed on NaOH treated (Ti-AAE) sam ples andthe bind ing en er gies peak po si tions dif fer from those of HA, as dis cussed above.The pri mary cal cium- and phos phate-rich lay ers ad sorbed on sur faces of Ti-AEand Ti-AAE sam ples could not be as signed more spe cific chem i cal for mu lae at area son able level of con fi dence. It is a con se quence of com pli cated sur face mor -phol ogy, as well as the fact that XPS spec tra of sev eral rep re sen ta tives from thecal cium-phosphates fam ily dif fer too lit tle to dis tin guish in our mea sure ments[42]. The atomic Ca/P ra tio of the crys tal lites, the co her ence in shifts in Ca 2p, P2p, C 1s and O 1s spec tra (48 and 74 h in SBF), and the Ca 2p and P 2p peak po si -tions lead us to a con clu sion that the cal cium-phosphate de vel oped to the crys tals on the Ti-AAE sur faces is most likely car bon ated hydroxy apatite. From the Ca/Pra tios and SEM ob ser va tion we con clude that the higher rel a tive abun dance ofcal cium at oms in the ad sorbed pre cur sor layer is es sen tial for later for ma tion ofhydroxyapatite crys tal lites.

SEM-EDS

SEM-EDS method was used to de tect the changes in the sur face struc ture dur ingex po sure in SBF and to an a lyze pre cip i tates formed on the sur face of sam ples. Fig -ure 11 shows the po rous struc ture of the Ti-AAE sam ples af ter ex po sure in SBF. Itcan be seen that af ter 2.5 min up to 24 h only small denser (lighter) ar eas areformed. These can rep re sent pos si ble nu cle ation sites of fu ture ap a tite crys tal



Fig. 11 The surface structure of the Ti-AAE samples after exposure in SBF fora – 2.5 min and b – 1 h c – 24 h d – 48 h e – 72 h (SEM, mag. 4000)(arrows mark the apatite crystallites)



Fig. 12 EDS analysis of the a – crystallite b – the substrate of the sample not coveredby the crystallites after exposure of Ti-AAE sample in SBF for 48 h

Fig. 13 Surface of the Ti-AE samples after a – 24, b – 48 and c – 72 h of soaking inSBF together with EDS analysis (below) after 48 h of soaking in SBF(SEM-EDS, magn. 4000)

growth. On the sam ple ex posed in SBF for 48 and 72 h for ma tion of crys tal lites wasob served. The SEM-EDS anal y sis of the Ti-AAE sam ple ex posed in SBF for 48 h isshown in Fig. 12. The anal y sis in di cates that the sur face of Ti-AAE sam ple (not cov -ered by the pre cip i tat ing crys tal lites) af ter ex po sure in SBF con tains a de tect ableamount of cal cium but not de tect able (by EDS) amount of phos pho rus. The EDS anal y -sis from the spher i cal crys tal lite (Fig. 12b) shows strong lines of cal cium and phos pho -rus in the atomic ra tio of 1.77, which is slightly higher than that one of stoichiometrichydroxyapatite (1.67). From EDS and XPS re sults we con clude that spher i cal bodiesconsist of hydroxyapatite.

The Ti-AE sam ples did not show any dif fer ence in sur face struc ture and EDS anal y sis dur ing the ex po sure in SBF and only ti ta nium was de tected af ter 48 h in SBF (Fig. 13).

Conclusions

The solution analysis during the dynamic SBF test detected calcium and phosphateionadsorption on the acid+alkali-etched samples in the first stages of interaction. Thismethod was not sensitive enough to detect less intense adsorption in case of Ti-AEsamples, where the adsorption layer was detected by XPS analysis. Already after theshortest exposure time of 2.5 min the adsorption layer on the acid+alkali treatedsamples contained larger amounts of phosphate and especially calcium (also Ca/P ratio was significantly higher than in the case of acid-treated samples). After the inductiontime of 24.2 h the acid+alkali treated titanium induced apatite crystal growth. Theonset of the crystal growth was clearly indicated by the solution analysis.

Charging of the spherulitic crys tal lites dur ing the XPS mea sure ments al lowedus to de ter mine the Ca/P ra tio sep a rately for the ad sorp tion layer and the crys tal lites.Using XPS and EDS anal y sis the crys tal lites were iden ti fied as car bon ated ap a titewith the Ca/P ra tio close to that of stoichiometric ap a tite. The de ter mi na tion of the in -duc tion time could serve as a mea sure of a sub strate abil ity to sup port ap a tite nu cle -ation and growth. The alkali-treated ti ta nium com pared to the acid-treated one in -duced ap a tite for ma tion in vi tro and could there fore pro vide a suit able sub strate forap a tite like cal cium phos phate pre cip i tates in the ce ment lines formed on the im -plants sur faces dur ing the pro cess of new bone for ma tion.

The study is con tin u a tion of the se ries of our com mu ni ca tion deal ing withglass-ceramic sub stance and sur faces [44, 45] per spec tive as mi metic ma te ri als forthe bone tis sue sub sti tu tion par tic u larly ap pli ca ble in the den tal prac tice.

This study was supported by the Ministry of Industry and Trade of CzechRepublic under the project number FB-CV/64 as well as by the Grant Agency ofAcademy of Sciences of Czech Republic under the project number A 4010101 andwas also a part of the research program MŠMT 113200002 financed by the Ministryof Education of Czech Republic.



References

1 P. I. Br¯nemark, Scand. J. Plast. Reconstr. Surg., 3 (1969) 81.

2 A. Wennerberg, T. Albrektsson, C. Johansson and B. Andersson, Biomaterials, 17 (1996) 15.

3 K. Gotfredsen, T. Berglundh and J. Lindhe, Clin. Im plant. Dent. Relat. Res., 2 (2000) 120.

4 M. Roccuzzo, M. Bunino, F. Prioglio and S. D. Bianchi, Clin. Oral Im plants Res.,

12 (2001) 572.

5 D. L. Cochran, D. Buser, C. M. Bruggenkate, D. Weingart, T. M. Tay lor, J. P. Bernard,

F. Pe ters and J. P. Simpson, Clin. Oral Im plants Res., 13 (2002) 144.

6 P. R. Klokkevold, P. John son, S. Dadgostari, A. Caputo, J. E. Davies and R. D. Nishimura,

Clin. Oral Im plants Res., 12 (2001) 350.

7 W. Khang, S. Feldman, C. E. Haw ley and J. Gunsolley, J. Periodontol, 72 (2001) 1384.

8 R. J. Lazzara, Bone re sponse to Dual Acid-Etched and Ma chined Ti ta nium Im plant Sur faces.

In: J. E. Davies (Ed.) ‘Bone en gi neer ing’ To ronto, Em squared Inc 2000, p. 381.

9 T. Yamamuro, L. L. Hench and J. Wilson, Handbook of Bioactive Ce ramics, CRC Press 1990.

10 T. Kokubo, S. Ito, M. Shigematsu, S. Sakka and T. Yamamuro, J. Ma ter. Sci., 22 (1987) 4067.

11 U. Gross, R. Kinne, H. J. Schmitz and V. Strunz, CRC Crit i cal Re views in Biocompatibility,

4 (1988) 2.

12 S. Vercaigne, J. G. Wolke, I. Naert and J. A. Jansen, Oral. Im plants Res., 9 (1998) 261.

13 L. Sun, C. C. Berndt, K. A. Gross and A. Kucuk, J. Biomed. Ma ter. Res., 58 (2001) 570.

14 Z. Strnad, J. Strnad, C. Povýšil and K. Urban, J. Oral and Maxillofacial Implants, 15 (2000) 483.

15 T. Kokubo, H. Kushitani, S. Sakka, T. Kitsugi and T. Yamamuro, Ma ter. Res., 24 (1990) 721.

16 P. Li, C. Ohtsuki, T. Kokubo, K. Nakanishi, N. Soga and K. de Groot, J. Biomed. Ma ter. Res.,

28 (1994) 7.

17 P. Li, X. Ye, I. Kangasniemi, J. M. de Blieck-Hogervorst, C. P. Klein and K. de Groot,

J. Biomed. Ma ter. Res., 29 (1995) 325.

18 L. L. Hench, Sur face re ac tion Ki net ics and Ad sorp tion of bi o log i cal moi eties: A mech a nis tic

Ap proach to tis sue At tach ment. In J. E. Davies (Ed.), Bone en gi neer ing, To ronto, Em. squared

Inc. 2000, p. 381.

19 L. L. Hench and O. Andersson, Bioactive glasses. In: L. L. Hench, J. Wilson (Eds) ‘An in tro -

duc tion to bioceramics’, Florida, World Sci en tific Pub lishing 1993, p. 41.

20 T. Hanawa, Ti ta nium and its ox ide film: A sub strate for for ma tion of ap a tite. In: J. E. Davies

(Ed.) Bone-biomaterial in ter face, To ronto, Em squared Inc. 2000, p. 49.

21 P. Li and P. Ducheyne, J. Biomed. Ma ter. Res., 41 (1998) 341.

22 K. Soballe, Acta Orthop. Scand. Suppl., 255 (1993) 1.

23 K. Soballe, E. S. Hansen, H. Brockstedt-Rasmussen, V. E. Hjortdal, G. I. Juhl, C. M. Pedersen,

I. Hvid and C. Bung er, Clin. Orthop., 272 (1991) 300.

24 J. E. Elingsen, On the prop er ties of sur face-modified ti ta nium. In: J. E. Davies, (Ed.) Bone en -

gi neer ing, To ronto, Em squared Inc. 2000, p. 183.

25 H. M. Kim, F. Miyaji and T. Kokubo, J. Ma ter. Sci.: Ma ter. in Med i cine, 8 (1997) 341.

26 H. M. Kim, F. Miyaji, T. Kokubo, S. Nishiguchi and T. Nakamura, J. Biomed. Mater. Res.,

45 (1999) 100.

27 T. Kokubo, H. M. Kim, S. Nishiguchi and T. Nakamura, In vivo for ma tion in duced on ti ta -

nium metal and its al loys by chem i cal treat ment in Bioceramics. In: Pro ceed ings of the 13th

Int. Symp. on Ce ramics in Med i cine, Bo lo gna, It aly, Trans. Tech. 2001, p. 3.



28 S. Nishiguchi, T. Nakamura, M. Kobayashi, W. O. Yan, H. M. Kim, F. Miyaji and T. Kokubo,

The ef fect of heat treat ment on bone bond ing abil ity of al kali-treated titanium.

In: Bioceramics, Vol. 10, Pro ceed ings of the 10th Int. Symp. on Ce ramics in Med i cine, Paris,

France, Elsevier 1997, p. 561.

29 A. Šimùnek, J. Strnad, J. Novák, Z. Strnad, D. Kopecká and R. Mounajjed, Clin. Oral. Impl. Res.,

12 (2001) 393.

30 L. Jonášová and J. Hlavá¹, Ef fect of chem i cal treat ment of ti ta nium on ap a tite formation.

In: H. Stallforth and P. Revell (Eds) ‘Ma te rials for med i cal en gi neer ing’, Euromat 99,

Weinheim, Wiley-VCH 2000, p. 126.

31 A. Šimùnek, J. Strnad and A. Štìpánek, Clin. Oral Impl. Res., 13 (2002) 4.

32 A. Helebrant, L. Jonášová and L. Šanda, Ce ramics/Silikáty (Prague), 46 (2002) 9.

33 K. Hata, T. Kokubo, T. Nakamura and T. Yamamuro, J. Am. Ceram. Soc., 78 (1995) 1049.

34 J. Strnad, A. Helebrant and J. Hamá¹ková, Glastech. Ber. Glass. Sci. Technol., 73 (2000) C1.

35 G. H. Nancollas and W. Wu, The con trolled nu cle ation and growth of se lected cal cium phos -

phate phases on mod i fied ti ta nium im plant surfaces. In: Pro ceed ings of Sixth World

Biomaterial Con gress, Ha waii: So ci ety for Biomaterials, USA 2000.

36 Y. Djaoued, S. Badilescu, P. V. Ashrit and J. Robichaud, Internat. J. Vibrational

Spectroscopy, 5 (2003) 6.

37 J. E. Davies and N. Baldan, J. Biomed. Ma ter. Res., 36 (1997) 429.

38 M. M. Hosseini, J. Sodek, R. P. Franke and J. E. Davies, The struc ture and com po si tion of the

bone im plant in ter face. In: J. E. Davies (Ed.) Bone en gi neer ing, To ronto, Em squared Inc

2000, p. 295.

39 E. D. Eanes, Dy nam ics of cal cium phos phate pre cip i ta tion. In: E. Bonucci (Ed.), ‘Cal ci fi ca tion

in Bi o log i cal Sys tems’, Boca Raton, CRC Press 1992, p. 1.

40 H. M. Kim, F. Miyaji, T. Kokubo and T. Nakamura, J. Biomed. Mat. Res., 32 (1996) 409.

41 M. C. de Andrade, M. R. T. Filgueiras and T. Ogasawara, J. Biomed. Mat. Res.,

46 (1999) 441.

42 S. Ka¹iulis, G. Mattogno, L. Pandolfi, M. Cavalli, G. Gnappi and A. Montenero,

Appl. Surf. Sci., 151 (1999) 1.

43 H. Takadama, H. M. Kim, T. Kokubo and T. Nakamura, J. Biomed. Ma ter. Res.,

55 (2001) 185.

44 J. Šesták, J. Therm. Anal. Cal., 61 (2000) 305.

45 N. Koga, Z. Strnad, J. Šesták and J. Strnad, J. Therm. Anal. Cal., 71 (2003) 927.



ARTICLE IN PRESS

0022-3697/$ - se

doi:10.1016/j.jp

�CorrespondiE-mail addre

[email protected] (

Journal of Physics and Chemistry of Solids 68 (2007) 841–845

www.elsevier.com/locate/jpcs

Bio-activated titanium surface utilizable for mimetic boneimplantation in dentistry—Part III: Surface characteristics

and bone–implant contact formation

Jakub Strnada, Zdenek Strnada, Jaroslav Sestakb,�, Karel Urbanc, Ctibor Povysild

aLASAK—Laboratory for Glass and Ceramics, Papırenska 25, CZ—16200 Praha 6, Czech RepublicbUniversity of West Bohemia, Faculty of Applied Sciences, Universitni 8, CZ-30614 Pilsen and Institute of Physics,

Academy of Sciences of the Czech Republic, Cukrovarnicka 10, CZ—162 53 Prague 6, Czech RepubliccOrthopedic Clinic, Teaching Hospital, Charles University, Simkova 870, CZ—500 01 Hradec Kralove, Czech Republic

dInstitute of Pathology, 1st Medical Faculty of Charles University and Institute for Postgraduate Studies,

U Nemocnice 4, CZ—128 52 Praha 2, Czech Republic

Abstract

This study was carried out to quantify the effect of an alkali-modified surface on the bone–implant interface formation during healing

using an animal model. A total of 24 screw-shaped, self-tapping, (c.p.) titanium dental implants, divided into test group B—implants with

alkali-modified surface (Bio surface) and control group M—implants with turned, machined surface, were inserted without pre-tapping

in the tibiae of three beagle dogs. The animals were sacrificed after 2, 5 and 12 weeks and the bone–implant contact (BIC%) was

evaluated histometrically. The surface characteristics that differed between the implant surfaces, i.e. specific surface area, contact angle,

may represent factors that influence the rate of osseointegration and the secondary implant stability. The alkali-treated surface enhances

the BIC formation during the first 2–5 weeks of healing compared to the turned, machined surface.

r 2007 Elsevier Ltd. All rights reserved.

1. Introduction

More than a quarter century ago, two material groupshave been found to be able to form a mechanically stableand functional interface with bone. One group consisted ofcertain soda–lime–silica glasses, with or without additionof phosphorus (V) oxide, and the first glass exhibiting thebone-bonding ability (discovered by Hench) [1,2] wasnamed and registered under the name Bioglass. The glassesexhibiting the bone-bonding ability were designated asbioactive by the following definition: ‘‘the bioactivity is thecharacteristics of an implant material which allows it toform a bond with living tissues’’.

Another material found to exhibit the bone-bondingability was machined titanium. This characteristic of

e front matter r 2007 Elsevier Ltd. All rights reserved.

cs.2007.02.040

ng author.

sses: [email protected] (J. Strnad),

J. Sestak).

titanium was first described by Branemark [3] and cow-orkers and the phenomenon of attachment to bone wasnamed osseointegration with the following definition:‘‘osseointegration represents the formation of a directcontact of a material with bone without intermediatefibrous tissue layer’’, often when observed by means of alight microscope. The major difference between the twomaterial groups was represented by the materials reactivityand related kinetics of the bone–material interface forma-tion. Very reactive bioactive glasses formed a stableinterface with bone within days where as machinedtitanium required healing periods of several months toreach the same bone–implant contact (BIC). Thanks tohigh reactivity of bioactive glasses, the course of theglass–body environment interaction could be investigatedmore easily, and the bone-bonding ability was describedand quantified in direct relation to the glass composition[4–7]. Soon other materials like hydroxyapatite, sol/gelprepared glasses or glass ceramics, which were found to

www.elsevier.com/locate/jpcs

dx.doi.org/10.1016/j.jpcs.2007.02.040

mailto:[email protected]

mailto:[email protected]

ARTICLE IN PRESSJ. Strnad et al. / Journal of Physics and Chemistry of Solids 68 (2007) 841–845842

exhibit similar bone-bonding ability, were included into the‘‘bioactive’’ group. All these materials had in common areaction step, which occurred during exposure to bodyenvironment and before the formation of the mechanicallystable interface with bone. This step was the precipitationof apatite mineral on the surface of the original material asa result of chemical interaction of the materials surfacewith body environment [1,8].

This apatite formation was accepted as a hallmark of thebioactive materials and it is assumed that this bone mineralformation enables quick formation of the mechanicallystable and functional interface of bone and a bioactivematerial. For initial assessment of potential bioactivityof a material, the apatite formation occurring in vivo wasreproduced in vitro in simulated body environment and agood correlation of in vivo and in vitro results was foundfor the above mentioned, highly bioactive materials [9–11].

The investigation of the osseointegration observed in thetitanium group was much more difficult because of the lowreactivity of titanium in body environment. Changesdetected in the passive oxide layer during the exposure tobody environment were not comparable in magnitude tothose occurring in bioactive materials [12,13]. With theintention to increase the implant surface available for boneingrowth and fixation and/or to increase the blood clotretention on the materials surface, machined titanium ismost often modified by sand blasting, plasma spraying oracid etching. Resulting roughened surfaces showed higherbone/implant interface strength (removal torque values)and higher BIC after the same healing time compared tomachined titanium [1,14–21]. It was suggested that surfaceswith the mean roughness parameter Sa (arithmetic averageheight deviation) in the interval of 1.0–1.5 mm showstronger bone response than smoother or rougher implantsurfaces [22]. These suggestions however do not considerthe chemical composition changes introduced to thecompared surfaces by the roughening procedures. It wasshown that the optimal roughness value varies according tothe chemical composition of the tested surfaces [13,23,24].Some authors also demonstrated the ability of titaniumoxide layer to attract preferentially calcium and phosphateions from simulated body solutions—a characteristicproperty of bioactive materials [12,25]. Although rough-ening surface treatments always change both surfacechemistry and morphology of the titanium substrate, theyare usually regarded as tools for optimization of surfaceroughness only. The mechanism of bone–titanium attach-ment of the moderately roughened surfaces is considered tobe solely morphology based and dependent on themechanical interlocking [22,26]. Acceleration of the bone–implant interface and the increase of the BIC in the earlyphases of healing [27–29], was successfully achieved byplasma spraying of a hydroxyapatite as a bioactivematerial on the titanium substrate, however, hydroxyapa-tite plasma sprayed coatings have been the subject of manycontroversies [30,31] regarding their long-term stability andthickness and low long-term success rates [32,33]. On the

other hand, some authors reported mid- and long-termclinical results showing high success rates of hydroxyapa-tite-coated implants [34–36].As lately as in the late 1990s, specific surface modifica-

tions of machined titanium have been developed with theintention to modify the reactivity of titanium chemically sothat it gains the best defined characteristics of well-provenbioactive materials—the ability to induce apatite mineralformation in vitro. Until now, to the knowledge of theauthors, only two surface modifications of titaniumsupporting the precipitation of bone mineral apatite havebeen developed and clinically introduced. In 1999, alkalitreatment [37] was used in combination with sand blastingand acid etching on LASAK implants (Bio surface) [38–41]and in 2000 fluoridated titanium surface (Osseospeed) wasintroduced by ASTRATECH [42,43].From the present scientific literature, it seems necessary to

evaluate both surface roughness and chemistry in order to beable to draw reliable conclusions regarding the effect of thesurface treatment on the bone–implant interface formation.A method of quantitative evaluation of the clinical benefit ofa surface treatment has not yet been established. Machinedsurface with defined surface roughness is often used as areference when evaluating the effect of a surface treatment.The aim of this study is to evaluate the physicochemicalproperties and the stability time dependence during healingof potentially bioactive titanium surface and to compare itwith the machined surface as a reference. The surfaceproperties of both surfaces were compared to those of othercommercially available implant surfaces.

2. Material and methods

Twelve pairs of screw shaped, self-tapping, (c.p.) titaniumdental implants, divided into test group B—implants withalkali-treated surface (Bio surface) and control group M—implants with turned, machined surface, were insertedwithout pre-tapping in the tibiae of three Beagle dogs. Theanimals were sacrificed after 2, 5 and 12 weeks and the BICwas evaluated histomorphometrically.Surface roughness of implants was determined using

scanning surface topography instrument a Talysurf CLI1000 with confocal CLA gauge (Taylor Hobson, Leicester,United Kingdom) that provides highly accurate non-contact 3D measurement. Dynamic contact angle measure-ment was performed using the Wilhelmy plate method,using Tensiometer K15 (Kruss GmbH, Germany). Thewetting angle values in water were determined from thedependence of the wetting force on the immersion depth.The mean values of the wetting angle were calculatedfrom four repeated measurements (Table 2). The surfacearea was calculated by the BET method from the results ofthe krypton gas absorption study. The surface area isexpressed in relation to unit geometric surface area of theimplant (Table 2). The determination was performedby absorption of krypton on a ASAP 2010M instrument(Micromeritics, USA).

ARTICLE IN PRESS

80

90

B

100

J. Strnad et al. / Journal of Physics and Chemistry of Solids 68 (2007) 841–845 843

A total of three beagle dogs (mean weight 16.2 kg) wereused in this study, which was approved by the ethicscommittee for work with experimental animals at theTeaching Hospital, Charles University, Hradec Kralove,Czech Republic. Under total anaesthesia, in a position onits back, following the usual preparation of the operationfield and toweling, a surgical cut with a length of 7 cm wasmade on the anteromedial surface of the tibia. A sharp cutwas made in the fascia and then in the periosteum, whichwas widened to the side with a raspatory. Followinguncovering of the surface of the tibia, the positions fordrilling the holes for implanting the tested implants weremarked. The implant sites were prepared with 1.5mm,2.0mm pilot and 3.0mm drills at low rpm with simulta-neous cooling with a physiological solution. Then acountersink drill was used. The three pairs of implantswith alkali etched surface and machined surface wereinserted in the tibia side by side. The implants were blindedwith screws and, following control of hemostasis, theoperation wounds were closed in layers and were finallycovered with a sterile bandage. Following completion ofthe experiment, the dogs were sacrificed after 2, 5 and 12weeks by an overdose of thiopental and the tibiae wereremoved, a contact X-ray was made and then they werefixed in 10% formaldehyde prior to histological evaluation.

The tibiae were dissected and blocks of 7mm thicknesscontaining one implant each were prepared. Non-decalci-fied (ground) sections were processed according to themethod of Donath and Breuner. Thin sections with athickness of 30–50 mm were stained with toluidine blue andexamined using an optical microscope (Olympus BX-60,Tokyo, Japan), equipped with an image system (QuickPHOTO Industrial 2.0, Prague, Czech Republic.). Bone–implant interfaces at the threaded part were histometricallyanalyzed by evaluating the percentage of BIC. The lengthof the bone tissue in direct contact with the implant (BC)and the total interface length (IL) were measured. Thepercentage of BIC is given by the ratio of the direct contactlength to the total interface length (BC/IL� 100 ¼ BIC%)(see Fig. 1). The presented mean values were calculatedfrom three measurements available for both types ofimplant surfaces (B,M).

Fig. 1. The histometric analyses. Calculation of the bone–implant contact

(BIC,%).

3. Results and discussion

Histological examination of the bone–implant interfacewas performed for both types of tested implant surfaces 2,5 and 12 weeks after implantation. Three pairs of implantsfrom each group were evaluated for each time interval. Thecervical part of the implants was mostly surrounded bycortical bone, while the thread part is surrounded bytrabecular bone. An intimate BIC was frequently observedat the cervical part of the implants with the Bio surface. Incontrast, the specimens with turned, machined surfacesshowed patchy implant–bone contacts and intermediatesoft tissue was indicated in some cases.The time development of the BICs% at the thread part of

the implant was evaluated histometrically. The results of thehistometric analysis (mean BIC%(B)7SD and meanBIC%(M)7SD) are presented in Fig. 2. Friedman ANOVArevealed statistically significant differences in BIC%(B) (testgroup) (p ¼ 0.029) as well as in BIC%(M) (control group)(p ¼ 0.032) throughout the measured intervals during a 12-week follow up. The BIC of the Bio surface increasedsharply during the first 2 weeks in contrast to the turned,machined surface, which exhibited a gradual increasestarting at a later follow-up time (cf. Fig. 2). Using theMann–Whitney U-test, statistically significant differencesbetween the test and control groups were observed after2(p ¼ 0.046), 5(p ¼ 0.049) and 12 (p ¼ 0.049) weeks.The time dependence of the differences [BIC%IC%(M)]

in the first 5 weeks of follow-up was evaluated by the methodof linear regressions and the parameters of the straight line(slope and intercept) were determined: BIC(B)�BIC(M) ¼7.93859*t (weeks)+10.9210. The positive slope D [BIC%(B)�BIC%(M)]/Dt ¼ 7.93859 BIC%/week)t ¼ 0.5 of the regres-sion line with correlation coefficient R ¼ 0.7561 was foundto be statistically significant (p ¼ 0.013).This study presents the results of measurement of

changes in the BIC during healing of implants with Bio

0

10

20

30

40

50

60

70

0 10 15

BIC

%

*

*

*

M

weeks

5

Fig. 2. Mean values of the bone–implant contact (BIC%67SD) for the—

Bio implants (test group B) and implants with machined surfaces (control

group M) 2, 5 and 12 weeks after implantation. (� significant difference

(po0.05) at 95% confidence level).

ARTICLE IN PRESS

Table 1

Mean roughness parameters of Bio and turned, machined implant surfaces measured at 3 different sites of implant threads

Surface treatments Sites of measurement Sa (SD) (mm) Sq (mSD) (mm) Sdr (mSD) (%) Sds (mSD) (mm�2)Gaussian filtre size 50� 50 mm

Bio surface Top 1.11(0.13) 1.50(0.19) 15.90(4.5) 0.019(0.002)

Valley 1.32(0.27) 1.72(0.36) 18.28(7.46) 0.022(0.003)

Flank 1.15(0.14) 1.55(0.20) 19.99(6.88) 0.031(0.003)

Turned, machined surface Top 0.57(0.14) 0.73(0.17) 8.35(3.06) 0.025(0.005)

Valley 0.28(0.03) 0.36(0.03) 2.5(0.04) 0.051(0.009)

Flank 0.50(0.11) 0.68(0.18) 8.29(4.10) 0.034(0.004)

Sa-arithmetic average height deviation; Sq— root mean square of height deviation; Sdr—developed surface ratio, Sds—the number of summits in a unit

sampling area.

Table 2

Contact angles and specific surface areas of Bio and turned, machined

surfaces

Surface

modification

Contact angle

G7SD(deg.)

Specific surface area G7SD

(mm2/mm2)

Turned,machined

surface

79.574.6 1.470.7

Bio-surface 27.276.9 138.0742.5

J. Strnad et al. / Journal of Physics and Chemistry of Solids 68 (2007) 841–845844

surface and a turned, machined surface as a reference.These findings demonstrated that the Bio surface morerapid formation of the BIC in the early stages of healing. Itcan be speculated that the differences in the rates ofosseointegration in the initial stages of healing for theBio and machined surface could be related to differentsurface reactivities [44] following from the different surfacematerial properties, e.g. surface area or surface wettability[45]. In general, the surface reactivity, which is a commoncharacteristic of bioactive materials, increases with increas-ing surface area. Therefore, the three-dimensional macro-,micro- and nano-structured Bio surface, which is morethan 100� larger (Table 1) compared to the machinedsurface, may significantly enhance the surface reactivitywith the surrounding ions, amino acids, and proteins,which determine the initial cellular events at the cell–material interface. In addition, easily wettable hydrophilicBio surface (Table 1) allows the establishment of thecontact between the body environment (blood) and thecomplicated rough and porous structure of the implant,and thus contribute to cell and biomolecule migration andadhesion [46]. Moderately, hydrophilic surfaces (20–401water contact angle) were also shown to promote thehighest levels of cell attachment [47] (Table 2).

The Bio surface, which is rich in hydroxyl groups, incontrast to the machined surface (cf. Fig. 1), rapidly inducesadsorption of calcium and phosphate ions on contact withthe ions of the blood plasma [48]. The calcium phosphate-rich layer promotes adsorption and concentration of proteins[49] and constitutes a suitable substrate for the first apatitestructures of the bone matrix, which are synthesized by the

osteogenic cells at the beginning of the formation of the newbone tissue. This mechanism can accelerate the formation ofa stable bone–implant interface, formed by fusion of thebiological cement line matrix with the reactive calciumphosphate layer on the surface.

4. Conclusion

It was demonstrated that sand-blasted, acid etched andalkali etched titanium surface (Bio) enhances the formationof BIC during the first 2–5 weeks of healing when comparedto the machined-turned titanium surface. This phenomenonis likely to be related to the increased roughness andchemical reactivity of the Bio surface. Further investigationis necessary to differentiate the contribution of both surfaceroughness and surface chemistry.

Acknowledgement

Study was supported by the Project no. FT-TA/087 ofthe Grant Agency of Industry and Trade Ministry of theCzech Republic as well as by the Project no. A100100639of the Grant Agency of Academy of Sciences of the CzechRepublic and by the research project of West BohemianUniversity by MSMT no. 4977751303.

References

[1] L.L. Hench, J. Wilson, An Introduction to Bioceramics, World

Scientific Publishing Co., New Jersey, London, Hong Kong, 1993,

Singapore, 1999.

[2] L.L. Hench, Bioceramics: from concept to clinic, J. Am. Ceram. Soc.

74 (1991) 1487–1570.

[3] P.I. Branemark, Intraosseous anchorage of dental prostheses, Scand.

J. Plast. Reconstr. Surg. 3 (1969) 81–93.

[4] P. Li, L.L. Hench, An investigation of bioactive glass, J. Appl.

Biomater. 2 (1991) 231–239.

[5] P. Li, Bioactive glass-ceramics, J. Biomed. Mater. Res. 29 (1995)

325–328.

[6] Z. Strnad, Role of the glass phase in bioactive glass-ceramics,

Biomaterials 13 (1992) 317–321.

[7] N. Koga, J. Strnad, Z. Strnad, J. Sestak, Thermodynamics of

non-bridging oxygen in silica bio-compatible glass-ceramics,

ARTICLE IN PRESSJ. Strnad et al. / Journal of Physics and Chemistry of Solids 68 (2007) 841–845 845

mimetic material for the bone tissue substitution, J. Thermal Anal.

Calorimetry 71 (2003) 927–937.

[8] Z. Strnad, J. Sestak, Bio-compatible Ceramics, in: J. Meel (Ed.),

invited plenary lecture at Third IPMM (Intelligent Processing and

Manufacturing of Materials) in Vancouver, 2001, Proceedings by

Vancouver University, Canada, 2001.

[9] E.A. Ohtsuki, Mechanism of apatite formation on CaO–SiO2–P2O5

glasses in SBF, J. Noncryst. Solids 143 (1992) 84–92.

[10] P. Li, Effects of ions in aqueous media on HA induction, J. Appl.

Biomater. 4 (1993) 221 AW.

[11] P.N. de Aza, Bioceramics–sbf interfaces, J Mater. Sci. Mater. Med. 7

(1996) 399.

[12] P.N. Hanawa, Titanium and its oxide film, a substrate for formation

of apatite, in: J.E. Davies (Ed.), The Bone–biomaterial Interface,

University of Toronto Press, 1990, p. 33.

[13] P. Li, P. Ducheyne, Quasi-biological apatite film induced by titanium

in a simulated body fluid, J. Biomed. Mater. Res. 5, 41(3) (1998)

341–348.

[14] A. Wennerberg, T. Albrektsson, C. Johansson, B. Andersson,

Experimental study of turned and grit-blasted screw-shaped implants

with special emphasis on effects of blasting material and surface

topography, Biomaterials 17 (1) (1996) 15–22.

[15] K. Gotfredsen, T. Berglundh, J. Lindhe, Anchorage of titanium

implants with different surface characteristics: an experimental study

in rabbits, Clin. Implant. Dent. Relat. Res. 2 (3) (2000) 120–128.

[16] D.L. Cochran, D. Buser, C.M. Bruggenkate, D. Weingart,

T.M. Taylor, J.P. Bernard, F. Peters, J.P. Simpson, The use of

reduced healing times on TI implants with a sandblasted and acid-

etched (SLA) surface: early results from clinical trials on ITI SLA

implants, Clin. Oral Implants Res. 13 (2) (2002) 144–153.

[17] P.R. Klokkevold, P. Johnson, S. Dadgostari, A. Caputo, J.E. Davies,

R.D. Nishimura, Early endosseous integration enhanced by dual acid

etching of titanium: a torque removal study in the rabbit, Clin. Oral

Implants Res. 12 (4) (2001) 350–357.

[18] W. Khang, S. Feldman, C.E. Hawley, J. Gunsolley, A multi-center

study comparing dual acid-etched and machined-surfaced implants in

various bone qualities, J. Periodontol. 72 (10) (2001) 1384–1390.

[19] R.J. Lazzara, Bone response to dual acid-etched and machined

titanium implant surfaces, in: J.E. Davies (Ed.), Bone Engineering,

Emsquared Inc, Toronto, 2000, pp. 381–388.

[20] J. Strnad, J. Sestak, Z. Strnad, Thermodynamic properties of

potentially bioactive materials: part III—surface characteristics,

J. Thermal Anal. Calor. 2007, in press.

[21] G. Corioli, Z. Majzoub, A. Piatelli, Romoval torque and histomo-

phometric investigation of 4 different titanium substrates: an

experimental study in the rabbit tiba, Int. J. Oral. Maxillofac.

Implants 15 (5) (2000).

[22] T. Albrektsson, A. Wennerberg, Oral implant surfaces: Part I—review

focusing on topographic and clinical properties of different surfaces

and in vivo responses, Int. J. Prosthodont. 17 (5) (2004).

[23] Y.T. Sul, C. Johansson, A. Wenneberg, L. Cho, B. Chang,

T. Albrektsson, Optimum surface properties of oxidized IMplants

for reinforcement of osseointegration surface chemistry, oxide

thickness, porosity, roughness and crystal structure, Int. J. Oral

Maxillofac. Implants 20 (2005) 349.

[24] J.R. Jones, Observing cell response to biomaterials, Materials Today

9, 2006, p. 34, Biomaterials 27, 2006, p. 964.

[25] P. Li, P. Ducheyne, Quasi-biological apatite film induced by titanium in

a simulated body fluid, J. Biomed. Mat. Res. 5, 41(3) (1998) 341–348.

[26] A. Wennerberg, T. Albrektsson, Suggested guidelines for the

topographic evaluation of implant surfaces, Int. J. Oral. Maxillofac.

Implants 5 (3) (2000).

[27] S. Vercaigne, J.G. Wolke, I. Naert, J.A. Jansen, Bone healing

capacity of titanium plasma-sprayed and hydroxylapatite-coated oral

implants, Oral Implants Res. 9 (1998) 261–271.

[28] L. Sun, C.C. Berndt, K.A. Gross, A. Kucuk, Material fundamentals

and clinical performance of plasma-sprayed hydroxyapatite coatings:

a review, J. Biomed. Mater. Res. 58 (2001) 570–592.

[29] Z. Strnad, J. Strnad, C. Povysil, K. Urban, Effect of plasma sprayed

hydroxyapatite coating on osteoconductivity of titanium implants,

Int. J. Oral Maxillofac. Implants 15 (2000) 483–490.

[30] H. Liao, B. Fartash, J. Li, Stability of hydroxyapatite coatingson

titanium oral implants (IMZ): 2 retrieved cases, Clin. Oral Implants

Res. 8 (1) (1997) 68–72.

[31] M.D. Rohrer, R.R. Sobczak, H.S. Prasad, H.F. Morris, Postmor-

temhistologic evaluation of mandibular titanium andmaxillary

hydroxyapatite-coated implants from one patient, Int. J. Oral

Maxillofac. Implants 14 (4) (1999) 579–586.

[32] M.S. Block, D. Gardiner, J.N. Kent, D.J. Misiek, I.M. Finger,

L. Guerra, Hydroxyapatite-coated cylindrical implants in the poster-

ior mandible: 10-year observations, Int. J. Oral Maxillofac. Implants

11 (5) (1996) 626–633.

[33] S.L. Wheeler, Eight-year clinical retrospective study of titanium

plasma-sprayed and hydroxyapatite-coated cylinder implants, Int.

J. Oral Maxillofac. Implants 11 (3) (1996) 340–350.

[34] N.C. Geurs, R.L. Jeffcoat, E.A. McGlumphy, M.S. Reddy,

M.K. Jeffcoat, Influence of implant geometry and surface character-

istics on progressive osseointegration, Int. J. Oral Maxillofac.

Implants 17 (6) (2002) 811.

[35] M.A. McGlumphy, L.J. Peterson, P.E. Larsen, M.K. Jeffcoat, Prospec-

tive study of hydroxyaptite-coated cylindric omniloc implants placed in

121 patients, Int. J. Oral Maxillofac. Implants 18 (1) (2003) 82–92.

[36] D. Schwartz-Arad, O. Mardinger, L. Levin, A. Kozlovsky,

A. Hirshberg, Marginal bone loss pattern around hydroxyapatite

coated versus commercially pure titanium implants after up to 12 years

of follow-up, Int. J. Oral Maxillofac Implants 20 (2) (2005) 238–244.

[37] J. Strnad, J. Protivınsky, D. Mazur, K. Veltruska, Z. Strnad,

A. Helebrant, J. Sestak, Interaction of acid and alkali treated

titanium with dynamic simulated body environment, J. Thermal

Anal. Cal. 76 (2004) 17–31.

[38] A. Simunek, J. Strnad, J. Novak, Z. Strnad, D. Kopecka,

R. Mounajjed, STI-Bio titanium implants with bioactive surface

design, Clin. Oral. Impl. Res. 12 (2001) 393–421.

[39] A. Simunek, J. Strnad, A. Stepanek, Bioactive titanium implants for

shorter healing period, Clin. Oral Impl. Res. 13 (4) (2002).

[40] A. Stepanek, A. Simunek, J. Strnad, Z. Strnad, Early loading

(4 weeks) of dental implants with bioactive surface in maxilla and

mandible monitoring of the healing process using resonance

frequency analysis, Quintessenz 14 (2005) 51–58 [in Czech].

[41] A. Simunek, D. Kopecka, J. Strnad, Shortening of the healing time

for implants with bioactive surface, Quintessenz 13 (2004) 34–38.

[42] J.E. Ellingsen, Improved retention and bone to implant contact with

fluoride modified titanium implants, Int. J. Oral Maxillofac. Implants

19 (2004) 659–666.

[43] J.E. Ellingsen, On the properties of surface-modified titanium, in: J.E.

Davies (Ed.), Bone Eng., Em squared Inc, Toronto, 2000, pp. 183–189.

[44] P. Ducheyne, Q. Qiu, Bioactive ceramics: the effect of surface

reactivity on bone formation and bone cell function, Biomaterials 20

(1999) 2287–2303.

[45] B.G. Keselowsky, D.M. Collard, A.J. Garcıa, Surface chemistry

modulates focal adhesion composition and signaling through changes

in integrin binding, Biomaterials 25 (2004) 5947–5954.

[46] M. Lampin, R. Warocquier-Cclerout, C. Legris, Correlation between

substratum roughness and wettability, cell adhesion and cell

migration, Biomed. Mater. Res. 36 (1997) 99–108.

[47] K. Webb, V. Hlady, P.A. Tresco, Relative importance of surface

wettability and charged functional groups on NIH 3T3 fibroblast

attachment, spreading, and cytoskeletal organization, J. Biomed

Mater. Res. 41 (1998) 422–442.

[48] J. Strnad, Kinetics of hydroxyapatite formation on inorganic

bioactive materials, Ph.D. Thesis, Prague Institute of Chemical

Technology, Czech Republic 2004.

[49] A. El-Ghannam, P. Ducheyne, I.M. Shapiro, Formation of surface

reaction products on bioactive glass and their effects on the

expression of the osteoblastic phenotype and the deposition of

mineralized extracellular matrix, Biomaterials 18 (1997) 295–303.

Introduction

More than a quarter century ago two material groups

have been found to be able to form a mechanically

stable and functional interface with bone. One group

consisted of certain soda-lime-silica glasses, with or

without addition of phosphorus oxide, and the first

glass exhibiting the bone-bonding ability was discov-

ered by Hench [1–3] and registered under the name

'Bioglass'. The glasses exhibiting the bone bonding

ability were designated as bioactive, with the follow-

ing definition: ‘the bioactivity is the characteristics of

an implant material which allows it to form a bond

with living tissues’.

Another material found to exhibit the

bone-bonding ability was machined titanium. This

characteristic of titanium was first described by

Branemark and coworkers [4] and the phenomenon of

attachment to bone was named osseointegration with

the following definition: ‘osseointegration represents

the formation of a direct contact of a material with

bone without intermediate fibrous tissue layer, when

observed using light microscope’.

The major difference between the two material

groups was represented by the materials reactivity and

related kinetics of the bone-material interface formation.

Very reactive bioactive glasses formed a stable interface

with bone within days where as machined titanium re-

quired healing periods of several months to reach the

same bone-implant contact. Thanks to the high reactiv-

ity of bioactive glasses, the course of the glass-body en-

vironment interaction could be investigated more easily,

and the bone bonding ability was described and quanti-

fied in direct relation to the glass composition [5–7].

Soon other materials like hydroxyapatite, sol/gel pre-

pared glasses or glass-ceramics, which were found to

exhibit similar bone-bonding ability, were included into

the ‘bioactive’ group [8–10]. All these materials exhib-

ited the precipitation of apatite mineral on their surfaces

as a result of chemical interaction of the materials sur-

face with body environment [1]. For initial assessment

of potential bioactivity of a material, the apatite forma-

tion occurring in vivo was reproduced in vitro in simu-

lated body environment [11] and a good correlation of

in vivo and in vitro results was found for the above men-

tioned, highly bioactive materials [1, 9].

The investigation of the osseointegration ob-

served in the titanium group was more difficult be-

cause of the low reactivity of titanium in body envi-

ronment. Changes detected in the passive oxide layer

during the exposure to body environment were not

comparable in magnitude to those occurring in

bioactive materials. With the intention to increase the

implant surface available for bone in-growth and fixa-

tion and/or to increase the blood clot retention on the

materials surface, machined titanium is most often

modified by sand-blasting, plasma spraying or acid

etching. Roughened surfaces showed higher bone/im-

plant interface strength (removal torque values) and

higher BIC (=bone/implant contact) after the same

1388–6150/$20.00 Akadémiai Kiadó, Budapest, Hungary

© 2007 Akadémiai Kiadó, Budapest Springer, Dordrecht, The Netherlands

Journal of Thermal Analysis and Calorimetry, Vol. 88 (2007) 3, x–x

PHYSICO-CHEMICAL PROPERTIES AND HEALING CAPACITY OFPOTENTIALLY BIOACTIVE TITANIUM SURFACE

J. Strnad1*, Z. Strnad1 and J. �esták2

1LASAK – Laboratory for Glass and Ceramics, Papírenská 25, 16200 Praha 6, Czech Republic2Institute of Physics, Academy of Sciences of the Czech Republic, Cukrovarnická 10, 162 53 Prague 6, Czech Republic

The aim of this study was to evaluate physico-chemical properties and the healing capacity of surface treated titanium. Surface

treatment combining sand-blasting, acid etching and alkaline etching (BIO surface) was evaluated together with machined titanium

as a reference surface. Hydration, wetting angle, surface area and roughness parameters were evaluated for both surfaces. Stability

of dental implants with both surfaces implanted in the tibia of dog was measured during the healing of twelve weeks. BIO surface

exhibited lower wetting angle, larger surface area, higher degree of hydration and higher average roughness compared to machined

titanium. Implants with the BIO surface maintained their stability during the whole healing period in contrast to those with ma-

chined titanium surface, which showed a statistically significant decrease in stability three and nine weeks after implantation.

Keywords: alkali-etched surface, bioactive surface, dental implants, hydroxyapatite, osseointegration,resonance-frequency analysis, secondary stability, titanium, wetting angle

* Author for correspondence: [email protected]

healing time compared to machined titanium [12–17].

Some authors suggested that surfaces with the mean

roughness parameter Sa in the interval of 1.0–1.5 �m

show stronger bone response than smoother or

rougher implant surfaces [18]. On the other hand it

was shown that the optimal roughness value varies

according to the chemical composition of the tested

surfaces [19, 20]. Acceleration of the bone-implant

interface and the increase of the bone-implant contact

in the early phases of healing [21–23] was success-

fully achieved by plasma spraying of a

hydroxyapatite as a bioactive material on the titanium

substrate, however, hydroxyapatite plasma sprayed

coatings have been the subject of many controversies

regarding their long-term stability and thickness as

well as low long-term success rates [24–27]. On the

other hand some authors reported mid and long term

clinical results showing high success rates of

hydroxyapatite coated implants [28–30].

As lately as in the late nineties specific surface

modifications of machined titanium have been devel-

oped with the intention to modify the reactivity of tita-

nium chemically so that it gains the best defined charac-

teristics of well proven bioactive materials - the ability

to induce apatite mineral formation in vitro. Until now,

to the knowledge of the authors, only two surface modi-

fications of titanium supporting the precipitation of

bone mineral apatite have been developed and clinically

introduced. In 1999, alkali treatment [31] was used in

combination with sand-blasting and acid etching on

LASAK implants (BIO surface) [32, 33] and in 2000

fluoridated titanium surface (Osseospeed) was intro-

duced by ASTRATECH [34, 35].

From the present scientific literature it seem nec-

essary to evaluate both surface roughness and chemis-

try in order to be able to draw reliable conclusions re-

garding the effect of the surface treatment on the

bone-implant interface formation. A method of quan-

titative evaluation of the clinical benefit of a surface

treatment has not yet been established. Machined sur-

face with defined surface roughness is often used as a

reference when evaluating the effect of a specific sur-

face treatment. The aim of this study is to evaluate the

physicochemical properties and the stability time de-

pendence during healing of potentially bioactive tita-

nium surface and to compare it with the machined

surface as a reference.

Experimental

Sample preparation

This study was designed as a comparative study of two

surfaces: turned machined titanium surface (Ti-M) and

potentially bioactive titanium surface (BIO). The BIO

surface is created by sand-blasting, acid etching and a fi-

nal treatment in an alkaline solution. Ti-M and BIO

surfaces were prepared on implants used for the stability

time dependence measurement.

Methods of surface characterization

Surface characterization was performed using a scan-

ning electron microscope (SEM, Hitachi, Japan) with an

accelerating voltage of 15–30 kV. Surface roughness

measurement was carried out using a Talysurf 6

profilometer (Taylor Hobson, Leicester, United King-

dom). The mean roughness (Ra) was measured and re-

corded at a traverse speed of 0.5 mm s–1 on a traverse

length of 2 mm with a diamond-tipped stylus. The spe-

cific surface area was calculated by the BET method

from the results of the krypton gas absorption study. The

surface area is expressed in relation to unit geometric

surface area of the implant. The determination was per-

formed by absorption of krypton on a ASAP 2010 M in-

strument (Micromeritics, USA). Diffuse.

Reflectance Infrared Fourier Transformed

(DRIFT) Spectroscopy was used to determine the de-

gree of surface hydration. The measurement was per-

formed on a Nicolet 740 instrument (Nicolet Madi-

son, USA) with resolution of 4 cm–1. The presence of

hydroxyl groups on the sample surface was quantified

using the absorption band height at 3400 cm–1 [36].

Dynamic contact angle measurement was performed

using the Wilhelmy plate method, using

Tensiometer K15 (Kruss GmbH , Germany). The wet-

ting angle values in water were determined from the

dependence of the wetting force on the immersion

depth. The mean values of the wetting angle were cal-

culated from four repeated measurements.

To evaluate the stability time dependence for im-

plants with TI-M and BIO surfaces the resonance fre-

quency analysis (RFA, Integration Diagnostics,

Gothenburg, Sweden) was performed using an animal

model and a commercially available F37 L5 trans-

ducer attached to the implant. The Implant stability

measurement was performed in the 1st, 3rd, 9th and 12th

week of healing time. A total of 6 implants per time

point and per treatment group were used for the mea-

surement. Implants were placed in a dog tibia with the

same average primary stability for both groups. The

experimental procedure was described earlier [37].

Results and discussion

BIO surface treatment alters the initial, low-rough-

ness surface of the machined titanium

(Ra=0.828�0.024 �m) to a rough surface

(Ra=2.264�0.311 �m) (Fig. 1, Table 1)) with micro-

and nano-porous gradient structure.

2 J. Therm. Anal. Cal., 88, 2007

STRNAD et al.

The surface area measurements showed that

combination of sand-blasting and acid and alkali

treatments used in the BIO surface preparation results

in a structured porous surface exhibiting an approx.

100-fold increase in surface area compared to the ma-

chined surface (Table 2).

Table 2 summarizes the results of the contact an-

gle measurements for the machined and BIO surface.

The machined surface exhibited hydrophobic charac-

teristics. On the other hand, the low contact angle of

the BIO surface (27.2°) indicates an easily wettable

hydrophilic surface.

The diagram in Fig. 2 shows that the number of

hydroxyl groups is greater on the BIO surface com-

pared to the machined surface. The level of

hydroxylation of the BIO surface was roughly an or-

der of magnitude greater (1.3·108) compared to the

other surfaces (Ti-Unite-2.4·107; SLA 3.0·107).

The time dependence of the stability for implants

with machined surfaces ISQt(M) (control group) ex-

hibits a statistically significant decrease after three

(p=0.0098) and nine (p=0.0082) weeks of healing and

this decrease in the stability ends in the 9th week

(Fig. 3). The stability increased again after the ninth

week and returned to almost the initial values in the

twelfth week. In contrast to this, the ISQt(B) values of

the BIO implants (test group) did not exhibit statisti-

cally significant changes (p>0.05) throughout the

measured intervals between 0 and 1, 3, 9 and

12 weeks of healing time (Fig. 3).

The time dependence (Fig. 3) indicates that the

BIO surface helps to maintain implant stability during

the early healing time compared to the machined sur-

face. A statistically significant difference between the

test and control groups was observed after 3

(p=0.0064) and 9 (p=0.00019) weeks.

The net contribution of the BIO surface to the

implant stability (with reference to machined sur-

faces) during the healing time (t=1; 3; 9 weeks) was

estimated as the difference between the BIO implant

stability ISQt(B) and the machined implant stability

ISQt(M). The time dependence of the stability differ-

ences exhibits a statistically significant positive slope

(0.84) of the linear regression line in the first nine

weeks of follow-up (Fig. 4).

The slope value (0.84 ISQ/week) represents the

difference in the rate of osseointegration between the

J. Therm. Anal. Cal., 88, 2007 3

POTENTIALLY BIOACTIVE TITANIUM SURFACE

200 m

a b

200 m

2 m 2 m

Fig. 1 Scanning electron microscopic images of the machined

surface a – TI-M and the b – BIO surface

Table 1 Roughness parameters of BIO and machined surfaces

Ra�SD/�m Sm�SD/�m Rku�SD

BIO surface 2.264�0.311 86�12 3.7�1.7

Machined surface 0.828�0.024 44�3 4.2�1.3

Ra=arithmetic mean of the profile departures from the mean line; Sm=mean spacing of adjacent local peaks; Rku=profile sharpness

Table 2 Contact angles and specific surface areas of BIO and machined surfaces

Surface modificationContact anglemean�SD/°

Specific surface areamean�SD/mm2 mm–2

machined surface 79.5�4.6 1.4�0.7

BIO-surface 27.2�6.9 138.0�42.5

–5.00E+07

0.00E+00

5.00E+07

1.00E+08

1.50E+08

2.00E+08

3050315032503350345035503650

BIO surface

Ku

bel

ka–

Mu

nk

un

its

Wavenumber/cm–1

Fig. 2 DRIFT spectra of the BIO surface (machined surface as

a reference)

implants with BIO surface and those with machined sur-

face (�ISQ(B)/�t–�ISQ(M)/�t), where �ISQ corre-

sponds to the changes in ISQ values and �t represents

the corresponding interval of the healing time of im-

plants in the bone. Taking machined implants as a refer-

ence, the slope value can be regarded as a stability

growth rate gained by the BIO surface modification.

It can be speculated that the differences in the

rates of osseointegration in the initial stages of heal-

ing for the BIO and machined surface could be related

to different surface reactivities [20] following from

the different surface material properties, e.g. specific

surface area, surface wettability, surface contact an-

gle or surface hydroxylation/hydration [38]. In gen-

eral, the surface reactivity, which is a common char-

acteristic of bioactive materials, increases with in-

creasing specific surface area. Therefore, the three-di-

mensional macro-, micro- and nano-structured BIO

surface, which is more than 100× larger (Table 2)

compared to the machined surface, may significantly

enhance the surface reactivity with the surrounding

ions, amino acids, and proteins, which determine the

initial cellular events at the cell-material interface.

In addition, easily wettable hydrophilic BIO sur-

face (Table 2) allows the establishment of the contact

between the body environment (blood) and the com-

plicated rough and porous structure of the implant,

and thus contribute to cell and biomolecule migration

and adhesion [39]. Moderately hydrophilic surfaces

(20–40° water contact angle) were also showed to

promote the highest levels of cell attachment [40].

The BIO surface, which is rich in hydroxyl groups,

in contrast to the machined surface (Fig. 2), rapidly in-

duces adsorption of calcium and phosphate ions on con-

tact with the ions of the blood plasma [41]. The calcium

phosphate-rich layer promotes adsorption and concen-

tration of proteins [42] and constitutes a suitable sub-

strate for the first apatite structures of the bone matrix,

which are synthesized by the osteogenic cells at the be-

ginning of the formation of the new bone tissue. This

mechanism can accelerate the formation of a stable

bone-implant interface [43–45], formed by fusion of the

biological cement line matrix with the reactive calcium

phosphate layer on the surface.

Conclusions

Implants with the BIO surface maintained their stabil-

ity during the whole healing period in contrast to

those with machined titanium surface, which showed

a statistically significant decrease in stability three

and nine weeks after implantation. The BIO surface

exhibits more favorable values of the major surface

characteristics compared to the machined surface.

Acknowledgements

Study was supported by the project No: FT-TA/087 of the

Grant Agency of the Industry and Trade Ministry of Czech

Republic and the project No A100100639 of the Grant Agency

of the Academy of Sciences of Czech Republic as well as by

the research project of FZU No. AV0210100521.

References

1 L. L. Hench and J. Wilson, Eds, An introduction to

bioceramics, World Scientific, Singapore 1999, pp. 1–23

and 41–62.

2 L. L. Hench, J. Am. Ceram. Soc., 74 (1991).

3 L. L. Hench, Thermochim. Acta, 280/281 (1996) 1.

4 P.-I. Branemark, B. E. Breine, R. Adell, B. O. Hansson,

J. Lindstrom and J. A. Olsson, Scand. J. Plast. Reconstr.

Surg., 3 (1969) 81.

5 N. Koga, Z. Strnad, J. �esták and J. Strnad, J. Therm.

Anal. Cal., 71 (2003) 927.

6 R. Li, A. E. Clark and L.L. Hench, J. Appl. Biomater., 2

(1991) 231.

7 Z. Strnad, Biomaterials, 13 (1992) 317.

8 P. Li, X. Ye, I. Kangasniemi,

J. M. A. de Blieck-Hogervorst, C. P. A. T Klein and

K. de Groot, J. Biomed. Mater. Res., 29 (1995) 325.

4 J. Therm. Anal. Cal., 88, 2007

STRNAD et al.

0 1 3 9 12

Time/week

62

64

66

68

70

72

74

76

78

80

BIO surfacemachined surface

ISQ

Fig. 3 The stability time dependence for the implant with

— – BIO surface and - - - – machined surface TI-M

(*p<0.05)

Fig. 4 Time dependence of the net contribution of the BIO sur-

face to the implant stability with the linear regression

line: ICQB–ICQM=0.84 weeks+1.01 with its boundaries

of reliability. Correlation coefficient R=0.76

9 R. Z. LeGeros, I. Orly, M. Gregoire and G. Daculsi,

The Bone-Biomaterial Interface, J. E. Davies, Ed., Univ.

Toronto, Toronto 1991, p. 76.

10 T. Kokubo, S. Ito, M. Shigematsu, S. Sakka and

T. Yamamuro, J. Mater. Sci., 22 (1987) 4067.

11 C. Ohtsuki, T. Kokubo and T. Yamamuro, J. Non-Cryst.

Sol., 143 (1992) 84.

12 A. Wennerberg, T. Albrektsson, C. Johansson and

B. Andersson, Biomaterials, 17 (1996) 15.

13 K. Gotfredsen, T. Berglundh and J. Lindhe, Clin. Implant

Dent. Relat. Res., 2 (2000) 120.

14 D. L. Cochran, D. Buser, C. M. Bruggenkate,

D. Weingart, T. M. Taylor, J. P. Bernard, F. Peters and

J. P. Simpson, Clin. Oral Implants Res., 13 (2002) 144.

15 P. R. Klokkevold, P. Johnson, S. Dadgostari, A. Caputo,

J. E. Davies and R. D. Nishimura, Clin. Oral Implants

Res., 12 (2001) 350.

16 W. Khang, S. Feldman, C. E. Hawley and J. Gunsolley,

J. Periodontol., 72 (2001) 1384.

17 R. J. Lazzara, Bone engineering, J. E. Davies, Ed.,

Em Squared Inc., Toronto 2000, p. 381.

18 T. Albrektsson and A. Wennerberg, Int. J. Prosthodont.,

17 (2004) 536.

19 Y. T. Sul, C. Johansson, A. Wennerberg, L. R. Cho,

B. S. Chang and T. Albrektsson, Int. J. Oral. Maxillofac.

Implants, 20 (2005) 349.

20 P. Ducheyne and Q. Qiu, Biomaterials, 20 (1999) 2287.

21 S. Vercaigne, J. G. Wolke, I. Naert and A. J. Jansen, Clin.

Oral Implants Res., 9 (1998) 261.

22 L. Sun, C. C. Berndt, K. A. Gross and A. Kucuk,

J. Biomed. Mater. Res., 58 (2001) 570.

23 Z. Strnad, J. Strnad, C. Pov��il and K. Urban, Int. J. Oral

Maxillofac. Implants, 15 (2000) 483.

24 H. Liao, B. Fartash and J. Li, Clin. Oral Implants Res.,

8 (1997) 68.

25 M. D. Rohrer, R. R. Sobczak, H. S. Prasad and H. F. Morris,

Int. J. Oral Maxillofac. Implants, 14 (1999) 579.

26 M. S. Block, D. Gardiner, J. N. Kent, D. J. Misiek,

I. M. Finger and L. Guerra, Int. J. Oral Maxillofac.

Implants, 11 (1996) 626.

27 S. L. Wheeler, Int. J. Oral Maxillofac. Implants,

11 (1996) 340.

28 N. C. Geurs, R. L. Jeffcoat, E. A. McGlumphy,

M. S. Reddy and M. K. Jeffcoat, Int. J. Oral Maxillofac.

Implants, 17 (2002) 811.

29 E. A. McGlumphy, L. J. Peterson, P. E. Larsen and

M. K. Jeffcoat, Int. J. Oral Maxillofac. Implants,

18 (2003) 82.

30 D. Schwartz-Arad, O. Mardinger, L. Levin, A. Kozlovsky

and A. Hirshberg, Int. J. Oral Maxillofac. Implants,

20 (2005) 238.

31 J. Strnad, J. Protivínsk�, D. Mazur, K. Veltruská,

Z. Strnad, A. Helebrant and J. �esták, J. Therm. Anal Cal.,

76 (2004) 17.

32 A. �im�nek, J. Strnad, J. Novák, Z. Strnad, D. Kopecká,

M. D. L. Cevallos and R. Mounajjed, STI-Bio Titanium

Implants with Bioactive Surface Design, Proceedings of

EAO Conference, Milano, Italy, 13–15.9.2001.

33 A. �im�nek, J. Strnad and A. �t�pánek, Clin. Oral

Implants Res., 13 (2002) 111.

34 J.E. Ellingsen, C. Johansson, A. Wennerberg and

A. Holmén, Int. J. Oral Maxillofac. Implants,

19 (2004) 659.

35 J. E. Ellingsen, Bone engineering, J. E. Davies, Ed.,

Em Squared Inc., Toronto 2000, p. 183.

36 Y. Djaoued, S. Badilescu, P. V. Ashrit and J. Robichaud,

Int. J. Vibr. Spec. [www.ijvs.com] 5, 6, 4 (2001).

37 J. Strnad, K. Urban and Z. Strnad, Clin. Oral. Implants

Res., 16 (2005) 111.

38 B. G. Keselowsky, D. M. Collard and A. J. García,

Biomaterials, 25 (2004) 5947.

39 M. Lampin, R. Warocquier-Clérout, C. Legris,

M. Degrange and M. F. Sigot-Luizard, J. Biomed. Mater.

Res., 36 (1997) 99.

40 K. Webb, V. Hlady and P. A. Tresco, J. Biomed. Mater.

Res., 41 (1998) 422.

41 J. Strnad, Kinetics of Hydroxyapatite Formation on

Inorganic Bioactive Materials, Ph.D. Thesis, Institute of

Chemical Technology, Prague 2004.

42 A. El-Ghannam, P. Ducheyne and I. M. Shapiro,

Biomaterials, 18 (1997) 295.

43 Z. Strnad and J. �esták, ‘Bio-compatible Ceramics’ in-

vited plenary lecture at the 3rd IPMM (Intelligent Pro-

cessing and Manufacturing of Materials) in Vancouver

2001, Proceedings by Vancouver University (J. Meel,

Ed.) Canada 2001, p. 123.

44 J. �esták, Heat, Thermal Analysis and Society, Nucleus,

Hradec Králové 2004, p. 318 (IBSN 80-86225-54-2).

45 B. Hlavá�ek, J. Strnad and J. �esták, Physics of Amor-

phous Materials, book in the course of preparation

(Pardubice University 2008).

DOI: 10.1007/s10973-006-8295-6

J. Therm. Anal. Cal., 88, 2007 5

POTENTIALLY BIOACTIVE TITANIUM SURFACE

BIOMEDICAL THERMODYNAMICS AND IMPLANTOLOGY ASPECTS OF

BIOCOMPATIBLE GLASS-CERAMICS AND OTHERWISE MODIFIED

INORGANIC MATERIALS AND SURFACES

J. Šesták1,3,a, Z. Strnad2,b, J. Strnad2,b , M. Holeček3,c and N. Koga4,d

1 Institute of Physics of the Academy of Sciences, Cukrovarnicka 10, CZ-16253 Praha,

Czech Republic 2 Laboratory for Glass and Ceramics (LASAK), Papírenská 25, 16000 Praha 6, 3 New Technology - Research Center in the West Bohemian Region, West Bohemian Uni-

versity, Universitni 8, CZ-30114 Pilsen 4 Chemistry Laboratory, Department of Science Education, Graduate School of Education,

Hiroshima University, 1-1-1 Kagamiyama, Hiroshima 739-8524, Japan [email protected], [email protected], [email protected], [email protected]

Keywords: bioactivity, ceramics, glass, surface, titanium, hydroxyapatite, osteointegration

Abstract: Some historical and recent attitudes aimed to bioactive inorganic materials are reviewed. The theory of bridging and non-bridging oxygen is reconsidered as well as the acid and alkali treatment of titanium surface necessary to achieve a good osteointegration. Some theoretical mod-els to characterize mechanical properties are also reviewed.

Bulk bio-glass-ceramics

Merely quarter century ago it was considered inconceivable that a man-made material [1-3]

could bond to living tissues because the selection of biomedical materials for use in the living body was dependent on those materials already available off the shelf and until a better understanding of body immune system many of them proved to be either pathogenic or even toxic. This understand-ing was irreversibly altered when a special composition of soda-lime-phosphate-silica glass was synthesized by Hench [4] and successfully implanted in the femurs of rats. About 6% of P2O5 was added to simulate the Ca/P constituents of hydroxyapatite, Ca10(PO4)6(OH)2, that is the inorganic mineral phase naturally existing in bones. The glasses exhibiting the bone bonding ability were des-ignated as bioactive, with the following definition: “the bioactivity is the characteristics of an im-

plant material which allows it to form a bond with living tissues”. The bioactivity [5-9] leads to both the osteoconduction and osteoproduction because of rapid reaction on the bioactive glass sur-face. The surface reactions [9-12] involve ionic dissolution of critical concentrations of soluble Si, Ca, Na and P ions that give rise to both the intracellular and extracellular responses at the interfaces of the glass with its physiological environment. The smartness of such a mimetic process is likely hidden in the activity of oxygen characterized by the action of silanol groups (Si-OH). They likely serve as nucleation sites for the bio-compatible interface formation capable to coexist between the original tissue and the implants, which can be expediently made from glass, glass-ceramics, ceram-ics or cements.

Empirical index of bioactivity, IB, was introduced by Hench [6, 13] as IB = 100/t0.5bb , where t0.5bb is the time for more than 50% of the implant interface to be bonded to bone. By reading the values of IB at various typical points of glass composition from the so far reported iso- IB plots in Na2O-CaO-SiO2 ternary system, the structural parameters, such as non-bridging X [O´] and bridg-ing Y [Oo] oxygens per polyhedron in the glass lattice can be derived on basis of the molar compo-sition. When X<1,5 (Y>2,5), the glass looses its bioactivity and bioactivity index comes close to IB~0. The value of X>1,5 (and Y<2,5) indicates the actual range of bioactive glasses where IB>0. Essentially linear dependence with negative intercept has been found between the mean number of non-bridging oxygen ions and bioactivity index. The compositional dependence of relative propor-

Advanced Materials Research Vols. 39-40 (2008) pp 329-334online at http://www.scientific.net© 2008 Trans Tech Publications, Switzerland

tion of non-bridging oxygen in the soda-lime-silica system was also calculated by applying methods currently used in the chemistry of organic polymers. It was assumed that the silicate ions are present as linear and branched chains see the equation below. The calculated relative proportion of [O´] in the Na2O-CaO-SiO2 glasses was compared with their ability to form the surface calcium phosphate layer in the simulated body liquid, which indicates its bioactivity [7]. The concept of non-bridging oxygens was already applied for various cases and treated under different theories [14, 15] which, we, earlier, evaluated when describing the oxygen states in the iron-rich borate glasses [16]. We used it as a simple and useful way for the description of bio-glass compositions and their structures [15]. These parameters can be readily calculated from the molar composition of glass assuming that Y = 2Z - 2R and X = 2R – Z where Z is the mean number of all types of oxygen per polyhedron, i.e., the mean coordination number of the glass-forming cations, and R is the ratio of the total number of oxygens to the total number of glass-forming cations in glass

The relative proportions of non-bridging oxygen ions O

-, bridging oxygen ions O

o and free oxygen

ions O2- in the binary (a) CaO-SiO2 and (b) Na2O-SiO2 systems [15].

A comparison of the relative proportion of non-bridging oxygen ions O

- in the Na2O-CaO-SiO2 ter-

nary system calculated in the reference [15] compared with the Iso- IB plots for the system enriched

by 6 wt% P2O5 as reported by Hench [16].

Surface treated inorganic substrates

It follows that the specific biological response is elicited at the interface with bioactive ma-terials with high surface reactivity resulting in the formation of a bond between the tissue and the material surface [17, 18]. Biomedical inorganic materials with well a adjusted bulk composition,

330 Glass – The Challenge for the 21st Century

such as bioactive glasses, glass-ceramics, hydrated silica and titanium gel-oxides or ordinary hy-droxyapatite, have the ability to provide a very rapid formation the stable interface with bone tissue through the creation of a calcium phosphate layer on their surfaces as a consequence of chemical interaction with body fluids. The mechanical strength of this interface usually exceeds the strength of the bone tissue to which bioactive material is bonded. However, because of their poor mechani-cal properties (glass brittleness), which prevent the use under load-bearing conditions; such bioac-tive materials (especially hydroxyapatite) are preferably applied to the surface of suitable supports (titanium) as to achieve better mechanical properties [19]. It means that the scientific interest moves from bulk materials to the surface layers, which can be formed by various methods on various sub-strates.

0 1 3 9 12

WEEKS

62

64

66

68

70

72

74

76

78

80

ISQ

BIO surface machined surface

Twelve pairs of screw were inserted without pre-tapping in the tibiae of three Beagle dogs. The resonance frequency analysis method was used to measure the implant stability quotient (ISQ). Time dependence of the mean stability for an implant with a special BIO surface (solid line, test group B) and analogous implants with mere machined surface TI-M (dashed line, test group M) at the instant of placement (0) and after 1, 3, 9 and 12 weeks of insertion.

To enhance secondary stability and accelerate the formation of stable and functional bone-implant interfaces, a number of implant surfaces have been developed. Surface morphology includ-ing deposition, etching or mere roughening is the most frequently studied property as a factor af-fecting secondary stability [17-22]. The surface chemistry is also thought to affect the secondary stability independently of the surface topography, although it is difficult to separate the effects of these two factors. There is ample experimental evidence that biomaterials with different chemical compositions trigger different biological responses.

M B B M B M

M B

Advanced Materials Research Vols. 39-40 331

Alternative material found to exhibit the bone–bonding ability became machined titanium, which shows to be most convenient dental material with suitable bone-implant contact

and mechanical properties. This characteristic of titanium was first described by Branemark and the phenomenon of attachment to bone was named osseointegration with the following defini-tion: “osseointegration represents the formation of a direct contact of a material with bone without

intermediate fibrous tissue layer, when observed using light microscope”. Surface modification of titanium by acid etching was introduced to support and speed-up the healing and bone formation processes around the implant. It creates a micro-rough texture that is able to retain the fibrin net-work of the blood clot, through which cells migrate to the surface of the implant, where the new bone is formed. Apparently, surface quality determines tissue reactions to an oral implant and its assets can be classified [23, 24] regarding: (1) mechanical (2) topographic (roughness, porosity) and (3) chemical properties. The latter bio-chemical bonding is related to bioactivity, which existence has often been questioned because there is not a clear evidence of separated effects of surface roughness and interfacial chemical reactions, which needs a clear definition. It can be speculated that the differences in the rates of osseointegration in the initial stages of healing for different sur-face can be related to different surface reactivity [5, 6] following from the different surface proper-ties, e.g. specific surface area, surface wettability, surface contact angle or surface hydroxyla-tion/hydration [23-25]. In general, the surface reactivity, as the most common characteristics of bio-active materials, increases with increasing specific surface area.

A specific procedure was developed to properly activate titanium surface by three-step sur-face treatment of implants that included sand-blasting, acid and alkali etching [23-25], abbreviated as BIO. Such a three-dimensionally macro-, micro- and nano-structured BIO surface, which is more than 100x larger than the machined surface [23], significantly enhances the surface reactivity with the surrounding ions, amino acids, and proteins, which determine the initial cellular events at the cell-material interface. In addition, more easily wettable hydrophilic BIO surface [24] allows the contact formation between the body environment (blood) and the complicated rough and porous structure of the implant, and thus contributes to cell and bio-molecule migration and adhesion. Moderately hydrophilic surfaces (20-40o water contact angle) were also showed to promote the highest levels of cell attachment [23]. The BIO surface, which is rich in hydroxyl groups in contrast to the machined surface, rapidly induces adsorption of calcium and phosphate ions on contact with the ions of the blood plasma. The calcium phosphate-rich layer promotes adsorption and concentra-tion of proteins and constitutes a suitable substrate for the first apatite structures of the bone matrix, which are synthesized by the osteogenic cells at the beginning of the formation of the new bone tissue. This mechanism can accelerate the formation of a stable bone-implant interface formed by fusion of the biological cement line matrix with the reactive calcium phosphate layer.

Modeling of mechanical properties

Hard and soft biological tissues are remarkable materials when regarding their ability to adapt the whole scale of external conditions by rapid changes of their physical properties. Concern-ing the mechanical properties, they are able to undergo considerable deformations without damag-ing their structure and return back to the initial state. Smooth muscle tissues, for example, are able to exert considerable internal forces changing mechanical properties of the tissue in several orders. It opens another important question, namely, how describe and model effectively important features of their design. The crucial role seems to play the cellular arrangement of tissues. Living cells are perfect controlling units with many remarkable properties. They are able to control precisely and reorganize quickly the local structural and mechanical properties of their cytoskeleton - a protein fiber network spanning any eukaryotic cell [26]. The cell construction resembles a perfect structure whose building blocks may change considerable and very quickly in dependence on external condi-tions. Though many physical and structural properties of cytoskeleton are well known, we are still very far from a detail understanding the cells' enormous ability to control very effectively the over-all properties of tissues. The main method of characterizing mechanical properties of living tissues


may be defined as follows. At the beginning, a macroscopic sample is in thermal equilibrium with-out an external load. Its state is defined by the temperature and a set of work parameters, A(0), defin-ing its configuration. Then the work parameters are changed with time by some external manipula-tion (e.g. the extension of the sample) during which the sample's surrounding serves as a heat reser-voir with the constant temperature. At the end of the process, the sample has the configuration A. The work performed on the sample during the process, W, depends essentially of the chemical com-position and the structure of the material and on a concrete realization of the process. Only if the process is reversible, W equals to the Helmholtz free energy difference ∆F = F (A) – F(A

(0)) be-

tween the initial and final state. Otherwise, we have only the inequality W ≥ ∆F. The dependence of the density of the free energy on the parameters A is thus an important characterization of the material because it measures a minimal value of performed work to reach a demand configuration of a unified material sample. The existence of a reversible path, however, is not trivial. Finite changes of configurations of many materials are unavoidably accompanied by irreversible processes, like motion of defects in polycrystallic materials. In living tissues, however, the existence of re-versible paths may be assumed because they function effectively in living organs, which are able to change considerably their configurations in many operating cycles. An estimation of the free energy of a living tissue in dependence on microstructural parameters is thus an effective tool in characteri-zation of their mechanical properties [27, 28].

The study was supported by the projects: No A100100639 of the Grant Agency of Academy of Sci-

ences, No 1M06031 of the Grant Agency MSMT (Ministry of Education) and No FT-TA/087 of the Grant

Agency of MOP (Ministry of Industry and Trade).

Literature

[1] Z. Strnad, “Glass-ceramic Materials”, Elsevier, Amsterdam 1984 [2] J. Šesták (ed) “Vitrification, Transformation and Crystallization of Glasses” Special issue of Thermochim. Acta, Vol. 280/281 (1996) [3] J. Šesták, “Art and horizon of non-equilibrated and disordered states and new phase formation” Glasstech. Berichte Glass Sci. Tech., 70 C (1997) 153 and ”Miracle of reinforced states of matter” J. Thermal Anal. Cal., 61 (2000) 305 [4] L.L. Hench, R.J. Splitner, W.C. Allen and T.K. Greenlee, J. Biomed. Mater. Res., 2 (1971) 117 and L.L. Hench and H.A. Paschal, J. Biomed. Mater. Res., 4 (1973) 25; 5 (1974) 49. [5] L.L. Hench, J. Wilson (eds.): “An introduction to bioceramics”. World Scientific, Singapore 1999 and L.L. Hench, J.M. Polak “Third-generation Biomedical Materials” Science 295 (2002) 1014 [6] L.L. Hench: “Bioceramics: From concept to clinic”. J. Amer. Ceram. Soc 74 (1991) 1487 [7] Z. Strnad, “Role of the glassy phase in bioactive ceramics” Bioceramics, 13 (1992) 317 [8] Z. Strnad, J. Šesták, Biocompatible glass-ceramics” key lecture at the 2nd Inter. Conference on Intelligent Processing and Manufacturing of Materials, Honolulu 1999, proc. p.123 [9] J. Šesták, “Heat, Thermal Analysis and Society”, Nucleus, Hradec Kralove 2004 [10] J. Šesták, Z. Strnad “Compositional Dependence of Nucleation” in Proc. 8th Int. Conf. React. of Solids, Gothenburg, Elsevier, Amsterdam 1976, p. 410 and in Proc. Inter. Congress on Glass’77, CVTS, Prague 1977, Vol. II, p. 410 [11] N. Koga and J. Šesták: “Crystal nucleation and growth in glasses by thermal analysis” J. Am. Ceram. Soc. 83 (2000) 1753-1760. [12] El-Ghannam A., Ducheyne P., Shapiro I. M.: “Formation of surface reaction products on bio-

active glass and their effects on the expression of the osteoblastic phenotype and the deposition of

mineralized extracellular matrix “. Biomaterials 18 (1997) 295 [13] L.L. Hench, in P. Ducheyne and J. Lemons (eds.), Bioceramics, Vol. 523, Annals of New York Academy of Sciences, New York, 1988, p. 54. [14] Z. Strnad, N. Koga key lecture “Compositional Dependence of Non-bridging Oxygen Ions in

Bioactive Glasses” in the Proc. Forum New Materials (Florence) 1998, p. 12

Advanced Materials Research Vols. 39-40 333

[15] N. Koga, Z. Strnad, J. Strnad, J. Šesták, Thermodynamics of non-bridging oxygen in silica bio-

compatible glass-ceramics for bone tissue substitution J. Thermal Anal Calor. 71 (2003) 927 [16] K. Závěta, J. Šesták “Structure and Magnetic Properties of Fe-rich Oxide Glasses” Amer. Cer. Soc. 55 (1972) 537 and in Proc. Int. Con. on Glass’77, ČVTS Publ. House, Vol. 1, p. 39, Prague 1977 [17] J.E. Davis: “Mechanisms of endosseous integration”. Int. J. Prosthodont. 11 (1998) 391 [18] Y.T. Su, C. Johansson, A. Wenneberg, L. Cho, B. Chang, T. Albrektsson “Optimum surface

properties of oxidized implants for reinforcement of Osseointegration: Surface Chemistry, Oxide

thickness, Porosity, Roughness and crystal structure” Int.J. Oral. Maxillofac. Implants 20 (2005) 349. [19] Z. Strnad, J. Strnad, C. Povýšil, K. Urban, Effect of Plasma Sprayed Hydroxyapatite Coating

on Osteoconductivity of cp Titanium Implants, Int. J. Oral Maxillofac. Implants 15 (2000) 483 [20] C. Ohtsuki, T. Kokubo, T. Yamamuro, Mechanism of apatite formation on CaO-SiO2-P205

glasses in a simulated body fluid, J. Non-Cryst. Sol. 143 (1992) 84 [21] S.L. Wheeler. Clinical retrospective study of titanium plasma-sprayed and hydroxyapatite-

coated cylinder implants. Int. J. Oral Maxillofac Implants 11 (1996) 340 [22] T. Albrektsson, A. Wennerberg, Oral implant surfaces: review focusing on topographic and

clinical properties of different surfaces and in vivo responses, Int. J. Prosthodont. 17 (2004) 536 [23] J. Strnad, Z. Strnad, J. Šesták , K. Urbanand, C. Povýšil: “Bio-activated Titanium Surface Uti-

lizable for Mimetic Bone Implantation in Dentistry: surface characteristics and bone-implant con-

tact formation”, J. Phys. Chem. Solids 68 (2007) 841-845 [24] J. Strnad, J. Protivínský, D. Mazur, Z. Strnad, A. Helebrant and J. Šesták, Interaction of acid

and alkali treated titanium with dynamic simulated body environment J. Thermal Anal Calor. 76 (2004) 17 [25] J. Strnad, Z. Strnad, J. Šesták, “Physico-chemical properties and healing capacity of poten-

tially bioactive titanium surface”, J. Thermal Anal. Calor. 8 (2007) 775-779 [26] A.R. Bausch, K. Kroy, Nature Physics, Vol. 2, April (2006), pp. 231 – 238 [27] G.A. Holzapfel, R.W. Ogden (Eds.), Mechanics of Biological Tissues, Springer 2006 [28] M. Holeček, F. Moravec « Hyperelstacic model of material microstructure formed by balls and

springs » :Int. J. Solids Structures 43 (2006) 7393


Journal of Thermal Analysis and Calorimetry, Vol. 71 (2003)

THERMODYNAMICS OF NON-BRIDGING OXYGEN INSILICA BIO-COMPATIBLE GLASS-CERAMICSMimetic material for the bone tissue substitution

N. Koga1, Z. Strnad2, J. Šesták3* and J. Strnad 2

1Chemistry Laboratory, Department of Science Education, Graduate School of Education,Hiroshima University, 1-1-1 Kagamiyama, Higashi-Hiroshima 739-8524, Japan2Laboratory for Glass and Ceramics (LASAK), Papírenská 25, CZ-16000 Praha 6, Czech Republic3Institute of Physics of the Academy of Sciences, Cukrovarnicka 10, CZ-16253 Praha and Instituteof Interdisciplinary Studies, te West Bohemian University, Husova 17, CZ-30114 Pilsen,Czech Republic

Abstract

Correlations between the structural properties of Na2O–CaO–SiO2 glasses characterized by the ac-tivity of oxygen ions and the bioactivity were examined by comparing the compositional de-pendence of the structural parameters calculated on the basis of a thermodynamic consideration withthat of the bioactivity. A simple model of characterizing the glass structure by considering the bridg-ing and non-bridging oxygen ions was employed as the first step for this purpose. Further detailedthermodynamic analysis on the anionic constitution in the glass was performed and thecompositional dependences of the relative proportions of bridging, non-bridging and free oxygenions were calculated. The bioactive region corresponded to the compositional region characterizedby the higher relative proportion of non-bridging oxygen ions with co-existing an appreciable con-centration of bridging oxygen ions, suggesting a possible important role of the non-bridging oxygenions on the surface chemical process of bone-like apatite layer formation.

Kevwords: bio-compatible, bone-like apatite, glass-ceramics, mimetic material, thermodynamics

Introduction

Degeneration of a human skeletal system in time results in dysfunction of bones,teeth and joints. Extensive bone defects, left after the removal of tumors, infections oras a result of injuries, are ideally replaced by autogenous bone tissue. As the amountof this material for the patient is limited and the use of allogenic bone is accompaniedby biological, mechanical and also sociological difficulties, there is a great need foralternate non-human synthetic sources.

Merely four decades ago it was considered inconceivable that a man-made materialcould bond to living tissues in view of the deep-rooted experience that it would result in a

1418–2874/2003/ $ 20.00

© 2003 Akadémiai Kiadó, Budapest



Yariv04

* Author for correspondence: E-mail: [email protected]

foreign body reaction and the formation of non-adherent scar tissue at the interface withinserted material. This understanding was irreversibly altered when a special composi-tion of soda-lime-phosphate-silica glass was synthesized by Hench [1] and successfullyimplanted in the femurs of rats. About 6% of P2O5 was added to simulate the Ca/P con-stituents of hydroxyapatite, Ca10(PO4)6(OH)2, that is the inorganic mineral phase naturallyexisting in bones. Therefore the bone-like apatite formation on the surface of implant isof a key importance during the physical and chemical processes leading to the formationof an enough firm connection between the implanted material and the newly formed bonetissue [2–6]. The bioactivity leads to both the osteoconduction and osteoproduction as aconsequence of rapid reaction on the bioactive glass surface. The surface reac-tions [7–16] involve ionic dissolution of calcium and sodium ions, phosphates andhydrated silica that give rise to both the intercellular and extracellular responses at the in-terfaces of the glass with its physiological environment. The smartness of such a mimeticprocess is likely hidden in the activity of oxygen characterized by the action of silanolegroups (Si–OH). They likely serve as the nucleation sites for the bio-compatible interfaceformation capable to coexist between the original tissue and the implants which can beexpediently made from glass, glass-ceramics, ceramics, cements and other composites aswell as from certainly treated metals (etched titanium) respecting the set-in condition ofits suitable surface reactivity.

We have taken part in the research progress of bioactive materials since early eight-ies [17–22]. This matured in their actual appliance in practical implantology under thetrademark IMPLADENT- (a system for oral implantology produced by LASAK, Co.



Fig. 1 Surface layer formed on a BAS-O implant after exposure in a simulated bodyfluid for 28 days (a) and the cross section of the actual body implant (c):a – View of the surface layer formed on the BAS-O material after exposure inSBF (SEM 1000×, fracture), b – Content of elements (P, Ca and SiO2) in thesurface layer, and c – Detail of the interface of the bone tissue with a BAS-Oimplant 6 months after implantation (SEM 1200×)

Ltd) and BAS-O, BAS-HA, BAS-R-bioactive bone tissue substitutes [23]. Figure 1 illus-trates the bone-bonding ability of BAS-O which is based on inorganic, polycrystallinematerial prepared by controlled crystallization of glass, whose main components areCaO, P2O5, SiO2 and MgO. During the crystallization process, the glassy material is con-verted to a glass-ceramic material whose main crystalline phases are apatite andwollastonite. BAS-O granules and ground material are used to fill cysts, defects left byinjuries, defects left by excochleation of benign tumors, and can be of help to reconstructextensive acetabular defects. Compact, wedge-shaped blocks (with various heights andsurfaces) became useful for, e.g., condyl elevation. Individually shaped implants can beused in neurosurgery to cover defects left from cranial trepanation and as onlays in plasticsurgery. In Fig. 2, there is revealed another case of biomaterial BAS-HA (hydroxy-apatite) which is synthesized from aqueous solutions under precisely defined pH, temper-ature and other physical parameters, which ensure reproducible preparation of a highlypure, crystallographically defined product, which does not contain any unwanted calciumphosphates. Its structure and composition are similar to bio-apatite, which is the main in-organic component of living bone tissue. Implants form a strong bond between the bonetissue and the implant material without any intermediate fibrous layer. Final product isthe BAS-R, which is the surface bioactive, resorbable, inorganic, crystalline materialbased on tricalcium phosphate.

Bioactivity has since attracted increased attention being aimed to further molecularmanipulation (doping surfactants, micro-additives of various organic molecules such asproteins, glyco-proteins and polysaccharides, useful in easier mine-realization) which in-telligent response by host organism is evaluated in order to achieve well-tailored im-plants. Bio-glass-ceramics that activate genes offer the possibilities of repairing, or per-haps even preventing, many disease states, such as osteoporosis, in which a large fractionof women lose a substantial amount of bone mass as they age. They can be also used as asecond phase in a composite that mimics the structure and properties of bone. In future



Fig. 2 a – Direct contact between the BAS-HA implant and bone tissue, 2 months afterimplantation. BAS-HA material is a very dense ceramics with apparent porosity of1.7%. The Ca/P molar ratio is 1.66. The material exhibits a bending strength of60 MPa and compression strength of 200 MPa. The strength of the junction withthe bone tissue measured by the push-out test (shear stress) is equal to 19 MPa(two months after implantation) and 29 MPa (4 months after implantation).b – Dental implant (‘Impladent’) with hydroxyapatite coating