ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2018

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology 1710

Spark plasma sintered ZrO2-SiO2 glass ceramics and Si3N4

bioceramics

LE FU

ISSN 1651-6214ISBN 978-91-513-0419-9urn:nbn:se:uu:diva-357128

Dissertation presented at Uppsala University to be publicly examined in Häggsalen, 10132,Ångström, Lägerhyddsvägen 1, Uppsala, Wednesday, 10 October 2018 at 09:00 for the degreeof Doctor of Philosophy. The examination will be conducted in English. Faculty examiner:Professor Jérôme Chevalier ( Université de Lyon-INSA de Lyon).

AbstractFu, L. 2018. Spark plasma sintered ZrO2-SiO2 glass ceramics and Si3N4 bioceramics. DigitalComprehensive Summaries of Uppsala Dissertations from the Faculty of Science andTechnology 1710. 70 pp. Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-513-0419-9.

This thesis focuses on elaboration and characterization of two types of bioceramics: oneis ZrO2-SiO2 nanocrystalline glass ceramic (NCGC) for dental application. The goal is todevelop new ZrO2-SiO2 NCGCs with a combination of high strength and high translucency;the other is biodegradable Si3N4 ceramics for spinal fusion. This project aims to improvethe osteointergration property of Si3N4 ceramics. Translucent glass ceramics typically sufferfrom impaired mechanical properties, compared to full-ceramics. We presented a method ofobtaining ZrO2-SiO2 NCGCs, with a microstructure of monocrystalline ZrO2 nanoparticles(NPs), embedded in an amorphous SiO2 matrix. Raw powders containing different ZrO2 contentswere prepared by the sol-gel method, followed by the spark plasma sintering (SPS). The NCGCwith a composition of 35%ZrO2-65%SiO2 (molar ratio, 35Zr) was transparent. TetragonalZrO2 NPs were spherical with a diameter of 20–40 nm. The average flexural strength of35Zr NCGC was 234 MPa. To improve the flexural strength, NCGCs with compositionsof 45%ZrO2-55%SiO2 (45Zr), 55%ZrO2-45%SiO2 (55Zr), 65%ZrO2-35%SiO2 (65Zr) werealso elaborated. All NCGCs showed high translucency. The flexural strength of the NCGCssignificantly increased with the increase of ZrO2 content, achieving as high as 1014 MPa for65Zr NCGC. ZrO2 NPs in 65Zr NCGC were ellipsoidal and had a core-shell structure witha thin Zr/Si interfacial layer as the shell. Some of the ZrO2 NPs were connected and formedZrO2 nanofibers. Moreover, the ZrO2 nanofibers were orderly stacked in short-range to formthe 3D nano-architecture. The high flexural strength of the 65Zr NCGC mainly originatesfrom synergistic strengthening effects of the thin Zr/Si interfacial layer and 3D stacked nano-architecture. Regarding biodegradable Si3N4 bioceramics, we used a ternary sintering additiveof SrO, MgO and SiO2. The mechanical properties of the developed Si3N4 bioceramicswere comparable to those of traditional Si3N4 ceramics. Sr2+, Mg2+, and Si4+ ions releasedfrom the intergranular glass phase after immersion in solution, indicating that the developedSi3N4 bioceramics showed certain biodegradable ability. These ions enhanced the proliferationand differentiation of preosteoblasts. Meanwhile, the ionic dissolution products did not showany toxic effects to the development or physiology of zebrafish embryos.

Keywords: ZrO2-SiO2 glass ceramic, high-strength, translucency, 3D nano-architecture,Si3N4 bioceramic, biodegradable, spark plasma sintering

Le Fu, Department of Engineering Sciences, Applied Materials Sciences, Box 534, UppsalaUniversity, SE-75121 Uppsala, Sweden.

© Le Fu 2018

ISSN 1651-6214ISBN 978-91-513-0419-9urn:nbn:se:uu:diva-357128 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-357128)

To my dear parents and grandma

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I. L. Fu, C. Wu, K. Grandfield, E. Unosson, J. Chang, H. Engqvist, W.

Xia, Transparent single crystalline ZrO2-SiO2 glass nanoceramic sintered by SPS, J. Eur. Ceram. Soc. 36 (2016) 3487–3494.

II. L. Fu, H. Engqvist, W. Xia, Highly translucent and strong ZrO2-SiO2 nanocrystalline glass ceramic prepared by sol-gel method and spark plasma sintering with fine 3D microstructure for dental restoration, J. Eur. Ceram. Soc. 37 (2017) 4067–4081.

III. L. Fu*, L.Xie*, W. Fu, S. Hu, K. Leifer, H. Engqvist, W. Xia, Ultra-strong translucent glass ceramic with nanocrystalline, biomimetic structure. Submitted.

IV. L. Fu, H. Engqvist, W. Xia, Spark plasma sintering of biodegradable Si3N4 bioceramic with Sr, Mg and Si as sintering additives for spinal fusion, J. Eur. Ceram. Soc. 38 (2018) 2110–2119.

V. L. Fu, Y. Xiong, G. Carlsson, M. Palmer, S. Örn, W. Zhu, X. Weng, H. Engqvist, W. Xia, Biodegradable Si3N4 bioceramic sintered with Sr, Mg and Si for spinal fusion:Surface characterization and biological evaluation, Appl. Mater. Today. 12 (2018) 260-275.

Reprints were made with permission from the respective publishers. *Authors contributed to the work equally.

Author’s Contributions

Paper I Part of experimental work and major part of the writing. Paper II Part of planning, major part of experimental work, and major

part of the writing.

Paper III Major part of planning, part of experimental work, and major part of writing.

Paper IV Major part of planning, all experimental work, and major part

of writing.

Paper V Major part of planning, part of experimental work, and major part of writing.

Also published

L. Fu, W. Xia, T. Mellgren, M. Moge, H. Engqvist, Preparation of High Percentage α-Calcium Sulfate Hemihydrate via a Hydrothermal Method. J. Biomater. Nanbiotechnol. 08 (2017) 36-49. W. Xia, L. Fu, H. Engqvist, Critical cracking thickness of calcium phosphates biomimetic coating: Verification via a Singh-Tirumkudulu model. Ceram. Int. 43 (2017) 15729–15734.

Contents

1. Introduction ............................................................................................... 13 1.1 Ceramics in dentistry .......................................................................... 13 1.2 Biomaterials in spinal fusion .............................................................. 15 1.3 ZrO2 ceramics ..................................................................................... 16

1.3.1 Structure and transformation ...................................................... 16 1.3.2 Stabilization ................................................................................ 17 1.3.3 Mechanical properties and biocompatibility ............................... 18 1.3.4 Low-temperature degradation ..................................................... 18 1.3.5 Translucency ............................................................................... 19

1.4 Glass ceramics (GCs) ......................................................................... 20 1.4.1 Definition .................................................................................... 20 1.4.2 Manufacture ................................................................................ 21 1.4.3 Properties .................................................................................... 21 1.4.4 GCs in dentistry .......................................................................... 22

1.5 Si3N4 ceramics .................................................................................... 22 1.5.1 Property, application and microstructure .................................... 22 1.5.2 Liquid phase sintering ................................................................. 23 1.5.3 Biocompatibility ......................................................................... 24

2. Aims of the thesis...................................................................................... 25

3. Material preparation and characterization methods .................................. 27 3.1 Sol–gel method ................................................................................... 27 3.2 Spark plasma sintering (SPS) ............................................................. 28 3.3 Mechanical tests ................................................................................. 30

3.3.1 Flexural strength ......................................................................... 30 3.3.2 Indentation microhardness .......................................................... 31 3.3.3 Elastic modulus (E) and nanohardness ....................................... 31 3.3.4 Fracture toughness ...................................................................... 31

3.4 Transmission Electron Microscopy (TEM) ........................................ 32 3.4.1 Imaging ....................................................................................... 32 3.4.2 Electron spectroscopy ................................................................. 33 3.4.3 Electron tomography (ET) .......................................................... 34

4. Results and Discussion ............................................................................. 35 4.1 ZrO2-SiO2 GCs ................................................................................... 35

4.1.1 Characterization of sol–gel powders ........................................... 35 4.1.2 Optical image .............................................................................. 37 4.1.3 XRD measurement ...................................................................... 37 4.1.4 Mechanical properties ................................................................. 38 4.1.5 SEM observation ........................................................................ 39 4.1.6 TEM observation ........................................................................ 40 4.1.7 Microstructure-mechanical strength relation .............................. 46 4.1.8 Summary for ZrO2-SiO2 GCs ..................................................... 47

4.2 Biodegradable Si3N4 bioceramics ....................................................... 48 4.2.1 Material preparation .................................................................... 48 4.2.2 α-Si3N4/β-Si3N4 ratio ................................................................... 49 4.2.3 Mechanical properties ................................................................. 49 4.2.4 Ion release ................................................................................... 50 4.2.5 In vitro bacterial infection assessment ........................................ 52 4.2.6 In vitro cytocompatibility evaluation .......................................... 53 4.2.7 Zebrafish study: toxicity and skeletal calcification .................... 54 4.2.8 Summary for Si3N4 bioceramics ................................................. 55

Conclusion and future perspectives .............................................................. 57

Sammanfattning på svenska .......................................................................... 59

Acknowledgement ........................................................................................ 61

References ..................................................................................................... 64

Abbreviations

GC Glass ceramic NCGC Nanocrystalline glass ceramic NP Nanoparticle t-ZrO2 Tetragonal zirconia m-ZrO2 Monoclinic zirconia c-ZrO2 Cubic zirconia YSZ Yttria-stabilized zirconia LTD Low-temperature degradation TEM Transmission electron microscopy HR-TEM High-resolution TEM BF-TEM Bright-field TEM HAADF High-angle annular dark-field STEM Scanning transmission electron microscopy EDS Energy dispersive X-ray spectroscopy EELS Electron energy-loss spectroscopy CLSM Confocal laser scanning microscopy SEM Scanning electron microscopy XRD X-ray diffraction TEOS Tetraethyl orthosilicate SPS Spark plasma sintering α-Si3N4 Alpha-silicon nitride β-Si3N4 Beta-silicon nitride

13

1. Introduction

A wide range of materials are being used in the medical field, and each of these interacts with the human body. These materials are generally described as biomaterials. As one of the most important biomaterials, bioceramics, represents around 50% of the world consumption of biomaterials1. The applications of bioceramics include maxillofacial reconstruction, dental restoration, spinal fusion, and drug delivery. This thesis focuses on developing high-strength and translucent ZrO2-SiO2 glass ceramics for dental application and Si3N4 ceramics for spinal fusion.

1.1 Ceramics in dentistry Humans have long been aware of the medical and esthetic benefits of tooth replacements. Over 3500 years ago, the ancient Egyptians attempted to close gaps in dentition by carving false teeth out of mulberry wood and tying them to adjacent teeth with gold wire. Etruscans produced more esthetic tooth replacements using bovine teeth. Ceramic materials were first used for dental restorations in the 18th century2. Compared to other materials, aesthetics (adequate translucency) and durability (adequate strength and chemical stability) are the two major advantages of ceramics as dental materials. In 1962, the first two US patents for porcelain-fused-to-metal (PFM) restorations were awarded, which consisted of gold alloy and feldspathic porcelain3,4. Since then, PFM restorations have set the standard for multiple teeth restoration. In recent decades, dental bridges were mostly metal-porcelain composite, consisting of a metallic framework for load bearing and coated porcelain for aesthetic appearance2,5,6. Despite the wide application of PFM restorations the suitability of using metals in the oral cavity has been disputed in recent years due to their biological incompatibility and some other concerns, such as the aesthetic limitations5. These are the main factors motivating continued research into all-ceramic restorations. All-ceramic dental restorations have been available since the 1980s6. Table 1 lists some examples of all-ceramic dental restorations. The development of processing techniques, such as heat-pressing, slip-casting, and computer-aided design and computer-aided manufacturing (CAD-CAM), facilitate the design and fabricate of dental ceramics to fulfill higher clinic requirements7. The overall performance of all-ceramic restoration can be considered as quite successful,

14

for instance, densely sintered alumina crowns (Procera, Nobel Biocare, Sweden) had a 5-year survival rate of 96.4%, quite similar to that of leucite-reinforced glass ceramic (95.4%) (Empress, Ivoclar Vivadent, Liechtenstein)8.

Table 1. Commercially available metal-ceramic and all-ceramic dental restorations. Manufacturer Brand Crystalline phase Flexural strength

(MPa) Ivoclar-Vivadent IPS Classic9 leucite 80±10 Ivoclar-Vivadent IPS d.SIGN9 apatite and leucite 80±25 Ivoclar-Vivadent IPS e.max Press10 lithium disilicate 303-466a Ivoclar-Vivadent IPS e.max

ZirCAD11 Zirconium oxide 600-1000a

Nobel Biocare Procera6 Alumina oxide 607±73 Vident In-Ceram Alumina12 Alumina oxide 594±52 Vident In-Ceram Spinell6 Spinel 378±65 a Measured by different methods.

Strength is one of the most important criteria for dental ceramic. ISO 6872:2015 stipulates a minimum value of 500 (class 4) or 800 (class 5) MPa for flexural strength of dental ceramic. The reported average chewing forces during normal mastication vary widely, from 40 to 440 N. Higher forces can readily be reached for brief periods (~500 to ~880 N)5. ZrO2-based ceramics are among the most promising candidates to meet the strength needs of dental ceramic13. In its dense sintered state, ZrO2-based ceramics exhibit higher strength and toughness than all other dental ceramics. Moreover, the biocompatibility of ZrO2-based ceramics is already established from their in-body use for femoral head balls in hip replacements14. Glass ceramics show adequate translucency, but relatively low bending strength and toughness. Although ZrO2-based ceramics show high strength and toughness, some concerns remain, such as reduced translucency and susceptible to low-temperature degradation6. Glass ceramics that combine high translucency and high strength and toughness would be promising candidates for dental ceramic.

15

1.2 Biomaterials in spinal fusion Spinal fusion is an orthopedic surgical technique that joins two or more vertebrae. This procedure can be performed at any level in the spine (cervical, thoracic, or lumbar) and prevents movement between the fused vertebrae15. Bone graft, either from the patient (autograft), the donor (allograft) or artificial bone substitutes, is needed to help the vertebrae heal together. Additional hardware (screws, plates, or cages) is often used to hold the bones in place while the graft fuses the two vertebrae together16. Spinal fusion is most commonly performed to relieve the pain and pressure on the spinal cord. Common pathological conditions that are treated by spinal fusion include degenerative disc disorder, spinal stenosis, and spinal fractures15,16. Over the last 30 years, reports have indicated that the number and cost of spinal fusion procedures have increased significantly17,18. The slope of the increase became steeper in 1996, coincident with FDA approval of intervertebral fusion cages in the U.S. Between 1998 and 2008, the annual number of spinal fusion discharges increased 2.4-fold (137%) from 174,223 to 413,17118; the national bill for spinal fusion increased 7.9-fold from approximately $4.3 billion in 1998 to $33.9 billion in 200818. Approximately 488,000 spinal fusions were performed during U.S. hospital stays in 201119, a 70 % growth in procedures from 200120.

Table 2. Merits and drawbacks of PEEK, Ti, and Si3N4 as biomaterials used for spinal fusion.

Materials Merits Drawbacks

PEEK Good radiolucency, good mechanical and wear properties, chemical inertness, and low elastic modulus25

Migration of the cages, relatively poor protein absorption and cell adhesion, lack of osteointegration25,26

Ti Good mechanical properties, high corrosion resistance, a long track record of biocompatibility22,23

Insufficient osteointegration, metal ions-related allergies and implant-associated bacterial infections27

Si3N4 Good mechanical properties, biocompatibility, imaging characteristics24, antibacterial and osteoinductive properties28,29

Stress shielding, complicated shaping process

Three biomaterials have been used for spinal implants: polyetheretherketone (PEEK), titanium (Ti) and its alloys, and Si3N4 ceramics. Table 2 list the merits and drawbacks for PEEK, Ti, and Si3N4 as biomaterials used for spinal fusion. PEEK was introduced as spinal cages in the 1990s by AcroMed (Cleveland, OH, now DePuy Spine, Raynham, MA)21. Titanium

16

(Ti) and its alloys have been widely used in orthopedic and dental implants22,23 before being introduced to spinal fusion. Si3N4 was approved by the US Food and Drug Administration to be used as a spinal vertebral body replacement device since 2006. Before being used as spinal cages, Si3N4 has more than 30 years of clinical history in joint replacements because of their excellent combination of mechanical properties, biocompatibility, and imaging characteristics under X-ray, CT, and MR imaging24. Amedica Corporation (Salt Lake City, US) is the only FDA-registered and ISO-certified silicon nitride medical device manufacturing facility in the world.

1.3 ZrO2 ceramics 1.3.1 Structure and transformation Zirconia (ZrO2), an oxide of the metal, has been used in medicine and dentistry since the 1970s, due to its favorable properties, including low cytotoxicity, corrosion potential, and low propensity to bacterial adhesion13. Zirconia is a polymorph that occurs in three forms (Fig.1-1a): monoclinic (m-ZrO2), tetragonal (t-ZrO2), and cubic (c-ZrO2)14. At room temperature, pure zirconia is present in the most stable phase, m-ZrO2. As the temperature rises to about 1170 °C, m-ZrO2 transforms into t-ZrO2, accompanied by a shrinkage in volume of approximately 3–4%14 (Fig.1-1b). t-ZrO2 converts into c-ZrO2 at about 2370 °C, with only minimal changes in volume30 (Fig.1-1b). The reversible lattice transformations in ZrO2 share three similarities with martensitic transformations as in austenite-martensite transformation in steels, which are (1) diffusion-less (i.e., without transport of atoms); (2) occur within a temperature range, not at a specific temperature (i.e., athermal); and (3) involve shifts in the coordination of lattice positions31. In 1975, Garvie et al. found that the t-to-m transformation significantly improved the strength and toughness of ZrO2 ceramics32. These authors came up with the term "ceramic steel" for partially stabilized zirconia because it shares features with strengthened steel: three allotropes, the metastable phases, and the martensitic transformation32. Fig.1-1c demonstrates the transformation toughening mechanism in ZrO2 ceramics. During crack propagation, the stress applied to the surrounding crack induces the t-ZrO2 to m-ZrO2 transformation. The volume expansion (3–4%) then exerts stress on the crack, which seals the crack and inhibits crack propagation33. The transformability of t-ZrO2 is influenced by temperature, vapor, particle size, micro- and macro-structure of the material, and the concentration of stabilizing oxides34.

17

1.3.2 Stabilization It is well-known that the monoclinic phase can be disfavored. Tetragonal and cubic phases can be stabilized as metastable phases at room temperature by incorporation of components such as yttrium oxide (Y2O3), calcium oxide (CaO) or magnesium oxide (MgO) into the ZrO2 lattice14,31,35. The stabilization mechanism is that the lower valence dopant ions (for instance, Y3+) substitute Zr4+ in the ZrO2 lattice, leading to oxygen vacancies. It is the oxygen vacancies that hinder the t-ZrO2 to m-ZrO2 transformation11. Tetragonal ZrO2 can also be stabilized by cerium, Ce4+, instead of Y3+ or Ca2+ cations, without inducing oxygen vacancies36. Partially or fully stabilized zirconia can be formed by adding different amounts of dopants. Fully stabilized zirconia (FSZ) is achieved by adding either 8 mol% Y2O3 or 16 mol% MgO or CaO14. Smaller amounts (2–5 mol%) of the same dopants lead to partially stabilized zirconia (PSZ).

Fig.1-1 a) Unit cell of monoclinic (ICCD, 04-004-4339), tetragonal (ICCD, 04-005-4212), and cubic ZrO2 (ICCD, 04-011-9021); b) Crystal structure of ZrO2 at different temperatures and transformation, redrawn from ref11; c) Transformation toughening mechanism in ZrO2 ceramics.

18

1.3.3 Mechanical properties and biocompatibility ZrO2-based ceramics have excellent intrinsic physical and chemical properties, including high strength and toughness, good wear resistance, chemical inertness, ionic conductivity, and low thermal conductivity33. These properties mean that ZrO2-based ceramics can be used as both structural and functional ceramics. To date, ZrO2 ceramics have been demonstrated to be the strongest and toughest (single-phase) oxide ceramics yet produced33. Table 3 lists mechanical properties of cortical bone and four biomaterials developed for load-bearing applications. ZrO2 ceramics show the highest flexural strength (900–1200 MPa). Biocompatibility refers to "the ability of a material to perform with an appropriate biological response in a specific situation"37. Generally, ceramics are inert materials, which have no adverse local or general tissue reactions. In vitro and in vivo studies have confirmed the high biocompatibility of ZrO2-based ceramics14,38.

Table 3. Mechanical properties of cortical bone and four biomaterials developed for load bearing application.

a Tensile strength; b Flexural strength.

1.3.4 Low-temperature degradation Biomedical grade ZrO2 ceramics were introduced to solve the problem of Al2O3 brittleness. Before 2000, ZrO2 ceramics were widely used as the ball head in total hip arthroplasty43

. However, failure events of Prozyrs femoral heads in 2001–2002 have generated controversy on the future of zirconia as a biomaterial; roughly 400 femoral heads failed within a very short period45. The problem is that ZrO2 ceramics are susceptible to so called low-

Property Ti6Al4V39,40 Stainless steel41,42

Al2O3 14,43

ZrO2 14,30,31

Cortical bone41,44

Density (g/cm3) ~4.4 ~8 ~4 ~6 --

Strength (MPa) ~600-~1100a

~465-~950a ~500-~700b

~900-~1200b

~50-~150

Compressive strength (MPa)

-- ~1000 ~4000-~5000

~2000-~3000

~100-~230

Young’s mudulus (GPa)

~110 ~200 ~380-~410

~200-~300

~7-~30

Hardness (HV) ~100 ~190 ~2200 ~800-~1200

--

Fracture toughness(MPa m-1)

58-66 55-95 ~4-~6 ~5-~10 2-12

19

temperature degradation (LTD), in which the metastable t-ZrO2 transforms into the stable m-ZrO2

46. The transformation starts at the surface in the presence of water at relatively low temperatures. The LTD process mainly involves two stages: (a) nucleation on a particular grain at the surface, leading to microcracking and stresses to the neighbors; and (b) growth of the transformed zone, leading to extensive microcracking and surface roughening45. Surface degradation caused by LTD, like pull-outs and micro-cracks, leads to strength degradation. For hip joint heads, surface degradation leads to increased wear and the release of wear debris in the body45.

1.3.5 Translucency Transparent/translucent ceramics/glass ceramics form a special group of materials that have been used in various applications, including laser hosts, infrared (IR) windows/domes, and transparent armors47. In most cases, ceramics or glass ceramics are opaque or show low translucency, because there are various sites to scatter light (Fig.1-2). The most significant factor for transparency of ceramics is porosity. The transmission gradually increases with the decrease of porosity48. Owing to sufficient glass phase, glass ceramics can achieve zero porosity, thus avoiding strong light scattering of pores49. Additional scattering of light arises at grain boundaries when the light travels from one grain to another50. On the basis of Mie theory, the relationship between the intensity of scattered light with physical characteristics of matrix and inclusions is given by:

(1)

where Is is the intensity of scattered light, I0 the intensity of the primary light, λ the wavelength of incoming light, r the radius of inclusion, N the number of inclusions per cm3, and m the relative refractive index ni/nm

51.

As is obvious from Eq. (1), the size of inclusions, their concentrations in the matrix, and their refractive indices are the main factors influencing scattering. Light scattering is strongly dependent on the size of the grain. When the grain size (< 100 nm) is much smaller than the wavelength of visible light (350–700 nm), light can “bypass” grains without much scattering. Thus, reducing grain size is an effective way to increase translucency48,52. External factors, such as surface finish and the sample thickness, also affect translucency of a ceramic sample. Significant diffuse

= 1283 − 1+ 1

20

scattering occurs on a rough surface, and the intensity of transmitted light decreases with the increase of sample thickness53. In summary, the key strategy to develop transparent ceramics is to eliminate all possible scattering sites of light, including least porosity (> 99.9 % of theoretical density), the absence of impurity, small grain size (< 100 nm), isotropic lattice structure, and high surface finish48,51,54,55. Compared to t- and m-ZrO2, it is relatively easy to manufacture transparent c-ZrO2 ceramics since the cubic structure has isotropic refraction index in different crystallographic direction, while tetragonal and monoclinic structures are birefringent52. However, the strength of c-ZrO2 ceramics is much lower than t-ZrO2 ceramics. A few studies have been focused on preparing transparent/translucent yttria-stabilized zirconia (YSZ), aiming for nanocrystalline YSZ with both desirable translucency and mechanical properties48,52,55.

Fig. 1-2 Light transmission through polycrystalline ceramic, with various light scattering sources: 1.Grain boundary; 2. Second phase or impurity; and 3. Residual pore.

1.4 Glass ceramics (GCs) 1.4.1 Definition Strictly speaking, the term ‘glass’ describes a state of matter where the atoms/molecules are randomly arranged, in other words, glass materials are amorphous. While ceramic materials are mainly composed of crystalline grains, with a small amount of glass phase at the grain boundaries. GCs are a special group of materials that consist of at least one crystalline phase and glass phase matrix54. The crystallinity varies most frequently between 30 % and 70 %. Fig.1-3 schematically describes the structural differences between glass, GC, and ceramic.

21

1.4.2 Manufacture Normally, GCs are produced by a so-called "controlled crystallization": First, glass is formed by a glass-manufacturing process; second, the glass is cooled down and is then reheated in a second step. In this heat treatment, the glass partly crystallizes. In most cases, nucleation agents (e.g., noble metals, fluorides, ZrO2, TiO2 or Fe2O3) are added to the base glass composition to aid and control the nucleation-crystallization process56. GCs can also be produced by concurrent sintering-crystallization of glass-particle compacts. A main advantage of the sintering-crystallization process is that nucleating agents are not necessary, since nucleation-crystallization starts at the glass-particle interfaces56.

1.4.3 Properties The properties of GCs principally depend on the characteristics of the dispersed crystalline phases and the residual glass phase, which can be tailored by the composition of the base glass, the content and type of crystalline phases, and heat treatment schedules. GCs provide a wide range of combinations of physical and mechanical properties as they are able to embrace a combination of the unique properties of sintered ceramics and glasses1. For instance, lithium disilicates GCs developed by Ivoclar-Vivadent for dental application combine high strength and toughness (350–400 MPa; KIc≈2.3–2.9 MPa∙m1/2) with translucency and biocompatibility10. Another example is β-quartz solid-solution GCs, which is an ideal material for radiant stovetop due to the combination of the near-zero coefficient of thermal expansion and high transparency in the visible wavelength range54. GCs show average fracture strength and toughness of 100–250 MPa and 1–2.5 MPa∙m1/2, respectively, which are generally lower than ceramics but higher than commercial glasses (50–70 MPa; KIc≈ 0.7 MPa∙m1/2)56. GCs possess zero porosity and high translucency, which are distinctive characteristics of glasses.

22

Fig.1-3 Schematic microstructures of glass (left), GCs (middle), and ceramic (right). Glass is amorphous, with randomly arranged atoms; GC is composed of at least one crystalline phase and an amorphous matrix, with crystallinity varies between 30 % and 70 %; Ceramics mainly consist of crystalline grains, with a small amount of glass phase at the grain boundaries.

1.4.4 GCs in dentistry GCs are attractive to dentists and patients because they provide good mechanical and aesthetic properties. Currently, in the dental market, there are many types of GCs with different compositions and manufacture processes. For example, lithium disilicates (Li2Si2O5) GCs, which allows a wide range of dental indications to be covered, ranging from single dental units to small three-unit bridges in the anterior region (IPS Empress® 2, Ivoclar Vivadent AG)57. The GCs blocks are machined with CAD/CAM equipment in an intermediate lithium meta-silicate (Li2SiO3) phase. Once the restoration has been milled, the Li2SiO3 is transformed into Li2Si2O5 by heat treatment. Another example is leucite (KAlSi2O6) GCs prepared by molding technology without any additional heat treatment after the molding step57. Controlled surface nucleation and surface crystallization were used to develop leucite GCs in the SiO2–Al2O3–K2O–Na2O system.

1.5 Si3N4 ceramics 1.5.1 Property, application and microstructure As one of the most important structural ceramics, Silicon nitride (Si3N4) ceramics have been studied intensively for more than 40 years, because of their excellent combination of physical and chemical properties, including high strength (~1100 MPa) and toughness (8-11 MP m1/2), high decomposition temperature (1900 °C), high oxidation resistance, low coefficient of friction, and good thermal shock properties58–60. These key properties make Si3N4 ceramics an ideal material for high-performance components in severe environments, such as bearings, cutting tools, wear parts, valves, and heat engine components59. Si3N4 ceramics occur in two

23

forms, α- and β-Si3N4, both having hexagonal crystal structures built by cor-corner-sharing SiN4 tetrahedra58. α-Sialon and β-Sialon are solid solutions of α- and β-Si3N4, respectively61. α-and β-Si3N4 have quite different grain morphology, with α-phase grains typically exist as equiaxed gains, while β-phase grains typically exist as elongated hexagonal prisms. The high strength and toughness of Si3N4 ceramics are attributed to the “self-reinforced” microstructure consisting of elongated β-Si3N4 grains embedded in an equiaxed-grained matrix 24,58. Analogous to whisker-reinforced ceramics, the reinforcing microstructure promotes toughening mechanisms, such as crack bridging and pullout and crack deflection62.

1.5.2 Liquid phase sintering The high atomic diffusion rate is one of the most important prerequisites of the sintering process. However, the activation energy required for solid-state diffusion of covalent bonded Si3N4 solids is extremely high, resulting in very low atomic self-diffusion coefficients63,64. This prevents densification of Si3N4 ceramics with normal solid-state sintering. Sintering additives (mainly oxides) need to be added to assist the densification process through “liquid phase sintering”59. Amongst all oxide sintering additives, the best-known system is the yttria-alumina (Y2O3-Al2O3)59. The sintering additives react with oxidized surface layers on silicon nitride powders to form a eutectic liquid at high temperature62,65. The high-temperature liquid phase provides the vehicle for rapid mass transport and the driving force for densification is the capillary pressure of the liquid phase between the fine solid particles. Furthermore, the liquid phase is needed for the α-β phase transformation, which is believed to occur through the dissolution of the fine α-phase starting powder and subsequent precipitation of the β-phase58. The liquid phase is usually retained as an inter-granular glass phase after sintering63. Fig. 1-4 schematically illustrates the liquid phase sintering process of Si3N4 ceramics, which can be approximately divided into three stages59,66: Stage I: particle rearrangement and large pore elimination take place once liquid phase forms, meanwhile α-phase or small β-phase grains start to dissolves into liquid phase. Stage II: elongated β-phase grains precipitate from the liquid phase, and rapid densification occurs; Stage III: α-β transformation almost completes, and grain growth occurs.

24

Fig. 1-4 A schematic drawing illustrating α-β phase transformation and microstructural evolution during liquid phase sintering of Si3N4 ceramics.

1.5.3 Biocompatibility Si3N4 ceramics exhibit a unique combination of bulk mechanical properties and imaging characteristics (under X-ray, CT, and MR imaging) that make Si3N4 ceramics an ideal biomaterial for orthopedic implants67. Si3N4 ceramics have been used as a fusion cage for spinal fusion since 2008, and approximately 25,000 Si3N4 spinal fusion cages have been implanted with few adversely reported events68. It has been reported that Si3N4 ceramics have substantially improved antibacterial and osteoinductive properties compared with other biomaterials used in spinal fusion (i.e., Ti and PEEK)28,29. The improved biocompatibility of Si3N4 ceramics has been attributed to the surface characteristics, for instances, the surface of Si3N4 ceramics is hydrophilic, which could promote protein adsorption28. A porous form of Si3N4 showed good bone ingrowth capability69. Due to the sufficient intergranular glass phase, Si3N4-based ceramics can be regarded as ceramic-glass composites60. And the bioactivities of Si3N4 ceramics can be improved by using sintering additives that contain bioactive elements44.

Starting powder Stage I and Stage II Stage III

25

2. Aims of the thesis

The work presented in this thesis was aimed at developing two types of bioceramics: high-strength and translucent ZrO2-SiO2 glass ceramics (GCs) for dental application and biodegradable Si3N4 ceramics for spinal fusion. Highly translucent GCs in dentistry, such as lithium disilicate, are generally weaker than full ceramics, which limits the application of GCs on long bridgeworks where high mechanical resistance is required. While full ceramics, for instance, ZrO2 ceramics, possess high mechanical strength, but less translucency. Thus, new glass ceramics which possess an excellent combination of high strength and high translucency would be a promising alternative ceramic in dentistry. Si3N4 bioceramics with standard sintering additives (Al2O3 and Y2O3) are being used for spinal fusion. Alternative Si3N4 bioceramics with biologically beneficial sintering additives could lead to improved osseointegrative properties. The present work focused on the following specific aims:

Paper I: To prepare transparent ZrO2-SiO2 GCs with sol-gel method and spark plasma sintering; to characterize the mechanical properties and microstructure of the GCs. The composition is 35%ZrO2-65%SiO2 (molar ratio).

Paper II: To improve the mechanical strength of ZrO2-SiO2 GCs by increasing ZrO2 content; to characterize the mechanical properties and microstructure of the GCs.

Paper III: To characterize the microstructure of the 65%ZrO2-35%SiO2 (molar ratio) GC via TEM; to correlate the microstructure with the high flexural strength of the 65%ZrO2-35%SiO2 GC.

Paper IV: To prepare Si3N4 bioceramics with biologically beneficial sintering additives (i.e., SrO, MgO, and SiO2) via spark plasma sintering; to characterize the mechanical properties and microstructure of the Si3N4 bioceramics.

Paper V: To characterize the surface properties of Si3N4 bioceramics; to evaluate the biological properties of the Si3N4 bioceramics.

26

27

3. Material preparation and characterization methods

3.1 Sol–gel method The sol–gel process is a wet-chemical technique used in the fabrication of both glass and ceramic materials70,71. Generally, the sol–gel process can be divided into three stages70–73: First, hydrolysis: precursors are mixed at the molecular level, followed by hydrolysis of precursors. Typical precursors are metal alkoxides and metal chlorides. Second, poly-condensation: sol evolves gradually towards the formation of a gel-like network (gel) containing both a liquid phase and a solid phase. Third, drying and firing: a drying process is required to remove the remaining liquid (solvent) phase contained in the gel. Thermal treatment in the range 500–800 °C desorbs chemisorbed hydroxyls, resulting in a stabilized gel. The sol–gel process offers many advantages compared to the conventional "powder" route: first, higher purity and homogeneity. This is because precursors are mixed at the molecular level70; second, fine control of the product’s chemical composition. Even small quantities of dopants, such as organic dyes, can be introduced in the sol and end up uniformly dispersed in the final product71,74; third, relatively low cost. Compared with other material preparation methods, such as the hydrothermal method; the sol-gel process is more economical since it is carried out under ambient atmospheres without special equipment70.

In the ZrO2-SiO2 GC project, tetraethyl orthosilicate (TEOS) and zirconium propoxide solution were used as the precursors. Sol-gel powders of different ZrO2 contents in the xZrO2·(100-x)SiO2 system were fabricated (x = 35, 45, 55, 65mol%). Details of the procedure can be found in Paper I and the supplemental information in Paper II. Briefly, the hydrolysis of TEOS and zirconium propoxide were initialized separately by adding HCl solution. The two hydrolyzed solutions (sol) were mixed and kept under stirring. The sol was then put into an oven (60 °C) for 3 days for gelation and aging, which results in a homogenous gel. A drying process was conducted at 120 °C for another 3 days. The xerogel was milled by planetary ball milling. Pre-calcination of the powder was conducted in a muffle furnace at 600 °C for 1 h with a ramping rate of 1 °C/min.

28

3.2 Spark plasma sintering (SPS) Spark plasma sintering (SPS), also known as field assisted sintering technique (FAST), is a low voltage, direct current (DC) activated, pressure-assisted sintering technique76,77. The basic configuration of a typical SPS system is shown in Fig.3-1a. The system consists of vertical single-axis pressurization, a water-cooled vacuum chamber, a DC pulse generator, a temperature measurement system, and an SPS controller. Fig.3-1b and c are two optical images of the SPS machine, showing the main components of the SPS unit. The main characteristic of SPS is that pulsed or unpulsed DC current directly passes through the graphite die, as well as the powder compact (in case of conductive powders76,78). Joule heating plays a dominant role in the densification of powder compacts. The heat generation is internal, in contrast to the conventional sintering techniques, where the heat is provided by external heating elements. The internal Joule heating allows the application of very high heating (up to 1000 °C/min) and cooling rates (up to 400 °C/min). Hence the sintering process is typically very fast (within a few minutes)77, see Fig.3-1d. The high heating rate makes SPS an excellent technique to manufacture nanocrystalline materials, avoiding grain coarsening, which accompanies standard densification routes79. It has been reported that the heating is accomplished by spark discharges in voids between the particles76. The particle surface is activated and purified, which assists heat-transfer and mass-transfer during sintering76. However, the intermediate process and mechanism for creating plasma are complicated, and it is unknown whether plasma is produced or not77,78. Anyway, the advantages of conventional sintering methods make SPS a widely used technique for the preparation of ceramics with enhanced thermoelectric80, magnetic81, optical82, and biomedical83 properties.

29

a)

Control and display

Vacuum chamber

Pyrometer

b)

Upper graphite support

Lower graphite support

Graphite die

c) d)

Fig.3-1 SPS technique. (a) A schematic drawing showing the basic configuration of SPS; (b) An optical image of SPS machine; (c) An optical image of the vacuum chamber and loaded sintering die; (d) Comparison of the heating cycle between SPS and conventional sintering techniques.

In this thesis, the SPS process was carried out on an SPS-825 machine (Fuji Electronic Industrial Co., Ltd, Japan). For the sintering of ZrO2-SiO2 GCs, a two-step heating process was applied: 120 °C/min from 366 °C to 900 °C, and 30 °C/min from 900 °C to holding temperature (between 1100 °C to 1250 °C). The temperature was monitored by an optical pyrometer. The dwelling time was 5 min, and the applied pressure was 60 MPa. Detail information can be found in Paper I and Paper II. For Si3N4 ceramics, the SPS process was conducted with a heating rate of 100 °C/min from 366 °C to holding temperature (~1700 °C). The holding temperature was 3–5 min. A pressure of 50 or 60 MPa was applied. The sintering process was carried out in an N2 atmosphere. More information on the SPS process can be found in Paper IV.

30

3.3 Mechanical tests 3.3.1 Flexural strength Strength is defined as the ultimate stress that is necessary to cause fracture or plastic deformation84. Flexural strength is commonly used to characterize the responses of ceramics to loading forces. Flexural strength test methods of ceramics are either uniaxial (e.g., 3- or 4-point bending)85 or biaxial flexural tests (e.g., piston-on-three-ball)86,87. Uniaxial bend tests (e.g., 3- or 4-point bending) are sensitive to flaws nearly perpendicular to the beam axis of the specimen, resulting in large variations88. The biaxial flexural test eliminates the effect of edges because they are not directly loaded, thus producing less variation in data88,89. The piston-on-three-ball test has been selected as the standard method (ISO 6782) to characterize the flexural strength of dental ceramics90. In the piston-on-three-ball test, a disk-shaped specimen is supported from below by three balls distributed in a circular pattern. The load is applied from above by use of a piston concentric with the supporting balls, see Fig.3-2 a and b. During the test, on the inside of the bend (concave face), the stress will be at its maximum compressive stress value, while on the outside of the bend (convex face) the stress will be at its maximum tensile value (Fig.3-2c). For brittle materials (such as ceramics), fracture normally occurs at the convex surface under tensile stress caused by flexure (Fig.3-2c). Detail information on the piston-on-three-ball test for ZrO2-SiO2 GC can be found in Paper I and Paper II. Those for Si3N4 ceramics can be found in Paper IV.

Fig. 3-2 Piston-on-three-ball test. (a) An optical image of the set-up; (b) A schematic representation of the set-up, front view and top view; (c) A schematic illustration of stress distribution during the test (front view).

Loading piston

Specimen

a) b) c)

31

3.3.2 Indentation microhardness The indentation hardness of a material defines the relative resistance that its surface imposes against the penetration of a harder body85. Indentation hardness is usually measured with conventional microhardness instruments, such as Knoop or Vickers diamond indenters. These instruments make impressions for which the diagonal size is measured with an attached optical microscope.

3.3.3 Elastic modulus (E) and nanohardness Elastic modulus, also known as Young’s modulus, represents the stiffness of a material. It is an indication of the amount of reversible deformation that will occur in a structure when a load is applied to it89. Elastic modulus can be measured by static methods that are based on measuring the deformation of a specimen loaded by a known static91. In 1992 an analytical method (instrumented indentation) was proposed by Oliver and Pharr to measure elastic modulus92,93. Instrumented indentation (nano-indentation) testing is a development of conventional hardness tests such as Rockwell, Knoop, and Vickers. This testing is similar to standard hardness testing in that a hard indenter is made to contact with the test material, with which the contact area can be determined from the force-displacement data, thus removing the necessity to image the residual impression93,94. Moreover, the displacement retained as the indenter is withdrawn and shows elastic recovery, which can be utilized to acquire elastic modulus (E)92. Detail information of elastic modulus and nanohardness for ZrO2-SiO2 GCs and Si3N4 ceramics can be found in Paper I and Paper II, and Paper IV, respectively.

3.3.4 Fracture toughness Fracture toughness is defined as the critical stress intensity level at which a given flaw starts growing, and indicates the ability of a material to resist rapid crack propagation and its consequent catastrophic failure85. Fracture toughness is typically reported as MPam1/2 and is a quantitative way of expressing a material’s resistance to brittle fracture when a crack is present. A material with high fracture toughness might undergo ductile fracture as opposed to brittle fracture.

Several methods, such as “single-edge-v-notched beam” (SEVNB) and the Vickers indentation method (IF), are available to evaluate fracture toughness. The IF method offers an approximate determination of relative fracture toughness, which is assessed by introducing flaws of controlled size, shape, and location, followed by direct measurement of the radial cracks. It is a simple technique; however, it results in high dispersion because of

32

difficulties inherent with the accurate measurement of crack length and sub-subcritical slow crack growth95. Thus, it has been pointed out that IF method is not an adequate method to exactly determine fracture toughness96. The SEVNB method, in which bar specimens are needed, is believed to be a more accurate methodology than the IF method96 and the SEVNB method was selected by ASTM as the method of determination of fracture toughness of advanced ceramics97. Due to the limitation of the graphite die used in SPS, only disc samples can be produced. Thus, the IF method was used to estimate the fracture toughness of ZrO2-SiO2 GCs (Paper I and Paper II). Although the indentation method’s reliability to evaluate fracture toughness is debatable, it still offers a first rough estimation.

3.4 Transmission Electron Microscopy (TEM) In this thesis, the TEM has been applied as a major analysis tool for characterizing the microstructure of ZrO2-SiO2 GCs (Paper I, II, and III). Modern TEM is capable of analyzing samples by a number of different imaging and spectroscopic signals, such as: atomically resolved images by high-resolution TEM (HR-TEM), structural information from electron diffraction, chemical composition analysis using electron energy-loss spectroscopy (EELS) or energy dispersive X-ray spectroscopy (EDS)98. In general, an electron beam with the desired size and dose is focused onto the specimen. The specimen is most often an ultrathin lamella (less than 100 nm thick) or a suspension on a grid. A wide range of signals can be generated from the electron-mater interaction, see Fig.3-3a. Elastically scattered, direct transmitted, inelastically scattered electrons and characteristic X-rays are the most commonly used signals. Mass-thickness contrast, diffraction contrast, and phase contrast are the major contrast mechanisms for TEM imaging99. In the following session, different imaging modes or spectroscopic signals that have been used in this thesis will be briefly introduced.

3.4.1 Imaging In the bright field (BF) mode of the TEM, an aperture is placed in the back focal plane of the objective lens, which allows only the direct beam to pass. Both mass-thickness contrast and diffraction contrast contribute to image formation. Thick areas and areas with enriched heavy atoms appear with a darker contrast. BF-TEM was used to give an overview of the microstructure of ZrO2-SiO2 GCs (Paper I and II). In HR-TEM mode, a parallel beam is used. The contrast of an HR-TEM image arises from the interference in the image plane of the electron wave with itself (phase contrast). Fast Fourier transformation (FFT) is a useful tool to assist the interpretation of HR-TEM images. FFT gives crystallographic information, similar to electron

33

diffraction98. The interface between ZrO2 and SiO2 has been investigated by HR-TEM in Paper II. In STEM mode, an electron beam is focused to a fine probe that is scanned across the sample in a two-dimensional (2D) raster100. The electrons will scatter due to strong Coulomb interaction with the atoms in samples. The scattering angle will be larger when interacting with a heav-ier atom. A High-Angle, Annular Dark Field detector (HAADF) is posi-tioned as a ring around the transmitted beam, collecting only electrons that have been scattered to higher angles (Fig.3-3b). Therefore, the HAADF sig-nal emitted by each atom scales proportionally with the square of the atomic number (Z)101. STEM-HAADF has been used to characterize ZrO2 NPs in Paper I, II and III. Fig.3-3 Electron-matter interaction, different imaging modes, and spectroscopy. (a) Scheme of electron-matter interaction. Elastically scattered, direct transmitted, ine-lastically scattered electrons, and characteristic X-rays are the most commonly used signals used in TEM; (b) Schematic illustration of STEM-HAADF, EELS, and STEM-EDS.

3.4.2 Electron spectroscopy In EELS, the energy that the transmitted electrons have lost due to inelastic scattering in the sample is recorded and measured by a spectrometer (Fig.3-3b). These energy losses are characteristic of the binding energy of the elec-trons, which give information of not just what elements are present, but also about their chemical surrounding (i.e., bonding state)98. The EELS technique was used in Paper I to identify the composition of NPs and matrix. EDS is easy to use and is particularly sensitive to heavier elements. The Titan The-mis TEM used in Paper III is equipped with a Super-X high sensitivity windowless EDS detector, which allows detection of atomic elemental in-formation. The EELS is considered as being complementary to EDS. But, EELS is a more difficult technique, and it is in principle capable of measur-

b)

a)

34

ing atomic composition, chemical bonding, valence, and conduction band electronic properties99.

3.4.3 Electron tomography (ET) All of the above techniques, in general, provide only 2D information in the form of images, thus neglecting the 3D information of samples101,102. This limits the full characterization of nanoscale materials with complex mor-phology and chemistry in 3D. To overcome this limitation, electron tomog-raphy (ET), developed originally for the biosciences, has been transferred to the field of material science to provide a 3D reconstruction of nanoscale objects from a series of images103. A host of possible imaging modes can be used for ET, among which the STEM-HAADF mode has been established as the most commonly used image mode in material science104. Briefly, a typi-cal ET process involves three steps (Fig. 3-4)103,105,106: first, 2D images are acquired at successive tilts. In many cases, the tilt intervals are regular (typi-cally 1° or 2°); second, 2D images are combined into an image stack ordered by viewing angle; third, the tilt series are aligned, and a reconstruction algo-rithm is applied to produce a 3D reconstruction of the object. ET technique has been used to characterize the 3D arrangement of ZrO2 NPs in Paper III.

Fig.3-4 Schematic diagram of the process of ET: first, 2D images are acquired at successive tilts about a single axis; second, 2D images are combined into an image stack ordered by viewing angle; third, the tilt series are aligned, and reconstruction algorithm is applied to produce 3D reconstruction of object. Redraw from ref103.

35

4. Results and Discussion

4.1 ZrO2-SiO2 GCs 4.1.1 Characterization of sol–gel powders Sol–gel powders of four compositions in the xZrO2·(100-x)SiO2 system were fabricated (x = 35, 45, 55, 65 mol%). In this study, the specimens were designated as 35Zr, 45Zr, 55Zr, and 65Zr, respectively. All sol-gel powders before calcination were amorphous. Only the XRD pattern of the 65Zr pow-der before calcination was presented in Fig.4-1. 35Zr and 45Zr powders remained amorphous after calcination, while 55Zr and 65Zr powders showed a weak peak at around 30°, indicating that a small amount of nanocrystalline ZrO2 formed from the amorphous raw powder during calcination (Paper I and II). It has been reported that the addition of SiO2 delayed the crystalliza-tion of ZrO2 because of the encasement of the SiO2 matrix107,108. The delayed effect of SiO2 increased with the increase of SiO2 content109. This explains why the 35Zr and 45Zr powders remained amorphous after calcination, while the 55Zr and 65Zr powders started to crystallize. The calcined 65Zr powder was characterized via HR-TEM (Paper II). Dif-fuse diffraction rings were found in the FFT image in Fig.4-2a, indicating that in the observed region the power was primarily amorphous. While in another region (Fig.4-2b), clear lattice fringes can be observed, and diffrac-tion spots appeared in the inserted FFT image. The spacings of the lattice fringes (2.97 Å) were comparable to the (111) plane spacing of tetragonal ZrO2 (2.964 Å). Together with the above XRD results (Fig.4-1), it can be concluded that the observed NPs (Fig.4-2b) were tetragonal ZrO2 NPs. In the ZrO2-SiO2 system, tetragonal ZrO2 was the first phase to crystallize from the amorphous state during heat treatment109,110. The sol-gel powders had similar morphology after calcination, consisting of relatively large irregularly-shaped particles (4–10 µm) and small particles (less than 1 µm) (Fig.4-2c).

36

Fig.4-1 XRD patterns of sol-gel powders after calcination at 600 °C for 1 h and the 65Zr powder before calcination.

Fig.4-2 TEM and SEM characterization of the 65Zr sol-gel powder: (a-b) HR-TEM images with fast Fourier transformation (FFT) images inserted; (c) SEM image showing that the powder consists of irregular particles.

10 20 30 40 50 60 70 80

55Zr

65Zr

2θ (°)

65Zr before calcination

35Zr

45Zr

a)

2.97 Å

b) c)

5 µm

37

4.1.2 Optical image As shown in Fig.4-3, 1 mm thick 35Zr sample was transparent. Text under the samples can be clearly seen (Paper I). 45Zr, 55Zr, and 65Zr samples showed high translucency (Paper II). The text under the 45Zr sample can be clearly recognized, while the text under the 65Zr sample was blurry. The stripes under the 65Zr sample can still be distinguished. According to Equa-tion (1), the intensity of scattered light is proportional to the radius of inclu-sion (r) and the number of inclusions per cm3 (N). The average size of ZrO2 NPs in the 65Zr (47.5 nm) and 55Zr (35.1 nm) samples were larger than those in the 45Zr (29.5 nm) and 35Zr (28.1 nm) samples. In addition, the volume fraction of ZrO2 NPs in 65Zr (58.7 %) and 55Zr (47.7 %) samples were larger than those in 45Zr (39.7 %) and 35Zr (28.1 %) samples, which resulted in more ZrO2 inclusion per cm3 in 65Zr and 55Zr samples. There-fore, more incoming light was scattered by the 65Zr and 55Zr samples com-pared to 45Zr and 35Zr, leading to higher translucency in the 35Zr and 45Zr samples and relatively lower translucency in 55Zr and 65Zr samples.

Fig.4-3 Optical images of ZrO2-SiO2 GCs after grinding and polishing (diameter: 20 mm, thickness: 1 mm). Samples were sintered under 1150 °C (holding temperature)-60 MPa (applied pressure)-5 min (holding time). (a) The 35Zr sample is transparent (Paper I); (b-d) 45Zr, 55Zr, and 65Zr samples are translucent (Paper II). The trans-lucency decreases with the increase of ZrO2 content.

4.1.3 XRD measurement All the GC samples showed almost the same XRD patterns (Fig.4-4) (Paper I and Paper II). The patterns exhibited strong peaks at 30°, 50°, and 60°, which corresponded to the (101), (112), and (211) planes of tetragonal ZrO2 (t-ZrO2), respectively. No peaks of monoclinic ZrO2 (m-ZrO2) or crystalline SiO2 were found. It has been reported that in the ZrO2-SiO2 system an amor-phous SiO2 matrix can greatly retard the tetragonal-monoclinic transfor-mation of ZrO2, thereby stabilizing t-ZrO2

108. The stabilizing effect of SiO2 on t-ZrO2 was attributed to the encasement and the constraint effect of the SiO2 matrix, and the formation of Zr-O-Si interphasic reaction layer around the ZrO2 particles107,108.

a) 35Zr b) 45Zr c) 55Zr d) 65Zr

38

Fig.4-4 XRD patterns of ZrO2-SiO2 GCs (Paper I and Paper II). Tetragonal ZrO2 was the only crystalline phase in all the glass ceramic samples.

4.1.4 Mechanical properties The mechanical properties of the studied ZrO2-SiO2 GCs are summarized in Table 4 (Paper I and Paper II). The micro-hardness of the glass ceramics varied from 7.16 GPa to 7.57 GPa. However, no significant differences were found between hardness values of 35Zr, 45Zr, and 55Zr samples, while the micro-hardness value of 65Zr sample was significantly higher than those of the 35Zr, 45Zr, and 55Zr samples. Nano-hardness of 65Zr sample was also significantly higher than those of 35Zr, 45Zr and 55Zr samples. The Young’s modulus of the studied ZrO2-SiO2 GCs increased with the increase of ZrO2 content, increasing from 78.9 GPa (35Zr) to 152.5 GPa (65Zr). Alt-hough the indentation method’s reliability to evaluate fracture toughness is debatable96, it still offers a first rough KIc estimation111. The average fracture toughness value of the ZrO2-SiO2 GCs also increased with the increase of ZrO2 content. The average flexural strengths of the 35Zr, 45Zr, 55Zr, and 65Zr samples were 234, 533, 737, and 1014 MPa, respectively, increasing with the increase of ZrO2 content. The flexural strength value of the 45Zr sample was close to that of commercial lithium disilicates dental GCs10, while the 45Zr sample showed higher translucency than lithium disilicates dental ceramics. The achieved flexural strength of the 65Zr sample was as high as 1014 MPa, which was closed to yttria-stabilized ZrO2 full ceramics (800–1200 MPa)14,30. To our knowledge, this is one of the highest reported flexural strength values for GCs.

10 20 30 40 50 60 70 80

45Zr

55Zr

2θ (°)

65Zr

(101)

(211)(103)(200)

(112)(110)(002) 35Zr

Tetragonal ZrO2

39

Table 4. Mechanical properties of the studied ZrO2-SiO2 GCs materials (average values ± standard deviation). Material Micro-

hardness (GPa)

Nano-hardness

(GPa)

Young’s modulusa

(GPa)

Fracture tough-nessc (MPa m-1/2)

Flexural Strengthb

(MPa) 35Zrd 7.42±0.20 7.0±0.40 78.9±1.3 3.58±0.08 234±44 45Zr 7.21±0.10 9.02±0.73 117.0±4.4 4.15±0.53 533±127 55Zr 7.16±0.03 9.58±0.59 142.0±4.7 5.48±0.23 737±166 65Zr 7.57±0.11 10.55±0.52 152.5±8.8 6.55±0.29 1014±185

a Measured by nano-indentation; b Measured by piston-on-three-ball test; c Vickers indentation method (IF); d 35Zr sample sintered under 1150 °C-60 MPa-5 min.

4.1.5 SEM observation Fig.4-5a demonstrates that there was a large number of spherical ZrO2 NPs (indicated by dash white arrows) on the fracture surface of the 35Zr sample. Corresponding voids (indicated by solid white arrows) left by pulled-out ZrO2 NPs were also observed (Paper I). A hackle region, a region with radi-ating ridges and valleys, was found on the fracture surface of 65Zr sample (Fig.4-5b), indicating the 65Zr sample exhibited brittle conchoidal fracture behavior during the piston-on-three-ball test (Paper II). Both ZrO2 NPs (in-dicated by dash white arrows, Fig.4-5c) and voids can be found on the frac-ture surface of the 65Zr sample, and the ZrO2 NPs were more densely dis-tributed compared to those in the 35Zr sample. The 45Zr and 55Zr samples showed similar fracture surface morphology characteristics with 35Zr and 65Zr samples (Supporting information in Paper II). As can be seen from Fig.4-5d, ZrO2 NPs in the 65Zr sample presented a homogenous size distri-bution.

40

\\\\

Fig.4-5 SEM images of 35Zr (Paper I) and 65Zr (Paper II) samples: (a) fracture surface of 35Zr after piston-on-three-ball test; (b-c) fracture surface of 65Zr at low (b) and high magnification (c), respectively; (d) plane surface of 65Zr sample afteretching with by hydrofluoric acid.

4.1.6 TEM observation The microstructure of the 35Zr sample was characterized via TEM (Paper I). In Fig.4-6a, the regions with dark contrast corresponded to crystalline ZrO2 NPs, while regions with bright contrast corresponded to the amorphous SiO2 matrix. Dispersive diffraction dots formed diffraction rings in the in-serted selected area electron diffraction patterns (Fig.4-6a, inset), revealing that the 35Zr sample was a polycrystalline material. EELS mapping (Fig.4-6b) showed the distinct elemental distribution of Zr (red) and Si (green). Thus, the microstructure of 35Zr GC was spherical ZrO2 NPs embedded in an amorphous SiO2 matrix. In the STEM-HAADF image (Fig.4-6c), the bright spheres were ZrO2 NPs with diameters between 20 to 40 nm. The ZrO2 NPs were isolated without much contact. A more detailed analysis (Fig.4-6d) revealed that the lattice fringes in the ZrO2 NPs were parallel and arranged in the same direction, suggesting that the ZrO2 NPs were monocrys-talline NPs. The microstructure of 65Zr sample was also characterized with TEM (Fig.4-6 e and f) (Paper II). ZrO2 NPs in the 65Zr sample were ellipti-cal with grain boundary and coalescence. By comparing TEM results of the 35Zr (Fig.4-6 a-d) and 65Zr (Fig.4-6 e and f) samples, one can discover the

d)

200 nm

c)

200 nm

200 nm

a) b)

20 µm

Hackle ridge

41

microstructure evolution of the ZrO2-SiO2 GCs with the increase of ZrO2 content.

Fig.4-6 TEM characterization of the 35Zr (Paper I) and 65Zr (Paper II) samples. (a) BF-TEM image and selected area electron diffraction patterns (inset, scale bar =2 nm-1); (b) EELS mapping shows the distinct elemental distribution of Zr (red) andSi (green); (c) STEM-HAADF images of ZrO2 NPs at low (c) and high-resolution(d); STEM-HAADF images of 65Zr sample at low (e) and high-resolution (f), show-ing that the ZrO2 NPs had an ellipse-like morphology, and that they were contactedwith each other.

a) b)

c)

5 nm

d)

100 nm

e)

10

Grain boundary

Grain coalescence

f)

42

The 65Zr sample was further characterized using the STEM-EDS technique (Paper III). The NPs with bright contrast were ZrO2 NPs, and they were distributed in all three dimensions (Fig.4-7a). The Zr map showed that ZrO2 NPs appeared inter-connected with each other (Fig.4-7b). The Si map (Fig.4-7c) and the overlapped EDS map (Fig.4-7d) showed that the SiO2 matrix formed a network-like structure. Fig.4-7e is a STEM-HAADF image of one ZrO2 NP that is located at the edge of the TEM sample. The region marked by two parallel lines showed slightly different contrast to the core and it was considered as a shell of the selected ZrO2 NP. The thickness of the shell was approximately 2 nm. The existence of the shell was further confirmed by intensity profiles. A straight line was drawn along four ZrO2 NPs (dash lines in Fig.4-7a-c). The width of STEM-HAADF intensity profile was approxi-mately 2–3 nm (marked as parallel black lines) larger than the width of the Zr intensity profile (Fig.4-7f). Also, the STEM-HADDF intensity was higher than the Zr EDS intensity, indicating that both the Zr and Si elements con-tributed to the STEM-HADDF intensity in these 2–3 nm interfacial regions between the ZrO2 NPs and SiO2 matrix. The formation of the thin interfacial layer was a result of Zr/Si inter-diffusion during the sintering process. Based on TEM results and the above analysis, a schematic illustration (Fig.4-7g) was proposed to reveal the core-shell structure of ZrO2 NPs where the shell was composed of a 2–3 nm thin interfacial layer. The origin of the interfacial layer can be explained by the classical nucleation and growth of NPs from the surrounding matrix. The driving force for a particle to precipi-tate and grow from the surrounding matrix consists of two parts: the de-crease of the interfacial free energy and the solute concentration gradient at the interface112. As a function of the particle radius, the concentration gradi-ent at the interface can be modeled by Fick’s first law of diffusion, and an interfacial layer between the precipitated particle and the surrounding matrix always exists when the system is at the non-equilibrium condition113. In the 65Zr sample, the formation of Zr/Si interfacial layer was due to the concen-tration gradient at the interfacial region. In thin film technology, the adhe-sion between film and substrate primarily depends on the interface layer, and the formation of diffusion interface layer results in strong bonding between film and substrate114. Thus, similarly, the Zr-O-Si interfacial layer in the GC provided strong bonding between ZrO2 NPs and the SiO2 matrix.

43

a)

80 nm

b)

80 nm

c)

80 nm

d)

80 nm 5 nm Shell

e)

Fig.4-7 Zr/Si interfacial layer in the 65Zr sample revealed by TEM (Paper III). (a) STEM-HAADF image showing ZrO2 NPs; (b-c) STEM-EDS maps of Zr and Si, respectively; (d) EDS overlapped map; (e) STEM-HAADF image of a ZrO2 NP. The NP shows a core-shell like structure, with an approximately 2 nm-thick shell (marked with parallel lines); (f) Overlapped intensity profiles extracted from the dash lines in the STEM-HAADF image (a), Zr (b) and Si (c) maps; (g) Schematic illustration of ZrO2 NP (core) and Zr-Si interfacial layer (shell) structure.

Two slices (Fig.4-8a) from the reconstructed tomogram obtained by ET are presented to illustrate the distribution of ZrO2 NPs. The colored contrast observed in the slices corresponded to ZrO2 NPs. In slice A, we observed that some of ZrO2 NPs were connected and formed ZrO2 nanofibers with a length of approximately 500 nm (white lines). These ZrO2 nanofibers were in situ formed during the sintering process and they were close to parallelly arranged with a center to center spacing of approximately 75 ± 5 nm. Mean-while, individual ZrO2 NPs can also be observed and they were equally

Zr Si HAADF

Interface layer Interface layer

f) g)

44

spaced (solid white cycles). Also, the center to center distances between these individual ZrO2 NPs were approximately 80 nm. In slice B (Fig.4-8b), the ZrO2 nanofibers presented in slice A disappeared. New ZrO2 nanofibers (dash yellow lines in Fig.4-8b) started to appear between the projections (white lines in Fig.4-8b) of ZrO2 nanofibers presented in slice A. The indi-vidual ZrO2 NPs also had the same stacking order [i.e., the ZrO2 NPs in slice B (dash yellow cycles in Fig.4-8b) located at the middle of the projections (white cycles in Fig.4-8b) of ZrO2 NPs presented in slice A]. Thus, both the ZrO2 nanofibers and individual ZrO2 NPs were not randomly distributed in 3D space, but rather in an ordered form. The ZrO2 nanofibers and individual NPs had an ABAB… stacking sequence in the X-Z plane. Fig.4-8c gives an overview of the 3D spatial distribution of ZrO2 NPs, from which it can be observed that ZrO2 NPs were randomly distributed without obvious ordering. To closely observe the arrangement of ZrO2 NPs, three snapshots from a trimmed volume (Movie S3 in Paper III) are presented in Fig.4-8d. On the bottom part of the snapshot A, near parallel ZrO2 nanofibers stacked in the Y-axis direction can be observed (dash yellow lines). On the top part of snapshot A, ZrO2 nanofibers (dash white lines) can also be found. The nano-fibers were connected in the Y-axis direction (orange arrows). However, the orientation of those two arrays of ZrO2 nanofibers (dash yellow and white lines) was oblique. The misorientation of ZrO2 nanofiber arrays broke the ABAB… stacking order in the long-range. Thus, ZrO2 nanofiber arrays were only orderly stacked in a short-range. The stacking order of ZrO2 nanofibers can be viewed as 3D nano-arcthitecture. One snapshot from Movie S4 (Pa-per III) is displayed in Fig.4-8e, in which the ZrO2 nanofiber stacked in the Z-axis direction can also be observed (dash yellow lines). ZrO2 nanofibers were composed of ZrO2 NPs connected by grain boundaries (white arrows). The ZrO2 NPs showed irregular and complex morphologies (Fig.4-8e)

45

Fig.4-8 Observation of orderly packed ZrO2 nanofibers (Paper III). (a-b) In slice A, both parallel ZrO2 nanofibers (solid white lines) and individual ZrO2 NP (solid white cycles) can be observed. In slice B, new ZrO2 nanofibers (dash yellow lines) or ZrO2 NP (dash yellow cycles) appear in the middle of the projections of ZrO2 nanofibers or individual ZrO2 NP in slice A; (c) An overview of the 3D spatial distribution of ZrO2 NPs; (d) Near parallel ZrO2 nanofibers stacked in the Y-axis direction (dash yellow and white lines in Snapshot A). The nanofibers were connected in the Y-axis direction (orange arrows); (e) ZrO2 NPs in nanofibers were connected by grain boundaries (white arrows).

a) Slice A

X

Y

b) Slice B

X

Y 80 nm 80 nm

c)

60 nm X

Z Y

e)

X Y

Z

30 nm

d) Snapshot C Snapshot B Snapshot A

30 nm Z Y

X

46

4.1.7 Microstructure-mechanical strength relation The average flexural strengths of the 35Zr, 45Zr, 55Zr, and 65Zr samples were 242, 533, 737, and 1014 MPa, respectively (Table 4). The correspond-ing nominal volume fractions of ZrO2 were 28.1 %, 39.7 %, 47.7 %, and 58.7 %, respectively. The flexural strength showed good positive correlation with the ZrO2 volume fraction, increasing with the increase of ZrO2 volume fraction (Fig.4-9a). As schematically demonstrated in Fig.4-9b, in the 35Zr sample, most of the ZrO2 NPs (28.1 vol%) were isolated without much con-tact. While in the 65Zr sample, as shown in the above electron tomography results and the Fig.4-9c, most of ZrO2 NPs (58.7 vol%) were interconnected, forming a complex 3D nano-architecture with in situ ZrO2 nanofibers orderly stacked in the short-range. The 65Zr sample can be viewed as a fiber-reinforced ceramic composite due to the formation of ZrO2 nanofibers. For ceramic composite composing of reinforcement and matrix, the difference between the Young's modulus of the reinforcement and the matrix gives rise to different stresses in the reinforcement and in the matrix when they are deformed to the same strain115. If the reinforcement is much stiffer than the matrix, then the reinforcement carries more load than the matrix115. The Young’s modulus of the tetragonal ZrO2 (~210 GPa31) is much larger than that of amorphous SiO2 (~70 GPa116). Meanwhile, the Zr/Si interfacial layer enables load transfer from the SiO2 matrix to ZrO2 3D nano-architecture. As schematically shown in Fig.4-9d, during the piston-on-three-ball test, both compressive stress on the upper sample surface and tensile stress on the lower sample surface were applied and dissipated on ZrO2 nanofibers or ZrO2 NPs. It is the ZrO2 3D nano-architecture that carries the majority of the external load, rather than the amorphous SiO2 matrix. Thus, the ZrO2 3D nano-architecture acted as hard bricks for primary load bearing and SiO2 matrix as soft mortar for load distribution and energy dissipation. As a result, the 65Zr sample achieved a flexural strength of as high as 1014 MPa (Table 4).

47

d)

Fig.4-9 Flexural strength and ZrO2 volume fraction. Stress situation of the 65Zr sample during piston-on-three balls test (Paper III). (a) The average flexural strength increases with the increase of ZrO2 nominal volume fraction; (b) Schematic diagram of the 35Zr sample in which most of ZrO2 NPs were distributed in 3D space without much contact; (c) Segmented 3D reconstruction of 65Zr sample where ZrO2 NPs form 3D nano-architecture; (d) A front-view schematic diagram of stress situation of 65Zr sample during piston-on-three balls test.

4.1.8 Summary for ZrO2-SiO2 GCs 1. Raw powder with a composition of 35%ZrO2-65%SiO2 (molar ratio, 35Zr) powder was synthesized by a sol-gel method, followed by SPS sinter-ing. The 35Zr samples were transparent. TEM results revealed that 35Zr samples had a homogenous microstructure, composing of tetragonal ZrO2 NPs embedded in an amorphous SiO2 matrix. ZrO2 NPs were spherical, and they were monocrystalline ZrO2 NPs with a size of 20–40 nm. The average flexural strength of 35Zr samples was 234 MPa.

35Zr 45Zr 55Zr 65Zr

200

400

600

800

1000

1200

Composition

Flex

ural

stre

ngth

(MPa

)

25

30

35

40

45

50

55

60

Volu

me

fract

ion

of Z

rO2 (

%)

b) 35Zr

c) 65Zr

a)

48

2. Translucent NCGCs with different ZrO2 contents [i.e., 45%ZrO2-55%SiO2 (45Zr), 55%ZrO2-45%SiO2 (55Zr), 65%ZrO2-35%SiO2 (molar ratio, 65Zr)] can also be prepared by sol-gel method and SPS technique. The flexural strength of the NCGC significantly increased with the increase of ZrO2 con-tent. The 65Zr sample achieved an average flexural strength of as high as 1014 MPa. 3. The increase in the ZrO2 content led to changes in the morphology and arrangement of the ZrO2 NPs. More specifically, ZrO2 NPs in the 35Zr sam-ple were spherical and isolated without much contact with each other, while ZrO2 NPs in the 65Zr sample were ellipsoidal and contacted with grain boundaries and coalescence. 4. In 65Zr NCGC, Si atoms diffuse into the edge of ZrO2 NPs, forming a thin (2–3 nm) Zr/Si interfacial layer. ZrO2 NPs had a core-shell structure. Elec-tron tomography results provided direct evidence of 3D nano-architecture constructed by the orderly stacked ZrO2 nanofibers. The high flexural strength of the 65Zr NCGC originates from synergistic strengthening effects of the thin Zr/Si interfacial layer and 3D nano-architecture.

4.2 Biodegradable Si3N4 bioceramics The results of the Si3N4 bioceramic project are reported in Paper IV and Paper V. In Paper IV, we prepared the Si3N4 bioceramics by SPS and char-acterized the mechanical properties and microstructures. In Paper V, we evaluated the surface characteristics and the biological properties of the de-veloped Si3N4 bioceramics.

4.2.1 Material preparation The raw materials were commercially available α-Si3N4, quartz-SiO2, MgCO3, and SrCO3. MgCO3 was calcined at 400 °C for 1 h to obtain MgO. SrCO3 decomposes to SrO at approximately 1300 °C during sintering. Two sets of raw materials with different amount of additives were prepared (Ta-ble. 5) [i.e., 2%SiO2-4%MgO-4%SrO-90%Si3N4 (weight ratio, 10wt% glass phase) and 2%SiO2-8%MgO-8%SrO-82%Si3N4 (weight ratio, 18wt% glass phase)]. The details of material designation and different sintering conditions are presented in Table 5.

49

Table 5. Material designation, starting powder composition, sintering parameters, α-Si3N4/β-Si3N4 ratio of sintered samples measured by XRD (Paper IV). Material designa-

tion Additive composition

(wt%) Sintering parame-

ters (°C-MPa-min)

α-Si3N4/β-Si3N4 ratio

10G-653 2SiO2-4MgO-4SrO 1650-50-3 54.0±6.3/46.0±6.3

10G-655 2SiO2-4MgO-4SrO 1650-50-5 19.1±4.5/80.9±4.5

18G-653 2SiO2-8MgO-8SrO 1650-50-3 30.4±5.9/69.6±5.9

10G-763 2SiO2-4MgO-4SrO 1720-60-3 17.9±10.6/82.1±10.6

18G-763 2SiO2-8MgO-8SrO 1720-60-3 5.1±3.3/94.9±3.3

4.2.2 α-Si3N4/β-Si3N4 ratio The β-Si3N4 content of Si3N4 bioceramics are listed in Table 5 (Paper IV). The amount of sintering additive, holding temperature, and holding time exert influence on the α-β transformation. Sample 10G-653 and sample 10G-655 were composed of 46.0 % β-Si3N4 and 80.9 % β-Si3N4, respectively, indicating that α-β transformation occurred rapidly at the holding tempera-ture. By comparing the β-Si3N4 content of sample 10G-653 (46.0 %) with sample 18G-653 (69.6 %), it could be concluded that high sintering additive content was beneficial for the transformation. This could be because the transformation took place via a dissolution-diffusion-precipitation process64,117. The higher amount of liquid phase offered more reaction sites, thus accelerating the phase transformation. As high as 94.9 % β-Si3N4 can be achieved in sample 18G-763.

4.2.3 Mechanical properties Mechanical properties of the developed Si3N4 bioceramics are listed in Table 6 (Paper IV). 10G samples (samples contain 10wt% glass phase) showed an average microhardness value of approximately 14 GPa and they were not significantly different between each other. The mircohardness of sample 18G-653 and sample 18G-763 were 12.6 GPa and 11.8 GPa, respectively, which are significantly lower than the 10G samples. The amount of inter-granular glass phase showed a significant effect on microhardness. The high amount of glass phase gave lower hardness values and vice versa. This is because the glass phase was much softer than the α-Si3N4 and β-Si3N4 grains118. The Young’s modulus values of 10G samples (301-316 GPa) were significantly higher than those of 18G samples (~260 GPa). Sample 18G-653 had the lowest flexural strength value (642 MPa) among all the samples.

50

This is because sample 18G-653 contained 18 wt% glass phase and only 69.9 % β-Si3N4. The flexural strength of other developed Si3N4 bioceramics was approximately 1 GPa. According the literature, hot pressed Si3N4 showed a flexural strength ranging from 450 to 1200 MPa59, Vickers hard-ness ranging 13 to 16 GPa59,118,119, and Young’s modulus ranging from 250 to 320 GPa59,118,119. Thus, our Si3N4 bioceramics showed comparable me-chanical properties with most of Si3N4 ceramics sintered with other sintering additives.

Table 6. Mechanical properties of the developed Si3N4 bioceramics (average values ± standard deviation) (Paper IV).

Sample Microhardness (GPa)

Flexural strength (MPa)

Young’s modulus (GPa)

10G-653 14.4±0.4 829±75 316±29

10G-655 14.1±0.9 905±82 331±21

18G-653 12.6±0.2 642±76 264±19

10G-763 14.1±0.6 1079±63 301±15

18G-763 11.8±0.8 928±134 262±18

4.2.4 Ion release Fig.4-10a and b demonstrate the ion release result of sample 10G-763 (Pa-per IV). Sr2+, Mg2+, and Si4+ ions showed similar real-time release behaviors (i.e., they started to leach out after 8 h immersion and the release rate kept increasing within the following 4 days, reaching the maximum value at 4 days’ immersion). After that, the release rate decreased. After 10 days’ im-mersion, the cumulative released concentrations of Sr2+, Mg2+, and Si4+ ions were 68.4, 46.3, and 61.9 mg/L, respectively (Fig.4-10b). The sample 10G-763 had a flat surface after polishing. After immersion for 1 day, intergranu-lar glass phase leached out, leaving shallow interparticulate voids (dash white cycles in 4-10c) between Si3N4 grains. More glass phase leached out after 4 days’ immersion, resulting in deep interparticulate voids (white dash cycles in Fig.4-10d).

51

Fig.4-10 Ion release test and surface morphology of sample 10G-763 (Paper IV). Real-time (a) and cumulative (b) release of Sr2+, Mg2+, and Si4+ ions from the sample 10G-763; (c-d) SEM images of sample surfaces after ion release.

In Paper V, samples with 10% additives (2%SiO2-4%SrO-4%MgO, wt%) and 18% additives (2%SiO2-8%SrO-8%MgO, wt%) were designated as 4Sr-4Mg and 8Sr-8Mg, respectively. Commercially available Si3N4 implants manufactured by Amedica (America Corporation, Salt Lake City, USA) use Al2O3 and Y2O3 as sintering additives. To compare the surface properties and biological properties of the newly developed Si3N4 ceramics with the traditional Si3N4 ceramics, Si3N4 samples with a composition of 4%Al2O3-6%Y2O3-90%Si3N4 (wt%) were sintered using SPS technique under 1700 °C (holding temperature) and 60 MPa (applied pressure) for 3 min (holding time). These Si3N4 samples were designated as 4Al-6Y. Medical grade pure Ti discs were used as a reference sample.

0 2 4 6 8 10

2

4

6

8

10

12

14

16

18Io

n co

ncen

tratio

n (m

g/L)

Soaking time (day)

Mg Sr Si

8h

a

0 2 4 6 8 10

0

10

20

30

40

50

60

70

Ion

conc

entra

tion

(mg/

L)

Soaking time (day)

Mg Sr Si

8h

b

Shallow Inter-particulate void

c) 1d

1 µm

Deep inter-particulate void

400 nm

d) 4d

52

4.2.5 In vitro bacterial infection assessment S. epidermidis is one of the most common etiological agents isolated in or-thopedic implant-associated infections120. The constructed 3D CLSM images showed that S. epidermidis bacteria on 4Sr-4Mg and 8Sr-8Mg surfaces colo-nized and formed isolated colonization (Fig.4-11 a and b) (Paper V). The average biofilm thickness (n = 3) on 4Sr-4Mg and 8Sr-8Mg samples were 13.1 and 15.3 µm, respectively. In contrast, bacteria on 4Al-6Y surface spread widely, colonizing in densely packed biofilms with an average thick-ness of 24.5 µm (Fig.4-11c). Bacteria on the Ti surface tended to form larg-er colonization and had a much thicker biofilm (44.1 µm) than the Si3N4 samples. Si3N4 bioceramics (both the newly developed Si3N4 and the tradi-tional Si3N4 bioceramics) were less susceptible to initial bacterial adhesion than Ti. 4Sr-4Mg and 8Sr-8Mg bioceramics showed less biofilm formation than 4Al-6Y.

Fig.4-11 In vitro bacterial infection assessment (Paper V). Constructed 3D images confocal laser scanning microscopy (CLSM) micrographs of S. epidermidis biofilm formed on Si3N4 samples (a-c) and Ti (d) after overnight incubation.

a) 4Sr-4Mg b) 8Sr-8Mg

c) 4Al-6Y d) Ti

53

4.2.6 In vitro cytocompatibility evaluation Viability analysis indicated (Fig.4-12a) that all the extracted solutions were cytocompatible during the first 24 h of exposure, with no obvious cell toxici-ty (Paper V). After 2 days of exposure to 4Sr-4Mg and 4Al-6Y extracts, cell viability remained higher than controls (p < 0.05), indicating that the extract-ed solutions had a positive effect on cell proliferation. In contrast, extracts from 8Mg-8Sr and Ti were non-cytotoxic and showed little effect on cell proliferation (Fig.4-12a). The cell viability of all samples was approximately 100 % after exposure for 3 days. The ALP activity was determined to evalu-ate the osteogenic potential of MC3T3-E1 cells in the presence of different extracts (Fig.4-12b). After exposure to leach for 7 days, extracts of 4Sr-4Mg (p < 0.01), 8Sr-8Mg (p < 0.05), and 4Al-6Y (p < 0.01) significantly en-hanced the ALP activity of MC3T3-E1 cells compared with extracts from Ti. The 4Al-6Y (p < 0.01) extracts achieved the greatest stimulatory effect on ALP activity, producing 100 % and 50 % more than both 4Sr-4Mg and 8Sr-8Mg leach, on 7 and 10 days, respectively. Throughout the 10 days of expo-sure osteoblasts in contact with extracts of 4Sr-4Mg (p < 0.05), 8Sr-8Mg (p < 0.01), and 4Al-6Y (p < 0.05) continually produced higher ALP activity than osteoblasts exposed to Ti extracts. No significant difference between 4Sr-4Mg extract and 8Sr-8Mg extracts (p > 0.05) were observed. Fig.4-12 In vitro cytocompatibility evaluation (Paper V). (a) Cell viability of MC3T3-E1 cells in the presence of extracts assayed after culture for 1, 2, and 3 days. The cells cultured in medium without any bioceramic extracts were treated as the control group. (mean ± SD, n = 4, *p < 0.05 compared with the control group); (b) ALP activity of MC3T3-E1 cells in the presence of extracts after being cultured for 4, 7 and 10 days. (mean ± SD, n = 4; * p < 0.05, ** p < 0.01).

a) b)

54

4.2.7 Zebrafish study: toxicity and skeletal calcification After leaching samples for 4 days, the cumulative released ions from 8Sr-8Mg samples of Sr2+, Mg2+, and Si4+ were 85, 54, and 74 mg/L, respectively. The extract and tap water were used to culture zebrafish larvae. Fig.4-13 a and b show the development of zebrafish embryos in controls and 8Sr-8Mg test solutions at 24-, 96-, and 144-hpf, respectively (Paper V). No increase in the rate of morphological malformations of the embryos was observed for the extract, indicating that even extract with high ion concentration did not disturb the basic development of the embryo. No obvious bone formed in the 96-hpf larvae. By 144-hpf, head skeletal elements, anterior vertebrae, and posterior vertebrae can be clearly visualized (Fig.4-13 c and d). However, there were slight developmental differences between larvae even in the same batch of larvae that were exposed to the same extract, which makes it diffi-cult to quantify the bone formation ability of larvae that exposed to different extracts difficulty. The staining results (Fig.4-13 c and d) are only for illus-tration of calcified skeletal structures. Interestingly, the vertebrae system can be divided into two distinct domains: an anterior domain (solid white arrows) and a posterior domain (yellow arrows).

55

Fig.4-13 Development of Zebrafish larvae exposed to 8Sr-8Mg extract and tap water (Paper V). (a-b) Photographs of developing zebrafish embryos held in solution prepared by tap water controls (a) and exposed to 8Sr-8Mg (c) test solutions at 24-hpf (left), 96-hpf (middle), and 144-hpf (right); (c-d) Fluorescence images of zebrafish larvae (144-hpf) exposed to tap water (c) and 8Sr-8Mg (d). The larvae were stained with fluorescent chromophore calcein before observation.

4.2.8 Summary for Si3N4 bioceramics 1. Si3N4 bioceramics can be prepared using a ternary sintering additive of SrO, MgO, and SiO2 with SPS technique. The mechanical properties of the newly developed Si3N4 bioceramics were comparable to those of classic Si3N4 ceramics with sintering additives of Al2O3 and Y2O3. 2. SrO, MgO, and SiO2 formed an intergranular glass phase after sintering. Sr2+, Mg2+, and Si4+ ions kept releasing during immersion in the solution, indicating that the developed Si3N4 bioceramics showed certain biodegrada-ble ability. 3. The newly developed Si3N4 bioceramics showed less in vitro bacterial affinity. The ionic dissolution products from these newly developed Si3N4 bioceramics enhanced proliferation and differentiation of MC3T3-E1 preos-

a) Tap water (control group)

b) 8Sr-8Mg

c) tap water d) 8Sr-8Mg

Posterior vertebrae

Anterior vertebrae

Yolk

56

teoblasts. Ionic dissolution products containing a high amount of Sr2+, Mg2+, and Si4+ ions were not toxic to the development or physiology of zebrafish embryos.

57

Conclusion and future perspectives

This thesis describes the preparation and characterization of two types of bioceramics: one is ZrO2-SiO2 nanocrystalline glass ceramic (NCGC) for dental application; the other is biodegradable Si3N4 ceramics for spinal fu-sion. For ZrO2-SiO2 NCGC, high purity raw powders with different ZrO2 contents were prepared by the sol-gel method, followed by the spark plasma sintering. The NCGC with a composition of 35%ZrO2-65%SiO2 (molar ra-tio, 35Zr) was transparent. TEM results revealed that the 35Zr NCGC had a microstructure of tetragonal ZrO2 NPs embedded in an amorphous SiO2 ma-trix. ZrO2 NPs were spherical, and they were monocrystalline ZrO2 NPs with a diameter of 20–40 nm. The average flexural strength is 234 MPa. To im-prove the flexural strength of ZrO2-SiO2 NCGC, NCGCs with compositions of 45%ZrO2-55%SiO2 (45Zr), 55%ZrO2-45%SiO2 (55Zr), and 65%ZrO2-35%SiO2 (65Zr) (all presented in molar ratio) were also prepared by the sol-gel method and SPS technique. These NCGCs showed high translucency. The flexural strength of the NCGCs significantly increased with the increase of ZrO2 content, achieving 533, 737, and 1014 MPa for 45Zr, 55Zr, and 65Zr, respectively. ZrO2 NPs in 65Zr sample were ellipsoidal and had a core-shell structure with a thin (2–3 nm) Zr/Si interfacial layer as the shell. Some of the ZrO2 NPs were connected and formed ZrO2 nanofibers. Also, the ZrO2 nanofibers were orderly stacked in the short-range to form a 3D nano-architecture. The high flexural strength of the 65Zr NCGC mainly orig-inates from synergistic strengthening effects of the thin Zr/Si interfacial lay-er and 3D nano-architecture. Regarding biodegradable Si3N4 bioceramics, we used a ternary sintering additive of SrO, MgO, and SiO2. The mechanical properties of the newly developed Si3N4 bioceramics were comparable to those of traditional Si3N4 ceramics sintered by Al2O3 and Y2O3. Sr2+, Mg2+, and Si4+ ions released from the intergranular glass phase after immersion in the solution, indicating that the developed Si3N4 bioceramics showed certain biodegradable ability. As expected, there ions showed biological benefits as we expected: i.e., they enhanced the proliferation and differentiation of MC3T3-E1 preosteoblasts. Meanwhile, the ionic dissolution products did not show any toxic effects to the development or physiology of zebrafish em-bryos. A number of interesting avenues remain to be investigated to further expand the findings presented herein.

58

Can we prepare the ZrO2-SiO2 NCGCs with conventional sintering tech-niques (e.g., hot pressing or even pressure-less sintering)? In this thesis, we used the SPS technique. The ZrO2-SiO2 NCGCs show great advantages over other commercial dental ceramics in terms of strength and translucency. It would be more favorable if the ZrO2-SiO2 NCGCs could be prepared by conventional sintering techniques.

Can we further increase the ZrO2 content? The flexural strength of the ZrO2-SiO2 NCGCs increases with the increase of ZrO2 content. Among all the prepared NCGCs, the highest ZrO2 molar fraction is 65%. Can we pre-pare NCGCs with 75% or even 80% ZrO2? What would the flexural strength of these two NCGCs be? Would they exceed the strength of 65Zr samples (1014 MPa)?

What about the low-temperature aging of the ZrO2-SiO2 NCGCs? It is well known that ZrO2-based bioceramics suffer from low-temperature aging. The spontaneous isothermal tetragonal to monoclinic transformation exerts detrimental effects on the mechanical properties and stability of the ZrO2-based bioceramics. Do our NCGCs show higher aging resistance compared to other ZrO2-based bioceramics? The tetragonal ZrO2 in the ZrO2-SiO2 NCGCs are NPs with a size smaller than 70 nm. It has been reported that it is much more difficult for these tetragonal ZrO2 NPs to transform into mono-clinic ZrO2. Moreover, there are not so much grain boundaries in our ZrO2-SiO2 NCGC, which could also help to stabilize tetragonal ZrO2. Thus, it is possible that our ZrO2-SiO2 NCGCs show higher aging resistance than other ZrO2-based bioceramics.

With regard to Si3N4 bioceramics, do the bioceramics show beneficial ef-fects on the osteointegration, and new bone formation with the large animal model? In this thesis, we only know that the ionic products from Si3N4 bioc-eramics have enhanced the proliferation and differentiation of MC3T3-E1 preosteoblasts. Thus, it is necessary to further investigate the beneficial ef-fects of the developed Si3N4 bioceramics using a large animal model.

59

Sammanfattning på svenska

Den här avhandlingen beskriver tillverkning och karakteriseringen av två typer av biokeramer: den första är en ZrO2-SiO2 nanokristallin glaskeram (NCGC) för användning inom tandvård; den andra är en biologiskt nedbryt-bar Si3N4-keram för användning inom steloperationer. För ZrO2-SiO2 NCGC framställdes råpulver av hög renhet med olika ZrO2-innehåll med hjälp av en sol-gel-metod, följt av gnistplasmasintring (förkortat SPS från engelskan spark plasma sintering). NCGC med en sammansättning av 35% mol ZrO2 - 65% mol SiO2 (35Zr) var transparent och TEM-resultaten visade att 35Zr NCGC bestod av tetragonala ZrO2 nanopartiklar (NP) inbäddade i en amorf SiO2-matris. NP av ZrO2 var sfäriska och monokristallina med en diameter av ca 20-40 nm. Den genomsnittliga böjhållfastheten hos 53Zr är 234 MPa. För att förbättra böjhållfastheten hos ZrO2-SiO2 NCGC framställdes NCGCs med sammansättningarna 45% mol ZrO2 - 55% mol SiO2 (45Zr), 55% mol ZrO2 - 45% mol SiO2 (55Zr) och 65% mol ZrO2 - 35% mol SiO2 (65Zr) även dessa med sol-gel-metod och SPS-teknik. NCGC-materialen uppvisade hög genomskinlighet. Böjhållfastheten hos NCGC-materialen ökade signifikant med ökande ZrO2-halt, och materialen uppnådde 533, 737 och 1014 MPa för 45Zr, 55Zr respektive 65Zr. ZrO2 NP i 65Zr-provet var ellipsoidala och hade en kärna av ZrO2 med ett tunt (2-3 nm) Zr/Si gränslager som angänsade till SiO2-matrisen. Några av ZrO2-NP var sammankopplade i en riktning och bildade ZrO2 nanofibrer. Och dessa nanofibrerna bildade tillsammans en 3D nano-arkitektur. Den höga böjhållfastheten hos 65Zr NCGC beror huvudsak-ligen på synergistiska förstärkningseffekter av det tunna Zr/Si gränssnittet och 3D nano-arkitekturen. När det gäller biologiskt nedbrytbara Si3N4 bioke-ramer, använde vi ett sintringsadditiv bestående av SrO, MgO och SiO2. De mekaniska egenskaperna hos de nyutvecklade Si3N4-biokeramerna var jäm-förbara med klassiska Si3N4-keramer sintrade med Al2O3 och Y2O3. Sr2+, Mg2+ och Si4+-joner som frigjordes från den intergranulära glasfasen efter att materialen hade sänkts ner i lösning indikerar att Si3N4-biokeramen visade viss biologisk nedbrytningsförmåga. Dessa joner uppvisade de biologiska fördelar vi förväntade oss, dvs de förstärkte proliferationen och differentie-ringen av pre-osteoblaster från MC3T3-E1. Samtidigt visade exponering av urlakningsvätskan innehållande joner inga toxiska effekter på utvecklingen eller fysiologin hos zebrafiskembryon.

60

Det kvarstår ett antal intressanta områden att undersöka för att komplet-tera de resultat som presenteras här.

Kan vi förbereda ZrO2-SiO2 NCGCs med konventionella sintringstekni-ker? Till exempel hetpressning eller till och med tryckfri sintring. I denna avhandling använde vi SPS teknik som har sina fördelar, såsom snabb upp-värmning. Tekniken är lämplig för laboratorieforskning, speciellt för till-verkning av nanokristallina keramer. Den är emellertid inte lika kostnadsef-fektivt som konventionella sintringstekniker, vilket begränsar den kommer-sialiserade produktionen av ZrO2-SiO2 NCGC. Våra ZrO2-SiO2 NCGCs visar stora fördelar jämfört med andra kommersiella dentala keramer med avseende på styrka och genomskinlighet. Det skulle vara fördelaktigare om ZrO2-SiO2 NCGC:erna kunde framställas med konventionella sintringstekni-ker.

Kan vi ytterligare öka ZrO2-innehållet? Böjhållfastheten hos ZrO2-SiO2 NCGCs ökar med ökning av ZrO2-halten. I de NCGC som vi redan har för-berett är den högsta ZrO2-molfraktionen 65%. Kan vi förbereda NCGC med 75 mol% eller till och med 80 mol% ZrO2? Vad är böjhållfastheten hos dessa två NCGC? Är den högre än böjhållfastheten för 65Zr (1014 MPa)? Eller lägre?

Vad sägs om åldring av ZrO2-SiO2 NCGCs vid låga temperaturer? Det är välkänt att ZrO2-baserade biokeramer lider av lågtemperaturåldring. Den spontana isotermiska omvandlingen från en tetragonal struktur till en mono-klinisk har en negativ effekt på de mekaniska egenskaperna och stabiliteten hos ZrO2-baserade biokeramer. Uppvisar våra NCGC högre åldringsbestän-dighet jämfört med andra ZrO2-baserade biokeramer? Den tetragonala ZrO2 i ZrO2-SiO2 NCGC-materialen är NP med en diameter mindre än 80 nm. Det har rapporterats att det är mycket svårare för dessa tetragonala ZrO2-NP att omvandlas till monoklinisk ZrO2. Dessutom finns det förhållandevis få korngränser i vår ZrO2-SiO2 NCGC, vilket också skulle kunna bidra till att stabilisera tetragonal ZrO2. Således är det möjligt att våra ZrO2-SiO2 NCGCs visar högre åldringsbeständighet än andra ZrO2-baserade biokeramer.

Vad beträffar Si3N4 biokeramer, visar biokeramen fördelaktiga effekter på ny benbildning med stordjursmodell? I denna avhandling vet vi bara att de joniska produkterna från Si3N4-biokeramik har förbättrat proliferationen och differentieringen av MC3T3-E1 preosteoblaster. Det är nödvändigt att ytter-ligare undersöka de fördelaktiga effekterna av den utvecklade Si3N4-biokeramiken med användning av en stordjursmodell.

61

Acknowledgement

I still remember the day that I arrived at Uppsala. It is September 5th, 2014. Now, four years have passed; how time flies! I learned a lot during the last four years, not just in terms of scientific knowledge and skills, but also life experience. “To be a better man” is always on my mind. I have special feel-ings for Uppsala and Sweden, regarding them as my second home not just because of the clean, quiet, and harmonious environment, but also because of the people that I met here. This is a great chance for me to express my gratitude to all of them.

First and foremost, my supervisors Wei and Håkan. Thank you for offer-ing me the chance to become a PhD student and join our group. Without your offer, I would be somewhere else, doing something else and living an-other type of life. People who have ever worked with them should know that they are a complementary team. Wei is responsible for scientific content and Håkan is more interested in turning scientific content into something real. I am lucky to have them as supervisors, which enables me to learn many dif-ferent things from them, simultaneously. With Wei, we have been able to design experiments, obtain and understand fantastic results, and also publish papers. Wei’s passion and optimism motivated me a lot. ‘What’s the value there?’ I learned a significant question from Håkan. Doing research is not just publishing paper. Papers do not make people’s life better. Products do. As material researchers, we need to turn our research into something real, something that the public can use. The question would definitely guide me throughout my academic life. Again, I am very lucky to have Wei and Håkan as my supervisors. I am thankful for your support.

My dear MIM group members: It is the MIM group members that I spend most of my time with. You all are the persons that I will never forget. Jun, you are the first person in this long list. I regard you as my classmate be-cause we met each other almost every working day and many weekends in the lab during the last four years. Thank for so much for caring for my up-coming girlfriend(s), flat, and job. Best wished to you and Jiaojiao; Michael, my dear officemate. You have asked me “how are you?” for at least one thousand times, but I am still enjoying being asked and will definitely miss this warm American-style greeting. Actually, we don’t talk much to each other in the office, because we don’t have much time for that! We love working more than we can say. Michael and Shiuli, a super nice and friendly couple full of romantic love. I am so jealous of you two. Céline, my dear

62

friend, I always feel very close to you. You are funny and lovely. I enjoyed sitting beside you and chatting with you. Dan, I always remember we took the same flight from Beijing to Arlanda airport. Not to forget the Chinese friends who had been in the group and are now off for their further adven-tures: Bing, Yixiao, Song, and Tao.

Sincere thanks to all the Swedish members in our group. All of you ena-ble me to know how nice Swedish people are. Erik, thank you for offering your valuable suggestions on glass ceramics; Cecilia, Caroline, and Marjam, so nice to have you as seniors in the group, guiding and offering us valuable advices; Gry, you are so enthusiastic and nice. I hope I could also have a nice house like yours in the future; Torbjörn, Ingrid, Camilla, and Susanne, I will remember the days that we run in the snow; Amina, Oscar, David, Christian, and Susan, it is great to have you in our group. Charlotte, thank you for your help with the nano-indentation and Swedish translation of thesis summary;

I also have many international group members who give me the chance to know nice people from all over the world. Sara, I really miss your warm Italian hug; Gemma, I really like your enthusiasm. Ale, you are the best lab manager. Thank you for organizing our lab; Luimar, a Brazilian guy who learned a lot of Chinese. It would my pleasure to show you around in China one day; Lee, when we were doing lab cleaning together I knew you are a responsible person who takes things seriously; Anna, you are a free and easy person. I should learn from you.

I received a lot of valuable support from the administration. I am thankful to Sara for paying so many invoices for me; Special thanks to Jonatan for the IT and laptop related support; I also pay my gratitude towards Maria, Ingrid for their effective help in all relevant matters.

I would like to express my thanks to all the people in the TMV division. Especially, Yuta, my first international friend and Federico, Sarah, Hugo… my volleyball friends from MST. Expert scientists at MSL have also been very helpful to me. Thanks Victoria for SEM and AFM training; Thanks Fredric and Jan-Åke for training me to use many instruments in the clean room.

I sincerely acknowledge all the collaborators. Ling and Klaus for TEM, I learned a lot from Ling, not just TEM knowledge, but also the attitude of taking details seriously; Yi for cell study; Gunnar and Stefan from SLU for the Zebrafish study …

I have met so many Chinese friends in Ångström lab. They are part of my life. I really appreciate their company. Liyang, I really enjoyed having tea and chatting with you every now and then; Yuan, Zhen, and Jiaojiao, we were together for four years. I appreciate our friendship; Rui, you are such as an industrious girl and you will definitely have a bright future; Changqing, my best badminton friend; Changgang, a kind-hearted Dongbei guy; Mingzhi, such a funny and nice friend; Xi, I regard you as a Chinese friend

63

even though your English is much better than Chinese; Wei, so nice to have you as a “master sister”. We used to have very happy period; Zhenhua, my fan; Shihuai, Wen, Peng, Dou, Hailiang, Huan, Hu, Chenjuan, Huiying, Huiming, Junxing, Lei, Bei, Jie, Bo… I hope I did not forget anyone here. Thank you so much for sharing magnificent memories with me in Ångström.

Some friends outside the department would definitely remain in my memories. I pay my gratitude to Xiaorong and Li who helped me a lot in the beginning of my life in Uppsala; Anshu, I really like your loud laugh and your absolute optimism; Daniel, my dear former roommate. You are always interested in my PhD work, which brings me a lot of confidences. I will miss the hang outs we had; Xiao and Cuiyan, we used to have really memorable journey together to Gotland and Kiruna. Best wishes to your life in Uppsala; Jiajie and Shijun, my travelling friends (‘‘lv you’’ in Chinese); Ao, Huijie, Ping, I miss you and the time we were together; Hao, Ruijun, and Gongbo, my Saturday badminton friends; ….

Special thanks to Mirva from Stockholm University for her kind help and discussion of SPS. Also, thanks for providing the nice cover page image.

I also acknowledge the financial support from China Scholarship Council (CSC), Carl Tryggers Stiftelse, and travel grant from Anna Maria Lundins.

In the end, I would like to thank my friends and relatives in China. They are proud of me, which brings me endless energy. I want to express my deepest gratitude to my parents and grandma for their constant support and unconditional dedications. I love you! I am also proud of myself (Le) for overcoming many difficulties in my private life and lab research in the last four years.

Thank you all! Le Fu July 19th, 2018 Ångström lab, Uppsala, Sweden

64

References

1. El-Meliegy, E. & Van Noort, R. Glasses and Glass Ceramics for Medical Applications, Springer New York Dordrecht Heidelberg London,. (2012).

2. Filser, F. Direct ceramic machining of dental restorations. Doctoral Thesis, ETH Zürich, (2001). doi:10.3929/ethz-a-004183626

3. Weinstein, M. & Weinstein, A. B. Porcelain covered Metal-Reinforced Teeth. United States Patent, US 3,052,983. (1962).

4. Weinstein, M. & Katz, S., Weinstein, A. B. Fused Porcelain-to-Metal Teeth. United States Patent Office, US 3,052,982. (1962).

5. Kelly, J. Ceramics in Restorative and Prosthetic Dentistry 1. Annu. Rev. Mater. Sci. 27, 443–468 (1997).

6. Denry, I. & Holloway, J. A. Ceramics for dental applications: A review. Materials (Basel). 3, 351–368 (2010).

7. Shenoy, A. & Shenoy, N. Dental ceramics: An update. J. Conserv. Dent. 13, 195 (2010).

8. Ha, C. H. F. A systematic review of the survival and complication rates of all-ceramic and metal–ceramic reconstructions after an observation period of at least 3 years . Part I : single crowns. Clin Oral Implant. Res. 3, 86–96 (2007).

9. Ivoclar-Vivadent, IPS e.max Classic®, Scientific Documentation. (2003) 10. Ivoclar-Vivadent, IPS e.max Press®, Scientific Documentation. (2011). 11. Ivoclar-Vivadent, IPS e.max ZirCAD, Scientific Documentation. (2017). 12. Guazzato, M., Albakry, M., Ringer, S. P. & Swain, M. V. Strength, fracture

toughness and microstructure of a selection of all-ceramic materials. Part I. Pressable and alumina glass-infiltrated ceramics. Dent. Mater. 20, 441–448 (2004).

13. Mussano, F., Genova, T., Munaron, L., Faga, M. G., Carossa, S. Dental Implantology and Biomaterial 64–90 (2016). doi:10.5772/62701

14. C. Piconi, G. M. Zirconia as a ceramic biomaterial. Compr. Biomater. 20, 1–25 (1999).

15. Richard A. Deyo, M.P.H. & Alf Nachemson, S. K. M. Spinal-Fusion Surgery—The Case for Restraint. Spine J. 4, 129S–142S (2004).

16. Fritz, J. M., Delitto, A., Welch, W. C. & Erhard, R. E. Lumbar spinal stenosis: A review of current concepts in evaluation, management, and outcome measurements. Arch. Phys. Med. Rehabil. 79, 700–708 (1998).

17. Bono, C. M. & Lee, C. K. Critical analysis of trends in fusion for degenerative disc disease over the past 20 years: influence of technique on fusion rate and clinical outcome. Spine. 29, 455–463 (2004).

18. Rajaee, S. S., Bae, H. W. & Kanim, L. E. A., Delamarter, R. B. Spinal Fusion in the United States. Spine. 37, 67–76 (2012).

19. Weiss A. J. & Elixhauser A, A. R. Characteristics of Operating Room Procedures in U.S. Hospitals. HCUP Stat. Br. #170. Rockville, MD Agency Healthc. Res. Qual. https://www.hcup-ahrq.gov/reports/statbriefs/sb (2011).

65

20. Weiss, A. J. & Elixhauser, A. Trends in Operating Room Procedures in U.S. Hospitals, 2001-2011. HCUP Stat. Br. #171. Rockville, MD Agency Healthc. Res. Qual. 1–14 (2014).

21. Kurtz, S. M. & Devine, J. N. PEEK biomaterials in trauma, orthopedic, and spinal implants. Biomaterials 28, 4845–4869 (2007).

22. Mei, S., Wang, H., Wang, W., Tong, L., Pan, H., Ruan, C., Ma, Q., Liu, M., Yang, H., Zhang, L., Cheng, Y., Zhang, Y, Zhao, L. & Chu, P. K. Antibacterial effects and biocompatibility of titanium surfaces with graded silver incorporation in titania nanotubes. Biomaterials 35, 4255–4265 (2014).

23. Puckett, S. D., Taylor, E., Raimondo, T. & Webster, T. J. The relationship between the nanostructure of titanium surfaces and bacterial attachment. Biomaterials 31, 706–713 (2010).

24. Mazzocchi, M. & Bellosi, A. On the possibility of silicon nitride as a ceramic for structural orthopaedic implants. Part I: Processing, microstructure, mechanical properties, cytotoxicity. J. Mater. Sci. Mater. Med. 19, 2881–2887 (2008).

25. Kersten, R. F. M. R., Van Gaalen, S. M., De Gast, A. & Öner, F. C. Polyetheretherketone (PEEK) cages in cervical applications: A systematic review. Spine J. 15, 1446–1460 (2015).

26. Zhao, Y., Wong, H.M., Wang, W., Li, P., Xu, Z., Chong, E.Y.W., Yan, C.H.Y., Yeung, K.W.K. & Chu, P.K. Cytocompatibility, osseointegration, and bioactivity of three-dimensional porous and nanostructured network on polyetheretherketone. Biomaterials 34, 9264–9277 (2013).

27. Jin, G., Qin, H., Cao, H., Qian, S., Zhao, Y., Peng, X., Zhang, X., Liu, X. & Chu, P.K. Synergistic effects of dual Zn/Ag ion implantation in osteogenic activity and antibacterial ability of titanium. Biomaterials 35, 7699–7713 (2014).

28. Gorth, D. J., Puckett, S., Ercan, B., Webster, T.J., Rahaman, M. & Bal, B.S. Decreased bacteria activity on Si3N4 surfaces compared with PEEK or titanium. Int. J. Nanomedicine 7, 4829–4840 (2012).

29. Webster, T. J., Patel, A. A., Rahaman, M. N. & Sonny Bal, B. Anti-infective and osteointegration properties of silicon nitride, poly(ether ether ketone), and titanium implants. Acta Biomater 8, 4447–4454 (2012).

30. Denry, I. & Kelly, J. R. State of the art of zirconia for dental applications. Dent. Mater. 24, 299–307 (2008).

31. Kelly, J. R. & Denry, I. Stabilized zirconia as a structural ceramic: An overview. Dent. Mater. 24, 289–298 (2008).

32. Smith, P. Ceramic steel? Nature. 253, 703–704 (1975). 33. Hannink, R. H. J., Kelly, P. M. & Muddle, B. C. Transformation Toughening in

Zirconia-Containing Ceramics. J. Am. Ceram. Soc. 83, 461–487 (2000). 34. Odi-owei, S. & Ologhadien, I. Advances in Ceramics-Electric and Magnetic

Ceramics, Bioceramics and Environment 398–420 (2011). doi:10.5772/20472 35. Piva, R. H., Piva, D. H., Montedo, O. R. K. & Morelli, M. R. Improving

physical properties of cubic InO1.5-stabilized zirconia by alloying with MoO3. J. Alloys Compd. 685, 593–603 (2016).

36. Turon-Vinas, M., Zhang, F., Vleugels, J. & Anglada, M. Effect of calcia co-doping on ceria-stabilized zirconia. J. Eur. Ceram. Soc. 38, 2621–2631 (2018).

37. Advances in Ceramics-Electric and Magnetic Ceramics, Bioceramics and Environment, Book chaper, 398–420 (2011).

38. Manicone, P. F., Rossi Iommetti, P. & Raffaelli, L. An overview of zirconia ceramics: Basic properties and clinical applications. J. Dent. 35, 819–826 (2007).

66

39. Geetha, M., Singh, A. K., Asokamani, R. & Gogia, A. K. Ti based biomaterials, the ultimate choice for orthopaedic implants - A review. Prog. Mater. Sci. 54, 397–425 (2009).

40. Niinomi, M. Mechanical properties of biomedical titanium alloys. Mater. Sci. Eng. A243 243, 231–236 (1998).

41. Katti, K. S. Biomaterials in total joint replacement. Colloids Surfaces B Biointerfaces 39, 133–142 (2004).

42. Kunčická, L., Kocich, R. & Lowe, T. C. Advances in metals and alloys for joint replacement. Prog. Mater. Sci. 88, 232–280 (2017).

43. Maccauro, G., Rossi, P., Raffaelli, L. & Francesco, P. Biomaterials Applications for Nanomedicine 300–308 (2011). doi:10.5772/23917

44. Amaral, M., Lopes, M. A., Silva, R. F. & Santos, J. D. Densification route and mechanical properties of Si3N4–bioglass biocomposites. Biomaterials 23, 857–862 (2002).

45. Chevalier, J. What future for zirconia as a biomaterial? Biomaterials 27, 535–543 (2006).

46. Chevalier, J., Gremillard, L. & Deville, S. Low-Temperature Degradation of Zirconia and Implications for Biomedical Implants. Annu. Rev. Mater. Res. 37, 1–32 (2007).

47. L.B. Kong, Y. Z. Huang, W. X. Que, T. S. Zhang, S. Li, J. Zhang, Z. L. & Dong, D. Y. T. Transparent Ceramics Materials. Springer (2015). doi:10.1016/j.optmat.2014.05.009

48. Tsukuma, K., Yamashita, I. & Kusunose, T. Transparent 8 mol% Y2O3–ZrO2 (8Y) Ceramics. J. Am. Ceram. Soc. 91, 813–818 (2008).

49. Suárez, M. & Fernández, A. Sintering to Transparency of Polycrystalline Ceramic Materials. Sintering of Ceramics - New Emerging Techniques (2012).

50. Apetz, R. & van Bruggen, M. P. B. Transparent Alumina: A Light-Scattering Model. J. Am. Ceram. Soc. 86, 480–486 (2003).

51. Khomenko, V. M., Langer, K. & Wirth, R. On the influence of wavelength-dependent light scattering on the UV-VIS absorption spectra of oxygen-based minerals: a study on silicate glass ceramics as model substances. Phys. Chem. Miner. 30, 98–107 (2003).

52. Zhang, Y. Making yttria-stabilized tetragonal zirconia translucent. Dent. Mater. 30, 1195–1203 (2014).

53. Kong, L., Huang, Y., Que, W., Zhang, T., Li S., Zhang, J., Dong, Z. & Tang, D. Transparent Ceramics. (Springer, 2015). doi:10.1111/jace.12362

54. Beall, G. H. & Pinckney, L. R. Nanophase glass-ceramics. J. Am. Ceram. Soc. 82, 5–16 (1999).

55. Klimke, J., Trunec, M. & Krell, A. Transparent tetragonal yttria-stabilized zirconia ceramics: Influence of scattering caused by birefringence. J. Am. Ceram. Soc. 94, 1850–1858 (2011).

56. Zanotto, E. D. A bright future for glass ceramics. Am. Ceram. Soc. Bull. 89, 609–612 (2011).

57. Ritzberger, C., Schweiger, M. & Höland, W. Principles of crystal phase formation in Ivoclar Vivadent glass-ceramics for dental restorations. J. Non. Cryst. Solids 432, 137–142 (2016).

58. Riley, F. L. Silicon Nitride and Related Materials. J. Am. Ceram. Soc. 83, 245–265 (2004).

59. Krstic, Z. & Krstic, V. D. Silicon nitride: The engineering material of the future. J. Mater. Sci. 47, 535–552 (2012).

60. Mazzocchi, M., Gardini, D., Traverso, P. L., Faga, M. G. & Bellosi, A. On the possibility of silicon nitride as a ceramic for structural orthopaedic implants.

67

Part II: Chemical stability and wear resistance in body environment. J. Mater. Sci. Mater. Med. 19, 2889–2901 (2008).

61. Zhang, Y. & Cheng, Y. B. Microstructural design of Ca-sialon ceramics: Effects of starting compositions and processing conditions. J. Eur. Ceram. Soc. 23, 1531–1541 (2003).

62. Regular, M., Pennycook, S. J., Gosnell, T. R., Painter, G. S., Shelton, W. A. & Becher, P. F. Observation of rare-earth segregation in silicon nitride ceramics at subnanometre dimensions. Nature 428, 2–5 (2004).

63. Liu, M. & Nemat-Nasser, S. The microstructure and boundary phases of in-situ reinforced silicon nitride. Mater. Sci. Eng. A 254, 242–252 (1998).

64. Liu, X. J., Huang, Z. Y., Ge, Q. M., Sun, X. W. & Huang, L. P. Microstructure and mechanical properties of silicon nitride ceramics prepared by pressureless sintering with MgO-Al2O3-SiO2 as sintering additive. J. Eur. Ceram. Soc. 25, 3353–3359 (2005).

65. Urakami, R., Sato, Y., Ogushi, M., Nishiyama, T., Goto, A., Yamada, K., Teranishi, R., Kaneko, & K., Kitayama, M. Phase transformation and interface segregation behavior in Si3N4 ceramics sintered with La2O3-Lu2O3 mixed additive. J. Am. Ceram. Soc. 100, 1231–1240 (2017).

66. Peng, H. Spark Plasma Sintering of Si3N4-Based Ceramics-Sintering mechanism-Tailoring microstructure-Evaluating properties. Doctoral Thesis, Stockholm University (2004). doi:10.1016/j.jallcom.2010.04.102

67. Bal, B. S. & Rahaman, M. N. Orthopedic applications of silicon nitride ceramics. Acta Biomater. 8, 2889–2898 (2012).

68. McEntire, B. J., Bal, B. S., Rahaman, M. N., Chevalier, J. & Pezzotti, G. Ceramics and ceramic coatings in orthopaedics. J. Eur. Ceram. Soc. 35, 4327–4369 (2015).

69. Anderson, M. C. & Olsen, R. Bone ingrowth into porous silicon nitride. J. Biomed. Mater. Res. - Part A 92, 1598–1605 (2010).

70. Hench, L. L. & West, J. K. The Sol-Gel Process. Chem. Rev. 90, 33–72 (1990). 71. Livage, J., Sanchez, C., Henry, M. & Doeuff, S. The chemistry of the sol-gel

process. Solid State Ionics 32–33, 633–638 (1989). 72. Owens, G. J. Sol-gel based materials for biomedical applications. Prog. Mater.

Sci. 77, 1–79 (2016). 73. Nogami, M. Glass preparation of the ZrO2-SiO2 system by the sol-gel process

from metal alkoxides. J. Non. Cryst. Solids 69, 415–423 (1985). 74. Branda, F. Advances in Nanocomposites-Synthesis, Characterization and

Industrial Applications (2011). doi:10.5772/15454 75. Zheng, K. & Boccaccini, A. R. Sol-gel processing of bioactive glass

nanoparticles: A review. Adv. Colloid Interface Sci. 249, 363–373 (2017). 76. Suárez, M., Fernandez, A., Menendez, J. L, Torrecillas, R., Kessel, H. U.,

Hennicke, J., Kichner, & R. Kessel, T. Challenges and Opportunities for Spark Plasma Sintering : A Key Technology for a New Generation of Materials. Sinter. Appl. 319–342 (2013). doi:10.5772/53706

77. Guillon, O., Julian, J. G., Dagatz, B., Kessel, T., Schierning, G., Rathel, J. & Herrrmann, M. Field-assisted sintering technology/spark plasma sintering: Mechanisms, materials, and technology developments. Adv. Eng. Mater. 16, 830–849 (2014).

78. Song, S. X., Wang, Z. & Shi, G. P. Heating mechanism of spark plasma sintering. Ceram. Int. 39, 1393–1396 (2013).

79. Nanko, M. & Dang, K. Q. Pulsed Electric Current Sintering of Transparent Alumina Ceramics, Sintering Techniques of Materials, Arunachalam Lakshmanan, IntechOpen, DOI: 10.5772/59170 (2015).

68

80. Wang, H., Li, J. F., Nan, C. W., Zhou, M., Liu, W., Zhang, B. P. & Kita, T. High-performance Ag0.8Pb18+xSbTe20 thermoelectric bulk materials fabricated by mechanical alloying and spark plasma sintering. Appl. Phys. Lett. 88, 1–4 (2006).

81. Aubert, A., Loyau, V., Mazaleyrat, F. & LoBue, M. Uniaxial anisotropy and enhanced magnetostriction of CoFe2O4 induced by reaction under uniaxial pressure with SPS. J. Eur. Ceram. Soc. 37, 3101–3105 (2017).

82. Kim, B. N., Hiraga, K., Morita, K. & Yoshida, H. Spark plasma sintering of transparent alumina. Scr. Mater. 57, 607–610 (2007).

83. Al Mansour, F., Karpukhina, N., Grasso, A., Wilson, R. M., Reece, M. J. & Cattell, M. J. The effect of spark plasma sintering on lithium disilicate glass-ceramics. Dent. Mater. 31, 1–10 (2015).

84. Kelly, J. R. Perspectives on strength. Dent. Mater. 11, 103–110 (1995). 85. Yilmaz, H., Aydin, C. & Gul, B. E. Flexural strength and fracture toughness of

dental core ceramics. J. Prosthet. Dent. 98, 120–128 (2007). 86. Kang, S.-H., Chang, J. & Son, H.-H. Flexural strength and microstructure of

two lithium disilicate glass ceramics for CAD/CAM restoration in the dental clinic. Restor. Dent. Endod. 38, 134–140 (2013).

87. Huang, C. W. & Hsueh, C. H. Piston-on-three-ball versus piston-on-ring in evaluating the biaxial strength of dental ceramics. Dent. Mater. 27, e117–e123 (2011).

88. Wagner, W. C. & Chu, T. M. Biaxial flexural strength and indentation fracture toughness of three new dental core ceramics. J. Prosthet. Dent. 76, 140–144 (1996).

89. Albakry, M., Guazzato, M. & Swain, M. V. Biaxial flexural strength, elastic moduli, and x-ray diffraction characterization of three pressable all-ceramic materials. J. Prosthet. Dent. 89, 374–380 (2003).

90. ISO 6872. Dentistry–ceramic materials. Geneva: International Organization for Standardization. (2006).

91. Cornelius, T. W. & Thomas, O. Progress of in situ synchrotron X-ray diffraction studies on the mechanical behavior of materials at small scales. Prog. Mater. Sci. 94, 384–434 (2018).

92. Oliver, W. C. & Pharr, G. M. An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. J. Mater. Res. 7, 1564–1583 (1992).

93. Oliver, W. C. & Pharr, G. M. Measurement of hardness and elastic modulus by instrumented indentation: Advances in understanding and refinements to methodology. J. Mater. Res. 19, 3–20 (2004).

94. Li, X. & Bhushan, B. A review of nanoindentation continuous stiffness measurement technique and its applications. Mater. Charact. 48, 11–36 (2002).

95. Albakry, M., Guazzato, M. & Swain, M. V. Fracture toughness and hardness evaluation of three pressable all-ceramic dental materials. J. Dent. 31, 181–188 (2003).

96. Fischer, H. & Marx, R. Fracture toughness of dental ceramics: Comparison of bending and indentation method. Dent. Mater. 18, 12–19 (2002).

97. Standard Test Methods for Determination of Fracture Toughness of Advanced Ceramics, ASTM, C1421-15, 2014. doi:10.1520/C1421-10.Annex

98. Gustavsson, F. Triboactive Component Coatings Tribological Testing and Microanalysis of. (2013).

99. David B.Williams, C. B. C. Transmission Electron Microscopy A Textbook for Materials Science. (2009).

69

100. Jarausch, K., Thomas, P., Leonard, D. N., Twesten, R. & Booth, C. R. Four-dimensional STEM-EELS: Enabling nano-scale chemical tomography. Ultramicroscopy 109, 326–337 (2009).

101. Alania, M., Altanzis, T., Backer, D., Lobato, I., Bals, S. & Aert, S. V. Depth sectioning combined with atom-counting in HAADF STEM to retrieve the 3D atomic structure. Ultramicroscopy 177, 36–42 (2017).

102. Ling Xie. Electron tomography analysis of 3D order and interfacial structure in nano-precipitates. Doctoral Thesis, Uppsala University (2016).

103. Levin, B. D. A. et al. Nanomaterial datasets to advance tomography in scanning transmission electron microscopy. Sci. Data 3, 1–11 (2016).

104. Friedrich, H., McCartney, M. R. & Buseck, P. R. Comparison of intensity distributions in tomograms from BF TEM, ADF STEM, HAADF STEM, and calculated tilt series. Ultramicroscopy 106, 18–27 (2005).

105. Collins, S. M. & Midgley, P. A. Progress and opportunities in EELS and EDS tomography. Ultramicroscopy 180, 133–141 (2017).

106. Genc, A., Kovarik, L., Gu, M., Cheng, H., Plachinda, P., Pullan, L., Freitag, B. & Wang, C. XEDS STEM tomography for 3D chemical characterization of nanoscale particles. Ultramicroscopy 131, 24–32 (2013).

107. Nagarajan, V. S. & Rao, K. J. Crystallization studies of ZrO2-SiO2 composite gels. J. Mater. Sci. 24, 2140–2146 (1989).

108. Wang, S., Huang, X. & Gou, J. Mechanical properties and microstructure of ZrO2–SiO2 composite. J. Mater. Sci. 2, 197–201 (1997).

109. Aguilar, D. H., Torres-Gonzalez, L. C., Torres-Martinez, L. M., Lopez, T. & Quintana, P. A Study of the Crystallization of ZrO2 in the Sol–Gel System: ZrO2–SiO2. J. Solid State Chem. 158, 349–357 (2001).

110. Nagarajan, V. S. & Rao, K. J. Crystallization studies of ZrO2-SiO2 composite gels. J. Mater. Sci. 24, 2140–2146 (1989).

111. Denry, I. L. & Holloway, J. a. Elastic constants, Vickers hardness, and fracture toughness of fluorrichterite-based glass-ceramics. Dent. Mater. 20, 213–219 (2004).

112. Xie, L., Jarolimek, K., Kocevski, V., Rusz, J., Zeman, M., Swaaij, R. & Leifer, K. Electron tomography analysis of 3D interfacial nanostructures appearing in annealed Si rich SiC films. Nanoscale 9, 6703–6710 (2017).

113. Thanh, N. T. K., Maclean, N. & Mahiddine, S. Mechanisms of nucleation and growth of nanoparticles in solution. Chem. Rev. 114, 7610–7630 (2014).

114. Pulker, H. K. Mechanical properties of optical films. Thin Solid Films 89, 191–204 (1982).

115. Xia, K. & Langdon, T. G. The toughening and strengthening of ceramic materials through discontinuous reinforcement. J. Mater. Sci. 29, 5219–5231 (1994).

116. Khan, A., Philip, J. & Hess, P. Young’s modulus of silicon nitride used in scanning force microscope cantilevers. J. Appl. Phys. 95, 1667–1672 (2004).

117. Knoch, H. & Gazza, G. E. On the α to β phase transformation and grain growth during hot-pressing of Si3N4 containing MgO. Ceramurg. Int. 6, 51–56 (1980).

118. Miyazaki, H., Hyuga, H., Yoshizawa, Y. I., Hirao, K. & Ohji, T. Correlation of wear behavior and indentation fracture resistance in silicon nitride ceramics hot-pressed with alumina and yttria. J. Eur. Ceram. Soc. 29, 1535–1542 (2009).

119. Miyazaki, H., Hyuga, H., Yoshizawa, Y. I., Hirao, K. & Ohji, T. Crack profiles under a Vickers indent in silicon nitride ceramics with various microstructures. Ceram. Int. 36, 173–179 (2010).

70

120. Pammi, M., Liang, R., Hicks, J. M., Barrish, J. & Versalovic, J. Farnesol decreases biofilms of staphylococcus epidermidis and exhibits synergy with nafcillin and vancomycin. Pediatr. Res. 70, 578–583 (2011).

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology 1710

Editor: The Dean of the Faculty of Science and Technology

A doctoral dissertation from the Faculty of Science andTechnology, Uppsala University, is usually a summary of anumber of papers. A few copies of the complete dissertationare kept at major Swedish research libraries, while thesummary alone is distributed internationally throughthe series Digital Comprehensive Summaries of UppsalaDissertations from the Faculty of Science and Technology.(Prior to January, 2005, the series was published under thetitle “Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-357128

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2018