Article Images (0)
×![]()
featured_play_listSave article to folder
X
Create new folder
Full Text
PDF
Table of Contents
Abstract
1. Introduction
2.2. Specimen Preparation
2.3. X-Ray Diffraction (XRD)
2.4. Flexural Strength
2.5. Fracture Toughness
2.6. Nanoindentation
2.9. Statistical Methods
3. Results
3.1. XRD Patterns
3.1.1. The Not-Fired, 530–590, 590–750, And 590–750 ◦C
H14 Groups
3.1.2. The 750–780 ◦C Group
3.1.3. The 750–840, 820–840, And 820–840 ◦C (H14) Groups
3.2. Physical Properties
3.2.1. Flexural Strength, Flexural Modulus, And Fracture Toughness
3.2.2. Nanoindentation—Elastic Modulus
3.2.3. Nanoindentation—Hardness
3.3. Microstructural Evolution
50–840 ◦C, (G) 820–840 ◦C (Recommended), (H) 820–840 ◦C (H
4. Discussion
4.1. Relationship Between Heating Schedules,
4.3. Comparison With Past Studies
5. Conclusions
Acknowledgements
Abstract
Background. Elucidating the microstructural responses of the lithium disilicate system like the popular IPS e.max® CAD (LS2), made specifically for computer-aided design and computer-aided manufacturing (CAD-CAM), as a temperature-dependent system unravels new ways to enhance material properties and performance. Objective. To study the effect of various thermal processing on the crystallization kinetics, crystallite microstructure, and strength of LS2. Methods. The control group of the LS2 samples was heated using the standard manufacturer heating-schedule. Two experimental groups were tested: (1) an extended temperature range (750–840 ◦C vs. 820–840 ◦C) at the segment of 30 ◦C/min heating rate, and (2) a protracted holding time (14 min vs. 7 min) at the isothermal temperature of 840 ◦C. Five other groups of different heating schedules with lower-targeted temperatures were evaluated to investigate the microstructural changes. For each group, the crystalline phases and morphologies were measured by X-ray diffraction (XRD) and scanning electron microscope (SEM), respectively. Differential scanning calorimeter (DSC) was used to determine the activation energy of LS2 under non-isothermal conditions. A universal testing machine was used to measure 3-point flexural strength and fracture toughness, and elastic modulus and hardness were measured by a nanoindenter. A one-way ANOVA/Tukey was performed per property (alpha = 0.05). Results. DSC, XRD, and SEM revealed three distinct microstructures during LS2 crystallization. Significant differences were found between the control group, the two aforementioned experimental groups, and the five lower-targeted-temperature groups per property (p < 0.05). The activation energy for lithium disilicate growth was 667 (±29.0) kJ/mol. Conclusions. Groups with the extended temperature range (750–840 ◦C) and protracted holding time (820–840 ◦C H14) produced significantly higher elastic-modulus and hardness properties than the control group but showed similar flexural-strength and fracturePlease cite this article in press as: Lien W, et al. Microstructural evolution an (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 toughness properties with occurred only when maxi
D M a W T a b c d e a A R R 2 A A K I L L G N M P H T D N C h 0 ARTICLE IN PRESSENTAL-2563; No. of Pages 13 d e n t a l m a t e r i a l s x x x ( 2 0 1 5 ) xxx–xxx Available online at www.sciencedirect.com ScienceDirect jo ur nal home p ag e: www.int l .e lsev ierhea l th .com/ journa ls /dema icrostructural evolution and physical behavior of lithium disilicate glass–ceramic en Liena,b,∗, Howard W. Roberts c, Jeffrey A. Plattb, Kraig S. Vandewalled, homas J. Hill e, Tien-Min G. Chub United States Air Force Institute of Technology, Wright-Patterson Air Force Base, OH, USA Indiana University School of Dentistry, Indianapolis, Indiana, USA United States Air Force, Keesler Air Force Base, MS, USA United States Air Force, Joint Base San Antonio, TX, USA Ivoclar Vivadent, Amherst, NY, USA r t i c l e i n f o rticle history: eceived 31 October 2014 eceived in revised form 0 March 2015 ccepted 7 May 2015 vailable online xxx eywords: PS e.max® CAD ithium disilicate ithium metasilicate lass–ceramic anoindentation icrostructure hase transformation eating schedule emperature threshold ifferential scanning calorimetry ucleation rystallization a b s t r a c t Background. Elucidating the microstructural responses of the lithium disilicate system like the popular IPS e.max® CAD (LS2), made specifically for computer-aided design and computer-aided manufacturing (CAD-CAM), as a temperature-dependent system unravels new ways to enhance material properties and performance. Objective. To study the effect of various thermal processing on the crystallization kinetics, crystallite microstructure, and strength of LS2. Methods. The control group of the LS2 samples was heated using the standard manufacturer heating-schedule. Two experimental groups were tested: (1) an extended temperature range (750–840 ◦C vs. 820–840 ◦C) at the segment of 30 ◦C/min heating rate, and (2) a protracted holding time (14 min vs. 7 min) at the isothermal temperature of 840 ◦C. Five other groups of different heating schedules with lower-targeted temperatures were evaluated to investigate the microstructural changes. For each group, the crystalline phases and morphologies were measured by X-ray diffraction (XRD) and scanning electron microscope (SEM), respectively. Differential scanning calorimeter (DSC) was used to determine the activation energy of LS2 under non-isothermal conditions. A universal testing machine was used to measure 3-point flexural strength and fracture toughness, and elastic modulus and hardness were measured by a nanoindenter. A one-way ANOVA/Tukey was performed per property (alpha = 0.05). Results. DSC, XRD, and SEM revealed three distinct microstructures during LS2 crystallization. Significant differences were found between the control group, the two aforementioned experimental groups, and the five lower-targeted-temperature groups per property (p < 0.05). The activation energy for lithium disilicate growth was 667 (±29.0) kJ/mol. Conclusions. Groups with the extended temperature range (750–840 ◦C) and protracted holding time (820–840 ◦C H14) produced significantly higher elastic-modulus and hardness properties than the control group but showed similar flexural-strength and fracture- Please cite this article in press as: Lien W, et al. Microstructural evolution an (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 toughness properties with occurred only when maxi Pu ∗ Corresponding author at: United States Air Force Institute of Technol E-mail address: wenlien2003@yahoo.com (W. Lien). ttp://dx.doi.org/10.1016/j.dental.2015.05.003 109-5641/Published by Elsevier Ltd on behalf of Academy of Dental Ma d physical behavior of a lithium disilicate glass–ceramic. Dent Mater the control group. In general, rapid growth of lithium disilicates mum formation of lithium metasilicates had ended. blished by Elsevier Ltd on behalf of Academy of Dental Materials. ogy, Wright-Patterson Air Force Base, OH, USA. Tel.: +1 2105087456. terials. ARTICLE IN PRESSDENTAL-2563; No. of Pages 13 s x x 2 d e n t a l m a t e r i a l
1. Introduction
Lithium disilicate glass–ceramics were first introduced into the dental community in 1998 by Ivoclar Vivadent. Since its inception, dental research on the lithium disilicate glass–ceramics have been based on the commercial product, IPS Empress® 2 (Ivoclar Vivadent, Schaan, Liechtenstein). It contained approximately 65% volume fraction of lithium disilicates, 34% volume fraction of residual glass, and 1% volume fraction of porosity after heat treatments [1]. Unlike the binary lithium disilicate system that was first developed by Stookey [2], the IPS Empress® 2 was derived from a multi-component system, formulated from SiO2–Li2O–K2O–ZnO–Al2O3–La2O3–P2O5 compositions [3,4]. Scanning electron micrographs of IPS Empress® 2 revealed that the microstructures of lithium disilicates were elongated crystals with a mean grain length and diameter of 5.2 and 0.8 m, respectively [1]. In contrast to IPS Classic®, for which uncontrolled devitrification of leucites occurred only on the surface [5,6], the controlled crystallization of IPS Empress® 2 ensured that nucleation and crystal growth of lithium disilicates propagated uniformly throughout the bulk structure during heat treatments [3,6]. The nucleation in IPS Empress® 2 was achieved with the aid of special additives (e.g., P2O5, TiO2 and ZrO2) [7,8]. Additionally, these additives could alter the eutectic composition and temperature of the IPS Empress® 2 glass–ceramic [9]. According to Headley and Loehman, at low temperature, P2O5 amassed and formed the crystalline nuclei of lithium orthophosphates. Then, lithium metasilicates, lithium disilicates, and cristobalites could be crystallized by epitaxial growth on those lithium orthophosphates [10]. Besides the special additives, the growth of lithium disilicate crystals could also be affected by a one- or twostage heating schedule. The one-stage heating schedule only involved a single heating rate and holding time. The twostage heating schedule typically entailed first and second heat treatments for nucleation then crystallization, respectively [7,11]. The initial heat-treatment stage was important to establish a kinetically favorable setting for stabilizing lithium metasilicates [11]. The second heat-treatment stage, usually at a higher temperature range than the initial, supplied the thermal energy to induce growth of lithium disilicates and to thermodynamically destabilize the lithium metasilicates [11]. According to Borom et al., the growth of lithium disilicate crystals was not dependent on the crystalline nuclei of lithium metasilicates [11]. Rather, lithium metasilicates kinetically competed with lithium disilicates but slowly diminished since it was thermodynamically less stable than lithium disilicates at high temperatures [11]. In contrast, Zheng et al. suggested an interdependence between lithium metasilicates and lithium disilicates, where lithium disilicates could be epitaxially grown on lithium metasilicates [7]. Past investigations have argued that a two-stage heating schedule precipitated more and larger lithium disilicate crystals than a single-stage Please cite this article in press as: Lien W, et al. Microstructural evolution an (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 heating schedule [7,11–13]. Even though the single-stage heating schedule might require less overall processing time, it tended to lack the appropriate thermal enrichment for maturation of lithium disilicate crystals [7,12]. Because of this, when x ( 2 0 1 5 ) xxx–xxx a DSC was used to investigate the non-isothermal crystallization kinetics of a lithium disilicates system, the separation between the exothermic peaks of lithium metasilicates and lithium disilicates were less distinguishable, which further indicated that amidst the glass–ceramic microstructures, the lithium metasilicates and lithium disilicates were less identifiable from one to another for the single-stage heating schedule that encourages a fast or ultrafast heating rate. Hence, in theory the timing of the sequential heat treatments and the heating rates were critical to the origin, as well as discontinuation, of the microstructural segregation and to the nucleation and propagation of lithium disilicate crystals. With the advent of CAD-CAM technology and phasing out of IPS Empress® 2, newer generations of glass–ceramic blocs were introduced to accommodate the ease of milling, to maximize cutting efficiency, and to prolong the life of the milling tools. Today, the insertion of a chair-side IPS e.max® CAD prosthesis involves three fabricating progressions: industrial casting of the blocs, CAD milling, and final thermal refinement for enriching lithium disilicate crystallization. First, according to the manufacturer, glass compositions (mainly SiO2, Li2O, P2O5, ZrO2, ZnO, and K2O) are incongruently melted, quenched, and annealed to form blue ingots, IPS e.max® CAD blocs [14]. The blue tint, acquired from the added colorants, is evidence that the bloc has undertaken a partially glassycrystalline transformation and signifies its readiness for the second process, CAD milling. In this partially crystallized state, these intermediates inherit a mild to moderate strength and hardness, which can be easily machined by any popular CADCAM system. Often, this second process can be conveniently done in a private dental practice. After milling, it is then transformed by a two-stage heat treatment into a dental prosthesis containing both glassy phase and lithium disilicate crystals. Different heating parameters can upset the driving force for growing lithium disilicates and can alter the overall percentage of residual glasses [11,15–17]. Theoretically, glass–ceramic prostheses, containing an extra residual glassy phase, are more likely to adversely impact a number of properties including load-bearing capacity, resistance to acidic attacks, and fracture toughness [8]. In contrast, amplifying crystallization lowers the coefficient of thermal expansion, improves the resistance to thermal shock, and increases prosthetic strength [18–20]. Although many studies have been conducted to evaluate the clinical performance and potential shortcomings of lithium disilicate glass–ceramics in comparison to other popular types of dental materials, only a few focus on the glass–ceramics’ properties from an intrinsic perspective of crystallization, phase assembly, thermal history, and kinetics. Additionally, most of that handful of studies has been confined within the erudite realms of the pure or binary Li2O–SiO2 systems [7,13,21–25]. Exploration on how a “multi-component” CAD-CAM bloc crystallizes has been very limited [26]. Further investigation in describing the intricate interplay between thermal treatments and crystalline architecture exhibited by these materials can offer insights on how their atomic-scale d physical behavior of a lithium disilicate glass–ceramic. Dent Mater behaviors can transcend to distress or to fortify their macroscopic material properties. Most importantly, clarification on why lithium metasilicates tend to evolve to form lithium disilicates needs to be addressed, so their desired clinical properties ARTICLE IN PRESSDENTAL-2563; No. of Pages 13 x x x ( 2 0 1 5 ) xxx–xxx 3 c o r g f c g T b o i u l o w t t o 8 t 3 o w s r m i o e p 8 g m 2 2 B s t 8 s t c s t a h a t o r w s m o t Tw o- st ag e h ea ti n g sc h ed u le s. N ot fi re d 53 0– 59 0 ◦ C 59 0– 75 0 ◦ C 59 0– 75 0 ◦ C (H 14 ) 75 0– 78 0 ◦ C 75 0– 84 0 ◦ C R ec om m en d ed 82 0– 84 0 ◦ C 82 0– 84 0 ◦ C (H 14 ) St ag e 1 St ag e 2 St ag e 1 St ag e 2 St ag e 1 St ag e 2 St ag e 1 St ag e 2 St ag e 1 St ag e 2 St ag e 1 St ag e 2 St ag e 1 St ag e 2 St ag e 1 St ag e 2 N ot ap p li ca bl e 40 3 40 3 40 3 40 3 40 3 40 3 40 3 0. 3 0. 3 0. 3 0. 3 0. 3 0. 3 0. 3 90 30 90 30 90 30 90 30 90 30 90 30 90 30 53 0 59 0 59 0 75 0 59 0 75 0 75 0 78 0 75 0 84 0 82 0 84 0 82 0 84 0 0: 10 7 0: 10 7 0: 10 14 5 14 0: 10 7 0: 10 7 0: 10 14 45 0 53 0 45 0 59 0 45 0 59 0 55 0 75 0 45 0 75 0 55 0 82 0 55 0 82 0 53 0 59 0 59 0 75 0 59 0 75 0 75 0 78 0 75 0 84 0 82 0 84 0 82 0 84 0 im e N ot ap p li ca bl e 10 .8 8 14 .8 8 21 .8 8 24 .1 6 14 .3 2 12 .7 7 19 .7 7 rn ac e st an d -b yte m p er at u re , S (m in ) = fu rn ac e d oo r cl os in g ti m e, t (◦ C /m in ) = h ea ti n g or ra m p ra te , T (◦ C ) = h ol d in g te m p er at u re , H (m in ) = h ol d in g ti m e, V 1 (◦ C ) = va cu u m -o n te m p er at u re , V 2 u m -o ff te m p er at u re , ( H 14 ) = h ol d fo r 14 m in . d e n t a l m a t e r i a l s an be deliberately manifested through the manipulation f heat treatments. Here, we studied the history-dependent esponse (thermal versus physical) of a multi-component lass–ceramic, named IPS e.max® CAD that is sold in the orm of a partially crystallized precursor, and endeavored to omprehend its kinetic process through analysis of its emerent microstructures and macroscopic physical properties. he aim of this study was to characterize the transformative ehavior, crystallizing kinetics, and microstructural evolution f a partially crystallized glass precursor (IPS e.max® CAD) nto lithium disilicate glass–ceramics. According to the manfacturer, the heating schedule for inducing crystallization of ithium disilicates within an IPS e.max® CAD bloc consisted f two (double) heating rates and two holding times, each of hich was initiated and held at a specific targeted temperaure (see Table 1 for the group labeled as 820–840 ◦C). Initially, he partially crystallized precursor was heated at a rapid rate f 90 ◦C/min from 403 ◦C (furnace stand-by-temperature) to 20 ◦C and held for 10 s at 820 ◦C (first targeted temperaure). This was followed by a slower, second heating rate of 0 ◦C/min. Then, it was held for a period of 7 min at 840 ◦C (secnd targeted temperature). In this study, we hypothesized that hen IPS e.max® CAD is thermally processed under a twotage heating schedule, an early onset of the second heating ate at a lower targeted temperature (750 ◦C) than the recom- ended (820 ◦C), which causes a time extension of the heating nterval for the second heating stage, will have an impact n the glass–ceramic’s flexural strength, fracture toughness, lastic modulus, and hardness. We also hypothesized that rotracting the holding time at the isothermal temperature, 40 ◦C, of the second heating stage will have an impact on the lass–ceramic’s flexural strength, fracture toughness, elastic odulus, and hardness. . Materials and methods .1. Heating schedules ased on past studies and manufacturer recommendations, even unique two-stage heating schedules were developed o evaluate the IPS e.max® CAD blocs. See Table 1. Group 20 840 ◦C represented the manufacturer’s recommended twotage heating schedule and was the control group. Here, the wo-stage heating schedule was designed to thermally proess a glass precursor in two successive stages, where each tage consisted of a unique heating rate, holding time, and argeted temperature. The targeted temperature was defined s the terminal temperature point at which the ramping of eat at a particular heating rate was ended and as the start of n additional ramping of heat at a new heating rate. Usually, he first heating rate was ramped much faster than the secnd heating rate. In this work, we followed the manufacturer’s ecommendation for which the first and second heating rates ere maintained at 90 and 30 ◦C/min respectively. The reaon behind this was for consistency, ease of comparison, and Please cite this article in press as: Lien W, et al. Microstructural evolution and physical behavior of a lithium disilicate glass–ceramic. Dent Mater (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 inimizing covariates. All heating schedules were derivatives of the recmmended two-stage heating schedule, but the targeted emperatures and the second holding times were modified. Ta bl e 1 – B (◦ C ) S (m in ) t (◦ C /m in ) T (◦ C ) H (m in ) V 1 (◦ C ) V 2 (◦ C ) H ea ti n g t (m in ) B (◦ C ) = Fu (◦ C ) = va cu ARTICLE IN PRESSDENTAL-2563; No. of Pages 13 s x x 4 d e n t a l m a t e r i a l The heating schedules for the 530–590, 590–750, 590–750 (H14), and 750–780 ◦C groups allowed us to study the evolutionary development of the lithium disilicate system. For the 750–840 ◦C group, the second heating rate (30 ◦C/min) began at a lower onset temperature than the control group (750 ◦C vs. 820 ◦C). This would protract the time for the second heat ramping to reach the final temperature of 840 ◦C since it was ramping at a speed of 30 ◦C/min instead of 90 ◦C/min. The control group would take less time to complete its second heat ramping as compared with the 750–840 ◦C group since it was ramping through a narrower temperature interval of 20 ◦C scale versus an interval of 90 ◦C scale for the 750–840 ◦C group. For the 820–840 ◦C (H14) versus the control group, their difference was the longer holding time of 14 min as opposed to the regular 7 min at 840 ◦C. For this study, furnace stand-by temperature, door closing time, and heating rates were held constant. Additionally, an ultra-short first holding time of 10 s was followed by a second holding time of either 7 or 14 min. Thus, the overall heating time was calculated by summing the time for closing the furnace door, the two two-stage ramp periods, and the holding times.
2.2. Specimen preparation
Following the ISO Specification 6872 [27], the IPS e.max® CAD blocs were sectioned into bars using a diamond saw (Isomet 1000, Buehler, Lake Forest, IL). The rectangular bars were randomly but equally divided into the eight groups of various firing schedules. See Table 1. Twelve rectangular bars per group were used (i.e., for the flexural test, n = 96, and for the fracture toughness test, n = 96). After firing, all surfaces of the bar were polished using silicon carbide paper of 600-, 800-, 1000-, and 1200-grit (EXAKT Technologies, Oklahoma City, OK, USA) under running water at 300 rpm on a polishing machine (EXAKT 400 CS, EXAKT Technologies, Oklahoma City, OK, USA). After polishing with each of the various grits, the specimens were rinsed with water. The specimens were stored dry until testing was performed.
2.3. X-ray diffraction (XRD)
The XRD data were collected from three representative specimens per group (obtained from the fragments of the 3-point flexure test) by using a D8 Discover X-ray diffractometer with two-dimensional VÅNTEC-500 detector (Bruker Instruments, Billerica, MA, USA). Using monochromatic radiation ( K = 1.5406 Å), each specimen was scanned in bulk over the 2 range, 16–82◦, with an angular resolution of 0.005◦ for identifying the crystalline phases.
2.4. Flexural strength
The 3-point flexure test was performed as recommended by ISO Specification 6872 [27], and the flexural strengths, FS (MPa), were calculated according to the following formula: FS = 3Fl/2bd2, where F was the breaking load (N); l was the test Please cite this article in press as: Lien W, et al. Microstructural evolution an (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 span (mm); b was the width of the specimen (mm); and d was the thickness of the specimen (mm). The 3-point flexure test fixture consisted of two cylinders with a radius of 0.8 mm (span distance, 15 mm) and a loading, cylindrical head with a radius x ( 2 0 1 5 ) xxx–xxx of 0.8 mm. The IPS e.max® CAD blocs were prepared into bars (1.3 mm × 4 mm × 18 mm) as described in the sample preparation section. Each specimen was loaded to failure (crosshead speed = 0.5 mm/min) using a universal testing machine (MTS Sintech ReNew 1123, MTS Systems, Eden Prairie, MN, USA), at room temperature. The flexural modulus was acquired from the slope of the best-fitted linear region of the load-deflection curve using TestWorks® software (MTS Systems, Eden Prairie, MN, USA). The mean and standard deviation were then calculated.
2.5. Fracture toughness
The fracture toughness values were determined by a single-edge notched-beam method, ISO Specification 6872 [27]. The IPS e.max® CAD blocs were prepared into bars (1.3 mm × 4 mm × 18 mm) as described in the sample preparation section. The notches of the specimens were prepared with a diamond saw (blade thickness = 0.3 mm, EXAKT 300, EXAKT Technologies, Oklahoma City, OK, USA). All root radii of the prepared notches were then manually refined using a singleedged razor blade and diamond polishing paste. The final notch depth and root radius were 1.0 ± 0.2 and 0.05 ± 0.02 mm, respectively, which was verified by using a stereomicroscope (Nikon Measurescope UM-2, Shinjuku, Tokyo, Japan). The KIC (MPa m0.5) values were calculated using the following equations: KIC = ( PS bw √ w )( 3 √ ˛ 2(1 − ˛)1.5 ) Y where Y = 1.9472 − 5.0247˛ + 11.8954˛2 − 18.0635˛3 + 14.5986˛4 − 4.6896˛5 and ̨ = a/w; and, where P, S, a, b, and w were peak load (MPa), test span length (m), notch depth (m), specimen thickness (m), and specimen width (m), respectively. The specimens were tested in a similar manner as flexural strength in a universal testing machine at a crosshead speed of 1 mm/min. The mean and standard deviation were then calculated.
2.6. Nanoindentation
A MTS Nanoindenter® XP (MTS Systems, Eden Prairie, MN, USA) equipped with TestWorks® software (MTS Systems, Eden Prairie, MN, USA) and fitted with a tetrahedral Berkovich diamond indenter tip (Serial # TB20128, MTS Systems, Eden Prairie, MN, USA) of 20 nm radius (faces 65.3◦ from vertical axis) was used to measure all specimens. A linear array of indents (100 indents/group) was diagonally imprinted on the polished surfaces obtained from the fragments of the 3-point flexure test. Each consecutive indent was spaced 30 m apart from each other to avoid any interference of residual stresses from adjacent imprints. Force–displacement curves for the indents were used to evaluate the elastic moduli. For each indent, elastic modulus was calculated using the standard methods of Oliver and Pharr [28]. The Elastic modulus, E (GPa), per group was computed with the following expression, d physical behavior of a lithium disilicate glass–ceramic. Dent Mater E = (1 − v2) [ 1 Er − v 2 i Ei ]−1 ARTICLE IN PRESSDENTAL-2563; No. of Pages 13 x x x w i E ( d d w i 2 M E T t r m 9 w e g i w t n u u 2 A C i T s s o r v t C t u h h i E w L b t a o U t p b l d e n t a l m a t e r i a l s here and Er (GPa) were the Poisson’s ratio of lithium disilcate glass–ceramic and reduced modulus [29–31], and i and i (GPa) were the Poisson’s ratio (0.07) and elastic modulus 1141 GPa) of the Berkovich indenter, respectively. The nanoinentation hardness was obtained from the indentation load ivided by the projected contact area, A (nm2), hardness = P/A, here P (mN) is the maximum contact force exerted by the ndenter onto the sample. .7. Scanning electron microscopy (SEM) icrostructural analyses were performed using a Field mission-SEM (Sigma VP, Carl Zeiss, Oberkochen, Germany). o study the microstructures of lithium disilicate crystals, he polished surfaces of the glass–ceramic specimens (three epresentative specimens per group obtained from the frag- ents of the 3-point flexure test) were etched with an aqueous % hydrofluoric acid (HF) for 1 min. This etching procedure as necessary to partially remove the glassy phase, thereby nhancing the image contrast between the crystalline and lassy phases under SEM. After the chemical etching, the specmens were washed several times using acetone and distilled ater. Next, they were placed in an ultrasonic bath at room emperature for 10 min to remove residuals of HF and exteral particles adhering to the surfaces. Then, they were imaged nder SEM after being sputter-coated with gold (Denton Vacum Desk II, Denton Vacuum, Moorestown, NJ, USA). .8. Differential scanning calorimetry (DSC) differential scanning calorimeter (DSC822e, Mettler-Toledo, olumbus, OH, USA) was used to investigate the non- sothermal crystallization kinetics of the IPS e.max® CAD. emperature and sensitivity calibrations were done in the ame experimental conditions as those used for the actual amples. The non-isothermal experiments were performed n forty IPS e.max® CAD specimens (10 specimens/heating ate) that were without any previous thermal treatment. Four ariable heating rates (5, 10, 15, 20 ◦C/min) in the temperaure range of 500–880 ◦C were done. Another ten IPS e.max® AD specimens were tested in the DSC that strictly adhered to he manufacturer’s recommended two-stage heating schedle, where each of the partially crystallized precursors was eated at a rapid rate of 90 ◦C/min from 403 to 820 ◦C and eld for 10 s at 820 ◦C, was followed by a slower, second heatng rate of 30 ◦C/min, and then was held for 7 min at 840 ◦C. ach specimen’s dimension was 2 mm × 3 mm × 4 mm, which as prepared by using a diamond saw (Isomet 1000, Buehler, ake Forest, IL), and was tested in a platinum crucible for etter thermal conductivity and under nitrogen atmosphere o prevent extensive thermal degradation. To determine the ctivation energy, the approach used in this study was based n the theoretical model formulated by Kissinger [32–34]. sing the Kissinger model, the relationship between a paricular heating rate, ˇi (e.g., 5, 10, 15, or 20 K/min), and the eak exothermic (crystallization) temperature, Tp (K), could Please cite this article in press as: Lien W, et al. Microstructural evolution an (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 e expressed as the following, n ( ˇi Tp ) = E RTp + constant ( 2 0 1 5 ) xxx–xxx 5 where E (kJ/mole) is the crystallization activation energy, and R is the gas constant [8.3145 J/(K mol)]. A plot of ln(ˇi/Tp) versus 1/Tp would then yield a straight line with slope E/R, whose terms could be rearranged to obtain the activation energy, E [35]. Also, the exothermic energies (peak area normalized against mass) were acquired from the DSC curves, and the mean and standard deviation were then calculated for both one-stage and two-stage heating schedules.
2.9. Statistical methods
The statistics of the measured properties was analyzed by oneway analysis of variance (ANOVA) and Tukey’s post hoc tests at alpha = 0.05 significance using SAS® 9.4 statistical software (SAS Institute Inc., Cary, NC, USA).
3. Results
3.1. XRD patterns
3.1.1. The not-fired, 530–590, 590–750, and 590–750 ◦C
H14 groups
The XRD patterns for the eight groups are presented in Fig. 1, and they are organized by their temperature intervals at the second heating stage of a two-stage heating schedule, from the lowest to the highest temperature intervals. Starting with the “not-fired” group at the bottom of Fig. 1, the diffraction pattern near the baseline, ranging from the 2-theta scale of 16–38◦, shows a widely distributed “hump”, which represents the glassy phase within the IPS e.max® CAD blocs. As the temperature was gradually elevated and the “glassy hump” slowly dwindled but did not disappear, its continual presence across all eight of the XRD patterns demonstrates the tenacity of residual glasses within the glass–ceramic matrix. This justifies that the heat-treated IPS e.max® CAD material can be categorized as a glassy-crystalline material. Groups that were treated within the second-stage thermal interval of 530–750 ◦C (i.e., 530–590, 590–750, and 590–750 ◦C H14) exhibited similar XRD patterns as compared to the not-fired group. Their major XRD peaks are identified to be the lithium metasilicate [Li2SiO3 or Li2O–SiO2] and lithium orthophosphates [Li3PO4], using ICCD 029-0829 and ICCD 025-1030 respectively.
3.1.2. The 750–780 ◦C group
As shown in Fig. 1, the different peaks that appeared in the 750–780 ◦C XRD pattern were indicative signs of a glass–ceramic that consisted of three major phases: lithium disilicates, cristobalite, and lithium orthophosphates; their identification was made by using ICCD 040-0376, 015-0637, and 039-1425. When the five XRD patterns, ranging from the bottom of Fig. 1 up to the 750–780 ◦C group, were simultaneously surveyed, they revealed a glass–ceramic that was being transformed from predominantly lithium metasilicates’ contents into a heterogeneous mixture of different phases. Since the XRD peak intensities have been used to qualita- d physical behavior of a lithium disilicate glass–ceramic. Dent Mater tively estimate the relative proportions of different phases in a glass–ceramic system by comparing peak intensities attributed to the identified phases [36], the relative peak intensities between the three major phases in the 750–780 ◦C ARTICLE IN PRESSDENTAL-2563; No. of Pages 13 6 d e n t a l m a t e r i a l s x x x ( 2 0 1 5 ) xxx–xxx ray d Fig. 1 – X- group suggested that lithium metasilicates continued to thrive within the glass–ceramic network while lithium disilicate and cristobalite crystals started to amass. Hence, for groups treated with temperature below 780 ◦C, including the not-fired group, lithium metasilicates were observed as the main crystalline phase.
3.1.3. The 750–840, 820–840, and 820–840 ◦C (H14) groups
For groups treated with the thermal ranges above 780 ◦C, the precipitations of lithium disilicates [Li2Si2O5 or Li2O–2SiO2] were seen as the main crystalline phase. This was determined by Fig. 1a–c for which the intensities of their three strongest peaks (23.8, 24.4, and 24.9) represented the (1 3 0), (0 4 0), and (1 11 ) crystallographic planes of the Li2O–2SiO2 monoclinic phase. In addition, the intensities for these three strongest peaks demonstrated a gradual increase in comparison to their infancy state when treated with the temperature interval between 750 and 780 ◦C. This showed that a greater amount of lithium disilicate crystallization developed for groups treated with the temperature intervals above 780 ◦C than the 750–780 ◦C group. Also, the XRD patterns exhibited other minor chemical species. For example, a discernable peak, fused at its baseline and comprised of three local maxima, settled at the 21.8 next to the 22.4 peak; this peak denoted the presence of cristobalite. Furthermore, the thermodynamically less stable remnants, lithium metasilicates and lithium orthophosphates, persisted at 41.4◦ and 72.6◦, respectively. The 820–840 (H14) and 750–840 ◦C groups have Please cite this article in press as: Lien W, et al. Microstructural evolution an (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 diffraction angles matching the recommended group, demonstrating a steady amount of Li2O–2SiO2 crystalline growth as well as the presence of lithium metasilicate remnants. Thus, the XRD patterns have revealed that the transformation from iffraction. lithium metasilicates to lithium disilicates was dependent on the heating temperature but independent of the overall heating time (See Table 1). A minimum threshold of 780 ◦C has to be crossed for growth and maturation of lithium disilicates.
3.2. Physical properties
Significant differences were found between groups per physical property. Fig. 2A–E graphically summarizes the measured results and statistics for all physical properties. Except for fracture toughness and nano-hardness, groups treated with temperatures surpassing 780 ◦C, which were 750–840, 820–840, and 820–840 ◦C (H14), significantly outperformed the groups treated with temperatures below 780 ◦C in every aspect of the tested properties. For this study, a generalized upward trend existed for Fig. 2A and C, such that the flexural strength and fracture toughness started at a minimum, then, gradually sloped upward, and finally reached a plateau. Furthermore, in Fig. 2A–E, the temperature interval, 750–780 ◦C, demarcated a transitional point, where a change, for better or worse, in physical properties was about to commence.
3.2.1. Flexural strength, flexural modulus, and fracture toughness
The glass–ceramic, IPS e.max® CAD, exhibited significant differences in flexural strength at three distinctive thermal ranges: below 590 ◦C, between 590 and 780 ◦C, and above 780 ◦C. The three highest flexural strength values were 350 ± 43.0, 367 ± 43.3, and 362 ± 78.6 MPa for groups, 750–840, ◦ d physical behavior of a lithium disilicate glass–ceramic. Dent Mater 820–840, 820–840 C (H14), respectively (see Fig. 2A). For the physical property of flexural modulus, the 820–840 ◦C group (66.6 ± 5.52 GPa) demonstrated significantly higher flexural modulus than all other groups while the next two highest were ARTICLE IN PRESSDENTAL-2563; No. of Pages 13 d e n t a l m a t e r i a l s x x x ( 2 0 1 5 ) xxx–xxx 7 Fig. 2 – n = 12 per group for (A) flexural strength, (B) flexural modulus, and (C) fracture toughness, n = 100 per group for (D) nanoindentation-elastic modulus and (E) nano-hardness. Groups with the same letter per column are not significantly d t ( a s fl s u t a n t t m ifferent (p > 0.05). he groups of 750–840 ◦C (60.9 ± 6.46 GPa) and 820–840 ◦C (H14) 57.6 ± 2.28 GPa). See Fig. 2B. For groups that were intentionlly not heated above 780 ◦C, they displayed no statistically ignificant differences from each other. But, interestingly, the exural moduli of the 590–750 and 750–780 ◦C groups are not ignificantly different from the 820–840 ◦C (H14) flexural modlus. Fig. 2C presents the changes in glass–ceramic’s ability o resist fracture as the heat-treatment temperatures were ltered. Statistically, groups with the same letter are not sig- Please cite this article in press as: Lien W, et al. Microstructural evolution an (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 ificantly different than each other, and a gradual shift from he letter, “a”, to the letter, “e”, demonstrating that fracture oughness could be significantly improved via heat treat- ent. Furthermore, the 820–840 ◦C (H14) (4.07 ± 0.733 GPa) and 820–840 ◦C (3.55 ± 0.572 GPa) groups exhibited significantly higher fracture toughness than all other groups, while the 590–750 (H14), 750–780, and 750–840 ◦C groups were not significantly different from one another in terms of their ability to resist fractures. Similarly, groups like 590–750 ◦C (H14) and 820–840 ◦C (H14) that were held at the second targeted temperature for a prolonged period of 14 min portrayed similar fracture resistance as those groups without the extra 14 min of heat treatment, specifically the 590–750 and 820–840 ◦C group, respectively. d physical behavior of a lithium disilicate glass–ceramic. Dent Mater
3.2.2. Nanoindentation—elastic modulus
Fig. 2D shows how elastic modulus could be tailored via various two-stage heating schedules. The two best temperature ARTICLE IN PRESSDENTAL-2563; No. of Pages 13 8 d e n t a l m a t e r i a l s x x x ( 2 0 1 5 ) xxx–xxx intervals for achieving the two highest elastic modulus values were the 750–840 ◦C and 820–840 ◦C (H14) groups, having the values of 99.0 ± 1.29 GPa and 98.9 ± 2.82 GPa, respectively. Even though these two groups were not statistically different than each other, they performed significantly better than all other groups, including the recommended group whose elastic modulus was ranked the next highest in comparison with all groups. For those groups that were deprived of heating above 780 ◦C, the 750–780 ◦C group exhibited a significantly lower elastic modulus (75.9 ± 6.99 GPa) than all groups. Interestingly, the one trait that the 750–780 and 530–590 ◦C groups have in common was their large standard deviations in comparison with the other groups. This variability in elastic moduli depicted that the microstructures of the 750–780 and 530–590 ◦C groups could be composed of heterogeneous phases rather than a homogeneous distribution of a single crystalline phase. Furthermore, these two temperature intervals could be considered as critical transitions in the overall development of lithium disilicate crystals.
3.2.3. Nanoindentation—hardness
Fig. 2E discloses the relationship between nanoindentation hardness and various two-stage heating schedules. Unlike Fig. 2A and C, where a generalized upward trend exists, the surface hardness for the lithium disilicate glass–ceramics demonstrates a relatively downward trend as the ranges of the firing temperatures at the second heating stage were incrementally elevated from room temperature to the 820–840 ◦C range. Fig. 2E also shows that the 590–750 ◦C group exhibited a higher surface hardness than all the other tested groups, while the surface hardness for the specimens in the 750–780 ◦C group was the lowest when compared to all the other groups. In addition, the 750–780 ◦C group has the largest standard deviation in comparison to all other groups, and this large variability is closely related to the ratio of the nano-scale radius of the indenter tip to that of the micro-scale size of the structure onto which the indents were imprinted. To clarify, because the nominal size of our Berkovich tip radius (approximately 20 nm) is much smaller than the characteristic sizes (approximately in the micrometer scale) of the heterogeneous structures (i.e., lithium metasilicates, lithium disilicates, cristobalites, etc.; see Fig. 1 and Table 2 for the various chemical phases), the nano-scale indent could be directly imprinting on top of a crystal or a partially crystal- Please cite this article in press as: Lien W, et al. Microstructural evolution an (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 lized precursor, be contacting in between crystals (possibly in glassy components), or be interacting on the boundary shared by two different phases. Thus, in principle, the probability for an indenter tip to encounter one or the other phases depends on the distribution and surface fraction of various phases that were uniquely evolved from the partially crystallized precursor at a specified thermal gradient and were randomly scattered on the indentation surface. Furthermore, due to the disparity between the indenter tip radius and the characteristic sizes of the heterogeneous structures, the nanoindentation behavior (Fig. 2D) differs from the flexural strength and fracture toughness performances (Fig. 2A and C).
3.3. Microstructural evolution
For this study, the strengthening of the glass–ceramic physical properties corresponded to the appearance and disappearance of lithium disilicate and lithium metasilicate crystals respectively. After HF etching and in the absence of the surrounding glassy continuum, the SEM micrographs (Fig. 3A–H) identified three major microstructures: (1) the porous and finely knitted mesh of lithium metasilicates existed below the 590 ◦C thermal range; (2) the ovoid- and spherical-like configurations of Li2SiO3 and Li3PO4 emerged within the thermal range of 590–780 ◦C; and (3) the irregularly rod-shaped or oblate-like crystals of lithium disilicates appeared above the 780 ◦C thermal range. For the two groups treated with the thermal ranges below 590 ◦C, both exhibited similar, less dense, mesh-like microstructures, in which lithium disilicate precipitates were not seen (Fig. 3A and B). For the thermal ranges between 590 and 780 ◦C, the sphericallike morphologies of the 590–750 and 750–780 ◦C groups appeared to be larger in size and more maturely grown than the 590–750 ◦C (H14) group. Even though the 590–750 (H14) and 750–780 ◦C groups were both held at the second targeted temperature for a prolonged period of 14 min, the 590–750 ◦C (H14) group acquired more of the knitted-mesh network, which could be possible remnants persisting from the thermal range below 590 ◦C, when compared with the 750–780 ◦C whose morphology was mostly spherical. However, the meshlike network of 590–750 ◦C (H14) group appeared to be less porous and much denser than the not-fired and 530–590 ◦C groups. Although the 590–750 ◦C (H14) and 750–780 ◦C groups had the two longest overall heating times, they received a thermal range below the minimum temperature threshold. This delivering of the insufficient thermal energy merely elicited a response of densification rather than crystallization (Fig. 3A–E). For the three groups treated with the thermal ◦ d physical behavior of a lithium disilicate glass–ceramic. Dent Mater ranges above 780 C, lithium disilicates were clearly observed as rod-like crystals (Fig. 3F–H), and their orientations were random, making the overall bulk properties behave in an isotropic manner. For these groups, the rod-shaped crystals ARTICLE IN PRESSDENTAL-2563; No. of Pages 13 d e n t a l m a t e r i a l s x x x ( 2 0 1 5 ) xxx–xxx 9 Fig. 3 – Representative SEM images for the (A) not-fired, (B) 530–590 ◦C, (C) 590–750 ◦C, (D) 590–750 ◦C (H14), (E) 750–780 ◦C, (F) 7 14) g n a o i p fl h 3 c F m i s a u L g t f e
50–840 ◦C, (G) 820–840 ◦C (recommended), (H) 820–840 ◦C (H
ot only interlocked with each other but also intertwined mongst the mesh-like, dendritic cavities, which were once ccupied by the glassy phase that were etched away for ncreasing SEM image contrast; and, these isotropic crystals layed a significant role in modifying the bulk properties like exural strength, fracture toughness, elastic modulus, and ardness of the material. .4. Non-isothermal kinetics for lithium disilicate rystallization or this study, the Kissinger approximation was used to odel the kinetics of lithium metasilicates and lithium dislicates [32–34]. The lines in Fig. 4A and B, attained from the tatistical linear regression, were the best fits between ln(ˇi/Tp) nd 1/Tp, whose slopes (E/R, no unit) yielded the vales of 45.8 for lithium metasilicate formation and 80.3 for i2O–2SiO2 crystals. These slopes were multiplied by the −1 −1 Please cite this article in press as: Lien W, et al. Microstructural evolution an (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 as constant (8.314 J K mol ) to obtain the effective activaion energies of Li2O–SiO2 (381 ± 8.20 kJ/mol) and Li2O–2SiO2 ormations (667 ± 29.0 kJ/mol), which were the minimum nergy barriers that must be overcome for nucleation and roups. crystallization of lithium metasilicates and lithium disilicates within an IPS e.max® CAD bloc to happen. Finally, past studies have shown that the release of the exothermic energies (peak area normalized against mass) was directly proportional to the number of nuclei and crystals that were formed [32–34]. Fig. 4C demonstrates that the exothermic energies released by a glass–ceramic processed through the two-stage heating method were significantly more than the single-stage heating process, which suggests that the recommended two-stage heating schedule significantly enhances lithium disilicate production more than a single-stage heating, whose schedule consists of no holding time and a heating rate that is less than or equal to 20 ◦C/min.
4. Discussion
4.1. Relationship between heating schedules,
d physical behavior of a lithium disilicate glass–ceramic. Dent Mater microstructures, and physical properties Within the limits of this study, we have found that the premature onset of the second heating rate at the targeted ARTICLE IN PRESSDENTAL-2563; No. of Pages 13 10 d e n t a l m a t e r i a l s x x x ( 2 0 1 5 ) xxx–xxx Fig. 4 – Non-isothermal crystallizing kinetics for (A) lithium metasilicates and (B) lithium disilicates. (C) Exothermic energies of single-stage vs. two-stage heating schedules. Groups with the same letter per column are not significantly different (p > 0.05). temperature of 750 ◦C rather than 820 ◦C did not have a statistically significant impact on the glass–ceramic’s flexural strength and fracture toughness. Given this evidence, we would be inclined to reject the first hypothesis, but our other outcomes such as the glass–ceramic’s flexural modulus, elastic modulus, and hardness were significantly altered by the perturbation provoked from our imposed heating condition. Similarly, for the 820–840 ◦C group versus the 820–840 ◦C (H14) group, our evidence suggested that protracting the holding time at the isothermal temperature, 840 ◦C, of the second heating stage did not have a statistically significant impact on the glass–ceramic’s flexural strength and fracture toughness, but we did find statistically significant differences in flexural modulus, elastic modulus, and hardness between those two groups. Therefore, we could neither fully reject nor accept our first and second hypotheses. However, we did show that the multi-component glass–ceramics were dependent on our imposed heating conditions whose thermal energies transcended into developing different mixtures of microstructural phases, which were further manifested into different macroscopic glass–ceramic solids that offered a combination of physical properties based on the benefits of those heterogeneous phases. For example, the wax and wane of each specific microstructure (finely knitted mesh, spherical-like intermediates, and irregularly oblate-like crystals) were respectively Please cite this article in press as: Lien W, et al. Microstructural evolution an (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 associated with the three successive thermal intervals (below 590 ◦C, between 590 and 780 ◦C, and above 780 ◦C). Amongst the microstructural phases, the three groups treated above 780 ◦C have the most distinctive microstructural separation of all other groups. Furthermore, the glass–ceramic continuum that develops after thermal processing exhibits a structural hierarchy featuring the macro-scale voids between the glass–crystal interfaces, the micro-scale shape and size of the crystals, and the nano-scale defects in the crystalline lattice. Because of this wide scale range, these structural configurations play a vital role in influencing the physical properties of a glass–ceramic [18,37]. As shown by our SEM images, the complexity of the spatial distribution of Li2O–2SiO2 crystals for groups above 780 ◦C significantly contributed to the enhanced strength, modulus, and fracture toughness of the CAD blocs. For groups below 780 ◦C, their weak physical properties were associated with the absence of the high Li2O–2SiO2 volume fraction. These results appeared to be in good agreement with the fracture theory proposed by Hasselman and Fulrath, which stated: the strength of a glass–ceramic with a high volume fraction of a continuous glassy matrix is only dependent on the volume fraction of its crystallinity (i.e., dispersed phase), but the strength of a glass–ceramic with a high crystalline volume fraction is a function of both the volume fraction and size of its crystalline phase [37]. Additionally, increasing the average distance between crystals dispersed in the matrix could have an impact on governing the average flaw size and on d physical behavior of a lithium disilicate glass–ceramic. Dent Mater how crack propagation could have been barricaded or possibly stopped to avoid crack bridging [37]. According to Mecholsky and Freiman, a glass–ceramic with a high volume fraction ARTICLE IN PRESSDENTAL-2563; No. of Pages 13 x x x o a w o d e w i t i s a c d l h p t i t 4 c i T fi m r t t w t n a p h c i d o T g a s t t f m o t 5 t n a p y a i o d e n t a l m a t e r i a l s f crystallinity not only could provide greater toughness but lso could incur more average flaw sizes than a glass–ceramic ith a low volume fraction of crystallinity [38]. Furthermore, ther factors like the crystalline size, shape, and its threeimensional spatial configuration with other crystallites could ither deter against fracture or incur flaws [1,39]. Although we ere able to easily distinguish the “crowded” distribution and sotropic orientation of the Li2O–2SiO2 crystals as opposed to he more porous and mesh-like network of lithium metasilcates, we were not able to attest as to how the different izes, shapes, and orientation of lithium disilicates and their verage dispersed distance between the crystalline particles ould influence the overall physical properties of the lithium isilicate glass–ceramics. However, as the volume fraction of ithium disilicates were increased through manipulation of eat treatments, our data (Fig. 2A and C) did reflect the basic rinciple that an increase in the volume fraction of crysallinity within a glassy matrix was usually accompanied by an ncrease of the glass–ceramic’s flexural strength and fracture oughness. .2. Glass–ceramic’s rystalline-density-saturation-gradient composition and ts hardness he relative downward trend of hardness as the ranges of the ring temperatures at the second heating stage were increentally elevated from room temperature to the 820–840 ◦C ange (see Fig. 2E) was atypical and could be explained by he process of nucleation and crystallization (i.e., devitrificaion). For example, the orientation and saturation of crystals ithin a glass–ceramic relied on the proximity between he nucleating sites and on the locations and numbers of ucleating agents, whose development could be induced t random or at the glass–ceramic’s center of mass or its eriphery and whose distribution might or might not be omogenous in bulk [3,40,41]. Therefore, quite possibly, a rystalline-density-saturation gradient, defined as the stratfication of different glassy–crystalline ratios at different epths or regions in a glass–ceramic, could have been develped across from the glass–ceramic’s center to its periphery. hus, each two-stage heating schedule could yield a unique lassy–crystalline ratio across the glass–ceramic’s surfaces nd throughout its outer-to-inner core, resulting in distinct urface-indentation-hardness values amongst the various wo-stage heat treatments and quite possibly resulting in disinct indentation-hardness values per depth (i.e., depth across rom the glass–ceramic’s periphery to its core or center of ass) and per two-stage heat treatment. Furthermore, based n our hardness, XRD, DSC, and SEM results, we hypothesized hat: at low heating temperature intervals (e.g., 530–590 and 90–750 ◦C), the transformation from lithium metasilicates o lithium disilicates was immature; the separation between ucleating-, crystallizing-, and glassy-phases was indistinct; nd, crystalline-density-saturation gradient through comositional segregation via epitaxial crystallization was not Please cite this article in press as: Lien W, et al. Microstructural evolution an (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 et apparent. Thereby, glass–ceramics that were processed t low heating temperature intervals should generate simlar saturations of un-evolved lithium metasilicates to that f unfired glass–ceramics. And, surface hardness remained ( 2 0 1 5 ) xxx–xxx 11 less altered for the not-fired, 530–590, and 590–750 ◦C groups than those groups treated above 780 ◦C because compositional segregation between lithium metasilicates, lithium disilicates, and residual glass was expected to be absent, and the distribution of lithium-metasilicate-to-glassy components throughout the glass–ceramic’s surface to its core were more homogenous versus heterogeneous for the not-fired, 530–590, and 590–750 ◦C specimens. While at high heating temperatures, due to thermodynamic influence on the diffusivity of chemical species, compositional segregation would be more evident. Because of this, we suspected that more condensations of lithium disilicate crystals were produced at the glass–ceramic center of mass than at its surfaces, which resulted in more residual glass than crystals being present on the glass–ceramic surfaces, thereby yielding low hardness value.
4.3. Comparison with past studies
Similar to past studies [3,24,26], our SEM indicated that the crystallizing scheme of the IPS e.max® CAD began with the condensation and growth of nuclei within the glassy matrix. When the temperature was gradually raised from 530 to 590 ◦C, the lithium metasilicate continued to be the dominant phase with no new type of crystalline precipitate. Other studies have reported the presence of a mixture of lithium metasilicate and lithium orthophosphate at a temperature range of 500–560 ◦C, where the precipitations of Li3PO4 acted as the first nano-particles or sites for crystallization prior to the manifestation of lithium disilicates [3,10]. For our case, the XRD patterns showed that Li3PO4 was already incorporated into the not-fired glass–ceramic blocs, but interestingly it disappeared when treating with the heating schedules of 530–590, 590–750, and 590–750 ◦C (H14). On the contrary, when heating was elevated to and beyond 750 ◦C, only then Li3PO4 precipitates reappeared and remained as a residual phase in the three glass–ceramic groups, 750–840, 820–840, and 820–840 ◦C (H14), that had the highest treated thermal ranges. One possible explanation to the occurrence of Li3PO4 at dissimilar thermal settings was owing to the difference in stoichiometric and elemental compositions between IPS e.max® CAD and the earlier glass–ceramics. Only specific stoichiometric compositions of alkali- and alkaline earth-metal silicate crystals were considered as suitable formulations for designing a glass–ceramic system with crystalline assemblage [8,42]. An alternative reason was that at the intermediate temperature intervals (530–750 ◦C), the Li3PO4 structures began reorganization, forming amorphous nano-size particles, and consequently escaped the XRD detection [43]. When the temperature surged beyond 780 ◦C, both our SEM images and XRD patterns revealed that the growth of lithium disilicates was abrupt, and this phenomenon was accompanied by the presence of lithium orthophosphates, possibly started as intermediates and ended as residual remnants. As mentioned in the earlier section, Headley and Loehman have shown that the success of lithium disilicate crystalline growth was powerfully influenced by d physical behavior of a lithium disilicate glass–ceramic. Dent Mater their ability to epitaxy on the Li3PO4 nuclei, whose assemblage was built by amassing with the agglomeration of nucleating agents like P2O5, TiO2, and ZrO2 along with the appropriate heating condition [10]. We could only suspect that Li3PO4 could ARTICLE IN PRESSDENTAL-2563; No. of Pages 13 s x x r 12 d e n t a l m a t e r i a l have played a role in the development of lithium disilicates. Finally, stable lithium disilicate assemblage was observed over the 750–840 ◦C range while lithium metasilicates disintegrated. See Table 2 for the possible events when IPS e.max® CAD was heat-treated. This activity was in accordance with earlier findings [3]. Generally, relatively quick growth of lithium disilicates occurs only when the maximum formation of lithium metasilicates has ended [3]. Contrariwise, less thermal energy was needed in favor of growing Li2SiO3 crystals, while an ample amount of thermal energy was compulsory to surpass the steep activation energy of the glassy-to-crystalline reaction so Li2O–2SiO2 growths could occur. As shown by our data, only temperatures exceeding beyond 780 ◦C could induce growth of Li2O–2SiO2 crystals, while formation of Li2SiO3 crystals necessitated less thermal energy, approximately in the temperature range of 530–750 ◦C. Furthermore, similar to past studies [7,11–13], our non-isothermal data have demonstrated that a single-stage heating schedule, having no holding time, in comparison to the recommended two-stage heating schedule could often retard the growth and maturation of lithium disilicates (Fig. 4C). However, the quest for finding an optimal heating schedule, whether it is single-, double-, or multiple-heating schedule, remains to be explored. For this study, the key challenge was to identify appropriate thermal gradients that could predict the microstructural changes of the IPS e.max® CAD blocs. Alternative heating schedules that involved different combinations of thermal gradients, curtailed-or-prolonged heating rates, and temperature holding times could have been evaluated to further understand the thermal responses of our materials. However, the heating schedule selection decisions should ideally have clinical performance in mind so that the optimal heating schedule would result in a final product that would offer the best survival probability for our glass–ceramic prosthesis. Also, we have not addressed the effect of restrictions other than temperature such as pressure and concentrations of the constituent components, which could also impose microstructural alterations. Another limitation has to do with the reliability of our laboratory measurements. In this study, we have chosen fracture toughness, flexural strength, and elastic modulus as reliable parameters because they have been commonly known as good clinical predictors from past literature even though there was no proven association, at least in clinical dentistry, between these parameters and their clinical outcomes [44,45].
5. Conclusions
The heat treatments carried out in this study fell into three categories, temperature ranges below 590 ◦C, between 590 and 780 ◦C, and above 780 ◦C. Consequently, from these three thermal categories, three major microstructures were identified: the finely knitted mesh [Li2SiO3] predominated below 590 ◦C; the spherical-like intermediates [Li2SiO3, Li2O–2SiO2, and Li3PO4] emerged between 590 and 780 ◦C; and, irregularly ◦ Please cite this article in press as: Lien W, et al. Microstructural evolution an (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 oblate-like crystals [Li2O–2SiO2] arose above 780 C. See Table 2 for the possible evolutionary process of the IPS e.max® CAD. At each of these three evolutionary stages, a glass–ceramic that was formed through controlled devitrification via distinctive x ( 2 0 1 5 ) xxx–xxx heating schedules often yielded a principal microstructure that possessed interesting, sometimes peculiar, combinations of glassy–crystalline properties. Additionally, the growth of Li2O–2SiO2 crystals within the IPS e.max® CAD blocs was independent of the overall heating time but dependent on a minimum temperature threshold (780 ◦C). Groups heated above the minimum temperature threshold (780 ◦C) exhibited enhanced flexural strength, fracture toughness, and elastic modulus than groups intentionally not heated above 780 ◦C. Groups with the extended temperature range (750–840 ◦C) and protracted holding time (820–840 ◦C H14) produced significantly higher elastic-modulus and hardness properties than the control group but showed similar flexural-strength and fracture-toughness properties with the control group. Finally, the effective activation energy of crystallization calculated from the non-isothermal measurements for the IPS e.max® CAD blocs was 667 ± 29.0 kJ/mol. Disclosure The views expressed in this study are those of the authors and do not reflect the official policy of the United States Air Force, the Department of Defense, or the United States Government. The authors do not have any financial interest in the companies whose materials are discussed in this article.
Acknowledgements
The authors would like to thank Dr. Angela Campbell and Dr. Gregory Ehlert at the United States Air Force Research Laboratory and Dr. Tao You and Dr. Todd Lincoln at the United States Army Institute of Surgical Research for their help and support. Also, special thanks goes to Dr. Shashikant Singhal at Ivoclar Vivadent (Schann, Liechtenstein) for supplying the IPS e.max® CAD blocs. This research was partially funded by the Dental Master’s Thesis Award Grant from the Delta Dental Foundation. e f e r e n c e s [1] Guazzato M, Albakry M, Ringer SP, Swain MV. Strength, fracture toughness and microstructure of a selection of all-ceramic materials Part I. Pressable and alumina glass-infiltrated ceramics. Dent Mater 2004;20:441–8. [2] Stookey SD. Catalyzed crystallization of glass in theory and practice. Ind Eng Chem 1959;51:805–8. [3] Höland W, Apel E, van ‘t Hoen C, Rheinberger V. Studies of crystal phase formations in high-strength lithium disilicate glass–ceramics. J Non-Crystal Solids 2006;352:4041–50. [4] Höland W, Beall G. Glass-ceramic technology. Hoboken, NJ: Wiley: American Ceramic Society; 2012. [5] Höland W, Frank M, Rheinberger V. Surface crystallization of leucite in glasses. J Non-Crystal Solids 1995;180:292–307. [6] Höland W, Rheinberger V, Apel E, van’t Hoen C. Principles and phenomena of bioengineering with glass-ceramics for d physical behavior of a lithium disilicate glass–ceramic. Dent Mater dental restoration. J Eur Ceram Soc 2007;27:1521–6. [7] Zheng X, Wen G, Song L, Huang XX. Effects of P2O5 and heat treatment on crystallization and microstructure in lithium disilicate glass ceramics. Acta Mater 2008;56:549–58. ARTICLE IN PRESSDENTAL-2563; No. of Pages 13 x x x 102–11. d e n t a l m a t e r i a l s [8] Beall G. Design and properties of glass-ceramics. Annu Rev Mater Sci 1992;22:91–119. [9] Xiao Z, Zhou J, Wang Y, Luo M. Microstructure and properties of Li2O-Al2O3-SiO2-P2O5 glass–ceramics. Open Mater Sci J 2011;5:45–50. [10] Headley TJ, Loehman RE. Crystallization of a glass–ceramic by epitaxial growth. J Am Ceram Soc 1984;67:620–5. [11] Borom MP, Turkalo AM, Doremus RH. Strength and microstructure in lithium disilicate glass-ceramics. J Am Ceram Soc 1975;58:385–91. [12] Bischoff C, Eckert H, Apel E, Rheinberger VM, Holand W. Phase evolution in lithium disilicate glass-ceramics based on non-stoichiometric compositions of a multi-component system: structural studies by 29Si single and double resonance solid state NMR. Phys Chem Chem Phys 2011;13:4540–51. [13] Wen G, Zheng X, Song L. Effects of P2O5 and sintering temperature on microstructure and mechanical properties of lithium disilicate glass-ceramics. Acta Mater 2007;55:3583–91. [14] Scientific Documentation IPS e.max® CAD. In: Ivoclar Vivadent AG RD, editor. FL-9494 Schaan, Liechtenstein; 2005. [15] Albakry M, Guazzato M, Swain MV. Biaxial flexural strength, elastic moduli, and X-ray diffraction characterization of three pressable all-ceramic materials. J Prosthetic Dent 2003;89:374–80. [16] Albakry M, Guazzato M, Swain MV. Fracture toughness and hardness evaluation of three pressable all-ceramic dental materials. J Dent 2003;31:181–8. [17] Gao J, Chen J-h Wang F, Deng Z-x Li F, Wu D. Effect of heat-pressing on the microstructure and properties of a novel lithium disilicate glass-ceramic. Adv Mater Res 2011;177:441–6. [18] Chen X, Chadwick TC, Wilson RM, Hill RG, Cattell MJ. Crystallization and flexural strength optimization of fine-grained leucite glass-ceramics for dentistry. Dental Mater 2011;27:1153–61. [19] Denry IL, Mackert JR, Holloway JA, Rosenstiel SF. Effect of cubic leucite stabilization on the flexural strength of feldspathic dental porcelain. J Dental Res 1996;75:1928–35. [20] Tinschert J, Natt G, Mautsch W, Augthun M, Spiekermann H. Fracture resistance of lithium disilicate-, alumina-, and zirconia-based three-unit fixed partial dentures: a laboratory study. Int J Prosthodontics 2001;14:231–8. [21] Burgner LL, Lucas P, Weinberg MC, Soares PC, Zanotto ED. On the persistence of metastable crystal phases in lithium disilicate glass. J Non-Crystal Solids 2000;274:188–94. [22] Deubener J, Bruckner R, Sternitzke M. Induction time analysis of nucleation and crystal growth in di- and metasilicate glasses. J Non-Crystal Solids 1993;163:1–12. [23] Iqbal Y, Lee WE, Holland D, James PF. Metastable phase formation in the early stage crystallisation of lithium disilicate glass. J Non-Crystal Solids 1998;224:1–16. [24] Soares PC, Zanotto ED, Fokin VM, Jain H. TEM and XRD study of early crystallization of lithium disilicate glasses. J Non-Crystal Solids 2003;331:217–27. [25] Zanotto ED. Metastable phases in lithium disilicate glasses. J Non-Crystal Solids 1997;219:42–8. Please cite this article in press as: Lien W, et al. Microstructural evolution an (2015), http://dx.doi.org/10.1016/j.dental.2015.05.003 [26] Huang SF, Zhang B, Huang ZH, Gao W, Cao P. Crystalline phase formation, microstructure and mechanical properties of a lithium disilicate glass-ceramic. J Mater Sci 2013;48:251–7. ( 2 0 1 5 ) xxx–xxx 13 [27] ISO 6872: 2008(E) dentistry—ceramic materials, 3rd Ed. Geneva. Switzerland International Organization for Standardization; 2008. [28] Oliver WC, Pharr GM. Improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. J Mater Res 1992;7:1564–83. [29] Schmelzer JWP, Potapov OV, Fokin VM, Muller R, Reinsch S. The effect of elastic stress and relaxation on crystal nucleation in lithium disilicate glass. J Non-Crystal Solids 2004;333:150–60. [30] Albakry M, Guazzato M, Swain MV. Biaxial flexural strength and microstructure changes of two recycled pressable glass ceramics. J Prosthodontics 2004;13:141–9. [31] Della Bona A, Mecholsky Jr JJ, Anusavice KJ. Fracture behavior of lithia disilicate- and leucite-based ceramics. Dental Mater 2004;20:956–62. [32] Kissinger HE. Reaction kinetics in differential thermal analysis. Anal Chem 1957;29:1702–6. [33] Ozawa T. Kinetics of non-isothermal crystallization. Polymer 1971;12:150–8. [34] Matusita K, Sakka S. Kinetic study on non-isothermal crystallization of glass by thermal analysis. Bull Inst Chem Res, Kyoto Univ 1981;59:159–71. [35] Xingzhong G, Wenyan L, Hui Y, Jiajie Z, Wenda Z. Effect of neodymium on the crystallization, microstructure and colorization of Li2O-Al2O3-SiO2 glass ceramics. New J Glass Ceram 2012;2:98–103. [36] Spurr RA, Myers H. Quantitative analysis of anatase-rutile mixtures with an X-ray diffractometer. Anal Chem 1957;29:760–2. [37] Hasselman DPH, Fulrath RM. Proposed fracture theory of a dispersion-strengthened glass matrix. J Am Ceram Soc 1966;49:68–72. [38] Mecholsky JJ, Freiman SW. Fracture surface analysis of glass ceramics. In: Proceedings XIth International Congress on Glass. 1977. [39] Bennison SJ, Lawn BR. Role of interfacial grain-bridging sliding friction in the crack-resistance and strength properties of nontransforming ceramics. Acta Metall 1989;37:2659–71. [40] El-Meliegy E, Noort R. Lithium disilicate glass ceramics. glasses and glass ceramics for medical applications. New York: Springer; 2012. p. 209–18. [41] Iqbal Y, Lee WE, Holland D, James PF. Crystal nucleation in P2O5-doped lithium disilicate glasses. J Mater Sci 1999;34:4399–411. [42] Pinckney LR, Beall GH. Microstructural evolution in some silicate glass-ceramics: a review. J Am Ceram Soc 2008;91:773–9. [43] Huang S, Cao P, Li Y, Huang Z, Gao W. Nucleation and crystallization kinetics of a multicomponent lithium disilicate glass by in situ and real-time synchrotron X-ray diffraction. Cryst Growth Des 2013;13:4031–8. [44] Anusavice KJ. Standardizing failure, success, and survival decisions in clinical studies of ceramic and metal-ceramic fixed dental prostheses. Dental Mater 2012;28: d physical behavior of a lithium disilicate glass–ceramic. Dent Mater [45] Kelly J. Clinically relevant approach to failure testing of all-ceramic restorations. J Prosthetic Dent 1999;81: 652–61.
Article Images (0)
×![]()
Bioz is the world’s most comprehensive AI search engine for scientific experimentation. The Bioz search engine offers researchers billions of data-driven product, technique, and protocol recommendations. Bioz taps into the latest advances in Artificial Intelligence (AI) – including Natural Language Processing (NLP), Machine Learning (ML) and Generative AI to mine and structure hundreds of millions of pages of complex and unstructured scientific papers. By harnessing the power of AI, Bioz provides researchers with an unprecedented amount of summarized scientific experimentation knowledge right at their fingertips, ultimately helping to speed up drug discovery and the rate of success in finding cures for diseases. Bioz is used by over 15 Million researchers from over 17,000 different universities and companies in 196 countries.
