- Email: [email protected]
Pressed ceramics onto zirconia. Part 2: Indentation fracture and influence of cooling rate on residual stresses
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
Available online at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/dema
Pressed ceramics onto zirconia. Part 2: Indentation fracture and influence of cooling rate on residual stresses Jung Eun Choi, J. Neil Waddell, Michael V. Swain ∗ Sir John Walsh Research Institute, Faculty of Dentistry, University of Otago, Dunedin, New Zealand
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objectives. The aim of this study was to evaluate the fracture toughness and surface residual
Received 13 October 2010
stresses present in various pressable ceramics to zirconia resulting from cooling induced
Received in revised form
temperature gradients.
14 June 2011
Materials and methods. Indentation fracture toughness was used to evaluate the residual
Accepted 17 August 2011
stress present in various pressable ceramics (Noritake CZR Press, Vita PM9, Wieland PressXzr and IPS e.max ZirPress) to zirconia when subjected to different cooling regimen. The cooling responses of two ceramics were evaluated by thermocouples embedded in the surface of
Keywords:
the porcelains and at the porcelain–zirconia interface.
Zirconia
Results. The effective Kc results obtained by indentation tests confirmed the presence of
All-ceramics
surface residual compressive stress for all-ceramic systems subjected to different cooling
Pressed ceramics
procedures. The residual stresses were quantified from the change in the radial crack size
Fracture toughness
and the values compared for each ceramic before pressing, pressed ceramic only and pressed
Residual stress
ceramic veneered on zirconia, from fast to slow cooling rates. A significant level of residual
Cooling curves
stress was found in the materials before pressing. Slow cooling significantly reduced the formation of residual stress for all pressed ceramics. From data produced by the thermocouples it was found that ‘slow cooling’ generated the least temperature difference between inner and outer surfaces of porcelain. A direct relationship was found for the cooling induced temperature difference between the surfaces, and interface thermocouples, and magnitude of the surface residual stresses. Significance. Leucite containing porcelains have higher intrinsic fracture toughness, and for all porcelains fast cooling generated significant residual stress within the veneering porcelain. To reduce development of residual stress, slow cool is recommended on the last heating cycle (e.g. glazing cycle). © 2011 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Patients’ demands for esthetic restorations have led to an increase in the use of all-ceramic restorations instead of porcelain-fused-to-metal (PFM), full metal crowns and resin composite restorations. This demand has led to application
∗
of glass-ceramics for a range of dental restorations. Along with development of glass ceramics, the heat pressing technique for the veneering of all-ceramic core materials is now widely used in dentistry as it offers esthetic outcomes and strength simultaneously [1]. In paper part 1 the strength and adhesion of four pressed ceramic materials for zirconia were investigated. Despite the growing usage of veneering porcelain
Corresponding author. Tel.: +64 3 479 4196; fax: +64 3 479 5079. E-mail addresses: [email protected], [email protected] (M.V. Swain). 0109-5641/$ – see front matter © 2011 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dental.2011.08.003
1112
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
to zirconia; clinical studies have reported veneering failure; namely chipping as an issue [2,3]. Moreover, research has shown that differences in failure rate not only exist between the hand veneering and the pressed veneering ceramics, but also between the different pressed ceramic materials [4,5]. Possible reasons postulated for chipping are: insufficient bond strength, excessive tensile stress due to a coefficient of thermal expansion mismatch with an un-equal thermal contraction between veneer layer and coping, and excessive load due to premature contacts, tensile thermal tempering residual stress introduced during cooling [6,7]. Ceramic materials used for the fabrication of all-ceramic restorations are sintered and glazed several hundred degrees above the glass transition temperature (Tg ). At the end of the firing process the dental technician removes the ceramic restoration from the furnace and it is cooled. Once the porcelain is removed from the furnace, its outer surface starts contracting and becomes rigid as it loses heat rapidly, while the interior part is still hot and somewhat viscous. The tensile stresses which would normally develop in an elastic body during such rapid cooling are relaxed while the interior of the porcelain is above the Tg , or more specifically above the softening temperature. However, temperature gradient results in the formation of surface compressive stresses and compensating internal tensile stresses, so called tempering stresses in the body. An indication of the magnitude of such stresses has been addressed by Asaoka et al. [8] and Swain [6] for the case of a bilayer material system. The tensile stresses associated with tempering may cause immediate cracking of the porcelain upon deformation of the restoration, and can increase the probability of fracture during functional loading of the restoration, manifested clinically as chipping of the veneering porcelain [9]. One problem associated with ceramic restoration failure is therefore the development and magnitude of residual stress within the structure. The residual stress developed may also influence the measured nominal fracture toughness of ceramic materials. Marshall and Lawn developed a simple approach using indentation fracture about a Vickers indenter to quantify the magnitude of these residual stresses [10]. Fracture toughness or critical stress intensity (K1c ) indicates the ability of a material to resist rapid crack propagation and is an indicator of clinical reliability and serviceability of a ceramic restoration. The present study thus has three primary objectives, namely; quantifying the fracture toughness of the four pressable veneering ceramic materials; determining the residual stress present in various pressable ceramics bonded to zirconia using the indentation fracture toughness technique and measuring the cooling curves when the ceramics are subjected to different cooling procedures.
2.1.
Vita In-Ceram YZ zirconia and the pressed ceramics were prepared to the dimension required for each test following the manufacturers’ instructions as listed in paper 1. Three groups of specimens were prepared (1) Bilayered group (1-1) 2 mm zirconia and 2 mm pressed porcelain (1-2) 2 mm zirconia and 4 mm pressed porcelain (2) One layer group (2-1) 2 mm of pressed porcelain ceramic only (2-2) 2 mm slice from each porcelain ingot before pressing The bilayered specimens were divided into 3 groups. Before glazing, the pressed ceramic surface of the specimens were all ground and polished to the desired dimension for each test, using 120–4000 grades of abrasive paper (Struers PSA backed Silicon carbide paper) on a metallographic lapping machine (Knuth Rother, Struers, Denmark) for the better observation of the cracks. For group (1) the specimens were self-glazed at the temperature recommended in the manufacturer’s instructions and then removed from the furnace as soon as the support plinth was low enough to remove the samples. The specimens were then force cooled with compressed air (2 bar) from two opposing sources 30 cm apart. Specimens for group (2) were glaze fired and cooled to mimic what dental technicians normally do when building crowns. The specimens were removed from the furnace when the temperature of the furnace drops to the starting temperature of the glazing cycle. Specimens for group (3) were glaze fired and slow cooled by stopping the firing plinth descending from the furnace and waiting until the temperature reduces to 100 ◦ C. The latter took approximately 30–40 min depending on the firing cycle used.
2.2.
Materials and methods
Materials used are presented in Table 1 in paper part 1.
Indentation tests
Two groups of specimens were used: (i) 2 mm pressed ceramic only and (ii) a 2 mm slice from an ingot before pressing. Specimens for group (i) were glaze fired and slowly cooled in the furnace to 100 ◦ C to measure the intrinsic fracture toughness without any tempering induced residual stress present. Indentation cracks on the veneer surface of all specimens were made with a Vickers indenter at a load of 25 N for samples with ceramic only and 35 N for the samples before pressing. Different loads were chosen after making preliminary indentations with loads ranging from 20 to 70 N and the loads mentioned above showed the most suitable crack patterns in terms of size and absence of chipping. Between 5 and 12 indents were made depending on the porosity of the specimen. Indentations were made approximately 3–4 mm apart and cracks were measured directly post indentation. The following equation was used to calculate the effective fracture toughness [9]: Kc =
2.
Specimen preparation
E 0.5 P H
c3/2
(1)
where H is the hardness, P the load, E the elastic modulus, c the length of radial crack, and is a constant 0.016.
1113
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
Table 1 – Summary of the materials used in the study with their coefficient of thermal expansion and glass transition temperatures. Materials used
CTE (ppm/K)
Tg (◦ C)
Vita Zahnfabrik Ivoclar Vivadent Wieland Dental + Technik Noritake Kizai Co.
9.2 9.75 9.3 10.1
640 530 620 615
Vita Zahnfabrik
10.5
Manufacturer
Pressed ceramics Vita PM9 IPS e.max Zirpress Wieland Xzr Noritake CZR Zirconia Vita In-Ceram YZ
Indentation cracks were introduced within the veneer surface of all specimens using a Vickers indenter at a load of 50 N for the bi-layered specimens. The high polish achieved on the sample surface minimised the inaccuracies associated with crack terminations due to surface scratches and porosity. An optical microscope with a high resolution digital stage (Nikon, Japan) was used to measure the length of radial cracks and to microscopically inspect the indentation surface of the specimens.
2.3.
Residual stress determination
The presence of residual stresses is very apparent by the extent of radial crack extension about the residual impression. Marshall and Lawn developed a simple analysis of this problem [10]. The presence of residual or applied tensile or compressive stress in the surface of the material being indented introduces an additional term in the expression for the stress intensity factor as given by Eq. (1). For the situation where the indentation crack size is small compared with the gradient of stress, the expression for the stress intensity factor is composed of two terms: K1c = KInd + KApp/Res
(2)
where KInd is the indentation stress intensity factor given by Eq. (1) and KApp/Res is the stress intensity factor for a crack in the residual or applied stress field and is given by: KApp/Res
√ = YApp/Res c
(3)
where Y is the shape factor for a half-penny surface crack and App/Res is the applied or residual stress present. Combining Eqs. (1)–(3) we have: K1c =
E 1/2 P H
c3/2
√ ± YApp/Res c
(4)
The second term in Eq. (4) may be positive or negative depending upon whether the stresses on the surface are tensile or compressive. The above approach was used to determine the tempering residual stresses developed in veneering porcelains as a consequence of rapid cooling [8,10].
2.4.
heating and cooling of the specimens. Two groups of samples were used: (1) 2 mm zirconia and 2 mm porcelain; (2) 2 mm zirconia and 4 mm porcelain. To illustrate the cooling response curves, two ceramic materials were chosen among the four; Noritake CZR Press and IPS e.max ZirPress to observe the difference between leucite containing and non-leucite containing glass ceramic materials and their respective glaze firing cycles. A small hole was made into the specimens to the interface in the middle of the sample with a high speed bur. The thermocouple wires were then inserted and covered by porcelain to hold the wires in place. Another thermocouple wire was placed on the surface of the sample and held in place with porcelain. The specimens with thermocouples attached were then taken through the final glaze firing cycle for each bi-layer system. The thermocouple wires were connected to a digital data logging system (PowerLab, Sydney, Australia). The specimens were cooled precisely as done for samples used for indentation fracture toughness determination. During this firing and cooling process, the data logging system recorded the inner (interface) and outer (surface) temperature of the veneering porcelain.
Thermocouple observations
Welded thermocouple wires (Type K) were placed on the surface and embedded at the porcelain–zirconia interface, allowing continuous measurement of temperature during
3.
Results
3.1.
Indentation fracture toughness test
The results obtained from indentation fracture toughness tests using Eq. (1) are presented in Fig. 3. The nominal fracture toughness of the four pressable ceramics in this study was in the range of 0.92–2.01 MPa m0.5 before pressing and 0.6–0.78 MPa m0.5 after pressing for ceramics not veneered to zirconia. The intrinsic fracture toughness of the materials is considered to be that for the 2 mm ceramic only specimens which were slowly cooled. It was found that Noritake CZR Press has the highest fracture toughness before and after pressing, and the leucite containing materials (Noritake CZR Press and Vita PM9) have higher values than non-leucite containing glass ceramics (Wieland PressXzr and IPS e.max ZirPress). The nominal fracture toughness Kc values obtained in this study were in the range of 0.8–1.74 MPa m0.5 (Fig. 3). When only the leucite containing ceramics are compared, Vita PM9 had a higher effective Kc than that of Noritake CZR when fast cooled. However, when medium and slow cooled, Noritake had a higher effective Kc than the rest of the materials. Vita PM9 and IPS e.max ZirPress had similar Kc values when they were slow cooled. Upon slow cooling, there was a slight increase in
1114
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
the effective fracture toughness values as the thickness of the veneering porcelain increases from 2 mm to 4 mm. Moreover, it was also found that the leucite containing (Vita PM9 and Noritake CZR Press) pressed ceramics generally have higher Kc in all situations than non-leucite containing glass pressed ceramics (Wieland PressXzr and IPS e.max ZirPress).
3.2.
The residual stresses
The residual stresses in the surface of the samples were determined based upon Eq. (4). The effective or intrinsic Kc values for the individual materials were chosen as the slowly cooled porcelain only 2 mm samples. With this value the residual stress results obtained from indentation fracture toughness are presented in Fig. 4. As indicated in Eq. (2), the values of effective fracture toughness are influenced by the tensile/compressive residual stresses present within the ceramic systems. The magnitude of the residual stresses present in the specimens under different cooling conditions followed the same trend as the effective fracture toughness values. For the samples before pressing; there was a large compressive stress present for all four materials. The residual stress present in the bilayered samples was directly related to the cooling rate, where fast cooling resulted in high stress levels and slow cooling reduced stress levels.
3.3.
Thermocouple determined cooling curves
The cooling curves obtained by the thermocouples for the three cooling conditions for both the Noritake and IPS e.max ZirPress materials are presented in Figs. 1a and 2a. The graphs showed a significant difference between cooling conditions for all materials, namely; the faster the cooling, the steeper the temperature gradient between the firing temperature and the Tg point (Table 1). Here the firing temperatures recommended by the manufacturers were used. For the Noritake CZR Press, the glaze temperature was 900 ◦ C while for IPS e.max ZirPress it was 770 ◦ C. The temperature difference between the interface and the outer surface of the porcelain during cooling is shown in Fig. 2a and b. The data in these figures have been fitted with a simple polynomial trend line of best fit and shows the Noritake CZR fired at a higher temperature, resulting in a higher temperature difference. The 4 mm bi-layered samples (not included) also showed the same trend however the temperature difference was greater than that of 2 mm bilayer samples.
4.
Discussion
The cooling curves show big differences between the three conditions evaluated (fast, normal and slow cooling) (Figs. 1 and 2). This is also clearly seen in the temperature differences between the inner and outer thermocouple readings (Fig. 2a and b). These observations are very similar to what Asaoka et al. observed when they measured the influence of cooling rate on porcelain and porcelain bonded to metal systems [8]. As shown in the cooling curves (Figs. 1 and 2), the trends of three different cooling conditions were similar for both
Fig. 1 – Cooling curves for (a) Noritake CZR Press (2 mm thickness) and (b) IPS e.max ZirPress (2 mm thickness) when subjected to slow, normal and fast cooling.
Noritake CZR Press and IPS e.max ZirPress. However, it is obvious that the cooling curves for the Noritake CZR Press are steeper than those of IPS e.max ZirPress, in addition to having a greater temperature difference between the inner and the outer surfaces of the material. This is anticipated from Newton’s cooling law; namely the rate of change of the temperature of an object is proportional to the difference between its current temperature and the ambient temperature. This rationalises the observations that for the same cooling heat transfer conditions the higher the firing temperature, the faster the cooling rates. Another observation was that as the thickness of the porcelain increased the temperature difference between the outer surface and the inner thermocouple increased as may be seen in Table 1. These observations are in accord with those of Asaoka et al. [8] and Swain [6]. Plots of the temperature difference for the 2 mm thick Noritake and IPS e.max samples as a function of the external temperature of the specimens for the three cooling conditions are shown in Fig. 2a and b. In the case of the Noritake CZR Press, the fast cooling condition resulted in a temperature difference greater than 80 ◦ C between the inner and outer thermocouple that persists over an extended cooling range from when the external temperature changes from 700 ◦ C to 250 ◦ C. Whereas the normal and slow cooling conditions resulted in similar shaped curves but the temperature differences were only 35 ◦ C and 10 ◦ C respectively. For IPS e.max ZirPress, the lower firing temperatures resulted in less temperature difference under the same nominal cooling conditions as the
1115
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
Table 3 – The average cooling rate between −50 ◦ C from the glaze temperature and the temperature difference when the external temperatures is at Tg .
Noritake CZR Press Tg = 615 (◦ C) 2 mm Slow cool Normal cool Fast cool 4 mm Slow cool Normal cool Fast cool IPS e.max ZirPress Tg = 530 (◦ C) 2 mm Slow cool Normal cool Fast cool 4 mm Slow cool Normal cool Fast cool
Fig. 2 – Difference between external and internal thermocouple measured temperature readings for the three cooling conditions shown in Fig. 1 for (a) Noritake and (b) IPS e.max.
Noritake samples. In this instance the temperature differences were 35 ◦ C for the fast, 18 ◦ C for the normal and 5 ◦ C for the slow cooling conditions. Table 1 contains the maximum temperature differences when the external temperature reaches Tg for Noritake and IPS e.max (Tables 2 and 3).
Cooling rate veneering porcelain (◦ C/s)
Temperature difference porcelain is at Tg (◦ C)
−1.1 −6.6 −16.6
10 31 85
−0.16 −5.8 −19.6
9 52 121
−0.02 −3.95 −10.29
7 28 35
−0.05 −2.8 −17.42
8 22 92
The effective indentation fracture toughness test results show that a wide range of Kc values were found for all the pressed veneering ceramics investigated (Fig. 3). The highest values were in two cases the samples cut from the “as received” ingots, namely for the Noritake and Weiland materials. However after disks of the ingot materials were refired to the pressing temperature and slowly cooled within the furnace, the effective Kc for all materials were the least. It is also clearly seen that with increasing the cooling rate and thickness of the porcelain, the effective Kc values also increased. Another feature evident was that in the case of slow cooling, irrespective of the thickness of the porcelain, or whether the porcelain was fused to zirconia or not, there appeared to be minimal difference in the effective Kc values measured for a specific material.
Table 2 – Pressing and glazing schedule for the pressed ceramics used in the study. Brand/name of product
(a) Pressing schedule Vita/PM9 Noritake/CZR Press IPS e.max ZirPress Wieland/PressXzr
Brand/name of product (b) Glazing schedule Vita/PM9 Noritake/CZR IPS e.max ZirPress Wieland/PressXzr
Heat up temp (◦ C)
Start temp (◦ C)
Heat rate (◦ C/min)
Vacuum hold time (min)
Pressing temp (◦ C)
Press time (min)
850 850 900 900
700 700 700 700
50 60 60 60
20 26 15 20
1000 1065 910 1060
6 6 6 8
Drying time (min)
Start temp (◦ C)
Heat rate (◦ C/min)
Firing temp (◦ C)
Hold time (min)
5 5 5 5
500 600 450 575
80 65 60 75
900 900 770 880
1 1 1 1
1116
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
Fig. 3 – Summary of the effective fracture toughness values of the four different pressed ceramics determined by the indentation fracture toughness technique.
Fig. 4 – Summary of surface residual stress values of the four different pressed ceramics determined by the indentation fracture toughness technique.
In this work we consider the values of Kc measured for the slow cooled and unbonded porcelain as the reference Kc values of the individual porcelains. This follows the approach of Marshall and Lawn [10] and industrial procedures namely that slow cooling rates are generally used to minimise the presence of residual stresses in glasses and visco-elastic materials such as porcelains, see for instance Kingery et al. [11] The values of the intrinsic Kc values for the four pressed porcelains range from 0.6 to 0.78 MPa m0.5 which is typical for porcelains. The leucite containing materials exhibited the highest Kc , whereas the primarily glass materials had the least. These results are in accord with a recent study by Cesar et al. [12] that showed an increasing trend of Kc values of veneering porcelains with increasing leucite content. From the indentation fracture toughness results, a significant trend was found; a decrease of nominal Kc values of all materials after pressing. While other materials Kc dropped by 5–15% due to pressing, Noritake dropped approximately 60% (2.01–0.78 MPa m0.5 ). According to the X-ray diffraction results from the companion paper [Choi et al.], Noritake had only a very minor change in leucite content in terms of volume fraction or crystal structure, which would not have been expected to result in the significantly lower Kc after pressing. Vita PM9 on the other hand, was expected to change significantly, as the X-ray diffraction results revealed that there were differences of leucite volume fraction and crystalline structure (from tetragonal to cubic) after pressing. However, Vita PM9 had the smallest change in Kc before and after pressing, 0.92–0.71 MPa m0.5 . This drop in Kc values after pressing may have been caused by the residual stress present in the material before pressing caused by the block manufacturing process. Despite having leucite as the main crystalline structure, the Kc values of Noritake and Vita materials had a slight difference after pressing. The Kc value of Noritake CZR Press was slightly higher than that of Vita PM9 after pressing for annealed monolithic samples. From the X-ray diffraction patterns obtained in part 1 of this study, it was found that the Noritake CZR Press has a higher volume fraction and retains the tetragonal leucite after pressing; whereas Vita PM9 has a
change in leucite structure after pressing from tetragonal to cubic as well as a reduction in volume fraction. Rasmussen et al. [13] and Denry et al. [14] showed that dental porcelains with cubic leucite at room temperature had significantly lower strength as well as fracture toughness than porcelains with tetragonal leucite. This may have also contributed to the reduction observed in this study. A similar decrease in Kc values after pressing was also found with Wieland PressXzr and IPS e.max ZirPress. Since these materials were found to be non-leucite glass ceramics it can be concluded that the drop in the Kc after pressing for these materials is mainly caused by the presence of the residual stress in the ingot before pressing due to the manufacturing process, such as rapidly cooling induced residual stresses developed. Utilizing the above estimates of the intrinsic Kc values for the different porcelains it is possible to estimate the residual stresses developed on the surface of the differently cooled porcelains from Eq. (4). The values of these stresses are plotted in Fig. 4. For most of the materials there appears to be a clear trend in the results, namely that with the faster cooling rates and thicker porcelain layer, the higher the surface residual compressive stresses. There are exceptions in that the Vita material appears to have higher residual stresses for the 2 mm thickness rather than the 4 mm when fast cooled and the values for normal cooling were very low, even lower than the slow cooled specimens. For the Noritake and IPS e.max materials where the temperature was measured during cooling for the 2 and 4 mm thick samples and the surface residual compressive stresses determined as shown in Fig. 4, it was possible to plot residual stress versus the temperature difference when the surface of the specimen was at Tg as shown in Fig. 5. A simple linear plot through these data points shows an excellent fit. This relationship is anticipated for viscous materials rapidly cooled from above Tg . As shown in the classic text by Kingery et al. when a viscous body experiences a temperature gradient during cooling the stresses that would nominally be developed if the material were elastic are able to be relaxed by viscous flow [11]. The ability for this relaxation decreases when the temperature
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
Fig. 5 – Graph of the magnitude of the surface compressive residual stresses versus the temperature difference between the inner and outer surface when the external surface reached Tg for the Noritake and IPS e.max materials.
of the object drops below the softening temperature Ts and completely below Tg . From the plots of the temperature difference between the inner and outer surfaces of the porcelain as shown in Fig. 2, it is evident that the temperature differences remains reasonably constant over the interval for the external and internal thermocouple positions, cooling well below the Tg temperature. This implies that the stress difference between the external and internal parts of the porcelain will reflect the temperature difference along with the E modulus and thermal expansion. According to Kingery et al. [11] the stress developed as a consequence of this temperature difference for a homogeneous material, is given by: =
E˛T 1−
in some instances had completely run through the porcelain, were found for fast cooled samples. No such change was observed for slow cooled samples. This is considered due to the subsurface component of indentation cracks in the rapid cooled samples being influenced by the tensile residual stresses present within the material compensating the compressive stresses developed on the surface. The cause of chipping of veneering ceramics on zirconia all-ceramic cores was reported to be multi-factorial [1–7]. Therefore the individual and the combined effects of such variables studied in the two parts of the study; the crystalline structure, bond strength, fracture toughness, flexural strength and the cooling behavior can influence the fracture behavior of pressed ceramics on to zirconia and therefore the clinical success of such restorations. The results from part 1 revealed two different types of pressed ceramics; leucite and glass (nonleucite) ceramics and how the presence of leucite influences the mechanical properties. The outcome of this paper leads to a better appreciation of the role of cooling on the residual stress formation in the different pressed ceramic materials. However, the similar trends found in cooling curve analysis and residual stress developed for both leucite and non-leucite types of materials places more emphasis on the cooling conditions regarding the residual stress formation along with the thickness of the porcelain; especially for zirconia bilayered pressed ceramics.
5.
Conclusions
Within the limitations of this study, the following conclusions were drawn:
(5)
where E is the elastic modulus, is Poisson’s ratio, ␣ is the coefficient of thermal expansion and T is the temperature difference. For a tempered object the internal tensile stress is half the magnitude of the external compressive stress. Thus the slope of line in Fig. 5 should be given by: 2E˛ = T 3(1 − )
1117
(6)
The value 2/3 arises from the fact that only the surface compressive stresses were determined using the indentation approach and are compared with the temperature difference. The slope from the data in Fig. 5 is approximately 0.4 MPa/◦ C whereas the predicted slope for a material with an E modulus of 70 GPa, 0.2 Poisson’s ratio and 9 × 10−6 /◦ C coefficient of thermal expansion is 0.5 MPa/◦ C. This simple approach does not include any relaxation as incorporated in more rigorous numerical models by DeHoff and Anusavice [15,16] and considers the temperature profile and resultant stresses in a bilayer system simply as for a homogeneous visco-elastic body. For some fast cooled samples, the indentation cracks formed were again observed 24 h after the indentation to see the effect of time delay between indenting and crack measurement. When the cracks were re-visited, longer cracks, which
• The intrinsic Kc results obtained from indentation showed a specific trend. For slow cooled samples, the leucite containing ceramics (Noritake and Vita) have higher fracture toughness values than non-leucite glass ceramics (Wieland and IPS e.max). • There was a significant presence of residual stress found in the material even before pressing which may have been caused by the ingot manufacturing process. • Fast cooled samples produced short cracks (higher effective Kc ) due to compressive residual stress formed on the surface to compensate the tensile stress formed inside the system. • Slow cooling almost completely eliminated formation of residual stress for all pressed ceramics.
Therefore the following general conclusions were drawn:
• Practitioners have choices of two types of ceramic materials when using pressing technique: Leucite containing and non-leucite glass ceramics. • To reduce the development of residual stress within the ceramic system, practitioners are advised to slow cool the restoration on the last heat treatment cycle (e.g. glazing cycle).
1118
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
Acknowledgment The authors wish to thank Dr Basil Al-Amleh for critically reading this paper.
references
[1] Gorman CM, Mcdevitt WE, Hill RG. Comparison of two heat-pressed all-ceramic dental materials. Dent Mater 2000;16:389–95. [2] Tsalouchou E, Cattel MJ, Knowles JC, Pittayachawan P, Mcdonald A. Fatigue and fracture properties of yttria partially stabilized zirconia crown systems. Dent Mater 2008;24:308–18. [3] Sailer I, Feher A, Filser F, Gauckler LJ, Luthy H, Hammerle CHF. Five-year clinical results of zirconia frameworks for posterior fixed partial. Int J Prosthodont 2007;20:383–8. [4] Christensen GJ. PFM vs. zirconia restorations—how are they comparing clinically? Clinician’s Report, vol. 1(11). CR Foundation; November 2008. [5] Taskonak B, Borges GA, Mecholsky J, Anusavice KJ, Moore KB, Yan J. The effects of viscoelastic parameters on residual stress development in a zirconia/glass bilayer dental ceramic. Dent Mater 2008;24:1149–55. [6] Swain MV. Unstable cracking (chipping) of veneering porcelain on all-ceramic dental crowns and fixed partial dentures. Acta Biomater 2009;5(5):1668–77.
[7] Fisher J, Stawarczyk B, Trottmann A, Hammerle CHF. Impact of thermal misfit on shear strength of veneering ceramic/zirconia composites. Dent Mater 2009;25(4):419–23. [8] Asaoka K, Kuwayama N, Tesk JA. Influence of tempering method on residual stress in dental porcelain. J Dent Res 1992;71:1623–7. [9] Isgro G, Wang H, Kleverlaan CJ, Feilzer A. The influence of multiple firing on thermal contraction of ceramic materials used for the fabrication of layered all-ceramic dental restorations. Dent Mater 2005;21:557–64. [10] Marshall DB, Lawn BR. Indentation technique for measuring stresses in tempered glass surfaces. J Am Ceram Soc 1977;60:86–7. [11] Kingery WD, Bowen HK, Uhlmann DR. Introduction to ceramics. 2nd ed. New York: John Wiley & Sons; 1975. [12] Cesar PF, Yoshimura HN, Miranda Jr WG, Miyazaki CL, Muta LM, Rodrigues Filho LE. Relationship between fracture toughness and flexural strength in dental porcelains. J Biomed Mater Res B Appl Biomater 2006;78:265–73. [13] Rasmussen ST, Groh CL, O’Brien WJ. Stress induced phase transformation of a cesium stabilized leucite porcelain and associated properties. Dent Mater 1998;14:202–11. [14] Denry IL, Mackert JR, Holloway JA, Rosentiel SF. Effect of cubic leucite stabilization on the flexural strength of feldspathic dental porcelain. J Dent Res 1996;75:1928–35. [15] DeHoff PH, Anusavice KJ. Viscoelastic stress analysis of thermally compatible and incompatible metal–ceramic systems. Dent Mater 1998;14:237–45. [16] Dehoff PH, Anusavice KJ. Tempering stresses in feldspathic porcelain. J Dent Res 1989;68:134–8.
Available online at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/dema
Pressed ceramics onto zirconia. Part 2: Indentation fracture and influence of cooling rate on residual stresses Jung Eun Choi, J. Neil Waddell, Michael V. Swain ∗ Sir John Walsh Research Institute, Faculty of Dentistry, University of Otago, Dunedin, New Zealand
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objectives. The aim of this study was to evaluate the fracture toughness and surface residual
Received 13 October 2010
stresses present in various pressable ceramics to zirconia resulting from cooling induced
Received in revised form
temperature gradients.
14 June 2011
Materials and methods. Indentation fracture toughness was used to evaluate the residual
Accepted 17 August 2011
stress present in various pressable ceramics (Noritake CZR Press, Vita PM9, Wieland PressXzr and IPS e.max ZirPress) to zirconia when subjected to different cooling regimen. The cooling responses of two ceramics were evaluated by thermocouples embedded in the surface of
Keywords:
the porcelains and at the porcelain–zirconia interface.
Zirconia
Results. The effective Kc results obtained by indentation tests confirmed the presence of
All-ceramics
surface residual compressive stress for all-ceramic systems subjected to different cooling
Pressed ceramics
procedures. The residual stresses were quantified from the change in the radial crack size
Fracture toughness
and the values compared for each ceramic before pressing, pressed ceramic only and pressed
Residual stress
ceramic veneered on zirconia, from fast to slow cooling rates. A significant level of residual
Cooling curves
stress was found in the materials before pressing. Slow cooling significantly reduced the formation of residual stress for all pressed ceramics. From data produced by the thermocouples it was found that ‘slow cooling’ generated the least temperature difference between inner and outer surfaces of porcelain. A direct relationship was found for the cooling induced temperature difference between the surfaces, and interface thermocouples, and magnitude of the surface residual stresses. Significance. Leucite containing porcelains have higher intrinsic fracture toughness, and for all porcelains fast cooling generated significant residual stress within the veneering porcelain. To reduce development of residual stress, slow cool is recommended on the last heating cycle (e.g. glazing cycle). © 2011 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Patients’ demands for esthetic restorations have led to an increase in the use of all-ceramic restorations instead of porcelain-fused-to-metal (PFM), full metal crowns and resin composite restorations. This demand has led to application
∗
of glass-ceramics for a range of dental restorations. Along with development of glass ceramics, the heat pressing technique for the veneering of all-ceramic core materials is now widely used in dentistry as it offers esthetic outcomes and strength simultaneously [1]. In paper part 1 the strength and adhesion of four pressed ceramic materials for zirconia were investigated. Despite the growing usage of veneering porcelain
Corresponding author. Tel.: +64 3 479 4196; fax: +64 3 479 5079. E-mail addresses: [email protected], [email protected] (M.V. Swain). 0109-5641/$ – see front matter © 2011 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dental.2011.08.003
1112
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
to zirconia; clinical studies have reported veneering failure; namely chipping as an issue [2,3]. Moreover, research has shown that differences in failure rate not only exist between the hand veneering and the pressed veneering ceramics, but also between the different pressed ceramic materials [4,5]. Possible reasons postulated for chipping are: insufficient bond strength, excessive tensile stress due to a coefficient of thermal expansion mismatch with an un-equal thermal contraction between veneer layer and coping, and excessive load due to premature contacts, tensile thermal tempering residual stress introduced during cooling [6,7]. Ceramic materials used for the fabrication of all-ceramic restorations are sintered and glazed several hundred degrees above the glass transition temperature (Tg ). At the end of the firing process the dental technician removes the ceramic restoration from the furnace and it is cooled. Once the porcelain is removed from the furnace, its outer surface starts contracting and becomes rigid as it loses heat rapidly, while the interior part is still hot and somewhat viscous. The tensile stresses which would normally develop in an elastic body during such rapid cooling are relaxed while the interior of the porcelain is above the Tg , or more specifically above the softening temperature. However, temperature gradient results in the formation of surface compressive stresses and compensating internal tensile stresses, so called tempering stresses in the body. An indication of the magnitude of such stresses has been addressed by Asaoka et al. [8] and Swain [6] for the case of a bilayer material system. The tensile stresses associated with tempering may cause immediate cracking of the porcelain upon deformation of the restoration, and can increase the probability of fracture during functional loading of the restoration, manifested clinically as chipping of the veneering porcelain [9]. One problem associated with ceramic restoration failure is therefore the development and magnitude of residual stress within the structure. The residual stress developed may also influence the measured nominal fracture toughness of ceramic materials. Marshall and Lawn developed a simple approach using indentation fracture about a Vickers indenter to quantify the magnitude of these residual stresses [10]. Fracture toughness or critical stress intensity (K1c ) indicates the ability of a material to resist rapid crack propagation and is an indicator of clinical reliability and serviceability of a ceramic restoration. The present study thus has three primary objectives, namely; quantifying the fracture toughness of the four pressable veneering ceramic materials; determining the residual stress present in various pressable ceramics bonded to zirconia using the indentation fracture toughness technique and measuring the cooling curves when the ceramics are subjected to different cooling procedures.
2.1.
Vita In-Ceram YZ zirconia and the pressed ceramics were prepared to the dimension required for each test following the manufacturers’ instructions as listed in paper 1. Three groups of specimens were prepared (1) Bilayered group (1-1) 2 mm zirconia and 2 mm pressed porcelain (1-2) 2 mm zirconia and 4 mm pressed porcelain (2) One layer group (2-1) 2 mm of pressed porcelain ceramic only (2-2) 2 mm slice from each porcelain ingot before pressing The bilayered specimens were divided into 3 groups. Before glazing, the pressed ceramic surface of the specimens were all ground and polished to the desired dimension for each test, using 120–4000 grades of abrasive paper (Struers PSA backed Silicon carbide paper) on a metallographic lapping machine (Knuth Rother, Struers, Denmark) for the better observation of the cracks. For group (1) the specimens were self-glazed at the temperature recommended in the manufacturer’s instructions and then removed from the furnace as soon as the support plinth was low enough to remove the samples. The specimens were then force cooled with compressed air (2 bar) from two opposing sources 30 cm apart. Specimens for group (2) were glaze fired and cooled to mimic what dental technicians normally do when building crowns. The specimens were removed from the furnace when the temperature of the furnace drops to the starting temperature of the glazing cycle. Specimens for group (3) were glaze fired and slow cooled by stopping the firing plinth descending from the furnace and waiting until the temperature reduces to 100 ◦ C. The latter took approximately 30–40 min depending on the firing cycle used.
2.2.
Materials and methods
Materials used are presented in Table 1 in paper part 1.
Indentation tests
Two groups of specimens were used: (i) 2 mm pressed ceramic only and (ii) a 2 mm slice from an ingot before pressing. Specimens for group (i) were glaze fired and slowly cooled in the furnace to 100 ◦ C to measure the intrinsic fracture toughness without any tempering induced residual stress present. Indentation cracks on the veneer surface of all specimens were made with a Vickers indenter at a load of 25 N for samples with ceramic only and 35 N for the samples before pressing. Different loads were chosen after making preliminary indentations with loads ranging from 20 to 70 N and the loads mentioned above showed the most suitable crack patterns in terms of size and absence of chipping. Between 5 and 12 indents were made depending on the porosity of the specimen. Indentations were made approximately 3–4 mm apart and cracks were measured directly post indentation. The following equation was used to calculate the effective fracture toughness [9]: Kc =
2.
Specimen preparation
E 0.5 P H
c3/2
(1)
where H is the hardness, P the load, E the elastic modulus, c the length of radial crack, and is a constant 0.016.
1113
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
Table 1 – Summary of the materials used in the study with their coefficient of thermal expansion and glass transition temperatures. Materials used
CTE (ppm/K)
Tg (◦ C)
Vita Zahnfabrik Ivoclar Vivadent Wieland Dental + Technik Noritake Kizai Co.
9.2 9.75 9.3 10.1
640 530 620 615
Vita Zahnfabrik
10.5
Manufacturer
Pressed ceramics Vita PM9 IPS e.max Zirpress Wieland Xzr Noritake CZR Zirconia Vita In-Ceram YZ
Indentation cracks were introduced within the veneer surface of all specimens using a Vickers indenter at a load of 50 N for the bi-layered specimens. The high polish achieved on the sample surface minimised the inaccuracies associated with crack terminations due to surface scratches and porosity. An optical microscope with a high resolution digital stage (Nikon, Japan) was used to measure the length of radial cracks and to microscopically inspect the indentation surface of the specimens.
2.3.
Residual stress determination
The presence of residual stresses is very apparent by the extent of radial crack extension about the residual impression. Marshall and Lawn developed a simple analysis of this problem [10]. The presence of residual or applied tensile or compressive stress in the surface of the material being indented introduces an additional term in the expression for the stress intensity factor as given by Eq. (1). For the situation where the indentation crack size is small compared with the gradient of stress, the expression for the stress intensity factor is composed of two terms: K1c = KInd + KApp/Res
(2)
where KInd is the indentation stress intensity factor given by Eq. (1) and KApp/Res is the stress intensity factor for a crack in the residual or applied stress field and is given by: KApp/Res
√ = YApp/Res c
(3)
where Y is the shape factor for a half-penny surface crack and App/Res is the applied or residual stress present. Combining Eqs. (1)–(3) we have: K1c =
E 1/2 P H
c3/2
√ ± YApp/Res c
(4)
The second term in Eq. (4) may be positive or negative depending upon whether the stresses on the surface are tensile or compressive. The above approach was used to determine the tempering residual stresses developed in veneering porcelains as a consequence of rapid cooling [8,10].
2.4.
heating and cooling of the specimens. Two groups of samples were used: (1) 2 mm zirconia and 2 mm porcelain; (2) 2 mm zirconia and 4 mm porcelain. To illustrate the cooling response curves, two ceramic materials were chosen among the four; Noritake CZR Press and IPS e.max ZirPress to observe the difference between leucite containing and non-leucite containing glass ceramic materials and their respective glaze firing cycles. A small hole was made into the specimens to the interface in the middle of the sample with a high speed bur. The thermocouple wires were then inserted and covered by porcelain to hold the wires in place. Another thermocouple wire was placed on the surface of the sample and held in place with porcelain. The specimens with thermocouples attached were then taken through the final glaze firing cycle for each bi-layer system. The thermocouple wires were connected to a digital data logging system (PowerLab, Sydney, Australia). The specimens were cooled precisely as done for samples used for indentation fracture toughness determination. During this firing and cooling process, the data logging system recorded the inner (interface) and outer (surface) temperature of the veneering porcelain.
Thermocouple observations
Welded thermocouple wires (Type K) were placed on the surface and embedded at the porcelain–zirconia interface, allowing continuous measurement of temperature during
3.
Results
3.1.
Indentation fracture toughness test
The results obtained from indentation fracture toughness tests using Eq. (1) are presented in Fig. 3. The nominal fracture toughness of the four pressable ceramics in this study was in the range of 0.92–2.01 MPa m0.5 before pressing and 0.6–0.78 MPa m0.5 after pressing for ceramics not veneered to zirconia. The intrinsic fracture toughness of the materials is considered to be that for the 2 mm ceramic only specimens which were slowly cooled. It was found that Noritake CZR Press has the highest fracture toughness before and after pressing, and the leucite containing materials (Noritake CZR Press and Vita PM9) have higher values than non-leucite containing glass ceramics (Wieland PressXzr and IPS e.max ZirPress). The nominal fracture toughness Kc values obtained in this study were in the range of 0.8–1.74 MPa m0.5 (Fig. 3). When only the leucite containing ceramics are compared, Vita PM9 had a higher effective Kc than that of Noritake CZR when fast cooled. However, when medium and slow cooled, Noritake had a higher effective Kc than the rest of the materials. Vita PM9 and IPS e.max ZirPress had similar Kc values when they were slow cooled. Upon slow cooling, there was a slight increase in
1114
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
the effective fracture toughness values as the thickness of the veneering porcelain increases from 2 mm to 4 mm. Moreover, it was also found that the leucite containing (Vita PM9 and Noritake CZR Press) pressed ceramics generally have higher Kc in all situations than non-leucite containing glass pressed ceramics (Wieland PressXzr and IPS e.max ZirPress).
3.2.
The residual stresses
The residual stresses in the surface of the samples were determined based upon Eq. (4). The effective or intrinsic Kc values for the individual materials were chosen as the slowly cooled porcelain only 2 mm samples. With this value the residual stress results obtained from indentation fracture toughness are presented in Fig. 4. As indicated in Eq. (2), the values of effective fracture toughness are influenced by the tensile/compressive residual stresses present within the ceramic systems. The magnitude of the residual stresses present in the specimens under different cooling conditions followed the same trend as the effective fracture toughness values. For the samples before pressing; there was a large compressive stress present for all four materials. The residual stress present in the bilayered samples was directly related to the cooling rate, where fast cooling resulted in high stress levels and slow cooling reduced stress levels.
3.3.
Thermocouple determined cooling curves
The cooling curves obtained by the thermocouples for the three cooling conditions for both the Noritake and IPS e.max ZirPress materials are presented in Figs. 1a and 2a. The graphs showed a significant difference between cooling conditions for all materials, namely; the faster the cooling, the steeper the temperature gradient between the firing temperature and the Tg point (Table 1). Here the firing temperatures recommended by the manufacturers were used. For the Noritake CZR Press, the glaze temperature was 900 ◦ C while for IPS e.max ZirPress it was 770 ◦ C. The temperature difference between the interface and the outer surface of the porcelain during cooling is shown in Fig. 2a and b. The data in these figures have been fitted with a simple polynomial trend line of best fit and shows the Noritake CZR fired at a higher temperature, resulting in a higher temperature difference. The 4 mm bi-layered samples (not included) also showed the same trend however the temperature difference was greater than that of 2 mm bilayer samples.
4.
Discussion
The cooling curves show big differences between the three conditions evaluated (fast, normal and slow cooling) (Figs. 1 and 2). This is also clearly seen in the temperature differences between the inner and outer thermocouple readings (Fig. 2a and b). These observations are very similar to what Asaoka et al. observed when they measured the influence of cooling rate on porcelain and porcelain bonded to metal systems [8]. As shown in the cooling curves (Figs. 1 and 2), the trends of three different cooling conditions were similar for both
Fig. 1 – Cooling curves for (a) Noritake CZR Press (2 mm thickness) and (b) IPS e.max ZirPress (2 mm thickness) when subjected to slow, normal and fast cooling.
Noritake CZR Press and IPS e.max ZirPress. However, it is obvious that the cooling curves for the Noritake CZR Press are steeper than those of IPS e.max ZirPress, in addition to having a greater temperature difference between the inner and the outer surfaces of the material. This is anticipated from Newton’s cooling law; namely the rate of change of the temperature of an object is proportional to the difference between its current temperature and the ambient temperature. This rationalises the observations that for the same cooling heat transfer conditions the higher the firing temperature, the faster the cooling rates. Another observation was that as the thickness of the porcelain increased the temperature difference between the outer surface and the inner thermocouple increased as may be seen in Table 1. These observations are in accord with those of Asaoka et al. [8] and Swain [6]. Plots of the temperature difference for the 2 mm thick Noritake and IPS e.max samples as a function of the external temperature of the specimens for the three cooling conditions are shown in Fig. 2a and b. In the case of the Noritake CZR Press, the fast cooling condition resulted in a temperature difference greater than 80 ◦ C between the inner and outer thermocouple that persists over an extended cooling range from when the external temperature changes from 700 ◦ C to 250 ◦ C. Whereas the normal and slow cooling conditions resulted in similar shaped curves but the temperature differences were only 35 ◦ C and 10 ◦ C respectively. For IPS e.max ZirPress, the lower firing temperatures resulted in less temperature difference under the same nominal cooling conditions as the
1115
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
Table 3 – The average cooling rate between −50 ◦ C from the glaze temperature and the temperature difference when the external temperatures is at Tg .
Noritake CZR Press Tg = 615 (◦ C) 2 mm Slow cool Normal cool Fast cool 4 mm Slow cool Normal cool Fast cool IPS e.max ZirPress Tg = 530 (◦ C) 2 mm Slow cool Normal cool Fast cool 4 mm Slow cool Normal cool Fast cool
Fig. 2 – Difference between external and internal thermocouple measured temperature readings for the three cooling conditions shown in Fig. 1 for (a) Noritake and (b) IPS e.max.
Noritake samples. In this instance the temperature differences were 35 ◦ C for the fast, 18 ◦ C for the normal and 5 ◦ C for the slow cooling conditions. Table 1 contains the maximum temperature differences when the external temperature reaches Tg for Noritake and IPS e.max (Tables 2 and 3).
Cooling rate veneering porcelain (◦ C/s)
Temperature difference porcelain is at Tg (◦ C)
−1.1 −6.6 −16.6
10 31 85
−0.16 −5.8 −19.6
9 52 121
−0.02 −3.95 −10.29
7 28 35
−0.05 −2.8 −17.42
8 22 92
The effective indentation fracture toughness test results show that a wide range of Kc values were found for all the pressed veneering ceramics investigated (Fig. 3). The highest values were in two cases the samples cut from the “as received” ingots, namely for the Noritake and Weiland materials. However after disks of the ingot materials were refired to the pressing temperature and slowly cooled within the furnace, the effective Kc for all materials were the least. It is also clearly seen that with increasing the cooling rate and thickness of the porcelain, the effective Kc values also increased. Another feature evident was that in the case of slow cooling, irrespective of the thickness of the porcelain, or whether the porcelain was fused to zirconia or not, there appeared to be minimal difference in the effective Kc values measured for a specific material.
Table 2 – Pressing and glazing schedule for the pressed ceramics used in the study. Brand/name of product
(a) Pressing schedule Vita/PM9 Noritake/CZR Press IPS e.max ZirPress Wieland/PressXzr
Brand/name of product (b) Glazing schedule Vita/PM9 Noritake/CZR IPS e.max ZirPress Wieland/PressXzr
Heat up temp (◦ C)
Start temp (◦ C)
Heat rate (◦ C/min)
Vacuum hold time (min)
Pressing temp (◦ C)
Press time (min)
850 850 900 900
700 700 700 700
50 60 60 60
20 26 15 20
1000 1065 910 1060
6 6 6 8
Drying time (min)
Start temp (◦ C)
Heat rate (◦ C/min)
Firing temp (◦ C)
Hold time (min)
5 5 5 5
500 600 450 575
80 65 60 75
900 900 770 880
1 1 1 1
1116
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
Fig. 3 – Summary of the effective fracture toughness values of the four different pressed ceramics determined by the indentation fracture toughness technique.
Fig. 4 – Summary of surface residual stress values of the four different pressed ceramics determined by the indentation fracture toughness technique.
In this work we consider the values of Kc measured for the slow cooled and unbonded porcelain as the reference Kc values of the individual porcelains. This follows the approach of Marshall and Lawn [10] and industrial procedures namely that slow cooling rates are generally used to minimise the presence of residual stresses in glasses and visco-elastic materials such as porcelains, see for instance Kingery et al. [11] The values of the intrinsic Kc values for the four pressed porcelains range from 0.6 to 0.78 MPa m0.5 which is typical for porcelains. The leucite containing materials exhibited the highest Kc , whereas the primarily glass materials had the least. These results are in accord with a recent study by Cesar et al. [12] that showed an increasing trend of Kc values of veneering porcelains with increasing leucite content. From the indentation fracture toughness results, a significant trend was found; a decrease of nominal Kc values of all materials after pressing. While other materials Kc dropped by 5–15% due to pressing, Noritake dropped approximately 60% (2.01–0.78 MPa m0.5 ). According to the X-ray diffraction results from the companion paper [Choi et al.], Noritake had only a very minor change in leucite content in terms of volume fraction or crystal structure, which would not have been expected to result in the significantly lower Kc after pressing. Vita PM9 on the other hand, was expected to change significantly, as the X-ray diffraction results revealed that there were differences of leucite volume fraction and crystalline structure (from tetragonal to cubic) after pressing. However, Vita PM9 had the smallest change in Kc before and after pressing, 0.92–0.71 MPa m0.5 . This drop in Kc values after pressing may have been caused by the residual stress present in the material before pressing caused by the block manufacturing process. Despite having leucite as the main crystalline structure, the Kc values of Noritake and Vita materials had a slight difference after pressing. The Kc value of Noritake CZR Press was slightly higher than that of Vita PM9 after pressing for annealed monolithic samples. From the X-ray diffraction patterns obtained in part 1 of this study, it was found that the Noritake CZR Press has a higher volume fraction and retains the tetragonal leucite after pressing; whereas Vita PM9 has a
change in leucite structure after pressing from tetragonal to cubic as well as a reduction in volume fraction. Rasmussen et al. [13] and Denry et al. [14] showed that dental porcelains with cubic leucite at room temperature had significantly lower strength as well as fracture toughness than porcelains with tetragonal leucite. This may have also contributed to the reduction observed in this study. A similar decrease in Kc values after pressing was also found with Wieland PressXzr and IPS e.max ZirPress. Since these materials were found to be non-leucite glass ceramics it can be concluded that the drop in the Kc after pressing for these materials is mainly caused by the presence of the residual stress in the ingot before pressing due to the manufacturing process, such as rapidly cooling induced residual stresses developed. Utilizing the above estimates of the intrinsic Kc values for the different porcelains it is possible to estimate the residual stresses developed on the surface of the differently cooled porcelains from Eq. (4). The values of these stresses are plotted in Fig. 4. For most of the materials there appears to be a clear trend in the results, namely that with the faster cooling rates and thicker porcelain layer, the higher the surface residual compressive stresses. There are exceptions in that the Vita material appears to have higher residual stresses for the 2 mm thickness rather than the 4 mm when fast cooled and the values for normal cooling were very low, even lower than the slow cooled specimens. For the Noritake and IPS e.max materials where the temperature was measured during cooling for the 2 and 4 mm thick samples and the surface residual compressive stresses determined as shown in Fig. 4, it was possible to plot residual stress versus the temperature difference when the surface of the specimen was at Tg as shown in Fig. 5. A simple linear plot through these data points shows an excellent fit. This relationship is anticipated for viscous materials rapidly cooled from above Tg . As shown in the classic text by Kingery et al. when a viscous body experiences a temperature gradient during cooling the stresses that would nominally be developed if the material were elastic are able to be relaxed by viscous flow [11]. The ability for this relaxation decreases when the temperature
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
Fig. 5 – Graph of the magnitude of the surface compressive residual stresses versus the temperature difference between the inner and outer surface when the external surface reached Tg for the Noritake and IPS e.max materials.
of the object drops below the softening temperature Ts and completely below Tg . From the plots of the temperature difference between the inner and outer surfaces of the porcelain as shown in Fig. 2, it is evident that the temperature differences remains reasonably constant over the interval for the external and internal thermocouple positions, cooling well below the Tg temperature. This implies that the stress difference between the external and internal parts of the porcelain will reflect the temperature difference along with the E modulus and thermal expansion. According to Kingery et al. [11] the stress developed as a consequence of this temperature difference for a homogeneous material, is given by: =
E˛T 1−
in some instances had completely run through the porcelain, were found for fast cooled samples. No such change was observed for slow cooled samples. This is considered due to the subsurface component of indentation cracks in the rapid cooled samples being influenced by the tensile residual stresses present within the material compensating the compressive stresses developed on the surface. The cause of chipping of veneering ceramics on zirconia all-ceramic cores was reported to be multi-factorial [1–7]. Therefore the individual and the combined effects of such variables studied in the two parts of the study; the crystalline structure, bond strength, fracture toughness, flexural strength and the cooling behavior can influence the fracture behavior of pressed ceramics on to zirconia and therefore the clinical success of such restorations. The results from part 1 revealed two different types of pressed ceramics; leucite and glass (nonleucite) ceramics and how the presence of leucite influences the mechanical properties. The outcome of this paper leads to a better appreciation of the role of cooling on the residual stress formation in the different pressed ceramic materials. However, the similar trends found in cooling curve analysis and residual stress developed for both leucite and non-leucite types of materials places more emphasis on the cooling conditions regarding the residual stress formation along with the thickness of the porcelain; especially for zirconia bilayered pressed ceramics.
5.
Conclusions
Within the limitations of this study, the following conclusions were drawn:
(5)
where E is the elastic modulus, is Poisson’s ratio, ␣ is the coefficient of thermal expansion and T is the temperature difference. For a tempered object the internal tensile stress is half the magnitude of the external compressive stress. Thus the slope of line in Fig. 5 should be given by: 2E˛ = T 3(1 − )
1117
(6)
The value 2/3 arises from the fact that only the surface compressive stresses were determined using the indentation approach and are compared with the temperature difference. The slope from the data in Fig. 5 is approximately 0.4 MPa/◦ C whereas the predicted slope for a material with an E modulus of 70 GPa, 0.2 Poisson’s ratio and 9 × 10−6 /◦ C coefficient of thermal expansion is 0.5 MPa/◦ C. This simple approach does not include any relaxation as incorporated in more rigorous numerical models by DeHoff and Anusavice [15,16] and considers the temperature profile and resultant stresses in a bilayer system simply as for a homogeneous visco-elastic body. For some fast cooled samples, the indentation cracks formed were again observed 24 h after the indentation to see the effect of time delay between indenting and crack measurement. When the cracks were re-visited, longer cracks, which
• The intrinsic Kc results obtained from indentation showed a specific trend. For slow cooled samples, the leucite containing ceramics (Noritake and Vita) have higher fracture toughness values than non-leucite glass ceramics (Wieland and IPS e.max). • There was a significant presence of residual stress found in the material even before pressing which may have been caused by the ingot manufacturing process. • Fast cooled samples produced short cracks (higher effective Kc ) due to compressive residual stress formed on the surface to compensate the tensile stress formed inside the system. • Slow cooling almost completely eliminated formation of residual stress for all pressed ceramics.
Therefore the following general conclusions were drawn:
• Practitioners have choices of two types of ceramic materials when using pressing technique: Leucite containing and non-leucite glass ceramics. • To reduce the development of residual stress within the ceramic system, practitioners are advised to slow cool the restoration on the last heat treatment cycle (e.g. glazing cycle).
1118
d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 1111–1118
Acknowledgment The authors wish to thank Dr Basil Al-Amleh for critically reading this paper.
references
[1] Gorman CM, Mcdevitt WE, Hill RG. Comparison of two heat-pressed all-ceramic dental materials. Dent Mater 2000;16:389–95. [2] Tsalouchou E, Cattel MJ, Knowles JC, Pittayachawan P, Mcdonald A. Fatigue and fracture properties of yttria partially stabilized zirconia crown systems. Dent Mater 2008;24:308–18. [3] Sailer I, Feher A, Filser F, Gauckler LJ, Luthy H, Hammerle CHF. Five-year clinical results of zirconia frameworks for posterior fixed partial. Int J Prosthodont 2007;20:383–8. [4] Christensen GJ. PFM vs. zirconia restorations—how are they comparing clinically? Clinician’s Report, vol. 1(11). CR Foundation; November 2008. [5] Taskonak B, Borges GA, Mecholsky J, Anusavice KJ, Moore KB, Yan J. The effects of viscoelastic parameters on residual stress development in a zirconia/glass bilayer dental ceramic. Dent Mater 2008;24:1149–55. [6] Swain MV. Unstable cracking (chipping) of veneering porcelain on all-ceramic dental crowns and fixed partial dentures. Acta Biomater 2009;5(5):1668–77.
[7] Fisher J, Stawarczyk B, Trottmann A, Hammerle CHF. Impact of thermal misfit on shear strength of veneering ceramic/zirconia composites. Dent Mater 2009;25(4):419–23. [8] Asaoka K, Kuwayama N, Tesk JA. Influence of tempering method on residual stress in dental porcelain. J Dent Res 1992;71:1623–7. [9] Isgro G, Wang H, Kleverlaan CJ, Feilzer A. The influence of multiple firing on thermal contraction of ceramic materials used for the fabrication of layered all-ceramic dental restorations. Dent Mater 2005;21:557–64. [10] Marshall DB, Lawn BR. Indentation technique for measuring stresses in tempered glass surfaces. J Am Ceram Soc 1977;60:86–7. [11] Kingery WD, Bowen HK, Uhlmann DR. Introduction to ceramics. 2nd ed. New York: John Wiley & Sons; 1975. [12] Cesar PF, Yoshimura HN, Miranda Jr WG, Miyazaki CL, Muta LM, Rodrigues Filho LE. Relationship between fracture toughness and flexural strength in dental porcelains. J Biomed Mater Res B Appl Biomater 2006;78:265–73. [13] Rasmussen ST, Groh CL, O’Brien WJ. Stress induced phase transformation of a cesium stabilized leucite porcelain and associated properties. Dent Mater 1998;14:202–11. [14] Denry IL, Mackert JR, Holloway JA, Rosentiel SF. Effect of cubic leucite stabilization on the flexural strength of feldspathic dental porcelain. J Dent Res 1996;75:1928–35. [15] DeHoff PH, Anusavice KJ. Viscoelastic stress analysis of thermally compatible and incompatible metal–ceramic systems. Dent Mater 1998;14:237–45. [16] Dehoff PH, Anusavice KJ. Tempering stresses in feldspathic porcelain. J Dent Res 1989;68:134–8.