Abstract
Objectives
To assess surface/subsurface damage after grinding with diamond discs on five CAD-CAM restorative materials and to estimate potential losses in strength based on crack size measurements of the generated damage.
Methods
The materials tested were: Lithium disilicate (LIT) glass-ceramic (e.max CAD), leucite glass-ceramic (LEU) (Empress CAD), feldspar ceramic (VM2) (Vita Mark II), feldspar ceramic-resin infiltrated (EN) (Enamic) and a composite reinforced with nano ceramics (LU) (Lava Ultimate). Specimens were cut from CAD-CAM blocs and pair-wise mirror polished for the bonded interface technique. Top surfaces were ground with diamond discs of respectively 75, 54 and 18 μm. Chip damage was measured on the bonded interface using SEM. Fracture mechanics relationships were used to estimate fracture stresses based on average and maximum chip depths assuming these to represent strength limiting flaws subjected to tension and to calculate potential losses in strength compared to manufacturer’s data.
Results
Grinding with a 75 μm diamond disc induced on a bonded interface critical chips averaging 100 μm with a potential strength loss estimated between 33% and 54% for all three glass-ceramics (LIT, LEU, VM2). The softer materials EN and LU were little damage susceptible with chips averaging respectively 26 μm and 17 μm with no loss in strength. Grinding with 18 μm diamond discs was still quite detrimental for LIT with average chip sizes of 43 μm and a potential strength loss of 42%.
Significance
It is essential to understand that when grinding glass-ceramics or feldspar ceramics with diamond discs surface and subsurface damage are induced which have the potential of lowering the strength of the ceramic. Careful polishing steps should be carried out after grinding especially when dealing with glass-ceramics.
1
Introduction
CAD-CAM restorative materials such as feldspar ceramics, leucite glass-ceramic, lithium disilicate glass-ceramic are commonly used for chairside made single crowns, inlays, onlays and veneers. Newer materials with interesting lower elastic moduli have been added to the available CAD-CAM chair-side blocs such as a feldspar ceramic resin-infiltrated and a resin composite reinforced with nano ceramic particles. The most popular CAD-CAM milling unit (Cerec) uses diamond burs of 64 μm grit (a step and a cylinder pointed bur 12S) operating at 42,000 rpm on sintered, presintered ceramic or fully polymerized composite blocs. Further 3 manual polishing steps are then recommended using existing polishing kits which are manufacturer depended containing 2 synthetic rubbers (coarse and fine) and a diamond paste 2–4 μm . A crystallization/glaze firing cycles are needed for some ceramic blocs after milling. Depending on the quality of the tooth preparation, the optical impression and the 3D design of the restoration, the computer-generated restoration may undergo additional manual adjustments. These are usually performed with a red color-coded diamond bur (45–50 μm) on inner angles, interproximal contact points and occluding surface. These adjustments on the outer surface are further polished using manufacturers recommended polishing kits. After cementation, the occlusal surface is again checked and adjusted if needed with the red ring diamond bur and polishing rubbers. The margins of the milled restoration are usually further polished after cementation on the tooth preparation. Dental ceramics and resin composites will suffer strength degradation when the surface is roughened . As a general rule, diamond grinding on brittle materials should be followed by polishing steps in order to remove the permanent surface and subsurface damage introduced during grinding . Strength degradation will come from the introduction of microcracks from the abrasive diamond grinding. Depending on the direction of grinding, these small cracks can run either parallel to the grinding direction or orthogonal . Radial cracks propagating in the plane of motion may also be generated due to increased tensile stresses occurring from tangential forces exercised during grinding .
Overall, the literature seems unanimous in suggesting that grinding ceramics induces surface/subsurface damage which should be eliminate by using finer grits and polishing steps. This is particularly true for Cerec CAD-CAM milling units operating with 64 μm diamond burs. Surface chipping and subsurface microcracking of 40–60 μm were reported using step-wise lapping after CAD-CAM milling a feldspar porcelain (Vita Mk2) . Maximum crack lengths of respectively 130 μm for feldspar porcelain (CAD-CAM blocs, Sirona Cerec) and 120 μm for a feldspar-based layered porcelain (Vita Omega) were measured after grinding with 100 μm coarse diamond burs at 45,000 rpm . The strength reduction estimates using fracture mechanics relationships were respectively of 30% and 7%. Finite Element Modelling could successfully model subsurface induced damage from resurfacing feldspar porcelain (Vita Mk2) and leucite glass-ceramic (Pro CAD) with coarse (106–125 μm) diamond burs showing a dependence on feed rate and depth of cut . Finer grit size diamonds (20–30 μm) on feldspar-based porcelain (Vita Mk2) induced subsurface damage below 18 μm but remained dependent on the bur depth of cut and feed rate . Diamond grits on feldspathic porcelain larger than 40 μm resulted in rough surfaces, chipped edges, and strength limiting surface and subsurface microcracks . Generally speaking, ceramics with increased fracture toughness will show decreased grinding chip damage .
In clinical dentistry, grinding adjustments of the occlusion, interproximal contacts followed by the needed ceramic surface polishing steps (up to 1 μm gloss finish) after CAD-CAM milling are not always performed to satisfaction by the dentist. Grinding damage remains on the surface and may extent critically into the subsurface. The diamond burs commonly used for working on dental ceramics by the dentist or dental technician have color-coded grit sizes. A black ring is defined as “supercoarse” (150–180 μm), a green ring “coarse” (125–150 μm), a blue ring “standard” (100–110 μm), a red ring “fine” (45–50 μm) and a yellow ring “superfine” (15–30 μm). High speed rotatory diamonds combined with high pressure will generate even more pronounced grinding damage. In order to obtain a visual assessment of machining or grinding induced surface/subsurface crack damage a bonded interface technique has been used combined with optical microscopy and SEM to analyze ceramic crack-damage . As near-surface damage has been recognized as a persistent problem in crowns obtained by CAD-CAM , there is a real need to assess the surface/subsurface damages induced by diamonds grits used for reshaping or adjustment in clinical dentistry.
With the increasing variety of available chairside CAD-CAM restorative materials it was the purpose of this research to assess, via the bonded interface technique , the surface/subsurface damages induced by diamonds of medium (75 μm), fine (54 μm) and superfine (18 μm) grits on five chairside CAD-CAM restorative materials. By measuring maximum and average damage chip depths and assuming the chip to act as a crack of critical size, estimates of fracture stresses based on fracture mechanics relationship were calculated and compared to the manufacturer’s reported strength value in order to estimate potential losses in strength as a result of grinding damage.
2
Materials & methods
2.1
Materials
The selected five CAD-CAM materials are listed in Table 1 including data from the manufacturer regarding fracture toughness, flexural strength (ISO 6872) and elastic modulus. (1) A lithium disilicate glass-ceramic (IPS e.max CAD; Ivoclar Vivadent) made of 70 vol% of ∼1 μm lithium disilicate crystals. (2) A leucite glass-ceramic (IPS Empress CAD; Ivoclar Vivadent) made of 35–45 vol% of 1–5 μm leucite crystals. (3) A feldspar ceramic (Vita Mark II; Vita Zahnfabrik) containing 30 vol% of ∼5 μm sanidine crystals. Further investigations on this feldspar ceramic revealed multiple crystalline phases including sanidine , feldspathoid nepheline and anorthoclase . (4) A hybrid ceramic (Enamic; Vita Zahnfabrik) made of a structure-sintered feldspar ceramic (86 wt%) containing pores which are infiltrated by 14 wt% UDMA/TEGDMA polymer resin. (5) A resin composite, defined by the manufacturer as “resin nano ceramic” (Lava Ultimate; 3 M), a misleading term as this is not a ceramic, is made of a highly cross-linked resin matrix reinforced with 80 wt% of silane treated nano zirconia–silica particles agglomerated to clusters (0.6–10 μm) and individual silane bonded nano silica or zirconia particles (<20 nm) reducing the interstitial space between the clusters.
| Brand name | Description | Manufacturer | E (GPa) | Flex. strength S (MPa) ISO 6872 | Fracture toughness K Ic (MPa√m) |
|---|---|---|---|---|---|
| IPS e.max CAD (LIT) | Lithium disilicate glass-ceramic | Ivoclar Vivadent Schaan, FL | 95 | 360 | 2.2 |
| IPS Empress CAD (LEU) | Leucite glass-ceramic | Ivoclar Vivadent Schaan, FL | 62 | 160 | 1.3 |
| Vita Mark II (VM2) | Feldspar porcelain | Vita Zahnfabrik Bad Säckingen, D | 63 | 113 | 1.7 |
| Enamic (EN) | Feldspar ceramic- polymer infiltrated | Vita Zahnfabrik Bad Säckingen, D | 30 | 155 | 1.5 |
| Lava Ultimate (LU) | Resin composite reinforced with nano ceramics | 3 M, Seefeld, D | 13 | 205 | 2.0 |
2.2
Specimen preparation
Pairs of 3 × 4 × 12 mm slices where cut from the original CAD-CAM blocs using a low speed rotatory fine cutting diamond disc under cooling water in an ISOMET machine (Buehler). All the CAD-CAM materials were already fully dense except lithium disilicate glass-ceramic (e.max CAD) provided in a partially crystalized state by the manufacturer and for which the cut specimens had to undergo additional sintering for dense crystallization. For the bonded interface technique, the cut pairs of specimens were tightly mounted together with sticky wax onto a metal support following a three step procedure as described below. In step 1, surface A (top) ( Fig. 1 ) was ground with hand pressure using diamond discs of decreasing grit size (75, 54, 18 μm) mounted on a turntable (RotoPol 22, Struers) rotating at 300 rpm with water irrigation for 5 min per grit. The ground surfaces were further polished with diamond pastes of 6, 3 and 1 μm at 150 rpm rotation speed for 3 min each paste. In step 2, both surfaces A were bonded together exposing surface B (step 3) for receiving the same grinding sequence as described in step 1. In step 4, pairs were again bonded together by their surface B, exposing the mirror polished surfaces A facing the top. Grinding damage of respectively 75, 54 and 18 μm diamond discs was performed on individual sets of pairs, ground longitudinally (i.e. parallel to the bonded interface) on their mirror polished surface A for 3 min at 150 rpm under water irrigation using hand pressure by one and same operator. After grinding, the bonded pairs were detached, cleaned ultrasonically in a 16% sodium hypochlorite solution for 3 min, a routine cleaning procedure in our laboratory before fractographic SEM analysis. The resulting damage on top surface A and interface B was viewed under the SEM and represented schematically in Fig. 1 showing edge chips and cracks running parallel or orthogonal crack to the ground surface.
2.3
SEM damage evaluation
For each material tested, a pair (N = 2) of specimen selected ground conditions (i.e. 75, 54 and 18 μm) was viewed under the SEM for surface and chip damage evaluation. Chip sizes were measured on the bonded surface B and the presence of extending cracks searched within the chip fractured surface. The top ground surface (A) was analyzed to assess the abrasion mode (ductile or brittle).
2.4
Stress estimates and loss of strength
Fig. 2 illustrates an example of a chip-damage on the bonded interface (B). The chip’s depth ( a ) and width (2c) is assessed over two thirds of the specimen’s length. Mean ( a mean ) and maximum crack depth ( a max ) were assessed for a minimum of 15 and maximum of 30 localized chips. The stress intensity shape factor Y was estimated from chip depth and width ratio . Hence, a semi-ellipse with c = 2 a will have a Y = 1.6. Using the general relationship of K Ic = Y σ f √ a (Eq. (1)) and manufacturer’s K Ic values ( Table 1 ), stress estimates were calculated assuming that the chip depth ( a ) is a strength limiting flaw. The estimated stress ( σ f ) for a given chip size was then compared to the strength values reported by the manufacturer ( Table 1 ) for each material and percentage of possible loss in strength reported.
2.5
Statistical analysis
One-way ANOVA was used to compare chip size damage. Fischer’s least significant difference test at 95% level of significance (p < 0.05) was performed for chip size differences among materials and diamond grit.
2
Materials & methods
2.1
Materials
The selected five CAD-CAM materials are listed in Table 1 including data from the manufacturer regarding fracture toughness, flexural strength (ISO 6872) and elastic modulus. (1) A lithium disilicate glass-ceramic (IPS e.max CAD; Ivoclar Vivadent) made of 70 vol% of ∼1 μm lithium disilicate crystals. (2) A leucite glass-ceramic (IPS Empress CAD; Ivoclar Vivadent) made of 35–45 vol% of 1–5 μm leucite crystals. (3) A feldspar ceramic (Vita Mark II; Vita Zahnfabrik) containing 30 vol% of ∼5 μm sanidine crystals. Further investigations on this feldspar ceramic revealed multiple crystalline phases including sanidine , feldspathoid nepheline and anorthoclase . (4) A hybrid ceramic (Enamic; Vita Zahnfabrik) made of a structure-sintered feldspar ceramic (86 wt%) containing pores which are infiltrated by 14 wt% UDMA/TEGDMA polymer resin. (5) A resin composite, defined by the manufacturer as “resin nano ceramic” (Lava Ultimate; 3 M), a misleading term as this is not a ceramic, is made of a highly cross-linked resin matrix reinforced with 80 wt% of silane treated nano zirconia–silica particles agglomerated to clusters (0.6–10 μm) and individual silane bonded nano silica or zirconia particles (<20 nm) reducing the interstitial space between the clusters.
| Brand name | Description | Manufacturer | E (GPa) | Flex. strength S (MPa) ISO 6872 | Fracture toughness K Ic (MPa√m) |
|---|---|---|---|---|---|
| IPS e.max CAD (LIT) | Lithium disilicate glass-ceramic | Ivoclar Vivadent Schaan, FL | 95 | 360 | 2.2 |
| IPS Empress CAD (LEU) | Leucite glass-ceramic | Ivoclar Vivadent Schaan, FL | 62 | 160 | 1.3 |
| Vita Mark II (VM2) | Feldspar porcelain | Vita Zahnfabrik Bad Säckingen, D | 63 | 113 | 1.7 |
| Enamic (EN) | Feldspar ceramic- polymer infiltrated | Vita Zahnfabrik Bad Säckingen, D | 30 | 155 | 1.5 |
| Lava Ultimate (LU) | Resin composite reinforced with nano ceramics | 3 M, Seefeld, D | 13 | 205 | 2.0 |
2.2
Specimen preparation
Pairs of 3 × 4 × 12 mm slices where cut from the original CAD-CAM blocs using a low speed rotatory fine cutting diamond disc under cooling water in an ISOMET machine (Buehler). All the CAD-CAM materials were already fully dense except lithium disilicate glass-ceramic (e.max CAD) provided in a partially crystalized state by the manufacturer and for which the cut specimens had to undergo additional sintering for dense crystallization. For the bonded interface technique, the cut pairs of specimens were tightly mounted together with sticky wax onto a metal support following a three step procedure as described below. In step 1, surface A (top) ( Fig. 1 ) was ground with hand pressure using diamond discs of decreasing grit size (75, 54, 18 μm) mounted on a turntable (RotoPol 22, Struers) rotating at 300 rpm with water irrigation for 5 min per grit. The ground surfaces were further polished with diamond pastes of 6, 3 and 1 μm at 150 rpm rotation speed for 3 min each paste. In step 2, both surfaces A were bonded together exposing surface B (step 3) for receiving the same grinding sequence as described in step 1. In step 4, pairs were again bonded together by their surface B, exposing the mirror polished surfaces A facing the top. Grinding damage of respectively 75, 54 and 18 μm diamond discs was performed on individual sets of pairs, ground longitudinally (i.e. parallel to the bonded interface) on their mirror polished surface A for 3 min at 150 rpm under water irrigation using hand pressure by one and same operator. After grinding, the bonded pairs were detached, cleaned ultrasonically in a 16% sodium hypochlorite solution for 3 min, a routine cleaning procedure in our laboratory before fractographic SEM analysis. The resulting damage on top surface A and interface B was viewed under the SEM and represented schematically in Fig. 1 showing edge chips and cracks running parallel or orthogonal crack to the ground surface.
2.3
SEM damage evaluation
For each material tested, a pair (N = 2) of specimen selected ground conditions (i.e. 75, 54 and 18 μm) was viewed under the SEM for surface and chip damage evaluation. Chip sizes were measured on the bonded surface B and the presence of extending cracks searched within the chip fractured surface. The top ground surface (A) was analyzed to assess the abrasion mode (ductile or brittle).
2.4
Stress estimates and loss of strength
Fig. 2 illustrates an example of a chip-damage on the bonded interface (B). The chip’s depth ( a ) and width (2c) is assessed over two thirds of the specimen’s length. Mean ( a mean ) and maximum crack depth ( a max ) were assessed for a minimum of 15 and maximum of 30 localized chips. The stress intensity shape factor Y was estimated from chip depth and width ratio . Hence, a semi-ellipse with c = 2 a will have a Y = 1.6. Using the general relationship of K Ic = Y σ f √ a (Eq. (1)) and manufacturer’s K Ic values ( Table 1 ), stress estimates were calculated assuming that the chip depth ( a ) is a strength limiting flaw. The estimated stress ( σ f ) for a given chip size was then compared to the strength values reported by the manufacturer ( Table 1 ) for each material and percentage of possible loss in strength reported.
2.5
Statistical analysis
One-way ANOVA was used to compare chip size damage. Fischer’s least significant difference test at 95% level of significance (p < 0.05) was performed for chip size differences among materials and diamond grit.
3
Results
3.1
Grinding damage evaluation on the bonded interface
For each material tested, average chip sizes measured on the bonded interface after 75 μm, 54 μm or 18 μm diamond grit grinding are summarized in Table 2 as well as in Fig. 3 . The high standard deviation encountered in all materials and for all three diamond grits is a reflection of the large variation in chip sizes formed within the same pairs of specimens. As expected, coarser diamond grits led to deeper chip damage. Hence, grinding with 75 μm diamond grain size on LIT, LEU and VM2 showed the formation of similar semi-elliptical chip defects reaching in average 100 μm in depths with no significant difference among these three ceramics. Table 3 which will be discussed further later lists maximum and average chip depths ( a max , a mean ) with corresponding Y, estimated stress σ at failure (MPa) and loss of strength (%). The deepest chips made with 75 μm diamond reached 176 μm for LIT, 159 μm for LEU and 243 μm for VM2. Softer materials (lower modulus of elasticity) such as EN and LU chipped significantly less with average chip depths of respectively 26 μm and 17 μm. Grinding with 54 μm diamond induced still important chip depths for LIT averaging 84 μm which was significantly higher than all the other materials. As expected, 18 μm diamond grinding induced significantly smaller chip damage than the rougher diamonds. The softer composite material, LU showed the smallest chips of all the materials tested. A graphical representation ( Fig. 3 ) of the grinding induced chip damage as a function of diamond grain size shows very similar behavior for the leucite glass-ceramic (Empress CAD) and the feldspar ceramic (Vita Mark 2). Lithium disilicate (e.max CAD) was highly susceptible to grinding induced chipping and significantly different than all the other materials after 54 and 18 μm diamond grinding ( Fig. 3 ).
| Ceramic brand | Damage 75 μm diamond grinding | Damage 54 μm diamond grinding | Damage 18 μm diamond grinding |
|---|---|---|---|
| Mean ± SD (μm) | Mean ± SD (μm) | Mean ± SD (μm) | |
| e.max CAD (LIT) | 100.2 ± 35.9 b,1 | 84.1 ± 44.7 c,1 | 42.8 ± 13.8 d,2 |
| Empress CAD (LEU) | 94.2 ± 34.0 b,1 | 59.6 ± 19.4 b,2 | 15.1 ± 7.6 c,3 |
| Vita Mark II (VM2) | 106.5 ± 53.1 b,1 | 51.3 ± 12.8 b,2 | 10.2 ± 3.6 b,3 |
| Enamic (EN) | 25.8 ± 10.8 a,1 | 18.3 ± 6.3 a,2 | 7.0 ± 1.3 a,b,3 |
| Lava Ultimate (LU) | 16.6 ± 7.5 a,1 | 7.1 ± 2.0 a,2 | 4.8 ± 1.5 a,2 |
Stay updated, free dental videos. Join our Telegram channel
VIDEdental - Online dental courses
