RESEARCH AND EDUCATION

Color stability of lithium disilicate ceramics after aging and

immersion in common beverages
Eleni-Sotiria Palla, DDS,a Eleana Kontonasaki, DDS, MS, PhD,b Nikolaos Kantiranis, DSc, PhD,c
Lambrini Papadopoulou, DSc, PhD,d Triantafyllia Zorba, DSc, PhD,e
Konstantinos M. Paraskevopoulos, DSc, PhD,f and Petros Koidis, DDS, MS, PhDg

ABSTRACT
Statement of problem. The color of an esthetic restoration and its color stability are important for long-term success. However, the impact of
common beverages on lithium disilicate ceramic is not well known.
Purpose. The purpose of this in vitro study was to investigate color variations of lithium disilicate ceramics after thermal cycling (TC) and
immersion in commonly consumed beverages.
Material and methods. A total of 288 specimens (1×10×10 mm) were fabricated from IPS e-max computer-aided design (CAD) (n=72), IPS
e-max CERAM ([CER] n=72), IPS e-max Press with glazing ([PG] n=72), and IPS e-max Press without glazing ([PNG] n=72) according to the
manufacturer’s instructions. Each group was divided into 4 subgroups (n=18): TC, coffee, black tea, and red wine. Thermal cycling was
performed at 21 900 cycles at 5! C, 37! C, 55! C, and 37! C (3 years’ clinical simulation), whereas the specimens were soaked in the staining
solutions for up to 54 hours (3 years’ clinical simulation). Color parameters L*, a*, and b* were assessed with an ultraviolet-visible
spectroscopy recording spectrophotometer. Color difference (DE) was calculated using the equation [DE= [(DL*)2 + (Da*)2 + (Db*)2]½].
Intraexaminer reliability was assessed by using the intraclass correlation coefficient. Two-way analysis of variance was used for the
analysis of DE, and the parameters L*, a*, and b* were analyzed with linear mixed models for repeated measurements and the Bonferroni
pair-wise comparison test (a=.05).
Results. Parameters DE, L*, a*, and b* were significantly affected by the interaction between material and treatment (P<.001). A DE >1 was
recorded for PG with tea, wine, and coffee, PG after TC and CER after TC. For L*, greater reduction was observed for PNG with tea and CER after
TC, whereas for a* significant changes were positive (to red shades), except for PNG with TC, where PNG with wine showed the greatest
positive change. For b*, significant changes were negative (to blue shades) except for PNG with tea and coffee and CAD with tea.
Conclusions. All groups demonstrated color changes below the clinically perceptible level (DE<3.7), except PNG in tea which showed sta-
tistically significant color differences (DE>4). CAD presented higher color stability compared with the nonglazed Press specimens. (J Prosthet
Dent 2017;-:---)

Changes in restorative treatment patterns, as well as the esthetic dentistry. The color of an esthetic restoration and
introduction of new and improved restorative materials its stability is important to its long-term success. Ce-
and techniques, have led to an increased demand for ramics are popular restorative materials because of their

Supported by the State Scholarships Foundation from the resources of the European program Education and Lifelong Learning of the European Social Fund of
the NSRF 2007-2013. Ivoclar Vivadent provided IPS e-max CAD and IPS e-max Press blocks. Presented orally at the 40th European Prosthodontic Association (EPA)
and 65th German Society for Prosthetic Dentistry and Biomaterials Congress, Halle, Germany, September 15-17, 2016 (won EPA Oral Presentation Prize).
a
Postgraduate student, Scholar of State Scholarships Foundation, School of Health Sciences, Department of Dentistry, Section of Prosthodontics, Aristotle University of
Thessaloniki, Thessaloniki, Greece.
b
Assistant Professor, School of Health Sciences, Department of Dentistry, Section of Prosthodontics, Aristotle University of Thessaloniki, Thessaloniki, Greece.
c
Assistant Professor, School of Geology, Aristotle University of Thessaloniki, Thessaloniki, Greece.
d
Specialized and Laboratory Teaching Staff, School of Physics, Aristotle University of Thessaloniki, Thessaloniki, Greece.
e
Professor, School of Physics, Aristotle University of Thessaloniki, Thessaloniki, Greece.
f
Professor, School of Health Sciences, Department of Dentistry, Section of Prosthodontics, Aristotle University of Thessaloniki, Thessaloniki, Greece.
g
Professor, Department of Fixed Prosthesis & Implant Prosthodontics, Aristotle University of Thessaloniki, Thessaloniki, Greece.

THE JOURNAL OF PROSTHETIC DENTISTRY 1

2 Volume
Clinical Implications
Nonglazed surfaces, such as after crown
recontouring or occlusal adjustments, should be
avoided by carefully designing the restorations and
ensuring that the surfaces are glazed before
definitive cementation. Compared with the Press,
the computer-aided design and computer-aided
manufacturing lithium disilicate seems to be a more
color stable material.
excellent esthetics,1 biocompatibility, and enhanced
strength and because their surface smoothness facilitates
plaque control.2,3 Color in ceramics is generally affected
by intrinsic factors such as the ceramic composition and
the glaze layer, whereas dietary habits, food, and liquid
colorants, as well as oral hygienic agents are extrinsic
factors that can influence color stability.4-11
Lithium disilicate has become popular for esthetic
rehabilitation. Preserving the color of lithium disilicate is
an essential determinant of the long-term clinical success
of such restorations.
Glazing before definitive cementation is important to
the color stability and stain resistance of ceramic resto-
rations.12-18 However, occlusal adjustments after the
cementation of ceramic restorations result in removal of
the glazed layer, thus increasing the surface roughness.
Rough surfaces reduce the amount of reflected light and
consequently affect the color of the restoration13-15 while
increasing extrinsic staining.16,19,20
Artificial accelerated aging with either ultraviolet (UV)
light or temperature and humidity variations has been
used to assess the color stability of various dental ma-
terials by simulating clinical conditions.21-23 Thermal
cycling is a popular in vitro procedure that simulates the
oral environment by causing artificial accelerated aging.
This method includes standardized thermal variations in
a humid environment with deionized water, which sim-
ulates the thermal variations in the oral cavity.24-29
The color of both teeth and esthetic restorations
can be evaluated with spectrophotometers or colori-
meters.7-10,30-33 Spectrophotometers measure the wave-
length that is reflected or transmitted from one object at a
time, without being affected by the subjective in-
terferences of the color,14,15,34-44 whereas colorimeters
provide an overall measurement of the light absorbed.
The Commission Internationale de l’Eclairage L*a*b*
(CIELab) system measures chromaticity and defines the
color of an object in a uniform 3-dimensional space. Color
difference (DE) is calculated through differences in the
color coordinates L*, a*, and b*.45 If DE=0, the material
tested is considered color stable, whereas values of DE
between 0 and 2 suggest negligible color differences5,46-49;
THE JOURNAL OF PROSTHETIC DENTISTRY
- Issue -
however, even DE values >0.4 may be perceivable by a
trained human observer.48
Although DE values between 3 and 5 have been
considered clinically acceptable, they are significant
because they are perceptible by even untrained ob-
servers. Finally, DE values of more than 5 are not
acceptable, suggesting the restoration requires replace-
ment.50-52 A DE value of >3.3 or 3.7 has been considered
unacceptable in various investigations,6-8,30,35,43,53-56
representing a clinically distinguishable color difference
in the intraoral environment. However, more recently,
Douglas et al52 used a regression model to predict that a
color difference of 2.6 DE units viewed in a patient’s
mouth would be perceptible to 50% of observing
dentists.
The purpose of this in vitro study was to investigate
the color variations of lithium disilicate ceramic speci-
mens with different manufacturing methods (computer-
aided design [CAD], Press, Press with glaze, and
Ceram) after TC and immersion in coloring solutions
and to qualitatively evaluate the relationship between
surface structure and color change. The null hypotheses
were that lithium disilicate ceramic groups would not
present color changes higher than the intraorally
perceivable DE value of 3.7, either after TC or immersion
in staining solutions and that no color differences would
be found in terms of DL*, Da*, Db*, and DE among the
ceramic groups after being subjected to the same aging
processes.
MATERIAL AND METHODS
A total of 288 square specimens (10×10×1 mm) were
fabricated in 4 groups: CAD group (n=72) from IPS
e-max CAD HTA2/I 12; Press without glazing (PNG)
group (n=72) as received from IPS e-max Press HT A2;
Press with glazing (PG) group (n=72) glazed specimens
of IPS e-max Press HT A2; and CER group (n=72) from
IPS e-max CERAM deep dentin A2/TI.
IPS e-max CAD blocks were sectioned in a cutting
device (Accutom-2; Struers) with an Al2O3 wheel at 300
to 3000 rpm under water spray. The specimens were
finished and polished with a series of silicon carbide
(SiC) papers and fired according to the manufacturer’s
instructions in a furnace (Programmat P300; Ivoclar
Vivadent AG) at 90! C/min with a final temperature of
820! C, in a vacuum at 415! C, with vacuum released at
850! C. For the IPS e-max Press groups, the specimens
were prepared from a square wax pattern (10×10×1 mm).
They were then heat pressed (IPS Empress EP600; Ivoclar
Vivadent AG) at 920! C. Half of them were used to create
the PG specimens, which were coated on one side with
a layer of neutral shade glaze and fired at 740! C (tem-
perature rate of 70! C/min; in a vacuum at 450! C, with
vacuum released at 720! C). The CER specimens were
Palla et al
- 2017 3

**
7

∆E
6

**
Mean + SD

4 *
**
**
**
3
*

2 *

0
Thermal cycling Coffee Tea Wine
CAD 0.85 0.63 0.67 0.7
PG 1.39 0.61 0.59 0.55
PNG 0.85 1.1 4.99 1.85
CER 1.61 1.31 0.77 0.76

Figure 1. Mean ±SD color difference (DE) values. Statistically significant differences are *P<.05 and **P".001. CAD, computer-aided design; CER,
CERAM; PG, Press with glazing; PNG, Press without glazing.

Table 1. Two-way ANOVA results for DE Table 2. Results of linear mixed model (algorithm MIXED) for a* (type III
Type III Partial Eta tests of fixed effects)
Sum of Mean Squared Numerator Denominator
Source Squares df Square F P (h2) Source df df F P
Corrected model 100.209 15 6.681 18.114 <.001 0.5 Intercept 1 272 31 653.363 <.001
Intercept 10.702 1 10.702 29.017 <.001 0.096 Time (pre-post) 1 272 449.113 <.001
Material (CAD, PG, 44.794 3 14.931 40.486 <.001 0.309 Material (CAD, PG, 3 272 3 412.205 <.001
PNG, CER) PNG, CER)
Treatment 4.204 3 1.401 3.799 .011 0.04 Treatment (TC, W, C, T) 3 272 46.689 <.001
(TC, W, C, T)
Material×time 3 272 78.952 <.001
Material×treatment 51.211 9 5.69 15.428 <.001 0.338
Treatment×time 3 272 143.795 <.001
Error 100.316 272 0.369
Material×treatment 9 272 23.015 <.001
Total 211.226 288
Material×treatment×time 9 272 63.157 <.001
Corrected total 200.524 287
CAD, computer-aided design; CER, CERAM; PG, Press with glazing; PNG, Press without
CAD, computer-aided design; CER, CERAM; PG, Press with glazing; PNG, Press without
glazing; TC, thermal cycling; W, red wine; C, coffee; T, black tea.
glazing; TC, thermal cycling; W, red wine; C, coffee; T, black tea.

fabricated by mixing ceramic powder (0.35 g of IPS e-max
CERAM deep dentin A2) with modeling liquid (0.2 g of
IPS e.max Ceram Build-Up Liquid). The slurry produced
was transferred into polyvinyl siloxane (Virtual; Ivoclar
Vivadent AG) molds and condensed using a vibrator to
remove water. The specimens were removed with gentle
hand pressure, sintered, and autoglazed according to the
manufacturer’s instructions (temperature rate of 50! C/
min, at a final temperature of 800! C, under vacuum at
450! C, with the vacuum released at 799! C).
Digital calipers were used to measure and verify the
thickness of all specimens. All specimens were cleaned in
an ultrasonic cleaner with distilled water for 10 minutes.
Then they were stored in an incubator (37 ±1! C, 5% CO2,
100% humidity) until the color parameters were
measured. Each group was divided into 4 subgroups,
depending on the type of aging process the specimens
would be subjected to (thermal cycling [TC], wine [W];
coffee [C]; tea [T]).
For TC, the specimens were subjected to repeated
cycles of hot and cold in water baths according to the
following sequence: 5! C to 37! C to 55! C to 37! C. In total,
21 900 cycles were completed that corresponded to 3
years of exposure in the oral cavity.25 Coffee (Jacobs
Kronung; the Kraft Heinz Co), black T (Lipton tea;
Unilever), and red W (Naoussa; Boutari) were used as the
staining liquids. Coffee and T were prepared according to
the manufacturer’s instructions (15 g in 300 mL of boiled,
distilled water). Staining liquids were prepared once,
using the same solution for each specimen. Specimens
were immersed individually in the staining liquids in 48-
well cell culture plates and remained soaking inside an
incubator for up to 54 hours. The incubation period was
selected based on the following correlation: 3 Cs per day

Palla et al THE JOURNAL OF PROSTHETIC DENTISTRY

4 Volume - Issue -

Table 3. Results of linear mixed model (algorithm MIXED) for b* (type III Table 4. Results of linear mixed model (algorithm MIXED) for L** (type III
tests of fixed effects) tests of fixed effects)
Numerator Denominator Numerator Denominator
Source df df F P Source df df F P
Intercept 1 272 73 999.53 <.001 Intercept 1 272 1 526 952.475 <.001
Time (pre-post) 1 272 39.023 <.001 Time (pre-post) 1 272 0.626 .429
Material (CAD, PG, 3 272 4479.245 <.001 Material (CAD, PG, 3 272 926.631 <.001
PNG, CER) PNG, CER)
Treatment (TC, W, C, T) 3 272 38.728 <.001 Treatment (TC, W, C, T) 3 272 5.212 .002
Material×time 3 272 92.72 <.001 Material×time 3 272 4.07 .007
Treatment×time 3 272 120.379 <.001 Treatment×time 3 272 13.989 <.001
Material×treatment 9 272 23.788 <.001 Material×treatment 9 272 6.086 <.001
Material×treatment×time 9 272 80.879 <.001 Material×treatment×time 9 272 5.104 <.001
CAD, computer-aided design; CER, CERAM; PG, Press with glazing; PNG, Press without CAD, computer-aided design; CER, CERAM; PG, Press with glazing; PNG, Press without
glazing; TC, thermal cycling; W, red wine; C, coffee; T, black tea. glazing; TC, thermal cycling; W, red wine; C, coffee; T, black tea.

100 100 **
L* ** L*
90 90 **
80 80 *
*
Mean ± SD

Mean ± SD
70 70
60 60
50 50
40 40
30 30
20 20
10 10
0 0
Thermal Coffee Tea Wine Thermal Coffee Tea Wine
cycling cycling
CAD PNG
Time pre 66.59 67.74 67.43 67.39 Time pre 73.95 73.51 72.52 73.34
Time post 67.34 67.58 67.18 66.89 Time post 74.13 74.22 71.06 73.19

100 100
L* L*
90 90
80 80
Mean ± SD

Mean ± SD

70 70
60 60
50 50
40 40
30 30
20 20
10 10
0 0
Thermal Coffee Tea Wine Thermal Coffee Tea Wine
cycling cycling
CER PG
Time pre 72.64 73.23 72.92 72.64 Time pre 67.04 66.9 67.31 66.86
Time post 73.61 73 73.19 73.08 Time post 67.6 66.67 67.06 66.92

Figure 2. Mean ±SD L* before (Pre) and after (Post) each aging test. Statistically significant differences are *,# for P<.05 and **,## for P<.001. Red
symbols are assigned to pre and post comparisons within each aging test, while black symbols to pair-wise comparisons between aging tests. CAD,
computer-aided design; CER, CERAM; PG, Press with glazing; PNG, Press without glazing.

corresponding to 1 minute of exposure per cup, which
equals a total of 1095 minutes or 18 hours of exposure per
year.4 After immersion, the specimens were washed in
distilled water, dried with filter paper, and subjected to
color determination.
Color data were recorded using a ultraviolet-visible
spectroscopy recording spectrophotometer (UV-2401 pc;
Shimadzu Co) with software (UVPROBE v2.21; Shi-
madzu Co). Data were retrieved relative to the illuminant
C standard, corresponding to average daylight with
2-degrees observation in the reflectance mode on a black
background. The spectrophotometer was equipped with
an ISR-240 A integrating sphere, with a 7-mm opening
directing the light at the specimen at an angle of 0 de-
grees. Color changes were examined using the CIELab
color system before and after each aging test. This system
consists of lightness (L*), red-green (a*) and yellow-blue
(b*) coordinates. The CIELab DE values were calculated
to provide a quantitative measurement of color change
by using the following equation: [DE=([DL*]2 + [Da*]2 +

THE JOURNAL OF PROSTHETIC DENTISTRY Palla et al

- 2017 5

0.5 0.5
a* a*
0 0
Mean ± SD

Mean ± SD
–0.5 –0.5
–1 –1
–1.5 –1.5
**
–2
** –2 ** ***
** **
–2.5 ** –2.5
Thermal Coffee Tea Wine Thermal Coffee Tea Wine
cycling cycling
CAD PNG
Time pre –2.02 –2.07 –2.08 –2.02 Time pre –1.48 –1.53 –1.44 –1.52
Time post –1.98 –1.92 –1.97 –1.69 Time post –1.7 –1.32 –1.06 –0.32

0.5 0.5
a* a*
0
0.25
Mean ± SD

Mean ± SD
–0.5
0 –1
–1.5
–0.25
–2
* **
–0.5 –2.5
Thermal Coffee Tea Wine Thermal Coffee Tea Wine
cycling cycling
CER PG
Time pre –0.07 –0.19 –0.08 –0.11 Time pre –1.74 –1.81 –1.8 –1.79
Time post –0.03 –0.18 –0.08 0.02 Time post –1.67 –1.66 –1.73 –1.6

Figure 3. Mean ±SD a* values before (Pre) and after (Post) each aging test. Statistically significant differences are *,# for P<.05 and **, ## for P<.001. Red
symbols are assigned to pre and post comparisons within each aging test, black symbols to pair-wise comparisons between aging tests. CAD,
computer-aided design; CER, CERAM; PG, Press with glazing; PNG, Press without glazing.

[Db*]2)½], where DL*=L1−L0; Da*=a1−a0; and
Db*=b1−b0. Terms L0, a0, and b0 were the color data
initially measured.
The surface was characterized using X-ray diffraction
(XRD) analysis and Fourier transform infrared (FTIR)
spectroscopy to reveal any structural changes on the
surfaces of the specimens after each aging test.57-60 X-ray
diffraction analysis was carried out using a diffractometer
(PW1710; Philips) with Ni-filtered CuKa radiation.
Diffraction patterns were obtained from 0 to 75 degrees
at a scan speed of 0.008 degrees/min. The FTIR reflec-
tance spectra of the specimens were obtained in the
spectral area of mid- and far-infrared (4000 to 120 cm−1)
by using an FTIR extended spectrometer (IFS113v;
Bruker) with a 2-cm−1 resolution. Topographical evalua-
tion was performed by scanning electron microscopy
(JSM 840A; Jeol).
The intraexaminer reliability concerning color mea-
surements was assessed in a pilot study with 5 speci-
mens. The intraclass correlation coefficient and method
error were computed by means of limits of disagreement
using software (MedCalc v12.5.0.0; MedCalc Software).
Two-way analysis of variance (ANOVA) was used for the
analysis of DE; the parameters L*, a* and b* were
analyzed using linear mixed models for repeated
measurements (SPSS Statistics v20.0; IBM Corp), and the
Bonferroni method was used for pair-wise comparisons
(a=.05 for all tests). Adequacy of the models was exam-
ined by inspecting the residuals and the predicted values.
Power analysis concerning differences in DE was per-
formed using software (G*Power v3.1; Heinrich-Heine-
Universität Düsseldorf). For 2-way ANOVA, a=.05 and
h2=0.1; with 288 specimens, the power was found to be
0.99 for both main effects and 2-way interactions be-
tween material and treatment.
RESULTS
The mean and standard deviation DE values were
significantly affected by the interaction between material
and treatment (P<.001) (Fig. 1). The effect of the aging
process on the color change varied among the groups.
For the CAD group, DE<1 was recorded for all specimens
regardless of the aging process applied. A DE value of
1.39>1 was recorded for PG specimens only after TC. The
CER group presented significant color changes after TC
(DE=1.61>1) and after immersion in C (DE=1.31>1). The
most chromatically unstable material was the PNG
group. It presented the higher color difference
(DE=4.99>1) after immersion in T. Color changes were

Palla et al THE JOURNAL OF PROSTHETIC DENTISTRY

6 Volume - Issue -

20 20 **
b* b*
18 18
16 16 *
**
Mean ± SD

Mean ± SD
14 14
12 12
10 10 *
8 8
6 6
4 4
2 2
0 0
Thermal Coffee Tea Wine Thermal Coffee Tea Wine
cycling cycling
CAD PNG
Time pre 6.31 6.34 6.17 5.96 Time pre 6.9 7.17 7.62 7.5
Time post 6.02 6.39 6.5 5.95 Time post 7.05 7.54 12.3 6.79

20 20
* b* b*
18 18
16 16

Mean ± SD
Mean ± SD

14 14
12 12
10 10
8 8
6 6
4 4
2 2
0 0
Thermal Coffee Tea Wine Thermal Coffee Tea Wine
cycling cycling
CER PG
Time pre 16.57 16.18 16.54 16.59 Time pre 7.77 7.27 7.64 7.33
Time post 16.58 15.84 16.37 16.32 Time post 7.26 7.27 7.76 7.14

Figure 4. Mean ±SD b* values before (Pre) and after (Post) each aging test. Statistically significant differences are *,# for P<.05 and **, ## for P<.001. Red
symbols are assigned to pre and post comparisons within each aging test, while black symbols to pair-wise comparisons between aging tests. CAD,
computer-aided design; CER, CERAM; PG, Press with glazing; PNG, Press without glazing.

also high after immersion in W (DE=1.85>1) and C
(DE=1.1>1). The 2-way ANOVA results for DE are pre-
sented in Table 1.
Results of the mixed linear model for the specific color
parameters L*, a*, and b* are presented in Tables 2-4,
whereas mean ±SD values and statistically significant
differences from the pair-wise comparisons are presented
in Figures 2-4. Mean ±SD differences for lightness, L*,
the greater reduction was observed for PNG with T
(mean, −1.46 ±0.23; P<.001) and for CER after TC (mean,
0.97 ±0.23; P<.001). The important color change for PNG
with T was in a negative direction along with the smaller
color change for CAD with W (mean, −0.5 ±0.23;
P=.006). For the a* parameter, significant changes were
positive (change from green to red shades) except for
PNG with TC (mean, −0.23 ±0.03; P<.001). The greatest
change for a* was found for the PNG group with W
(mean, 1.20 ±0.03; P<.010). Finally, for b*, significant
changes were negative (change from yellow to blue
shades) except for PNG with T (mean, 4.68 ±0.13;
P<.001) and C (mean, 0.37 ±0.13; P=.005) and CAD with
T (mean, 0.33 ±0.13; P=.010). The most significant pos-
itive color change of b*was observed for PNG after
immersion in T.
Surface analysis with FTIR (Fig. 5) and XRD (Fig. 6)
demonstrated no significant alterations on the surface
structure of the CAD and CER specimens, either after
TC or immersion in T, C, or W (Fig. 5A, C, Fig. 6A, C).
For CAD specimens, the characteristic peaks at 473, 554,
635, 753, 782, 1021, and 1110 wave numbers (cm−1)
were attributed to the dominant phase of lithium dis-
ilicate (Li2Si2O5)56 as verified also by XRD (w85%wt),
whereas the peaks at 933 and 960 cm−1 indicate the
coexistence of a small amount of lithium silicate
(Li4SiO4) as verified by XRD (w1%wt).57-59 The major
crystalline phase for the PNG specimens was also
lithium disilicate, whereas the peaks at 759 and 1047
cm−1, and the shoulder at 995 cm−1 in the XRD patterns
reveals the coexistence of lithium trisilicate (Li2Si3O7).57
The CER specimens were highly amorphous (95-97%
wt). X-ray diffraction demonstrated a reduction of the
surface crystallinity of the PG specimens after W and TC
(41% and 49%wt, respectively, from 52.5%wt) but no
other alteration. For the PNG specimens, FTIR
demonstrated that immersion in beverages resulted in a
significant decrease of the reflectance peaks’ intensity,
broadening of the respective peaks, and a slight shifting
to lower wave numbers, a finding that was further

THE JOURNAL OF PROSTHETIC DENTISTRY Palla et al

- 2017 7

CAD control CAD coffee CAD tea PNG control PNG coffee PNG tea
CAD wine CAD TC PNG wine PNG TC
Reflectance

Reflectance
1400 1200 1000 800 600 400 1400 1200 1000 800 600 400
Wavenumbers (cm–1) Wavenumbers (cm–1)
A B

CER control CER coffee CER tea PG control PG coffee PG tea

CER wine CER TC PG wine PG TC
Reflectance

Reflectance

1400 1200 1000 800 600 400 1400 1200 1000 800 600 400
Wavenumbers (cm–1) Wavenumbers (cm–1)
C D
Figure 5. FTIR spectra of representative specimens from each subgroup. A, CAD. B, PNG. C, CER. D, PG specimens. CAD, computer-aided design; CER,
CERAM; FTIR, Fourier transform infrared spectroscopy; PG, Press with glazing; PNG, Press without glazing.

confirmed by XRD by the slightly higher amounts of the
amorphous phase.
Representative backscattered scanning electron mi-
crophotographs of all specimens before and after each
aging procedure are presented in Figure 7. Significant
morphological changes of the surface of each specimen
were not observed for any group after aging. The PG
and CAD specimens presented smooth surfaces,
whereas the roughest surface was observed for the PNG
specimens.
DISCUSSION
Based on results of the present investigation, the first null
hypothesis was accepted for the CAD, PG, and CER
groups but rejected for the PNG group, and the second
was rejected in most cases and accepted in a few. All
aging processes (TC and staining solutions) affected the
final color of the lithium disilicate specimens. However,
almost all groups, independently of the aging process,
presented with DE values below 2, suggesting acceptable
color stability, except for the PNG group after immersion
in T, which presented the highest color difference,
DE=4.99>3.7, and which was regarded as clinically
unacceptable.20
In the present in vitro study, a spectrophotometer was
selected because of its ability to produce reproduc-
ible measurements free of the subjective effect of
color.15,35,37,43 The device’s reproducibility in the present
study was evaluated from 5 specimens of each group by
calculating the DE value after 2 color measurements of
the same surface at different time intervals (2 hours). The
threshold was set at DE<1, as this was the upper limit for
reproducible measurements according to the spectro-
photometer’s manufacturer. All measurements were well
below the threshold (0.22 for the CAD, 0.24 for the CER,
0.58 for the PNG, and 0.32 for the PG), suggesting a high
degree of reproducibility.

Palla et al THE JOURNAL OF PROSTHETIC DENTISTRY

8 Volume - Issue -

Figure 6. X-ray diffraction patterns of representative specimens from each subgroup. CAD, computer-aided design; CER, CERAM; PG, Press with glazing;
PNG, Press without glazing.

THE JOURNAL OF PROSTHETIC DENTISTRY Palla et al

- 2017
Figure 7. Representative backscattered scanning electron micrographs of specimens before and after each aging procedure (original magnifica-
tion ×500). CAD, computer-aided design; CER, CERAM; PG, Press with glazing; PNG, Press without glazing.
A variety of aging methods can be applied to inves-
tigate the color stability of dental ceramics. Most methods
refer to extrinsic factors, including exposure to environ-
mental conditions. In previous investigations, ambient
and ultraviolet irradiation (UV-B light)55; food and drink
colorants such as C, T, red W, cola, orange juice4-10,44;
and other pigments such as salbutamol sulfate54 and
methylene blue16,19,48 have been tested. However, clin-
ical discoloration from commonly consumed beverages
such as C, T, W, and orange juice are known to be more
detrimental than ultraviolet light or distilled water
alone.11,44 Thus, in the present investigation, we
preferred immersion in commonly consumed staining
solutions, including C, T, and red W as an aging method,
along with TC, which is one of the most widely used
procedures for simulating the physiological aging of
biomaterials.
Numerous in vitro studies that used TC devices to test
dental materials’ resistance to aging25-29 have been
published following the review by Gale and Darwell24
more than 10 years ago. However, great variation exists
among studies, and no standardized protocol exists for
the number of appropriate cycles for aging of the various
dental materials. Accordingly, in the present investiga-
tion, the number of cycles was selected based on the
Palla et al
9
initial study by Gale and Darwell24 and on the logical
assumption that the number of cycles when the thermal
gradient occurring in vivo is 20 per day, leading to 7300
per year and 21 900 over 3 years. Furthermore, for an
optimized clinical situation, the thermal sequence used
included the reference in vivo resting temperature of
37! C.
Often minor occlusal adjustment and/or crown
recontouring is required after cementation. These ad-
justments can remove the surface glaze, leaving a
roughened surface capable of absorbing more colorants
and having inferior ability to reflect the light. This ex-
plains the observation that the unglazed e-max Press
(PNG) specimens presented the highest color change
after immersion in T and W than in all other groups (T,
DE=4.99; W, DE=1.85). Although restorations should be
polished after occlusal adjustment, glazing yields a
smoother ceramic surface than other finishing
methods.39
The lowest DE value after all aging processes occurred
for the CAD group, followed by the PG group. This
finding was accompanied by a lack of surface structural
changes, as recorded by both XRD and FTIR, suggesting
a superior stability of the material, and especially CAD
lithium disilicate, which was the least affected by
THE JOURNAL OF PROSTHETIC DENTISTRY
10
handling during the manufacturing process. In the PNG
group, the rough surface might have facilitated water
penetration and the consequent silica network dissolu-
tion, resulting in a reduction of crystallinity and greater
absorption of coloring pigments. Furthermore, the pres-
ence of other than lithium disilicate phases, such as
lithium trisilicate, as revealed by XRD, might have
induced additional grain boundaries, facilitating water
diffusion and rendering the material prone to dissolution.
On the contrary, the glazed surfaces of the PG specimens
might have healed surface irregularities or microcracks,
prohibiting water diffusion and subsequent silica network
dissolution.
Limitations of the present in vitro study include
limited investigation time. Immersion of 54 hours
equaled 3 years in the oral cavity, representing a limited
lifespan for a prosthetic restoration and restricted clinical
simulation. As TC was conducted in water and not in
saliva, there was no UV exposure, and only standard
temperatures were investigated. Finally, the additive ef-
fect of other common aging procedures such as tooth
brushing was not investigated. Future investigations
should explore color stability for prolonged aging times
and in other beverages such as carbonated beverages or
fruit juices, whereas for optimized clinical simulation, the
combined effect of other contributing factors such as
sunlight, presence of saliva proteins/enzymes, and
smoking also should be evaluated.
CONCLUSIONS
Within the limitations of this in vitro study, the following
conclusions were drawn:
1. The majority of specimens presented slight color
changes (DE<2).
2. Only the nonglazed pressed lithium disilicate group
after immersion in tea presented clinically unac-
ceptable color change (DE=4.99>3.7), which was
accompanied by a rough surface, reduced reflec-
tance, and surface crystallinity after immersion.
3. Glaze is important for the color stability of IPS
e-max Press restorations. The presence of non-
glazed surfaces, as after excessive occlusal adjust-
ments, should be avoided by careful design of the
restorations and thorough occlusal checking prior to
definitive cementation.
4. Computer-aided design and computer-aided
manufacturing derived lithium disilicate appears to
be a very color stable material.
REFERENCES
1. Fischer H, Marx R. Improvement of strength parameters of a leucite-
reinforced glass ceramic by dual-ion exchange. J Dent Res 2001;80:336-9.
2. Leinfelder KF. Porcelain esthetics for the 21st century. J Am Dent Assoc
2000;131(suppl):47S-51S.
THE JOURNAL OF PROSTHETIC DENTISTRY
Volume - Issue -
3. Dalkiz M, Sipahi C, Beydemir B. Effects of six surface treatment methods on
the surface roughness of a low-fusing and a ultra low-fusing feldspathic
ceramic materials. J Prosthodont 2009;18:217-22.
4. Gawriołek M, Sikorska E, Ferreira L, Costa A, Khmelinskii I, Krawczyk A,
Sikorski M, Koczorowski R. Color and luminescence stability of selected
dental materials in vitro. J Prosthodont 2012;21:112-22.
5. Karagoz-Motro P, Kursoglou P, Kazazoglu E. Effects of different surface
treatments on stainability of ceramics. J Prosthet Dent 2012;108:231-7.
6. Samra A, Pereira S, Delgado L, Borges C. Color stability evaluation of
aesthetic restorative materials. Braz Oral Res 2008;22:205-10.
7. Mutlu-Sagesen L, Ergun G, Ozkan Y, Bek B. Color stability of different
denture teeth materials an in vitro study. J Oral Science 2001;43:193-205.
8. Ghahramanloo A, Madani A, Sohrabi K, Sabzevari S. An evaluation of color
stability of reinforced composite resin compared with dental porcelain in
commonly consumed beverages. J Calif Dent Assoc 2008;36:673-80.
9. Gunay Y, Ozkan Y, Akyil S, Karayazgan B, Toksoy F. Effect of colored bev-
erages on the color stability of feldspathic porcelain subjected to various
surface treatments. Quintessence Int 2009;40:41-8.
10. Koksal T, Dikbas I. Color stability of different denture teeth materials against
various staining agents. Dent Mater J 2008;27:139-44.
11. Burrow MF, Makinson OF. Color change in light-cures resins exposed to
daylight. Quintessence Int 1991;22:447-52.
12. Jagger DC, Harrison A. An in vitro investigation into the wear effects of
unglazed, glazed, and polished porcelain on human enamel. J Prosthet Dent
1994;72:320-3.
13. Al-Wahadni A, Martin DM. Glazing and finishing dental porcelain: A liter-
ature review. J Can Dent Assoc 1998;64:580-3.
14. Sarac D, Sarac S, Yuzbasıoglu E, Bal S. The effects of porcelain polishing
systems on the color and surface texture of feldspathic porcelain. J Prosthet
Dent 2006;108:122-8.
15. Al-Wahadni AM, Martin DM. An in vitro investigation into the wear effects
of glazed, unglazed and refinished dental porcelain on an opposing material.
J Oral Rehabil 1999;26:538-46.
16. Esquivel JF, Chai J, Wozniak WT. Color stability of low-fusing porcelains for
titanium. Int J Prosthodont 1995;8:479-85.
17. Kawai K, Urano M, Ebisu S. Effect of surface roughness of porcelain on
adhesion of bacteria and their synthesizing glucans. J Prosthet Dent 2000;83:
664-7.
18. Fuzzi M, Zaccheroni Z, Vallania G. Scanning electron microscopy and pro-
filometer evaluation of glazed and polished dental porcelain. Int J Prostho-
dont 1996;9:452-8.
19. Yılmaz C, Korkmaz T, Demirkoprulu H, Ergun G, Ozkan Y. Color stability of
glazed and polished dental porcelains. J Prosthet Dent 2008;17:20-4.
20. Atay A, Karayazgan B, Ozkan Y, Akyil MS. Effect of colored beverages on the
color stability of feldspathic porcelain subjected to various surface treatments.
Quintessence Int 2009;40:e41-8.
21. El-Araby AM, Talic YF. The effect of thermocycling on the adhesion of self-
etching adhesives on dental enamel and dentin. J Contemp Dent Pract
2007;8:17-24.
22. Catalbas B, Uysal T, Nur M, Demir A, Gunduz B. Effects of thermocycling on
the degree of cure of two lingual retainer composites. Dent Mater J 2010;29:
41-6.
23. Özel Bektas Ö, Eren D, Herguner Siso S, Akin GE. Effect of thermocycling on
the bond strength of composite resin to bur and laser treated composite resin.
Lasers Med Sci 2012;27:723-8.
24. Gale MS, Darwell BW. Thermal cycling procedures for laboratory testing of
dental restorations. J Dent 1999;27:89-99.
25. Doerr CL, Hilton TJ, Hermesch CB. Effect of thermocycling on the micro-
leakage of conventional and resin-modified glass ionomers. Am J Dent
1996;9:19-21.
26. Schuckar M, Geurtsen W. Proximo-cervical adaptation of class II-composite
restorations after thermocycling: a quantitative and qualitative study. J Oral
Rehabil 1997;24:766-75.
27. Wegner SM, Gerdes W, Kern M. Effect of different artificial aging conditions
on ceramic-composite bond strength. Int J Prosthodont 2002;15:267-72.
28. Bedran-de-Castro AKB, Pereira PNR, Pimenta LAF, Thompson JY. Effect of
thermal and mechanical load cycling on nanoleakage of Class II restorations.
J Adhes Dent 2004;6:221-6.
29. D’Amario M, Campidoglio M, Morresi AL, Luciani L, Marchetti E, Baldi M.
Effect of thermocycling on the bond strength between dual-cured resin ce-
ments and zirconium-oxide ceramics. J Oral Sci 2010;52:425-30.
30. Saygili G, Sahmali S, Demirel F. Colour stability of porcelain repair materials
with accelerated ageing. J Oral Rehabil 2006;33:387-92.
31. Turgut S, Bagis B. Color stability of laminate veneers: an in vitro study. J Dent
2011;39:57-64.
32. Razzoog M, Lang B, Russell M, May K. A comparison of the color stability of
conventional and titanium dental porcelain. J Prosthet Dent 1994;72:453-6.
33. Dikicier S, Ayyildiz S, Ozen J, Sipahi C. Effect of varying core thicknesses and
artificial aging on the color difference of different all-ceramic materials. Acta
Odontol Scand 2014;72:623-9.
34. Paul S, Peter A, Pietroban N, Hammerle CHF. Visual and spectrophotometric
shade analysis of human teeth. J Dent Res 2002;81:578-82.
Palla et al
- 2017
35. Macentee M, Lakowski R. Instrumental color measurement of vital and
extracted human teeth. J Oral Rehabil 1981;8:203-8.
36. Seghi RR, Johnston WM, O’Brien WJ. Spectrophotometric analysis of
color differences between porcelain systems. J Prosthet Dent 1986;56:
35-40.
37. Um CM, Ruyter IE. Staining of resin-based veneering materials with coffee
and tea. Quintessence Int 1991;22:377-86.
38. Crispin BJ, Hewlett E, Seghi R. Relative color stability of ceramic stains
subjected to glazing temperatures. J Prosthet Dent 1991;66:20-3.
39. Rosenstiel SF, Baiker MA, Johnston WM. Comparison of glazed and polished
dental porcelain. Int J Prosthodont 1989;2:524-9.
40. Lee SY, Nathanson D, Giordano R. Color stability of a new light-cured
ceramic stain system subjected to glazing temperature. J Oral Rehabil
2001;28:457-62.
41. Wyszecki G, Stiles WS. Color Science: concepts and methods, quantitative
data and formulas. 2nd ed. New York: John Wiley & Sons; 1982:117-243.
42. Kuehni RG, Marcus RT. An experiment in visual scaling of small color dif-
ferences. Color Res Appl 1979;4:83-91.
43. Ishikawa-Nagai S, Sato RR, Shiraishi A, Ishibashi K. Using a computer color
matching system. A newly developed spectrophotometer designed for clinical
application. Int J Prosthodont 1994;7:50-5.
44. Dos Santos DM, da Silva EV, Vechiato-Filho AJ, Watanabe D,
Bitencourt SB, Guiotti AM, et al. Effect of different acidic solutions on the
optical behavior of lithium disilicate ceramics. J Prosthet Dent 2017 Feb 20.
doi: 10.1016/j.prosdent.2016.10.023. [Epub ahead of print].
45. Bilmeyer FW, Saltsman M. Principles of color technology. 2nd ed. New York:
John Wiley & Sons; 1981:1-110.
46. Wozniak WT. Proposed guidelines for the acceptance program for dental
shade guides. Chicago: American Dental Association; 1987:1-2.
47. O’Brien WJ, Groh CL, Boenke KM. A new small color difference equation for
dental shades. Dent Res 1990;69:1762-4.
48. Goldstein GR, Schmitt GW. Repeatability of a specially designed intraoral
colorimeter. J Prosthet Dent 1993;69:616-9.
49. Esquivel-Upshaw J, Chai J, Sansano S, Shonberg D. Resistance to staining,
flexural strength, and chemical solubility of core porcelains for all-ceramic
crowns. Int J Prosthodont 2001;14:284-8.
50. Johnston WM, Kao EC. Assessments of appearance match by visual obser-
vation and clinical colorimetry. J Dent Res 1989;68:819-22.
51. Ebeid K, Wille S, Hamdy A, Salah T, El-Etreby A, Kern M. Effect of changes
in sintering parameters on monolithic translucent zirconia. Dent Mater
2014;30:419-24.
Palla et al
11
52. Douglas RD, Steinhauer TJ, Wee AG. Intraoral determination of the tolerance
of dentists for perceptibility and acceptability of shade mismatch. J Prosthet
Dent 2007;97:200-8.
53. Ertan A, Sahin E. Color stability of low fusing porcelains: an in vitro study.
J Oral Rehabil 2005;32:358-61.
54. Heydecke G, Zhang F, Razzoog M. In vitro color stability of double-layer
veneers after accelerated aging. J Prosthet Dent 2001;85:551-7.
55. Ayaz EA, Bagis B, Turgut S. Effect of antiasthmatic medication on the surface
roughness and color stability of dental restorative materials. Med Princ Pract
2014;23:24-8.
56. De-Souza F, Casemiro L, Garcia L, Cruvinel D. Color stability of dental ce-
ramics submitted to artificial accelerated aging after repeated firings.
J Prosthet Dent 2009;101:13-8.
57. Fuss T, Mogu!s-Milankovi"c A, Ray CS, Lesher CE, Youngmane R, Daya DE.
Ex situ XRD, TEM, IR, Raman and NMR spectroscopy of crystallization of
lithium disilicate glass at high pressure. J Non Cryst Solids 2006;352:4101-11.
58. West AR, Glasser FP. Crystallization of lithium trisilicate, Li2Si307, from Li20-
SiO2 melts Mat Res Bull 1970;5:837-42.
59. Yin Z, Wang K, Zhao P, Tang X. Enhanced CO2 chemisorption properties of
Li4SiO4, using a water hydration−calcination technique. Ind Eng Chem Res
2016;55:1142-6.
60. Wang JX, Chen KT, Wu JS, Wang PH, Huang ST, Chen CC. Production of
biodiesel through transesterification of soybean oil using lithium orthosilicate
solid catalyst. Fuel Process Tech 2012;104:167-73.
Corresponding author:
Dr Petros T. Koidis
Department of Fixed Prosthesis & Implant Prosthodontics
Aristotle University of Thessaloniki, University Campus
Dentistry Bldg, GR 54124, Thessaloniki
GREECE
Email: pkoidis@dent.auth.gr
Acknowledgments
The authors thank mathematician Dr Vasilios Karagiannis for statistical analysis;
and dental technicians Diamantis Tsifoutakos and Giorgos Nikolaou for
manufacturing specimens. The experimental procedures were performed at the
Department of Basic Dental Sciences, Division of Dental Tissues Pathology and
Therapeutics, School of Dentistry, Aristotle University of Thessaloniki, Greece.
Copyright © 2017 by the Editorial Council for The Journal of Prosthetic Dentistry.
THE JOURNAL OF PROSTHETIC DENTISTRY