In vitro color stability of double-layer veneers after accelerated aging

Guido Heydecke, DDS, Dr Med Dent,a Feimin Zhang, DDS, MS,b and
Michael E. Razzoog, DDS, MS, MPHc
School of Dentistry, University of Michigan, Ann Arbor, Mich., and School of Dentistry, Nanjing
Medical University, Nanjing, China
Statement of problem. Porcelain laminates made from thin alumina shells veneered with feldspathic
porcelain could be a promising alternative to conventional veneers. Long-term shade stability is critical
for esthetics.
Purpose. This study compared changes in CIE L*a*b* color coordinates of simulated veneers made
from aluminum oxide core material veneered with feldspathic porcelain after 300 hours of accelerated
photothermal aging (weathering).
Material and methods. Fifteen aluminum oxide disks (Procera) were divided into 3 groups. Each of
the 5 disks was veneered with All-Ceram porcelain of the Vita shades A1 and B4, respectively. Five disks
remained unfinished as controls. The disks were bonded to composite substrates simulating stained teeth.
The color of the specimens was measured with a colorimeter. All specimens were subjected to 300 hours
of accelerated aging under light exposure and thermocycling. Color measurements were repeated, and the
data were statistically evaluated with multiple paired t tests.
Results. Color changes in the test groups involved an increase in lightness and a decrease in chroma.
The calculated total color differences were not statistically significant compared with a level of 3 E units.
Conclusion. This simulation of double-layer veneers appears to demonstrate acceptable color stability
for this type of restoration. Clinical research is necessary to substantiate these in vitro findings. (J Prosthet
Dent 2001;85:551-7.)

CLINICAL IMPLICATIONS
The results of this study indicate that aluminum oxide shells can be used as the core of
double-layer veneers. This veneer system provides the clinician and patient with a lasting and therefore predictable shade match.

or decades, 1 objective of anterior dental restorations has been to replace tooth structure lost by dental
disease in an esthetic manner. Improvements in dental
techniques and materials in the last 15 years have led
to the development of porcelain veneers,1 which are
well suited to change the shape and color of teeth.2-5
Clinical research demonstrates that the overall success
rates of veneer restorations are very high.1,6,7
However, the challenge with ceramic laminates is to
achieve maximum esthetics, which frequently includes
color change with limited preparation of the enamel.
Several studies have evaluated the effect of shade,
thickness, amount of opaque porcelain, and color of
luting resins on the final appearance of laminates.8-10
Chpindel and Cristou11 described a 2-layer porcelain
aVisiting

Assistant Professor, Department of Prosthodontics,

University of Michigan.
bClinical Assistant Professor, Nanjing Medical University.
cProfessor, Department of Prosthodontics, University of Michigan.
JUNE 2001

technique that introduces opacity with the first layer

and achieves the impression of depth and translucency
with the second layer. These authors also reported an
increase in strength with the double-layer technique.
An all-ceramic system that uses densely sintered
aluminum oxide as a high-strength core with a flexural strength of 601 73 MPa12 has been developed for
the creation of single crowns (Procera, Nobel Biocare,
Gteborg, Sweden).13-15 This system lends itself well
to the double-layer technique. The Procera core is
semitranslucent, and its shade closely matches Vita
A1.16 The aluminum oxide core is veneered with a
specific porcelain (AllCeram, Ducera, Germany) that
has fluorescent properties that correspond to those of
the natural dentition. These properties minimize problems of metamerism and improve the lifelike
appearance of the restoration.13,14,17
The use of densely sintered aluminum oxide shells,
which are subsequently veneered with feldspathic
porcelain to fabricate porcelain veneers, could potenTHE JOURNAL OF PROSTHETIC DENTISTRY 551

THE JOURNAL OF PROSTHETIC DENTISTRY

Fig. 1. Aluminum oxide shell (0.2 mm thick) manufactured

with Procera technique for veneer restoration of left lateral
incisor.

HEYDECKE, ZHANG, AND RAZZOOG

tem is commonly used in dental research. In this system, the location of a particular shade in the color
space is defined by 3 coordinates: L*, a*, and b*. L*
describes the lightness of the object being assessed.
The a* value defines the color on the red-green axis
and b* on the yellow-blue axis. The measure of the
total color difference between 2 objects is described by
E. The magnitude of E, however, gives no information on the character of the color difference between
specimens because it does not indicate the quantity
and direction of the individual CIE L*a*b* components.18-20 A color difference of up to 3.7 E units has
been described as an acceptable clinical shade match in
dentistry, whereas color differences above 3.7 E units
are visually detectable.21
No quantitative data are available for the color stability of veneer restorations.1,22 This experiment was
designed to evaluate the color stability of double-layer
porcelain veneers fabricated with the Procera technique in an in vitro weathering apparatus.

MATERIAL AND METHODS

Fig. 2. After veneering process, finished veneer was luted

with composite cement.

tially optimize both strength and esthetics. A clinical

example of a veneer restoration that used the described
technique is shown in Figures 1 and 2.
The widely recognized CIE L*a*b* color order sys552

An aluminum mold was machined as a master die

for the fabrication of multiple substrates. An autopolymerizing acrylic resin filling material (Sevriton,
Amalgamated Dental Trade Distributions, Ltd,
London, United Kingdom) was mixed with amalgam
powder to impart a gray shade and poured into an
impression of the master die to form the substrates (for
dimensions, see Fig. 3). The ratio of the amalgam was
recorded so that consistent mixtures of the substrate
resin could be fabricated. Twenty substrates were prepared and divided into 4 groups, and the initial color
of each substrate was measured and recorded. Three
groups were used for the experimental procedures, and
1 group served as the control.
Fifteen Procera aluminum oxide disks, each
13.50 mm in diameter and 0.2 mm in thickness, were
used for the experiment. Before the test, each disk was
measured in 3 locations with a Digimatic Caliper
(Mitutoyo Corp, Tokyo, Japan) to determine the actual thickness. Specimens in group 1 were not veneered;
the disks in group 2 were veneered with shade A1 and
in group 3 with shade B4 (AllCeram, Ducera, Rosbach,
Germany). The colors were chosen to represent 1 frequently requested light shade (A1) and 1 darker shade
(B4). The initial thickness of the veneering layer was
slightly more than 0.4 mm; the excess was ground with
a diamond bur to achieve a uniform thickness of
0.4 mm. All veneered surfaces were polished with 600grit polishing paper (Ecomet 3, Buehler Ltd, Lake
Bluff, Ill.) to produce a smooth surface, which was
microscopically examined to ensure lack of porosity.
The final thickness of the veneer portion was verified to
be 0.4 mm (20 m) with a Digimatic Caliper.
The disks in groups 2 and 3 were self-glazed at
VOLUME 85 NUMBER 6

HEYDECKE, ZHANG, AND RAZZOOG

THE JOURNAL OF PROSTHETIC DENTISTRY

Fig. 3. Schematic diagram of specimen consisting of simulated stained tooth substrate and
veneered porcelain disk. Chromameter head in place.

Fig. 4. Overview of the experimental procedures by group.

880C. All 15 disks were cleaned with distilled water

in an ultrasonic cleaner, and the nonveneered surfaces
were silica coated (Rocatec, ESPE USA, Norristown,
Pa.) and silanated (Silane Primer, Kerr, Orange,
Calif.). The surface of the substrates was cleaned by
etching with phosphoric acid for 15 seconds (Gel
Etchant, Kerr). Bonding was applied to the substrate
and the nonveneered surface of each disk, which was
JUNE 2001

subsequently bonded to the substrate with a neutralcolored dual-polymerizing composite cement

(NEXUS, Kerr). Light polymerization was performed
in the center of each disk for 120 seconds. The remaining 5 substrates served as controls. All samples were
stored in darkness, except during the testing procedures. An overview of the experimental process is
presented in Figure 4.
553

THE JOURNAL OF PROSTHETIC DENTISTRY

HEYDECKE, ZHANG, AND RAZZOOG

Table I. Statistical evaluation of the color differences between the 3 groups of substrates (1-way ANOVA)
Coordinate

Source of variation

L*

Between groups
Within groups
Total
Between groups
Within groups
Total
Between groups
Within groups
Total

a*

b*

df

3
16
19
3
16
19
3
16
19

Mean squares

Mean

SD

0.475
0.619

.529

51.54

0.77

0.102
0.034

.06

5.54

0.21

0.137
0.068

.153

0.99

0.28

ANOVA = Analysis of variance.

All 20 specimens were subjected to 300 hours of

accelerated aging by exposure to ultraviolet light and
water spray in a Weather-O-Meter (Atlas Electronic
Devices, Chicago, Ill.). The veneered surface of each
specimen was continuously exposed to the light
source. The irradiance level of the xenon arc light
source was 0.55 W/m2. The back panel temperature
varied between 38C (dark) and 70C (light), and the
relative humidity was 95% (dark) and 50% (light). The
dry bulb temperature was 38C in the dark and 47C
in the light stage. The testing cycle consisted of 40
minutes of light only, 20 minutes of light with front
water spray, 60 minutes of light only, and 60 minutes
in dark stage with back water spray. The total exposure
energy was 389.2 kJ/m2. The accelerated aging conditions have been previously described.23,24
A colorimeter (Minolta Chroma Meter II, Minolta
Inc, Osaka, Japan) was used for all measurements of
the color coordinate variables. The colorimeter has a
3.0-mm head and a diffuse illumination vertical viewing (0 degrees) geometry for color measurement. CIE
L*a*b* notations were used for all color measurements, and the standard illuminant D65 was selected
for all measurements. Before the experimental measurements, the colorimeter was calibrated to a white
standard tile supplied by the manufacturer.25 The
L*a*b* color notation of each specimen and the control substrates was measured 3 times consecutively, and
an average of the 3 readings was calculated to give the
initial color of the specimen (Fig. 3). All measurements were repeated after the weathering process.
Differences were calculated between the color coordinate measurements (L*, a*, and b*) of each
specimen, and the total color difference E was
obtained with the following formula:
E= ([L*1 L*2]2 + [a*1 a*2]2 + [b*1 b*2]2)

1/2

where L*1, a*1, and b*1 represent the preweathering

and L*2, a*2, and b*2 represent the postweathering
color coordinates of each specimen.20
The mean values for the color coordinates L*, a*,
b* of the disk-substrate specimens for each porcelain
554

shade were compared with the nonveneered disks with

the use of multiple paired t tests. The total color difference was statistically evaluated by comparison of the
group mean against a hypothesized mean of 3 E with
t tests. A 1-way analysis of variance (ANOVA) was
used to compare the 4 groups of substrates with each
other to detect baseline color coordinate differences.

RESULTS
The 1-way ANOVA of the color differences among
the 3 groups of substrates did not reveal any significant
differences. The mean color coordinate values are presented in Table I.
The evaluation of the color differences for the substrates alone revealed a highly significant change in the
b* (blue-yellow) coordinate. Although a significant
change also was observed in the a* (red-green) coordinate, the total difference of 12.47 E units was a
strong function of the shift toward yellow, as indicated
by the b* coordinate (Table II).
In the 3 test groups, a shift in lightness (L*) was
observed for the specimens with A1 and B4 veneering
porcelain. However, the level of statistical significance
(E =3) was not reached in any group. In all 3 test
groups, a statistically significant increase of the a*
value was observed (Fig. 5), relating to a decrease in
the green component of the specimens. The observed
changes in the b* coordinate were inconsistent among
the groups; none reached the level of significance.
Overall, an increase in lightness and a decrease in chromaticity were observed (Table II).
Resulting total color differences (E) in the 3 test
groups (disk, A1, and B4) were not clinically significant with 1 exception. In the B4-veneer group, the
change in lightness and shift in the green-red axis contributed to a change of 3.34 E units (Table II).

DISCUSSION
A new procedure has been described for the fabrication of porcelain laminate veneers with the use of a
thin aluminum oxide base material veneered with
feldspathic porcelain. The colorimetric technique used
VOLUME 85 NUMBER 6

HEYDECKE, ZHANG, AND RAZZOOG

THE JOURNAL OF PROSTHETIC DENTISTRY

Table II. Statistical evaluation of the color coordinate differences (L*,a*,b*) before and after weathering (paired t tests)
Substrate
Coordinate

Weathering

L*

Before
After
Difference
df
P
Before
After
Difference
df
P
Before
After
Difference
df
P
Before/after
Hypothesized
df
P

a*

b*

5
5
5
4
.199
5
5
5
4
.001
5
5
5
4
<.0001
5

Disk

Mean

SD

51.44
50.38
1.06

0.92
1.32
1.54

5.80
6.80
1.00

0.18
1.27
1.28

0.93
13.23
12.30

0.47
1.66
1.26

12.47
3.00

1.18

4
<.0001

Mean

5
63.31
5
63.29
5
0.03
4
.980
5
6.73
5
5.84
5
0.89
4
.001
5
8.33
5
7.09
5
1.24
4
.014
5
2.33
3.00
4
.278

A1
SD

1.19
1.03
2.19

0.20
0.13
0.20

0.43
0.50
0.66

1.20

Mean

5
66.96
5
69.16
5
2.20
4
.032
5
5.72
5
5.04
5
0.68
4
<.0001
5
9.46
5
10.13
5
0.67
4
.292
5
2.66
3.00
4
.635

B4
SD

1.30
1.08
1.52

0.10
0.08
0.05

1.25
0.35
1.24

1.49

Mean

5
63.79
5
66.61
5
2.82
4
.012
5
6.36
5
5.63
5
0.74
4
<.0001
5
15.96
5
17.24
5
1.28
4
.076
5
3.34
3.00
4
.639

SD

1.48
2.82
1.44

0.15
0.18
0.15

1.35
1.81
1.21

1.51

For the evaluation of E, a threshold level of E = 3 was defined.

has been previously validated25 and is widely used in

dental research.8,23,26 A model simulating tetracycline
staining was developed to study the color stability of
the veneer system; the model included a simulation of
the bonding interface and a dark background.
Although color measurements of stained teeth are not
available in the literature, such teeth have been described
as appearing yellow, brown, or gray to black.4,27
Therefore, the model used seems realistic for the
described scenario.
However, the described discoloration of the unfinished substrates is not comparable with the degradation
found in other composite materials. Generally, an
increase in lightness and a decrease in chroma have
been observed in conventional composites subjected to
accelerated aging.23,24 In this experiment, an increase
in the b* coordinate was apparent as a yellow discoloration of the substrates, which may have resulted from
the use of the resinamalgam powder mixture.
However, for the specimens with simulated veneers
bonded to the substrates, this discoloration could not
be detected visually. An instrumental measurement of
the underlying substrates was not possible because the
bonding material could not be removed without damage to the specimens. This type of discoloration is not
likely to occur in natural teeth and may have affected
the results. The increase in the b* coordinate of the
substrates alone, however, was not reflected in the specimens with the nonveneered disk. As a decrease in b*
was observed in the latter specimens, the discoloration
of the substrates appears negligible.
JUNE 2001

Dental restorative materials must withstand widely

varied conditions in the mouth, including temperature
changes, continuous exposure to moisture, and
mechanical use of the restoration. Controlled clinical
trials are necessary to substantiate the validity of treatment procedures. Unfortunately, they are expensive
and time-consuming. Thus, to estimate the reliability
of a new restorative technique, in vitro testing can precede clinical studies.28 Efforts have been made to
simulate, as closely as possible, clinical parameters with
in vitro testing procedures. This has led to the development of various chewing simulators.29,30 The
additional parameters of light exposure and humidity
changes can be simulated in a so-called weathering
process, which has been widely used for the testing of
dental composite and porcelain materials.23,24,31 The
manufacturer of the Weather-O-Meter used in this
study claimed that 300 hours of accelerated weathering are equivalent to 1 year of clinical service. Clinical
validation of this claim was not available.
A previous study presented very small color changes
in simple porcelain specimens after accelerated weathering.26 In this study, we chose to evaluate the color
changes of the whole system, including cementation
composite and the simulated tooth. Although slight
changes in color were observed after the weathering
process for all simulated veneers, only in group 4 was
the color change larger than 3 E units, indicating some
clinical significance. All changes were larger than in the
study by Razzoog et al26; however, it should be noted
that, in the current study, a near-clinical system that
555

THE JOURNAL OF PROSTHETIC DENTISTRY

HEYDECKE, ZHANG, AND RAZZOOG

Fig. 5. Coordinate (L*, a*, b*) and total color differences (E) before and after accelerated
aging of samples.

involved a nonuniform background was investigated.

Johnston and Kao21 stated that differences in E below
3.7 units are a clinical match. Therefore, the presented
color differences can be termed acceptable. In both
veneered groups (A1 and B4), the changes involved a
notable increase in lightness (L*) and slight decreases in
the chroma coordinates (a*, b*). Because it was not
observed in the nonveneered disk specimens, this
increase in lightness can be attributed to the veneering
porcelain. This correlates well with the reported mechanism of degradation of composites.23,24 The
discoloration observed in the group of substrates alone
did not affect the final result, as a nonsignificant shift to
yellow was observed in the groups with nonveneered or
veneered disks. Most likely, the veneers protected the
substrate resin from the weathering influence, although
a masking effect was also possible.
It is interesting that the color stability of porcelain
veneers has never been assessed quantitatively, either
in vitro or in vivo. Although a clinical evaluation of
the color match to adjacent teeth has been presented
in most clinical trials involving veneers, only relatively broad classifications were made. In light of the
data by Johnston and Kao,21 quantitative evaluations
seem advisable for future clinical studies. Whether
the presented double-layer technique has esthetic
advantages over conventionally layered veneers is

556

unknown at this stage. Further studies are needed to

compare the color stability of different systems.

CONCLUSIONS
Within the limitations of this study, the following
conclusions were drawn:
1. After weathering simulated double-layer veneers
on stained teeth, color changes were statistically
insignificant. Recorded color differences were caused
mainly by an increase in lightness.
2. A clinically stable shade can be achieved with double-layer veneers, even for severely discolored teeth.
REFERENCES
1. Peumans M, Van Meerbeek B, Lambrechts P, Vanherle G. Porcelain
veneers: a review of the literature. J Dent 2000;28:163-77.
2. Nixon RL. Masking severely tetracycline-stained teeth with ceramic laminate veneers. Pract Periodontics Aesthet Dent 1996;8:227-35; quiz 37.
3. Belser UC, Magne P, Magne M. Ceramic laminate veneers: continuous
evolution of indications. J Esthet Dent 1997;9:197-207.
4. Faunce F. Management of discolored teeth. Dent Clin North Am
1983;27:657-70.
5. McLaughlin G. Porcelain veneers. Dent Clin North Am 1998;42:653-6,
ix.
6. Kihn PW, Barnes DM. The clinical longevity of porcelain veneers: a 48month clinical evaluation. J Am Dent Assoc 1998;129:747-52.
7. Fradeani M. Six-year follow-up with Empress veneers. Int J Periodontics
Restorative Dent 1998;18:216-25.
8. Yaman P, Qazi SR, Dennison JB, Razzoog ME. Effect of adding opaque
porcelain on the final color of porcelain laminates. J Prosthet Dent
1997;77:136-40.

VOLUME 85 NUMBER 6

HEYDECKE, ZHANG, AND RAZZOOG

THE JOURNAL OF PROSTHETIC DENTISTRY

9. Davis BK, Papcum LJ, Johnston WM. Effect of cement shade on final
shade of porcelain veneers. J Dent Res 1991;70:297 [Abstract 250].
10. Davis BK, Aquilino SA, Lund PS, Diaz-Arnold AM, Denehy GE.
Colorimetric evaluation of the effect of porcelain opacity on the resultant
color of porcelain veneers. Int J Prosthodont 1992;5:130-6.
11. Chpindel P, Cristou M. Tooth preparation and fabrication of porcelain
veneers using a double-layer technique. Pract Periodontics Aesthet Dent
1994;6:19-28; quiz 30.
12. Wagner WC, Chu TM. Biaxial flexural strength and indentation fracture
toughness of three new dental core ceramics. J Prosthet Dent
1996;76:140-4.
13. Andersson M, Oden A. A new all-ceramic crown. A dense-sintered,
high-purity alumina coping with porcelain. Acta Odontol Scand
1993;51:59-64.
14. Andersson M, Razzoog ME, Oden A, Hegenbarth EA, Lang BR. Procera:
a new way to achieve an all-ceramic crown. Quintessence Int
1998;29:285-96.
15. Oden A, Andersson M, Krystek-Ondracek I, Magnusson D. Five-year
clinical evaluation of Procera AllCeram crowns. J Prosthet Dent
1998;80:450-6.
16. Razzoog ME, Lang LA, McAndrew KS. AllCeram crowns for single
replacement implant abutments. J Prosthet Dent 1997;78:486-9.
17. Hegenbarth EA. Procera aluminum oxide ceramics: a new way to
achieve stability, precision and esthetics in all-ceramic restorations.
QDT 1996;19:21-34.
18. Sproull RC. Color matching in dentistry. I. The three-dimensional nature
of color. J Prosthet Dent 1973;29:416-24.
19. Rosenstiel SF, Land MF, Fujimoto J. Contemporary fixed prosthodontics.
2nd ed. St. Louis (MO): Mosby; 1995 p. 490-5.
20. Hunter RS, Harold RW. The measure of appearance. 2nd ed. New York
(NY): John Wiley; 1987. p. 178-85.
21. Johnston WM, Kao EC. Assessment of appearance match by visual observation and clinical colorimetry. J Dent Res 1989;68:819-22.
22. Kreulen CM, Creugers NH, Meijering AC. Meta-analysis of anterior
veneer restorations in clinical studies. J Dent 1998;26:345-53.
23. Douglas RD. Color stability of new-generation indirect resins for
prosthodontic application. J Prosthet Dent 2000;83:166-70.

24. Powers JM, Fan PL, Raptis CN. Color stability of new composite restorative materials under accelerated aging. J Dent Res 1980;59:2071-4.
25. Seghi RR, Johnston WM, OBrien WJ. Performance assessment of colorimetric devices on dental porcelains. J Dent Res 1989;68:1755-9.
26. Razzoog ME, Lang BR, Russell MM, May KB. A comparison of the color
stability of conventional and titanium dental porcelain. J Prosthet Dent
1994;72:453-6.
27. Haywood VB, Leonard RH, Dickinson GL. Efficacy of six months of
nightguard vital bleaching of tetracycline-stained teeth. J Esthet Dent
1997;9:13-9.
28. Kern M, Strub JR, L XY. Wear of composite resin veneering materials in
a dual-axis chewing simulator. J Oral Rehabil 1999;26:372-8.
29. Krejci I, Reich T, Lutz F, Albertoni M. An in vitro test procedure for evaluating dental restoration systems. 1. A computer-controlled mastication
simulator [in German]. Schweiz Monatsschr Zahnmed 1990;100:95360.
30. DeLong R, Douglas WH. Development of an artificial oral environment
for the testing of dental restoratives: bi-axial force and movement control. J Dent Res 1983;62:32-6
31. Powers JM, Dennison JB, Koran A. Color stability of restorative resins
under accelerated aging. J Dent Res 1978;57:964-70.
Reprint requests to:
DR MICHAEL E. RAZZOOG
SCHOOL OF DENTISTRY
UNIVERSITY OF MICHIGAN
ANN ARBOR, MI 48109
FAX: (734)763-3453
E-MAIL: merim@umich.edu
Copyright 2001 by The Editorial Council of The Journal of Prosthetic
Dentistry.
0022-3913/2001/$35.00 + 0. 10/1/115385

doi:10.1067/mpr.2001.115385

Bound volumes available to subscribers

Bound volumes of The Journal of Prosthetic Dentistry are available to subscribers (only) for
the 2001 issues from the publisher at a cost of $92.00 ($106.00 international) for Vol. 85
(January-June) and Vol. 86 (July-December). Shipping charges are included. Each bound volume
contains a subject and author index, and all advertising is removed. Copies are shipped within
30 days after publication of the last issue in the volume. The binding is durable buckram with the
journal name, volume number, and year stamped in gold on the spine. Volumes 83 and 84 are
also available. Payment must accompany all orders. Contact Mosby, Subscription Customer
Service, 6277 Sea Harbor Dr, Orlando, FL 32887, or call 800-654-2452 or 407-345-4000.
Subscriptions must be in force to qualify. Bound volumes are not available in place of
a regular Journal subscription.

JUNE 2001

557