Journal of Oral Rehabilitation 2001 28; 95–101
Effect of different acid treatments on a porcelain surface1
Ş. CANAY, N. HERSEK & A. ERTAN
Ankara, Turkey
SUMMARY The objective of this study was to deter-
mine the effect of selected surface treatments on
the surface texture of a feldspathic porcelain. The
three different etchant treatments were, acidulated
phosphate fluoride (APF) applied for 10 min and
hydrofluoric acid (HF) applied for 1 and 4 min.
After acid treatment, half of the specimens from
each group were cleansed with water and others
were subjected to ultrasonic cleaning and then
dried. Half of the specimens cleansed with two
different methods were treated with silane. Scan-
ning electron microscopy (SEM) and energy disper-
sive spectroscopy (EDS) were used to characterize
the effects of such treatments. Etching with APF
displays shallow patterns. Etching for 1 min with
Introduction
The increasing use of porcelain restorations in dentistry
has resulted in the need for repair of fractured porce-
lain. Various methods have been introduced to repair
fractured porcelain with composite resin. Successful
repair requires the conditioning of porcelain surfaces.
In new repair systems, in spite of physical alterations
of porcelain surface, some chemical agents are also
used to help promote the chemical bond of composite
resin to the ceramic surface (Swift, Le Valley & Boyer,
1992; Thurmond, Barkmeier & Wilwerding, 1994).
Many studies have reported enhanced bond strength
with application of silane to the etched porcelain sur-
face. In addition, silane treatment will improve the
longevity of repairs (Stokes, Hood & Tidmarsh, 1988;
Sorenson et al., 1991; Aida, Hayakawa & Mizukawa,
1
This work was presented in part at the annual meeting of the
International Association for Dental Research in Nice, France
24– 28 June 1998.
© 2001 Blackwell Science Ltd
Hacettepe University, Faculty of Dentistry, Department of Prosthodontics,
HF displays deep channels, pores and precipitates
on the surface and as the etching time increased
these channels were replaced by larger channels.
EDS analyses show that the crystalline precipitates
on the etched surfaces, which were not readily
soluble in water, were the reaction products of Na,
K, Ca, Al, etc. HF displayed a more roughened
surface than the APF gel. However, the precipitates
remain on the surface after acid application, they
can only be removed by ultrasonic cleaning and
cannot be removed by rinsing.
KEYWORDS: porcelain etching, porcelain repair, si-
lane agents
1995; Shahverdi et al., 1998). Organo-silane coupling
agents are bifunctional, with one end of the molecule
capable of reacting with an inorganic surface such as
porcelain and the other with an organic surface such as
a resin (Fan, 1991).
Various surface treatment methods to increase the
bond between porcelain surface and composite resin
has brought the bond strength values to clinically ac-
ceptable level.
Etching of dental ceramics with hydrofluoric (HF)
acid or acidulated phosphate (ADP) fluoride gel is a
common procedure to achieve a clean microretentive
surface before bonding or repairing of ceramic restora-
tions (Bailey & Bennett, 1988; Lacy et al., 1988; Shen,
Phoenix & Anusavice, 1994; Tylka & Stewart, 1994;
Kupiec et al., 1996).
Both acids can etch glass and porcelain and thus
create a mechanically retentive surface. Type of
etchant, concentration of etchant and time of etching,
all influence bond strength (Tylka & Stewart, 1994).
95
96 Ş . C A N A Y et al.
For this reason, acid concentrations and etching times
should be adjusted with specific porcelains to optimize
bond strength.
HF acid etching is a well accepted pre-treatment
regimen that produces an irregular surface topography
(Al Edris et al., 1990). The main crystalline component
of dental porcelain is leucite (K2O.Al2O3.4SiO2).
Leucite dissolves more rapidly than the surrounding
glass in HF acid, so etching produces microretentive
channels in the porcelain where leucite had been. The
number and size of such irregularities has been associ-
ated with the increase in bond strength (Stangel,
Nathanson & Hsu, 1987). The hazards of HF acid are
well recognized, so the question is whether APF acid
gel might serve as a safe and effective substitute for
etching porcelain surfaces bonded to composite or not
(Lacy et al., 1988). After acid treatment many grooves
and channels of varying depths and sizes are formed on
the porcelain surface for the penetration of composite
resin. However, acid precipitates remain at the bottom
surface of these grooves and channels. The question
still remains: if the precipitate on the surface could be
removed by rinsing is ultrasonic cleaning still needed?
The purpose of this study, was to characterize treated
porcelain surfaces by means of scanning electron mi-
croscope (SEM) and examine the effect of the etching
time, silane treatment and ultrasonic cleaning upon
surfaces, by energy dispersive spectroscopy (EDS).
Materials and methods
Porcelain discs were condensed in a stainless steel mold
with an internal diameter of 7 mm and a height of
3 mm with Ceramco II porcelain*: excess fluid was
removed with facial tissues. After removal from the
mold the disks were fired according to the manufactur-
ers’ directions in a Ney Centurion Q 100† fully auto-
mated computerized dental porcelain furnace with the
pre-programmed ‘main vacuum’ cycle.
Following sintering, the disks were examined for
voids. Twenty-four acceptable specimens were abraded
and polished with a no. 600 grit silicon– carbide paper
to flatten the surface and to remove the glaze after
which they were steam cleaned. After preparation, the
specimens were immersed in 37 °C water for a period
of 24 h.
* Ceramco Inc., Six Terri Lane, Burlington, NJ 08016, U.S.A.
† ¶
Ney Dental Int., Yucaipa, CA, U.S.A.
© 2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 95–101
Porcelain samples were divided into three groups.
One group was etched with 1·23% APF§ gel for
10 min. The second group was etched with 9·5% HF
solution§ for 1 min and the third group was etched
with 9·5% HF for 4 min. In each group four samples
were etched and subsequently rinsed with distilled
water for 10 s, the other four samples were etched and
cleaned by ultrasonic wave for 1 min. After rinsing
with distilled water and ultrasonic cleaning, respec-
tively, two out of four samples of each group were
treated with silane Scotchprime¶. In this way 12 groups
of porcelain samples with different surface treatments
were obtained. A summary of sample preparations is
presented in Table 1. The silane agent was applied to
the porcelain surfaces with a minisponge and was
blown gently with air to a thin layer, then air-dried for
10 s.
In the first part of the study the photomicrographs of
the etched patterns created by the HF and APF acid gels
were examined. The surface-texture of the porcelain
samples was observed by means of a SEM. In the
second part of the study, additional X-ray area scans
were performed to determine any changes in the ele-
mental distribution profiles of the specimens. Compo-
nents of fluorine were analysed by X-ray analyses. The
effect of rinsing with distilled water or ultrasonic clean-
ing on the removal of acid precipitates was evaluated.
Then the effect of silane application on the surfaces
was examined.
Table 1. Porcelain surface treatments
Groups Treatments
1 10 min APF+distilled water
2 10 min APF+distilled water+silane
3 10 min APF+ultrasonic cleaning
4 10 min APF+ultrasonic cleaning+silane
5 1 min HF+distilled water
6 1 min HF+distilled water+silane
7 l min HF+ultrasonic cleaning
8 1 min HF+ultrasonic cleaning+silane
9 4 min HF+distilled water
10 4 min HF+distilled water+silane
11 4 min HF+ultrasonic cleaning
12 4 min HF+ultrasonic cleaning+silane
‡
Nupro, Johnson & Johnson Dental Care Co., New Brunswich,
NJ, U.S.A.
§
IPS Ceramic Aetzgel Sortiment, Ivoclar, Schaan, Liechtenstein.
3M, Dental Products, St Paul, MN, U.S.A.
 ACID TREATMENTS ON PORCELAIN SURFACES 97

Specimens for SEM were mounted as chips on alu-

minum stubs and gold coated. A JEOL** SEM equipped
with a TRACOR Northern 5502 EDS system allowing
semiquantitative analyses was used to determine the
morphology and chemical characteristics of the miner-
als. After all the peaks on the EDS spectrum were
identified, the relative concentrations of the elements
were then compared with the crystal morphology. The
peak heights of Si, Al, K, Na, Ca and F are proportional
to their concentrations.

Results
In this study etch pattern as a function of time and
etchant type was investigated with SEM. The contents
of any precipitates and the means to remove them,
according to the type of etchant employed, were
analysed through EDS analysis. The acid that creates
more irregularities on the surface will promote good
adhesion of composite resin to the porcelain surface.
There was a remarkable difference between the
etchants and the time of exposure to the etchant.
The APF clearly affected the porcelain surfaces but
appeared to be insufficient for creating a microme-
chanical bonding surface in comparison with HF acid
treated specimens. The samples etched with APF acid
gel revealed a relatively smooth homogenous surface,
whereas the samples etched with HF acid revealed an
irregular surface showing a three dimensional lattice of Fig. 1. Porcelain surfaces treated via 1·23% APF acid gel (origi-
voids and channels. Figure la represents the surface nal magnification ×1500): (a) after rinsing with distilled water
topography of samples treated with APF, which only (group 1); and (b) after ultrasonic cleaning (group 3).
roughens the surface. Figure lb demonstrates that ul-
trasonic cleaning of the surface could remove the sonication. The surface is uniformly porous with nu-
residue and surface roughness becomes more clear. merous distinguishable microundercuts. The
The application of silane on the etched surface, as photomicrographs of samples etched with HF for 4 min
would be noticed with other groups, causes a decrease were represented in Fig. 3a and b. This procedure
of sharpness, leaving a hazy appearance. creates much larger and deeper channels. The etched
HF acid etchant patterns appeared more pronounced surface was both intricate and complicated. After 4 min
and aggressive. Voids and channels appeared larger and of etching with HF acid the surface is mostly covered
deeper in samples treated with HF for 4 min than the by a thick, porous layer of crytalline precipitate. Precip-
samples treated with HF for 1 min and more numerous itates in Fig. 3a, as would be seen clearly in Fig. 3b
than the samples treated with APF for 10 min. Figure have been removed after ultrasonication.
2a and b depict the effect of 1 min application of HF
acid. This treatment resulted in the production of small
channels throughout the ceramic surface. Small chan- EDS analysis
nels, which were approximately 3 – 4 mm wide, were There was no noticeable difference in the EDS profiles
clearer in Fig. 2b indicating the effectiveness of ultra- of samples treated with APF acid gel. Typical examples
** JSM-840 A, Tokyo, Japan. of the EDS analysis after treatment with HF for 1 and

© 2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 95– 101
98 Ş . C A N A Y et al.

4 min are shown in Fig. 4 and Fig. 5. Figure 4a and b

compares the specimens which were rinsed with water
and cleaned ultrasonically after 1 min etching with HF.
EDS analysis revealed a decreased concentration of Si,
Al, Ca, and Na in the etched zone. The decrease is
more prominent in the samples treated with HF for
4 min (Fig. 5a and b). The compositional profiles of
samples rinsed only with distilled water after acid etch-
ing indicate fluorine peaks whereas those cleaned ul-
trasonically shows no fluorine peak at all. The fluorine
peaks in the samples treated with HF for 4 min are
more numerous than those treated for 1 min and
meantime Si, Al, K, Ca and Na peaks have been dis-
tinctly decreased.

Fig. 3. Porcelain surfaces treated via 4 min etch with HF acid gel
(original magnification × 1500). (a) Surfaces after rinsing with
distilled water (group 9), the patterns have been obscured by
reaction products. (b) Surfaces after ultrasonic cleaning (Group
11), note the open structure with deep channels and pores.

Fig. 2. Porcelain surfaces treated via 1 min etch with 95% HF
acid gel (original magnification × 1500). (a) Surfaces after rinsing
with distilled water (group 5), etching with HF for 1 min causes
the appearance of shallow patterns. (b) Surfaces after ultrasonic
cleaning (group 7), the debris could be readily removed by
ultrasonic cleaning
Discussion
Different types of treatment methods have been used
to prepare porcelain surfaces for resin bonding. Among
these, etching procedures are used to facilitate me-
chanical interlocking thereby enhancing porcelain
composite resin adhesion (Thurmond et al., 1994;
Phoenix & Shen, 1995; Kupiec et al., 1996).
Many researchers reported that SEM findings of the
etched porcelain surfaces with APF and HF acid
showed that etching by APF may be adequate and
produce pronounced morphological changes compared
with HF (Lacy et al., 1988; Swift et al., 1992; Tylka &
Stewart, 1994). According to our results and as a result
of the fact that bond strength was not tested, the
samples etched with APF may have an inferior

© 2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 95–101

 ACID TREATMENTS ON PORCELAIN SURFACES 99

micromechanical retentive surface. APF has hydrogen
and fluoride ions but produces only a low concentra-
tion of hydrofluoric acid, which acts more superficially
on the surface as seen in Fig. 1a and b.
APF acid gel presents little risk compared with HF
which is caustic to soft tissues. When working with HF
acid, an isolated workspace with adequate ventilation
is necessary. HF does not cause tissue damage by being
caustic per se, but by being a metabolic poison. HF is
not a particularly strong acid, but it diffuses into cells
and kills them by disrupting their metabolism. When
HF is spilled onto skin, it may not appear immediately
to have caused a burn; it often takes 24– 48 h to see the
effects, which can often be very deep tissue necrosis. A
2% aqueous solution of HF can cause corneal erosion
(Beiran, Miller & Bentur, 1997). During the intra-oral
use of HF for porcelain etching some precautions
should be taken into consideration.
The results of this study clearly indicate that HF acid
applied either for 1 or 4 min, produced the desired
surface as determined by many other researchers (Al
Edris et al., 1990; Tylka & Stewart, 1994). The micro-
retentiveness of the HF etched surfaces appeared re-
markably superior than APF etched surfaces. Longer
etching time produced increased surface roughness of
porcelain (Fig. 3a and b). HF selectively dissolves the

Fig. 4. Energy dispersive spectroscopic analysis of porcelain surfaces treated with 1 min HF acid. (a) Surfaces rinsed with distilled water
(group 5), elements were identified as Si, Al, K, Ca, Na, Ba and F. (b) Ultrasonic cleaning (group 7), note that there is no F peak.

© 2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 95– 101
100 Ş . C A N A Y et al.

Fig. 5. Energy dispersive spectroscopic analysis of porcelain surfaces treated with 4 min HF acid. (a) Rinsed with distilled water (group
9), note that the same elements are found in the spectra of Fig. 4 a but in different proportions because of the longer etching time. F
peak can easily be seen on the spectra. (b) Ultrasonic cleaning (group 11) removed the F from the surface.

glass matrix and produces a porous irregular surface
which increases the surface area and facilitates the
penetration of repair materials into the microundercuts
of the etched porcelain surfaces (Lu et al., 1992).
The poorly adherent precipitates that are deposited
on the surface after acid treatment, may weaken
resin– porcelain bonds, and lead to clinical failure
(Phoenix & Shen, 1995). EDS analysis shows that these
are the reaction products of fluorosilicate of Na, K, Ca
and Al. According to the results of this study it can be
concluded that ultrasonication on the surface will be
effective in removing these precipitates, whereas only
rinsing with distilled water will result in the fluorine-
containing precipitates remaining on the surface. Dur-
ing intra-oral repair procedure, ultrasonic cleaning is
impossible, so the precipitates remain on the surface.
Further studies are needed to investigate the effects of
HF acid remaining on the surface for a long time. Aida
et al. (1995) reported that ultrasonic cleaning after
etching did not markedly affect the bond strength.
Although we did not test the bond strength, when
SEM and EDS results of the present study are analysed,
it could be stated that the best surface would be ob-
tained by immersion of the specimens in an ultrasonic
bath because this procedure would produce a more
penetrative surface. Glass attacked by HF, produces

© 2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 95–101

 ACID TREATMENTS ON PORCELAIN SURFACES
insoluble silica– fluoride salts, which remains as residue
or deposit on the surface (Bailey & Bennett, 1988). On
the other hand, application of silane to the etched
surface may cause these salts to dissociate. This was
achieved through the hydrolysis and adsorption of si-
lane on a ceramic surface. In addition, silane promotes
wetting of the porcelain surface and, thus, enhances
the flow of the resin thereby improving the bond
strength. It was concluded that after activation silane
condenses to form a high molecular mass of poly-
methacryloxypropyl– trimethoxysilane oligomer (Lu et
al., 1992). Silane application forms an organosilane
layer that was not visible under SEM. This silica layer
may be thinner than the coating layer of the gold used
for preparation of the specimen surfaces for the SEM
observation. Based upon the quality of etched surface,
we can conclude that an etching time of 1 min can
provide adequate roughness. Increasing the etching
time to 4 min produces sharp irregularities on the
porcelain surface, which causes stress concentration
that can lead to fracture within the porcelain substrate.
However, the hazard of keeping HF acid in the mouth
for longer time should be taken into consideration as
well.
References
AIDA, M., HAYAKAWA, T. & MIZUKAWA, K. (1995) Adhesion of
composite to porcelain with various surface conditions. Journal
of Prosthetic Dentistry, 73, 464.
AL EDRIS, A., JABR, A.A., COOLEY, R.L. & BARGHI, N. (1990) SEM
evaluation of etch patterns by three etchants on three porce-
lains. Journal of Prosthetic Dentistry, 64, 734.
BAILEY, L.F. & BENNETT, R.J. (1988) Dicor surface treatments for
enhanced bonding. Journal of Dental Research, 67, 925.
BEIRAN, I., MILLER, B. & BENTUR, Y. (1997) The efficacy of calcium
gluconate in occular hydrofluoric acid burns. Human and Ex-
perimental Toxicology, 16, 223.
© 2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 95– 101
101
FAN, P.L. (1991) Porcelain repair materials. Journal of American
Dental Association, 122, 124.
KUPIEC, K.A., WUERTZ, K.M., BARKMEIER, W.W. & WILWERDING,
T.M. (1996) Evaluation of porcelain surface treatments and
agents for composite to porcelain repair. Journal of Prosthetic
Dentistry, 76, 119.
LACY, A.M., LA LUY, J., WATANABE, B.S. & DELLINGES, M. (1988)
Effect of porcelain surface treatment on the bond to composite.
Journal of Prosthetic Dentistry, 60, 288.
LU, R., HARTCOURT, J.K., TYAS, M.J. & ALEXANDER, B. (1992) An
investigation of the composite resin/interface. Australian Dental
Journal, 37, 12.
PHOENIX, R.D. & SHEN, C. (1995) Characterization of treated
porcelain surfaces via dynamic contact angle analysis. Interna-
tional Journal of Prosthodontics, 8, 187.
SHAHVERDI, S., CANAY, Ş., ŞAHIN, E. & BILGE, A. (1998) Effect of
different surface treatment methods on the bond strength of
composite resin to porcelain. Journal of Oral Rehabilitation, 25,
699.
SHEN, C., PHOENIX, R.D. & ANUSAVICE, K.T. (1994) Effects of
fluorine– based etchants upon representative porcelain sur-
faces. Todays FDA Scientific, 6, 4C.
SORENSON, J.A., ENGELMAN, M.J., TORRES, T.J. & AVERA, S.P.
(1991) Shear bond strength of composite resin to porcelain.
International Journal of Prosthodontics, 4, 17.
STANGEL, I., NATHANSON, D. & HSU, C.S. (1987) Shear strength of
the composite bond to etched porcelain. Journal of Dental
Research, 66, 1460.
STOKES, A.N., HOOD, J.A.A. & TIDMARSH, B.G. (1988) Effect of 6
month water storage on silane treated resin/porcelain bonds.
Journal of Dentistry, 16, 294.
SWIFT, E.J., LE VALLEY, B.D. & BOYER, D.B. (1992) Evaluation of
new methods for composite repair. Dental Materials, 8, 362.
THURMOND, J.W., BARKMEIER, W.W. & WILWERDING, T.M. (1994)
Effect of porcelain surface treatments on bond strengths of
composite resin bonded to porcelain. Journal of Prosthetic Den-
tistry, 72, 355.
TYLKA, D.F. & STEWART, G.P. (1994) Comparison of acidulated
phosphate fluoride gel and hydrofluoric acid etchants for
porcelain-composite repair. Journal of Prosthetic Dentistry, 72,
121.
Correspondence: Dr Ş. Canay, Bağış sok. 17/10, 06660 Kocatepe
Ankara, Turkey. E-mail: atilae@yahoo.com