European Journal of Orthodontics, 2019, 1–6

doi:10.1093/ejo/cjz024
Original article

Downloaded from https://academic.oup.com/ejo/advance-article-abstract/doi/10.1093/ejo/cjz024/5476400 by Stockholm University Library user on 23 April 2019

Original article

A study of the mechanical properties of

as-received and intraorally exposed single-crystal
and polycrystalline orthodontic ceramic brackets
Eleni Alexopoulou1, Georgios Polychronis2, Dimitrios Konstantonis3,
Iosif Sifakakis3, Spiros Zinelis2 and Theodore Eliades1,
Clinic of Orthodontics and Pediatric Dentistry, Center of Dental Medicine, University of Zurich, Switzerland and
1

Departments of 2Dental Biomaterials and 3Orthodontics, School of Dentistry, National and Kapodistrian University of
Athens, Greece

Correspondence to: Theodore Eliades, Clinic of Orthodontics and Pediatric Dentistry, Center of Dental Medicine, University
of Zurich, Plattenstrasse 11, Zurich 8032, Switzerland. E-mail: [email protected]

Summary
Background: Although ceramic brackets have been extensively used for decades in orthodontics
there is not till today any study focusing on the possible deterioration of mechanical properties
after in vivo ageing.
Objectives: To determine whether the mechanical properties of alumina orthodontic brackets
change after intraoral ageing thereby assessing the validity of a theoretical model established for
the performance of ceramics in wet environments.
Materials and methods: Two alumina brackets, one single crystal (Radiance, American
Orthodontics, Sheboygan, WI) and one polycrystalline (Clarity, 3M, St. Paul, MN) were included in
this study. Ten brackets for each group were collected from different patients after a minimum of
3-month intraoral exposure, whereas as-received brackets of the same manufacturers were used
as controls. The specimens were subjected to Raman spectroscopy and were then embedded in
epoxy resin and metallographic ground and polished. The mechanical properties of four groups
(radiance control: RAC, radiance-retrieved RAR, clarity control: CLC and clarity-retrieved CLR) were
determined using instrumented indentation testing according to ISO 14577-2002. The mechanical
properties tested were Martens hardness (HM), indentation modulus (EIT), the ratio of elastic to
total work, commonly known as elastic index (ηIT), and fracture toughness (KIC). The numerical
results were statistically analysed employing two-way analysis of variance (ANOVA) and Tukey
multiple comparison test at a = 0.05.
Results: Raman analysis revealed that both brackets are made of a-Al2O3 (corundum). No
statistically significant differences were found for HM (N/mm2): RAC = 7249 (1507), RAR = 6926
(1144), CLC = 8052 (1360), CLR = 7390 (2393), or for EIT (GPa): RAC = 141 (27), RAR = 139 (23),
CLC = 139 (28), CLR = 131 (47). However, significant differences were identified between the two
alumina brackets tested for ηIT (%): RAC = 55.7 (4.2), RAR = 54.0 (3.5), CLC = 62.5 (4.4), CLR = 61.8 (4.7),
while KIC was measured only for the polycrystalline bracket (Clarity) because of the complicated
fractured pattern of the single-crystal bracket. Both brackets share equal HM and EIT before and
after orthodontic intraoral ageing.
Limitations: Whereas the study assessed the changes after intraoral exposure per theoretical
model, which describes the reduction of critical stress to induce fracture after wetting, long-term
intraoral ageing could have induced more pronounced effects.

© The Author(s) 2019. Published by Oxford University Press on behalf of the European Orthodontic Society.
1
All rights reserved. For permissions, please email: [email protected]
2 European Journal of Orthodontics, 2019

Conclusions/Implications: The results of this study indicate that 3 months of intraoral ageing do
not change the mechanical properties of single-crystal and polycrystalline orthodontic brackets
tested, thus indicating that the Griffith theory may not be applied to the case of manufactured
ceramic brackets owing possibly to internal defects.

Downloaded from https://academic.oup.com/ejo/advance-article-abstract/doi/10.1093/ejo/cjz024/5476400 by Stockholm University Library user on 23 April 2019

Introduction The foregoing theory predicts that the decreased surface energy
associated with immersing ceramics in the intraoral environment
After almost four decades from their appearance in orthodontic clin-
would cause a reduction in the magnitude of the critical stress that is
ical practice, single-crystal and polycrystalline alumina and zirconia
the stress required to introduce fracture. This degradation in struc-
ceramic brackets possess a central position in the armamentarium of
ture integrity should be reflected in mechanical properties of used
aesthetic orthodontic materials (1–4). Although ceramic brackets are
brackets, and thus, the aim of this study was to test whether the
free of stain and discolouration complications of their plastic coun-
mechanical properties of alumina brackets show inferior character-
terparts, the increased fracture incidence and complicated debonding
istics after intraoral exposure. The null hypothesis tested was that
process of alumina brackets precluded their wide-scale application
no significant differences will be found between the as-received and
in everyday practice (5,6). The use of zirconia brackets is limited due
clinically aged alumina brackets.
to increased friction coefficient, worse aesthetics and decreased bond
strength compared with alumina ones (2,7,8).
Despite the enthusiastic debut, their clinical application revealed Materials and methods
that alumina brackets are prone to higher incidence of bracket fail-
ure during orthodontic treatment and at debonding process, while Sample collection
their most important complication soon surfaced in the form of iat- Two brands of commercially available alumina brackets were
rogenic trauma on enamel during debonding (4,9). The aforemen- included in this study: a single-crystal (Radiance, American
tioned complications have been appended to the brittle nature of Orthodontics, Sheboygan, WI) and a polycrystalline one (Clarity,
ceramics, which combined with low KIC, which indicates a low resist- 3M, St. Paul, MN). Ten brackets from each brand were collected
ance to crack propagation (10), lead to an unfavourable response from different patients after exposure in the oral cavity of patients
to loading during orthodontic treatment. In as much, enamel crack for at least 3 months; another group of 10 specimens per material
formation during debonding drew the attention of many studies, was used in the as-retrieved (dry) condition serving as a control.
which adopted various approaches ranging from engineering to epi- Briefly, patients undergoing orthodontic treatment received ceramic
demiology of complications. The percentage of the ceramic bracket brackets ligated to wires with elastomeric moduli but not bonded
base pad covered by enamel during debonding was found to range as to enamel. The reason for avoiding conventional bracket placement
much as three times higher than the corresponding rate of debonded involving bonding was the necessity of retrieving the brackets with-
metallic bracket. The extent of damage for alumina brackets was out applying stresses during debonding and without additional frac-
found up to 46%, which however may be an underestimation as tures. By bonding the brackets to the enamel, there would be a need
fragments of enamel may not be retained by adhesive on the base to debond them and that means that stress should have been applied
bracket pad (11). along with potential fractures of the wing. Both these parameters
Although iatrogenic damage of enamel is the most serious would have adverse effects on measuring the mechanical proper-
complication, the fracture of alumina brackets during orthodontic ties of brackets. For the same reason, brackets were ligated under
therapy (12,13) or debonding is considered also as a major clini- the wire in a way to facilitate clearance from occlusal contacts with
cal complication prolonging the treatment (14–16). In addition, the opposing teeth. The 3-month period was considered an adequate
removal of broken alumina brackets is difficult and requires exten- time frame to initiate wetting and ageing as intraoral ageing has
sive chair time for polishing (17) while the use of a rubber dam is been shown to set early after the placement of materials in the oral
recommended (18). This procedure is carried out by using diamond cavity. Specifically, relevant research has demonstrated that what
stones to remove the remaining part of alumina bracket (19) jeop- is considered ageing in the broader dental ceramic field consists of
ardizing the enamel integrity (15). Based on clinical data, a study 3-month immersion (20) in laboratory environment in liquid media,
has shown that alumina bracket has a fracture rate of 14.6% dur- which bear no comparison with the actual presence of the brackets
ing debonding alone (11). However, a fractographic analysis of frac- in the oral cavity. All patients provided consent and the protocol
tures of single-crystal alumina brackets occurring during treatment was approved by the Ethics Committee School of Dentistry approval
reported that internal defects were the primary cause of fracture number 338/15.5.2017, National and Kapodistrian University of
(47.5%), followed by external defects induced by machining the Athens.
single-crystal during manufacturing (42.5%) (13). At the best of our knowledge, there are no previously published
Given that the mechanism of fracture in ceramics is associated data of mechanical properties tested, and thus an assessment of sam-
with the propagation of internal cracks, it is rational to assume that ple size was not feasible. Therefore, sample size calculation, which
this phenomenon will be facilitated under intraoral ageing. The requires that previously published data on the properties tested are
Griffiths equation describes the relationship between applied nomi- used to define the sample size required, was not appropriate to con-
nal stress and crack length at fracture, i.e. when it becomes energeti- duct. At any rate, retrieval analysis do not follow this rule because the
cally favourable for a crack to grow. When a brittle body is subjected intraoral ageing of materials in different oral cavities of patients does
to loading and experiences a crack propagation, there is a change of not constitute a standardized treatment on specimens; as a result, the
energy taking place, which comprises changes in the potential energy emphasis is on the actual clinical conditions and their effects and not
of the body and the surface energy of the new fracture surfaces. on standardization of the effect of ageing (21).
E. Alexopoulou et al. 3

Brackets from all four groups (single crystal and polycrystalline ratio of Al2O3, and a1 = 0.170204, a2 = −0.157669, a3 = 0.110937,
before and after intraoral exposure) were then subjected to Raman a4 = −0.048401, a5 = −0.005516 and a6 = 0.007625.
spectroscopy and instrumented indentation testing (IIT) and fracture
toughness (discussed in detail in the following sections). Then, η is provided by the formula:

E/(1 − 0.ν 2 )

Downloaded from https://academic.oup.com/ejo/advance-article-abstract/doi/10.1093/ejo/cjz024/5476400 by Stockholm University Library user on 23 April 2019

Raman spectroscopy η=
Ei /(1 − νi2 ) (5)
One Raman spectrum was acquired from the wing region of all brack-
ets tested. The brackets were placed on the stage of a microscope
(LEICA BME, Leica Microsystems Ltd, Heerbrugg, Switzerland), and Finally, the fracture toughness KIC is calculated by the formula:
the regions of interest were determined employing optical lens at ×40
Ç å î  2   ó
nominal magnification. A special device (MicroViewer-785, Raman 1 + η 0.0068 ac − 0.1118 ac + 0.8295
10−6 Pm
Microscope Adaptor) was used to attach the Raman Probe to the KIC = ∗ Ä ä2 Ä ä
c3/2 0.0757 W − 0.1956 W
Wt + 0.1285
e e
microscope. Spectra were acquired with an EZ Raman-I, high-sensitiv- Wt
ity portable Raman analyser equipped with a Laser (Soliton, Laser Und
(6)
Messtechnik, Gliching, Germany) operating at 480-mW output power,
785-nm emitted wavelength and a nominal resolution of 4.5–6.5 cm−1.
where the unit KIC is in MPa m1/2, length of c and a is in metres, and
force is expressed in Newton.
IIT and fracture toughness
All brackets were embedded in epoxy resin and subjected to met-
allographic grinding and polishing. The mechanical properties of Statistical analysis
four groups (radiance control: RAC, radiance-retrieved RAR, clar- The results of HM, EIT, ηIT and KIC were statistically analysed by
ity control: CLC and clarity-retrieved CLR) were determined using two-way ANOVA employing brackets and condition as discriminat-
IIT according to ISO 14577-2002 (22). The mechanical properties ing variables, while significant differences were identified by Tukey
tested were Martens hardness (HM), indentation modulus (EIT) and multiple comparison test at a = 0.05. The normality and homosce-
the ratio of elastic to total work, commonly known as elastic index dasticity of all data were initially checked by Kolmogorov–Smirnov
(ηIT). Additional information for the methodology and mathemati- and equal test variance, respectively.
cal formulas can be found in recently published papers (23,24). In
addition, a recently introduced analytical technique was employed
for the first time in relevant research to estimate the fracture tough-
Results
ness of orthodontic brackets combining the results of IIT and crack Raman spectroscopy
length around Vickers indentation (25). The first step is the acquisi- Figure 1 demonstrates representative Raman spectra from all groups
tion of force indentation depth curve with an instrumented indenta- tested. The characteristic peaks at 379, 418, 429, 578, 644 and
tion tester equipped with Vickers indenter at a preselected maximal 750 cm−1 correspond to α-Al2O3, which is the most common form
load, Pm. Then the nominal half diagonal length (a) and nominal of crystalline aluminium oxide (corundum). Both brackets showed
hardness (Hn) were calculated according to the following equations: identical spectra before and after in vivo ageing.
2
a = 3.5 hm (1) and Hn = Pm /(24.5 hm )(1)
IIT and fracture toughness
where hm stands for the maximum indentation depth. Figure 2 illustrates representative force-indentation depth curves
The elastic index is determined by dividing the elastic We to total from all groups tested. It may be worth noting that despite the vari-
work Wt; both of them are calculated by integrating the areas under ations found within groups, curves from same bracket demonstrated
the unloading and loading curves, respectively. The length of four good reproducibility. All data were drawn from normally distributed
cracks originated from the four corners of impression (Figure 3) were populations and share equal standard deviations as determined after
measured, and the average length was determined as c = (c1 + c2 + Kolmogorov–Smirnov and equal variance tests (P > 0.05).
c3 + c4)/4. Then, one of the following conditions must be satisfied: In Figure 3, representative images of Vickers impression from the
surfaces of both brackets are presented. All cracks are originated
1.5 ≤ c/a ≤ 6 with 0.3 ≤ We /Wt ≤ 0.71(2)

or

1.05 ≤ c/a ≤ 1.5 with 0.591 − 0.194 (c/a)

≤ We /Wt We /Wt ≤ 0.71(3)

Should neither of the aforementioned conditions was satisfied, a dif-

ferent Pm must be selected and the abovementioned procedure must
be repeated. The modulus of elasticity E is given by the formula:

1 − ν2
E= 6 We m
m= 1 [ a m ( Wt ) ] 1.32(1−νi2 )
−
Hn Ei (4)
Figure 1 Representative Raman spectra of as received and retrieved ceramic
where Ei (1141 GPa) and vi (0.07) are the modulus of elasticity and brackets. For the sake of Clarity, only one spectrum is presented for each
Poisson ratio of Vickers indenter, respectively, ν (0.22) the Poisson bracket.
4 European Journal of Orthodontics, 2019

from the four corners of Vickers impression as expected for poly- possibility that differences in mechanical properties are related to
crystalline alumina bracket (Clarity) (Figure 3A). In contrast, an phase transformation such in case of tetragonal stabilized zirconia
irregular crack distribution pattern was found for the single-crystal (26). However, both brackets were found to consists of the stable
alumina bracket (Radiance) (Figure 3B), which precluded the meas- corundum alumina phase, and the structure of both brackets is unaf-
urement of KIC, because of the inability to define crack lengths. fected by intraoral ageing as shown in Figure 1.

Downloaded from https://academic.oup.com/ejo/advance-article-abstract/doi/10.1093/ejo/cjz024/5476400 by Stockholm University Library user on 23 April 2019

The mean values and standard deviations of HM, EIT, ηIT and KIC All mechanical properties studied (HM, EIT, ηIT and KIC) showed
are presented in Figure 4. No statistically significant differences were no significant differences before and after intraoral ageing imply-
found between brackets or between before and after in vivo age- ing that, contrary to what is predicted by the theoretical model, the
ing with the exception of ηIT between brackets (Figure 4). The effect mechanical properties are not affected by the short-term intraoral
of different brackets (single crystal versus polycrystalline) did not exposure. In addition, the findings indicated that both bracket
seem to be depended on the treatment or state (control versus in vivo groups have similar mechanical properties (apart from elastic
aged) as there was no statistically significant interaction between index), a finding that is appended to the fact that they are made of
brackets and condition (P > 0.05) for all properties tested. the same material. The higher elastic index of single-crystal bracket
(Clarity) denotes a more brittle material compared to polycrystal-
line one (Radiance). To the best of authors’ knowledge, there has
Discussion been no previously reported assessment of the properties reported in
Based on the results of this study, the null hypothesis must be this study and thus no comparison can be made with data derived
accepted as no significant differences were found between the con- from published studies as those were just provided data of the raw
trol and retrieved alumina brackets tested. material (alumina) derived from relevant tables found in technical
The usefulness of Raman analysis as applied to the study of bulletins.
the specimens was twofold: it aided to identify the structure of From the early introduction of ceramic brackets to the profes-
alumina and allocates possible differences between single-crystal sion, the paramount importance of KIC in clinical behaviour was
and polycrystalline brackets, and also, it assists in excluding the highlighted (6,10,27). Given that KIC of alumina is considered
20–40 times lower than stainless steel (13,28), the fracture of alu-
mina brackets is more likely to occur compared with metallic ones.
Although a range of values has been given for single-crystal and
polycrystalline alumina brackets, these values have never been meas-
ured experimentally as noted previously. The reason was that KIC
measurement requires bulky specimens with special notches (25),
which are impossible to be induced in brackets because of the size
and complex morphology of orthodontic appliances. An alternative
is to measure the KIC with the indentation method, which has been
widely used for this purpose mainly due to superiority in efficiency
and convenience (25).
KIC for the polycrystalline Clarity was found to be 1.9 MPa m1/2,
which is lower than the value of 99.9% hot-pressed or sintered,
pure, fine-grained alumina (3–4 MPa m1/2), sintered or glass-bonded
alumina (2.5–5 MPa m1/2), or coarse-grained alumina (3–5 MPa
m1/2) (29). This lower value may be attributed to material quality
Figure 2 Representative force indentation depth curves from all groups and/or manufacturing processes, which is considered proprietary.
tested. The curves from each bracket showed good reproducibility, but all the Testing of the single-crystal alumina bracket (Radiance) provided
groups showed great variation in tested mechanical properties. an irregular pattern of cracks around Vickers indentations, which is

Figure 3 (Α) Representative image of Vickers impression from the surface of Clarity. All cracks are originated from the four corners of Vickers impression. The
C1 and C2 denote the crack length from the centre of impression. The other two (C3 and C4) are not shown for the sake of clarity. (B) Representative image from
the surface of Radiance. The longest cracks are not originated from the corners of impression (white arrows), while a crack perpendicular to the left corner is
presented. Chipping of adjacent area at the right corner is easily identified (bar: 50 μm).
E. Alexopoulou et al. 5

Downloaded from https://academic.oup.com/ejo/advance-article-abstract/doi/10.1093/ejo/cjz024/5476400 by Stockholm University Library user on 23 April 2019

Figure 4 Mean values and standard deviations of all properties tested. Numerical values are presented within the bars. Horizontal lines connect mean values
with significant statistical differences.

typical for single-crystal alumina as the crack propagation is domi- to explain the failure of brittle materials, which showed a paradox:
nated by crystal orientation (30). The estimation of KIC, in this case, the stress required to fracture bulk glass ranged around 100 MPa,
was not feasible, and this might constitute a limitation of the inden- whereas the theoretical stress estimated to fracture it was 100 times
tation technique, reported for the first time in the application of this higher. Therefore, there was a necessity to introduce a theory to
methodology to orthodontic ceramics. explain the contradicting evidence. Griffith suggested that the low
The calculation of KIC gives the ability to estimate the critical size fracture strength observed in experiments, as well as the size-depend-
of crack (α), which would lead to catastrophic failure according to ence of strength, was due to the presence of microscopic flaws in the
equation (10): bulk material. Because the expression derived by Griffith’s work pro-
vided the critical stress as a function of the surface energy through
K2IC critical surface tension of the material, and since this term is reduced
a= = 0.6 µ m
π∗σF2 (7) when the brittle material is wetted, the critical stress required to
induce failure was hypothesized to be reduced at wet conditions.
Although this has been already demonstrated in dental ceramics
where σF is the fracture strength.
(31,32), it was not verified in the case of ceramic brackets mainly
The abovementioned value is almost 10 times lower than the 5.5 μm
critical crack size (10), which has been calculated employing the most because the effect of internal defects in the form of cracks arising
widely reported value of KIC (5.8 MPa m1/2). This implies that the mate- from the manufacturing of ceramic brackets seems to modify the
rial may be more crack susceptible than previously considered, since loading pattern and initiation of stress-induced cracks.
even minute cracks can lead to catastrophic failure if loaded at σF. The The results of this study also have significant clinical implica-
aforementioned calculations also demonstrate the reason for the tremen- tions. Given that none of the tested properties showed a reduction
dous impact of this property on the performance of brittle materials and after intraoral exposure, the fracture of orthodontic brackets during
explain the frequent fracture of ceramic brackets during routine ortho- deboning is not associated with the ageing-related inferior mechani-
dontic mechanics, especially at debonding. cal properties but probably with variations in their mechanical
Fracture mechanics was developed during World War I by an properties as a result of manufacturing-induced crack formation and
English aeronautical engineer Griffith, hence the term Griffith crack, distribution in the material.
6 European Journal of Orthodontics, 2019

Conclusions 13. Viazis, A.D., Chabot, K.A. and Kucheria, C.S. (1993) Scanning electron
microscope (SEM) evaluation of clinical failures of single crystal ceramic
The mechanical properties of single-crystal and polycrystalline alu- brackets. American Journal of Orthodontics and Dentofacial Orthope-
mina brackets are not affected by exposure to a wet environment in dics, 103, 537–544.
contrast to the prediction of theoretical model, possibly because of 14. Rhodes, R.K., Duncanson, M.G. Jr, Nanda, R.S. and Currier, G.F. (1992)
the formation of internal cracks during the manufacturing process. Fracture strengths of ceramic brackets subjected to mesial-distal archwire

Downloaded from https://academic.oup.com/ejo/advance-article-abstract/doi/10.1093/ejo/cjz024/5476400 by Stockholm University Library user on 23 April 2019

Single-crystal and polycrystalline brackets share equal mechani- tipping forces. The Angle Orthodontist, 62, 67–76.
cal Martens hardness and elastic modulus, but single crystal demon- 15. Matasa, C.G. (1999) Impact resistance of ceramic brackets according
strated higher brittleness. to ophthalmic lenses standards. American Journal of Orthodontics and
Dentofacial Orthopedics, 115, 158–165.
16. Holt, M.H., Nanda, R.S. and Duncanson, M.G. Jr (1991) Fracture resist-
ance of ceramic brackets during arch wire torsion. American Journal of
Funding
Orthodontics and Dentofacial Orthopedics, 99, 287–293.
No funding was obtained for this study from any source. 17. Wool, A.L. (1992) A better debonding procedure. American Journal of
Orthodontics and Dentofacial Orthopedics, 102, 84–86.
18. Chate, R.A. (1993) Safer orthodontic debonding with rubber dam. Ameri-
Conflicts of interest can Journal of Orthodontics and Dentofacial Orthopedics, 103, 171–174.
None to declare. 19. Britton, J.C., McInnes, P., Weinberg, R., Ledoux, W.R. and Retief, D.H.
(1990) Shear bond strength of ceramic orthodontic brackets to enamel.
American Journal of Orthodontics and Dentofacial Orthopedics, 98, 348–
References 353.
1. Harris, A.M., Joseph, V.P. and Rossouw, P.E. (1992) Shear peel bond 20. Drummond, J.L., King, T.J., Bapna, M.S. and Koperski, R.D. (2000)
strengths of esthetic orthodontic brackets. American Journal of Ortho- Mechanical property evaluation of pressable restorative ceramics. Dental
dontics and Dentofacial Orthopedics, 102, 215–219. Materials, 16, 226–233.
2. Eliades, T., Eliades, G. and Brantley, W. (2001) Orthodontic brackets. In 21. Eliades, T., Eliades, G., Brantley, W.A. and Watts, D.C. (2005) In Vivo
Brantley, W., Eliades, T. (eds), Orthodontic Materials. Thieme, Stuttgart, Ageing of Dental Materials: Aging and Related Phenomena. Quintessence,
Germany, pp. 151–169. Chicago, IL.
3. Eliades, T. (2007) Orthodontic materials research and applications: part 22. ISO14577-1 2002 Metallic Materials – Instrumented Indentation Test for
2. Current status and projected future developments in materials and bio- Hardness and Materials Parameters. International Organization for Stand-
compatibility. American Journal of Orthodontics and Dentofacial Ortho- ardization, Geneva, Switzerland.
pedics, 131, 253–262. 23. Zinelis, S., Al Jabbari, Y.S., Gaintantzopoulou, M., Eliades, G. and Elia-
4. Winchester, L.J. (1991) Bond strengths of five different ceramic brackets: des, T. (2015) Mechanical properties of orthodontic wires derived by
an in vitro study. European Journal of Orthodontics, 13, 293–305. instrumented indentation testing (IIT) according to ISO 14577. Progress
5. Karamouzos, A., Athanasiou, A.E. and Papadopoulos, M.A. (1997) Clinical in Orthodontics, 16, 19.
characteristics and properties of ceramic brackets: a comprehensive review. 24. Zinelis, S., Sifakakis, I., Katsaros, C. and Eliades, T. (2014) Microstruc-
American Journal of Orthodontics and Dentofacial Orthopedics, 112, 34–40. tural and mechanical characterization of contemporary lingual orthodon-
6. Bishara, S.E. and Fehr, D.E. (1997) Ceramic brackets: something old, tic brackets. European Journal of Orthodontics, 36, 389–393.
something new, a review. Seminars in Orthodontics, 3, 178–188. 25. Ma, D., Wang, J., Sun, L. (2017) Methodology for measuring frac-
7. Keith, O., Kusy, R.P. and Whitley, J.Q. (1994) Zirconia brackets: an eval- ture toughness of ceramic materials by instrumented indentation test
uation of morphology and coefficients of friction. American Journal of with Vickers indenter. Journal of the American Ceramic Society, 100,
Orthodontics and Dentofacial Orthopedics, 106, 605–614. 2296–2308.
8. Springate, S.D. and Winchester, L.J. (1991) An evaluation of zirconium 26. Zhuang, Y., Zhu, Z., Jiao, T. and Sun, J. (2019) Effect of aging time and
oxide brackets: a preliminary laboratory and clinical report. British Jour- thickness on low-temperature degradation of dental zirconia. Journal of
nal of Orthodontics, 18, 203–209. Prosthodontics, 28, e404–e410.
9. Jeiroudi, M.T. (1991) Enamel fracture caused by ceramic brackets. Ameri- 27. Scott, G.E. Jr (1988) Fracture toughness and surface cracks – the key to
can Journal of Orthodontics and Dentofacial Orthopedics, 99, 97–99. understanding ceramic brackets. The Angle Orthodontist, 58, 5–8.
10. Kusy, R.P. (1988) Morphology of polycrystalline alumina brackets and its 28. Ghafari, J. (1992) Problems associated with ceramic brackets suggest lim-
relationship to fracture toughness and strength. The Angle Orthodontist, iting use to selected teeth. The Angle Orthodontist, 62, 145–152.
58, 197–203. 29. Quin, G. (2007) Fractography of Ceramic and Glasses. National Institute
11. Cochrane, N.J., Lo, T.W.G., Adams, G.G. and Schneider, P.M. (2017) of Standards and Technology. Washington, DC
Quantitative analysis of enamel on debonded orthodontic brackets. 30. Wang, K., Jiang, F., Yan, L., Xu, X., Wang, N., Zha, X., Lu, X. and
American Journal of Orthodontics and Dentofacial Orthopedics, 152, Wen, Q. (2019) Study on mechanism of crack propagation of sapphire
312–319. single crystals of four different orientations under impact load and static
12. Gibbs, S.L. (1992) Clinical performance of ceramic brackets: a survey load. Ceramics International, 45, 7359–7375.
of British orthodontists’ experience. British Journal of Orthodontics, 19, 31. Drummond, J.L., Novickas, D. and Lenke, J.W. (1991) Physiological aging
191–197. of an all-ceramic restorative material. Dental Materials, 7, 133–137.