Basic Research—Technology

Effects of Root Canal Irrigants on the Push-out Strength and

Hydration Behavior of Accelerated Mineral Trioxide
Aggregate in its Early Setting Phase
Seong-Tae Hong, DDS, MSD,* Kwang-Shik Bae, DDS, PhD,† Seung-Ho Baek, DDS, PhD,†
Kee-Yeon Kum, DDS, PhD,† Won-Jun Shon, DDS, PhD,† and WooCheol Lee, DDS, PhD†

Abstract
Introduction: The purpose of this study was to evaluate
the effects of endodontic irrigants on the push-out
strength and hydration behavior of accelerated mineral
M ineral trioxide aggregate (MTA) has been widely used as a promising biomaterial
trioxide aggregate (MTA) in its early setting phase.
Methods: In an in vitro perforation model, MTA with
or without 10% CaCl2 was condensed and allowed to
initial set for 10 minutes. The samples were divided
into four groups (n = 10) to be immersed into either
3.5% sodium hypochlorite (NaOCl) or 2% chlorhexidine
gluconate (CHX) for 30 minutes and then allowed to set
for 48 hours. In the control group, a wet cotton pellet was
placed over MTA. The maximum force applied to the
set MTA mixture before dislodgement was recorded.
Irrigant-treated surfaces were examined using a scanning
electron microscope (SEM). Chemical elements of these
surfaces were also analyzed by energy dispersive x-ray
spectroscope (EDS). Results: The push-out strength of
group A1 (NaOCl-treated accelerated MTA) was the
highest of all groups. When compared with nonacceler-
ated MTA, CaCl2-accelerated MTA showed significantly
higher push-out strength (p < 0.05). NaOCl-treated
groups showed significantly higher push-out strength
than CHX-treated groups (p < 0.05). Scanning electron
microscopic examination and EDS analysis showed that
the formation of calcium hydroxide crystals on acceler-
ated MTA exposed to NaOCl was increased compared
with those of the control group. Conclusion: These find-
ings imply that the use of accelerated MTA under the
NaOCl irrigation was effective in perforation repair
without altering its hydration behavior even in the early
setting phase. (J Endod 2010;36:1995–1999)
Key Words
Accelerated mineral trioxide aggregate, calcium hydro-
xide crystal, hydration behavior, perforation model,
push-out strength
to repair root perforation because of its excellent biocompatibility, superior seal-
ing, and ability of setting in the presence of blood (1-3). Despite the numerous favorable
properties that support its clinical use, the prolonged setting time has been a critical
drawback of MTA. This slow setting property results in a challenge to the handling
and washout of newly placed MTA in the repaired area by immediate irrigation
procedure (4, 5). Therefore, it is recommended to place a moist cotton pellet over
MTA for at least 4 hours to allow its complete setting. Then, patients have to be
recalled for another visit for the completion of the treatment. This protocol has
several disadvantages, such as an increased number of appointments and the risk of
coronal leakage. In this regard, Kogan et al (6) reported that MTA mixed with sodium
hypochlorite gel could be recommended for single-visit procedures because it
improved the working properties and decreased the setting time of this material.
Nandini et al (7) also showed that the placement of glass-ionomer cement over partially
set MTA did not affect its setting reaction and could be used for a single-visit procedure.
MTA is composed of hydrophilic powder, which reacts with water and sets into
a hard mass by forming calcium hydroxide and calcium silicate hydrate gel (8).
During the hydration process, various physiologic environments and solutions can
affect the physical properties of MTA (9, 10). With respect to this, Lee et al (9) re-
ported that an acidic environment of pH 5 adversely affected both the physical prop-
erties and the hydration behaviors of MTA. The microhardness of the pH 5 group was
significantly weaker compared with the other groups, and the erosion of the cubic
crystal surfaces was detected by a scanning electron microscope (SEM). It is
supposed that the hydration process of MTA is adversely affected by physiologic envi-
ronments because of its long setting time. Therefore, if the setting reaction of MTA is
accelerated, the more predictable result can be expected without the influence of the
environmental factors.
Recently, a number of studies have been introduced to decrease the setting time of
MTA, which showed not only the sealing ability but also its physicochemical property
was improved by the addition of a setting accelerator, such as 10% calcium chloride
(CaCl2) (11–16). However, there has been no study regarding the influence of the
environmental factors on the physical properties and hydration behavior of
accelerated MTA. We hypothesized that the early use of the endodontic irrigants can
weaken the bond strength of accelerated MTA-dentin interface and interfere with the
hydration behavior of accelerated MTA. Therefore, the purpose of this study was to eval-
uate the push-out strength and hydration behavior of accelerated MTA when it was in
contact with the endodontic irrigants in its early setting phase.

From the * Department of Dentistry, School of Medicine, Ajou University, Suwon, Korea; and † Department of Conservative Dentistry, School of Dentistry & Dental
Research Institute, Seoul National University, Seoul, Korea.
Address requests for reprints to Dr WooCheol Lee, Department of Conservative Dentistry, School of Dentistry & Dental Research Institute, Seoul National University,
275-1 Yeongeon-Dong, Jongno-Gu, Seoul, Republic of Korea 110-768. E-mail address: jimin525@snu.ac.kr.
0099-2399/$ - see front matter
Copyright ª 2010 American Association of Endodontists.
doi:10.1016/j.joen.2010.08.039

JOE — Volume 36, Number 12, December 2010 Effects of Root Canal Irrigants on MTA in Early Setting Phase 1995
Basic Research—Technology
Figure 1. A schematic illustration of the perforation model design. (This figure is available in color online at www.aae.org/joe/.)
Materials and Methods
Preparation of the Perforation Model
Single-rooted, extracted human teeth were used for this study. The
coronal part of the tooth was removed, and midroot dentin was horizon-
tally sectioned into 1.5-mm thick slices with a diamond saw microtome.
To make a perforation model, the existing canal space of the dentin slice
was instrumented with Gates Glidden burs (Mani, Utsunomiya, Japan)
#2 through #5 to a diameter of 1.3 mm. The dentin slice was then set
into a cylindrical steel frame, and the inner space of the frame was filled
with resin that was allowed to set for 24 hours. After polymerization, the
sample was extracted from the frame with the tooth section flush with
the upper surface and with the perforation in the axis of the cylinder
(Fig. 1). Finally, the perforation was enlarged with a diamond bur
(SF-12; Mani, Tochigi, Japan) to a standardized diameter of 1.5 mm.
The MTA mixtures (ProRoot MTA; Dentsply Tulsa, Tulsa, OK), with
or without 10% CaCl2, were prepared as shown in Table 1. These
mixtures were placed inside the perforations and were condensed
with S-Kondenser (Analytic Endodontics, Glendora, CA) on saline-
moistened Oasis. This setup provided a model simulating periapical
tissue. Excess material was trimmed from the surface of the samples
with a scalpel. These samples were wrapped in wet gauzes and sealed
TABLE 1. Preparation of Mixtures and Experimental Groups
Used
Cements Mixtures Irrigants
Accelerated 1.0 g MTA, 0.1 g CaCl2, 3.5% NaOCl
MTA 0.25 ml DW 2% CHX
MTA 1.0 g MTA, 0.3 ml DW 3.5% NaOCl
2% CHX
MTA, mineral trioxide aggregate; DW, distilled water; NaOCl, sodium hypochlorite; CHX, chlorohex-
idine gluconate.
1996 Hong et al.
in the plastic bags, placed in an incubator, and allowed to initial set
for 10 minutes at 37 C with 100% humidity. Then, the samples were
divided into four groups (n = 10) to be immersed into either 3.5%
sodium hypochlorite or 2% chlorhexidine gluconate in glass plates
for 30 minutes (Table 1). The samples were removed from the irrigants
and rinsed with distilled water and then allowed to set for 48 hours at
37 C with 100% humidity. As a control group, a wet cotton pellet was
placed over MTA and allowed to set for 48 hours (n = 10).
Push-out Strength Test
The samples were placed in an aligning device with a cylindrical
steel holder that held it centered below a punch of the steel piece.
These samples made by pouring resin into a frame could be accurately
placed in the aligning device because they have uniform cylindrical
shapes. This aligning device was then fixed to a materials testing
machine (LF Plus; Lloyd instruments Ltd, Hampshire, UK) that moved
one piece against the other with a constant speed of 1 mm/min so that
the punch of the upper piece pushed the set MTA mixture in the spec-
imen. The underlying space free of resin could allow the free motion
of the punch and the dislodgement of the mixture. The maximal force
applied to mixture before dislodgement occurred was recorded in
MPa.
TABLE 2. Push-out Strengths of the Samples after 48 Hours (MPa)
Groups Push-out Value
Groups A1 98.06  9.18a
A1 A2 82.18  13.68ad
A2 M1 63.13  18.03b
M1 M2 31.33  13.40c
M2 Control 66.34  6.74bd
Values are means  standard deviations. Groups identified by the same superscript letters are not
significantly different (p > 0.05). Different letters identify significantly different groups (p < 0.05).
JOE — Volume 36, Number 12, December 2010
 Basic Research—Technology

Figure 2. SEM examinations of the samples (10,000). (A) A typical cluster of hexagonal crystalline was observed on the surface of MTA in the control group. (B)
Group A1 showed surface crystalline outgrowths, and their size and amount were increased compared with the control group. (C) Group M1 showed the unde-
veloped cubic crystals with uneven surfaces, and the amount of crystals was lower than that of the control group. (D) Group A2 showed little surface crystalline
formation; the surface crystals had thin plate structures, and their size was reduced almost to one tenth of those of the control group. (E) Group M2 showed the
similar appearances with group A2.

The Statistical Package for Social Sciences program (SPSS 12.0;
SPSS Inc, Chicago, IL) was used to perform the data analysis. Two-
way analysis of variance was used to analyze whether there were differ-
ences in the push-out strength on the basis of the endodontic irrigants
and CaCl2 accelerator. Scheffe multiple comparison testing was used to
compare the significant differences found between the groups. A prob-
ability value of less than 0.05 was considered to be significant.
Scanning Electron Microscopic Examination
The irrigant-treated surface of MTA was examined with an SEM (S-
4700; Hitachi, Tokyo, Japan). Two samples were prepared from each of
the different groups, and they were mounted on aluminum stub and
sputter-coated with a 300-Å gold layer. Magnification considered
adequate to characterize the microstructure (10,000) was selected,
and the surface characteristics of the samples were recorded with the
digital imaging system.
Energy Dispersive X-ray Spectroscopic Analysis
Energy dispersive x-ray spectroscopic (EDS) analysis was
executed using EMAX (Horiba, London, UK), which was attached to
an SEM, for chemical element analysis of the irrigant-treated surface
of MTA. The high magnification of 10,000 was selected to discern
chemical compositions of specific crystal types within a sample. Via
this process, a spectrum will be obtained, and elements can be identi-
fied. A semiquantitative standardless analysis was performed from these
spectra to derive atomic percent concentrations of the elements.
Results
Push-out Strength Test
Table 2 shows the mean push-out strength in each group. The
push-out strength of group A1 was the highest of all the groups, and
it was significantly higher than the nonaccelerated groups (M1 and
M2) (p < 0.05). When compared with the control group, the push-
out strength of the group A1 was significantly increased (P < .05)
and that of the M2 group was significantly reduced (p < 0.05).
When comparing accelerated MTA (A1 and A2) with nonacceler-
ated MTA (M1 and M2), accelerated MTA showed significantly higher
push-out strength than nonaccelerated MTA (p < 0.05). In addition,
sodium hypochlorite (NaOCl)-treated groups (A1 and M1) showed
significantly higher push-out strength than chlorhexidine gluconate
(CHX)-treated groups (A2 and M2) (p < 0.05).
Scanning Electron Microscopic Examination
A typical cluster of hexagonal crystalline was observed on the
surface of MTA in the control group (Fig. 2A). When they were con-
tacted with NaOCl, accelerated MTA samples showed a different surface
morphology with the control. On the surface of NaOCl-treated acceler-
ated MTA, crystalline outgrowths were detected and their size and
amount were increased although the surfaces of the cubic crystals ap-
peared uneven (Fig. 2B). On the contrary, the undeveloped cubic crys-
tals were shown on the surface of nonaccelerated MTA when it was
exposed to NaOCl (Fig. 2C). When they were contacted with CHX, accel-
erated MTA samples showed similar appearances as the nonaccelerated
MTA samples. In general, there was little crystalline formation on the
surface of the CHX-treated samples, and it was smaller in size and
less in amount than the control and was covered with an amorphous
layer that was rarely washed with the distilled water (Fig. 2D and E).
EDS Analysis
The EDS spectrums for elemental identification showed that the
crystals that were observed on the surface of MTA in the control group
were composed of calcium and oxygen (Fig. 3A). When the same irri-
gants were used, there was no distinct difference between accelerated
and nonaccelerated MTA. However, the samples had different chemical

JOE — Volume 36, Number 12, December 2010 Effects of Root Canal Irrigants on MTA in Early Setting Phase 1997
Basic Research—Technology

Figure 3. EDS analysis of the samples. (A) The typical hexagonal crystals on the surface of MTA in the control group were composed of calcium and oxygen. (B)
The surface crystals of group A1 showed similar chemical distributions with the control group, and they were composed of calcium, oxygen, and carbon. EDS
analysis of group M1 has not appeared here because it had the similar pattern with that of group A1. (C) Group A2 showed the different chemical distributions
with the control group, and silicon was detected along with calcium, oxygen, and carbon. EDS analysis of group M2 does not appear here because it had a similar
pattern to that of group A2.

elements according to the used irrigants. The NaOCl-treated samples
showed similar patterns with the control group, and they were
composed of calcium, oxygen, and carbon (Fig. 3B). On the contrary,
the crystals on the surfaces of the CHX-treated samples had different
chemical distributions, and they consisted of calcium, oxygen, carbon,
and silicon (Fig. 3C).
Discussion
We designed our study to evaluate the effects of the early exposure
of endodontic irrigants on accelerated MTA in its early setting phase.
First, we investigated the resistance of accelerated MTA to push-out
force after exposing accelerated MTA to endodontic irrigants before
its final setting. If the irrigants inhibited the hydration and setting of
MTA, we would get a low value of push-out force. This may be an impor-
tant factor determining the restoration success after root canal therapy
(17). Therefore, we hypothesized that the early exposure of the irrigants
would weaken the bond strength of accelerated MTA-dentin interface.
In this study, the push-out strength of the accelerated MTA was
significantly increased compared with nonaccelerated MTA (p <
0.05). A possible explanation to this result is that the setting accelerator
(CaCl2) may alter the chemical composition, surface area, and charac-
teristics of the pores of cements, providing the advantages of increased
resistance to compression (18). The NaOCl-contacted groups also
showed a significantly higher push-out strength than the CHX groups.
The accelerated MTA exposed to NaOCl resulted in a significantly higher
push-out strength than the control group. This means that NaOCl did not
interrupt the accelerated MTA from hardening and rather enhanced its
setting process. Therefore, our first hypothesis was rejected. In this re-
gard, Kogan et al (6) investigated the effects of various additives on
setting properties of MTA and showed that MTA mixed with NaOCl gel
improved the setting process. NaOCl is also known as a halogenated
compound that can cause mineral accumulation in human root dentin
(19). With respect to this, Uyanik et al (20) reported that the sealing
ability of furcal perforation repaired with MTA was affected by exposure
to the irrigation regimens and the samples irrigated with NaOCl had the
lowest leakage values.
Accelerated MTA exposed to CHX showed a higher push-out value
than the control group, but there was no significant difference.
However, nonaccelerated MTA exposed to CHX had a significantly lower
value than the control group (p < 0.05). These results were consistent
with those of Nandini et al (21) who showed that 2% CHX reduced the

1998 Hong et al. JOE — Volume 36, Number 12, December 2010

surface hardness of set white MTA significantly after 24 hours and sug-
gested that CHX irrigation within 24 hours of placement of white MTA
should be avoided.
Our second hypothesis was that the early exposure of irrigants
would interfere with the hydration behavior of accelerated MTA. We
observed the surfaces of the completely hardened MTA after treating
with endodontic irrigants in the early setting phase. We focused espe-
cially on investigating whether calcium hydroxide formation was in-
hibited by endodontic irrigants because calcium hydroxide crystal as
a reaction product on the surface of MTA could play an important
role in maintaining the high alkalinity of MTA (8, 22).
The results of the scanning electron microscopic observations in
this study showed that typical hexagonal crystal structure was observed
on the surface of MTA in the control group. This surface crystal turned
out to be calcium hydroxide by EDS analysis, and this finding was
consistent with other studies (9, 10). When the accelerated MTA was
exposed to NaOCl in the early setting phase, the formation of calcium
hydroxide was not inhibited by NaOCl, and our second hypothesis
was also rejected when NaOCl was used as an irrigant. The surface
crystals were fairly well maintained, and their size and amount were
increased compared with the control group. On the contrary,
nonaccelerated MTA exposed to NaOCl produced round crystal
structures with uneven surfaces, and the amount of crystals was
lower than that of the control group. Thus, NaOCl increased the size
and amount of calcium hydroxide crystals in accelerated MTA,
whereas it had a detrimental effect on the formation of calcium
hydroxide in nonaccelerated MTA.
When CHX was used as an irrigant, we could not observe the
apparent crystal structure on the surface of both accelerated and non-
accelerated MTA. The surface crystals had the thin plate structures, and
their size was reduced almost to one tenth of those of the control group
(Fig. 2D and E). In EDS analysis, silicon was detected along with
calcium, oxygen, and carbon, which proved that they were not the
typical calcium hydroxide crystals (Fig. 3C). These findings may explain
why the push-out strength of the CHX groups was significantly reduced
compared with the NaOCl groups. To our knowledge, this is the first
report that showed that CHX had the adverse influence on physical prop-
erties and hydration behaviors of MTA when it was in contact with MTA
before the initial setting.
When accelerated MTA was used for perforation repair and it was
exposed to NaOCl irrigation in its early setting phase, the push-out
strength of accelerated MTA was improved, and the formation of
calcium hydroxide crystals on the surface of accelerated MTA was
increased. Therefore, we presume that, with its fast setting action and
improved physical property, the use of accelerated MTA could make
it possible to perform the endodontic procedure and the perforation
repair in a single visit.
However, we could not exactly reproduce clinical situations when
using endodontic irrigants. Therefore, the possibility of washout by
injection pressure of the irrigants may not be excluded. Further studies
JOE — Volume 36, Number 12, December 2010
Basic Research—Technology
are indicated to investigate the sealing ability of accelerated MTA, which
is one of the important factors in perforation repair, and the initial solu-
bility of accelerated MTA by endodontic irrigants.
References
1. Apaydin ES, Shabahang S, Torabinejad M. Hard-tissue healing after application of
fresh or set MTA as root-end-filling material. J Endod 2004;30:21–4.
2. Ferris DM, Baumgartner JC. Perforation repair comparing two types of mineral
trioxide aggregate. J Endod 2004;30:422–4.
3. Hardy I, Liewehr FR, Joyce AP, et al. Sealing ability of One-Up Bond and MTA with
and without a secondary seal as furcation perforation repair materials. J Endod
2004;30:658–61.
4. Chng HK, Islam I, Yap AU, et al. Properties of a new root-end filling material.
J Endod 2005;31:665–8.
5. Wiltbank KB, Schwartz SA, Schindler WG. Effect of selected accelerants on the phys-
ical properties of mineral trioxide aggregate and Portland cement. J Endod 2007;33:
1235–8.
6. Kogan P, He J, Glickman GN, et al. The effects of various additives on setting prop-
erties of MTA. J Endod 2006;32:569–72.
7. Nandini S, Ballal S, Kandaswamy D. Influence of glass-ionomer cement on the inter-
face and setting reaction of mineral trioxide aggregate when used as a furcal repair
material using laser raman spectroscopic analysis. J Endod 2007;33:167–72.
8. Camilleri J, Montesin FE, Brady K, et al. The constitution of mineral trioxide aggre-
gate. Dent Mater 2005;21:297–303.
9. Lee YL, Lee BS, Lin FH, et al. Effects of physiological environments on the hydration
behavior of mineral trioxide aggregate. Biomaterials 2004;25:787–93.
10. Tingey MC, Bush P, Levine MS. Analysis of mineral trioxide aggregate surface when
set in the presence of fetal bovine serum. J Endod 2008;34:45–9.
11. Abdullah D, Ford TR, Papaioannou S, et al. An evaluation of accelerated Portland
cement as a restorative material. Biomaterials 2002;23:4001–10.
12. Bortoluzzi EA, Broon NJ, Bramante CM, et al. Sealing ability of MTA and radiopaque
Portland cement with or without calcium chloride for root-end filling. J Endod
2006;32:897–900.
13. Bortoluzzi EA, Broon NJ, Duarte MAH, et al. The use of a setting accelerator and its
effect on pH and calcium ion release of mineral trioxide aggregate and white Port-
land cement. J Endod 2006;32:1194–7.
14. Ber BS, Hatton JF, Stewart GP. Chemical modification of ProRoot MTA to improve
handling characteristics and decrease setting time. J Endod 2007;33:1231–4.
15. Hong ST, Bae KS, Baek SH, et al. Microleakage of accelerated mineral trioxide aggre-
gate and Portland cement in an in vitro apexification model. J Endod 2008;34:56–8.
16. Bortoluzzi EA, Broon NJ, Bramante CM, et al. The influence of calcium chloride on
the setting time, solubility, disintegration, and pH of mineral trioxide aggregate and
white Portland cement with a radiopacifier. J Endod 2009;35:550–4.
17. Sluyk SR, Moon PC, Hartwell GR. Evaluation of setting properties and retention char-
acteristics of mineral trioxide aggregate when used as a furcation perforation repair
material. J Endod 1998;24:768–71.
18. Ramachandran VS. Concrete Admixtures Handbook. Park Ridge, NJ: Noyes Pub;
1984.
19. Inaba D, Ruben J, Takagi O, et al. Effect of sodium hypochlorite treatment on remi-
neralization of human root dentine in vitro. Caries Res 1996;30:218–24.
20. Uyanik MO, Nagas E, Sahin C, et al. Effects of different irrigation regimens on the
sealing properties of repaired furcal perforations. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod 2009;107:e91–5.
21. Nandini S, Natanasabapathy V, Shivanna S. Effect of various chemicals as solvents on
the dissolution of set white mineral trioxide aggregate: an in vitro study. J Endod
2010;36:135–8.
22. Torabinejad M, Hong CU, McDonald F, et al. Physical and chemical properties of
a new root-end filling material. J Endod 1995;21:349–53.
Effects of Root Canal Irrigants on MTA in Early Setting Phase 1999