Mineral trioxide aggregate with anti-washout gel – Properties and microstructure

Mineral trioxide aggregate with anti-washout gel – Properties and microstructure

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306 Available online at www.sciencedirect.com journal homepage: www.intl.elsevierhealth.com/journa...

2MB Sizes 2 Downloads 48 Views

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306

Available online at www.sciencedirect.com

journal homepage: www.intl.elsevierhealth.com/journals/dema

Mineral trioxide aggregate with anti-washout gel – Properties and microstructure L.M. Formosa a , B. Mallia a , J. Camilleri b,∗ a b

Department of Metallurgy and Materials Engineering, Faculty of Engineering, University of Malta, Malta Department of Restorative Dentistry, Faculty of Dental Surgery, University of Malta, Malta

a r t i c l e

i n f o

a b s t r a c t

Article history:

Objective. One of the problems encountered clinically when using mineral trioxide aggregate

Received 23 April 2012

(MTA) as a root-end filling material is washout immediately after placement. A novel MTA is

Accepted 10 November 2012

supplied with an anti-washout gel that replaces the mixing water. The aim of this research was to characterize and assess the properties of a novel MTA mixed with an anti-washout liquid.


Methods. MTA Plus mixed with either water (MTA-W) or an anti-washout gel (MTA-AW)


was investigated. Un-hydrated and set materials were characterized by scanning electron


microscopy (SEM), energy X-ray dispersive analysis (EDX), X-ray diffraction analysis (XRD)

Chemical properties

and Fourier transform infrared spectroscopy (FT-IR) after being stored dry or immersed

Dental materials

in Hank’s balanced salt solution (HBSS). The chemical and physical properties of the set

Mineral trioxide aggregate

materials were then investigated.

Physical properties

Results. The MTA Plus was composed of tricalcium silicate, dicalcium silicate and bismuth

Root-end filling materials

oxide. The anti-washout gel used was water-based and FT-IR plots showed the presence of an organic additive. Both materials immersed in HBSS displayed the presence of reaction by-product with MTA-W exhibiting a high-intensity calcium hydroxide peak on X-ray diffraction. The X-ray diffractograms of all materials following hydration demonstrated the reduction in peak intensity of the tri- and dicalcium silicate. Hydroxyapatite deposits were evident on the surfaces of both materials in contact with HBSS. The pH of the leachate was similar for both materials. MTA-AW exhibited lower levels of calcium ions in solution and reduced fluid uptake in the early stages of reaction. The anti-washout gel reduced the setting time of the cement and enhanced the compressive strength. The radiopacity of both materials was approximately 8 mm aluminum. Significance. The use of the water-based anti-washout material instead of the standard water with MTA affects the hydration and properties of the set material. © 2012 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.



White mineral trioxide aggregate (MTA) is a dental cement with numerous applications including pulp-capping, apexification, repair of root perforations, root-end filling [1] and

others [2]. One of the drawbacks of MTA is washout [3], which refers to the tendency of freshly prepared cement paste to disintegrate upon contact with blood or other fluids [4]. Washout of a root-end filling material can occur when rinsing an osteotomy site [5] resulting in a compromised root-end seal and its associated repercussions.

∗ Corresponding author at: Department of Restorative Dentistry, Faculty of Dental Surgery, University of Malta, Medical School, Mater Dei Hospital, Msida MSD 2090, Malta. Tel.: +356 2340 1174. E-mail address: [email protected] (J. Camilleri). 0109-5641/$ – see front matter © 2012 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dental.2012.11.009

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306

New-generation endodontic materials have been investigated as developments of MTA. Prototype cements Generex-A and Capasio have displayed superior washout resistance compared to MTA, whilst having similar compressive strength, radiopacity and alkaline pH [5], making them potentially suitable replacements. Washout-resistant injectable cements based on different material systems have also been investigated; injectable calcium-phosphate cement with the addition of sodium hyaluronate [6], sodium alginate [4,7] chitosan or modified starch [4], were found to have good washout resistance. Whilst calcium phosphate is primarily indicated for use as a synthetic bone substitute, it has also shown potential as a dental cement [8]. MTA is composed primarily of Portland cement [9]. Portland cement is used as a binder in concrete. Cements based on calcium silicates (the basic components of the Portland cement in MTA) have been shown to exhibit an increase in washout resistance by the addition of carboxymethyl chitosan [10] or gelatin [11]. In the construction industry, concrete employed in building underwater structures has historically had an antiwashout admixture (based on a water-soluble polymer) added to modify its rheological properties and make it more resistant to washout [12]. Since both concrete and MTA use Portland cement as a binder [13], a similar admixture could be expected to impart the same enhanced washout resistance to MTA. A novel mineral trioxide aggregate (MTA PlusTM , Prevest Denpro, Jammu City, India) which is claimed to have a finer particle size than the MTAs currently available for clinical use is provided with either water or an anti-washout gel [14]. The gel is intended to improve its washout resistance [15], but little information about the properties of this new material is available in the literature. An ideal anti-washout agent should (i) inhibit the decay of cement paste in liquid; (ii) not interfere with the hydration reaction or significantly reduce the bioactivity of the cement; (iii) not decrease the mechanical strength of the cement once set; (iv) not worsen the handling properties of the cement paste; (v) not significantly extend the setting time; and (vi) not reduce the radiopacity [4]. The purpose of this study was to characterize and investigate the chemical and physical properties of un-hydrated MTA Plus as well as set cements composed of MTA Plus mixed with the anti-washout gel (MTA-AW), and to compare these properties to MTA Plus mixed with only distilled water, as control (MTA-W).



The materials used in this study included MTA Plus (compounded by Prevest Denpro, Jammu, India for Avalon Biomed Inc., Bradenton, FL, USA) lot #2011022801, mixed with either distilled water at a water to cement ratio of 0.35 or antiwashout gel (compounded by Prevest Denpro, Jammu, India for Avalon Biomed Inc., Bradenton, FL, USA). The gel was dosed by weight (0.350 g gel per gram MTA Plus powder). The testing was performed at 1 day and at 28 days and the materials were stored either dry or immersed in Hank’s balanced salt solution (HBSS; H6648, Sigma Aldrich, St. Louis, MO, USA).


2.1. Characterization of un-hydrated cements, washout gel and set materials 2.1.1.

X-ray diffraction analysis

Phase analysis was carried out on un-hydrated MTA Plus, the anti-washout gel and set cements cured in a sealed, dry container or immersed in HBSS for either 1 or 28 days using X-ray diffraction (XRD). The diffractometer (Rigaku Ultima IV, Rigaku Corporation Tokyo, Japan) used Cu K␣ radiation at 40 mA and 45 kV. Samples were presented in powder form and the detector was rotated between 5◦ and 45◦ , with a sampling width of 0.05◦ and 2◦ /min scan speed. Phase identification was accomplished by use of search-match software utilizing the ICDD database (International Center for Diffraction Data, Newtown Square, PA, USA).


Fourier transform infra-red spectroscopy

Un-hydrated MTA Plus powder and the set cements were analysed using a Fourier transform infrared spectrophotometer (Shimadzu IRAffinity-1; Shimadzu Corp., Kyoto, Japan) using and ATR window diameter of 7 mm, spectral resolution of 4 cm−1 and 45 scans per spectrum. The anti-washout gel was first dried for 48 h in an incubator at 90 ± 1 ◦ C to produce a thin uniform sheet of dry polymer. A portion of the sheet was then cut out with scissors, loaded into the spectrophotometer and analysed in a similar way to the un-hydrated MTA powder.


Chemical analysis of anti-washout gel

The chemical analysis of the anti-washout gel was performed using Energy Dispersive X-ray Fluorescence (EDXRF; Bruker S2 Ranger, Bruker Corporation, Madison USA) using water as the matrix and 4 ␮m liquid prolene film.


Scanning electron microscopy of hydrated cements

Cement cube specimens of side 7 mm, stored dry or immersed in HBSS for 28 days at 37 ± 1 ◦ C, were analysed using scanning electron microscopy (SEM). Microstructural analysis was performed with X-ray energy dispersive analysis (EDX). At the end of the curing period the materials were desiccated and embedded in epoxy resin (Struers Epofix, Struers, Ballerup, Denmark). This was followed by grinding them with progressively finer grits of abrasive paper, polished with diamond paste, washed with isopropanol alcohol, dried and carbon coated (Agar auto carbon coater; Agar Scientific, Essex, England). The materials were viewed under the scanning electron microscope (SEM; Zeiss MERLIN Field Emission SEM, Carl Zeiss NTS GmbH, Oberkochen, Germany) and cement microstructure was assessed in backscatter electron mode. EDX analysis was performed on each phase identified. In addition, plots of calcium–silicon ratios of the cement particle center, the margin of the cement particle and the cement matrix were plotted. The bismuth to calcium ratios of these areas in relation to the bismuth particle were also plotted. This was repeated for cubes stored in HBSS. In addition, cement discs 15 mm in diameter and 1 mm thick were cured for 24 h as described, immersed in HBSS for 28 days, dried and finally carbon coated without polishing the surfaces. The unpolished surfaces were then observed under the scanning electron microscope.


d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306


Chemical properties


pH and calcium ion release in physiological solution

Discs of diameter 15 ± 0.1 mm and height 2 ± 0.5 mm were cast and allowed to cure for 24 h in a dry incubator at a temperature of 37 ± 1 ◦ C. They were then immersed in individual sealed polycarbonate containers containing 20 ± 0.01 ml of HBSS. Three replicate samples of each material were made. Using a pH/mV/ISE meter (Hanna HI 3221, Hanna Instruments, Woonsocket, RI, USA) with a single-junction (Ag/AgCl) ceramic pH electrode (Hanna HI 1131, Hanna Instruments, Woonsocket, RI, USA) the pH of each solution was measured after 1, 7, 14, 21 and 28 days had elapsed from the moment of immersion. Temperature compensation was accomplished by simultaneously immersing a temperature probe (HI 7662, Hanna Instruments, Woonsocket, RI, USA) in the measurement solution. The pH meter was calibrated using three standard calibrating solutions (pH 4.01, 7.01 and 10.00) prior to each set of measurements. The calcium ion concentration in solution was measured at 1, 7, 14, 21 and 28 days using an ISE Calcium Electrode (consisting of Hanna HI 4000-50 Sensor handle and Hanna HI 4004-51 calcium module) and a reference electrode (Hanna HI 5315). Temperature compensation was accomplished by immersing a temperature probe (HI 7662, Hanna Instruments, Woonsocket, RI, USA) in the measurement solution simultaneously. The meter was calibrated using two standard calibrating solutions (100 ppm and 1000 ppm) prior to each set of measurements.

millimeter thickness of the cement under test, was calculated using the method described in a previous publication [16]. Briefly, an image manipulation program (Microsoft Paint; Microsoft Corp., Redmond, WA, USA) was used to determine the gray pixel values (a measure of the “shade of gray” or “color”) of the steps of the step wedge. A spreadsheet program (Microsoft Office Excel 2003; Microsoft Corp., Redmond, WA, USA) was used to plot a chart of gray pixel value vs. thickness of aluminum, and the best-fit exponential trendline was drawn through the points. The equation of this trendline was then used to calculate the radiopacity of the cement discs from their gray-pixel values on the same radiograph.


2.3.3. 2.3.

Physical properties



Cement discs of diameter 10 ± 0.1 mm and height 2 ± 0.5 mm were prepared and incubated for 24 h at 37 ± 1 ◦ C to allow the cement to cure, after which they were either stored in a dry sealed container, or immersed in HBSS inside a sealed container, in each case at 37 ± 1 ◦ C. Radiographs were taken after one day and after 28 days under these conditions. Three replicates of each test were made. To take the radiographs, the specimens were taped to a plastic sleeve containing a photostimulable phosphor (PSP) plate of the same size. An aluminum step-wedge (Everything X-ray, High Wycombe, UK) with steps of incremental thickness 3 mm, was placed beside the discs covering part of the sleeve such that it appeared in the radiographs alongside the specimens. The samples and the wedge were then irradiated with X-rays using a dental X-ray machine (GEC Medical Equipment Ltd., Middlesex, UK) using an exposure time of 0.4 s, tube current and voltage of 10 mA and 65 kV respectively, and a cathode-to-target film distance of 400 ± 10 mm. The radiographs were processed in an automatic processing machine (Clarimat 300, Gendex Dental Systems, Medivance Instruments Ltd., London, UK) and a digital image of the radiograph was obtained. The radiograph was in the form of a greyscale image file. The radiopacity (measured in millimeters of aluminum), denoting the thickness of aluminum required to block X-rays to the extent of one

Setting time

Setting time was evaluated using the procedure set out in ISO 9917-1:2007 [17]. The cement pastes were packed into stainless steel moulds with internal cross-section 10 mm by 8 mm and depth 5 mm. A stopwatch was started and the moulds were immediately placed in an incubator at 37 ± 1 ◦ C, either in air, or immersed in HBSS. Testing for setting was done using a modified Vicat apparatus, consisting of a weighted needle of square cross-section of side 1 ± 0.01 mm with a total mass of 400 ± 5 g. The cement was considered to have set when the needle was lowered gently onto the cement surface and did not leave a complete square indentation on it. The cement was tested for setting initially at 15 min time intervals, gradually shortening to around 1 min intervals as time progressed and the cement was visibly close to being completely set.

Compressive strength

Two-part cylindrical brass moulds with internal diameter 4 ± 0.1 mm and length 6 ± 0.1 mm were used. The moulds were placed on a glass plate and lubricated with mould release agent (Sika Separol, Switzerland). The cement was then compacted into each mould using a spatula, after which it was further compacted using a dental plugger to ensure a dense uniform sample with minimal porosity. Once filled, the excess was scraped off with the edge of a glass microscope slide to leave a flat uniform surface. The filled moulds were incubated for 24 h at 37 ± 1 ◦ C, after which they were removed from the incubator and the screws holding the two halves together were loosened to allow the set cement cylinder to be gently pushed out without damaging it. The cylinders were then left for 28 days in either a sealed, dry container, or immersed in HBSS contained in a sealed container, in each case in an incubator kept at 37 ± 1 ◦ C. 10 replicates of each test were made. After 28 days had elapsed, the specimens were taken out of the solution and dried using filter paper. The top and bottom were lightly sanded using moistened 600-grit sandpaper and each sample’s diameter was measured with a digital micrometer (Draper PM025D, Draper Tools Ltd., Hants, UK). They were then individually loaded with a compressive strength testing machine (Controls 50-C0050/CAL, Controls spa, Milan, Italy) at a loading rate of 50 ± 10 N/min as per ISO 9917-1:2007 [17], until failure of the cylinder occurred. The compressive loading before failure was recorded and the compressive strength of each cylinder calculated using Eq. (1), which assumes specimens of perfectly circular cross-section:


d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306


Maximum force applied before specimen failure Cross sectional area of specimen

= F×

4 d2


where , compressive strength (MPa); F, maximum force applied before failure (N); d, diameter of specimen cylinder (mm).



Specimens 15 ± 1 mm diameter and 1 ± 0.1 mm thick were prepared and allowed to cure for 24 h at 37 ± 1 ◦ C. The specimens were then removed from the moulds and immersed in 20 ± 0.001 ml of HBSS. After immersion in HBSS for 28 days the specimens were dried in a desiccator. The surface of each disc was ground using progressively finer grits of silicon carbide paper, from 180-grit to 1200-grit, and finally polished using 3-micron polycrystalline diamond paste. The polished discs were sprayed with a thin layer of graphite spray (Graphit 33, CRC Industries, Iffezheim, Germany) to aid visibility under the indenter. Measurement of micro-hardness was performed with a Micro-Vickers hardness-testing machine (MVK-H2, Mitutoyo Asia Pacific Ltd., Kanagawa, Japan) using a Vickers diamond indenter with 100 gf load (equivalent to 0.9807 N). The machine automatically applied the load, held it for 10 s and released it, with the loading and unloading rates being automatically controlled. The length of the two diagonals of the impression left by the indenter were measured, and Vickers hardness was calculated using Eq. (2): F HV = 0.1891 × 2 d


where HV, Vickers hardness number; 0.1891, Vickers constant (from indenter geometry); F, force in Newtons (=0.9807 N for the 100 g weight used); d, arithmetic mean of the two diagonals, in mm.

2.3.5. Fluid uptake, sorption, solubility and estimated porosity by mass measurement Specimens 15 ± 0.1 mm diameter and 1 ± 0.1 mm thick [18] were cast by placing the pastes in the moulds and allowing them to cure for 24 h at 37 ± 1 ◦ C. The specimens were then removed from the moulds and weighed to an accuracy of ±0.0001 g. This initial dry mass was recorded as m1 . The mean diameter of each specimen and the thickness of each specimen were measured to an accuracy of ±0.01 mm used to calculate the volume V of each disc. The specimens were then immersed in 20 ± 0.001 ml of HBSS. After 1 day of immersion, the specimens were removed, dried using filter paper and weighed (while they were still fully saturated with fluid). This mass was recorded as m. The fluid uptake, Fup of each specimen (a measure of the amount of fluid taken up by the specimens, expressed in ␮g/mm3 ) was calculated using Eq. (3) [19] and repeated to measure the fluid uptake after 1, 7, 14, 21 and 28 days of immersion Fup =

m − m1 V


where m is the mass, in micrograms, of the fluid-saturated specimen at 1, 7, 14, 21 or 28 days; m1 is the mass, in micrograms, of the specimen prior to immersion in fluid (i.e. at day zero); V is the volume of the specimen, in cubic millimeters. After 28 days, the mass of the specimens (fully saturated with fluid) was measured and recorded as m2 . The specimens were then dried by placing in a desiccator under vacuum for 24 h, using silica gel as desiccant, to constant mass as defined in the standard [18]. This final dry mass was recorded as m3 . The sorption, Fsp , in micrograms per cubic millimeter, for each sample was calculated using Eq. (4) [18]. Fsp =

m2 − m3 V


where m2 is the mass of the of the fluid-saturated specimen, in micrograms, after 28 days of immersion; m3 is the mass of the dried specimen following 28 days of immersion, in micrograms. Solubility, Fsl , for each sample was calculated using Eq. (5) [18]. Fsl =

m1 − m3 V


The percentage porosity of each specimen at 28 days was estimated using Eq. (6) [19]: Porosity (%) =

m − m  1 2 m1

× 100


This equation only provides an estimate of porosity because it assumes that all pores are connected and thus able to absorb water – whereas in reality some pores are unconnected and thus would not be accounted for by this equation due to their inability to absorb water.


Statistical analysis

The data were evaluated using SPSS (Statistical Package for the Social Sciences) software (SPSS Inc., Chicago, IL, USA). Parametric tests were performed as K–S tests on the results indicated that the data were normally distributed. Analysis of variance (ANOVA) with P = 0.05 and Tukey post hoc test were used to perform multiple comparison tests.



3.1. Characterization of un-hydrated cement, anti-washout gel and set cements The XRD and FT-IR results of the un-hydrated powder and AW solution are shown in Fig. 1a and b respectively. The XRD plot for the AW solution showed that there were no crystalline constituents. In all other traces, phases identified included tricalcium silicate (ICDD: 31-0301) with the strongest peaks at 2 (Cu-k␣): 29.6◦ , 32.2◦ , 32.7◦ ; dicalcium silicate (ICDD: 31-0299) with the major peaks at 2 (Cu-k␣): 32◦ , 37.5◦ , 41.3◦ ; and bismuth oxide (ICDD: 41-1449) with peaks at 2 (Cu-k␣): 27.4◦ , 28.0◦ .


d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306

Fig. 1 – Characterization of un-hydrated MTA Plus and anti-washout gel: (a) X-ray diffractogram showing calcium hydroxide (CH), bismuth oxide (BO), tricalcium silicate (C3S) and dicalcium silicate (C2S) peaks; and (b) absorption Fourier-transform infrared spectroscopy, showing peaks of tricalcium silicate (C3S).

The results for the XRD and FT-IR analysis for the hydrated cements after 1 day and 28 days stored dry or in HBSS for both materials are shown in Fig. 2. In the hydrated cements the major peaks for tricalcium silicate, dicalcium silicate and bismuth oxide were identified. The tricalcium silicate, dicalcium silicate and bismuth oxide peak intensities were noticeably reduced for all 28-day samples compared to the respective 1-day samples of the same material. In addition, calcium hydroxide (Portlandite, ICDD: 44-1481) with a major peak at 2 (Cu-k␣): 18.0◦ was identified in the XRD plots of the crushed hydrated cements. MTA-W had a larger calcium hydroxide peak than MTA-AW, particularly for the specimens immersed in HBSS (both after one and 28 days). For the dry samples, the difference in intensity of calcium hydroxide peaks between MTA-W and MTA-AW was only evident at 28 days. These observations were supported by the FT-IR results. Weak peaks of calcium carbonate (at 1400 cm−1 ) were identified in all samples. The un-hydrated MTA displayed a tricalcium silicate peak at ∼875 cm−1 . Tricalcium silicate (peaking at ∼875 cm−1 ) was also identified in all the set cements with the peak reduced in intensity when compared to the un-hydrated material, and calcium-silicate-hydrate gel (peaking at ∼975 cm−1 ) was only clearly identified on the 28-day specimens. In addition, weak peaks of ettringite (at

900 cm−1 ) were identified in all the dry samples and in the 1-day old HBSS samples. The results for the quantitative analysis of the antiwashout gel are shown in Table 1. The gel was water-based containing minor quantities of silicon, potassium, calcium and chlorine. The scanning electron micrographs of the set cements and the calcium–silicon and bismuth–calcium plots for cements stored dry and in HBSS are shown in Figs. 3 and 4 respectively. The MTA mixed with water and stored dry (Fig. 3a) exhibited little or no reaction rims around the cement particles. On the other hand the MTA mixed with anti-washout gel exhibited large reaction rims around the un-hydrated cement particle.

Table 1 – Quantitative analysis of anti-washout gel using X-ray fluorescence. Oxide composition H2 O SiO2 K2 O Cl CaO

wt% composition 97.8 2.08 0.08 0.02 0.01

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306


Fig. 2 – Characterization of set and crushed cement discs by (a) X-ray diffraction, showing calcium hydroxide (CH), bismuth oxide (BO), tricalcium silicate (C3S) and dicalcium silicate (C2S) peaks; and (b) absorption Fourier-transform infrared spectroscopy, showing tricalcium silicate (C3S), calcium-silicate-hydrate gel (CSH), ettringite (E) and calcium carbonate (CC).

The calcium–silicon ratio plots indicate a low calcium–silicon ratio in the MTA-AW signifying the presence of calcium silicate hydrate around the un-hydrated cement particle. The sizes of unhydrated cement grains varied somewhat, with most particles being on average smaller than 2 ␮m × 2 ␮m. Both MTA-W and MTA-AW stored in HBSS exhibited reaction by-product around the cement particles. This reaction by-product has a low calcium–silicon ratio. For all materials the bismuth to calcium ratio was reduced at the margin and lower still in the cement matrix when compared to the levels displayed in the center of the bismuth particle. However there was the presence of bismuth in the cement matrix. The scanning electron micrographs and the EDX analysis of the surfaces of cements stored in HBSS are shown in Fig. 5. The materials surfaces contained spherical particles that exhibited strong peaks for calcium and phosphorus when analysed using EDX analysis.


Chemical properties


pH and calcium ion release

The results are shown in Fig. 6a. Both materials tested exhibited an alkaline pH, with the pH level generally increasing over time. MTA-AW had slightly lower pH than MTA-W. This was only statistically significant for the 7-day and 21-day readings (P < 0.001, P = 0.04 respectively). The results for calcium ion release are shown in Fig. 6b. All materials leached calcium in solution, with the calcium ion concentration increasing over time. Since the solutions were not changed for the duration of the immersion period, the values reported here are cumulative concentrations and thus a decrease over time would be indicative of re-absorption of calcium ions. MTA-AW had slightly less calcium ion release than MTA-W. This was statistically significant at 14-days (P < 0.001), 21-days (P < 0.001) and 28-days (P = 0.002).


d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306

Fig. 3 – Left: Scanning electron micrographs of polished cross-sections of (a) MTA Plus mixed with water and (b) MTA plus mixed with anti-washout solution stored dry. Right: corresponding calcium–silicon and bismuth–calcium plots for different locations indicated in the micrographs.

3.3. 3.3.1.

Physical properties Radiopacity

The results are shown in Fig. 7. Each material exceeded the 3 mm limit recommended by ISO 6876 [20] under all conditions and time periods. MTA-AW appeared to have greater mean radiopacity than MTA-W but this was statistically significant only for the 1-day dry specimens (P = 0.038). For each individual material, environmental conditions and age made no statistically significant difference (P > 0.05 in all cases).


Setting time

The results are shown in Fig. 8. The anti-washout gel reduced the setting time of MTA by 65 min. MTA did not set in HBSS, even after 8 h, however MTA-AW set in HBSS even though it was immersed immediately after mixing with the gel. HBSS had the effect of retarding setting time. These changes were all statistically significant (P < 0.05 in all cases).


3.3.5. Fluid uptake, sorption, solubility and estimated porosity by mass measurement The fluid uptake results are given in Fig. 10a. Statistically significant differences were only observed for the 1-day readings (P < 0.001) where MTA-AW took up less fluid than MTA-W. At all other time periods no statistically significant differences between the materials were observed. The results for sorption and solubility, and estimated porosity are shown in Fig. 10b and c respectively. No statistically significant differences between MTA-W and MTA-AW were observed for all tests.

4. 3.3.3.


The micro-hardness results were (mean ± 1 SD): MTAW = 66.3 ± 6.25 HV; MTA-AW = 69.8 ± 5.86 HV. No statistically significant differences existed between the results of the two materials.


Compressive strength

The results are shown in Fig. 9a. MTA-W dry specimens were very fragile and failed as soon as the machine platen came in contact with the specimen surface, and thus no readings could be obtained. The anti-washout agent imparted a statistically significant increase in strength both when stored dry and immersed in HBSS (P < 0.001).

MTA PlusTM is a new version of mineral trioxide aggregate which is claimed to have a finer particle size than other commercially available versions (50% of the particles finer than 1 ␮m; C.M. Primus, unpublished results). MTA Plus is sold with water as dispensing liquid or salt-free polymer gel in place of water as the mixing vehicle to improve its washout resistance

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306


Fig. 4 – Left: Scanning electron micrographs of polished cross-sections of (a) MTA Plus mixed with water and (b) MTA plus mixed with anti-washout solution immersed in Hank’s balanced salt solution. Right: corresponding calcium–silicon and bismuth–calcium plots for different locations indicated in the micrographs.

[15]. The material is also claimed to have a setting time of 1.2 h [15]. This novel material is characterized using a variety of techniques. Scanning electron microscopy of ground cement surfaces and micro-structural analysis is a standard method used for Portland cement-based materials. X-ray diffraction analysis is a very useful tool for assessment of crystalline materials. MTA contains largely crystalline phases, with the calcium silicate hydrate being the only amorphous phase. The latter phase was identified using FT-IR. Un-hydrated MTAPlus powder gave the peaks normally associated with mineral trioxide aggregate, namely tricalcium silicate and dicalcium silicate, and bismuth oxide, indicating that from a qualitative point of view, this material is made using the same basic constituents as other MTA formulations [13]. The anti-washout gel was shown to be water-based from the XRF analysis and it did not display distinct peaks when characterized using X-ray diffraction. This shows an absence of crystalline constituents. It is consistent with what would be expected of a solution made primarily from water-soluble polymer dissolved in water. FT-IR analysis of the anti-washout gel indicated the presence of a peak in the region of 1200–1000 cm−1 . The XRF analysis also displayed traces of inorganic salts including silicon, potassium, chlorine and calcium. Since it is difficult to detect elements of low atomic number (such as the carbon content in the polymer) with XRF, the percentages reported by the XRF analysis will be artificially inflated due to the mass

of the elements in the sample that went undetected being assigned to the elements actually detected. A preliminary investigation of the anti-washout gel revealed that when the gel was dried in an oven at 100 ◦ C, 7 ± 0.0001 g of liquid gel yielded 0.56 ± 0.0001 g of solid polymer-like residue – a mass percentage of 8% of the original. Thus the mass of water in the gel is closer to 92% than the 97.8% reported by XRF. The silicon content in the gel is likely caused from the prolene film. The other components are likely impurities from the production process. It should be noted here that the machine gives results as oxides. This does not mean that oxides were detected (since XRF is sensitive to atoms and not molecules) – rather, any oxygen detected is first assigned to the other detected elements, and the remainder is assumed to be bound to water molecules and thus assigned to hydrogen atoms. In this way the XRF machine is able to provide an estimate for the percentage of water in the sample even though it cannot detect the presence of hydrogen due to its low atomic mass. Recent investigation of the anti-washout characteristics of MTA Plus mixed with anti-washout gel indicated that the gel drastically increased the washout resistance of the MTA [21]. In underwater concrete, anti-washout additives function by increasing the viscosity of the water mixed with the cement powder. This increased viscosity increases the resistance to segregation of the cement paste under the washing action of external water solutions. A similar mechanism is likely responsible for the results of the anti-washout admixture


d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306

Fig. 5 – (a, c) Backscatter electron micrographs of specimen surfaces and (b, d) X-ray energy dispersive analysis of area of (a, b) MTA Plus mixed with water and (c, d) MTA Plus mixed with anti-washout gel.

on the MTA. When in contact with water, the anti-washout admixtures produce a branched polymer network which controls the movement of water and reduces the tendency for dilution with external water during and after placing of the cement in the surgical site. Increased paste viscosity may also be the result of increased adhesion between the grains. This is consistent with the mechanisms of action for anti-washout additives proposed by other researchers [4–8]. The hydrated cements exhibited the formation of calcium silicate hydrate which was observed around the cement particles in the scanning electron micrographs, and from the FT-IR analysis. Since calcium silicate hydrate is amorphous its presence is not detected on XRD but the flattening of the tricalcium and dicalcium silicate peaks was evident in the 28-day old diffractograms. The calcium to silicon ratio plots exhibited a

lower level of calcium at the periphery of the cement particle. The MTA mixed with anti-washout gel exhibited larger reaction rims than MTA mixed with water when both materials were left in dry environmental conditions. When stored in a physiological solution both materials reacted in a similar way; however the MTA mixed with water had an intense calcium hydroxide peak whereas MTA-AW had consistently less intense CH peaks than MTA-W for corresponding curing conditions and time periods. The difference was much more pronounced in HBSS. This indicated that the anti-washout gel affected the reaction of the cement with environmental fluid, rather than with the water introduced during mixing. This corroborates with the fluid uptake results (Fig. 10a) where the rate of fluid uptake for the MTA mixed with anti-washout gel was much slower initially. This suggests a denser, more compact


d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306


Time (minutes)










MTA 180


MTA -A W >720



Fig. 8 – Setting times of MTA mixed with water or anti-washout gel tested dry or after immersion in Hank’s balanced salt solution (±1 SD).

and less porous structure in the case of the anti-washout MTA as this material appeared to be less sensitive to environmental conditions. This hypothesis is consistent with the results of the physical tests that show that MTA-AW had a lower rate of fluid uptake and estimated porosity. Additionally, in dry conditions MTA-AW had reaction by product whilst MTA-W had almost none (again suggesting that MTA-AW is less dependent on curing conditions). EDX plots show similar elemental constituents and percentages for both materials. The surface micrographs of both materials, immersed in HBSS showed spherical particles characteristic of apatite. However no apatite peaks were evident in the XRD and FT-IR results, presumably because the apatite precipitates mostly on the surface (where the sample is in contact with the physiological fluid) and this apatite-rich layer was mixed with the bulk of the material when the specimens were

Fig. 6 – Chemical analysis of MTA mixed with water or anti-washout gel; (a) pH values and (b) calcium ion release in HBSS tested over a 28 day period (±1 SD).


R adiopacity (mmAl)









1-day dry 7.34



1-day HBSS 7.05



28-day dry 7.11



28-day H BSS 7.75


Fig. 7 – Radiopacity values of MTA plus mixed with water or anti-washout gel tested after 1 day and 28 days stored either in dry environment or immersed in Hank’s balanced salt solution at 37 ◦ C (±1 SD).


d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306

90 80

Strength (MPa)

70 60 50 40 30 20 10 0








Fig. 9 – 28-day compressive strength of MTA Plus mixed with water or anti-washout gel stored dry or immersed in Hank’s balanced salt solution (±1 SD).

crushed prior to characterization, thus making the peaks too weak to observe. This is consistent with findings from other literature [22] where apatite peaks were only found on FT-IR of sample surfaces and not of the inner sides of the same samples. In addition, apatite peaks are difficult to observe on XRD because its peaks overlap with those of other phases. FT-IR is a useful test in addition to XRD as it allows the tracking of amorphous phases, particularly calcium-silicatehydrate gel. As hydration progresses, tricalcium silicate and dicalcium silicate are used up, and more calcium hydroxide and calcium-silicate hydrate is produced. This trend is evident when comparing 1-day and 28-day samples of the same material. In the case of XRD, the relative intensity of the CH peak increases compared to the intensities of the tricalcium silicate and dicalcium silicate peaks, whilst in the case of FT-IR, the broad tricalcium silicate peak centered around 875 cm−1 shifts to the left to give way to the broad peak of C–S–H centered around 975 cm−1 . This was observed for both MTA-W and MTA-AW, though the shift was stronger in the HBSScured materials than the dry ones. Weak peaks of ettringite

Fig. 10 – (a) Fluid uptake after different periods of immersion in HBSS, (b) estimated porosity after 28 days in HBSS and (c) sorption and solubility after 28 days in HBSS (±1 SD).

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306

(another hydration product) were also observed, however this did not appear to develop appreciably over time, and in fact became invisible for the 28-day specimens immersed in HBSS. The presence of apatite could not be confirmed because once again its strongest peaks, which should be centered around the 1000–1100 cm−1 region, overlap with those of other phases, in this case the large broad peaks of either calcium silicate hydrate or tricalcium silicate. It is a well-established fact that MTA releases calcium ions and promotes an alkaline pH in physiological solution [23]. The basis for the biological properties of MTA have been attributed to the production of hydroxyapatite when the calcium ions released by the MTA came into contact with tissue fluid [24]. Both materials exhibited an alkaline pH and released calcium ions into solution, indicating that both are bioactive. The MTA-W paste did not set when immediately immersed in HBSS. The interference of HBSS with setting of MTA and related materials has already been reported [16,25]. This could be due to the fluid penetrating into the paste and causing washout. In the case of MTA-AW, the gel prevented washout and thus explains why the cement was able to set even though it was immersed immediately. Alternatively the HBSS could also have affected the setting of the material by action of its various additives including retardants like sugars and phosphates. The polymer present in the anti-washout gel could also have reduced the effect of the HBSS. MTA-AW set faster than MTA-W. This may be due to the different liquid to cement ratios used. In the case of MTA-W, 0.35 ml/g water was added, whilst for MTA-AW gel was added in the ratio 0.35 g/g. Although the density of pure water is close to 1 g/ml, approximately 8% of the gel was composed of dissolved solids, meaning that MTA-AW effectively had a water-tocement ratio of 0.32, lower than that of MTA-W. Lowering the water to cement ratio is known to decrease setting time [26], which would partially explain this result. In addition, calcium chloride is known to decrease the setting time of MTA [27], although the effect of the low traces found may not have had a significant effect. Finally, the method used for testing setting time was an indentation method. Thus whether the indenter penetrates the surface or not is dependent on the stress value (or pressure) at which the cement surface yields. For the Vicat needle used, the surface area was 1 mm2 and the force applied was 3.92 N. Therefore lack of visible deformation after lowering the needle would indicate that the paste has reached a yield strength of at least 3.92 MPa. The anti-washout gel changed the rheology and handling properties of the material. In particular, it was noticed that whilst MTA mixed with water had a sandy consistency, the MTA mixed with antiwashout gel had a far more viscous and rubbery consistency, almost dough-like. This increased viscosity may explain, from a purely physical standpoint, why MTA-AW developed the threshold strength of 3.92 MPa sooner than MTA-W did – it had a higher viscosity (and thus strength level) to begin with, and as water was absorbed by the MTA granules for the hydration reaction, the branched polymer network was drawn out of solution and remained as a solid, thus bearing a portion of the load from the needle and contributing to early development of strength in the cement paste. The setting time of MTA-AW was found to be 115 min in dry conditions, which is longer than the 1.2 h (72 min) reported by the manufacturer


[15]. This may be due to using a higher gel-to-cement ratio in this study, or different setting time measurement apparatus. The anti-washout gel added to the MTA did not affect the radiopacity of the resultant material. An increase in compressive strength of MTA-AW compared to MTA-W was observed. The lower water–cement ratio used in the MTA-AW mixes could be causing the higher strength. However other factors could also play a role in the higher strength exhibited by the MTA mixed with anti-washout gel. The greater viscosity of MTA-AW results in specimen cylinders with smaller pores when cast, as the paste is more easily compacted with the dental plugger due to its dough-like consistency (whereas MTA-W has a runny consistency similar to wet sand). This is confirmed by the slightly lower fluid uptake and estimated porosity calculated for MTA-AW compared to the water-only MTA. Previous work [16] has shown that curing conditions have an effect on the physical properties of cement-based materials. In this case, the same trends were observed, namely that setting time lengthens and compressive strength increases when cured in HBSS. The micro-hardness followed the same trend as the compressive strength, but the disparity between MTA-W and MTA-AW was not so great. This may be because microhardness does not depend on macroscopic defects (such as porosity caused by entrapment of air bubbles during casting), but rather measures the local yield strength of a small volume of material and is thus more dependent on the material’s intrinsic properties than on the overall condition of the entire sample.



The anti-washout gel was found to significantly increase the washout resistance of MTA and enhance some of the physical properties. It reduced the setting time, increased the compressive strength and reduced fluid uptake and estimated porosity. Radiopacity, fluid sorption, fluid solubility and micro-hardness were not affected. With regards to chemical properties, both exhibited a similar microstructure, elemental make-up and hydration reaction. Both materials produced an alkaline pH and released calcium ions in solution, indicating that both are expected to be bioactive.

Acknowledgements The Faculty of Dental Surgery and the Research Grant Committee University of Malta for funding; Dr. Carolyn Primus for the materials. Ing. J. Camilleri of the Department of Metallurgy and Materials Engineering, Faculty of Engineering, and Mr. J. Spiteri and Mr. J. Grech of the Department of Chemistry, Faculty of Science, University of Malta for their technical expertise; ERDF (Malta) for the financing of the testing equipment through the project: “Developing an Interdisciplinary Material Testing and Rapid Prototyping R&D Facility (Ref. No. 012), “Furnishing and Equipping of the Chemistry and Biology Extensions (Ref. No. 011), and “Strengthening of Analytical Chemistry, Biomedical Engineering and Electromagnetics RTDI Facilities (Ref. No. 018)”. The research work disclosed in this publication is funded by the Strategic Educational Pathways Scholarship (Malta). The scholarship is part-financed by the European


d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 294–306

Union – European Social Fund (ESF) under Operational Programme II – Cohesion Policy 2007–2013, “Empowering People for More Jobs and a Better Quality of Life”.


[1] Torabinejad M, Chivian N. Clinical applications of mineral trioxide aggregate. Journal of Endocrinology 1999;25: 197–205. [2] Parikoth M, Torabinejad M. Mineral trioxide aggregate: a comprehensive literature review – part I: chemical physical, and antibacterial properties. Journal of Endocrinology 2010;36:16–27. [3] Bortoluzzi EA, Broon NJ, Bramante CM, Garcia RB, Moraes IG, Bernardineli N. Sealing ability of MTA and radiopaque Portland cement with or without calcium chloride for root-end filling. Journal of Endocrinology 2006;32:897–900. [4] Wang X, Chen L, Xiang H, Ye J. Influence of anti-washout agents on the rheological properties and injectibility of a calcium phosphate cement. Journal of Biomedical Materials Research Part B: Applied Biomaterials 2007;81:410–8. [5] Porter ML, Berto A, Primus CM, Watanabe I, Physical. Chemical properties of new-generation endodontic material. Journal of Endocrinology 2010;36:524–8. [6] Kai D, Li D, Zhu X, Zhang L, Fan H, Zhang X. Addition of sodium hyaluronate and the effect on performance of the injectable calcium phosphate cement. Journal of Materials Science 2009;20:1595–602. [7] Miyamoto Y, Ishikawa K, Takechi M, Toh T, Yuasa T, Nagayama M, et al. Histological and compositional evaluations of three types of calcium phosphate cements when implanted in subcutaneous tissue immediately after mixing. Journal of Biomedical Materials Research 1999;48:36–42. [8] Komath M, Varma HK. Fully injectable calcium phosphate cement – a promise to dentistry. Indian Journal of Dental Research 2004;15:89–95. [9] Torabinejad M, White DJ. Tooth filling material and use. US patent number 5,769,638; 1995. [10] Lin Q, Xianghui L, Li Y, Yu Y, Ni Y, Lu C, et al. Anti-washout carboxymethyl chitosan modified tricalcium silicate bone cement: preparation, mechanical properties and in-vitro bioactivity. Journal of Materials Science 2010;21: 3065–76. [11] Chen CC, Lai MH, Wang WC, Ding SJ. Properties of anti-washout-type calcium silicate bone cements containing gelatin. Journal of Materials Science 2010;21:1057–68. [12] Khayat KH. Effects of antiwashout admixtures on fresh concrete properties. ACI Materials Journal 1995;92: 164–71.

[13] Camilleri J, Montesin FE, Brady K, Sweeney R, Curtis RV, Pitt Ford TR. The constitution of mineral trioxide aggregate. Dental Materials 2005;21:297–303. [14] Avalon Biomed, Inc. MTA Plus|Avalon Biomed. Available from: http://www.avalonbiomed.com/mta plus/ [accessed April 2012]. [15] Qi Y, Li N, Niu L, Primus CM, Ling J, Pashley DH, et al. Remineralization of artificial dentinal caries lesions by biometrically modified mineral trioxide aggregate. Acta Biomaterialia 2012;8:836–42. [16] Formosa LM, Mallia B, Camilleri J. The effect of curing conditions on the physical properties of tricalcium silicate cement for use as a dental biomaterial. International Endodontic Journal 2012;45:326–36. [17] International Standards Organization. Dentistry – water-based cements. BS EN ISO 9917-1; 2007. [18] International Standards Organization. Dentistry: polymer-based restorative materials. ISO 4049; 2009. [19] Cutajar A, Mallia B, Abela S, Camilleri J. Replacement of radiopacifier in mineral trioxide aggregate; characterization and determination of physical properties. Dental Materials 2011;27:879–91. [20] International Standards Organization. Dental root canal sealing materials. BS EN ISO 6876; 2002. [21] Formosa LM, Mallia B, Camilleri J. A quantitative method for determining the anti-washout characteristics of cement-based dental materials including mineral trioxide aggregate. International Endodontic Journal, in press, http://dx.doi.org/10.1111/j.1365-2591.2012.02108.x [22] Gandolfi MG, Taddei P, Siboni F, Modena E, Ciapetti G, Prati C. Development of the foremost light-curable calcium-silicate MTA cement as root-end in oral surgery. Chemical–physical properties, bioactivity and biological behavior. Dental Materials 2011;27:e134–57. [23] Duarte MA, Demarchi AC, Yamashita JC, Kuga MC, Fraga SdeC. pH and calcium ion release of 2 root-end filling materials. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 2003;95:345–7. [24] Sarkar NK, Caicedo R, Ritwik P, Moiseyeva R, Kawashima I. Physicochemical basis of the biologic properties of mineral trioxide aggregate. Journal of Endocrinology 2005;31:97–100. [25] Kim Y, Kim S, Shin YS, Jung IY, Lee SJ. Failure of setting of mineral trioxide aggregate in the presence of fetal bovine serum and its prevention. Journal of Endodontics 2012;38:536–40. [26] Stefanou GD, Larsinos Ch. Influence of mixing water on the setting time of concrete. International Journal of Cement Composites and Lightweight Concrete 1981;3:45–8. [27] Ber BS, Hatton JF, Stewart GP. Chemical modification of ProRoot MTA to improve handling characteristics and decrease setting time. Journal of Endocrinology 2007;33:1231–4.