Chemical analysis of powder and set forms of Portland cement,

gray ProRoot MTA, white ProRoot MTA, and gray

MTA-Angelus
Jin-Seon Song, BDS, MS, FRACDS,a Francis K. Mante, DMD, PhD,b William J. Romanow,c
and Syngcuk Kim, DDS, PhD,d Philadelphia, PA
UNIVERSITY OF PENNSYLVANIA

Objective. To evaluate the chemical composition and crystalline structures of Portland cement, gray ProRoot MTA
(gray MTA), white ProRoot MTA (white MTA), and gray MTA-Angelus.
Study design. X-ray diffraction analysis was used to identify and characterize crystalline phases, and energy dispersive
x-ray spectrometer was used to determine the chemical composition of the test materials. Both powder form and set
form were examined.
Results. The crystalline structure and chemical composition of gray and white MTA were similar except for the
presence of iron in gray MTA. Both were composed mainly of bismuth oxide and calcium silicate oxide. Portland
cement was composed mainly of calcium silicate oxide and did not contain bismuth oxide. Gray MTA-Angelus had a
lower content of bismuth oxide than ProRoot MTA. There were no noticeable differences in the chemical composition
and crystalline structures between the powder and set forms of any of the material tested.
Conclusion. Portland cement differed from the MTA by the absence of bismuth ions and presence of potassium ions.
Gray MTA contained a significant amount of iron when compared with white MTA. In addition, gray MTA-Angelus
had a lower content of bismuth oxide than ProRoot MTA. (Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2006;
102:809-15)

Since its introduction and its approval in 1998 by the tial uses of MTA in other dental and medical proce-
US Food and Drug Administration, mineral trioxide dures are continually being explored. Some authors
aggregate (MTA) (ProRoot MTA, Dentsply Tulsa, have suggested using MTA as an obturating material
Tulsa, OK) has been used widely in clinical endodon- for the entire root canal system,15 and a possible appli-
tics. It has been shown to be superior to other materials cation of MTA in orthopedic setting has been explored
in terms of sealing ability,1,2 biocompatibility,3-5 and with positive results.16
enhancing regeneration of the periradicular tissues.6 It MTA is based on Portland cement; both gray and
also possesses ideal properties including antimicrobial white Portland cements are manufactured from sim-
effect,7,8 radiopacity, dimensional stability, and toler- ilar raw materials except that a fluxing agent is used
ance to moisture. Clinically, MTA is being used in for production of the white version to remove the
dental procedures such as vital pulp therapy,9,10 apexi- ferrite phase during the clinkering process. Accord-
fication,6,11 repair of root perforation,12,13 root-end fill- ing to the manufacturer, the principal components of
ing,14 internal bleaching, and resorption repair. Poten- the gray-colored formula are tricalcium silicate, bis-
muth oxide, dicalcium silicate, tricalcium aluminate,
tetracalcium aluminoferrite, and calcium sulfate de-
hydrate, and the white-colored formula lacks the
a
tetracalcium aluminoferrite. Both formulae are 75%
Department of Endodontics, School of Dental Medicine, University
of Pennsylvania. Portland cement, 20% bismuth oxide, and 5% gyp-
b
Director of Biomaterials Restorative Denistry, Department of Re- sum by weight.17,24 Another product, gray MTA-
storative Dentistry, University of Pennsylvania. Angelus (Angelus Soluções Odontológicas, Lond-
c
Undergraduate Instruction & Laboratory Coordinator, Department rina, Brazil), consists of 80% Portland cement and
of Materials Science & Engineering, University of Pennsylvania.
d
Louis I. Grossman Professor and Chairman, Department of Endo- 20% bismuth oxide.
dontics, School of Dental Medicine, University of Pennsylvania. A number of previous studies have compared MTA
Received for publication Nov. 30, 2005; returned for revision Nov. with Portland cement and unequivocally indicated that
30, 2005; accepted for publication Nov. 30, 2005. they are similar in chemical composition, if not identi-
1079-2104/$ - see front matter
© 2006 Mosby, Inc. All rights reserved. cal, except for the inclusion of bismuth oxide in
doi:10.1016/j.tripleo.2005.11.034 MTA.18,19 In vitro20,21 and in vivo22,23 studies have

809
 OOOOE
810 Song et al. December 2006

also shown the Portland cement to exhibit properties Portland cement sample was also mixed in the same
similar to MTA. manner. The mix was stored in an incubator at 37°C
Recently, white MTA has been developed and mar- and 100% humidity for 3 days. The set samples were
keted to replace gray MTA to address esthetic con- then mounted onto the XRD apparatus for analysis.
cerns. A number of studies have been performed on this
new material to determine if it exhibits the same prop- XRD analysis
erties as gray MTA. These studies comparing white and The prepared sample material was mounted onto the
gray MTA have yielded conflicting results in terms of XRD apparatus (Geigerflex Horizontal diffractometer
biocompatibility, sealing ability, and ability to induce with a graphite crystal monochrometer; Rigaku/MSC,
tissue regeneration. A leakage study by Ferris and Woodlands, TX). The x-ray beam angle 2␪ range was
Baumgarnter24 and a biocompatibility study by Holland set between 3 degrees (3000) to 70 degrees (70000) and
et al.25 have shown them to have similar properties. scanned at 2 degrees per minute. The Cu x-ray source
Camilleri et al.26 reported that osteoblasts had the same was set at accelerating voltage of 45 KV and the current
biocompatible reaction to both types of MTA. An in in the electron beam at 30 mA and on continuous scan
vivo study by Menezes et al.10 showed that when white mode. The peaks on the diffraction pattern were
MTA, gray MTA, and white Portland cement were marked using the Rigaku software (version 2.8). Then
applied as a pulpotomy agent, all the samples resulted the peaks were compared and matched with that of the
in healing of the pulp and a hard tissue bridge forma- standard material in the powder diffraction file (JCPDS
tion. International Center for Diffraction Data 1998, Penn-
However, some in vitro studies, such as by Perez sylvania) using a micro powder diffraction search and
et al.,27 have reported that osteoblastic cell line matching analysis program.
grown on gray MTA adhered and differentiated bet-
ter than the cells grown on white MTA. A leakage Preparation of the sample for EDS
study by Matt et al.28 showed that gray MTA dem- Each sample powder was placed on an aluminum
onstrated significantly less dye leakage when com- stop with double-sided carbon tape. Carbon coating
pared to white MTA. was applied before running the scanning electron mi-
In light of the conflicting findings in comparing the croscopy (SEM).
sealing ability, biocompatibility, and tissue-regenerat-
ing ability of gray MTA and white MTA between these EDS
studies, there is a need to compare the chemical com- Analytical scanning electron microscopy was per-
position of the MTA products available. The purpose of formed on JEOL 6400 SEM (Tokyo, Japan). This mi-
this study was to analyze and compare the chemical croscope was equipped with an Oxford energy disper-
composition of Portland cement, gray MTA, white sive x-ray spectrometer (EDS) and wavelength
MTA, and gray MTA-Angelus. Both the powder and dispersive x-ray spectrometer (WDS). The EDS system
the set forms were examined. X-ray diffraction (XRD) was used to determine the chemical composition of the
was used to identify and characterize crystal phases, examined materials.
and an energy dispersive x-ray spectrometer system
(EDS) was used to determine the chemical composition RESULTS
of the examined materials. XRD
The chemical composition and its crystal structure
MATERIALS AND METHODS of gray and white MTA were similar (Fig. 1, Table
Preparation of the sample (powder) I). The only exception was that gray MTA contained
MTA powder was mixed with acetone and applied significant amount of iron when compared with white
onto the frosted surface of a glass slide and left in air MTA. Both materials were composed mainly of bis-
for 30 seconds until the acetone had evaporated com- muth oxide crystalline structure and calcium silicate
pletely, leaving the MTA powder attached to the slide. oxide. Although present, other crystal phases con-
The prepared slide was mounted onto the XRD appa- tributed to a very small proportion of the material.
ratus. Three samples of each material, Portland cement, Portland cement did not contain bismuth oxide but it
gray MTA, white MTA and gray MTA-Angelus, were was composed of many different crystal phases. Gray
investigated by XRD. MTA-Angelus differed from the gray ProRoot MTA
in that it contained less bismuth oxide and more of
Preparation of the sample (set) other crystalline phases, many of which could not be
Each MTA sample was mixed with distilled water identified (Fig. 1; Table I).
according to the manufacturer’s direction for use. For each of the cement types, there were no notice-

www.endoexpert.net
OOOOE
Volume 102, Number 6 Song et al. 811

Fig. 1. X-ray diffraction patterns of gray Portland cement, gray MTA, white MTA, and gray MTA-Angelus showing peaks
representing the crystal phases present in each material.

able differences in the composition and crystalline bismuth ions and the presence of potassium ions
structure between powder and set form. (Figs. 2-4; Table II).

EDS DISCUSSION
The chemical composition of Portland cement, Several studies have compared the chemical com-
gray MTA, and white MTA were very similar. The position, surface characteristics, sealing ability, bio-
only difference observed between the 2 types of compatibility, and ability to regenerate original tis-
MTA was the lack of iron ions in white MTA. The sues of Portland cement, gray MTA, and white MTA.
Portland cement differed from MTA by the lack of The surface characteristics of Portland cement and
 OOOOE
812 Song et al. December 2006
Table I. X-ray diffraction analysis result showing the chemical composition and semiquantitative crystalline composition of Portland cement, gray MTA, white
gray MTA have been shown to be irregular, consist-
Gray MTA-Angelus ing of either cuboidal discrete crystals or areas of
Powder (%) granular material with a coral-like characteristic
38.8
30.1

18.6
ND
ND

ND

ND
ND
ND
ND
ND
3.4
3.9

1.0
4.2
when viewed under high magnification.16,18,20 This
was not the case for the white MTA, which showed
different surface roughness when viewed under
SEM.27
White MTA

Holland et al.22 clinically observed 2 differences

Set (%)
57.8
30.2
ND
ND

ND

ND

ND
ND
ND
ND
ND
ND
3.9
2.7

5.4
with these materials: the color and the setting time,
which was greater in white MTA.
Ferris and Baumgartner24 compared white and
Powder (%)
White MTA

gray MTA for their sealing ability of furcal perfora-

56.7
34.1

ND

ND
ND
ND
ND
ND
ND
0.9
0.3
3.7

1.6
0.9

1.7

tion using a bacterial leakage model. Their results

demonstrated no difference between the 2 in allow-
ing passage of F. nucleatum. However, a leakage
study by Matt et al.28 showed that gray MTA dem-
Gray MTA
Set (%)

onstrated significantly less dye leakage than white

67.8
21.8

ND

ND

ND
ND
ND
ND
ND
ND
1.4

3.2

2.3
1.1

2.5

MTA.
Abdullah et al.,20 Holland et al.,22 and Saidon et al.23
showed that the biocompatibility and the potential to
Powder (%)
Gray MTA

promote bone healing of Portland cement are compa-

58.8
30.3

ND

ND

ND

ND
ND
ND
ND
ND
ND
2.3

5.3

1.0

2.9

rable to that of gray MTA in their in vitro studies. There

are not many studies published comparing the new
white MTA and the extensively studied gray MTA.
Portland cement

Perez et al.27 compared their effects on osteoblasts.

Set (%)

Under the conditions of their study, primary osteoblasts

65.7

21.7
ND

ND
ND

ND
ND

ND
ND
ND
ND
ND
ND
8.5

1.8

bound and survived on gray MTA and initially bound to

white MTA but did not survive on the surface. On the
other hand, Holland et al.22,25 implanted white MTA
and gray MTA on the connective tissue of rats and
Portland cement
Powder (%)

reported that the mechanisms of action of white MTA

72.6

are very similar to those of gray MTA despite the

ND

ND
ND

ND

ND
6.1
1.0

6.7
5.6

1.8
1.6
2.1
1.4
1.4

differences in the material.

This study examined the presence and intensity of
the peaks from the XRD patterns. Over 50% of the
crystalline structure present in ProRoot MTA was in the
Ca2MgO. 2AlFeO. 6SiO.2O5

form of bismuth oxide, and calcium silicate oxide crys-

Chemical formula

talline contributed to approximately 30%. Gray MTA-

CaAl2Si6O16 5H2O

Angelus consisted of about 40% of the bismuth oxide

Cu5Si6 O17.7H2O

Ca(SO4)(H2O)2

phases and 30% of calcium silicate oxide. This study is

BaZn2(PO4)2
Ca3(SiO4)O

in agreement with that by other authors who compared

Mg3(PO)4

Ca2MgSi2

Ca3Al2O6
Unknown

Ca4P2O9
Ca2SiO4

CaSnO3

the chemical composition of Portland cement and gray

CaCO3
Bi2O3

MTA and showed that they contain the same chemical

MTA, and gray MTA-Angelus.

elements except that gray MTA also contains bismuth

oxide.18,19,21 This study is also in agreement with that
Calcium aluminum silicate hydrate

of Diamannti et al.,18 who found that iron oxide was

Calcium magnesium aluminum

Calcium magnesium silicate

absent in white MTA; however, their observation of

Calcium aluminum oxide

absence of calcium sulfate (plaster anhydride) only in

Calcium sulfate hydrate
Component

Copper silicate hydrate

Barium zinc phosphate

Calcium silicate oxide
Magnesium phosphate

⬎ND, None detected.

gray MTA was not observed in this study. Asgary et

Calcium phosphate
Calcium carbonate

Calcium tin oxide

al.29 found that white MTA contained significantly less

Calcium silicate
Bismuth oxide

iron silicate

iron oxide as well as aluminum oxide and magnesium

Unknown

oxide than gray MTA. Our results also showed that

there was more of the magnesium compound phase in
gray MTA than in white MTA. Metal oxides such as

www.endoexpert.net
OOOOE
Volume 102, Number 6 Song et al. 813

Fig. 2. Energy dispersive x-ray spectrum of Portland cement.

Fig. 3. Energy dispersive x-ray spectrum of gray MTA.

Fig. 4. Energy dispersive x-ray spectrum of white MTA.

aluminum and iron oxides have been known to cause compounds detected in the Portland cement and gray
abnormal tissue reactions equivalent to a chemical in- MTA-Angelus may or may not play a significant role in
sult.31 The effect of iron oxide in gray MTA at this the short- and long-term clinical application, and this
point is not clear. Also, the trace amount of other warrants further studies.
 OOOOE
814 Song et al. December 2006

Table II. Energy dispersive x-ray spectrometer results ties of tooth-colored ProRoot mineral trioxide aggregate. J Endod
showing the composition of Portland cement, gray 2004;30:429-31.
9. Ford TR, Torabinejad M, Abedi HR, Bakland LK, Kariyawasam
MTA, and white MTA
SP. Using mineral trioxide aggregate as a pulp-capping material.
Portland cement Gray MTA White MTA J Am Dent Assoc 1996;127:1491-4.
Ca Yes Yes Yes 10. Menezes R, Bramante CM, Letra A, Carvalho VG, Garcia RB.
Si Yes Yes Yes Histologic evaluation of pulpotomies in dog using two types of
K Yes No No mineral trioxide aggregate and regular and white Portland ce-
C Yes Yes Yes ments as wound dressing. Oral Surg Oral Med Oral Pathol Oral
O Yes Yes Yes Radiol Endod 2004;98:376-9.
Fe Yes Yes No 11. Hayashi M, Shimizu A, Ebisu S. MTA for obturation of man-
Mg Yes Yes Yes dibular central incisors with open apices: case report. J Endod
Al Yes Yes Yes 2004;30:120-2.
S Yes Yes Yes 12. Pitt Ford TR, Torabinejad M, McKendry D, Hong CU, Kariya-
Bi No Yes Yes wasam SP. Use of mineral trioxide for repair of furcal perfora-
tions. Oral Surg Oral Med Oral Pathol Oral Radiol Endod
1995;79:756-62.
13. Nakata TT, Bae KS, Baumgartner JC. Perforation repair com-
paring mineral trioxide aggregate and amalgam using an anaer-
CONCLUSION obic bacterial leakage model. J Endod 1998;24:4-186.
The difference observed between the 2 types of 14. Torabinejad M, Hong CU, Lee SJ, Monsef M, Pitt Ford TR.
MTA was the lack of iron ions in white MTA. They Investigation of mineral trioxide aggregate for root-end filling in
were similarly composed of bismuth oxide and cal- dogs. J Endod 1995;21:603-8.
15. O’Sullivan SM, Hartwell GR. Obturation of a retained primary
cium silicate oxide crystalline structures, and other mandibular second molar using MTA: a case report. J Endod
crystalline components comprised a very small pro- 2001;27:703-5.
portion of the material. Portland cement differed 16. Koh ET, Torabinejad M, Pitt Ford TR, Brady K, McDonald F.
from MTA by the absence of bismuth ions and pres- Mineral trioxide aggregate stimulates a biological response in
ence of potassium ions. In addition, ProRoot MTA human osteoblasts. J Biomed Mater Res 1997;37:432-9.
17. Literature from the manufacturer. Tulsa (OK): Dentsply Tulsa
appeared to have more homogeneous composition
Dental, Tulsa, OK. Material Safety Data Sheet; 1998.
than Portland cement and gray MTA-Angelus. For 18. Diamanti E, Kerezoudis NP, Gakis DB, Tsatsas V. Chemical
each of the cement types, there were no noticeable composition and surface characteristics of grey and new white
differences in the composition and crystalline struc- ProRoot MTA. J Endod 2003;Abstract R81.
ture between powder and set form.30 19. Funteas UR, Wallace JA, Fochtman EW. A comparative analysis
of mineral trioxide aggregate and Portland cement. Aust Endod
We thank Dr. Wojciech J Grzesik for reviewing this J 2003; 29:43-4.
manuscript and giving us very valuable advice. 20. Abdullah D, Ford TR, Papaioannou S, Nicholson J, McDonald F.
An evaluation of accelerated Portland cement as a restorative
material. Biomaterials 2002;23:4001-10.
REFERENCES
21. Estrela C, Bammann LL, Estrela CR, Silva RS, Pecora JD.
1. Wu MK, Kontakiotis E.G., Wesselink PR. Long-term seal pro-
Antimicrobial and chemical study of MTA, Portland cement,
vided by some root-end filling materials. J Endod
calcium hydroxide paste, Sealapex and Dycal. Braz Dent J
1998;24:557-60.
2. Torabinejad M, Rastegar AF, Kettering JD, Pitt Ford TR. Bac- 2000;11:3-9.
terial leakage of mineral trioxide aggregate as a root-end filling 22. Holland R, Souza V, Nery MJ, Faraco IM Jr, Bernabe PF,
material. J Endod 1995;21:109-12. Otoboni Filho JA, Dezan E Jr. Reaction of rat tissue to
3. Koh ET, McDonald F, Pitt Ford TR, Torabinejad M. Cellular implanted dentin tube filled with mineral trioxide aggregate,
response to mineral trioxide aggregate. J Endod 1998;24:543-7. Portland cement or calcium hydroxide. Braz Dent J
4. Torabinejad M, Pitt Ford TR, McKendry DJ, Abedi HR, Miller 2001;12:3-8.
DA, Kariyawasam SP. Histologic assessment of mineral trioxide 23. Saidon J, He J, Zhu Q, Safavi K, Spangberg LS. Cell and
aggregate as a root-end filling in monkeys. J Endod tissue reactions to mineral trioxide aggregate and Portland
1997;23:225-8. cement. Oral Surg Oral Med Oral Pathol Oral Radiol Endod
5. Mitchell PJ, Pitt Ford TR, Torabinejad M, McDonald F. Osteo- 2003;95:483-9.
blast biocompatibility of mineral trioxide aggregate. Biomateri- 24. Ferris DM, Baumgartner JC. Perforation repair comparing two
als 1997;20:167-73. types of mineral trioxide aggregate. J Endod 2004;30:422-4.
6. Shabahang S, Torabinejad M, Boyne P, Abedi H, McMillan P. A 25. Holland R, Souza V, Nery MJ, Faraco Junior IM, Bernabe PF,
comparative study of root-end induction using osteogenic pro- Otoboni Filho JA, Dezan Junior E. Reaction of rat connective
tein-1, calcium hydroxide, and mineral trioxide aggregate in tissue to implanted dentin filled with a white mineral trioxide
dogs. J Endod 1999;25:1-5. aggregate. Braz Dent J 2002;13:23-6.
7. Torabinejad M, Hong CU, Pitt Ford TR, Kettering JD. Antibac- 26. Camilleri J, Montesin F, Papaioannou S, McDonald F, Pitt Ford
terial effects of some root end filling materials. J Endod TR. Biocompatibility of two commercial forms of mineral triox-
1995;21:403-6. ide aggregate. Int Endod J 2004;37:699-704.
8. Stowe TJ, Sedgley CM, Stowe B, Fenno JC. The effects of 27. Perez AL, Spears R, Gutmann JL, Opperman LA. Osteoblasts
chlorohexidine gluconate (0.12%) on the antimicrobial proper- and MG-63 osteostcoma cells behave differently when in

www.endoexpert.net
OOOOE
Volume 102, Number 6 Song et al. 815

contact with ProRoot MTA and white MTA. Int Endod J 31. Steinemann SG. Titanium—the material of choice? Periodontol
2003;36:564-9. 2000 1998;17:7-21.
28. Matt GD, Thorpe JR, Strother JM, McClanahan SB. Compara-
tive study of white and gray mineral trioxide aggregate (MTA) Reprint requests:
stimulating a one- or two-step apical barrier technique. J Endod
2004;30:876-9. Dr. Syngcuk Kim
29. Asgary S, Parirokh M, Eghbal MJ, Brink F. Chemical differences Department of Endodontics
between white and gray mineral trioxide aggregate. J Endod School of Dental Medicine
2005;31:101-3. University of Pennsylvania
30. Torabinejad M, Hong CU, McDonald F, Pitt Ford TR. Physical 240 S. 40th Street
and chemical properties of a new root-end filling material. J Philadelphia, PA 19104-6030
Endod 1995;21: 349-53. [email protected]