Journal of Oral Rehabilitation, 1990, Volume 17, pages 599-609 < i

An in vitro evaluation of a prototype CFRC

prefabricated post developed for the restoration of
pulpless teeth
p . A . K I N G and D . J . S E T C H E L L Department of Conservative Dentistry, Eastman
Dental Hospital, and Institute of Dental Surgery, University of London, London, U.K.

Summary . ;
Carbon fibre reinforced carbon (CFRC) in the form of a prefabricated post* has
recently been developed and is theoretically acceptable for consideration in an endo-
dontic post-retained crown system. This study compared four different types of post-
core system cemented into 40 extracted anterior human teeth. The test groups
consisted of CFRC posts cemented with a composite resin luting agent, and used with
either a cast gold alloy core (Group B) or a composite resin core (Group C). Two
existing post-core techniques were used as controls for comparison with the CFRC
groups. One control was a prefabricated wrought precious alloy post having a cast
gold alloy core, and cemented with zinc phosphate cement (Group A). The other was
a prefabricated stainless steel post with a composite resin core, and cemented with a
composite resin luting agent (Group D). All specimens were restored with a gold alloy
crown and tested to failure with an obliquely applied compressive load at 130° in an
Instron using a cross-head speed of 5 cm min~\ The results showed that post-retained
crowns using a prefabricated CFRC post exhibited properties comparable with, and in
some cases better than, those of existing prefabricated posts. The mode of failure of
specimens restored with a CFRC post was more favourable to the remaining tooth
tissue than was that of specimens restored with a metallic post.

Introduction
Recent technology has produced a rigid non-metallic composite in the form of carbon
fibre reinforced carbon (CFRC), which can be formed in various configurations in
order to make maximum use of its properties.
The manufacturers claim that the mechanical properties are within the range stated
by Adams, Williams and Hill (1978). When the fibres are aligned at an angle to the
principal axis of the composite, high transverse strength and moduli of elasticity,
similar to those of bone, are achieved. Longitudinal and transverse fiexural moduh of
60 GNm"^, longitudinal and transverse flexural strengths of 500 GNm"^, and an
interlaminar shear strength of 18 MNm"^ are typical. The failure of CFRC in bending
stresses is initiated by exceeding the shear strength between the fibre layers, and not

* Dunlop Medical Products, Leyland, Lancashire, U.K. .;.•••:, . . :' ;:' J ;•.>';

Correspondence: Mr P.A. King, Department of Conservative Dentistry, Eastman Dental Hospital,

Institute of Dental Surgery, 256 Gray's Inn Road, London WCIX 8LD, U.K.

599
600 P.A. King and D.J. Setchell - . , /vv . .-

by cracking under tension. Short-term or local overload of CFRC only results in

cracking of some fibres, and not in catastrophic failure of the whole structure.
The biocompatibility of CFRC has been investigated by Hobkirk (1982). He
studied the tissue reaction to intra-bony implants of CFRC in rabbits over an 18-
month period. The implants were well tolerated but were usually surrounded by a
fibrous capsule. Lovell (1983) has demonstrated the compatibility of composite resins
with CFRC and suggested that they may be suitable for a post and core system.
Pilot studies have shown that it is possible to cast noble and non-noble alloys on to
CFRC without any evident disruption of the fibre pattern. However, Thompson,
Buchanan and Lemons (1979) emphasized that carbon is an electrical conductor and a
noble element electrochemically, and they demonstrated pitting corrosion of 18—8
stainless steel when galvanically coupled with carbon.
Post-retained crowns have a relatively high failure rate (Roberts, 1970) and no
one system is ideal. The failures are due either to the lack of strength of the post,
the tooth, or more commonly to insufficient retention of the post within the root
(Ruemping, Lund & Schnell, 1979). The choice is influenced by design features and
the type of material employed.
A number of in vitro studies (CoUey, Hampson & Lehman, 1968; Charlton, 1970;
Johnson & Sakumura, 1978; Standlee, Caputo & Hanson, 1978) concluded that:
(i) threaded screw-in posts were more retentive than unthreaded cemented posts,
but imparted greater installation stresses and therefore increased the risk of
root fracture;
(ii) of the cemented posts, the serrated, parallel-sided post was the most retentive;
(iii) increased post length gave increased retention and decreased stress to the root;
(iv) an increase in post diameter increased the retention but also increased the
likelihood of root fracture;
(v) the ideal material should have a high tensile, compressive, transverse-impact
and fatigue strength. The modulus of elasticity and the coefficient of linear
expansion should be similar to that of tooth tissue.
Few investigations have determined the clinical significance of post and core design
features and materials used. Sorenson and Martinoff (1984), in a restrospective study,
observed the clinical performance of 1273 endodontically-treated teeth. Of those teeth
restored with intracoronal reinforcement, the parallel-sided cast gold alloy post and
core, Parapost*, proved the most successful, with no failures recorded over a period
of 1-25 years. The tapered cast dowel and core displayed a higher failure rate than
teeth treated without intracoronal reinforcement.
The prefabricated wrought precious alloy post with a cast custom gold alloy core is
considered by many authors to be the best available post-core system (Gerstein &
Brunell, 1964). Its main disadvantages are preparation time and expense. An alterna-
tive, quicker and less expensive method, using a prefabricated stainless steel post
combined with a composite resin core, has been advocated by many clinicians (Steele,
1973; Federick, 1974). The fracture strength of most commonly used dental cements is
less than that of dentine or conventional post materials. Failure of the system usually
occurs cohesively within the cement or at its interface with dentine (CoUey et al.,
1968; 0ilo, 1978). The new breed of composite resins introduced as luting agents.

* Whaledent International, New York, U.S.A.

 Prototype CFRC prefabricated post 601

combined with their adhesion promoters, goes some way toward resolving the problem
(Assif & Ferber, 1982; Gontar et al, 1986). . /.; vav^i.i '^ -• ^.-v / ^.^ -.-Hi /^.:
The load and direction of forces applied to teeth during masticatory function are
complex. Bates, Stafford and Harrison (1975) have reviewed the literature on move-
ment of the mandible, and forces developed during mastication.
The majority of in vitro tests have used a tensile force to assess the relative
retention capabilities of post design, length, diameter and surface characteristics
(CoUey et al, 1968; Hanson & Caputo, 1974, Standlee et al, 1978; Johnson &
Sakumura, 1978). However, laboratory studies involving the evaluation of fully re-
stored post-retained crown restorations in extracted human teeth have usually em-
ployed compressive loads at angles simulating tooth contact (Kantor & Pines, 1977;
Lovdahl & Nicholls, 1977; Hoag & Dwyer 1982).
The purpose of this study was to compare the resistance to fracture and the mode
of failure of extracted human teeth restored with either prefabricated CFRC posts or
existing metallic prefabricated posts in a post-retained crown system.

Materials and methods

Freshly extracted maxillary central incisors and maxillary and mandibular canines
were collected for experimentation. All teeth were stored in 5% formol/saline solution
until required for the experiment. From the total number collected, 40 teeth were
selected for use on the basis of dimensional requirements and lack of fractures, cracks,
caries or abrasions in the usable portions. The maximal facio-lingual and mesio-distal
dimensions of the tooth at the level of the facial cemento-enamel junction were
recorded, using a metal gauge* accurate to within 0-1 mm.
Teeth were mounted vertically in standard brass electrical conduit fittingsf to a
level 2 mm below the facial cemento-enamel junction, using a highly filled acrylic
resin, 'Formatray't. The specimens were numbered and stored in distilled water at
room temperature (21°C).
All teeth were prepared in a standardized manner to receive parallel posts 10 mm
long, of diameter 1-5 mm. The crown of the tooth was removed 2 mm coronal to the
facial cemento-enamel junction. A shoulder 1 mm in height and depth was prepared
at this level around the circumference of the tooth. The antirotational feature was
achieved by enlarging the eccentric shape of the coronal root canal. As much coronal
dentine was preserved as possible whilst still providing a sufficient anti-rotational
dimension. Full water spray was used with all rotary instruments. Increasing sizes of
Parapost twist drills were used to prepare the post hole. The latter was initially
prepared to a length of 10-5 mm in order to accommodate a layer of zinc polycarboxy-
late cement, 'Poly-F plus'§, to seal the apical root canal. Finally, the whole preparation
was smoothed, and all sharp angles other than the shoulder margin were rounded.
The teeth were then randomly assigned and restored using one of four post-core
systems, with 10 teeth in each group.

* Iwansson, Stockholm, Sweden.

t Logic, M.K. Ltd, London, U.K.
i Kerr/Sybron, Basel, Europe.
§ AD International Ltd, Weybridge, U.K.
602 P.A. King and D.J. Setchell

Group A. Wrought precious alloy post and gold alloy core

A Parapost P-52 non-oxidizing wrought precious alloy post size No. 6 (1-5 mm in
diameter) was combined with a custom type III gold alloy core, 'Trucast Hard'*. A
standard 'lost wax' method was used to cast the core on to the post. The post and core
were heat-hardened at 350°C for 10 min followed by quenching in water at room
temperature. The post, core and tooth were cleaned and dried using water, compressed
air and paper points. Zinc phosphate cement (Ash C.A.S.)t mixed on a cool glass slab
in a powder:hquid ratio of l-8g:l ml was used to cement the post and core into the
tooth with firm finger pressure.

Group B. CFRC post and gold alloy core : .

A smooth, parallel-sided CFRC post (1-5 mm in diameter) was used as supplied by
the manufacturer in an airtight sealed package. The post was combined with a custom
type III gold alloy core. The same 'lost wax' technique as described for Group A was
used to cast the core on to the post. No heat hardening of the CFRC post-gold core
system was necessary.
The tooth, post and core were thoroughly cleaned and dried. A composite resin
luting cement and an adhesion promoter (Conclude, Scotchbond)$ were proportioned
and mixed according to the manufacturer's instructions, and used to cement the post
and core into the tooth. Two coats of Scotchbond were applied to the fitting surface of
the tooth and post and core prior to cementation.

Group C. CFRC post and composite resin core

A CFRC post was cemented into the post hole using a composite luting cement
(Conclude)t and adhesion promoter (Scotchbond)t. Excess cement was removed
from the antirotational feature using hand instruments. A light-cured, heavily filled,
composite resin ('P30')t was used to form the core portion. Material was first con-
densed and cured into the antirotational feature. Further increments of composite
resin placed within a cylindrical cellulose matrix§ were used to form the remainder of
the core. The composite resin was cured using a 15V 150W tungsten halogen light
source through a fibre-optic light guide.
Following the removal of the matrix, the core was shaped and smoothed with
rotary instruments.

Group D. Stainless steel post and composite resin core

A Parapost stainless steel post size No. 6 (1-5 mm in diameter) was cemented into the
post hole using Conclude composite luting cement and Scotchbond adhesion promoter.
The same technique as described for Group C was used to form the P30 composite
resin core.
Prior to the provision of a full gold alloy crown, all the post and core specimens
were coated with copal ether varnish and stored in distilled water. Direct wax patterns
were formed on the specimens and cast in Trucast type III gold alloy using the 'lost
wax' technique. Ash C.A.S. zinc phosphate cement (powder:liquid ratio 1-8 g:l ml)
was used to cement the crowns on to the post and cores, and they were maintained in
* Engelhard, Chessington, Surrey, U . K . : : . .:
t A D International Ltd, Weybridge, U . K . • . . '
t 3 M C o . , St Paul, M N , U . S . A . -• ;;:.•.•:• - . - < - ^r ;• , . . 4
§ O d u s , Zurich, Switzerland. - , H - : .* - . ' r- ^; . - ,- ; •.•
 Prototype CFRC prefabricated post 603

position with firm finger pressure. The restored teeth were stored in distilled water at
room temperature (21°C). A minimum period of 7 days elapsed between cementation
and testing. Details of the posts and specimen specifications are shown in Figs 1 and 2.

Test procedures
An Instron Universal testing machine was used to apply controlled loads to the
specimens. A stainless steel mounting block was designed and machined to accept the
threaded specimen holder. A 50°-angled face to the mounting block produced a 130°
angle between the long axis of the tooth specimen and the compressive head. A steel,
ball-ended, compressive head, 1 mm in diameter, was used to apply the load from a
lingual direction. The cross-head speed was 5 cm min"' for all specimens. The
direction, size and speed of the compressive head were selected so as to simulate as
far as possible inter-incisal forces that might be encountered in the mouth. The head
was positioned for all specimens in the mid-line at a point measured 5 mm towards the
incisal edge, from the start of the lingual slope (Fig. 3).
Displacement and load magnitude were registered by a compressive load cell, and
recorded by plotting time against force on the calibrated drive chart of the Instron
machine. Failure threshold was determined as the point at which the specimen could
no longer withstand an increase in load. All specimens were stored in distilled water
for subsequent analysis of the mode of failure.
Multiple comparisons of the specimen mean peak stress values at failure were

(a)B (b) (d).

Fig. 1. Prefabricated posts used in experiment: (a) carbon fibre reinforced carbon; (b) wrought
precious alloy; (c) stainless steel; (d) 1-5 mm twist drill.
604 P.A. King and D.J. Setchell

Fig. 2. Specifications of restored teeth in mounting block.

•U,

Fig. 3. Instron and specimen assembly.

made using Student's t-tesi. The mode of specimen failure was analysed between the
groups using the Chi-squared test on a fourfold contingency table.

Results
The compressive load (kgf) recorded at failure of the specimen was converted into a
stress value (MNm"^) by using the cross-sectional area of the tooth at the facial
 Prototype CFRC prefabricated post 605

Table 1. Arithmetic means and standard deviations for the peak stress at failure of specimen groups

Group Restoration type Sample Mean stress Standard

size i* at failure deviation
(MNm~^) (MNm^^)

A Precious alloy wrought 10 . 16-24 2-60

post and gold alloy core
B CFRC post and gold alloy 10 15-75 .. 2-08
core
C CFRC post and composite 10 14-42 3-00
resin core
D Stainless steel post 10 13-00 2-53
and composite resin core

cemento-enamel junction. The area was calculated as an ellipse using the facio-lingual
and mesio-distal coordinates.
The mean peak stress at failure and the standard deviations are shown in Table 1.
These are presented graphically in Fig. 4.
There was a significant difference (P < 0-05) between mean stress values at failure
for the stainless steel post and composite resin core specimens compared with either

standard deviation

16

E 12

Ol

d)
to
c

Group A Group B Group C Group D ..•:•: /. . :.

Fig. 4. Comparison of mean stress at failure for specimen groups: Group A, wrought precious alloy
post/gold core; Group B, CFRC post/gold core; Group C, CFRC/composite resin core; Group D,
stainless steel post/composite resin core. . , ., , . . -x - . x^ > . > .
606 P.A. King and D.J. Setchell

Fig. 5. Characteristic mode of failure of specimens subjected to an obliquely applied compressive load
at 130° in the Instron.

Table 2. Statistical analysis to compare the mode of failure of CFRC with that of metaUic post
restored teeth using the Chi-square test

Restoration type Site of tooth fracture

Above acrylic Below acryhc

CFRC posts (Groups B and C) 13 7

Metallic posts (Groups A and D) 2 18

i-" = 12-9; P < 0-001 (1 degree of freedom).

the precious alloy post and the gold core or the CFRC post and gold core specimens.
No significant difference (F > 0-05) was demonstrated between mean stress values
at failure for any of the remaining specimen groups.
Dislodgement of the post crown restoration and fracture of the tooth occurred at
failure in all specimens (Fig. 5). However, the degree of tooth fracture varied from
minor (above the acryhc resin) to major (below the acrylic resin). This variable was
used to compare the mode of failure between the CFRC post and metallic post
groups. Statistical analysis of the mode of specimen failure is presented in Table 2.
There was a highly significant difference (P < 0-001) between the CFRC post and the
metallic post specimens with regard to the level of tooth fracture.

Discussion •:...'.-' -
Extracted human teeth, despite their heterogenous nature, were considered to be
acceptable for this form of investigation, in agreement with Lovdahl and Nicholls
(1977). Due to difficulty in attaining enough suitable teeth in the earlier experimental
period, it was necessary to use maxillary central incisors in addition to maxillary and
 , Prototype CFRC prefabricated post 607

mandibular canines. However, these were evenly distributed between the groups.
Although silicone rubber was used by Guzy and Nicholls (1979) to simulate the
periodontal ligament, its ability to reproduce the complex visco-elastic properties
exhibited by the ligament in vivo is limited. In order to control the number of variable
factors, no attempt was made to simulate this ligament. The teeth were therefore
mounted in a heavily filled acrylic resin which had a modulus of elasticity similar to
that of human bone (range 7—20 GNmm~^).
A standard chnical technique, as described by Shillingburg and Kessler (1982), was
employed to produce the post-retained crown tooth preparation. The principal variable
in the preparations was the facio-lingual dimension of the anti-rotational feature,
which was dependent on the individual tooth morphology.
The quality of the CFRC posts and the consistency of their diameter has been
improved since the earlier reports on the material by Lovell (1983). The diameter of
the CFRC post was well matched to that of the twist drill. However, some samples
exhibited voids between the surface fibres at selected areas along the post. For the
purpose of this study no posts showing this phenomenon were used. CFRC can be
shaped and adjusted by conventional dental rotary instruments, but any alteration to
the surface could lead to separation of the surrounding fibres. The post does not
withstand excess force from dental instruments without the possibility of disrupting its
fibre pattern. The material was also rapidly eroded by an aluminium oxide/air abrasive.
An increasing compressive load at a fixed angle was chosen as the method of
testing the post-retained crown specimens to failure. The angle of 130° between the
compressive head and the long axis of the specimen was chosen to represent the
average inter-incisal angle as reported by Ballard (1948). Although the velocity of the
mandibular movement is variable and will tend to reduce nearer tooth contact, a
steady impact velocity of 5 cm min~^ was chosen to simulate this movement.
It is clear that this type of in vitro loading does not represent the complete
situation in vivo. However, it has formed the basis for tests previously reported in the
literature involving post and core systems (Kantor & Pines, 1977; Lovdahl & Nicholls,
1977; Guzy & Nicholls, 1979). Other forms of laboratory testing, such as cyclic
loading, may provide additional information concerning the failure characteristics of
post-retained crowns. Application of this type of in vitro loading is currently under
investigation by the authors.
The exacting specifications, and consequently the time required to produce each
specimen, meant that the sample size per group was limited. Taking this into account,
the results show that there was relatively little to choose between the groups. Statisti-
cally the stainless steel post and composite resin core specimens were slightly inferior
to either the precious alloy post and gold core or the CFRC post and gold core
specimens.
Observation of the specimens for the mode of failure showed that there was a
failure of the CFRC/composite resin interface, particularly in the core portion. This
was partly due to the materials themselves, and possibly due to the smooth rather than
serrated post surface. The CFRC post exhibited a 'greenstick' fracture due to the
progressive failure of the individual fibres, in contrast to the bending which occurred
with the wrought metallic posts. The CFRC post resulted in less damage to the tooth
structure at failure. It was difficult to deduce from the present study whether this
tendency was due to properties of the CFRC, the post shape, or the bond with
composite resin. , - . . . . , .,
608 P.A. King and D.J. Setchell •

One distinct advantage of a fractured CFRC post, compared to a metallic post, is

its relative ease of removal from the post hole by conventional rotary instruments.
This factor, combined with the less destructive nature of the tooth damage, would
probably allow salvage and repreparation with the minimum of complex treatment.
If the CFRC post is to be considered as a potential commercial product, improve-
ment of the quality control is required. The CFRC post does not possess a high abuse
factor, and caution is required when handling the material. However, it can be
conventionally used with a standard prefabricated post kit.
Although at present the CFRC post cannot be machined to produce a serrated
surface configuration without seriously altering its mechanical properties, the addition
of a retention feature at one end of the post is worthy of consideration by the
manufacturers.
This study tested isolated single-rooted teeth, so caution must be exercised when
applying these results to multi-rooted teeth or abutment teeth used for fixed prostheses.

Conclusions
The results of the experiment showed that post-retained crowns using a prefabricated
CFRC post exhibited properties comparable with, and in some cases better than,
those of existing prefabricated posts.
The mode of failure of specimens restored with a CFRC post was more favourable
to the remaining tooth tissue than was that of specimens restored with a metallic post.
On the basis of the limited experimental results obtained to date, CFRC prefabricated
posts could be considered for a restricted clinical trial.

Acknowledgements
The authors wish to thank Dunlop Medical Products for providing the experimental
CFRC material, Mr E. Davies of the BioMaterials Unit, Institute of Dental Surgery,
Mr J. Morgan of the Photographic Department, Eastman Dental Hospital for his
photographic assistance, and Miss K. Reynolds for her help in preparation of the
manuscript.

References
ADAMS, D . , WILLIAMS, D.F. & HILL, J. (1978) Carbon fibre-reinforced carbon as a potential implant
material. Journal of Biomedical Materials Research, 12, 35.
AssiF, D. & FERBER, A . (1982) Retention of dowels using a composite resin as a cementing medium.
Journal of Prosthetic Dentistry, 48, 292.
BALLARD, C.F. (1948) Some bases for aetiology and diagnosis in orthodontics. Dental Record, 68, 133.
BATES, J.F., STAFFORD, G.D. & HARRISON, A. (1975) Masticatory function — a review of the
literature. Journal of Oral Rehabilitation, 2, 349.
CHARLTON, G.A. (1970) The design of posts and cores for porcelain jacket crowns; Ph.D. Thesis,
University of Bristol.
CoLLEY, I.T., HAMPSON, E.L. & LEHMAN, M.L. (1968) Retention of post crowns. British Dental
Journal, 124, 63.
FEDERICK, D.R. (1974) An application of the dowel and composite resin core technique. Journal of
Prosthetic Dentistry, 32, 420.
GERSTEIN, H . & BRUNELL, S.C. (1964) Prefabricated precision dowels. Journal ofthe American Dental
Association, 69, 787.
GoNTAR, G . , LtBERMAN, R . , U R S T E I N , M . , FITSIG, S . & B E N - A M A R , A . (1986) R e t e n t i o n of dowels
using conclude composite resin as a luting m e d i u m . Dental Materials, 2, 118.
G U Z Y , G . E . & NICHOLLS, J . I . (1979) In vitro comparison of intact endodontically treated teeth with
and without endo-post reinforcement. Journal of Prosthetic Dentistry, 42, 39. . ' ; / • -^'rn; v
 Prototype CFRC prefabricated post 609

HANSON, E.C. & CAPUTO, A.C. (1974) Cementing mediums and retentive characteristics of dowels.
Journal of Prosthetic Dentistry, 32, 551.
HoAG, E.P. & DwYER, T.G. (1982) A comparative evaluation of three post and core techniques.
Journal of Prosthetic Dentistry, 47, 177.
HOBKIRK, J.A. (1982) The response of bone to carbon fibre reinforced carbon implants. In: Clinical
Applications of Biomaterials, p. 109. J. Wiley & Sons Ltd, Chiehester.
JOHNSON, J . K . & SAKUMURA, J.S. (1978) Dowel form and tensile force. Journal of Prosthetic Dentistry,
40, 645.
KANTOR, M.E. & PINES, M.S. (1977) A comparative study of restorative techniques for pulpless teeth.
Journal of Prosthetic Dentistry, 38, 405.
LOVDAHL, P . E . & NICHOLLS, J.L (1977) Pin-retained amalgam cores vs. cast gold dowel cores. Journal
of Prosthetic Dentistry, 38, 507.
LOVELL, M.J. (1983) The bond between CFRC and restorative materials. M.Sc. Thesis, University of
London.
0ILO, G. (1978) Sealing and retentive ability of dental luting cements. Acta Odontologica Scandinavica,
36, 317.
ROBERTS, D.H. (1970) The failure of retainers in bridge prostheses. British Dental Journal, 128, 117.
RuEMPiNG, D.R., LUND, M.R. & SCHNELL, R.J. (1979) Retention of dowels subjected to tensile and
torsional forces. Journal of Prosthetic Dentistry, 41, 159.
SHILLINGBURG, H.T. & KESSLER, J.C. (1982) Restoration of the Endodontically Treated Teeth, p. 13.
Quintessence Publishing Co. Inc., Chicago, Illinois.
SoRENSON, J.A. & MARTINOFF, J.T. (1984) Clinically significant factors in dowel design. Journal of
Prosthetic Dentistry. 52, 28.
STANDLEE, J.P., CAPUTO, A.A. & HANSON, E.C. (1978) Retention of endodontic dowels: effects of
cement, dowel length, diameter and design. Journal of Prosthetic Dentistry, 39, 401.
STEELE, G.D. (1973) Reinforced composite resin foundations for endodontically treated teeth. Journal
of Prosthetic Dentistry, 30, 816.
THOMPSON. N.G., BUCHANAN, R.A. & LEMONS, J.E. (1979) In vitro corrosion of Ti-6A1-4V and Type
316L stainless steel when galvanically coupled with carbon. Journal of Biomedical Materials
Research, 13, 35.