Rapid Prototyping Journal

Emerald Article: Stereolithography of ceramic suspensions

G. Allen Brady, John W. Halloran

Article information:
To cite this document: G. Allen Brady, John W. Halloran, (1997),"Stereolithography of ceramic suspensions", Rapid Prototyping
Journal, Vol. 3 Iss: 2 pp. 61 - 65
Permanent link to this document:
http://dx.doi.org/10.1108/13552549710176680
Downloaded on: 08-10-2012
References: This document contains references to 6 other documents
Citations: This document has been cited by 18 other documents
To copy this document: permissions@emeraldinsight.com

Users who downloaded this Article also downloaded: *

Sandy Bond, (2011),"Barriers and drivers to green buildings in Australia and New Zealand", Journal of Property Investment &
Finance, Vol. 29 Iss: 4 pp. 494 - 509
http://dx.doi.org/10.1108/14635781111150367

Hui Chen, Miguel Baptista Nunes, Lihong Zhou, Guo Chao Peng, (2011),"Expanding the concept of requirements traceability: The role
of electronic records management in gathering evidence of crucial communications and negotiations", Aslib Proceedings, Vol. 63
Iss: 2 pp. 168 - 187
http://dx.doi.org/10.1108/00012531111135646

Franois Des Rosiers, Jean Dub, Marius Thriault, (2011),"Do peer effects shape property values?", Journal of Property
Investment & Finance, Vol. 29 Iss: 4 pp. 510 - 528
http://dx.doi.org/10.1108/14635781111150376

Access to this document was granted through an Emerald subscription provided by UNIVERSITY OF GUELPH

For Authors:
If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service.
Information about how to choose which publication to write for and submission guidelines are available for all. Please visit
www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
With over forty years' experience, Emerald Group Publishing is a leading independent publisher of global research with impact in
business, society, public policy and education. In total, Emerald publishes over 275 journals and more than 130 book series, as
well as an extensive range of online products and services. Emerald is both COUNTER 3 and TRANSFER compliant. The organization is
a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.

Stereolithography of
ceramic suspensions
G. Allen Brady and
John W. Halloran
The authors
G. Allen Brady and John W. Halloran are both in the
Material Science and Engineering Department, College of
Engineering, The University of Michigan, Ann Arbor,
Michigan, USA.
Abstract
Rapid prototyping of ceramics is accomplished with
stereolithography by using an SLA machine to build the
ceramic green from a UV-curable suspension of ceramic
powders a "ceramic resin". Objects are later sintered in
a separate furnace to complete the process. Aluminum
oxide resins based on hexanediol diacrylate are character-
ized for curing behaviour by photo-rheology and differen-
tial photo calorimetry with a UV lamp, and with a HeCd
laser using "windowpanes", single strings, and walls.
Rapid Prototyping Journal
Volume 3 Number 2 1997 pp. 6165
MCB University Press ISSN 1355-2546
61
Introduction
Free form fabrication of ceramics can be
accomplished by stereolithography (SLA) by
using as the resin a suspension of ceramic
powder in a UV-curable medium (Griffith
and Halloran, 1996). Several ceramic resins
for aluminum oxide, silicon dioxide, and
hydroxyapatite in acrylate and aqueous media
have been explored (Brady et al., 1996).
These very concentrated suspensions of
ceramic powder, although nearly opaque due
to scattering of the UV radiation by the
ceramic particles, do have a useful depth of
cure when irradiated by a UV laser (Griffith
and Halloran, 1997).
This article concerns the behaviour of a
simple resin consisting of aluminium powder
suspended in hexanediol diacrylate (HDDA).
The cure depth as a function of dose from the
HeCd laser in the SLA 250 is examined. The
kinetics of the polymerization reaction are
examined by thermal analysis, using UV
radiation from a Hg-vapour lamp in a differ-
ential photo calorimeter (DPC). Gelation
(i.e. the solidification due to curing)is charac-
terized by monitoring the storage modulus
(G)in a rheometer during irradiation with a
UV lamp (photo-rheometry).
Experimental procedures
Ceramic resin
The ceramic resin consists of 50 volume per
cent Al2O3 powder (RCHP DBM, Malakoff
Industries, Malakoff, Texas, USA; median
particle size = 0.3 microns) in 1,6 hexanediol
diacrylate (HDDA; Photomer 4017, Henkel
Corp., Kankakee, Indiana, USA), stabilized
to create a fluid suspension using a quaternary
ammonium acetate dispersant (Emcol CC-
55, Witco Corp., Organics Division, New
York, USA). The photoinitiator was 1-
hydroxy-cyclohexyl phenyl ketone (Irgacure
184, Ciba-Geigy, Ciba Additives, Tarrytown,
New York USA) at a concentration of 0.5
weight per cent of monomer. The ceramic
This research was supported by the US Office of
Naval Research under grant N00014-95-1-0527,
through Dr S. Fishman and R. Wachter. Univer-
sity of Nottingham. This paper was presented at
the Sixth European Conference on Rapid Proto-
typing and Manufacturing, University of Notting-
ham, 1-3 July 1997.
resin is stable against sedimentation with a viscometer for measuring photosensitive
nearly Newtonian viscosity of 200 mPa-s, polymers. The purpose of this measurement
which is similar to a standard commercial technique is to probe the change in the liquid-
epoxy resin (Cibatool SL5170, Ciba-Geigy, solid transition during UV-curing of these
Ciba Specialty Chemicals, Performance ceramic suspensions.
Plastics Div., Lansing, Michigan, USA). The differential photo calorimetry (DPC)
Stereolithography experiments were conduct- experiments were conducted in a PerkinElmer
ed in a 3D Systems, Inc. SLA 250/40 appara- DSC-7 apparatus (Perkin-Elmer Corp.,
tus, modified with a mini-vat (Brady et al., Norwalk, Connecticut, USA) which was
1996), which allows the apparatus to function modified to introduce UV radiation from the
with only two litres of experimental resin. The Pen-Ray lamp at 7.3mJ/cm2 sec as shown in
behaviour of the diacrylate-based ceramic Figure 2. The experiments were run at 40C
resin is compared with the commercial Ciba-
with a sample size of 5-10mg, for which
tool SL5170 epoxy resin typically used with
isothermal conditions could be satisfactorily
the SLA 250.
maintained by the DPC. The data reported in
this paper are the average of three to five runs.
Photo-rheology and differential photo
The sample chamber is contained in a flowing
calorimetry
N2 atmosphere to eliminate the effect of
Photo-rheological measurements were taken
with a Bohlin CS-50 rheometer (Bohlin oxygen inhibition.
Reologi, Lund, Sweden) in a plate-plate
configuration. The bottom plate of the Results and discussion
rheometer was made from UV-transparent
silica glass. Ultraviolet radiation at a dose rate The results from DPC appear in Figure 3 as
of 2.5mJ/cm2sec was introduced through the plots of heat flow versus dose. The pure
silica glass plate from a Pen-Ray lamp (Pen- HDDA (0 per cent aluminium oxide) reacts
Ray lamp, 11SC-1 L, UVP Inc., Upland, very rapidly at a low dose, reaching a peak
California, USA) at a rate of 2.5mJ/cm2sec.
This lamp emits the spectrum characteristic Figure 2 Schematic of differential photo calorimetry
of a medium pressure Hg lamp, with the measurement
largest emission at 365nm. The apparatus is
illustrated in Figure 1. The excitation condi-
tions were oscillation at 10Hz at an applied
shear stress of 500Pa and an equilibrium time
of 30 seconds. For each measurement, the UV
lamp was turned on at about 16 seconds and
remained on throughout the test. All the data
presented here were collected at a gap thick-
ness of 0.2mm.
This technique monitors the transition Figure 3 Heat flow versus dose from DPC measurements of (1) HDDA (0%
between viscous liquid ( ~ 200mPa-s) to fully aluminium oxide); (2) volume per cent aluminium oxide in HDDA; and (3)
cured solid (G ~ 1MPa) during oscillation in Cibatool SL5170
a plate-plate geometry. Otsubo et al. (1984a)
describe a similar plate-plate oscillatory

Figure 1 Schematic of photo rheology measurement

62
 heat evolution rate of 120J/gm-sec. Addition
of 50 volume per cent aluminium oxide (or 83
weight per cent) greatly moderates the rate of
the reaction, so that the peak heat evolution
rate drops to 17.5J/gm-sec. This effect
appears to be greater than a simple dilution
with inert filler, so probably represents the
effect of significant scattering by the alumini-
um oxide particles. The reactivity of the
aluminium oxide-HDDA resin is higher than
the SL5170 epoxy system.
The DPC data can be expressed as conver-
sion of monomer versus dose by assuming
that the unfilled resins are fully reacted. These
data are presented in Figure 4. Note that
unfilled HDDA (0 per cent aluminium oxide)
is 75 per cent reacted at a dose at which the 50
per cent aluminium oxide-HDDA resin and
the epoxy are only 3-5 per cent reacted. The
aluminium oxide-filled resin is comparable in
behaviour up to about 70 per cent conversion.
Beyond that point the curing of the ceramic
resin is greatly retarded. About 20 per cent of
the HDDA monomer in the aluminium-oxide
system remains unreacted at the end of the
experiment.
Photo-rheological measurements provide
in situ monitoring of UV-curing films. As the
resin cures to a gel point, the storage modulus
increases over many orders of magnitude.
This change is easily detected under oscillat-
ing stress conditions applied by the rheome-
ter. The dose required to begin curing, or
cure dose, is indicated by a sudden large
change in the storage modulus. This cure
dose varies predictably with the gap thickness,
following a Beer-Lambert type relation where
Figure 4 Per cent of monomers conversion versus dose for (1) HDDA (0 per
cent aluminium oxide); (2) 50 volume per cent aluminium oxide in HDDA; and
(3) Cibatool SL5170
63
Log (cure dose) varies linearly with gap
thickness. Other investigators (Otsubo et al.,
1984b) have observed such a relation between
cure dose and the depth at which films cure.
From this, resin parameters of Dp = 156
microns and an Ec = 21 mJ/cm2 are inferred
for the 50 per cent aluminum oxide resin as
measured by this technique.
The mechanical response of curing the 0.2
mm layer in the photo rheometer is shown in
Figure 5, which plots the storage modulus as a
function of dose. The two HDDA-based
resins display a sudden increase in storage
modulus at a dose of about 60-80mJ/cm2.
The storage modulus reaches a limiting value
around 90 MPa after a dose of about
100mJ/cm2. The onset of curing for the epoxy
SL 5170 is also about 80mJ/cm2, but the
storage modulus builds gradually with dose,
in what appears to be a two stage process. A
dose of 800 mJ/cm2 is required to achieve a
limiting value of modulus of about 8.8MPa.
Measurements of the loss modulus and shear
viscosity versus dose also show a sudden
increase occurring at the same cure dose as
observed with the storage modulus. There is
essentially no change in viscosity of the resins
at exposures smaller than the cure dose.
While the storage modulus versus dose is
similar for the HDDA resin and the 50 per
cent aluminium oxide-HDDA resin, there is a
large difference in the modulus versus
monomer conversion. Nearly 70 per cent of
the monomer in the unfilled HDDA is reacted
at the cure dose, so a large fraction of polymer
forms before gelation. The aluminum oxide-
filled HDDA has about the same cure dose,
Figure 5 Photo-rheological measurements comparing storage modulus
versus dose for: (1) HDDA (0 per cent aluminium oxide); (2) 50 volume per
cent aluminium oxide in HDDA; and (3) Cibatool SL5170
 but only about 2 per cent of the monomer has
reacted at the cure point. Very little polymer
has formed at the point of gelation, certainly
not enough for polymeric gelation. Recall that
the 50 per cent aluminium oxide-HDDA resin
is a quite concentrated colloidal suspension of
submicron powders, kept in dispersion by the
action of the dispersant. Without the disper-
sant, HDDA containing 50 volume per cent
aluminium oxide powder is a stiff paste a
colloidal gel. Thus the initial curing of the
ceramic resin might be by a photochemically-
induced colloidal gelation.
The penetration depth and the critical
exposure energy of the 50 per cent aluminium
oxide-HDDA resin was characterized on the
SLA 250 using the windowpanes tech-
nique. The thickness of windowpanes (cure
depth) versus dose appears in Figure 6, from
which the build parameters Dp = 62 microns
and Ec = 33mJ/cm2 can be derived. These
values are quite different from the typical SL
5170 epoxy resin, so building with this partic-
ular ceramic resin requires much longer
exposures.
Stereolithography builds objects from
strings (cured lines), defined as the scan-
ning laser beam traverses the surface of the
resin. The cross section of a string in a non-
scattering resin is a bullet-shaped parabolic
curve (Jacobs, 1992), typically much deeper
than it is wide. Cured line profiles in the
highly-scattering ceramic resin are much
flatter. Figure 7 is an optical micrograph of a
single string in this aluminium oxide resin. It
is about 230 microns deep and 520 microns
wide, flattened by attenuation of the forward
beam by ceramic filler, and widened by side
scatter. The edges are slightly scalloped; a
Figure 6 Depth of cure versus UV laser dose from SLA windowpanes
experiment
64
Figure 7 Cured line profile for SLA laser-cured line for 50
volume per cent aluminium oxide in HDDA
consequence of side scatter of the UV. These
effects are sensitive to the difference in refrac-
tive index of the filler and the medium, so are
different for other ceramic powders and
monomers (Brady et al., 1996).
After the SLA build is completed, the
object is a ceramic green body with the
photocured resin as its binder. An optical
photograph of mechanical test specimen is
shown in Figure 8. This specimen is shown
as-cured at about 38 mm in diameter and
4.5mm high with 26 layers. At this point, the
polymeric binder is removed by heating in air
at 1C/min, a typical rate for binder removal.
By about 600-800C in air, all of the polymer
residue and char is removed. This grade of
aluminium oxide powder begins to sinter
above 1,200C, and sinters to a pore-free state
at 1,600C. Figure 9 is a scanning electron
micrograph of the surface of the a sintered
aluminium part, showing excellent densifica-
tion of the ceramic.
Conclusions
The photocuring of the ceramic resins has
been characterized with photo-differential
Figure 8 As-cured mechanical test specimen produced from aluminium
oxide/HDDA resin. Plan view (A) on left. Perspective view (B) on right.
Figure 9 High magnification SEM micrograph of sintered
aluminium oxide
scanning calorimetry and photo-rheometry.
Photo-DSC reveals that the aluminium oxide
filled HDDA system polymerizes more slowly
than simple HDDA. Photorheometry shows
that the aluminium oxide-filled HDDA
system physically cures at nearly the same
exposure dose as the pure HDDA. Gelation
occurs at only 3 per cent reaction of the
monomer, so onset of rigidity could be due to
colloidal gelation, rather than polymeric
gelation.
For the aluminium oxide-HDDA resin at
this photoinitiator content the penetration
depth Dp = 62 microns and the critical expo-
sure energy Ec = 33mJ/cm2 as measured by
the windowpanes technique. Owing to scat-
tering of the UV radiation by the ceramic
particles, the cured line profile for this
65
ceramic resin is flattened, more than twice as
wide as it is deep. Vertical walls produced by
stacking these lines are therefore scalloped.
Pore free aluminium oxide ceramics can be
built by stereolithography using a concentrat-
ed suspension of ceramic powder in a UV-
curable medium. Despite side scattering,
sufficient penetration and subsequent curing
of the aluminium oxide-HDDA resin allows
for significantly-sized parts to be built quite
easily. Subsequent applications for this resin
system include direct shell mould and core
fabrication for investment cast products as
well as functional ceramic prototypes.
References
Brady, G.A., Chu, T.-M. and Halloran, J.W. (1996), Curing
behavior of ceramic resin for stereolithography, in
Beaman, J.J., Barlow, J.W., Bourell, D.L. and Craw-
ford, R.H. (Eds), Solid Free Form Fabrication Pro-
ceedings, SFF Symposium, Austin, TX, pp. 403-10.
Griffith, M.L. and Halloran, J.W. (1996), Free form
fabrication of ceramics via stereolithography,
Journal of the American Ceramic Society, Vol. 79
No. 10, pp. 2601-8.
Griffith, M.L. and Halloran, J.W. (1997), Scattering of
ultraviolet radiation in turbid ceramic suspensions,
Journal of Applied Physics (in press).
Jacobs, P.F. (1992), Rapid prototyping and manufacturing:
fundamentals of stereolithography, Society of
Manufacturing Engineers, Dearborn, MI, p. 87.
Otsubo, Y., Amari, T. and Watanabe, K. (1984a), Dynamic
viscoelastic measurements of photosensitive
polymers, Journal of Applied Polymer Science,
Vol. 29, pp. 57-66.
Otsubo, Y., Amari, T. and Watanabe, K. (1984b), Rheologi-
cal behaviour of epoxy acrylate prepolymer during
UV-Curing, Journal of Applied Polymer Science,
Vol. 29, pp. 4071-80.