1070-4272/05/7809-1508C2005 Pleiades Publishing, Inc.
Russian Journal of Applied Chemistry, Vol. 78, No. 9, 2005, pp. 1508!1511. Translated from Zhurnal Prikladnoi Khimii, Vol. 78, No. 9,
2005, pp. 1533!1536.
Original Russian Text Copyright C 2005 by Beloshenko, Konstantinova, Borzenko, Glazunova.
AND POLYMERIC MATERIALS
Effect of Filler Aggregation on Properties of a Composite
Based on Epoxy Polymer
V. A. Beloshenko, T. E. Konstantinova, A. P. Borzenko, and V. A. Glazunova
Galkin Physicotechnical Institute, Ukrainian National Academy of Sciences, Donetsk, Ukraine
Received December 22, 2004; in final form, May 2005
Abstract-The plasticity of a composite based on epoxy polymer was studied as influenced by the concentra-
tion of the filler, nanosized powders containing aggregates of strongly bound ZrO
Composites based on epoxy polymers (EPs) are
widely used as building and functional materials. The
operation properties of composites can be efficiently
controlled either by changing epoxy oligomer3curing
agent system or by adding appropriate filler. It is
known  that the properties of these systems are
determined by various factors, i.e., nature and con-
centration of the filler, its morphology, particle size,
adhesion power, procedure of addition into the poly-
mer matrix, etc.
Ultradispersed fillers (nanosized particles) allowing
formation of composites with unique physicomech-
anical properties are actively studied today .
However, preparation of polymeric nanocomposites is
a complex procedure, because nanoparticles strongly
tend to aggregation. The effect of the resulting ag-
gregates on the properties of epoxy polymers is poorly
In this study we examined the effect of aggregation
of zirconium dioxide nanoparticles on the deformation
and strength properties of an epoxy composite used
for preparing thermally shrinkable adapter fittings
[8, 9]. These fittings are used to obtain strong and
hermetically sealed adhesive joints of tubes and rods
of various, even dissimilar, materials [9, 11].
As polymeric matrices we used the epoxy com-
pound containing diglycidyl ester (DGE) of hydro-
phthalic acid, block oligomer (BO) of aliphatic epoxy
resin and acidic oligoester, epoxy3diphenylolpropane
resin (ED) with molecular weight of 3903430, iso-
methyltetrahydrophthalic anhydride (IMTHPA), and
tris(dimethylaminomethyl)phenol (TDAP)  with the
following component ratio, weight fraction: 77.5
DGE, 20 BO, 2.5 ED, 68 IMTHPA, and 0.7 TDAP.
The filler powders were prepared by precipitation
from zirconium oxychloride solutions: no. 1, from
chemical reagents in laboratory; no. 2, from industrial
intermediate product (hydroxide) with subsequent
treatment in laboratory; and no. 3 was commercial
product. The filler concentration was varied from 0 to
30 wt % (with EP content taken as 100%).
The experimental samples were prepared by mixing
of the epoxy composite components with ZrO
powder in a given ratio at 60370oC on a mechanical
stirrer with subsequent curing in appropriate molds
at 120oC for 4 h. Curing was performed with rotation
of the molds to prevent sedimentation. Depending
on the filler, three types of polymeric composites were
Thermally shrinkable fittings (couplers) were pre-
pared by rolling of tubular billets. The initial size of
fittings was as follows: 20 mm internal diameter d
35 mm height, and 3.5 mm thickness. Curing was
carried out in the hyperelastic state with subsequent
cooling of the deformed billet on a mandrel to the
temperature lower than the glass transition point T
of the polymer matrix. The degree of deformation was
determined from the expression e =(d 3 d
d is the mandrel diameter.
In accordance with GOST (State Standard) 112623
80, the tensile strength tests were carried out with
370 5 mm plates with thickness of 2 mm on a tensile-
testing machine equipped with a thermal chamber;
the rate of a mobile clamp was 10 mm min
The microhardness H
was measured using sam-
ples cut off from the fittings after thermal shrink-age.
The load on the indenter was 5 N; the average H
calculated from 10315 marks.
Differential scanning calorimetry (DSC) and ther-
momechanical analysis (TMA) were performed on