Biomedical Engineering Reference
In-Depth Information
6. It is then reasonable that the PDMS molecules would adsorb, through an elec-
trostatic binding of the negatively charged carboxylate groups, onto the cationic
sites of the nanoparticle surface. On the other hand, the nonfunctional end of the
PDMS oligomer would provide a brush layer that extended outwards from the
nanoparticle surface to prevent particle aggregation. Finally, the DCM was removed
under vacuum and the Fe
3
O
4
/PDMS particles were decanted, washed with water
and methanol, and then dried overnight at 40 °C under reduced pressure. In the
following descriptions, this sample will be referred to as
0 - pass
.
In order to remove aggregates and large particles, a dispersion of the Fe
3
O
4
/
PDMS nanoparticles in chloroform (CHCl
3
) was passed through magnetic columns
(i.e.,
B
magnet, placed approximately 10 cm apart from the columns, was used to generate
a magnetic fi eld of 0.24 T. The basic idea of the magnetic separation was that larger
particles and aggregates would be entrapped by the separation column because of
their magnetic moments. After sonication, the chloroform dispersion of the par-
ticles was passed through a second column at a fl ow rate of approximately 20 ml
min
− 1
(
1 - pass
sample). Alternatively, a portion of the chloroform dispersion of
nanoparticles was passed through fi ve freshly prepared separation columns (
5 - pass
sample). The collected dispersions were dried under vacuum and weighed to
determine the amount of material that had been retained in the separation
columns. The particles/polymer compositional ratios were determined using ther-
mogravimetric analysis (TGA) under an inert gas (N
2
). The materials that had been
magnetically separated had lower magnetite concentrations compared to those of
the original materials [115]. Since the separations preferentially removed aggre-
gates and larger particles, those complexes that passed through the columns
should have higher magnetite specifi c surface areas, leading to higher mass frac-
tions of polymer in the eluted materials. TEM investigations indicated that mag-
netically separated samples had a lower fraction of aggregates and larger particles,
leading to a decrease in mean particle size. An accurate statistical analysis of the
TEM data confi rmed that the magnetic separation had resulted in particles of a
smaller size and with narrower size distributions. Moreover, multiple passes
through the separation columns produced smaller average sizes and a narrower
size distribution.
The thermal dependence of magnetization was measured using the ZFC
and FC protocols.
T
max
in ZFC measurements was seen to decrease from
0 - pass
to
5 - pass
samples, due to the decrease in particle size, as showed in the statistical
analysis of TEM data. The irreversibility temperature in ZFC-FC curves (see
Section 12.2.3.1 ) for
0 - pass
and
1 - pass
samples was higher than 300 K, indicating
the presence of large particles and aggregates, in agreement with TEM results.
T
irr
for the
5 - pass
sample was lower than room temperature, indicating that the
magnetic separation technique was an effi cient method for removing larger
particles and aggregates. The dependence of magnetization on the external
magnetic fi eld was also studied. The value of the magnetic moment has been
normalized for the magnetite content which, in turn, was extracted from the
thermal analysis data. The
M
s
value decreased with decreasing in particle size,
∼
6 g of soft iron granules fi rmly packed into 3 ml plastic syringes). A Nd
−
Fe
−
