In 2000, we developed a Coulter counter constructed
using a membrane containing a single multiwall carbon nanotube (MWNT)
(JACS, 2000, 122, 12340). This single-pore MWNT membrane
was mounted on a Si/Si3N4 support structure and permitted detection
of polystyrene nanoparticles having diameters in the range of 60-100
nm. Statistical data analysis of the frequency of particle transport
events provided quantitative information on hydrodynamic and electrophoretic
mass transport rates. However, the signal-to-noise (S/N) ratio of this
system was not very good, and this made it difficult to determine particle
size because the transport time resolution was only about 1 ms.
Recently, research group members Dr. Takashi Ito and Dr. Li Sun showed
that the S/N ratio of the Coulter counter could be substantially improved
by replacing the Si/Si3N4 membrane support structure with one fabricated
from PDMS (Figure 1a). Because of its higher electrical resistance,
the PDMS support reduced the background noise level by a factor of >
20 compared to the Si/Si3N4 support structure, which made it possible
to accurately measure the height and width of resistive-pulse signals
resulting from transport of individual particles through the MWNT channel.
The immediate benefit of this approach is that both the size and the
surface charge of individual particles can now be determined. They also
show that polystyrene nanoparticles having nearly the same size, but
different surface charge, can be distinguished in mixed solutions of
varying composition.
Figure 2 shows data obtained with the device (experimental setup: Figure
1b) for solutions containing two types of polystyrene nanoparticles
having similar diameters (60 nm) but different surface charge density
(120 vs. 24220 COOH/particle). Figure 2a shows a typical current-vs.-time
plot for a 4:1 ratio of the two types of particles. The two pulses on
the left are wider and correspond to the particles having lower charge
density, and the pulses on the right are narrower and correspond to
particles having high charge density. Figures 2b and 2c show distribution
plots of particle diameter and transport time, respectively. It is clear
that it is not possible to distinguish between the two probes based
on their diameters (Figure 2b), but they can be differentiated based
on differences in their surface charge density (Figure 2c). The experimentally
determined particle diameters (ca. 60 nm) are the same as the diameters
measured by TEM, indicating that the Coulter counter accurately determines
particle size. In addition, particle surface charge density for the
wider signals was found to be ~120 charges/particle, which is the same
value determined by titration. In contrast, particle surface charge
for the narrower signals was determined to be ~800 charges/particle.
This value is substantially different from that determined by titration
(24,220 COOH/particle). We believe this discrepancy is a consequence
of ionic screening.
The improved S/N level described here brings us closer to our goal of
using carbon nanotubes having even smaller diameters to directly detect
large molecules such as polymers and DNA. A research paper describing
these results will be published in Analytical Chemistry on the ACS website
in April.
