Biomedical Engineering Reference
In-Depth Information
conventional loading loops were not practical. Instead, the crossed-T injector and appropri-
ate high-voltage electrophoretic control systems evolved. his allowed many separate picoliter
samples to be accurately positioned and sent down the CE channels under computer control.
he other problem was one of geometry. To put one or more long chromatographic separa-
tion columns onto a inite-sized (perhaps 5-in.-diameter) glass wafer, the channels cannot be
straight. To pack detection zones into a small region, which is also critical, it proved necessary
to use a radial packing of the chromatographic columns (e.g., see
Figure 4.3a
). However, the
radius of the wafers was not adequate for suicient separation, so the channels had to typically
have a few hairpin (180 degree) turns. hese turns badly smeared out the bands, costing as much
in resolution as had been gained by allowing the channels to be more than one wafer radius in
length. here were two reasons, one of which was a concentration of current on the inner edge
of the turn, and the other was the longer path taken by analyte molecules on the outer edge of
the channel. It was ultimately shown by the Mathies and Santiago groups that (some rather
unintuitive) modiications of the geometry of the turn itself allowed restoration of almost all
the resolution.
4.4.2 Continuous-Flow and “Free-Flow” Electrophoresis
his electrophoresis modality has found less efective uses than CE but was historically devel-
oped much earlier than CE. In fact, this idea can be traced back to the dawn of microluidics,
well before the irst microtechnologies had been invented. (Note that it is possible to construct
rudimentary microluidic devices with anything that conines luid low to a small space—a thin
glass tube, a sheet of paper, or two glass plates separated by a spacer will do.) Electrophoretic
separation does not need to be performed along the length of the capillary. For certain applica-
tions (in particular, where the sample can be guaranteed to low continuously for as long as data
needs to be gathered), it may be advantageous to apply the ield perpendicular to the channel
(
Figure 4.4a
). In July 1939, J. St. L. Philpot, a biochemist from Oxford University, submitted a
report entitled “he use of thin layers in electrophoretic separation” to the
Transactions of the
Faraday Society
in which he used laminar low principles to layer ive input lows (the central
one containing protein solution); the voltage delected the protein solution and separated the
“wanted” fractions to one output and the “unwanted” fractions to other outputs. To describe
laminar low, he reported that “the ive solutions low quite smoothly on top of each other.”
he idea of
continuous-low electrophoresis
was picked up by a few researchers ater Philpot,
most notably by Kurt Hannig from the Max Planck Institute (Munich) who in 1950 designed an
electrophoresis device in which gravity-fed low was sustained by ilter paper while subjected to
Sample
Buffer
a
b
Buffer
Sample
Electric
field
pH
gradient
x
y
Pos.
0
Neg.
High pI
Low pI
FIGURE 4.4
Continuous-low. electrophoresis. and. continuous-low. isoelectric. focusing.. (From.
Nicole. Pamme,. “Continuous. low. separations. in. microluidic. devices,”.
Lab Chip
. 7,. 1644-1659,.
2007.. Reproduced. with. permission. from. The. Royal. Society. of. Chemistry.. Figure. contributed. by.
Nicole.Pamme.)
