Biomedical Engineering Reference
In-Depth Information
arranged normal to the surface and with one negative charge per base pair),
b
is the dis-
tance between the nearest unit charges along the DNA,
N
is the density of the immo-
bilized probe molecules,
δ
is the hybridization effi ciency and
θ
is the fraction of DNA
charge compensated by condensed cations (
1 corresponds to a neutral molecule).
The calculations using expression (4) and typical values of
C
dl
θ
F cm
2
,
C
i
20
µ
F cm
2
,
10
12
molecules cm
2
,
0.35
µ
λ
D
1 nm (for a 0.1 M electrolyte solution),
N
b
0.34 nm and
θ
0.76 show that the hybridization signal
ϕ
will be on the order of
about 3 mV for a hybridization effi ciency of
δ
0.5 (i.e. 50%), and about 6 mV for a
hybridization effi ciency of
1.0 (i.e. 100%). The calculated values of the expected
hybridization signal are in good accordance with the estimations performed in [14, 43]
using the Graham equation (
δ
10
12
hybridized 12-mer DNA cm
2
in an electrolyte solution with an ionic strength of 23 mM [43], and
3 mV from the 3
0.8 mV from the
10
11
hybridized 20-mer DNA cm
2
in a solution with an ionic strength of 1 mM
[14]) as well as with some experimentally observed results reported in the literature
(see Table 7.1). The model calculations for an Al
2
O
3
-gate FED with charged macro-
molecules also predict surface potential changes of several millivolts when the charge
density of molecules is doubled after the hybridization [52]. On the other hand, the
same model predicts larger signals of several tens of mV for FEDs with an uncharged
gate-insulator surface [52].
To reduce the counter-ion screening effect and thus to enhance the sensitivity of the
sensor, FEDs must be operated in:
4
●
very low ionic-strength solutions, and
●
have to use a high densitiy of the immobilized probe molecules (
N
10
12
molecules/cm
2
).
For these cases, however, a reduced probability of hybridization and therefore an
extended hybridization time and reduced sensor signals can be expected. Thus, the
theoretical basis of the sometimes experimentally observed “large” sensor signals (see
Table 7.1) still remains unclear.
A further task for the correct functioning of FEDs for DNA detection by its intrinsic
charge is given by the fact that the surface interaction should only occur between the
immobilized ssDNA and its complementary cDNA. There should be no interference
by any background interaction of small ions with the underlying gate surface of the
FED. Ideally, to “insulate” the underlying gate insulator from the solution, the immo-
bilized ssDNA molecules should form a perfectly homogeneous and tightly packed
monolayer without any pores or interstitial spaces. This demand, however, is contro-
versy to reported values for the theoretical maximum surface coverage of 25-30% for
a random-sequential adsorption of rod-like molecules [59]. Many experimental results
support this fact, i.e. the molecular layers are much less dense allowing an interstitial
penetration of small inorganic ions and water molecules to the underlying layer.
In summary, a practical realization of FEDs for the pure electrostatic detection of
charged macromolecules by their intrinsic molecular charge, especially in high ionic-
strength solutions such as physiological conditions, seems to be problematic. All the
above discussed “disturbing” factors, together with a possible undesired adsorption or
Search WWH ::
Custom Search