Imaging and Molecular Motors - Single Molecule Dynamics in Life Science - page 47

sin 2 (x)/x 2 [22]. It is similar to a two-dimensional Gaussian shape with small fringes
added around the periphery. The minimum between the main peak and the rst
fringe has a diameter of d 0 m, which fits comfortably within

the NA e limits of 1.33 and 1.45 as shown in Figure 3.2B and C. The portion of the
outer fringes at NA e lower than 1.33, resulting in light propagated into the sample
compartment contains approximately 2% of the total energy. Similarly the outer
fringe scattered by the aperture is

¼ 3.28 l f L1 /d L1 ¼ 116 m

2%. In this example, then, 96% of the energy is

totally re ected as intended. Objective lenses with NA o even higher than 1.45 are

desirable to bring the incident beam farther away from the 1.33 NA c critical radius,
but some of them require special coupling fluids and special materials for the

specimen slides. Lenses from Olympus, Inc. and the Nikon, Inc, with NA o values
1.45 - 1.49 have come into broad use for single molecule TIRF experiments.
The component of the incident radiation that scatters off of optical surfaces in
the microscope toward the detector is several orders of magnitude greater than the
single-probe fluorescence emission. Care must be taken to block this back-scattered
excitation from reaching the detector. The dichroic mirrors and interference lters
commonly used in standard epi- uorescence microscopy discriminate wavelengths
suf ciently well that background radiation at the excitation wavelength is usually not
a dif culty. Fluorescence from the microscope objective, glass slides and other optics
in the excitation path, can sometimes give signi cant background intensity. The
optical train should be designed with these factors in mind.
The intensity of the evanescent wave decays exponentially with distance, z,
from the re ecting interface, I I 0 exp(z/d), where I 0 is the intensity at the slide

surface and the spatial decay constant d, also termed the penetration depth, is given
by [11]

¼

q

n 1 sin 2

q

NA e

n 2 n 2

d ¼ l= 4 p q ¼ l= 4 p ð 3 : 1 Þ

1

where isthefree-spacewavelengthandNA e istheeffectivenumericalapertureof

the excitation rays. The decay parameter is plotted versus NA e in Figure 3.3A,

l

Figure 3.3 Relationships between the effective numerical
aperture (NA e ) of the incident light and the penetration depth
(A) and the light intensity in the evanescent wave at z = 0 (B). The
upper and lower lines in B are plotted for p- and s-polarized
illumination.