Biomedical Engineering Reference
In-Depth Information
avalanche current causes the diode reverse bias voltage to drop below break-
down; thus pushing the junction to linear avalanching and even pure accumu-
lation mode. After quenching, the device requires a certain recovery time to
return to the initial state. The quenching and recovery times are collectively
known as
dead time
.
Recently, SPADs have been integrated in CMOS achieving timing res-
olutions comparable to those of PMTs [7]. Current developments in more
advanced CMOS technologies have demonstrated full scalability of SPAD
devices, a 25
m pitch, and dead time as low as 32 ns. The sensitivity, char-
acterized in SPADs as photon detection probability (PDP), can exceed 25-
50%. The noise, measured in SPADs as dark count rate (DCR), can be as
low as a few hertz [8, 9]. Thanks to these properties, SPCs based on SPADs
have been proposed for imaging where speed and/or event timing accuracy
are critical. Such applications range from fluorescence-based imaging, such as
Forster Resonance Energy Transfer (FRET), fluorescence lifetime imaging mi-
croscopy (FLIM) [10], and fluorescence correlation spectroscopy (FCS) [11], to
voltage sensitive dye (VSD) based imaging [12,13], particle image velocimetry
(PIV) [14], instantaneous gas imaging [15, 16], etc.
In the following sections we explore some applications of SPCs in com-
parison to conventional sensors, including potential fields of imaging where
SPADs can be a compelling implementation aspect of SPCs. We also look
at performance and architectural issues that the designer needs to take into
account when approaching real problems involving the use of SPCs.
ยต
13.2 Spectroscopy
Spectroscopy-based imaging has many incarnations. One, very successful one,
is known as fluorescence correlation spectroscopy (FCS); a technique used to
measure transitional diffusion coe
cients of macromolecules to count fluores-
cent transient molecules or to determine the molecular composition of a fluid
being forced through a bottleneck or a gap. In FCS, a femtoliter volume is
exposed to a highly focused laser beam that causes the molecules in it to
emit light in a well-defined spectrum. Figure 13.1 shows an example of optical
molecular response depending on the size and diffusion pattern of the mole-
cule. The physical causes of this behavior are related to the mobility of the
ligands. In the first case, free fluorescent ligands are continuously entering and
leaving the detection volume. In the second case, macromolecule ligands are
less mobile; thus producing slower but highly correlated intensity fluctuations.
Figure 13.2 shows an example of typical autocorrelation functions simulated
for different molecules [17].
A tighter correlation is observable in the case of low molecular weight
ligands. A macromolecule ligand generates a much more relaxed correlation.
A mixture of free and bound ligands is shown in the middle curve.
