Environmental Engineering Reference
In-Depth Information
larger number of transistors could be put together in a single integrated
circuit. Today, silicon-based integrated circuits are mainly fabricated using
CMOS processes, with a small slice devoted to bipolar technologies.
With respect to MOS, bipolar transistors have the advantage of a larger
transconductance factor, and a larger output resistance, so they exhibit
better performance in terms of current driving capability and achievable
voltage gain. Unfortunately, unlike MOS transistors, their base resistance is
finite and drastically decreases at high frequencies.
Recent bipolar transistors have a unity-gain frequency in the order of 20-
50 GHz. Since MOS unity-gain frequencies are one order of magnitude less,
BJTs are mainly used in RF or high-speed digital integrated circuits.
A typical simplified BJT cross-section is shown in Fig. 1.16, where the
so-called npn vertical transistor is depicted. It can be seen as two back-to-
back diodes because it is made up of two n-regions separated by a p-region
called base. The actual base region is the gray p-region in the figure whose
width, is small with respect to the other proportions and in modern
bipolar processes is between This region has a medium doping
concentration, in the order of The emitter is the heavily
doped n+ region in the figure. It has a width of a few and its doping
concentration is in the order of Finally, the actual collector
region is the gray n- epitaxial layer in the figure. The collector doping
concentration is in the order of To reduce the resistive path
that connects the actual collector region to the collector contact, a heavily
doped buried layer is grown below the device. The gray area represents the
region where the so-called transistor effect takes place and is the actual npn
transistor. Since this area extends vertically, the transistor is said to be
vertical. Finally, note that, unlike for MOSFETs, the transistor is not
symmetric.
Search WWH ::




Custom Search