Intel Corporation
2200 Mission College Blvd.
P.O. Box 58119
Santa Clara, CA 95052-8119
Glossary
INTEL HIGH-K, METAL GATE TRANSISTOR GLOSSARY OF TERMS
crystalline silicon substrate
gate electrode
source drain
gate dielectric
channel
Basic CMOS transistor
Transistor – A simple on/off switch that processes the ones and zeroes of electrical data. Digital
chips, such as microprocessors, consist of millions of such transistors connected together by
copper wires in a specific pattern. The current Intel® Core™2 quad-core processor has more
than 500 million tiny transistors. As technology advances, the objective is to make these
transistors smaller, faster, cheaper and less energy-hungry, all of which leads to more powerful
chips. Current flow from the source to the drain is determined by whether the gate is at high or
low voltage, much as a light switch’s state (up or down) controls whether or not current flows to
a light bulb.
Source – The part of the transistor from which the current flows. It consists of doped silicon –
meaning silicon containing impurities that lower resistance.
Drain – The part of the transistor to which the current flows. It is doped with impurities in the
same way as the source. A transistor is completely symmetrical, meaning current can flow from
source to drain, or drain to source.
Gate (also known as gate electrode) – A region at the top of the transistor whose electrical state
determines whether the transistor is on or off. Traditionally, the gate is made of polycrystalline
silicon (polysilicon) – or silicon whose atoms are randomly placed and are not in a grid-like
structure.
-- more -- Intel/Page 2
Channel – The region between the source and drain, where current flows when the transistor is
in the ‘on’ state. It consists of silicon in the crystalline state, that is, silicon in an orderly grid-like
(lattice) structure.
Intel/Page 2
Gate dielectric – A thin layer underneath the gate that isolates the gate from the channel. In
today’s chips, it consists of silicon dioxide.
Silicon dioxide – Molecules consisting of one silicon and two oxygen atoms, which form a good
insulator (non-conductor of electricity). For a gate dielectric, a thin silicon dioxide layer is
desirable for high performance. The problem is that the thinner the layer, the higher the leakage
through it – hence the effort to replace it with new materials that preserve its properties but don’t
need to be so thin.
High-k material – A material that can replace silicon dioxide as a gate dielectric. It has good
insulating properties and creates high-field effect (hence the term “high-k”) between the gate and
channel. Both are desirable properties for high-performance transistors. “k” (actually the Greek
letter kappa) is an engineering term for the ability of a material to hold electric charge. Think of
a sponge: it can hold a lot of water. Wood can hold some, but not as much. Glass can’t hold any
at all. Similarly, some materials can store charge better than others, and hence have a higher “k”
value. Also, because high-k materials can be thicker than silicon dioxide – while retaining the
same desirable properties – they greatly reduce leakage.
Leakage – Current flowing through the gate dielectric. In an ideal situation, the gate dielectric
acts as a perfect insulator, but as it is made ever thinner (in Intel’s 65nm process, it is a mere 5
atomic layers thick!), current leaks through it. This leads to undesirable results. The transistor
doesn’t behave as it should, and it consumes more power than it should. In comparison, think of
the waste caused by a leaky faucet.
NMOS transistor (also known as n-type transistor) – A transistor that is on when its gate is at
high voltage, and off when its gate is at low voltage.
PMOS transistor (also known as p-type transistor) – A transistor that is opposite to an NMOS
transistor, meaning it is off when its gate is at high voltage and on when its gate is at low voltage.
CMOS transistor (complementary metal oxide semiconductor) – A process technology in
which both NMOS and PMOS transistors exist. All modern logic chips such as microprocessors
and chipsets use CMOS due to its ability to deliver a combination of high performance and low
power, all at a low cost.
Threshold voltage – The voltage level between high and low that distinguishes whether a
transistor is on or off. For an NMOS transistor, if its gate is above the threshold voltage, it is
“on.” If it is below the threshold voltage, it is “off.” A PMOS transistor exhibits complementary
behavior. Transistors are designed to have a low threshold voltage, as this leads to high
performance (think of a racing car with a low center of gravity).
-- more --
Intel/Page 3
-- more --
Threshold voltage pinning (also known as Fermi level pinning) – One of two undesirable
effects when a high-k gate dielectric is combined with a polysilicon gate electrode. Due to some
defects that arise at the gate dielectric/gate electrode boundary, it becomes difficult to adjust the
threshold voltage to a low value, which is needed for high performance. The problem goes away
when the gate electrode is a specific metal, rather than polysilicon. The choice of metal is
different for NMOS and for PMOS transistors.
Phonon scattering – The second undesirable effect when a high-k gate dielectric is combined
with a polysilicon gate electrode. This phenomenon limits electron mobility and hence degrades
performance. The problem goes away when a gate made from a specific metal replaces the
polysilicon gate, and the right process recipe is applied.
Related terms
Moore’s Law – A prediction (not truly a law) made by Intel co-founder Gordon Moore that the
number of transistors on a chip double every two years. Intel’s microprocessors have followed
this law very closely, beginning with the 4004 in 1971, with just over 2,000 transistors, and
leading up to today’s Itanium® 2 processor that has 410 million transistors. In general, transistor
density is roughly doubled with each new process generation, which occurs every two years.
Strained silicon – A technique for speeding up transistors. As described above, the silicon atoms
in the channel are packed neatly in a grid-like (lattice) structure. It has been known for decades
that stretching the grid so the silicon atoms are slightly farther apart than in their natural state
makes NMOS transistors switch faster (similarly, compressing the lattice slightly speeds up
PMOS transistors). This stretching/compressing is known as straining. Intel uses special
techniques to strain its 90nm process NMOS and PMOS transistors to improve their
performance.
Low-k dielectric – Low-k dieletrics are used to insulate on-chip interconnects and should not be
confused with the high-k dielectrics. In transistor gate dielectrics, high-k is desirable as it gives
high performance with low leakage. In interconnects, low-k is desirable as it leads to faster
signal transmission times.
Silicon-on-insulator (SOI) – SOI refers to the use of a layered silicon-insulator-silicon substrate
on which transistors are built, rather than a simple (bulk) silicon substrate. Some companies
claim to get some performance and/or power benefits from SOI over bulk silicon. Intel’s analysis
shows that such benefits, if any, are marginal, and do not justify the substantial cost increase of
SOI wafers. Intel has never used, nor does it plan to use, partially depleted SOI (PD-SOI) that
others are using. There is another type, however, called fully-depleted SOI (FD-SOI) that is
under investigation at Intel and is not being used by any chip makers today.
Tri-gate transistor – A new type of transistor that Intel has designated a potential candidate for
its future process technology generations. The transistors described earlier in this document are
planar transistors. That is, they have a single flat gate that is parallel to the surface of the silicon
substrate. A tri-gate transistor employs a novel three-dimensional structure where the gate wraps
around three sides of the silicon channel. A traditional planar transistor could be likened to a
highway on top of a mesa (a flat-top mountain with vertical sides) with the electronic signals Intel/Page 4
traveling like cars across the flat surface of the mountaintop. With the new, elevated 3D design,
the signals travel not only across the flat top, but along both vertical sidewalls as well. Hence, the
‘tri-gate’ name.
Static Random Access Memory (SRAM) – SRAM is a type of memory that is faster and more
reliable than the more common DRAM (dynamic RAM). The term static is derived from the fact
that it doesn't need to be refreshed like dynamic RAM. While DRAM supports access times of
about 60 nanoseconds, SRAM can give access times as low as 10 nanoseconds. In addition, its
cycle time is much shorter than that of DRAM because it does not need to pause between
accesses.
45nm (45 nanometer) – The next milestone in semiconductor technology. The 45nm process
enables chip circuitry with higher performance-per-watt than the most advanced processes in
production today. In the future, using the 45nm process will allow chips to be made with twice as
many transistors in a given area. Forty-five nm technology will enable more than 20 percent
improvement in transistor switching speed and more than a five-fold reduction in transistor
current leakage. Intel will use its 45nm process technology to manufacture the next generation of
its leading Intel® Core™ 2 Duo and Intel Xeon™ processor families.
-- 30 --
Intel, Itanium, Xeon and Core are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other
countries.
*Other names and brands may be claimed as the property of others.
