Lasers are the stuff of science fiction: big, heavy boxes that make blazing blasts
If you've ever seen an ordinary laser in a laboratory, you'll know
it's quite a hefty beast: typically about as long as your forearm,
fairly heavy, quite hot, and capable of producing a very intense beam of light.
But if lasers are that big, how come we can use them in small things like
portable CD players and handheld barcode scanners?
The answer is that we don't! These things use a very different kind of laser that's
about the same size as (and works in a similar way to) an ordinary LED
(light-emitting diode). Known as
semiconductor lasers (also called diode lasers or injection lasers),
they were developed in the early 1960s by
Robert N. Hall and, largely because they're so compact and inexpensive, are now the most
widespread lasers in the world. Let's take a closer look!
Artwork: Diode lasers are tiny, light, and compact—perfect for generating precision light beams inside small electronic appliances.
Chances are you've used a semiconductor laser in the last few days without even knowing
it. If you've watched a DVD, you've "looked through" one; if
you've been into a grocery store and had a barcoded product swiped through the
checkout you've bought with one; if you've made a long-distance telephone
call by fiber-optic cable you've "talked through" one; and if
you've printed something with a laser printer your printout has passed very near one.
Semiconductor lasers make powerful, precise beams of light (like
ordinary lasers), but they're about the same size as simple LEDs—the
little colored lamps you see on electronic instrument panels.
If you've read our article on diodes, you'll already have an idea how LEDs work.
Essentially, an LED is a semiconductor sandwich with the "bread"
made from slices of two different kinds of treated silicon known as
p-type (rich in "holes" or, in other words, slightly lacking electrons, the tiny negatively charged particles inside atoms) and n-type
(with slightly too many electrons). Put the two slices together and you make what's
called a p-n junction diode that has all kinds of
In an ordinary diode, the p-n junction works like a turnstile that allows
electric current to flow in only one direction (known as
forward-biased operation). As electrons flow across this
barrier, they combine with holes on the other side and give out
energy in the form of phonons (sound vibrations) that
disappear into the silicon crystal. In an LED, much the same process
takes place but the energy is given out not as phonons but as
photons—packets of visible light.
Photo: The smaller circle on the bottom left of this photo is a semiconductor laser diode in a CD player. The larger, blue-tinted circle on the top right is a lens that reads the reflected light bouncing down off the CD. Never attempt to look at the laser light in a CD player: you could easily blind yourself.
How diode lasers make light
In a laser diode, we take things a stage further to make the emerging light more pure
and powerful. Instead of using silicon as the
semiconductor, we use a different material, notably an alloy of
aluminum and gallium arsenide (indium gallium arsenide phosphide is
another popular choice). Electrons are injected into the diode, they
combine with holes, and some of their excess energy is converted into
photons, which interact with more incoming electrons, helping to
produce more photons—and so on in a kind of self-perpetuating
process called resonance. This repeated conversion of incoming
electrons into outgoing photons is analogous to the process of
stimulated emission that occurs in a conventional, gas-based
Artwork: The basic setup of a laser diode. Laser light is produced when electrons and photons interact in a p-n junction arranged in a similar way to a conventional junction diode or LED. One end of the diode is polished so the laser light can emerge from it. The other ends are left roughened to help confine the light.
In a conventional laser, a concentrated light beam is produced by "pumping" the
light emitted from atoms repeatedly between two mirrors. In a laser
diode, an equivalent process happens when the photons bounce back and
forth in the microscopic junction (roughly one micrometer wide)
between the slices of p-type and n-type semiconductor, which is technically known
as a Fabry-Perot resonant cavity (a kind of interferometer). The amplified laser light eventually emerges from the polished end of the gap in a beam parallel to the junction. From there, it goes on to read music from your CD, scan the price on your cornflakes, print out your college dissertation, or do a thousand other useful things!
Stacked laser diodes
Early laser diodes could fire out only a single, relatively puny beam, but ingenious electronics engineers soon found ways to make them considerably more powerful. Since the 1990s, one common approach has been to mount a number of laser diodes on top of one another (like an apartment building) and then focus their individual beams into a single output beam using a
collimator and/or lens. This kind of arrangement is
variously called a semiconductor laser stack, stacked laser diode, or just a diode stack. Apart from making more power
than a single laser diode, a stack opens up the possibility of generating multiple different wavelengths at the same time
(because each laser in the stack can make a different one). Instead of a single P-N junction, there are multiple ones, and the laser light beams emerge from the active layers in between them; typically, there's also at least one tunnel junction between the stacked layers. A single pair of terminals (sometimes called Ohmic contacts) feeds electrical power to the entire stack.
Artwork: A simple stacked laser diode, comprising two diode lasers one on top of the other,
and made from doped layers of aluminum gallium arsenide. The
terminals (Ohmic contacts) are shown in gray at the top and bottom, the substrate (base material) is green, P-type layers are shown in blue, and N-type layers in red. The tunnel junction is labeled J2. Artwork courtesy of US Patent and Trademark Office,
from US Patent #5,212,706: Laser diode assembly with tunnel junctions and providing multiple beams by Faquir C. Jain, University of Connecticut, May 18, 1993.
Who invented semiconductor laser diodes?
Artwork: Robert Hall's original laser diode patent, courtesy of US Patent and Trademark Office.
Who do we thank for this fantastic invention? General Electric's Dr Robert N. Hall, who filed his
patent for the idea ("Stimulated emission semiconductor devices") on October 24, 1962 (it was granted as US Patent #3,245,002 on April 5, 1966). Here's one of the drawings from that patent, showing the basic arrangement of the parts described up above. The numbering is Hall's original, but I've added the coloring and simplified descriptions to make it easier to follow:
A typical P-N semiconductor laser diode.
P-type region (blue).
N-type region (red).
P-N junction region (resonant cavity) where light is produced by stimulated emission. This isn't drawn to scale!
In Hall's original patent, it's described as being 0.1 micron (0.1 millionths of a meter, 0.1μm, or 1000 Angstroms) thick.
Solder fixing upper electrode to p-type region.
Solder fixing lower electrode to n-type region. (This covers the whole of the base of the n-type region, not just the gray outer outline shown here.)
Upper electrode connector.
Lower electrode connector.
Highly polished front surface.
Highly polished rear surface, which must be precisely parallel to the front surface to ensure standing waves of electromagnetic radiation (laser light) are produced and emitted efficiently in the resonant cavity between the p-type and n-type regions. Surfaces 11 and 12 may be covered with mirrors or a metallic coating to improve the resonant effect.
Side surface cut at an angle (or roughened) to prevent waves of light forming in other directions.
Other side surface cut at a similar angle or roughened in a similar way.
You can read much more detail in Robert Hall's patent, listed in the references below.
Ordinary lab lasers are big beasts, as we've already seen—not so different from the one Goldfinger famously used in the James Bond film of the same name. Putting it another way, anything remotely compact that needs a laser to power it is likely to use a diode laser rather than a "Goldfinger laser." CD players, barcode scanners, fiber-optic phone lines, dental tools, laser hair-removal devices, laser pointers, and laser printers all use diode lasers because they're small, compact, and inexpensive. It doesn't follow that they're low-powered and puny, however—for three reasons.
Diode lasers can be surprisingly powerful (hundreds of watts is a perfectly achievable output).
As we saw up above, you can stack diode lasers to make devices with far higher output (in the kilowatts).
You can team diode lasers up with solid-state lasers (to make what are called diode-pumped solid-state lasers) or with conventional lasers, using them as "optical pumps" (instead of traditional flash tubes) to excite things like ruby rods (giving output in the megawatts).
The recent development of
deep ultraviolet laser diodes points the way to smaller and cheaper biosensors
and numerous other exciting biological applications, including cheap food and water sterilization.
Diode Lasers Jump to the Deep Ultraviolet by Jeff Hecht. IEEE Spectrum. January 9, 2020. A much more ultraviolet semiconductor diode laser paves the way to new applications such as smaller and cheaper biosensors.
Please do NOT copy our articles onto blogs and other websites
Articles from this website are registered at the US Copyright Office. Copying or otherwise using registered works without permission, removing this or other copyright notices, and/or infringing related rights could make you liable to severe civil or criminal penalties.