Terahertz waves penetrate the world of imaging

Foiling terrorists, detecting cancer and designing new drugs are just
some of the potential applications for equipment that exploits the
unique properties of terahertz waves. Oliver Graydon looks behind the
scenes at one of the hottest sectors in photonics research today.

From Opto & Laser Europe October 2002.
Detecting weapons concealed underneath clothing, analysing the
contents of suspicious-looking envelopes, or even spotting the onset
of cancer: these are just some of the exciting prospects that have
been turning terahertz wave research into one of the most talked-
about topics in photonics.
Terahertz waves are electromagnetic waves that have a frequency of
between 100 GHz and 30 THz and lie between the infrared and microwave
parts of the spectrum. What makes these waves so fascinating to
scientists is their ability to penetrate materials that are usually
opaque to both visible and infrared radiation.
For example, terahertz waves can pass through fog, fabrics, plastic,
wood, ceramics and even a few centimetres of brick - although they
can be blocked by a metal object or a thin layer of water. The way in
which terahertz waves interact with living matter has potential for
highlighting the early signs of tooth decay and skin or breast
cancer, or understanding cell dynamics.
Growing demand
The list of potential customers for terahertz wave technology is
growing all the time. The military wants high terahertz and
millimetre-wave imagers that are able to "see through" bad weather;
chemists want spectroscopy equipment for analysing the structure of
new drugs; and airports arguably need better security-screening
equipment.
As a result, research on terahertz waves is fast evolving beyond
being a mere scientific curiosity. There is now a buzz of activity as
researchers around the world race to build the first practical
terahertz imaging and spectroscopy equipment.
"The field has exploded - it's incredible," said Paul Planken from
Delft University of Technology, who is an active researcher in the
field. "A number of years ago I tried to make a list of all the
papers published on terahertz generation and detection. I started
with 1969 and there was just one paper. Now I see so many papers
being published that I refuse to believe that 10 years from now there
will be no real-life applications for terahertz technology."
The Delft team has built a state-of-the-art system for generating and
detecting terahertz waves and has already applied it to various
imaging tasks, such as filming the diffusion of a gas through
polystyrene foam. The team is now investigating biomedical
applications and has recently demonstrated a method of dramatically
improving the resolution of a terahertz imaging system.
Powerful potential
Although the majority of the initial research on terahertz imaging
was carried out by Martin Nuss and colleagues at Bell Labs in the
early 1990s, today the leading research group in the field is
undoubtedly Xi Cheng Zhang's group at Rensselaer Polytechnic
Institute, US. For more than a decade the group has been pushing back
the frontiers of terahertz technology and has published a huge number
of papers on the topic. The Institute has recently established a
dedicated facility for terahertz studies, the Center for Terahertz
Research, and has received a $1m (€1.03m) donation from the W M Keck
Foundation.
At this year's CLEO conference in May in California, Zhang and co-
workers from the University of Adelaide, Australia, and the New York
State Department of Health, US, reported initial results on the use
of terahertz waves to screen potential biohazards.
The team found that terahertz waves could detect and classify unknown
powders that were sealed inside an envelope. Using a transmission
terahertz imaging technique operating at 0.3 THz, the team
successfully detected and distinguished between samples of flour,
salt, baking soda and bacterial spores placed inside a paper
envelope.
Outside academia, one of the first companies attempting to cash in on
the market potential and commercialize terahertz technology is
Teraview, a start-up based in Cambridge, UK. With a headcount of 13,
the firm was spun out of Toshiba Europe's Research Laboratory in
April 2001 to develop equipment for medical imaging, drug development
and security screening.
The start-up has not wasted any time. Having built prototype medical-
imaging equipment, the Cambridge-based firm has tested its terahertz
technique for skin-cancer detection in field trials at UK hospitals,
as well as using it to image semiconductor chips for electronics
companies.
"Although there are lots of universities carrying out research in the
field, we are the first company dedicated to commercializing
terahertz technology," claimed Don Arnone, Teraview's chief executive
officer. "We're starting to bring products to market and will have a
product launch shortly."
Teraview's prototype medical imager, the TPI Scan, resembles a
photocopier on wheels. It squeezes the laser, optics and electronics
needed for terahertz imaging into a self-contained 1 m long, 1 m high
and 60 cm wide trolley that weighs 150 kg. The TPI Scan can scan a
sample of up to 25 x 25 mm in size in less than 1 min, and features
an integral camera that simultaneously generates a visible image of
the sample. The resolution of the scanning is 200 µm in one axis and
20 µm in the other.
The equipment is likely to retail for around £250,000 (€395,000) and
will be used to help diagnose skin lesions and plan surgery. It
features an external CD writer, USB port and network connection for
data transfer.
As for customers, Teraview has already signed a distribution
agreement with Bruckner, which makes analysis equipment for the
pharmaceuticals industry. "We're looking to align ourselves with
established players, such as Bruckner, in each of our key markets,"
said Arnone. "We'll produce a terahertz engine [a module for emitting
and detecting terahertz waves] and then supply it to commercial
collaborators for them to market."
Teraview is also working closely with the semiconductor group at
Cambridge University to develop a semiconductor laser that operates
in the terahertz region. To date, terahertz waves are usually
generated by illuminating a piece of semiconductor, such as gallium
arsenide, with femtosecond pulses from a solid-state laser such as a
Ti:sapphire laser. Although this approach works well and femtosecond
lasers are getting smaller and cheaper all the time, it is still a
relatively bulky and expensive solution. Ultimately, the preferred
source for many commercial applications would be a compact
semiconductor laser.
Quantum leap
The development of such emitters has recently taken a leap forward,
thanks to the invention at Bell Labs, US, of the quantum-cascade
laser, which emits in the mid-infrared at around 4 µm. Semiconductor
scientists are now adapting the technology to design lasers that are
operational in the far-infrared and terahertz regions.
Earlier this year, researchers from Teraview, the University of
Cambridge, and the National Institute for the Physics of Matter
(INFM) in Italy made a series of quantum cascade lasers that operate
in pulsed and continuous-wave mode at 4.4 THz (wavelength 68 µm).
Although the lasers can only currently work at low temperatures of up
to 50 K, they emit up to 2 mW of singlemode terahertz radiation. The
challenge now is to raise the operational temperature, which,
although it may take several years, is definitely feasible.
"I think the development of a quantum cascade laser is a major
milestone, certainly - if they are able to make one that works at
room temperature," said Planken. "Up until recently most people
generated terahertz radiation by illuminating GaAs with femtosecond
laser pulses or by optical rectification with telluride electro-optic
crystals. Now they have a third candidate."
Whether it will be its applications in medicine, security or
pharmaceuticals that take off first is hard to predict, but once
terahertz technology has a toehold in one market it is likely to
quickly spread to others.
"Some people say that terahertz radiation will replace X-rays, but
that's not true," said Planken. "It will not happen because the human
body is made up mostly of water, so you do not get much penetration.
Water is a very good terahertz application killer."


T-rays explained
The powerful nature of terahertz analysis stems from the fact that it
is a coherent technique that can make both amplitude and phase
measurements. Unlike common optical spectroscopic techniques that
only measure the intensity of light at specific frequencies,
terahertz experiments often measure the temporal electric field of
terahertz pulses that have interacted with (i.e. reflected off or
passed through) a sample.
A Fourier transformation of this time-domain data discloses the
amplitude and phase of the pulse and reveals a wealth of information
about the sample. For example, it allows precise measurements of the
refractive index and absorption coefficient of a sample. Molecules
also have unique rotation and vibration resonance lines in the
terahertz spectrum that can be used as terahertz fingerprints.
The most popular way to generate terahertz waves is to illuminate a
carefully engineered semiconductor crystal, such as GaAs, with
femtosecond pulses of visible light. This bombardment creates
ultrashort pulses of terahertz radiation (typically of the order 100
fs) that can be used for imaging and spectroscopy.
As the pulses reflect from different depths within an object,
an "image slice" at a desired depth can be built up by carefully
controlling the timing of the pulse detection. A 3D image can be
constructed by putting together a number of these sliced images.
Coherent detection of the pulses is achieved by illuminating a second
crystal with both the terahertz pulses and a portion of the visible
femtosecond pulses that is split off from the original beam and has
undergone a suitable time delay.


Breakthrough in resolution
Paul Planken and Nick van der Valk at Delft University of Technology
in the Netherlands have improved the resolution of a terahertz
imaging system by a factor of almost 1000. The development is
significant because it could open the door to terahertz studies of
tiny structures, such as living cells.
Imaging systems conventionally have a diffraction-limited resolution
(spot size) that is equal to about half of the optical wavelength. In
the case of terahertz waves, which have a wavelength of about 1 mm,
this corresponds to a resolution of about 0.5 mm - far too big for
cell imaging. However, the Delft team has now demonstrated that it
can create and measure terahertz spot sizes as small as 8 µm, and
that 1 µm may ultimately be possible.
The Dutch researchers achieve such small focal spots by using a sharp
metal tip to locally bend and concentrate the electric field of the
terahertz beam near the surface of a GaP crystal. The mechanism is
analogous to how a lightning rod acts to bend electrical field lines.
In the future the method could be used with a terahertz pump to probe
experiments by placing thin samples on the crystal underneath the
tip. The sample is illuminated by the pump beam from above and a
tightly focused probe beam from below. The crystal is then raster
scanned under the tip to generate a 2D terahertz image.