杏吧原创

Winnipeg, Canada

MICHAEL RUBNER鈥橲 sandwiches are tiny. For American sandwiches this is unusual
indeed, but there is something else odd about them. When Rubner flicks a switch
to send electricity coursing through them, they glow orange or bluey-green
depending on the filling. You wouldn鈥檛 want to eat them though鈥攖hey鈥檙e
made from layers of plastic.

At the engineering department of the Massachusetts Institute of Technology,
Boston, Rubner and other researchers are using an incredibly quick and simple
technique to stick ultra-thin slices of alternating positively and negatively
charged polymers together. They are building complex, multilayered materials
that could one day be used to make anything from portable computer screens to
artificial organs.

The technique is dirt-cheap, environmentally friendly and amazingly flexible,
allowing the researchers to dream up a whole range of novel devices. The choice
of fillings is not limited to polymers鈥攁nything with a positive or
negative charge can be added. Dyes can be used to create a surface that changes
colour according to pH and proteins can be incorporated to make
biologically active devices. And by controlling where the positive or negative
polymers are deposited, researchers are beginning to build intricate devices
such as electronic circuits and diffraction gratings.

Rubner and his colleagues at MIT made their first glowing sandwiches last
year. Starting with an indium-tin oxide base, they lay down alternate layers of
a negatively charged polymer, or polyanion, and a positively charged polymer, or
polycation, called poly(phenylene vinylene) (PPV) which glows brightly when
electricity is passed through it. After building between ten and fifty layers of
each polymer, making a sandwich just a few nanometres thick, the team tops it
off with an aluminium layer.

These layered materials can be used as tiny light-emitting diodes (LEDs).
Passing an electrical current between the indium-tin oxide and aluminium layers
makes the PPV glow
(see Diagram).
Rubner says it is remarkable that they
can make such minute LEDs so easily and with so few defects. If you tried this
using any other technique, he believes the layers would almost certainly contain
pinholes, which might fill up with aluminium and short-circuit the device.

Different colours can be produced by changing the polyanion. Adding
poly(methacrylic acid) layers creates an LED that glows blue-green, while
poly(styrene sulphonate) makes an orange light. Rubner is not certain why this
happens, but he thinks the polyanion layers may interact with the PPV layers to
change the wavelength of light they emit.

Bright future

The LEDs could be used to make full-colour displays that are thinner and
cheaper than is possible with liquid crystals. Jim Sheats of Hewlett Packard
in Palo Alto, California, sees a bright future for the new technology.
鈥淎pplications will likely start at the 1000 to 50 000-pixel range,鈥� he says,
鈥渋ncluding displays of small portable appliances such as cellphones.鈥�

Research into the remarkable sandwich-making technique began in 1990. Gero
Decher, a chemist now at the Louis Pasteur University in Strasbourg, France, was
using electrostatic attraction to stick molecules onto solid surfaces. At first,
his aim was to find a simpler way of creating thin coatings than the
Langmuir-Blodgett technique, in which a film of molecules forms on the surface
of water and is then transferred to a solid surface that is dipped into the
water. The trouble with this process is that special equipment is needed, and it
only works with a limited range of molecules.

Decher soon found that he could form thin films from virtually any charged
polymer and build up as many layers as he wanted. The technique is
straightforward. He treats the surface of a material, such as glass, with a
chemical containing a positively charged group. The molecules react with the
glass and the surface becomes covered with positive charges. He then dips the
glass into a solution of a polyanion such as poly(styrene sulfonate). PSS
has hundreds of negative charges along its chain, and so it sticks to the
positive glass surface and forms the first layer of the sandwich. It takes just
10 minutes to create a complete layer.

Now the beauty of the technique becomes clear. Once one layer of polyanion
molecules is in place, the surface bristles with negative charges and other
polyanions are repelled. So only a very thin film is formed. After a rinse with
water to remove any loose molecules, Decher dips the glass into a solution of a
polycation, such as poly(allylamine hydrochloride). A layer of this polymer now
sticks on top of the layer of polyanion. Next, after a quick rinse, it鈥檚 back to
the solution of polyanion鈥攁nd so on until the sandwich is as thick as
Decher wants it.

鈥淭he great advantage of this technique is that we can create very uniform
thin films in a simple manner,鈥� says Rubner, who started making his sandwiches
in 1993. The only equipment needed is three beakers and a robotic arm to dip the
glass. What鈥檚 more, industry likes it because it is all water-based鈥攏o
expensive or hazardous solvents are involved. Despite the simple processing, the
sandwiches are surprisingly durable. Decher has heated and artificially aged his
layers and seen no permanent changes in structure.

Such simplicity makes the technology very flexible and easy to scale up.
Decher has built sandwiches on bases of silica, glass, quartz, diamond,
polystyrene and metal up to a few centimetres thick. 鈥淲e could do whole window
panes, but we haven鈥檛 tried yet,鈥� he says.

Many different charged polymers can be used, too. 鈥淎ny water-soluble
polyelectrolyte will do,鈥� says Rubner. The key to producing the desired layer of
polymer lies in the solution in which the polymer is dissolved. 鈥淭he trick is
understanding the critical solution parameters such as pH and ionic
concentration,鈥� he says. 鈥淭hen you can control with extreme precision the
properties of the layers.鈥�

Decher can tune the thickness of a polymer layer to an accuracy of 0.1
nanometres simply by adding a particular concentration of salt to the polymer
solution. This works by increasing or decreasing the repulsion between charges
on the same polymer molecule. In a solution of polyanion, the negative charges
on a polymer repel each other and force the polymer into a flat, chain-like
shape. These polymers form thin layers. But add lots of salt and the negative
charges are hidden from each other by swarms of positive sodium ions. The
charges no longer repel each other and the polymers coil up, creating a thicker
layer
(see Diagram).

The layers formed by the polymers are not completely smooth. 鈥淭hese are not
ordered monolayers of homogeneous thickness,鈥� says Decher. 鈥淭hey have surface
roughness and loops.鈥� By contrast, Langmuir-Blodgett films tend to be highly
ordered鈥攚ith the result that where the order breaks up, defects occur.
鈥淭hese polymers are amorphous, like spaghetti,鈥� says Rubner. 鈥淭hey cover over
defects.鈥� Since defects in ordinary glass trigger water condensation, a
defect-free water-repellent layer of polymer could act as a new kind of
anti-condensation coating for windows and vehicle windscreens.

Rubner is also making sandwiches that conduct electricity, by adding layers
of a conductive polymer. This could be just the thing for making networks of
see-through wires and electrodes to connect arrays of light-emitting diodes.
Electrodes normally block or absorb some of the light from LEDs, but because
Rubner鈥檚 polymer layers are so thin, they will be virtually transparent and so
the LED will appear brighter.

Dropping charges

Rubner and his colleagues made some of these conductive sandwiches last year
by depositing layers of polyaniline (PA) and PSS on glass slides. They then
dipped the slides in strong acid to 鈥渄ope鈥� the PA鈥攄oping creates sites
along the polymer chain between which electrons can travel thereby making it
conductive. The resulting layer has the same conductivity but is much thinner
and contains fewer defects than layers made by the conventional technology of
spin-casting. In this process, molten polymer is poured into a mould, which is
then spun very fast. The centrifugal force causes the polymer to form a thin
layer.

In making PA layers, Rubner showed that a little creative thinking can vastly
expand the range of fillings that can be used in the sandwiches. He proved that
materials don鈥檛 have to contain lots of charges. PA is not a true polycation and
would not naturally be attracted to a negative surface. To get round this,
Rubner partially dopes the polymer with acid, which creates enough positive
sites to stick it to a negative surface.

The sandwich-makers are also using more than just polymers. Earlier this
year, Rubner鈥檚 team spread thin layers of negatively charged dyes, such as Congo
red, between layers of the polycation poly (allylamine hydrochloride). These
materials could be used as cheap optical filters to cut out specific wavelengths
of light, or as electrochromic or electroluminescent devices. Rubner also plans
to coat glassware with layers of pH-sensitive dyes, which would change
colour according to the acidity of whatever was poured into them. For example,
the water in a swimming pool could be tested simply by scooping some up in a
test tube.

Over in the chemistry department at Pennsylvania State University, Thomas
Mallouk is even more ambitious. He hopes to make sandwiches that mimic
photosynthesis. To do this he is taking a different approach. Rather than
building up polymer layers, he adds ready-made layers of solids such as mica.
Mallouk dissolves mica layers just 1 nanometre thick in water. 鈥淭hey are like
sheets of paper floating around in solution,鈥� he says.

The mica sheets replace the layers of polyanion in the sandwich-making
process. Meanwhile, Mallouk鈥檚 polycations contain groups that use energy from
light to transfer electrons and build up a potential difference, mimicking
photosynthesis. The mica sheets are important because they line up all the
light-harvesting groups in one layer so that they work in unison. The sheets
also prevent layers of polycation interfering with each other.

鈥淟iving鈥� devices can be made by interspersing synthetic layers with layers of
biological molecules. 鈥淢ost biomolecules are already dispersed in water and
charged,鈥� says Rubner. Yuri Lvov and Jim Rusling of the University of
Connecticut at Storrs and Toyoki Kunitake at Kyushu University in Fukuoka,
Japan, have been building layers of proteins for the past five years. 鈥淭hese are
ordered functional proteins,鈥� says Lvov. 鈥淭his has never been done before except
in living organisms.鈥�

Using layers of the muscle protein myoglobin, Lvov has built devices that
breakdown pollutants such as organochlorine compounds. Myoglobin contains iron,
which, in Lvov鈥檚 device, reacts with the pollutants and splits them into pieces.
Lvov has also made a filter containing the enzymes glucose oxidase and glucose
amylase which can 鈥渄igest鈥� starch. The enzymes break down starch molecules into
simple sugars.

Nonstick sandwiches

In the future, Lvov can even see materials containing layers of proteins
being used as artificial organs. 鈥淭his is just fantasy now,鈥� he says, 鈥渂ut we
will be talking to the biologists.鈥� Sandwiches for medical devices would need to
be made from inert polymers such as the fluoropolymers. These materials, which
include Teflon, are notoriously non-sticky and so are reluctant to coat other
polymers and form layers. However, by using a radio frequency plasma to generate
negatively charged sites over their surface, Rubner has managed to add layers of
polycations to fluoropolymers.

By masking certain areas on the fluoropolymer surface to stop the plasma
creating negative sites here, it should be possible to add layers of polymers in
specific patterns. The first layer of polycation would attach only to negatively
charged areas that had been exposed to the plasma. Subsequent layers would go on
top to build up the pattern. If a conductive polymer were used, the result would
be a pattern of conducting material surrounded by insulating
fluoropolymer鈥攖he makings of a simple circuit board.

Paula Hammond, also at MIT, is developing an even more precise way to make
patterned polymer layers in a project funded by the Office of Naval Research.
鈥淲e鈥檙e taking the layer-by-layer assembly process and adding a new dimension,鈥�
she says. Hammond creates her patterns using a technique called micro-contact
printing, developed by George Whitesides at Harvard University in 1993. First,
she etches a pattern with features as small as one micrometre across into a
mould using standard photolithography. Next, a 鈥渟tamp鈥� made from polysilicone is
cast from the mould.

Rubber stamped

Just as the rubber stamps used in banks are used to print patterns in ink
onto paper, the polysilicone stamp can be used to print its pattern onto a gold
wafer. For ink, Hammond uses chemicals called alkane thiols which bind to the
gold surface. She prints the pattern of the mould using an alkane thiol that is
chemically attractive to polycation or polyanion molecules. The rest of the gold
is then covered with an alkane thiol that doesn鈥檛 attract polymers.

When the gold wafer is dipped in a beaker of polycation or polyanion, a layer
only forms over the pattern. Hammond then adds more layers to build the pattern
up. 鈥淭his leads to some interesting possibilities for devices,鈥� says Hammond,
鈥渁nd you don鈥檛 have to use expensive processes to produce the patterns.鈥� One
possible application is in making patterned electrodes for miniature diodes and
circuits, built from conductive polymer. And by tailoring the refractive index
of the polymer sandwiches, Hammond hopes to make waveguides for transporting
light around optical systems.

Such complex devices are still a few years off. For now, the researchers are
busy perfecting their sandwich-making technique and trying more fillings to see
what works. But their work has already attracted a fair amount of commercial
interest. To anyone wanting a cheap, clean and simple way to make tiny
components and devices, the growing menu looks very appetising.