LET there be light

Transistors and light-emitting diodes (LEDs) have become two of the most well-known and useful devices, essential to everything from mp3 players to the future of low-wattage lighting. Now, a hybrid device – a light-emitting transistor (LET) – has been invented by Nick Holonyak Jr and Milton Feng of the University of Illinois at Urbana-Champaign. This device could become the fundamental element in optoelectronics devices for faster telecommunications and many other applications.

We have demonstrated light emission from the base layer of a heterojunction bipolar transistor, and showed that the light intensity can be controlled by varying the base current, explains Holonyak, who is the John Bardeen Professor of Electrical and Computer Engineering and Physics at Illinois. Holonyak was the inventor of the first practical light-emitting diode (LED) and the first semiconductor laser to operate in the visible spectrum.

The transistor

The transistor

This work is still in the early stage, so it is not yet possible to say what all the applications will be, he adds. But a light-emitting transistor opens up a rich domain of integrated circuitry and high-speed signal processing that involves both electrical signals and optical signals.

Transistors usually have two ports: one is for an input, the other for an output. The light-emitting transistor, on the other hand, has three ports: an input, an electrical output, and an optical output. This means that we can interconnect optical and electrical signals for display or communication purposes, notes Feng, who is ironically Holonyak Professor of Electrical and Computer Engineering at Illinois and creator of the world’s fastest bipolar transistor, a device that operates at a frequency of 509 gigahertz.

Nick Holonyak Jr

Nick Holonyak Jr

Graduate student Walid Hafez fabricated the light-emitting transistor in the university’s Micro and Nanotechnology Laboratory. Where traditional transistors utilise crystals of silicon and germanium, Hafez instead used indium gallium phosphide and gallium arsenide to build the light-emitting transistor. In a bipolar device, there are two kinds of injected carriers: negatively charged electrons and positively charged holes, Holonyak explains. Some of these carriers will recombine rapidly, supported by a base current that is essential for the normal transistor function.

Milton Feng

Milton Feng

By using these alternative semiconductor materials, the researchers were able to control the recombination process in indium gallium phosphide and gallium arsenide materials so that infrared photons are produced. In the past, this base current has been regarded as a waste current that generates unwanted heat, Holonyak adds. We’ve shown that for a certain type of transistor, the base current creates light that can be modulated at transistor speed.

Walid Hafez

Walid Hafez

This recombination process is almost the same as that exploited in LEDs to produce visible, rather than infrared, light; however, there is an important difference. In a conventional LED, the photons are produced much more rapidly. The Illinois transistor produces infrared in phase with the transistor’s base current, so it can be modulated – or switched on and off – at an operating frequency of 1 megahertz and perhaps much faster, which is a switching speed impossible to attain with an LED but perfectly suited to high-speed optical circuitry of the kind used to transmit data across fibre optic networks as well as in other applications.

At such speeds, optical interconnects could replace electrical wiring between electronic components on a circuit board, Feng explains. This work could be the beginning of an era in which photons are directed around a chip in much the same way as electrons are manoeuvred on conventional silicon chips, but at the speed of light! In retrospect, we could say the groundwork for this was laid more than 56 years ago with John Bardeen and Walter Brattain and their first germanium transistor, adds Holonyak, who was Bardeen’s first graduate student. But the direct recombination involving a photon is weak in germanium materials, and John and Walter just wouldn’t have seen the light – even if they had looked. If John were alive and we showed him this device, he would have to have a big grin.

Further reading

Appl. Phys. Lett., 2004, 84(1), 151-153
http://dx.doi.org/10.1063/1.1637950

Nick Holonyak Jr
http://www.ece.illinois.edu/people/profile.asp?nholonya

Milton Feng
http://www.ece.illinois.edu/people/profile.asp?mfeng

Suggested searches

transistors
light emitting diodes

Naked pores

German researchers have developed a new type of molecular sieve based on tiny capsules featuring the metal molybdenum. They say that their new materials can be tuned to trap different molecules depending on their size and shape and so could be useful as nanoscale filters, catalysts, or sensor materials.

Natural and synthetic porous minerals have been used by chemists for years as molecules sieves and filtration materials, as catalysts and as the molecular recognition units for sensors. The ability of these compounds, which includes the zeolites, with their pores and channels to trap molecules of specific shape and size has been well researched. However, Achim Mueller and colleagues at Bielefeld University in Germany believe discrete structures, that carry none of the mineral scaffolding of the zeolites and their chemical cousins could be much more useful. They have now built a well-defined, discrete, or molecular, compound that is essentially a naked pore.

(Credit: Wiley)

(Credit: Wiley)

These naked pores are based on a spherical skeleton composed of several molybdenum atoms and interconnecting chemical bridges. The molecule has entry points for smaller chemical species, such as cations. By fine-tuning the chemical groups around the sphere and in particular the entry points, the team can make their naked pore have affinity for specific chemicals, which can enter and become trapped inside. The skeleton – with chemical structure (pent)12(linker)30.{(Mo)Mo5O21(H2O)6}12{Mo2O4(ligand)}30 – can be crystallised to make solids that have what the team describes as sizeable pores and channels with finely sculpturable interiors. Most importantly, the chemical functionality of the channels and the size and charge of the individual capsules can be extensively varied.

The researchers have now shown that they can position different chemicals at well-defined positions above, below, and inside the channels. This situation allows us to study, in principle, new types of molecular transport phenomena, including osmotic-type ones, on the nanoscale, and shows properties of a nano-ion chromatograph, says Mueller.

Such a molecular device could be used to separate different chemicals in much the same way as a laboratory-scale chromatograph can but on a much smaller scale and with minute quantities of materials. Small quantities of cations can be separated according to their size and charge, Mueller told Spotlight, larger species cannot enter, but get trapped in the pores.

Our capsules can be considered as nanoscale laboratories, enthuses Mueller, with a variety of functionalities that enable the positioning of substrates and cations under controlled conditions. He adds that cation trapping in the pores and channels is possible in steps so that each cation trapped subsequently is affected by the decreased negative charge on the system induced by its predecessors. Such control is, Mueller says, one of the hallmarks of nanotechnology. Further on, there is the possibility of obtaining information about electrolytes under confined conditions, which is very important in biological systems, he adds.

Further reading

Angew. Chem. Int. Ed. 2003, 42, 5039-5044
http://dx.doi.org/10.1002/anie.200352358

DOI: 10.1002/anie.200352358

Suggested searches

Molecular sieves
Molybdenum elements
Nanotechnology

DNA and chips

The secret ingredient in a future biological computer is to add a little DNA. But, making hybrid devices from a silicon chip and a strand of genetic material means mixing hard-wired microelectronics technology with the softer world of molecular biology.

Now, chemists at the University of Newcastle upon Tyne have come up with a solution that could lead to new lab-on-a-chip devices and biological sensors for use in medicine and environmental analysis. It might even one day allow biology to compute or provide an interface between electronic devices and living things.

Ben Horrocks

Ben Horrocks

Newcastle chemists Benjamin Horrocks and Andrew Houlton and their colleagues have devised a way to automate the solid-phase synthesis of DNA on a semiconductor chip. They believe their method could readily be adapted to the conventional fabrication techniques of photolithography used in the microelectronics industry to pattern the microscopic transistors and circuitry on a computer chip.

The team recently reported how it has found a way to attach a DNA sequence of just seventeen nucleotides to a silicon surface modified with organic molecules. The key to unlocking hybrid DNA chips lies in the team’s use of bifunctional organic molecules. At one end the molecule has the right chemistry to allow it to be attached to an oxide-free silicon surface. The other end of the molecule has a functional chemical group on which a DNA strand can be grown using an automated DNA synthesizer of the kind found in biotech laboratories the world over.

Andrew Houlton

Andrew Houlton

The team is working with two aims in mind – first, the development of chemical sensors and secondly the synthesis of DNA on silicon surfaces for nanoscale molecular architecture. This addresses the projected reduction in the size of electronic components which by 2015 are predicted to be of the order of nanometres (i.e. built from molecules), explains Houlton.

Gel electrophoresis reveals DNA is attached to the silicon

Gel electrophoresis reveals DNA is attached to the silicon

Previous endeavours in this area have generally used glass in preference to silicon wafers and those that have focused on silicon have applied organic molecules to an oxidised surface rather than the naked silicon chip. The Newcastle team has now confirmed that it is possible to cover the surface of a silicon chip with DNA strands. Moreover, the 17-base DNA strands can be coupled with the complementary DNA strand making the familiar DNA double helix. From the nanodevice perspective the importance of the team’s work lies in their ability to pattern the surface of the silicon rather than simply randomly deposit DNA strands. Patterning using the printing and etching techniques of microelectronics fabrication means they can tightly control the arrangement of the DNA on the surface and so produce what might one day become molecular circuitry.

A DNA-patterned silicon surface

A DNA-patterned silicon surface

Sequential modification of a silicon surface with DNA

Sequential modification of a silicon surface with DNA

Further reading

Angew. Chem. Int. Ed., 41, 615 (2002)
http://www3.interscience.wiley.com/cgi-bin/abstract/90512278/ABSTRACT

DOI: 10.1002/1521-3773(20020215)41:4<615::AID-ANIE615>3.0.CO;2-Y

Benjamin Horrocks
http://www.ncl.ac.uk/chemistry/staff/profile/b.r.horrocks

Andrew Houlton
http://www.ncl.ac.uk/chemistry/staff/profile/andrew.houlton

Suggested searches

Molecular Electronics
Nanotechnology