Carbon nanotubes may point the way toward “smart” brain materials


by Roger Allan, Contributing Editor

Think small, think smart–that is the essence of research into carbon nanotubes (CNTs) currently underway in Italy and Switzerland that has tremendous implications in the search to find ways to “bypass” faulty brain wiring. Their research has shown that CNTs improve nerve responsiveness, potentially making them good candidates for the design of “smart materials” in biomedical applications such as brain repair.

Carbon exists in many forms, the best known of which are diamond and graphite. CNTs have recently received a lot of attention because they feature extraordinarily strong carbon molecules with unique electrical properties. They have proven to be versatile and valuable materials for a wide range of functions and applications. In the bio-medical field CNTs are being used in drug delivery, stem cell research, tissue engineering, and a large number of neurological studies. CNTs also have been used in the engineering of mechanical memory, nanoscale electric motors, a hydrogen sensor, touch screens, and flexible displays. A radio receiver consisting of a single CNT was developed in 2007, and in 2008 a sheet of CNTs was used to operate a loudspeaker. Research into their use in energy storage also has shown interesting results.

The new work by the Italian-Swiss team was funded by the European Union (EU) under a 36-month, ~€1.8 million project called Specific Targeted Research Project (STREP) that began August 2006, financed in part by the EU’s Sixth Framework Programme (EP6) as part of its NEURONANO program targeting development of neuro-implantable devices integrating neurons with carbon nanotubes.

Lead researchers in the work are Henry Markram, head of the Laboratory Microcircuitry at the école Polytechnique Federale de Lausanne (EPFL) in Lausanne, Switzerland; Michele Giugliano, also at EPFL (and now an assistant professor at the University of Antwerp in Belgium); and Laura Ballerini, a professor of physiology, and Maurizio Prato, a professor of organic chemistry, both at the University of Trieste in Italy. Other researchers also were involved at both locations as well as at Italy’s University of Milan, The International School for Advanced Studies and the ELETTRA multidisciplinary laboratory in Trieste, and Ireland’s Trinity College.

Prato was the first to design a chemical functionalization procedure, allowing a “friendly” manipulation of these electrically conductive nanoparticles. The chemical reaction, known as Prato’s reaction, was the first step toward the creation of planar substrates for the growth of neural networks ex vivo (external to the human body) as an in vitro (internal to the human body) model of a rudimentary brain-machine interface.

This same team two years ago showed that the collective activity emerging from the interactions between neurons in a network also appeared modulated. This shed some light on the cellular basis of such network-level interactions, holding the promise for researchers to engineer or reshape the function and excitability of individual neurons and networks using smart materials for neuroprosthetics.

In this latest effort, the researchers looked at the relationship between the electrical properties of CNTs and the way they interfere with single neurons. They measured the electrical activity of single nerve cells, used scanning tunneling electron microscopy (STM) for analysis, and applied theoretical modeling to see how nanotubes actually improve neuron responsiveness.

“Our findings show that CNTs, which like the nervous cells of our brain, are excellent electrical signal conductors and form intimate mechanical contacts with cellular membranes, thereby establishing a functional link to neuronal structures,” according to Bellerini and Prato. “Such a functional and mechanical link might favor electrical shortcuts between the proximal and distal compartments of the neuron, therefore improving neuronal performance.”

Figure 1: A scanning-electronic microscopic view of multi-walled carbon nanotubes about 150 nm in diameter. These are part of a batch being prepared for drug delivery. (Source: Annie Cavanagh, Wellcome Images.)
Click here to enlarge image


Electrochemical wave

In a brain’s synaptic transmission system, an electrochemical wave known as an “action potential” travels along the axon of a neuron. A neuron works by summing together inputs from its network of dendrites connected to other neurons. When the wave reaches a synapse, it provokes release of a puff of neurotransmitter molecules, which bind to chemical receptor molecules located in the membrane of another neuron on the opposite side of the synapse (Figure 1).

Axon potentials might occasionally backpropagate electrically to dendrites–that is, move against the direction of flow; the interaction with nanotubes favors this backpropagation. The resulting backpropagating current induces a voltage change that increases the concentration of an ionic solution of calcium (Ca++) in the dendrites, an event known as “calcium electrogenesis.” This can be measured through the presence of a slow membrane depolarization following repetitive action potentials.

For their experiment, the team deposited single-wall or multi-wall CNTs on a glass substrate and subsequently defunctionalized them by thermal treatment to form glass slides covered with a purified and mechanically stable thin film of about 50-70nm thickness. STM conductivity measurements revealed that both nanotube layers act as a largely resistive network. This dense network permits long-range electrical connectivity.

A standard stimulation protocol was used to probe the regenerative and excitable properties in the proximal and distal compartments of the neuron. By injecting a brief current pulse of 1nA and 4ms into the soma (a basic cell body that contains a nucleus and dendrites), the researchers observed that the neuron was forced to fire a regular train of six action potentials at frequencies ranging from 20-100Hz (Figure 2).

These results are exciting, according to the researchers, because they represent significant progress in addressing what Markram calls “the three fundamental obstacles to developing reliable neuroprosthetics:” creating an interface between nerve tissue and a device, understanding how best to stimulate the neural tissue, and figuring out which nerve signals should be recorded by the device so that it can make an automatic and appropriate decision.

Figure 2: The effects of carbon nanotubes (CNTs) on neural excitability can be seen. (a) CNTs induced significantly larger after-potential depolarization (ADP), following a single spike train. Current pulses at 80Hz (upper trace) evoke precisely timed action potentials in the control and CNT samples. Average voltage trajectories are superimposed for comparison. Areas below the voltage traces (grey shading) are quantified over 100ms, 50ms after the last action potential. (b) Area distributions are shown for 48 control cells and 66 CNT cells.
Click here to enlarge image

“The new CNT-based interface technology discovered, together with state-of-the-art simulations of brain-machine interfaces, is the key to developing all types of neuroprosthetics,” Markram added. “This includes sight, sound, smell, motion, vetoing epileptic attacks, and spinal bypasses, as well as repairing and even enhancing cognitive functions.”

The precise mechanisms for the observed effect of CNT substrates in this study are not yet totally understood. The results obtained so far using transmission-electron microscopy (TEM) suggest that one mechanism might rely on the detected discontinuities and tight contact between nanotubes and membranes. It is believed that the morphology of such contacts is indicative of the development of hybrid nanotubes or neuronal units. What is not clear is how these units are functionally different from membrane areas, if at all.

The results of this work were published Dec. 21 2008 in the advance online edition of the journal Nature Nanotechnology. For more information about this project, go to

Roger Allan is a 40-year electronics journalism veteran, with 15 of those years as Electronic Design’s executive editor. His specialties include MEMS and nanoelectronics technologies. Email: