Spin physics

DNA spintronics sees the light

A thin layer of double-stranded DNA on a gold surface can act as a spin filter.

Using the spin rather than the charge of electrons to carry and process information is what the field of spintronics is all about1. Spintronics offers a number of potential advantages compared with conventional electronics, such as low power consumption, but it is more difficult to control and manipulate spins than charges. In general spins are manipulated in one of two ways: with magnetic fields that couple directly with the intrinsic magnetic moments of the electrons (that is, with their spins); or through the spin–orbit interaction, which is stronger for heavier atoms. As a consequence, one would not expect a material that is neither magnetic nor made of heavy atoms to be relevant to spintronics. However, writing in Science, Ron Naaman and co-workers report that a layer of double-stranded DNA deposited on a gold surface acts as a filter for electron spins2.

Naaman and co-workers — who are based at the Weizmann Institute in Israel and the Westfälische Wilhelms University in Münster, Germany — shine ultraviolet light through the DNA layer towards the gold surface; the photons have enough energy to kick electrons out of the surface but the light intensity is not high enough to damage the DNA (Fig. 1). These electrons travel back through the DNA and are collected at a detector that can distinguish whether the spin points in the same direction as the velocity (let us call these spin-up electrons), or in the opposite direction (spin-down electrons). The surprising result is that they detect a lot more spin-down than spin-up electrons. Depending on the thickness of the DNA layer, Naaman and colleagues show that it is possible to achieve up to about 60% spin polarization2 (defined as the difference in the number of spin-up and spin-down electrons divided by their sum).

Figure 1: Spin selectivity with DNA.

When photons with enough energy strike a gold surface they eject electrons that move upwards away from the surface. If the incident light has a clockwise circular polarization (so that the electric field E rotates clockwise with respect to the wave vector k), the electron spins (red arrows) tend to point opposite to the direction of motion (spin down), whereas light with anticlockwise circular polarization produces more spin-up electrons. However, when the gold surface is covered with a monolayer of double-stranded DNA (dsDNA), most of the ejected electrons are spin down, irrespective of the polarization of the incident light.

The idea of using DNA in the field of spintronics is not new3. For example, there have been proposals to build spin valves by sandwiching a layer of DNA, which is an insulator, between two ferromagnetic layers. However, the structural properties of DNA — most importantly, the fact that it is a chiral molecule: in other words, it can be left-or right-handed — have not been used until now. Although the exact mechanisms at work in the experiments by the Weizmann–Münster collaboration remain unclear, chirality seems to have a major role.

The more common form of DNA found in cells has a double helix structure that is right-handed. Naaman and colleagues have demonstrated that even without DNA, circularly polarized light incident on the gold surface would eject spin-polarized electrons2. In particular, clockwise circularly polarized light — light whose electric field rotates clockwise during propagation — ejects more spin-down electrons. Anticlockwise circularly polarized light, however, ejects more spin-up electrons. Linearly polarized light — for which the electric field always points in the same direction — does not, on average, select any particular spin polarization.

The deposition of a monolayer of double-stranded DNA on gold changes the situation dramatically. Irrespective of the polarization of the light — clockwise, anticlockwise, or linear — most of the ejected electrons are spin down2. Moreover, by increasing the thickness of the DNA monolayer, the spin-polarization effect is enhanced2. Furthermore, in control experiments, preferential ejection of spin-up electrons (with a smaller spin polarization 9%) is observed in systems with single-stranded or ultraviolet-damaged double-stranded DNA for any light polarization2. This raises the fundamental question: what causes this remarkable effect? So far we do not have a definitive answer.

DNA is made of very light atoms such as hydrogen, carbon, nitrogen and phosphorous, so — as mentioned above — the spin–orbit interaction is unlikely to be the mechanism responsible for these results. Support for this argument comes from recent theoretical calculations4 that fail to predict the strong level of spin polarization measured experimentally2. One could argue that the electrons, while moving along the DNA spiral, create a magnetic field that can act on their spin and lead to the observed polarization. However, it is easy to estimate such a field and show that it is too small (10−4 T) to explain the results of Naaman and co-workers.

So what else remains that could explain these experiments? Here, we point out that the monolayers were prepared through the self-assembly of thiolated double-stranded DNA. It is well known5,6 that thiols (groups that contain sulphur atoms) interact strongly with gold atoms, leading to significant redistributions of electric charge and corresponding changes in the physical properties of gold structures and surfaces. In particular, permanent magnetism has been reported in thiol-capped gold5,6, silver and copper6 nanoparticles. It is possible, therefore, that the modification of surface states in the gold surface by the double-stranded DNA (and corresponding changes to the optical selection rules) is responsible for the effects observed by the Weizmann–Münster collaboration.

Also, we cannot completely rule out the interaction of light with chiral DNA molecules. The DNA used in these experiments is presumably right-handed (which is the more common form of DNA), so it could filter out light with anticlockwise polarization before it hits the gold surface. However, this alone cannot explain the results when the incident light has anticlockwise polarization.

The various hypotheses above could be readily verified by further experiments. In particular, it would be informative to investigate the transmission of polarized light through a layer of double-stranded DNA deposited on a transparent surface, or to repeat the experiment with left-handed DNA. Finally, it would be interesting to use a spacer, or a different chemical bond, between the gold surface and the DNA layer to better understand the role of the interactions at the interface between them. Irrespective of these suggestions, the work of Naaman and co-workers2 has opened up an intriguing new direction for research that is well worth pursuing. There will be a lot to learn along the way.


  1. 1

    Žutić, I., Fabian, J. & Das Sarma, S. Rev. Mod. Phys. 76, 323–410 (2004).

    Article  Google Scholar 

  2. 2

    Göhler, B. et al. Science 331, 894–897 (2011).

    Article  Google Scholar 

  3. 3

    Zwolak, M. & Di Ventra, M. Appl. Phys. Lett. 81, 925–927 (2002).

    CAS  Article  Google Scholar 

  4. 4

    Yeganeh, S. J. Chem. Phys. 131, 014707 (2009).

    Article  Google Scholar 

  5. 5

    Crespo, P. et al. Phys. Rev. Lett. 93, 087204 (2004).

    CAS  Article  Google Scholar 

  6. 6

    Garitaonandia, J. S. et al. Nano Lett. 8, 661–667 (2008).

    CAS  Article  Google Scholar 

Download references

Author information



Corresponding authors

Correspondence to Massimiliano Di Ventra or Yuriy V. Pershin.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Di Ventra, M., Pershin, Y. DNA spintronics sees the light. Nature Nanotech 6, 198–199 (2011). https://doi.org/10.1038/nnano.2011.48

Download citation

Further reading


Quick links

Find nanotechnology articles, nanomaterial data and patents all in one place. Visit Nano by Nature Research