Molecular electronics

Charged with manipulation

The ability to control charge transport through individual molecules sandwiched between electrodes could lead to further miniaturization of electronics. A better understanding of how such junctions work is crucial.

In its simplest form, a molecular junction consists of a single organic molecule sandwiched between two much larger electrodes. More efficient and more precise control of current flow through such elements, and a detailed understanding of the factors that influence that flow, are essential to take electronics from the microscale of conventional, silicon-based technology down to the nanoscale. On page 658 of this issue1, Wolkow and colleagues demonstrate how to control the onset of conduction through a molecular junction by using the charge state of an atom on the surface of a silicon electrode.

The difficulty with structures such as molecular junctions is that, although the molecule consists of a series of discrete states in a small, finite entity, the electrodes contain a very dense set of states in a macroscopic structure. Understanding how the electrostatic environment of the molecule modifies the transport process — in other words, the influence of the electrode — is, in many ways, the most vexing problem in dealing with such junctions.

In the more typical metal–molecule–metal type of molecular junction2,3,4,5,6, extra complexity results from the geometric disorder inherent in the usual sorts of metal–molecule coordination bonding. Typically, these bonds are between a coinage metal (copper, silver or gold) and a thiol group, or between palladium and a cyano group7. Geometric changes can strongly affect charge transport through these junctions and give rise to the random switching phenomena often seen8,9,10,11 in such structures.

One way to avoid this geometric uncertainty is to exploit transport in junctions built not on a metal but on a semiconductor (particularly silicon), where the molecule–electrode interface can be provided by two atoms ‘sharing’ the electrons in a covalent bond. This bond can be created in several ways12,13,14,15,16, but perhaps most directly by linking a free radical at the end of the molecule — in this case, a carbon atom with an unpaired electron — to a ‘dangling bond’ on the surface of the silicon electrode. This dangling bond can arise through the removal of a hydrogen atom originally attached to each silicon atom at the surface of the passivated electrode, leaving behind an unpaired electron that hangs free.

Wolkow and colleagues1 take advantage of two remarkable properties of silicon surfaces to characterize how changes in the charge state of a silicon surface atom influence the effective field over the molecular junction. First, the ‘polymerization’ of the molecules — the process by which they attach themselves covalently along ‘dimer rows’ on the silicon surface to form a line — ends abruptly at a dangling-bond site (Fig. 1). Second, the dangling-bond site itself can become more or less charged depending on the doping level of the silicon (that is, the deficiency or excess of electrons that is induced by adding an ‘impurity element’ with an intrinsic number of valence electrons different from silicon's four).

Figure 1: Dangling potential.

The polymerization of a molecule (here an organic styrene-derived molecule, not to scale) on a silicon substrate stops abruptly at a dangling-bond site. The blue and purple lines indicate the height of the molecules, as seen by a scanning tunnelling microscope (STM) — a measure of the charge transport across the molecules. At higher bias (blue line), all molecules are ‘turned on’, and appear bright in the STM picture. At lower bias (purple line), all molecules should appear dark. Wolkow et al.1, however, discover that the electrostatic potential of the negatively charged dangling bond causes the nearest molecules to remain bright. This suggests that such structures could be used to manipulate charge transport through molecular junctions. (Figure adapted from Fig. 1 of ref. 1.)

Wolkow and colleagues1 used a scanning-probe microscope to examine the effective charges along the length of the polymers originating from a given dangling-bond site. They found that when the dangling-bond site is charged, a ‘slope’ structure is seen that is absent when the site is uncharged (Fig. 1): molecules attached to the silicon electrode close to a charged dangling bond appear to stand out farther from the electrode than those at a greater distance from the dangling-bond site. Wolkow and colleagues interpret this as the effect of a local electrostatic charge on charge transport through the polymeric wire. This interpretation is supported by calculations showing changes in molecular orbital states caused by the charged dangling-bond site. This result constitutes direct evidence that localized charges profoundly affect charge transport in single-molecule structures on silicon surfaces at room temperature.

The silicon–organic interface investigated by Wolkow and co-workers has many advantages over the more conventional metal–molecule interface (at least as the latter is usually implemented). One is the tight geometric constraint provided by a single covalent bond. Another is the bandgap character of the silicon (requiring an electron to acquire additional energy to contribute to conduction), which can be changed by doping to adjust the conductivity. This bandgap can result in useful phenomena, such as the negative differential resistance12,17 (equivalent to an increase in voltage leading to a decrease in current) observed in transport through molecules bonded to dangling-bond sites.

More generally, changing the electrostatic potential on a molecule will change its conduction characteristics3,5,6,17,18. In ordinary transistor geometries, this electrostatic control is provided by one of three electrodes, the ‘gate’, which regulates the amount of current that can flow from ‘source’ to ‘drain’ through the main channel of the transistor. In most single-molecule transport measurements, for example using ‘break junctions’9,19,20, such gating can also be attained, but it requires very large voltages. In effect, this is because the molecular entities are a long way from the gate compared with the source–drain distance. This new work suggests that more effective electrostatic control could perhaps be obtained using localized charges.

The article by Wolkow and colleagues is one of a series of contributions1,12,13,16,17 that underline the potential of molecular transport junctions using silicon electrodes, and explore the different control mechanisms and mechanistic transport behaviours that can be observed in them. Using a semiconductor electrode substantially widens the perspective on single-molecule transport structures, the basic unit of molecular electronics.


  1. 1

    Piva, P. G. et al. Nature 435, 658–661 (2005).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Bowler, D. R. J. Phys. Condens. Matter 16, R721–R754 (2004).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Nitzan, A. & Ratner, M. A. Science 300, 1384–1389 (2003).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Reed, M. A. & Takhee, L. (eds) Molecular Nanoelectronics (American Scientific, Stevenson Ranch, CA, 2003).

  5. 5

    Nitzan, A. Annu. Rev. Phys. Chem. 52, 681–750 (2001).

    ADS  CAS  Article  Google Scholar 

  6. 6

    Datta, S. Quantum Transport: Atom to Transistor (Cambridge Univ. Press, 2005).

    Google Scholar 

  7. 7

    Tour, J. M. Molecular Electronics: Commercial Insights, Chemistry, Devices, Architecture and Programming (World Scientific, River Edge, NJ, 2003).

    Google Scholar 

  8. 8

    Xiao, X. Y., Xu, B. Q. & Tao, N. J. Nanoletters 4, 267–271 (2004).

    ADS  CAS  Article  Google Scholar 

  9. 9

    Mayor, M. & Weber, H. B. Angew. Chem. Int. Edn Engl. 43, 2882–2884 (2004).

    CAS  Article  Google Scholar 

  10. 10

    Basch, H., Cohen, R. & Ratner, M. A. Nanoletters (in the press).

  11. 11

    Lewis, P. A. et al. J. Am. Chem. Soc. 126, 12214–12215 (2004).

    CAS  Article  Google Scholar 

  12. 12

    Guisinger, N. P. et al. Nanoletters 4, 55–59 (2004).

    ADS  CAS  Article  Google Scholar 

  13. 13

    Tong, X., DiLabio, G. A. & Wolkow, R. A. Nanoletters 4, 979–983 (2004).

    ADS  CAS  Article  Google Scholar 

  14. 14

    Hamers, R. J. et al. Acc. Chem. Res. 33, 617–624 (2000).

    CAS  Article  Google Scholar 

  15. 15

    Filler, M. A. & Bent, S. F. Prog. Surf. Sci. 73, 1–56 (2003).

    ADS  CAS  Article  Google Scholar 

  16. 16

    Hersam, M. C., Guisinger, N. P. & Lyding, J. W. Nanotechnology 11, 70–76 (2000).

    ADS  CAS  Article  Google Scholar 

  17. 17

    Rakshit, T. et al. Nanoletters 4, 1803–1807 (2004).

    ADS  CAS  Article  Google Scholar 

  18. 18

    Xue, Y. Q. & Ratner, M. A. Int. J. Quant. Chem. 102, 911–924 (2005).

    ADS  CAS  Article  Google Scholar 

  19. 19

    Liang, W. J. et al. Nature 417, 725–729 (2002).

    ADS  CAS  Article  Google Scholar 

  20. 20

    Park, J. et al. Nature 417, 722–725 (2002).

    ADS  CAS  Article  Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Ratner, M. Charged with manipulation. Nature 435, 575–576 (2005).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.