Localisation of Switching Region

The prior experiments indicated switching near the interface of Cu(TCNQ) and the contacts. Flowever, from the results with the symmetric device we can not conclude at which of the contacts (positively or negatively biased) the switching occius. Flence, the previous planar structure was extended by an additional contact to facilitate the localisation of the switching. This electrode was placed into the gap between the two metal contacts and also covered with Cu(TCNQ) as described above. 

The extended device was fabricated as follows. After substrate cleaning a 100 nm Au-layer was deposited by e-beam evaporation. After photolithographic patterning an Au middle contact stripe of 5 am was patterned. Through the next deposition, photolithography and etching steps two A1 contact stripes were formed. The two A1 stripes were positioned parallel to the previous Au 

The additional centre contact allows us to use one common contact in the switching process of two different devices. The device with contacts I and II should operate as two separate device parts with one centre electrode with a common Cu(TCNQ)-layer. Applying a suitable voltage to one of the two device parts should switch this part from one conducting state into the other. If the switching would occur at the interface between the common centre electrode and the Cu(TCNQ), it should be possible to influence the properties of 

Indeed, the A1II deviee remained in the low condueting state and no switching was noticed. This behaviour was observed many times and no influence between the two deviees with an Au eommon electrode was noted. Therefore, the /- V charaeteristies of the two symmetrical device parts indicate that the switching does not appear at the interfaee of the Au contaet and the Cu(TCNQ) 

---