The authors report a systematic study of the bias-dependent contact resistance in rubrene single-crystal field-effect transistors with Ni, Co, Cu, Au, and Pt electrodes. They show that the reproducibility in the values of contact resistance strongly depends on the metal, ranging from a factor of 2 for Ni to more than three orders of magnitude for Au. Surprisingly, field-effect transistors with Ni, Co, and Cu contacts exhibit an unexpected reproducibility of the bias-dependent differential conductance of the contacts once this has been normalized to the value measured at zero bias. This reproducibility may enable the study of microscopic carrier injection processes into organic semiconductors. © 2007 American Institute of Physics. ͓DOI: 10.1063/1.2741411͔ Considerable improvements in the material control of organic thin films are now enabling the reproducible, lowcost fabrication of organic field-effect transistors ͑FETs͒ with mobility values in the range of 0.1-1 cm 2 /V s. 1 These values are sufficient for the development of applications in the field of plastic electronics. 2 However, the fabrication of highquality electrical contacts for organic transistors has not progressed comparably, 3 and contacts are now posing limits to the performance of organic FETs. Specifically, with mobility values in between 0.1 and 1 cm 2 / V s, the contact resistance-typically larger than 1 k⍀ cm even in the best devices-limits the transistor performance as soon as the channel length becomes smaller than Ӎ10 m, 4 preventing the possibility of device downscaling. Irreproducibility makes the situation even worse: for gold-contacted pentacene thin-film FETs, for instance, the spread in contact resistance values was recently observed to exceed three orders of magnitude ͑from 2 k⍀ cm to more than 1 M⍀ cm͒. 5 The current lack of understanding of the microscopic carrier injection 6 processes from a metal electrode into an organic semiconductor does not help us in determining the causes of the observed irreproducibility and more systematic experiments are needed.Here we report systematic transport measurements of rubrene ͑C 42 H 28 ͒ single-crystal FETs with electrodes made of five different metals ͑Ni, Co, Cu, Au, and Pt͒. All the transistors have been fabricated with a sufficiently short channel length, so that the total device resistance is entirely dominated by the contacts. By studying more than 250 contactdominated devices, we have collected enough statistics to determine the average contact resistance, its spread in values, as well as its bias dependence. We find significant differences between the different metals. In particular, for nickel-which exhibits the lowest resistance-the spread is only a factor of 2, for cobalt and copper slightly more than one order of magnitude, and for gold more than three orders of magnitude ͑platinum seems to behave similarly to gold, but the number of devices tested was not sufficient to make more quantitative statements͒. We also find that for Ni, Cu, and Co ͑but nor for Au and Pt͒ the bias depe...