Hector de los Santos recently proposed a logic circuid based on plamons. And I have a question concerning the interpretation of the last graph on the article he made his proposal. http://ieeexplore.ieee.org/xpl/freepre_abs_all.jsp?isnumber=4359107&arnumber=5233891 [Broken] Pag. 13, fig. 3C. On channel length of 200nm the value of exp(-k_im*L) is between 0.8 and 1.0, but in appedix D, on the same page, the author writes: "Fig. C.3 indeed shows that for NFL operation, an wt product of at least 7 would be required for detecting a plasmon at an amplitude of 20% its maximum 200 nm away from the input". So, in the graph it is written a value around 0.9, but he says 0.2. What am I missing?