Supplementary material for the paper

method in construction the two kinds of contact in figure 4(a) and (b)

We constructed the different electrical setups by precisely controlling (not by chance) the movement of the tubes and verified by more than 10 times of similar experiments (considering both the images and I-V measurement). We make it sure following the below steps in construction the two kinds of connect: firstly, the CNT-fullerene-CNT contact is constructed, the I-V curve(s) is obtained; then the CNT on the movable probe side was pushed forward carefully, step by step (movement resolusion: 20 pm). The CNT-fullerene-CNT contact is maintained during whole manipulation process (following the mechanism described in figure 1(b)). Secondly, the fullerene is pushed aside and the two CNTs contacted with each other. This process also can be precisely performed under controlling the movable probe. This process can be performed under carefully controlled manner and can be prepared repeatedly, which insures the results given in figure 4.

Supplementary references
[r1] A. A. Talin, F. Léonard, B. S. Swartzentruber, X. Wang, and S. D. Hersee, Phys. Rev. Lett., 101(2008)076802

[r2] A. Carbone, B. K. Kotowska, and D. Kotowski, Phys. Rev. Lett., 95(2005) 236601

[r3] Y. Gu and L. J. Lauhon, Appl. Phys. Lett., 89(2006)143102

[r4] A. Rose, Phys. Rev. 97(1955)1538

[r5] Y. L. Lin and F. Nori, Phys. Rev. B, 49(1994)5020

[r6] M. Pudlak and R. Pincak, Phys. Rev. A, 79(2009)033202

[r7] F. Börrnert, C. Börrnert, S. Gorantla, X. J. Liu, A. Bachmatiuk, J. O. Joswig, F. R. Wagner, F. Schäffel, J. H. Warner, R. Schönfelder, B. Rellinghaus, T. Gemming, J. Thomas, M. Knupfer, B. Büchner, and M. H. Rümmeli, Phys. Rev. B, 81(2010)085439

[r8] A. Nagataki, T. Kawai, Y. Miyamoto, O. Suekane, and Y. Nakayama, Phys. Rev. Lett., 102(2009)176808