Anisotropically grown (b-axis short) single-nano TiO2 (B), uniformly hyper-dispersed on the surface of multiwalled carbon nanotubes (MWCNT), was successfully synthesized via an in situ ultracentrifugation (UC) process coupled with a follow-up hydrothermal treatment. The uc-TiO2 (B)/MWCNT composite materials enable ultrafast Li(+) intercalation especially along the b-axis, resulting in a capacity of 235 mA h g(-1) per TiO2 (B) even at 300C (1C = 335 mA g(-1) ).

1. Introduction The next step in the high energy density Nanohybrid™ Capacitor (nanocrystalline Li4Ti5O12/activated carbon hybrid supercapacitors)1 is to substitute the negative electrode with other potential candidates with higher voltage (lower reaction potential) and high capacity. In the presentation, we focus on bronze-type TiO2 (TiO2(B)) which has a much higher electric conductivity (~10-2 S cm-1) compared to other TiO2 polymorphs such as anatase and rutile (10-14 ~10-13 S cm-1).2 TiO2(B) shows a theoretical capacity of 335 mAh g-1 during Li+ ion intercalation, where Li+ diffusion proceeds along the b-axis tunnel giving poor Li+ diffusion coefficient of 10-14 ~ 10-16 cm2 s-1.3 Using our original ultracentrifugation (UC) treatment, we have successfully synthesized the anisotropic-grown (b-axis short), single-nano TiO2(B), which is uniformly and highly dispersed within carbon matrix. Electrochemical performances of the hybrid supercapacitor composed of uc-TiO2(B) negative and activated carbon positive electrodes were tested and compared with those for the conventional EDLC and the uc-Li4Ti5O12-based hybrid supercapacitors. 2. Experimental In order to optimize the uc-TiO2(B) hybrid supercapacitor system, we prepared three uc-TiO2(B) negative electrodes under different conditions; i) untreated uc-TiO2(B) (case 1)), ii) pre-cycled uc-TiO2(B) to cancel its irreversible capacity (case 2)), and iii) uc-TiO2(B) with Li pre-doping (case 3)). Then, the laminate-type test cells (3 cm * 4 cm) were assembled using the uc-TiO2(B) negative electrodes combined with activated carbon (AC) positive electrodes, and the additional Ag reference electrode was used when necessary. The electrolyte was 1M LiPF4 in propylene carbonate. Test cells of EDLC and uc-Li4Ti5O12-based hybrid supercapacitors were assembled using the same AC positive electrode and electrolyte as uc-TiO2(B) system. Electrochemical characterizations (charge-discharge curves, rate capability, and cycleability etc…) were performed on the assembled test cells. 3. Results and Discussion The hybrid supercapacitor case 1) did not reach the targeted cell voltage (3.0V) appropriately and its capacity degraded within few cycles. Charge discharge profiles of positive and negative electrodes for the case 1) showed that the AC positive electrode was overcharged due to the characteristic behavior of the untreated uc-TiO2(B) negative electrode, its irreversible capacity during initial cycling and large voltage hysteresis especially in the high depth of discharge (DOD). Better cell performance was obtained for the case 2), thanks to canceling the irreversible capacity of uc-TiO2(B). The case 3) showed the best cell performance among three cases, when the uc-TiO2(B) was pre-lithiated and its operation potential was confined within the range between 1.0 V and 1.5 V vs. Li/Li+. For the optimized uc-TiO2(B) hybrid supercapacitor (case 3), the estimated volume energy density was tripled from that of EDLC and higher compared to the uc-Li4Ti5O12 hybrid supercapacitor. References 1) K. Naoi, S. Ishimoto, J. Miyamoto and W. Naoi., Energy & Environ. Sci., 5, (2012) 9363. 2) N. Taniguchi, M. Kato and K. Hirota, J. Jpn. Soc., 2012, 59, 326. 3) C. W. Mason, I. Yeo, K. Saravanan and P. Balaya, RSC Adv., 2013, 3, 2935.