DOI: 10.1002/adma.200801810 Nanotube–Silicon Heterojunction Solar Cells** By Yi Jia, Jinquan Wei, Kunlin Wang, Anyuan Cao, * Qinke Shu, Xuchun Gui, Yanqiu Zhu, Daming Zhuang, Gong Zhang, Beibei Ma, Liduo Wang, Wenjin Liu, Zhicheng Wang, Jianbin Luo, and Dehai Wu* The 1D nanoscale structure, high mobility, and excellent physical and electronic properties of carbon nanotubes [1–3] offer great promise in the development of high-efﬁciency solar cells. However, use of nanotubes in this area has been primarily focused on hybrid structures based on conjugated polymers, where the potential of nanotubes has not been fully explored, and the cell efﬁciency remains at relatively low levels (<1%). [4–9] Here, we report on carbon-nanotube-on-silicon heterojunction solar cells, with efﬁciencies of 5–7% under one solar illumination at 100 mW cm 2 and excellent air stability. The nanotubes serve as a heterojunction component for charge separation, as a highly conductive percolated network for charge transport, and as a transparent electrode for light illumination and charge collection. Our results represent a route for developing solar cells with modest to moderate efﬁciency and reduced cost, by integrating nano- and silicon technologies. Despite intensive efforts in involving carbon nanotubes as active components in polymer-based solar cells, [4–9] current results for such hybrid cells are still far from achieving modest to moderate (e.g., 5–10%) efﬁciencies or good environmental stability. Coexistence of metallic and semiconducting nano- tubes, phase segregation between nanotubes and polymers, and aggregation of nanotubes at higher concentrations (for reaching percolation) [7–10] are major challenges to be over- come for substantially improving cell efﬁciencies. While nanotubes are able to deliver dissociated charge carriers (e.g., electrons) to the electrode at high speed, polymers having poor mobility could severely limit the hole transport to the opposite electrode. [11] The polymer-based solar cells also suffer from degradation in air even at room temperature, although nanotubes are stable at much higher temperatures. Several other recent reports used high-conductivity nanotube ﬁlms as transparent electrodes for replacing indium tin oxide (ITO). However, the nanotubes were not involved into the photogeneration process, and cell efﬁciency was mainly limited by the polymeric materials forming the active layers. [12–14] On the other hand, silicon-based photovoltaics have been well established, and occupy more than 90% of the market. [15] The moderate efﬁciency (routinely >10%), stability, and material abundance of silicon maintained its position of dominance over other materials. [16] One of the major challenges is still the manufacturing cost of silicon solar cells, which remains high. [17] Despite that, development of amor- phous and polycrystalline silicon materials offers opportunities for second-generation solar cells, with much reduced cost and modest efﬁciencies. [18,19] Recently, carbon nanotubes have been deposited on silicon wafers to produce optoelectronic devices with appreciable photocurrent response. [20–22] Here, we combined carbon nanotubes with silicon to form heterojunction solar cells with modest efﬁciencies and high air stability. By simply coating thin nanotube ﬁlms onto silicon wafers, we have achieved power-conversion efﬁciencies of about 7% and negligible degradation of current density after hundreds of hours of exposure to air. The manufacturing process is simple and scalable, involving solution transfer of a porous, single-layer ﬁlm of double-walled nanotubes to a silicon surface to form heterojunctions with silicon, and does not require separation of metallic and semiconducting nanotubes. Compared with metal-on-Si Schottky cells, with an efﬁciency of 9.5%, [23] our method utilized the distinct properties of both crystalline silicon (e.g., diffusion length) and nanotubes (e.g., high mobility) to construct efﬁcient solar cells, and could potentially lead to lower-cost approaches based on nanomaterial–semiconductor heterojunction structures. The heterojunction solar cells consist of an n-type mono- crystalline silicon wafer coated by a thin ﬁlm of double-walled carbon nanotubes (DWNTs) through a solution transfer process (Fig. 1a, see Experimental section). Here, the DWNT ﬁlm is involved in three key processes for energy conversion. COMMUNICATION [*] Prof. A. Cao, Y. Jia Department of Mechanical Engineering University of Hawaii at Manoa Honolulu HI 96822 (USA) E-mail: anyuan@hawaii.edu Y. Jia, Dr. J. Wei, Prof. K. Wang, Q. Shu, X. Gui, Prof. D. Zhuang Dr. G. Zhang, Prof. W. Liu, Dr. Z. Wang, Dr. J. Luo, Prof. D. Wu Key Laboratory for Advanced Materials Processing Technology Ministry of Education, Department of Mechanical Engineering Tsinghua University, Beijing 100084 (PR China) E-mail: wdh-dme@tsinghua.edu.cn Prof. Y. Zhu Advanced Materials Group School of Mechanical, Materials and Manufacturing Engineering University of Nottingham University Park, Nottingham NG7 2RD (UK) B. Ma, Prof. L. Wang Department of Chemistry, Tsinghua University Beijing 100084 (PR China) [**] We thank C. M. Lieber and E. Miller for helpful discussions. A. Cao acknowledges funding support from the US National Science Foundation (NSF) grant CMMI-0728197. J. Wei thanks NSFC (grant no.: 50672047) for funding support. Supporting Information is available online from Wiley InterScience or from the author. 4594 ß 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Mater. 2008, 20, 4594–4598