Production of Photocurrent due to Intermediate-to-Conduction-Band Transitions: A Demonstration of a Key Operating Principle of the Intermediate-Band Solar Cell A. Martı ´, 1 E. Antolı ´n, 1 C. R. Stanley, 2 C. D. Farmer, 2 N. Lo ´pez, 1 P. Dı ´az, 2 E. Ca ´novas, 1 P. G. Linares, 1 and A. Luque 1 1 Instituto de Energı ´a Solar, Universidad Polite ´cnica de Madrid, E.T.S.I.Telecomunicacio ´n, Ciudad Universitaria s/n Madrid, Madrid 28040, Spain 2 Department of Electronics and Electrical Engineering, University of Glasgow, Glasgow, G12 8QQ, United Kingdom (Received 15 August 2006; published 13 December 2006) We present intermediate-band solar cells manufactured using quantum dot technology that show for the first time the production of photocurrent when two sub-band-gap energy photons are absorbed simulta- neously. One photon produces an optical transition from the intermediate-band to the conduction band while the second pumps an electron from the valence band to the intermediate-band. The detection of this two-photon absorption process is essential to verify the principles of operation of the intermediate-band solar cell. The phenomenon is the cornerstone physical principle that ultimately allows the production of photocurrent in a solar cell by below band gap photon absorption, without degradation of its output voltage. DOI: 10.1103/PhysRevLett.97.247701 PACS numbers: 84.60.Jt, 72.40.+w, 73.50.Pz, 85.60.Dw The basic principles of operation of the intermediate- band solar cell (IBSC) were originally described by Luque and Martı ´ in 1997 [1]. This novel cell relies on so-called intermediate-band material which is characterized by the existence of a band (the intermediate, IB) located between the conventional conduction and valence bands of a semi- conductor [Fig. 1(a)]. The presence of the IB leads to the generation of one net electron-hole pair when two below band gap photons are absorbed. One photon (photon ‘‘1’’) pumps an electron from the valence band (VB) to the IB while a second (photon ‘‘2’’) pumps an electron from the IB to the con- duction band (CB). This electron-hole pair adds to those produced conventionally by photons with above band gap energy E G that excite electrons directly from the VB to the CB (photon ‘‘3’’). A necessary requirement is that the IB is half-filled with electrons to provide both empty states to receive electrons from the VB and filled states to provide electrons to be supplied to the CB [2]. A further condition for successful operation of the IBSC is that the carrier population in each band is described by its own quasi-Fermi level [Fig. 1(b)]. This follows from the fact that the bands are isolated from each other by a zero density of states, and hence, the lifetimes associated with carrier recombination from one band to another are ex- pected to be much longer than carrier relaxation times within each band. These quasi-Fermi levels are E FC , E FI , and E FV for the CB, IB, and VB, respectively. The output voltage, V , of the cell is related to the quasi-Fermi levels by eV E FC E FV where e is the electron charge. To effec- tively obtain this split, the material containing the intermediate-band has to be sandwiched between two single gap semiconductors, one p-type and the other n-type. The current-voltage characteristic of the device is modeled by solving the electron and hole continuity equa- tions, as described in detail in Ref. [2,3]. It can be demonstrated rigorously [1,2] that the output voltage is limited by the total band gap E G and not by the lower of the band gaps E L or E H , as shown in Fig. 1(b). In consequence, the photocurrent gain achieved from the absorption of below band gap photons can occur without photo-voltage degradation. Within this framework, a limit- ing efficiency of 63.2% has been predicted for the IBSC, compared to 40.7% for conventional single gap solar cells and 55.4% for 2-junction tandem solar cells. These limits have been extended by Green et al. [4] to 86.8% (the same as the absolute limit for photovoltaic energy conversion [5]) for the case in which multiple bands exist within the semiconductor band gap. Essentially, these efficiency fig- ures are derived using the approximations in the Shockley and Queisser detailed balance [6] calculations which im- ply, in particular, that recombination to and from the IB takes place in the radiative limit. Luque et al. [7] have FIG. 1. Basic operation of an intermediate-band solar cell (a) structure of the intermediate-band material showing the absorption processes involved [1]; (b) intermediate-band mate- rial sandwiched between p and n type semiconductors showing the relation between voltage and quasi-Fermi level separation [2]. PRL 97, 247701 (2006) PHYSICAL REVIEW LETTERS week ending 15 DECEMBER 2006 0031-9007= 06=97(24)=247701(4) 247701-1 2006 The American Physical Society