Predicting and rationalizing the effect of surface charge distribution and orientation on nano-wire based FET bio-sensors† Luca De Vico, * a Lars Iversen, bc Martin H. Sørensen, d Mads Brandbyge, d Jesper Nyg  ard, ec Karen L. Martinez bc and Jan H. Jensen * a Received 25th March 2011, Accepted 19th May 2011 DOI: 10.1039/c1nr10316d A single charge screening model of surface charge sensors in liquids (De Vico et al., Nanoscale, 2011, 3, 706–717) is extended to multiple charges to model the effect of the charge distributions of analyte proteins on FET sensor response. With this model we show that counter-intuitive signal changes (e.g. a positive signal change due to a net positive protein binding to a p-type conductor) can occur for certain combinations of charge distributions and Debye lengths. The new method is applied to interpret published experimental data on Streptavidin (Ishikawa et al., ACS Nano, 2009, 3, 3969–3976) and Nucleocapsid protein (Ishikawa et al., ACS Nano, 2009, 3, 1219–1224). 1 Introduction Nano-BioFET sensors 1–5 are made of electrically contacted, semiconducting nano-wires working as field effect transistors (FET), which can sense binding events taking place at their surface. A bio-functionalization layer is linked to the surface of the nano-wire and is responsible for the recognition of the ana- lyte of interest. The analyte immobilization creates a change in the charge distribution in proximity to the surface (modulated by the ionic species present in solution through a Debye screening), 6 with a consequent perturbation of the current flow in the nano- wire. Nano-BioFET sensor importance lies in (i) the high binding selectivity inherited from the bio-molecules constituting the bio- functionalization layer, (ii) the absence of a need to label the analyte or receptor, and (iii) the high sensitivity for analytes with concentrations as low as femto molar. 7,8 While theoretical models of nano-BioFETs are available, 8–15 most interpretations of experimentally observed signal changes are done by using what can be called a single charge model: it is verified that the direction of the signal is the one expected based on the total charge of the analyte, taking into account whether the nano-wire is an n- or p-type semiconductor (e.g. a decrease in conductance due to a negative analyte binding to an n-type nano- wire). This assumption works satisfactorily in many cases, and we have recently shown that the single charge model can also form the basis for a semi-quantitative interpretation of the magnitude of the nano-wire conductance signal change and its dependence on various experimental conditions, such as ionic strength and pH. 16 However, a few reports have emerged where the observed signal change did not have the expected direction based on the total charge of the protein analyte. 17,18 One possible reason for this is that the total charge of the protein is too simple a repre- sentation of the effect of the many charged groups distributed throughout the protein, which are subject to different screening effects from the buffer ions depending on their different distances from the nano-wire surface. To test this hypothesis an improved model is needed. In this paper we (i) present an extension of our single charge model for nano-BioFET conductance sensitivity 16 to multiple charges, (ii) demonstrate under which conditions it is possible to expect counter-intuitive changes in the sign of the signal, by considering a few prototypical charge distributions, and (iii) apply the multiple charges method to the interpretation of experimentally observed signal changes that cannot be explained by the single charge model. 2 Computational methodology In the current method we want to predict the change in conductance sensitivity  DG G 0  when N systems (biomolecules) with m charged groups are sensed close to the surface of a Department of Chemistry, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen, Denmark. E-mail: luca@chem.ku.dk; jhjensen@chem.ku.dk; Fax: +45 35 32 02 14; Tel: +45 35 32 02 39 b Bio-Nanotechnology Laboratory, Department of Neuroscience and Pharmacology, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen, Denmark c Nano-Science Center, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen, Denmark d DTU Nanotech, Department of Micro and Nanotechnology, Technical University of Denmark, DTU-Building 345 East, DK-2800 Kongens Lyngby, Denmark e Niels Bohr Institute, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen, Denmark † Electronic supplementary information (ESI) available: A detailed description of the SMSCSL model is given, along with the Streptavidin and N protein treatment. Additional material is provided in Tables 1S–3S and Fig. 1S–13S. See DOI: 10.1039/c1nr10316d This journal is ª The Royal Society of Chemistry 2011 Nanoscale, 2011, 3, 3635–3640 | 3635 Dynamic Article Links C < Nanoscale Cite this: Nanoscale, 2011, 3, 3635 www.rsc.org/nanoscale PAPER