Page 271 - Organic Electronics in Sensors and Biotechnology
P. 271

248    Cha pte r  S i x

               complexity of the fabrication process, the panels are expensive, and
               there is significant interest in alternative technologies that would per-
               mit larger panels to be fabricated at substantially reduced cost.
                   OPV devices are especially appealing in this respect due to their
               excellent optoelectronic characteristics and their amenability to large-area,
               low-cost printing. The use of flexible plastic substrates raises the possi-
               bility of creating low-cost, large-area conformable panels that can adapt
               to the shape of the object or the radiation source, offering superior reso-
               lution and imaging capabilities. With these benefits in mind, Blakesley
                         87
               and Speller  undertook a conceptual feasibility study to determine
               whether OPV panels could provide a viable alternative to existing panel
                          87
               technologies.   They modeled the behavior of an OPV array on top of an
               a-Si backplane of thin-film transistors as shown in Fig. 6.28. The operat-
               ing principle of the OPV panels is most easily understood by considering
               a single pixel, which comprises a photodiode and an associated transis-
               tor. The anode of the photodiode is held at a fixed negative potential V ,
                                                                        b
               and the cathode is connected to the source terminal of the transistor. The
               drain terminal is connected to the inverting input of an op-amp integra-
               tor. The non-inverting input of the op-amp is permanently grounded,
               and so the inverting input behaves as a virtual earth. To start, the pixel is
               reset by opening the gate, causing a reverse bias of approximately V  to
                                                                      b
               be applied to the photodiode and thereby charging the capacitor to this
               value. The pixel is then disconnected from the integrator by closing the
               gate. The pixel is illuminated, which generates a negative photocurrent
               that partially discharges the capacitor. The pixel is then reconnected to


                  Column i, row 1     Column i, row 2
                                                             V b



                          R sh                R sh










                                                                      ADC


                  Row select

               FIGURE 6.28  A simple passive pixel circuit for an organic photodiode array.
               (Adapted with permission from Ref. 87, American Association of Physicist in
               Medicine, 2008.)
   266   267   268   269   270   271   272   273   274   275   276