Uncertainty in CCD detectors with and without cooling devices when used for molecular fluorescence measurements

The performance of CCD detectors for fluorescence measurements is evaluated through uncertainty studies, mainly as a function of both the signal intensity and the detector temperature. Two CCD detectors, one furnished with a cooling device and the other without a cooling device, were used, and the results were compared. The dependence of uncertainty on the instrumental signal was evaluated simultaneously for both detectors at temperatures ranging between -23 and 23 degrees C. The tested detectors required temperature increments between 30 and 50 degrees C to double dark noise (the random part of dark current). The detector temperature affects uncertainty, but this is basically related to dark noise in fluorescence measurements. The consequence of this is that the temperature does not significantly affect the signal-to-noise ratio (S/N); thus, the presence of a cooling device does not significantly improvement the performance, and the limit of detection (LOD) does not depend strongly on the detector temperature. The quality of the two-dimensional array does affect uncertainty, the value of S/N, and the LOD. A good CCD detector can perform at a level similar to a spectrofluorometer furnished with a photomultiplier tube. Laboratory data are given to show how the three components of uncertainty (dark noise, shot noise and flicker noise) behave at different signal intensities and temperatures. Dark noise is the most important factor, shot noise has relevance only at high signal values, and flicker noise is practically irrelevant. The classical model of the dependence of S/N on the fluorescence signal is applied, and the uncertainty constants that govern the performance of the apparatus used are given.