Difference between CCD and PMT

 Difference Between Charge couple device  & Photo multiplier Tubes :

Ø  Photomultiplier tubes (PMTs) and CCDs both give spectra. The difference is the PMT is used with a small slit in front of it to control the bandwidth of light being detected. The CCD takes advantage of the dispersed light fully. The pixel columns will each correspond to a wavelength (resolution and range depend on the grating used). A PMT requires scanning of the Monochromator to collect a spectra. The CCD takes a single snap shot and you have a spectrum. The CCD sensitivity and dynamic range is lower than a PMT.

Ø  A photomultiplier tube is a detection device that is made from a glass vacuum tube with a series of metal plate electrodes. A CCD is a solid state detector made from semiconductor materials

Ø  The main difference is one of sensitivity. Generaly speaking the better the spectral resolution of the instrument the lower the amount of light reaching the detector and so you need more sensitivity in your detector. A PMT measures a single point in the spectrum at a time whereas with a CCD the complate spectrum is imaged across the CCD and so can be measured all at the same time. 
An instrument with a CCD is usually much faster and cheaper but will not have as good a spectral resolution (the ability to resolve absorbance peaks very close to each other).

Ø  CCDs and photomultipliers vary in a number of aspects. One difference is gain, a photomultiplier has gain whereas a CCD does not (hence the multiplier bit of PMT). The PMT gain may be up to 10,000,000 and is available at high speeds and for large area detectors, which means that one can usually get close to the theoretical noise floor. On the other hand, PMTs have poor quantum efficiency compared to CCDs (25% typ against 85% typ) so you can sometimes get better performance with a CCD if you can go slowly enough.

Ø  PMTs are also typically single channel devices, although 16 channel linear arrays are available. CCDs are usually linear or 2D arrays.

Ø  In a dispersive spectrometer a linear CCD array can capture the entire spectrum in one measurement. A single channel PMT must have the spectrum scanned across it sequentially to produce the entire spectrum.

Ø  PMT's are typically preferable to CCD's on spectroscopic application for several reasons. The ability to adjust the gain of each PMT allows a manufacturer to adjust the response of each PMT to the specific signal being measured, so every element you are trying to detect can be analyzed at optimum conditions. Solid state CCD's are a compromise. Every element detected has the same conditions, so most are compromised. 

Ø  Also, PMT's can be heated and held at constant temperature (in well made instruments) to prevent drift caused by variation in temperature. If you try to heat a CCD, the noise level will go up, and the signal to noie ratio will degrade as a result. CCD's are sometimes cooled to try to improve their s/n ratio, but usually not cooled enough to really help much due to condensation issues that arise. 

Ø  A third advantage of PMT's is that they can be used in a vacuum chamber without long term degradation for decades of use. The surface of a CCD will degrade under vacuum over a few (3-5) years. Most manufacturers making CCD based instruments opt for a Nitrogen or Argon flush, rather than vacuum to displace the oxygen from the detector chamber. This method results in decreased performance compared to PMT's, and is used in lower performance less expensive spectrometers.