The short answer:

Energy dissipated in the dead layer does generate electron-hole pairs, but the dead region has such low resistance it acts as a conductor and there is no capacitive depletion region. It is charge motion in the depletion region that produces a measurable signal. Although there is some capacitance associated with the dead layer, it is shorted by the low material resistance. Consequently, the dead region acts as part of the electrical contact rather than the detecting region. Gamma-rays interacting in the “dead” region may excite energetic electrons capable of leaving the dead region and entering the sensitive region, but the measured energy with be lower due to the energy absorbed/lost in the dead region. This same problem will appear for charged particles. For instance, an alpha particle transiting the dead region will have some energy loss Delta E in this same region, causing a slight downshift in the pulse height spectrum equal to Delta E. Further, the energy resolution will decrease due to the added statistical uncertainly related the variance in Delta E.

The long answer:

A high-purity germanium (HPGe) detector is a semiconductor diode. There are multiple configurations, but the two most popular configurations are coaxial and planar geometries. The material is extremely pure, processed by zone refining to remove background impurities. Yet, even with shallow impurity carrier concentrations on the order of 10^10 cm^-3, at the usual operating temperature of 77K, approximately half of these shallow impurities are still ionized, and these charge carriers can produce problematic leakage current. Consequently, reverse biased pn junctions must be used to block the injection of charge carriers. P-type contacts are applied to n-type material and n-type contacts applied to p-type material. Under reverse bias voltage, the free charge carriers in the diode are removed at a rate much faster than the bulk generation rate. The supply of minority carriers is diminished in the actual junction contact, that is, holes are of low density in n-type material and electrons are of low density in p-type material, that the small supply of replacement charges can not inject enough charges to make up for the rapid removal of free carriers. Consequently, a region in the device becomes relatively depleted of charge carriers, and this depleted region is where gamma rays are detected. The width of the depleted region is generally a function of K(V/N)^(1/2) where K is constant, V is the applied reverse voltage, and N is the background concentration of the bulk material. This depletion region barely extends into the contact region (insignificant); hence the contact does not sustain an electric field. In summary, the doped junctions are required to block leakage current, but also form an insensitive region where the depletion region is hardly developed at all.

The measured signal is consequence of charge induction, predicted either by Green’s theorem or Green’s reciprocal theorem. An often quoted theorem used to explain charge induction in a semiconductor detector, a reduced form of Green’s theorem, is the Ramo theorem. Overall, it states that an independent electric field is produced by a charge between two conductive contacts, and this produces “mirror” charges on the contacts due to attraction and repulsion of electrons at those conductive surfaces. The electric field line density terminating at the contacts is a function of contact distance, shape, and size. As the charge moves between the contacts, the electric field lines terminated at the contacts also change. It is this change in electric field of each specific moving charge (not the applied voltage) that produces a measurable induced charge at the contacts, and this change, manifested as an induced current, is essentially what the detector electronics are measuring. The contact does not sustain an electric field. Hence, if the dead region is conductive, then it will act as part of the contact and not the detector region.

Further, the HPGe diode is encapsulated in a mounting cup that is mounted to a cold finger stem, and further encapsulated in an evacuated cap. These additional containers add absorbers in the path of the gamma-rays, and ultimately attenuate the flux, especially at low energies. Gamma-rays absorbed in these components surrounding the detector will not be absorbed in the detector. Hence, the efficiency curve of HPGe detectors reduces at low photon energies. The photon absorption efficiency in the detector, by photoelectric and Compton scattering, reduces with increasing energy. The efficiency generally also decreases as the photon energy increases. Consequently, there is an optimum energy where the device has the highest gamma-ray detection efficiency, usually somewhere between 50 keV – 100 keV. (Note that at high enough energies, above several MeV, pair production increases such that it becomes dominant.)

I hope that this helps.

Thank you very much Dr. Douglas S. McGregor for your request, please, I have another question. Taking into consideration the complexities and diversity of procedures *patterns* that gamma rays can follow from its entrance to the detector until they pass the dead layer region. What is the best method that can be considered for a correct simulation of the gamma transport into the dead layer some researchers consider a certain thickness of a dead layer in the simulation model and they consider that all interactions (or conversion of energy) into this region does not produce any type of signal. then they do the same with a new thickness of dead layer.... until to get to the same experimental efficiency.

what is your opinion, this method is correct?

Best Regards