Do you mean a situation when n type germanium is grown on p type silicon? If yes, parameters of such diode will be influenced by strain relaxation in germanium because these two materials have different lattice constants.

There would be band bending at interface for sure.

There are different energy levels through which electrons flows or reside. The energy levels for silicon and germanium are different, which in case the electron from n type germanium diffuses towards p type silicon, the recombination may be a problem, because the electron energy level should match with the p type carriers.

Now as we know the band gap energy for germanium is less than silicon, hence when there is rise in the temperature, the number of intrinsic carrier will also be different, creating a rise or fall in carriers.

The other problem is that growing p type silicon on n type germanium will itself create problems as the chemistry would not allow to grow them uniformly or properly.

What I feel is I dont see any particular advantage in such a structure. Please share your thoughts on the advantage of such a structure

An extrinsic semiconductor (SMC) is one which contains a dopant and has different electrical properties than the intrinsic SMC. n- and p- types of extrinsic SMC are SEGREGATED as follows:

Si/Ge (group14 elements with 4valence electrons)) can form four bonds with four more Si/Ge atoms. p- type Si/Ge SMC is formed when a group 13 element (B.Al.Ga, In) called dopant having only 3 valence electrons is added as impurity as follows:

This electron deficient dopant forms three bonds with each one of the three Si/Ge and leaves ONE BOND UNSATISFIEDor a hole is created which otherwise would have contained one electron.In a way, this hole being empty is analogous to positive hole . With the presence of a definite mole % of the dopant, a large number of such holes are created in the otherwese non conducting Si/Ge crystals. When potential differenc is applied, such holes can move across the crystal and can conduct electricity to some extent. As the electricity is conducted by the movement of these POSITIVE HOLES, hence such SMC are called p-type SMC.

n- type Si/Ge (SMC) are formed when a group 15 element dopant(P,As,Sb) having 5 valence electrons is added as impurity as follows:

These electron rich dopants form four bonds with each one of the four Si/Ge and leave one electron FREE. With the presence of a definite mole % of the dopant, a large number of such free electrons now become available in the otherwese non conducting Si/Ge crystals. When potential differenc is applied, such free electrons can move across the crystal and can conduct electricity in the NORMAL way, i.e., as happens in metals where free electrons are the conductors. As the electricity is conducted by the movement of the electrons (in the normal way), hence such SMC are called n-type SMC.

As in n-type Si/Ge SMC, the electricity is conducted the smallest free nonrestrictive electrons while in p-type Si/Ge SMC, it is conducted the movements of finite sized positive holes whose free movement can be restricted by the neighbouring Si/Ge /dopant species, p- tpye Si SMC are expected to be inferior .

Now about their combination (n-and p- types of extrinsic SMC placed in JUNCTION): Three most commonapplications are:

[1] Formations of DIODE: Electrons have a negative charge. Holes have a positive charge. Electrons and holes are not static: they can move. Holes move more slowly than electrons. When electrons move in one direction, holes move in the opposite direction. The p-type SiCMC and n-type Ge SMC have different values of band gaps whose exacact value depends upon the mole% of the dopant. So the resultant moment of the electrical conduction will be will depend upon their difference and so will allow current only in one direction to form a Diode.

[2] Formation of TRANSISTORS: They are the devices that enable current stitching and are of two types:

[a] Bipolar junction transistors (BJT) are of NPN and PNP type. NPN transistors have two layers of n-type semiconductors sandwiching a p-type semiconductor. PNP transistors have two layers of p-type semiconductors sandwiching an n-type semiconductor.

[b] Field effect transistors (FET) are opposed to BJTs, they are unipolar and considered either N-channel or P-channel. FETs are divided into two families, junction gate (JFET) (JFET) and insulated gate(IGFET)

[3] The p-type Si and n-type Ge CMC are two wide gap SMC and form heterojunction structure which been found to enhance charge separation, leading to photocatalytic activity which can be attributed to the reverse movement of majority carriers beyond interface of two different SMC under UV illumination.This helps us to fabricate the new SMC to improve their photocatalytic activity.

In fact your question is a bit confusing. Generally if you consider pure semiconductor in ideal situation, where [e-]=[h+], then you are right it is confusing to know if it is n-type or p-type. But as we know, there exist NO ideal case, so at some stage [e-]>[h+] OR [e-]