You have take a look at the energy level diagram of the materials involved.

For any material to be a perfect HTL (or EBL), the energy barrier between the conduction band of the HTL layer and LUMO of the donor should be higher (this will block the electrons from donor to reach the electrode). At the same time, the energy barrier between the valence band and HOMO of donor should be lower (this will help the holes to pass from donor to electrode) 

To read in details please refer to: doi:10.3390/electronics3010132

Dear Waranaha,

I would like to add to the answer of the research gate college Ghosh. the energy differences between the conduction bands and the valence band of two contacting materials are a necessary but not sufficient condition  to act as an extractor of e.g. holes as in the case of  PEDOT:PSS. The second condition is that the material must easily transport the holes, which means it must be good conductor of holes to deliver them to the metallic electrode. To demonstrate the argument assume the material satisfies the energy difference condition but it is insulating,it will not function properly. So, the thoughts of Waranatha are okay. 

wish you success

I think the above two comments have more or less answered the question. PEDOT:PSS is preferred due to several reasons. When there are no other additives (such as IPA, DMSO) PEDOT:PSS has a sufficiently high work function to form and Ohmic contact to P3HT enabling hole extraction. The energy level difference between the LUMO of PCBM and the PEDOT work function is too high resulting in a barrier for charge extraction. I am not sure as to whether PEDOT:PSS can be considered as a preferential hole transporter. placed between two metal contacts such as Au, the transport should be due to electrons. This is also seen in the case of ITO where it can act either as the anode or cathode contact depending on the nature of the layer in contact with it. So, the reason for using PEDOT:PSS as an HTL is more likely due to its work function. 

Thank you all for your detailed responses.

Summarizing what you have said, it is mainly the work function alignment with energy levels that make the PEDOT:PSS preferential for extracting holes and block the electrons. Not so much the preferential transport property such as a higher hole mobility than electron mobility. In this case, with proper alignment of energy levels, PEDOT:PSS could also act as a electron extracting layer? Does anybody know such an example?

Thanks.

Yes, provided you can have the proper energy alignment, you can have electron extraction through PEDOT. For example, the following work:

http://www.sciencedirect.com/science/article/pii/S1566119914004182

Suggests that adding sufficient amount of IPA to PEDOT lwoers the work function so that it now extracts electrons from PCBM. An interesting material to note is MoO3 (its really MoOx when evapourated) (http://pubs.acs.org/doi/abs/10.1021/nl2015729) which is regarded as an n type material but used as a hole extraction layer in OPVs.

I would like to stress generally that the material must be able to conduct the extracted holes across it. Here i am not comparing its ability to conduct electrons or holes since it rejects the electron in the given structure.I am speaking about concept  It is so that good collector of  charges must have accelerating potential gradient and good conduction. Other wise it will at least act a relatively high resistance causing losses in the generated electrical energy of he solar cell.Specifically, it causes the decrease in he fill factor of the cell.

In PEDOT, only holes contribute to the charge transport. The electrons will immediately recombine with oxidized states of PEDOT layer (page 147, PEDOT: Principles and Applications of an Intrinsically Conductive Polymer ). 

For Electron blocking layer/or Hole transport layer, we normally use materials having very high LUMO energy level than the LUMO energy level of electron donor material. The property of electrons is that they always move from higher to lower energy level. The LUMO energy level of PEDOT:PSS is very high around -2.2 eV compared to LUMO energy level of any other electron donor materials. Due to this property of PEDOT:PSS electrons cannot surpass form lower to higher energy level and will be blocked.

PEDOT:PSS is a p-type semiconducting material having high band gap around 2.8 to 3.1 eV which makes it very bad for absorption and make it very good for transmittance of incoming solar photons to the active layer.

The holes always tries to move from lower to higher energy level. PEDOT:PSS has HOMO energy level of around -5.0 eV, which is much higher than the most of the widely available electron donor materials and hence making it very good for hole extraction.

Yes, it is the HOMO-LUMO alignment of PEDOT:PSS with that of the HOMO-LUMO of the light absorbing material (e.g perovskite). For a substance to act as hole transport material, the valence band (HOMO) of the HTM must be higher than the valence band of the light absorbing material. This enables the HTM to act as electron blocking layer, being poor in electron affinity or extraction. On the other hand, for a material to act as HTM, it's conduction band (LUMO) must be higher than the conduction band of the perovskite or light absorbing material. To change this alignment so that PEDOT:PSS becomes ETM is a great task that should not be thought of or embarked on. Instead, use existing ETM such as TiO2 or PCBM.