**3.2. Morphology optimization by device engineering**

PCBM at roughly \$500-1000 per m2

368 Solar Cells - New Approaches and Reviews

**Figure 6.** Chemical structures of PC61BM and PC71BM.

**3.1. Characterization of morphology**

**3. Morphology and device engineering**

still cannot surpass that of fullerene based derivatives.

[42] makes the BHJ based PV technology with PCBM

problematic for commercialization. Use of technical grade PCBM (~80% PC61BM and 20% PC71BM) [115] might help relieve the stress but is far away from the desired price range. In fact, various types of small molecule based [116] or polymer based [117] acceptors have been tested to replace fullerene derivatives in OPV cells. But yet the efficiency of these acceptors

Along with the research effort to prepare D-A type conjugated polymers in BHJ solar cells to achieve the world record efficiency value, studies revealing the importance of the morphology of the polymer/fullerene blend have been carried out and the experience gained on controlling the morphology has become a valuable tool to explore the full potential of a new polymer as light harvesting materials. The thin film morphology characterization tools include grazing incidence wide-angle X-ray scattering, grazing incidence small angle X-ray scattering, resonant soft X-ray scattering, small-angle neutron scattering, transmission electron microscopy, atomic force microscopy, solid-state nuclear magnetic resonance, dynamic secondary ion mass spectrometry, near-edge X-ray absorption fine structure and scanning transmission X-ray microscopy. These analysis techniques are comprehensively summarized by Huang et al. [21]. As so much work has been done to investigate the morphology of the thin film, a rational question to ask is: *what is the best morphology?* Unfortunately, so far a precise answer to this question has not been achieved. One reason is because every characterization technique only sees the film from one aspect and a thorough mapping of the material distribution in the film still remains a challenge [118-119]. Another reason may be due to the fact that the reported A series of parameters that will influence the morphology of the polymer/fullerene thin film are listed in Figure 7. The physical properties of the polymer such as the π-backbone, sidechains, the molecular weight, the identity of the fullerene acceptors used and the mass ratio between the two etc., are factors related to the materials. The determination of the ideal D:A ratio for a new polymer material has been a matter of trial and error, with the ratio 1:1 to 1:4 most commonly used. The solubility of the polymer and the fullerene derivative should be sufficient in the processing solvent. Halogenated solvents (chloroform, chlorobenzene, 1,2 dichlorobenzene, 1,2,4-trichlorobenzene) are routinely used as they possess high solubility for both the donor polymer and the fullerene acceptors. For many reported polymers [20,21], judicious choice of the processing solvent has to be made in order to achieve the optimal morphology and power conversion efficiency. For a specific new polymer, or a new polymer/ fullerene blend, the choice of optimal processing solvent is not trivial, normally based on trialand-error investigations. Recently, more research has been focused on the replacement of halogenated processing solvents with more environmentally friendly solvents, such as toluene, xylenes and long alkanes [120]. This is especially important for the industrial pro‐ duction of OPV modules [121].

Processing additives with low vapor pressure and high boiling point are commonly added in the solvent to optimize the morphology of the polymer/fullerene blend [85]. Commonly used additives include 1,8-diiodooctane, 1,8-octanedithiol and 1-chloronaphthalene (Figure 8). General guideline for selection of the additive is that the solvent additive should be less volatile with higher boiling points than the host solvents. The mechanism how the additive influence the morphology and the overall performance of the device has been discussed by Lee et al. [122]. It should be noted that solvent additives added during the fabrication process might remain in the solar cell and behave as contaminants to hamper the performance of the solar cell. In some case, addition of additives shows no effect [123,124] while in some case the addition of additives is detrimental to the performance [125]. These results indicate that the solvent additive is not an elixir to enhance the PCE performance of all polymers and the detailed effect and mechanism should be made case by case. Moreover, a specific process step must be added to remove the residual deleterious compounds which are obviously unfavor‐ able in the industrial application. Hence, it is more desirable to design high performance polymer/fullerene system with no such additives needed to achieve the good performance. During the production of OPV modules, some other additives, such as rheology modifiers, anti-foaming agents and surface tension modifiers would be added into the ink formulation to make it more suitable for printing technology. The effect of these additives on the solar cell

**Figure 7.** Summary of parameters that influence the morphology of the BHJ active thin film.

performance has not yet been well studied. There are research attempts to incorporate various non-solvent additives, such as nanoparticles [126], carbon nanotubes [127], small molecules [128] and polymers [129], to create a ternary blend BHJ solar cell. These strategies have demonstrated how a third component in the polymer/fullerene blend influences the overall morphology and performance of the solar cell, which could be used as a potent routine method to enhance the performance of the device [130].

Thermal annealing is an alternative method for controlling the BHJ morphology and improv‐ ing the PCEs. By applying thermal energy to the thin film, it helps the reorganization of the polymer/fullerene blends and increases the crystallinity of the film [131]. This processing method has been routinely used to optimize the thin films for transistors [132] and BHJ solar cells [20,21]. The annealing temperature, the annealing duration and the cooling rate are key parameters to optimize the performance of the device and the optimal combination of the thermal treatment is material-dependent. Even for the same material, e.g., P3HT/PCBM blend, the optimal annealing condition differs from lab to lab [20], presumably due to different device structure, different solvent, different purity, different molecular weight, or even different operator. So far there is no general guideline to predict the optimal condition for a new OPV material and the optimal condition is obtained via a tedious trial-and-error approach.

**Figure 8.** Chemical structures of solvent additives for BHJ solar cells.

Solvent annealing refers to the treatment of the BHJ thin film with solvent vapor, typically in a petri dish contained with the solvent. Parameters to optimize include the solvent type and the treatment duration. By exposing the coated thin film with solvent vapors, there is reor‐ ganization and further morphological evolution over time. This method has been demonstrat‐ ed to enhance the morphology, the hole mobility of P3HT [133] and the PCE performance of the P3HT/PCBM solar cell [134]. Treatment of the thin film with polar solvents, such as methanol and ethanol, is also found to improve the morphology and render higher PCE values for P3HT/PCBM [135].

In summary, the morphology of the active polymer/fullerene blend in the BHJ solar cell can be tuned by a number of factors, such as the materials, choice of solvent, solvent additives, annealing condition, etc. The optimal condition to achieve the best device performance is typically material dependent and achieved in a trial-and-error approach. Sometimes, a minor modification of the processing condition can introduce magnificent enhancement of the device performance, for example, addition of processing additives. The complexity of the morphology control and the tedious optimization process would account for the phenomenon that why so many promising polymer materials in the literature with suitable absorption, energy level and solubility possess inferior device performance. Even for the same polymer motif, the OPV performance would vary significantly by different processing methods [136]. This again highlights the importance of device engineering work to explore the full potential of a new polymer in the BHJ solar cell.
