**2.1. Chromophores with new π-structure**

In the concept of BHJ solar cells, the active layer is comprised of a blend of electron-donating material and electron-withdrawing material. The electron-donating material can be small molecules [12, 13] or polymeric materials [10,11,14-19] while polymeric materials are more commonly used in the literature. Both types of materials typically follow a Donor-Acceptor (D-A) design principle, in which the conjugated backbone is constructed by covalent linkage of a series of electron rich moieties and electron deficient moieties in an alternative way. This design strategy is especially useful in tuning the physical properties (absorption, frontier orbital energy levels, etc.) of the final materials due to the vast stock of electron rich and electron deficient building blocks. Guo and co-workers [44] have presented a thorough summary of current prevailing donor and acceptor species, which include about 45 donor and 60 acceptor backbones. Note that there are also variations on the solubilizing chains and spacer groups. Hence the actual number of such building blocks would be much larger than the summarized numbers. Given the wide choice of building blocks, there is no surprise that a huge structural diversity of OPV materials exists and a large amount of new materials are coming out every year in the literature. The design principle of the donor material in the BHJ blend has been summarized [45]. Basic considerations include light absorption range, frontier orbital energy levels, charge carrier mobility, favorable blend morphology, stability and solubility.

Concurrently new building blocks are being synthesized to provide more possibilities to further optimize the optical and electronic properties of the final material. Take benzothiadia‐ zole (BT) as an example (Figure 2). Benzothiadiazole (BT) is one of the most commonly used building blocks for the construction of D-A type conjugated polymers. By carrying out structural modification of the backbone, a series of new electron-deficient moieties with different electronic properties can be prepared. One strategy is to replace the sulfur atom in the thiadiazole hetero-cycle with other elements such as carbon [46], oxygen [47,48], nitrogen [49], selenium [50,51], etc. The new building blocks have various electron withdrawing properties and hence are useful in tuning the properties of the polymer materials. The second strategy is to introduce substitutions on the BT unit, such as fluorine atoms [52-60], alkoxy groups [61-65], or replacement of C-H with imine nitrogen [66-69]. More building blocks can be prepared with the combination of these two strategies [70-75]. The third strategy is to extend the π-conjugation of the BT unit to prepare π-extended moieties [76-81]. Extension of the πbackbone is a versatile route to tune the electronic properties of the BT unit; however, the new building blocks typically exhibit poorer solubility and have to be prepared in longer synthetic steps. For example, by fusion of one more thiadiazole ring to the BT unit, bis-benzothiadiazole (BBT) can be prepared and possess much higher electron deficiency but poorer solubility.

**Figure 2.** Evolution of benzothiadiazole (BT) based electron withdrawing moieties for the preparation of D-A type light absorbing polymers.

Out of all these strategies, the substitution of fluorine atoms on the aromatic backbone turns out to be a very efficient approach to achieve high-performance OPV materials. Introduction of fluorine substitutions has minor influence on the absorption behavior of the polymer; however, it induces a decrease of the frontier orbital energy levels due to its strong electrone‐ gativity and consequently the fluorinated polymers typically exhibit higher open circuit voltage (*V*oc). The advantages of fluorine in OPV polymer have been demonstrated by Zhou et al. [82] Polymer **PBnDT-DTBT** (Figure 3) exhibits a HOMO and LUMO energy level at-5.20 eV and-2.92 eV, respectively. After addition of two fluorine atoms on the BT unit, the HOMO and LUMO energy level of **PBnDT-DTffBT** decreases to-5.30 eV and-2.97 eV, respectively. **PBnDT-DTBT**/PC61BM based solar cell device exhibits the best PCE=5.0% with *J*sc=10.03 mA cm-2, *V*oc=0.87 V and FF=0.57. For **PBnDT-DTffBT**, the best device exhibits PCE=7.2% with *J*sc=12.91 mA cm-2, *V*oc=0.91 V and FF=0.61. It is found that after attachment of the fluorine atoms on the repeating unit, the short circuit current, the open circuit voltage and the factor are all enhanced.

**Figure 3.** Chemical structures of **PBnDT-DTBT** and **PBnDT-DTffBT**.

Concurrently new building blocks are being synthesized to provide more possibilities to further optimize the optical and electronic properties of the final material. Take benzothiadia‐ zole (BT) as an example (Figure 2). Benzothiadiazole (BT) is one of the most commonly used building blocks for the construction of D-A type conjugated polymers. By carrying out structural modification of the backbone, a series of new electron-deficient moieties with different electronic properties can be prepared. One strategy is to replace the sulfur atom in the thiadiazole hetero-cycle with other elements such as carbon [46], oxygen [47,48], nitrogen [49], selenium [50,51], etc. The new building blocks have various electron withdrawing properties and hence are useful in tuning the properties of the polymer materials. The second strategy is to introduce substitutions on the BT unit, such as fluorine atoms [52-60], alkoxy groups [61-65], or replacement of C-H with imine nitrogen [66-69]. More building blocks can be prepared with the combination of these two strategies [70-75]. The third strategy is to extend the π-conjugation of the BT unit to prepare π-extended moieties [76-81]. Extension of the πbackbone is a versatile route to tune the electronic properties of the BT unit; however, the new building blocks typically exhibit poorer solubility and have to be prepared in longer synthetic steps. For example, by fusion of one more thiadiazole ring to the BT unit, bis-benzothiadiazole (BBT) can be prepared and possess much higher electron deficiency but poorer solubility.

> N N N

N S N

F F

R

N N N

N O N N Se N N Te N

<sup>S</sup> <sup>N</sup> <sup>N</sup>

RO OR

N S N

> N S N

**Figure 2.** Evolution of benzothiadiazole (BT) based electron withdrawing moieties for the preparation of D-A type

Out of all these strategies, the substitution of fluorine atoms on the aromatic backbone turns out to be a very efficient approach to achieve high-performance OPV materials. Introduction of fluorine substitutions has minor influence on the absorption behavior of the polymer; however, it induces a decrease of the frontier orbital energy levels due to its strong electrone‐

N S N

N

N N R R

N

N N N

N N N

N O N

RO OR

F

R

R

F F

N S N

F OR

N S N N S N

R

S N

NN

F

N S N

> N S N

N S N

N S N

362 Solar Cells - New Approaches and Reviews

H H

light absorbing polymers.

The longer synthesis steps for the more complicated building blocks are also a concern if they are to attract industrial attention. Although achieving materials in a cost-effective way is generally neglected in the academic publication and in many cases the complicated synthesis of a monomer is considered as *novelty* of the work, a simple, high yield and easily scalable synthesis of materials is highly desirable in the industry from the application point of view. For example, 4,7-dibromobenzo[c][1,2,5]thiadiazole can be prepared in three steps from 1,2 phenylenediamine (Scheme 1). As this building block is so commonly used nowadays, it has become commercially available by vendors such as Sigma Aldrich. While for alkylated 4,8 dibromo-[1,2,5]thiadiazole[3,4-f]benzotriazole [83], which is a BT unit fused with a triazole hetero-cycle, is prepared with four more synthetic steps from 4,7-dibromobenzo[c] [1,2,5]thiadiazole. According to a recent estimation [84], the cost per gram of the final material increases linearly with the number of synthetic steps needed for the synthesis. The extended synthesis would definitely reduce the potential applicability of the material, as the value of a material is a compromise between the performance and the cost.

A list of D-A type polymers that have demonstrated PCE values > 7% are shown in Figure 4 [85]. It should be highlighted that the high performance of the polymers does not necessarily mean that the embedded building blocks are superior. The power conversion efficiency is determined by a number of factors and the chemical structure of the polymer is just one of them. Even for a classical polymer P3HT, after careful optimization of the device condition, the PCE can also reach 7.4% [86]. Many other factors, like the fabrication conditions, also play

**Sheme 1.** Synthetic routes of 4,7-dibromobenzo[c][1,2,5]thiadiazole and alkylated 4,8-dibromo-[1,2,5]thiadiazole[3,4 f]benzotriazole.

a significant role in determining the overall efficiency of the cell. These factors will be discussed in the following text.

**Figure 4.** Chemical structures of polymers that exhibit PCE > 7%.
