Organic semiconducting materials have become the cornerstone of organic electronics, including photovoltaic cells, light-emitting diodes, field effect transistors, and electrochromic devices. The synthesis of new organic semiconducting materials and the development of new synthetic methods for preparing semiconducting organic materials are two important issues that have attracted great attention. In this chapter, we mainly focus on the new chemistry for the synthesis of conjugated polymers used for organic photovoltaics.

Palladium-mediated cross-coupling reactions such as Suzuki–Miyaura, Sonogashira, Heck, and Stille reactions have been widely used in the synthesis of π-conjugated semiconducting materials. Recently, some new π-conjugated donor–acceptor type copolymers have shown great prospects for photovoltaic cell applications. Power conversion efficiencies (PCEs) above 9% have been achieved for bulk heterojunction (BHJ) polymer solar cells (PSCs).¹  Conjugated polymers synthesized by Heck and Sonogashira coupling reactions are seldom used for organic photovoltaic (OPV) applications. Most high efficiency conjugated polymers for OPV are synthesized by Stille and Suzuki cross-couplings. The classical synthetic routes toward donor–acceptor (D–A) type copolymers are palladium catalyzed AA/BB-type (hetero)aryl–(hetero)aryl cross-couplings of dihaloarylene monomers and suitably functionalized bifunctional aromatic counterparts, mostly arylene diboronic acids/diboronic esters (Suzuki-type coupling) or distannyl arylenes (Stille-type coupling). Recently, the C–H arylation cross-coupling reaction, the so-called direct arylation for the syntheses of π-conjugated polymers, has been reported.²  Direct C–H arylation has been expected as an alternative route to replace the widely used Stille and Suzuki reactions. Only limited examples of D–A type conjugated polymers synthesized by the direct C–H arylation have been reported as donor materials to achieve high PCE in PSCs. In this chapter, we summarize the synthetic methods for D–A conjugated polymers, namely, Stille, Suzuki–Miyaura, and direct C–H arylation polycondensation. In addition, D–A conjugated polymers synthesized by these methods and used for photovoltaic applications will be described.

Eaborn and Kosugi developed the first examples of cross-coupling reactions between organostannanes and electrophilic partners between 1976 and 1977.³  Soon the body of work was well known, when it became established as the title of the Stille coupling in 1978.⁴  Stille and co-workers reported the preparation of ketones from acyl chlorides and organostannanes by the use of palladium-catalyzed cross-coupling.⁵  Following this, the Stille reaction quickly took its place as one of the most useful protocols for forming sp² carbon–carbon bonds in organometallic chemistry. Yu and co-workers further developed this methodology in 1993 for use in polycondensation reactions for heteroaromatic diblock copolymers.⁶  They optimized reaction conditions and prepared high molecular weight copolymers.

The reaction mechanism itself is known to be far more complex and has been the subject of extensive work.⁷  The generally accepted process involves an oxidative addition step, a transmetalation step, and a reductive elimination step, as shown in Scheme 1.1. The Pd(0) species is the active catalyst. Thus, the whole reaction includes Pd(0)-mediated cross-coupling of organohalides, triflates, and acyl chlorides with organostannanes. The Pd(ii) catalysts used in Stille reactions are reduced to Pd(0) by the organostannane monomers, enabling entry into the catalytic cycle. The detailed catalytic cycle steps are as follows: (1) oxidative addition: the organohalide or triflate oxidatively adds to the Pd(0) active catalyst and forms a Pd(ii) intermediate [PdL₂R¹X] (L = ligand; R¹ = alkenyl, alkynyl, aryl; X = Br, I, Cl, or OTf); (2) transmetalation, which is generally regarded as the rate-determining step and is the most complex and thus has been the subject of much debate. It is generally accepted as a process of cleavage of the Sn–C bond by an electrophilic Pd(ii) complex and ligand substitution on a Pd(ii) complex; (3) reductive elimination, the final step in the process, which generates the desired product and allows the palladium catalyst to reenter the catalytic cycle.

The effects of catalytic systems on cross-coupling reactions have been extensively studied. Pd(PPh₃)₄ is the most commonly employed catalyst in the Stille reaction. For Pd(PPh₃)₄, ligand PPh₃ is easily oxidized by traces of oxygen in reaction system to its oxide, Ph₃PO.⁸  Excess PPh₃ can inhibit the Stille reaction process. Researchers have developed a more air-stable source of palladium-(0), [Pd₂(dba)₃], which has been widely used in Stille cross-couplings. Additionally, the other catalyst systems have also been widely used in the Stille reactions, such as benzyl(chloro)bis(triphenylphosphine) palladium-(ii), bis(acetonitrile)palladium(ii) dichloride, 1,1′-bis(diphenylphosphino)ferrocene palladium(ii) dichloride and allylpalladium(ii) chloride dimer. Meanwhile, the ligand in catalyst systems has been widely studied because it plays a critical role in the kinetics of the Stille reaction. Ligands such as (4-MeO-C₆H₄)₃P, PPh₃, tri(2-furyl)phosphine, and AsPh₃ were employed with Pd₂(dba)₃ as the catalyst precursors to optimize the reaction conditions.

Benzene, toluene, xylene, mesitylene, tetrahydrofuran (THF), N,N-dimethylformamide (DMF), N-methyl pyrrolidone (NMP), dimethyl sulfoxide (DMSO), dioxane, and chloroform are widely used solvents for Stille cross-coupling reactions. Toluene is generally a good choice for Stille polymerization carried out at temperatures above 120 °C. The higher boiling chlorobenzene can be used in some microwave-assisted Stille polycondensations owing to the need for higher temperatures, in excess of 200 °C. Highly polar solvents such as DMF and NMP can help solubilize the resulting polymer, which can function as a catalyst-stabilizing effect for the palladium center. Therefore, they can be used in mixed solvent systems with toluene or another cosolvent, for example toluene/DMF or toluene/NMP. For the mixed solvent systems, a high yield of high molecular weight polymers can be obtained.^6b  Solvents such as THF and dioxane can work as catalyst stabilizers and solvents for the resulting polymers.

Difunctional monomers, such as diorganohalides/ditriflates and distannanes, are generally used in Stille couplings. According to the reaction mechanism, diorganohalides or ditriflates carrying electron-withdrawing groups can facilitate the first oxidative addition step. For the second transmetalation step, the process may also be facilitated by organotin compounds with electron-rich properties. Thus, high molecular weight polymers can be prepared by electron-rich organotin compounds and an electron-deficient halide or triflate. Different halides exhibit discrepant reactivity. Generally, diiodo monomers show higher reactivity than dibromo monomers and dichloro monomers.

The major advantages of Stille polycondensation are the tolerance of many functional groups and the mild reaction conditions. These features are especially important for the synthesis of conjugated polymers bearing functional groups. The organotin monomers can be conveniently prepared, and they are far less sensitive to oxygen and moisture than many other organometallic compounds, e.g., Grignard reagents, organozinc and organolithium reagents.^7a  Stille polycondensation is broadly applicable to the synthesis of a wide variety of donor–acceptor conjugated polymers for OPVs, and provides a facile route to prepare high molecular weight, narrowly dispersed polymers under mild conditions.

First, the purification of many tin compounds is difficult because of their instability under silica gel column chromatography. Difficulties in the purification of monomers can pose a substantial problem in the synthesis of high molecular weight polymers, because the purity of monomers is crucial to achieving a precise stoichiometry between two monomers. Second, Stille couplings require the use of highly toxic organostannyls, which are not environmentally friendly. Generally, the organotin compounds are prepared using chlorotrimethylstannane or chlorotributylstannane; chlorotrimethylstannane is more reactive than chlorotributylstannane.⁹  Trimethyltin compounds are often more easily purified by recrystallization than their tributyltin counterparts, but the toxicity of chlorotrimethylstannane is 100 times greater than that of chlorotributylstannane. Moreover, the trimethyltin derivatives tend to be more volatile when they are exposed to air. The difficulty in purification and the high toxicity of organostannyls make the use of Stille cross-couplings an unwelcome choice.

The Stille polycondensation mainly involves the coupling reactions of ditin compounds with dihalide compounds and can be performed using conventional heating. Significantly, microwave irradiation was found to improve the number average molecular weight (M_n) and yield, and decrease the polydispersity index (PDI). This section will describe several examples in which D–A low bandgap conjugated polymers synthesized by Stille cross-couplings were used as donor materials in single junction bulk heterojunction polymer solar cells to achieve high PCE (>7%).

Liang et al. reported the synthesis of polymer P₁ by Stille polycondensation between a dibromo compound and ditin compound.¹⁰  Pd(PPh₃)₄ was used as the catalyst, toluene and DMF (4 : 1) were used as the solvent, and the polymerization was carried out at 120 °C for 12 h under N₂ protection. PTB7 was obtained with average molecular weight (M_w) of 97.5 kDa and a PDI of 2.1. A PCE over 7% was obtained for BHJ PSCs based on PTB7. Later, the PCE was increased to 9.2% after device optimization (Scheme 1.2).¹¹ 

P₂ was prepared by Wei et al. with Stille polymerization of a dibromo compound and ditin compound using [Pd₂dba₃/P(o-tolyl)₃] as the catalyst precursor.¹²  The M_n of 9.7 kg mol⁻¹ with a PDI of 1.4 was determined by gel permeation chromatography (GPC) using chloroform as the eluent. P₂ is readily soluble in hot chlorinated solvents such as chloroform, chlorobenzene, and dichlorobenzene. With the blends of P₂:PC₇₁ BM (1:1.5, w/w) as the active layer, a PCE of 7.3% was achieved using 1,6-diiodohexane (DIH) as the processing additive (Scheme 1.3).¹² 

Chu et al. reported a new alternating copolymer of dithienosilole and thienopyrrole-4,6-dione (P₃), which was synthesized by Stille coupling of a ditin compound and dibromo compound in refluxing toluene/dimethylformamide (10 : 1) with Pd(PPh₃)₄ as the catalyst.¹³  The purified polymer has a M_n of 28 kDa and a PDI of 1.6, as determined by GPC using chlorobenzene (CB) as the eluent. P₃ could be readily dissolved in chlorinated solvents even at room temperature. When blended with PC₇₁BM, P₃ exhibited a PCE of 7.3% (Scheme 1.4).

You et al. reported the first fluorinated D–A conjugated polymers applied in PSCs with an exceptional performance. P₄ was synthesized by Stille coupling of a ditin compound and dibromo compound.¹⁴  The polymerization was carried out at 120 °C for 20 min under microwave irradiation with Pd₂dba₃ and P(o-tolyl)₃ as the catalyst precursors and o-xylene as the solvent. P₄ was obtained in a yield of 89% with M_n of 33.8 kDa. A PCE of 7.2% was obtained for P₄ T:PC₆₁BM BHJ based PSCs (Scheme 1.5).

Fluorinated polymer P₅ was synthesized in a similar method by You et al.¹⁵  PSCs based on P₅ showed a PCE above 7% when blended with PC₆₁BM as the active layer, and a PCE above 6% is still maintained at an active layer thickness of 1 µm (Scheme 1.6).

The copolymer P₆ was obtained through Stille coupling polymerization of two monomers with a yield of 86%.¹⁶  P₆ has a M_n of 40.5 kDa with a PDI of 3.20. PSCs based on P₆ showed a PCE of 6.00% when blended with PC₇₁BM as the active layer. A PCE of 8.4% was achieved from the inverted PSC by using a PFN–Br interfacial layer to modify the ZnO electron extraction layer (Scheme 1.7).¹⁷ 

Amb et al. reported the synthesis of the first dithienogermole (DTG)-containing conjugated polymers by Stille polycondensation and their photovoltaic performance.¹⁸  When P₇:PC₇₀BM blends are utilized in inverted bulk heterojunction solar cells, the cells display a PCE of 7.3%. In inverted PSCs, when surface-modified ZnO–polymer nanocomposites were used as the electron-transporting layer, a PCE of 7.4% was achieved (Scheme 1.8).¹⁹ 

Huo et al. developed new poly[benzo(1,2-b:4,5-b′)dithiophene-co-thieno(3,4-b)thiophene] (PBDTTT) derivatives having the thienyl substituted [benzo(1,2-b:4,5-b′)dithiophene] (BDT) and the alkylcarbonyl-substituted thieno-[3,4-b]thiophene (TT-C).²⁰  The polymer P₈ was prepared by Stille coupling of the bis(trimethyltin) BDT monomers and the bromides (TTC) in a solvent mixture of toluene and DMF (5 : 1) with Pd(PPh₃)₄ as the catalyst. PSCs based on P₈ and PC₇₀BM reached a PCE of 7.59%. A PCE of 8.79% for a single-junction BHJ PSC was obtained with metallic nanoparticles (NPs) embedded in the active layer.²¹  The PCE was further increased to 9.13% in inverted PSCs by using 1,8-diiodooctane (DIO) as the processing additive (Scheme 1.9).

Xu et al. reported the synthesis of P₉ by Stille coupling of IDTT-di-Tin and diiodo-DFBT with Pd₂dba₃ and P(o-tol)₃ as catalyst precursors.²²  PSC devices based on P₉ showed an improved PCE of 7.03% without the use of any additives or post-solvent/thermal annealing processes (Scheme 1.10).

Wu et al. reported the synthesis of polymer P₁₀ through Stille coupling of the bis(trimethyltin) monomers and the dibromo monomers.²³  P₁₀:PC₇₁BM based PSCs exhibit a PCE above 7.79% without any further treatment such as the use of additives or annealing in device fabrication (Scheme 1.11).

Dong et al. prepared a copolymer P₁₁ by Stille coupling. The P₁₁-based device showed an impressive PCE of 7.11% in inverted PSCs (Scheme 1.12).²⁴ 

P₁₂ was synthesized by microwave-assisted Stille coupling using Pd₂dba₃ and P(o-tolyl)₃ as the catalyst precursor, and chlorobenzene (CB) as the solvent. PSCs based on P₁₂:PC₇₁BM blends afforded PCE up to 7.2% without thermal annealing or the use of processing additives (Scheme 1.13).²⁵ 

Son et al. reported the synthesis of polymer P₁₃ by Stille coupling using Pd(PPh₃)₄ as the catalyst and DMF and anhydrous toluene (1 : 4) as the reaction cosolvent. After optimization of the polymer’s solubility and morphological compatibility with PC₇₁BM, PSCs achieved a PCE of 7.6% (Scheme 1.14).²⁶ 

Liao et al. reported the synthesis of P₁₄ by Stille coupling with Pd(PPh₃)₄ as the catalyst in a solvent mixture of toluene and DMF (5 : 1). Fullerene derivative (PCBE–OH)-doped ZnO nanometer-thick film (40 nm) was used as the cathode interfacial layer for the effective collection of electrons. PSCs with the P₁₄:PC₇₁ BM active layer and the ZnO–C₆₀ modified cathode give a PCE of 9.35% (Scheme 1.15).¹ 

Wang et al. reported the synthesis of P₁₅ by Stille coupling with Pd(PPh₃)₄ as the catalyst. P₁₅ can dissolve in chloroform, toluene and 1,2,4-trichlorobenzene (TCB) at room temperature. The best device performance, with a relatively high PCE of 8.30%, was obtained for P₁₅:PC₇₁BM (1 : 1.5) (Scheme 1.16).²⁷ 

Li et al. prepared polymer P₁₆ using Stille coupling. A toluene/DMF (10 : 1, v/v) solvent mixture was used to obtain high molecular weight materials. PSCs based on P₁₆:PC₇₁BM furnished a PCE of 7.1% (Scheme 1.17).²⁸ 

Hou et al. prepared a new copolymer, P₁₇, with Stille coupling. PSCs based on P₁₇:PC₇₁BM (1 : 1.5, w/w) gave the best PCE of 8.07% (Scheme 1.18).²⁹ 

Chen et al. reported the synthesis of P₁₈ by Stille coupling.³⁰  With P₁₈ as the donor and PC₇₁BM as the acceptor in inverted PSCs, the highest PCE of 7.64% was achieved with an active layer 230 nm thick (Scheme 1.19).

Over the past two decades, Suzuki polycondensation has become one of the most efficient methods for the synthesis of conjugated polymers. As another important cross-coupling protocol, the Suzuki–Miyaura cross-coupling reaction was invented by Suzuki and co-workers in 1979.³¹  The scope of the Suzuki reaction for synthetic applications has been surveyed in several excellent reviews by Kotha, Lahiri, Kashinath, Miyaura and Fu.³²  The Suzuki–Miyaura cross-coupling reaction provided deeper insights into how to connect two specific sp²-hybridized C-atoms more efficiently and under milder conditions. The Suzuki–Miyaura cross-coupling reaction was first used by Schlueter et al. to prepare poly(para-phenylene)s.³³  Since then, Suzuki polycondensation (SPC) has become one of the most powerful and widely used methodologies for the synthesis of conjugated polymers.

The catalytic cycle of Suzuki coupling is thought to follow a sequence involving the oxidative addition of an aryl halide to a Pd(0) complex to form an arylpalladium(ii) halide intermediate, the transmetalation with a boronic acid, and reductive elimination of the resulting diarylpalladium complex to afford the corresponding biaryl and to regenerate the Pd(0) complex. Oxidative addition is often the rate-limiting step, and it is not surprising that the relative reactivity of aryl halides decreases in the order I > Br > Cl. The role of the base in these reactions is to facilitate the transmetalation of the boronic acid by forming a more reactive boronate species that can interact with the Pd center and transmetalate in an intramolecular fashion (Scheme 1.20, path A).³⁴  Alternatively, it has also been proposed that the base replaces the halide in the coordination sphere of the palladium complex and facilitates an intramolecular transmetalation (path B).³⁵  In fact, the exact nature of the actual catalyst remains ambiguous.³⁶ 

Suzuki polycondensation is generally believed to be of the step-growth type involving so-called AA/BB and AB approaches. In the AA/BB case, two different monomers are required, each of which carries either two boronic acids (or esters) or two leaving groups such as halogen or triflate. When two aromatic monomers are combined, polyarylene backbones which contain the two aromatic residues in an alternating fashion are obtained. In the AB case, the monomer carries both functional groups at the same time.

Almost all Suzuki polycondensations published to date in the literature use 1–3 mol% of catalyst, mostly Pd[P(p-tolyl)₃]₃, Pd(PPh₃)₄ or in situ prepared Pd[P(o-tolyl)₃]₂. Most catalysts for the Suzuki polycondensation employ triarylphosphine ligands. New ligands, which include Buchwald’s biaryl-based phosphines,³⁷  Beller’s diadamantyl phosphines,³⁸  Fu’s tri(tert-butyl)phosphine,³⁹  and Hartwig’s pentaphenylated ferrocenyl phosphines,⁴⁰  have been developed for Suzuki–Miyaura cross-coupling reactions. Buchwald-type ligand has been applied to polymerize dichloro monomers using Suzuki polycondensation.

Because of the hydrolytic deboronation of 2,5-thiophenebis(boronic acid pinacol ester) under standard Suzuki–Miyaura cross-coupling conditions, most attempts to synthesize thiophene-containing conjugated polymers from electron-rich 2,5-thiophene bis(boronic acid pinacol ester) and aryl dibromides by Suzuki polycondensation have failed to afford high molecular weight polymers. Bo designed and synthesized a new thiophene-containing phosphorous compound, L1, which was used as the ligand for a zerovalent palladium catalyst for Suzuki polycondensation of 2,5-thiophenebis(boronic acid pinacol ester) and aryl dihalides.⁴¹  High molecular weight thiophene-containing conjugated polymers were successfully synthesized for the first time from thiophene based diboronic acid ester monomers. The new catalytic system can enhance the SPC reaction rate and cope with steric hindrance imparted by the monomers. With the new ligand SPC proceeded very rapidly, and high molecular weight fourth generation dendronized polymers could be obtained in a short time. This method should be also of great interest for the synthesis of pharmaceutical and agrochemical compounds and natural products (Scheme 1.21).

The solvent systems will affect the progress of the polycondensation. Most Suzuki polycondensation are carried out in biphasic mixtures of organic solvents such as toluene, xylene, THF, or dioxane and an aqueous medium containing the base. The commonly used bases include K₃PO₄, K₂CO₃, NaHCO₃, KOH, KF, and sodium tert-butoxide. However, the choice of base is still empirical, and no general rule for their selection has been established at present. The other solvent systems, in particular homogeneous ones, have been less explored. Phase transfer catalysts (PTCs) such as tetraalkylammonium salts (tetraethylammonium hydroxide) have also been tried.

Aryl halides (bromides or iodides) and triflates substituted with electron-withdrawing groups are suitable substrates for the cross-coupling reaction. Aryl triflates and sulphonates are regarded as the synthetic equivalents of aryl halides. Triflates are, however, thermally labile, prone to hydrolysis and expensive to prepare. Aryl sulphonates are an attractive option because they are easily prepared from phenols, are more stable than triflates and are cheap and easily available starting materials. Suzuki polycondensations are carried out mainly between aryl (heteroaryl) halides and aryl (heteroaryl) boronic acids (esters). Similar to other step-growth polymerization, monomer purity is a key issue for Suzuki polycondensation, especially when the AA/BB approach is used. Boronic acids easily form partially and fully dehydrated products, which makes it difficult to reach the correct 1 : 1 stoichiometry. Alternatively, the corresponding cyclic boronic esters are widely used, because the commonly used boronic pinacol esters can be easily purified by silica gel column chromatography. Free boronic acids tend to be more reactive than their ester analogs. The solubility of boronic acids vs. esters in solvents also influences relative reactivity. The higher reactivity of the acids can be counteracted by their lower solubility.

The palladium catalyzed Suzuki–Miyaura coupling reaction is one of the most efficient methods for the construction of C–C bonds. The coupling can be carried out under mild reaction conditions, which will not be affected by the presence of water and heat. The reaction can tolerate a broad range of functionality and yield non-toxic byproducts. The boron-containing byproducts are easily separated from the reaction mixture and handled when compared with other organometallic reagents, especially in large-scale production. Additionally, the diverse boronic acids are commercially available and environmentally friendly. Consequently, the cross-coupling reaction has been realized in diverse applications, not only in academic laboratories but also in industry. These desirable features make the Suzuki–Miyaura reaction an important tool in medicinal chemistry as well as in the large-scale synthesis of pharmaceuticals and fine chemicals.

The Suzuki–Miyaura reaction suffers from a few key drawbacks, the first of which is the requirement for basic conditions. A number of monomers may be unstable in basic conditions, thus rendering this methodology impractical for these applications, or require more complex protection–deprotection strategies. Also, the Suzuki–Miyaura reaction requires a two-phase system; thus, polymers that rapidly decrease in solubility as molecular weight increases may form precipitate in poor yields or display very low molecular weights and high polydispersities under Suzuki–Miyaura conditions, which is disadvantageous for photovoltaic application.

We will provide a class of low bandgap polymers applied in BHJ PSCs that are synthesized by Suzuki polycondensation. This section cannot possibly cover all these different polymers. Emphasis will be placed upon important classes of conjugated polymers based on bridged phenylenes, for example poly(2,7-fluorene), poly(2,7-carbazole), and poly(2,7-dibenzosilole) based D–A conjugated polymers.

In 2003, Svensson et al. reported a low bandgap polymer, P₁₉, and its application in PSCs. P₁₉ was synthesized by Suzuki polycondensation of 2,7-fluorenediboronic acid pinacol ester and benzothiadiazole based dibromide using Pd(PPh₃)₄ as the catalyst and toluene and 20% aqueous tetraethylammonium hydroxide as the reaction media (Scheme 1.22).⁴² 

Similarly, Blouin et al. reported the synthesis of P₂₀ by Suzuki coupling of 2,7-carbazolediboronic acid pinacol ester and benzothiadiazole based dibromide using Pd₂dba₃ and P(o-tol)₃ as the catalyst precursors. PSCs based on P₂₀ showed a PCE of 3.6%.⁴³  After device optimization by Heeger et al., the PCE was enhanced to 6.1% (Scheme 1.23).⁴⁴ 

Bo et al. reported the synthesis of P₂₁ by Suzuki polycondensation of 2,7-carbazolediboronic acid pinacol ester and 5,6-bis(octyloxy)benzothiadiazole based dibromide using Pd(PPh₃)₄ as the catalyst precursor and a biphasic mixture of THF–toluene (5 : 1)/aqueous NaHCO₃ as the reaction medium. P₂₁ based PSCs exhibit a PCE of 5.4% (Scheme 1.24).⁴⁵ 

He et al. reported a copolymer, P₂₂, which was synthesized by Suzuki polycondensation. With P₂₂:PC₇₁BM (1:4) as the active layer, solar cells with a PFN/Al bilayer cathode displayed PCEs up to 6.07% (Scheme 1.25).⁴⁶ 

P₂₃ was first reported as donor material for PSCs by Leclerc et al. in 2007.⁴⁷  P₂₃ was prepared with a M_n of 15 kDa by Suzuki polycondensation of 2,7-silafluorenediboronic acid pinacol ester and benzothiadiazole based dibromide. Preliminary device experiments with this copolymer gave a PCE of 1.6%. In parallel, Cao et al. reported the synthesis of the same polymer with a high molecular weight (M_n = 79 kDa). The PCE of P₂₃ based PSCs was increased to 5.4% with a polymer:fullerene ratio of 1 : 2 (Scheme 1.26).⁴⁸ 

Bo et al. reported the synthesis of a series of D–A alternating conjugated polymers P_24–31 with 5,6-bis(octyloxy)benzothiadiazole as the acceptor unit by Suzuki polycondensation of dibromo monomers and 5,6-bis(octyloxy)benzothiadiazole based diboronic acid pinacol ester. P₂₄, with a M_n of 102 kg mol⁻¹ and a PDI of 1.66, gave a PCE of 5.08% in devices with P₂₄:PC₇₁BM as the active layer. When TiO_x was used as an electron-blocking layer, the PCE was further increased to 6.05%.⁴⁹  The 9-arylidene-9H-fluorene based D–A conjugated polymers P₂₅ and P₂₆ were synthesized to investigate the influence of alkyloxy position on the performance of PSCs. High molecular weight P₂₅ (HMW-P₂₅) and low molecular weight P₂₅ (LMW-P₂₅) were used to investigate the influence of molecular weight on the performance of PSCs. HMW-P₂₅:PC₇₁BM-based PSCs showed a PCE of 6.52%; LMW-P₂₅:PC₇₁BM based PSCs showed poor photovoltaic performance, with a PCE of only 2.75%; and P₂₆:PC₇₁BM based PSCs gave a PCE of only 2.51%.⁵⁰  Polymer solar cells with P₂₇:PC₇₁BM as the active layer demonstrate a PCE of 2.23% with a high open circuit voltage (V_OC) of 0.96 V (Schemes 1.27–1.30).⁵¹ 

P₂₈, with a M_n of 27.7 kg mol⁻¹ and a PDI of 3.1, gave a PCE of 4.48% when a P₂₈:PC₇₁BM blend (1 : 3, by weight) was used as the active layer.⁵²  P₂₉, with 3,6-difluorocarbazole as the donor unit, has a M_n of 9.1 kg mol⁻¹ and a PDI of 2.63. Polymer solar cells based on P₂₉ and PC₇₁BM demonstrate a PCE of 4.8%.⁵³  P₃₀, with 9-alkylidene-9H-fluorene as the donor unit and 5,6-bis(octyloxy)benzothiadiazole as the acceptor unit, is of planar structure. PSCs with a blend of P₃₀ and PC₇₁BM as the active layer demonstrate a PCE of 6.2%.⁵⁴  P₃₁, with spirobifluorene as the donor unit and 5,6-bis(octyloxy)benzothiadiazole as the acceptor unit, was synthesized and applied in PSCs. PSCs based on the blend films of P₃₁ and PC₇₁BM show a high open-circuit voltage of 0.94 V and a PCE of 4.6% without any post-treatment.⁵⁵ 

Conventional synthesis of the π-conjugated polymers mainly relies on transition metal-catalyzed cross-couplings, such as Stille and Suzuki couplings (vide supra). The preparation of organotin or organoboron monomers used for Stille or Suzuki coupling requires multistep reactions and tedious purification. Additionally, Stille coupling requires the use of very toxic reagents as well as generating toxic byproducts, which is not environmentally friendly. In particular, some important classes of heteroaryl organotin or organoboron reagents are not readily accessible and may even be too unstable to undergo the coupling processes, which may confine the variety of polymer libraries to some extent. In recent years, transition metal-catalyzed direct C–H arylation of non-preactivated arenes with aryl halides or pseudohalides, called “direct arylation”, has attracted much attention and worldwide interest. The direct arylations, with the advantages of synthetic simplicity and atom economy, without the use of troublesome and toxic (hetero)aryl organometallic intermediates, would be one of the most ideal routes toward π-conjugated polymers. These reactions are mostly developed for the synthesis of small molecules. Indeed, up until now, only a few publications have reported the use of conjugated polymers synthesized by direct arylation in PSCs.

In the last decade, Fagnou et al. detailed efficient coupling reactions between electron-deficient aromatic rings and arylhalides catalyzed by palladium complex in the presence of phosphine ligand and base, a synthetic reaction now termed “direct arylation”.⁵⁶  Recently, this reaction has been applied to the synthesis of conjugated polymers, for example regioregular poly(3-hexylthiophene) (P3HT), although it required rigorous heating in a THF solution at 120 °C and under high pressure.

The mechanism of C–H activation has been studied experimentally and computationally, and possible pathways include electrophilic aromatic substitution, Heck-type coupling and concerted metalation–deprotonation (CMD).⁵⁷  Most heterocycles, such as thiophenes and indoles, are believed to follow a base-assisted CMD pathway. Two catalytic cycles for a CMD coupling of bromobenzene and thiophene using a palladium/phosphine catalytic system and cesium carbonate are shown in Schemes 1.31 and 1.32. Scheme 1.31 depicts a carboxylate-mediated process, while Scheme 1.32 represents the reaction process without a carboxylate additive. In the presence of carboxylate as the additive (Scheme 1.31), first the carbon–halogen bond is formed by oxidative addition, and then complex 1 is formed by exchange of the halogen ligand for the carboxylate anion. Complex 1 then deprotonates the thiophene substrate while simultaneously forming a metal–carbon bond by going through transition state 1-TS. The phosphine ligands, or the solvent, can recoordinate to the metal center, following Pathway 1, or the carboxylate group can remain coordinated throughout the entire process (Pathway 2). Finally, reductive elimination renders the aryl coupled product. In the case of no carboxylate additive, after oxidative addition, the reaction follows one of the two pathways shown in Scheme 1.32. If a bidentate phosphine is employed, C–H activation of thiophene can follow Pathway 1, where deprotonation is assisted intermolecularly (2-TS). When a monodentate phosphine is used, the reaction may follow Pathway 1 or Pathway 2. The latter mechanism most closely resembles Pathway 2 in Scheme 1.31 where the carbonate coordinates to the metal center to give the zwitterionic species. Here, deprotonation occurs intramolecularly through transition state 1′-TS. Reductive elimination then renders 2-phenylthiophene.

In most polymerization examples, Pd(OAc)₂ was used as the palladium catalyst. Some research groups have exploited the highly stable Herrmann–Beller catalyst. Catalytic addition of pivalic acid can aid the C–H activation. Some groups have found that polymerization in the absence of phosphine ligand can give high molecular weight materials.

Polar (dimethylacetamide (DMAC), DMF, NMP, THF) and nonpolar (toluene) solvents are suitable for these reactions and should be selected according to the polymer solubility. DMAc is not a suitable solvent for polymerization and always gives brown soluble material. Addition of carboxylic acid may be beneficial in nonpolar solvents owing to the high polarity of the C–H bond transition states. However, C–H bond activations have been accomplished in toluene and xylenes without carboxylic acids.⁵⁸ 

Several electron-rich monomers and electron-deficient monomers can be used for direct arylation polymerization. Direct arylation of 2-halo-3-alkylthiophenes for the preparation of polymers (P3RT) was first published in the late 1990s.⁵⁹  Monomers and dimers can be used to synthesize poly(3,4-ethylenedioxythiophene) (PEDOT) and poly(3,4-propylenedioxythiophene) (PProDOT) by direct arylation polymerization with aryl bromides.^58a  Interestingly, functional groups were tolerant to direct arylation polymerization conditions and polymers were obtained in relatively low molecular weights (M_n = 3–4 kDa) when EDOT monomers were substituted with different functional groups.⁶⁰  In addition, 2,2′-bithiophene or fused bithiophene based monomers can undergo facile C–H bond activation.

Electron-deficient thiophene-based monomers including thieno[3,4-c]pyrrole-4,6-dione (TPD), furo[3,4-c]pyrrole-4,6-dione (FPD) monomers, diketopyrrolopyrrole and isoindigo can be copolymerized with brominated thiophene-based electron-rich monomers such as benzodithiophene, dithienosilole, and dithienogermole by direct arylation polymerization. Additionally, electron deficient 4,4′-dinonyl-2,2′-bithiazole and 1,2,4,5-tetrafluorobenzene are highly reactive toward direct arylation polymerization.⁶¹ 

Direct arylation should bring significant benefits in the synthesis of donor–acceptor polymers by decreasing the number of monomer preparation steps, producing only acid as a byproduct, and freeing of undesired waste from the metal-containing reagents. This is particularly significant because the traditional Suzuki and Stille couplings inherently possess a number of problems: synthesis of monomers is time-consuming, purifying the organotin monomers is difficult, and organotin compounds are highly toxic. Therefore, it is highly desirable to develop new direct arylation methodology that is environmentally friendly, scalable, and high yielding for mass production, to meet the requirement for synthesis of donor–acceptor polymers used for PSCs.

For it to be a reliable synthetic methodology with a wide application range, there are still several problems related to direct arylation polycondensation reactions that need to be solved. First, the low molecular weight of the polymer obtained is disadvantageous for the photovoltaic application. Second, the poor selectivity of the C–H bond in direct arylation will resulted in branched and cross-linked polymer structures that affect the solubility and optoelectronic properties of polymers.

The first example of synthesis of poly(3-alkylthiophene)s by direct arylation was reported in 1999.⁵⁹  Recently, the syntheses of high molecular weight poly(fluorene-alt-tetrafluorobenzene), poly(fluorene-alt-dithiophene), and poly(thienopyrroledione-alt-dithiophene) by direct arylation were published.^61b,62  Here, we provide a few examples to show the use of direct arylation to synthesize low bandgap conjugated polymers that are promising donor materials for PSCs.

A series of DTDPP-based copolymers with an alternating D–A sequence and homopolymer P₃₂ were synthesized by the direct arylation polycondensation. The polymer was endowed with the common features of excellent π-conjugation and ideal planarity, which lead to remarkably low band gaps (1.22 eV). Photovoltaic properties of these polymers were not reported (Scheme 1.33).⁶³ 

P₃₃, with a higher molecular weight, up to 70 kDa, was synthesized via palladium catalyzed direct arylation between dithiophene derivative and dibromo compound with K₂CO₃, Pd(OAc)₂ (10 mol%) and pivalic acid (30 mol%) in a small volume of N-methylpyrrolidone (NMP) in a 10 mL Schlenk tube. For the first time, OPV characteristics of polymers synthesized via direct arylation were compared to those synthesized via Suzuki coupling. A blend ratio P₃₃:PC₆₁BM of 1 : 3 was used to fabricate OPV devices and give the highest PCE of 2.24%, which represents a moderate enhancement in performance obtained via Suzuki-coupled polymer (2.01%) (Scheme 1.34).⁶⁴ 

Direct arylation polycondensation using the phosphine-free catalytic system can be adapted to the synthesis of bithiazole-based alternating copolymers (P₃₄). In comparison with conventional polycondensation via other cross-coupling reactions, the polycondensation proceeded with a reduced amount of Pd catalyst (2 mol%) in a short reaction time (10 min to 3 h). Owing to the difference in reactivity of the C–H bond, controlling the reaction time was effective for suppressing the side reaction at the unexpected C–H bond (Scheme 1.35).⁶⁵ 

High molecular weight P₃₅, using nonactivated dithiophene derivative and dibromo derivative as coupling monomers, was obtained in good yields (up to 80%) with a M_n of up to 40 kDa (Scheme 1.36).⁶⁶ 

Wang et al. reported that direct arylation polycondensation of 2-bromo-3-hexylthiophene using Herrmann’s catalyst and a triarylphoshine ligand yielded regioregular poly(3-hexylthiophene) with a high molecular weight.⁶⁷  This catalytic system also can be used to direct arylation polycondensation of a thieno[3,4-c]pyrrole-4,6-dione derivative with a dibromobithiophene derivative, to give D–A polymer P₃₆. Photovoltaic characterization was not tried (Scheme 1.37).^62b 

A 2,2′-bithiophene-based monomer has more than one type of reactive C–H bond, although reactions typically occur in the 5- and 5′-positions first. Attempts to copolymerize 2,2-bithiophene with dibromo compound rendered materials with low solubility due to cross-linking in the 3,3′- and 4,4′-positions of the bithiophene monomer. To evade these side reactions, the other reactive C–H bonds on the thiophene monomers were blocked with methyl groups. P₃₇ and P₃₈ were synthesized by direct heteroarylation polymerization. However, the film absorption onsets were hypsochromically shifted compared to the “unprotected” bithiophene analogs. The methyl groups indeed cause twisting in the polymer backbone and disrupt conjugation. Although protection of the 3,3′,4,4′-positions of bithiophene circumvented cross-linking on this monomer, the planarity of the copolymers was disrupted and optimal packing properties diminished (Scheme 1.38).⁶⁸ 

A D–A conjugated polymer, P₃₉, with a M_n of 41 kDa was synthesized by Heeger et al. using a direct heteroarylation procedure. A relatively high PCE of over 6% was obtained from BHJ solar cells based on a photoactive film comprising a composite of P₃₉ and PC₇₁BM. This is the best result for polymers prepared via the direct heteroarylation method and used for BHJ solar cells (Scheme 1.39).⁶⁹ 

A selenophene–TPD copolymer, P₄₀, with a M_n of 36 kDa and a PDI of 1.97 was synthesized using the direct heteroarylation polymerization in a 94% yield. PSCs with the blend P₄₀ and PC₇₁BM as the active layer gave a PCE of about 5.8% (Scheme 1.40).⁷⁰