Chinese Chemical Letters  2016, Vol. 27 Issue (08): 1250-1258 PDF
Heavy metal complex containing organic/polymer materials for bulk-heterojunction photovoltaic devices
Liu Ya-Nan, Wang Shi-Fan, Tao You-Tian, Huang Wei
Abstract: The application of heavy-metal complexes in bulk-heterojunction (BHJ) solar cells is a promising new research field which has attracted increasing attention, due to their strong spin-orbit coupling for efficient singlet to triplet intersystem crossing. This review article focuses on recent advances of heavy metal complex containing organic and polymer materials as photovoltaic donors in BHJ solar cells. Platinum-acetylide containing oligomersor and polymers have been firstly illustrated due to the good solubility, square planar structure, as well as the fairly strong Pt-Pt interaction. Then the cyclometalated Pt or Ir complex containing conjugated oligomers and polymers are presented in which the triplet organometallic compounds are embedded into the organic/polymer backbone either through cyclometalated main ligand or the auxiliary ligand. Pure triplet small molecular cyclometalated Ir complex are also briefly introduced. Besides the chemical modification, physical doping of cyclometalated heavy metal complexes as additives into the photovoltaic active layers is finally demonstrated.
Key words: Platinum complex     Iridium complex     Bulk-heterojunction     Photovoltaic devices     Organic/polymer solar cells     Triplet
1. Introduction

Over the past few years, organic solar cells (OSCs) have attracted intense attention as renewable energy sources in both academia and industry, owing to their remarkable features such as flexibility, low cost, light weight, and large area mass production [1-5]. To date, the most efficient device structure for OSCs is utilized BHJ-type as the active layer [6-11], which consists of an interpenetrating blends of $\pi$-conjugated small molecules or polymers as electron donor (D) materials and typically fullerene derivatives [6, 6]-phenyl-C61-butyric acid methyl ester (PC61BM) (Fig. 1) as electron acceptor (A) materials, providing efficient charge separation of the photogenerated excitons as well as favorable charge transport at the D/A interfaces [12-14]. The most common donor material in BHJ solar cells is regioregular poly(3- hexylthiophene) (P3HT) (as shown in Fig. 1), which can be used to achieve a power conversion efficiency (PCE) over 6% [15]. To acquire high performance BHJ solar cells, one of the most successful methods is to modify the chemical structure of the donor materials to achieve the desired electronic properties. Although a large amount of new photovoltaic donor materials have been developed, small molecular OSCs with PCE over 9% have been rarely reported [16, 17]; on the other hand, only a few polymers show PCE exceeding 10% in single-junction [18-20] and ~12% in multi-junction BHJ polymer solar cell devices [21-25]. In spite of the aforementioned success, after decades development of extraordinarily large number of organic/polymer donor materials, the possibility of further enhancements through chemical modification of designing new materials becomes a bottleneck [26, 27]. Therefore, from the materials point of view, explore of new synthetic strategies affording extra approach to break through the PCE is particularly important. Since the mainstream of the previously reported organic/polymer materials all mainly based on singlet dominated pure organic materials, the introduction of organometallic heavy metal component with intersystem crossing is expected to better utilize the triplet excitons in devices to further improve the device performance. However, triplet concerning materials in organic photovoltaics (OPVs) have not been well investigated.

In general, the mechanism of a typical BHJ solar cell involves four steps: light absorption and exciton generation at the photoactive materials, exciton diffusion to D/A interface, charge separation at D/ A interface, and charge transport and collection [28]. Therefore, there are many aspects that can influence the performance of BHJ solar cells on the basis of the driving mechanism. Among, the average exciton diffusion length and charge separation are themost common factors to impact the power conversion efficiency. However, on one hand, the diffusion length (5-10 nm) [29, 30] of singlet excitons in the most majority photovoltaic materials is usually significantly shorter than the absorption depth required for efficient light absorption. To increase the exciton diffusion length, the introduction of organicmaterials with high mobility and/or long lifetime is allowed [31, 32]. On the other hand, in order to achieve high efficiency, eliminating geminate recombination of charge pairs that are formed by charge transfer (CT) at the donor-acceptor interface, which is competitive with charge separation should also be taken into consideration [33-36]. Besides, a major loss mechanism reported to limit the PCE of polymer solar cells is the "back reaction" in which the reverse energy transfer occurred from the photo-generated triplet charge transfer state (3CT) to the triplet state (1T) of donor material when the triplet energy level values of CT state was lower than that of donor polymer [33, 37]. Therefore, to reduce geminate charge recombination and facilitate charge separation, one possible way is to efficiently utilize the 3CT, as it is demonstrated that under strong magnetic field, the dark current of organic photovoltaic (OPV) devices could be increased by 45% due to the formation of triplet CT state [38]. In addition, 3CT with sufficiently long lifetime (micro-second scale or even longer) could increase the exciton diffusion lengths, making CT state more easier to separate and dissociate into free charges instead of recombination compared with the relatively shorter nano-second scale 1CT state [31, 33, 39, 40]. Thus, the formation ofmore 3CT based on heavy metal complexes is beneficial to increase the current and improve photovoltaic performance [31, 39, 40].

In the past decades, organometallic complexes have attracted extensive interests, owing to their specific interesting optoelectronic properties as the advanced characteristic to be widely applied as emitters in light-emitting devices [41, 42], bioimaging [43, 44] and sensitizers in dye-sensized solar cells [45, 46]. Whereas, organometallic complexes have been less investigated in organic thin film photovoltaics, and the initial studies have been focused on the donor/acceptor (p/n type) bilayer heterojunction device. For example, Tang et al. fabricated the first organic photovoltaic cell with an conversion efficiency of 1% in a two-layer-heterojuction device by using phthalocyanine (CuPc) (Fig. 2) as the p-type active material [47]. In a recent development, Roy and co-workers reported the synthesis of tetramethyl-substituted Cu(II) phthalocyanine (CuMePc) (Fig. 2) nanocrystals and BHJ ternary solar cells based on CuMePc:P3HT:PCBM, displayed excellent photovoltaic performance with a conversion efficiency of 5.3% [48]. Besides, Shao and Yang demonstrateda simple multilayer heterojunction photovoltaic device with the configuration of ITO/PEDOT/ PtOEP(30 nm)/C60(30 nm)/BCP(8 nm)/Al(100 nm) [31]. The work based on typical triplet heavy metal complex 2, 3, 7, 8, 12, 13, 17, 18- octaethyl-21H, 23H-porphineplatinum(II) (PtOEP) as p-type donor material, achieved relatively high performance with 2.1% of power conversion efficiency. Furthermore, heavy metal based cyclometalated Ir complex mer-bis(4'6'-difluorophenylpyridinato-N, C2') iridium(III) azaperylene (APIr) was also applied as the electron donor in a bilayer heterojunction device with a structure of ITO/ APIr(5 nm)/C60(30 nm)/N, N0-dihexyl-perylene-3, 4, 9, 10-bis(dicarboximide) (PTCDI)(10 nm)/BCP(14 nm)/Al [49]. A fairly good PCE of 2.8% and considerably high open-circuit voltage (VOC) up to 1.0 V was obtained.

 Download: larger image Figure 2. Molecular structures of CuPc, CuMePc, PtOEP and APIr

In this review, recent progress on the development and application of heavy-metal complex containing photovoltaic donor materials for the bulk-heterojunction solar cells are presented. Low-bandgap platinum-acetylide [50-70] and the cyclometalated Pt or Ir containing polymers, oligomersor small molecules [71-77] as donor materials for the active layer of BHJ solar cells are successively demonstrated, in which the heavy metal complex was chemically bonded to the molecular backbone. Finally, physically doping of heavy-metal complexes as additives into the BHJ solar cells to enhance the photovoltaic performance are also discussed [78-81].

2. Heavy metal complex involved through chemical bonding 2.1. Platinum-acetylide containing polymers and oligomers

Platinum-acetylide containing derivatives is reported to be one of the effective p-type materials for organic solar cells, due to the square planar structure of Pt complex and fairly strong Pt-Pt interaction as well as excellent optoelectronic properties which is favorable to the performance of the photovoltaic devices [54, 82]. The most promising and widely used platinum alkynyls have been a popular candidate for application into $\pi$-conjugated organometallic polymeric backbone [50-70]. The $\pi$-electron delocalization and the intra-chain charge transport along the entire polymer backbone is facilitated to be enhanced due to the d-orbitals (dxy and dxz) of the Pt atom overlaps with the p-orbitals ($\pi$y* and $\pi$z*) of the C≡C unit to construct the one dimensional conjugated polymer chain, when the alkyne is coordinated with the electron-rich Pt(II) ions [83]. Additionally, the presence of bulky PBu3 ligands which prevents strong aggregation improves the solubility of platinum-containing polyynes in organic solvents. Furthermore, by introduction of donor-acceptor (D-A) architecture to the platinum centered molecular backbone, broad absorption bands (even up to near-infrared region) attributed from the intramolecular charge transfer (ICT) between the electron donating and accepting units for small bandgaps are allowed to be achieved, making this kind of materials suitable for photovoltaic devices [84]. Therefore, the emerging use of low-bandgap soluble platinum-acetylide polymers and oligomers, which represent potential candidates for light weight solar energy conversion donor materials in OPVs, providing a new and versatile avenue to harvest sunlight for efficient solar power generation.

In 2006, Schanze and co-workers demonstrated the first report of a polymer-based photovoltaic device which involved a triplet excited state in the process of photoinduced charge separation [53]. Through blending a blue-violet absorbing platinum-acetylide polymer p-PtTh (polymer 1 as shown in Fig. 3) with PCBM to serve as the active layer in polymer photovoltaic device, a PCE of 0.27% was achieved. Due to the introduction of heavy metal Pt center in the platinum-acetylide, spin-orbit coupling enhanced, and thus rapid and efficient intersystem crossing from singlet to triplet occurred. The results from photophysical measurements provided the evidence that the charge carriers were produced in a photoinduced electron transfer (PET) process involving the triplet excited state of p-PtTh, which contributed to the device efficiency. The rather low efficiency could be attributed to the relatively wide bandgaps, and possibly unfavourable energy levels as well as poor charge-transport properties when blended with PCBM.

Then in 2007, Wong et al. reported 4, 7-di-2'-thienyl-2, 1, 3- benzothiadiazole (TBT) containing $\pi$-conjugated platinum-acetylide containing polymer 2 (Fig. 3) [54]. The novel platinum metallopolyyne is based on planar conjugated segments with extended $\pi$-electron delocalized system throughout the polymer chain. Internal D-A functions between the electron-rich Pt-ethynyl groups and the electron-deficient benzothiadiazole rings in combination with two electron-donating thiophenes allowed a lowbandgap of ~1.85 eV with extended absorption that features unique D-A characteristics of thematerial. BHJ polymer solar cells based on metallopolyyne/PCBM blends as the photoactive layer, showed an average PCE of 4.1%. The significantly improved efficiency clearly demonstrated the potential of metallated conjugated polymers for efficient photovoltaic devices. Wong group also presented four pconjugated platinum(II) polyynes (3-6) containing bithiazole-oligo (thienyl) rings with solution-processable and strongly visible-light absorption [56]. By varying the oligothienyl chain length in the polymer backbone, optical and charge transport properties aswell as the polymer solar cell efficiency could be tuned. Through gradually increasing the number of thienyl ring, the polymer bandgap reduced from 2.40 eV to 2.06 eV, thus the corresponding light-harvesting capacity enhanced and the intrachain charge carrier mobility improved. Therefore, in the polymer/PCBM blended systems, both polymer 5 and 6 exhibited carrier mobility on the orders of 10-4 cm2 V-1 s-1 and comparable photovoltaic efficiencies, with high PCE of up to 2.7% and peak external quantumefficiency (EQE) of 83%for polymer 5 and PCE of~2.5%for 6.Unfortunately, heavymetal effect responsible for the intersystem crossing was reduced by the highly extended heteroaryl rings in the ligand. Therefore, it is described that the charge-transfer excited state mostly contributed to the efficient photoinduced charge separation in the energy conversion for 3-6 insteadof the triplet state. Thiswas different from the monothiophene based Pt-polyyne 1 blended films in which charge separation occurs via the triplet excited state of polymer platinum-acetylide.

Similar molecular design strategies have also been employed to synthesize a series of $\pi$-conjugated organometallic platinumcontaining polymetallayne polymers which were functionalized with fluorine-oligothienylring (7-10), phenothiazine-oligothiophene (11-13) and triphenylamine-2, 1, 3-benzothiadiazole hybrids (14 and 15) [57-59], as shown in Fig. 3. The highest PCE of 2.88, 1.29 and 1.61% in each series were achieved for polymer 10, 13and 15 based BHJ devices respectively, owing to the presence of the most extended $\pi$-electron delocalized system along the polymer chain to efficiently harvest the solar energy. For example, the bandgaps of 7-10 successively reduced from 2.93, 2.60, 2.43 to 2.33 eV, accomplished by the gradual increase of PCE from 0.36, 1.54, 2.47 to 2.88%. The best performance of 10 based solar cell device was attributed to its considerably high charge mobilities and favorable light absorption characteristics [57].In general, the design and synthesis of $\pi$-conjugated platinum-acetylide containing polymers via altering the number of thienyl ring, indicating a promising road to tune their optical absorption, electrochemical and electronic properties as well as the photovoltaic performance when applied as donor materials in BHJ solar cell devices. However, the utilization of the triplet excited states to promote an efficient photoinduced charge separation was not detected in these D-A platinum-acetylide polymers.

Besides, Wong et al. also synthesized a new D-A type organometallicpolyyne 16 containing the more electron-rich 3, 4-ethylenedioxythiophene (EDOT) and the electron-deficient benzothiadiazole hybrid spacer. The polymer exhibited a narrow bandgap of 1.76 eV, when used as the electron donor in the BHJ active layer only a relatively low PCE of 0.3% was obtained [60]. In addition, three amorphous $\pi$-conjugated organometallic polymers 17-19 (as shown in Fig. 3) were demonstrated by Jen and coworkers [61], which showed field-effect hole mobility as high as 1.0×10-2 cm2 V-1 s-1, therefore a relatively high level of performance was expected in BHJ photovoltaic devices based on the blends of organometallic polymers and PCBM. The best performance with VOC of 0.79 V, fill factor (FF) of 51.4%, short-circuit current density (JSC) at 10.12 mA/cm2 and PCE up to 4.13% was achieved in a 1:4 weight ratio of polymer 18:PC71BM blends. Based on the 2, 1, 3-benzothiadiazole (BT) containing model polymer 2 (BT-BTPt), Luscombe et al. [63] reported a series of organometallic conjugated polyplatinynes comprising different electron-acceptors, such as pyrido[3, 4-b]pyrazine (HPP-BTPt, 20), thieno[3, 4-b]pyrazines (TP-BTPt 21, MTP-BTPt 22, PTP-BTPt 23). The optical bandgaps reduced from 1.96 of 20 to 1.86 (2), 1.66 (22), 1.54 (21) and 1.49 eV (23). BHJ solar cells based on the blends of polymers/ PC71BM gave PCE of 0.68% for 20 and 2.41% for 2.

[3, 4-i] Near-infrared (NIR) absorbing platinum(II) metallopolyynes functionalized with fluorene and benzo[1, 2-c:4, 5-c']bis[1, 2, 5] thiadiazole (24) or [1, 2, 5]thiadiazolo[3, 4-i]dibenzo[a, c]phenazine (25) hybrid spacer were also presented [64]. Through the weak electron-donor and strong electron-acceptor strategy, both polymers exhibited narrow bandgaps of 1.54 and 1.65 eV, respectively, with absorption spectra extended downto 800-850 nmandup to 1% of PCE for bulk heterojunction solar cells. Besides, Pt(II) containing metallopolyynes 26 and 27 based the electron-deficient 4Hcyclopenta[2, 1-b:3, 4-b']dithiophen-4-one were also reported [66], with extremely low bandgaps of 1.44-1.53 eV, representing the lowest optical bandgap yet reported for platinum(II) metallopolyynes to date. Similar to previous study, no triplet excited state occurs in photoinduced charge separation of the platinum-acetylide polymers/PCBM blends based BHJ solar cells. However, the work offers an attractive way towards conjugated heavy-metal complex containing photovoltaic materials with broad solar absorption and demonstrates the potential for both visible and NIR light power conversion.

Since the molecular weight of polymer active materials shows large impact on the performance and BHJ solar cells [20, 76], compared to polymeric systems which intrinsically exhibits large structural variations in molecular weight, polydispersity, regioregularity and batch-to-batch reproducibility, small molecules or oligomers with well-defined structures are relatively easier to synthesize and purify. Kirk S. Schanze and co-workers designed and synthesized a novel fulleropyrrolidine end-capped platinum acetylide donor-acceptor triad assembly of 28 (Fig. 4) [85]. Employing 28 as the active layer consisting solely of a donor-acceptor dyad or triad assembly in the OPV devices, the PCE and JSC as low as 0.056% and 0.5 mA/cm2 was obtained, respectively, due to the singlet and triplet states of the platinum-acetylide were strongly quenched in the triad assembly. Then later in 2012, Wong et al. reported four $\pi$-conjugated smallmolecular platinum(II)-bis(aryleneethynylene) complexes 29-32 which consist of the electron-accepting benzothiadiazole moiety and the electron-donating triphenylamine and/or thiophene units [84]. It was discovered that incorporating stronger electron-donor moieties into the p-conjugated metal-organic molecules could not only broad the absorption bands, but also enhance the intramolecular charge transfer between the donor and acceptor units.Therefore, the best performance was achieved in oligomers 29 and 30, with the hole mobility of up to 0.7 and 2.4×10-4 cm2 V-1 s-1, respectively. Solution-process BHJ photovoltaic devices by using them as donor active materials exhibited PCE of 2.37 and 2.34%, respectively.

2.2. Cyclometalated Pt/Ir complex containing p-conjugated polymers

Besides on the linear platinum acetylides comprising photovoltaic materials, heavy metal organometallic polymers containing cyclometalated Pt or Ir complexes which was attached via the C.N or O.O diketonate ligands have been attracted gradually increasing attention due to better involvement from the orbitals of Pt or Ir center and the cyclometalated organic ligand to polymer backbone in both the ground and the excited states of the material [71-75]. Strong spin-orbit coupling on the heave metal center allows for the mixing of singlet metal-to-ligand charge transfer state (1MLCT) with the formally forbidden triplet metal-to-ligand charge transfer (3MCLT) and 3$\pi$-$\pi$* states. Therefore, cyclometalated Pt/Ir complex containing conjugated polymer donor materials are worth to investigate on better understanding the role of heavy atom effects on triplet formation in polymer photovoltaics. It is reported that materials with large triplet yields may provide favorable access to increase the current of OPV devices [31, 51]. This enhancement in photovoltaic performance was attributed to the decrease of geminate recombination and increase of exciton diffusion length due to the forbidden nature of recombination fromthe triplet state. And it is demonstrated that triplet-forming polymers is beneficial to hinder geminate pair recombination and to improve the exciton diffusion length in BHJ solar cell devices [40, 53, 86].

Compared to the polyplatinynes, the cyclometalated Pt or Ir complex containing polymers demonstrated higher flexibility for tuning the optoelectronic properties through themodification of the chelated main ligands. In 2009, Fréchet et al. reported the two conjugated cyclometalated platinum-containing polymers Pt-T1 (33) andF8TZPt (34) (Fig. 5), inwhichthe platinumcenter is adjacent to the conjugated backbone via a cyclometalated 2-(2'-thienyl)thiazole C.N ligand to forma coplanar conformation [40]. The polymers exhibited optical bandgaps of 2.1 and 1.65 eV according to different comonomers of fluorene and thiophene, thus both the two cyclometalated polymers showed a large overlap with the solar spectrum. The space charge limited current (SCLC) mobilities of 34 and 33 were measured to be 2.5×10-9 and 1×10-5 cm-2 V-1 s-1, respectively. The SCLC mobility of Pt-T1 is significantly higher than polyplatinynes (10-8 to 10-7 cm-2 V-1 s-1) [39]. The improved charge transport properties of the cyclometalated platinumcontaining conjugated polymers indicated an extra attractive route to organometallic polymers for photovoltaics. Triplet exciton formation was discovered indirectly by conducting the photosensitized emission behavior of singlet oxygen in both solution and solid film. Photovoltaic devices fabricated by blending the tow polymer materials 34 or 33 with PCBMas the active layer gave PCE up to 1.3% and 0.4%, respectively, suggesting that cyclometalated platinum polymers are potential candidates of photovoltaic materials.

 Download: larger image Figure 5. Molecular structures of cyclometalated Pt/Ir containing polymers

In 2013, Chien-Hong Cheng and co-workers synthesized two cyclometalated platinum polymers (35 and 36) and two Pt-free control polymers (37 and 38) as the comparison to characterize their photophysical properties and the photovoltaic performance [87]. It is found that the cycloplatinated polymers 35 and 36 exhibited significantly red-shifted absorption spectra and lower bandgap with respect to the Pt-free counter parts of 37 and 38. The performance of the photovoltaic devices fabricated by Pt-based polymers and PCBM blends demonstrated PCE approximately more than twice of the corresponding Pt-free polymers, the best performance was achieved to be 2.9% for compound 35. However, it was demonstrated that the higher current and PCE of cycloplatinated polymers based device than the Pt-free polymers was attributed to their significant red-shifted (80-90 nm) absorption bands to better harvest solar energy of the Pt-containing polymers instead of the triplet contribution.

Schanze group also presented a $\pi$-conjugated organometallic isoindigo-based polymer 39 (Iin-C18-Poly) that contains cyclometalated platinum unit was developed and applied as electron donor into BHJ solar cells with PCBM [74]. Isoindigo is an excellent accepter in D-A type photovoltaic materials, attribute to its intrinsic electron-deficient nature [88-90]. The triplet excited state occurs in 39 with introducing cyclometalated platinum complex into the polymer backbone, could induce spin-orbit coupling and enhance singlet-triplet intersystem crossing.Since the singlet energy of the polymer (1.89 eV) is higher than the charge-separated states (Iin-C18-Poly+/PC61BM•-, Ecs:~1.82 eV), therefore, PET from the singlet state of polymer to its charge separated state is favorable. Unfortunately, as the singlet and triplet energy levels of the charge transfer state are very similar, it is evidenced that the triplet state (1.09 eV) of the polymer was more than 0.5 eV below the charge separated state, and thus charge transfer for charge separation occurred via the triplet state is not at all likely to happen, while the triplet state energy loss from the triplet charge separation state to the triplet of donor polymer might exist [33, 37]. Together with the close lowest unoccupied molecular orbital (LUMO) (~4.2 eV) of the polymer to that of PCBM which lead comparatively low driving force for charge separation at the donor-acceptor interface, BHJ photovoltaic device performed a rather low PCE of 0.22%.

Besides on the cycloplatinated polymers, oligomers (40-42) of cyclometalated platinum complex by introducing the widely reported high performance diketopyrrolopyrrole (DPP) or Isoindigo (IID) moiety were also reported [71, 74, 75]. For example, when DPP chromophores were end-functionalized with platinum containing auxochromes, photophysics study revealed that compared with the platinum acetylide units, orthometalated platinum auxochrome 40 is more likely to induce a greater extent of spin- orbit coupling. Unfortunately, no photovoltaic performance has been conducted on these cyclometalated heavy metal containing oligomers.

By chemically attaching the Ir complex into the $\pi$-conjugated poly(fluorine-co-phenylpyridine) (43) backbone through the cyclometalation of phenylpyridine with Ir center [86], the first cyclometalated heavy metal Ir complex containing polymer (44) for photovoltaics was presented by Holdcroft group. Blending the donor polymer with the electron acceptor of PCBM, a significant increase of EQE from 1.1% for 43 to 10.3% of 44 was observed. Despite the PCE of 44 based devices is considerably low as just 0.07%, whereas it is 35 times higher than the control Ir-free polymer 43, exhibiting an enormously great improvement. It is demonstrated that introducing triplet forming Ir complexes into the poly(fluorine-co-phenylpyridine) polymer to function as donor materials for OSC, considerably enhanced photovoltaic conversion efficiencies was attributed to the formation of the triplet state and thus to extend the longer diffusion length. The rather low PCE of the Ir containing poly(9, 9-dioctylfluorene-co-tris(2-phenylpyridine) iridium (III)) could be assigned to its unfavorable light harvesting ability and poor charge transport property.

According to the above mentioned issues, most of the BHJ polymersolar cellsemploying heavymetal complexes exhibited PCE no more than 3% [50-70]. It should be noted that BHJ OPVswith PCE over 6% reported so far are all based on singlet dominated pure organic materials [81-98]. Therefore, it is hypothesized that incorporating heavy-metal complexes into the previously reported high efficiency p-conjugated D-A type polymer backbones, is expected to promote the generation of triplet CT state and/or reduce the triplet CT loss, thus to improve the device efficiency. Based on this concept, our group reported a series of $\pi$-conjugated cyclometalated polymers (45-51) [76], by embedding various very low concentrations of triplet iridium complexes (dfppy)2Ir(dbm) (from0, 0.5, 1, 1.5, 2.5 to 5mol%) through the ancillary ligand to the polymer backbone of the outstanding famous donor material PTB7 which was first reported by Yu and co-workers, and two batches of polymers were synthesized to consider the molecular weight effect which could affect the photovoltaic performance [28, 91, 99, 100]. By blending with PC71BM as the active layer of polymer photovoltaic devices, compared to the Ir free control PTB7, enhanced JSC, EQE and PCE have been achieved under various repeated experiments at lower Ir content of 1, 2.5% (batch 1) and 0.5, 1.5% (batch 2). Devices basedon1%of Ir embeddedpolymer46, thePCEwas increasedby45, 39 and 31% at three different device fabrication conditions. Fromthe materials point of view, besides on the exploitation of new conjugated photovoltaicmaterials, chemically bonding lowcontent triplet heavy metal complexes to current high efficiency polymers could offer an extra approach to improve the PCE of BHJ polymer solar cells.

2.3. Phosphorescent triplet metal complexes

Cyclometalated heavy metal complexes bearing with large bandgap ligands have been extensively investigated as the phosphorescent emitter in high efficiency organic light emitting diodes [101-104]. However, the application of these small molecular pure triplet organometallic complexes as sole donor materials in BHJ organic solar cells has been rarely studied owing to the relatively poor charge transport and poor film-forming ability as well as narrow and weak light absorption capability. Zhen et al. designed and synthesized two red light-emission cyclometalated Ir complex 52 and 53 via harvesting triplet excitons in OSCs (Fig. 6) [77]. The film morphology, carrier mobility and light absorption were allowed to improve by introducing the 2, 2, 6, 6-tetramethyl-3, 5-heptanedione, p-methoxybenzene and triphenylamine into the ligand. The hole mobilities of 52 and 53 measured from field effect transistor (FET) are determined to be 3×10-6 and 3×10-5 cm2 V-1 s-1, respectively, which is in good agreement with SCLC method of 2.0×10-6 and 4.3×10-5 cm2 V-1 s-1. A JSC of 6.5 mA/cm2, VOC of 0.74 V, FF of 0.42, and overall PCE of 2.0% was obtained in the conventional solution processed BHJ organic photovoltaic device by using the Ir complex 53 as the electron donor and PC71BM as electron acceptor. The results indicated the possibility of involving phosphorescent dyes as single electron donors in solution processed BHJ OSCs, enriched the donor materials of OSCs and provided important guidance for the molecular design on new heavy metal complexes with low bandgaps, desired highest occupied molecular orbital (HOMO) and LUMO energy levels as well as suitable molecular packing to enhance the absorption and charge transport for triplet OSCs.

 Download: larger image Figure 6. Molecular structures of phosphorescent triplet metal complexes as sole donor materials

Besides, Hong and co-workers also synthesized three isoquinoline- based Ir complexes (54-56) by changing the main ligands and studied their photovoltaic performance through fabricating BHJ solar cells [81]. By blending Ir complex with PCBM in a 1:4 ratio for the spin coated devices, 55 exhibited the highest PCE of 0.50%, compared to 0.43 and 0.34% for 54 and 56, respectively. The higher PCE obtained from complex 55 mainly derived from the superior hole transport ability of thetriphenylamine moiety.

Morphology property of donor/acceptor blends in the active layer which includes domain sizes, D/A materials miscibility and molecular orientation, crystallinity etc., directly affects the photovoltaic device performance especially JSC and FF [105, 106]. If the interfacial area between the electron donor and acceptor is too indistinct or the domains of the two phases within of the active composite layer are too large, exciton dissociation is unbeneficial. Therefore, several strategies including solvent additives, solvent annealing, thermal annealing or mixed solvent have been carried out to improve the morphology of the D/A blends to reach an optimal length scale of phase-separated interpenetrating networks between the electron donor and acceptor to maximize exciton dissociation and simultaneously minimize charge recombination, and thus to enhance photovoltaic performance [107-111].

Besides, incorporating of the heavy metal complexes as additives into the photoactive layer of the BHJ solar cells can also improve the JSC and enhance the PCE of the devices. Loo and coworkers [78] reported the incorporation of fractional amounts of charged heavy metal Ir complex as nonvolatile additives to the photoactive layer of P3HT/PCBM blends. It was initially expected to induce the generation of triplet excitons which possess longer lifetimes and longer diffusion lengths from the heavy metal complex to enhance charge-carrier generation. However, detailed characterization suggested that it is not the generation of triplet excitons from the transition metal complex, whereas the Ir-based additives selectively partitioned into and remained in the P3HT phase, and thus to effectively increase the chemical incompatibility between the electron donor P3HT and acceptor PCBM. The strategy of involving of the nonvolative additives efficiently altered the thermodynamics of phase separation between the electron donor and acceptor, resulting in the formation of more PCBM clusters. It was demonstrated that PCE could be enhanced from 1.01% to 1.81% through physically doping of 5 wt% Ir complex 57 to P3HT/PCBM blended photoactive layer in inverted bulk-heterojunction organic solar cell.

Kim and co-workers introduced a multifunctional Ir complex 58 (Fig. 7) as additives and polyethylene oxide (PEO) as an ion channel to strengthen the ionic mobility [112, 113] to enhance the nanophase segregation of the photovoltaic active layer film and the compatibility of P3HT/PCBM blends [79]. Increased JSC from 8.57 to 10.24 mA/cm2 and PCE from 1.6% to 3.4% were observed. Besides on the influence of the additives to the morphology of the photoactive layer, they also discovered the obvious enhancement in JSC was primarily attributed from efficient triplet to singlet energy transfer from the Ir complex 58 to P3HT.

We have summarized recent advances on the $\pi$-conjugated heavy metal complex containing small molecules, oligomers and polymers for their application in BHJ solar cells. Platinum- acetylide containing oligomers or polymers as donor materials exhibited PCE up to 5%. And the introduction of 1 mol% of Ir complex to the champion donor polymer PTB7 backbone, the PCE was up to 8.7% which was increased by 31%-45% compared to the control polymer PTB7 under the same device fabrication condition. Involving pure triplet phosphorescent dyes as sole electron donors in solution-processed BHJ solar cells, gave overall PCE of 2%. It is believed that introducing appropriate long exciton lifetime organic materials is a very promising approach to improve the photovoltaic performance. The current pure triplet materials used in OPV displayed rather poor light harvesting ability and low charge carrier mobility. Therefore, high-mobility triplet materials with wide and efficient solar light absorption capability, high carrier mobility and well matched energy levels are highly desired to improve the efficiency of organic photovoltaic devices in the future. Furthermore, the confirmed detail mechanism of triplet content on the influence of organic solar cells is desired for future investigation, either for materials with very low ratio in current high efficiency D-A polymers or broad absorption of pure triplet organic small molecules or oligomers.