*3.3.6 N,N′-bis-(heptafluorobutyl)-2,6-dichloro-1,4,5,8-naphthalene tetracarboxylic diimide*

Würthner et al. manufactured single-crystal transistors based on *α*-phase crystals of N,N′-bis-(heptafluorobutyl)-2,6-dichloro-1,4,5,8-naphthalene tetracarboxylic diimide (Cl2-NDI), which showed electron mobility up to 8.6 cm2 V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup> ,

#### **Figure 7.**

*(a) and (d) Crystal color, size, and shape of the LT red and HT yellow polymorphs of TMS-DBC. (b) and (c) Side and top views of the crystal packing in the red LT polymorph. (e) and (f) Side and top views of the crystal packing in the yellow HT polymorph. The directions corresponding to the largest calculated electronic couplings are indicated with arrows [58]. Copyright 2015, American Chemical Society.*

**77**

1.26 cm2

 V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup>

*Crystal Polymorph Control for High-Performance Organic Field-Effect Transistors*

achieving the best performance for air-stable n-type OFETs reported till now [107]. The *α*-phase polymorph of Cl2-NDI was grown on n-octadecyl triethoxysilanemodified substrates resulting in ribbon-shaped crystals by drop casting via CHCl3 solution. In a study by He et al., the ribbon-shaped *β*-phase crystals were grown on various substrates like Si/SiO2 by sublimation in air ambient pressure [108]. The single-crystal transistors of *β*-phase crystals exhibited a maximum electron mobil-

In the *α*-phase crystal, Cl2-NDI molecules adopt a herringbone packing motif, where nearly half of each molecular skeleton overlaps with the adjacent molecule at a close π-stack distance of 3.27 Å. In contrast, molecules in the *β*-phase crystal exhibit a two-dimensional brick-wall packing arrangement with a π-stack distance of 3.29 and 3.32 Å. The selected area electron diffraction (SAED) studies revealed that single-crystal transistors for both the *α*-phase and *β*-phase crystals were measured along the π-π stacking direction. Compared to the *α*-phase crystal, the *β*-phase crystal possesses a weaker electronic coupling along the π-π stacking direction and a longer percolation pathway for electrons to cross the unit cell, which

In an investigation by He et al., dihexyl-dibenzo[d, d']thieno[3, 2-b; 4,5-b'] dithiophene (C6-DBTDT) was synthesized efficiently [47]; two polymorphs of C6-DBTDT were obtained by drop casting of solutions with different concentration in chlorobenzene or toluene. The platelet-like *α*-phase single crystals were prepared through drop casting from a high concentration chlorobenzene solution (5.0 mg/ mL). In contrast, the micro-ribbonlike *β*-phase single crystals were formed from a relatively diluted chlorobenzene solution (0.3 mg/mL). Single-crystal transistors were fabricated, where the *α*-phase and *β*-phase crystals exhibited hole mobilities of

It is generally believed that only the HOMO level contributes to hole charge carrier transport from one molecule to another adjacent molecule. The electronic couplings of adjacent molecules in the *α*-phase and *β*-phase crystals were calculated by the quantum chemical calculations. The results indicated that the electronic couplings of HOMO between adjacent molecules in *α*-phase crystal are larger than that in *β*-phase crystal. On the other hand, the electronic couplings of HOMO demonstrate that the *α*-phase crystals may facilitate charge transport, which is in opposition to the experimental results. The authors noted that the electronic couplings of (HOMO-1)s for *β*-phase could be much larger than that of *α*-phase, which is near to the HOMO level. It was thought that (HOMO-1) level plays an important role in the

*3.3.8 2,8-Bis(butyl(methyl)amino)-indeno[1,2-b]fluorene-6,12-dione (BMA-IFD)*

In a recent investigation by Fan et al., a new molecule, indeno[1,2-b]fluorene-6,12-dione derivative, i.e., BMA-IFD, was designed and synthesized [55]. Two polymorphs of BMA-IFD were easily obtained by crystallization from solution. The ribbon-shaped *α*-phase crystal (**Figure 8a**) was obtained from chloroform (1 mg/ml), while the flake-shaped *β*-phase crystal (**Figure 8b**) was obtained from xylene solution (0.2 mg/ml). Single-crystal OFETs were fabricated and measured. The results showed that the β-phase polymorph exhibits hole mobility up to

, much higher than that of the α-phase crystals (0.21 cm2

The α-phase and the *β*-phase crystals, whose structures were experimentally determined, were investigated by quantum chemical calculations to associate the

 V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup> ).

*DOI: http://dx.doi.org/10.5772/intechopen.91905*

ity of 3.5 cm2

8.5 and 18.9 cm2

 V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup> .

explains lower carrier mobility.

 V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup>

charge transporting behavior.

*3.3.7 Dihexyl-dibenzo[d, d']thieno[3, 2-b; 4,5-b']dithiophene*

, respectively.

#### *Crystal Polymorph Control for High-Performance Organic Field-Effect Transistors DOI: http://dx.doi.org/10.5772/intechopen.91905*

achieving the best performance for air-stable n-type OFETs reported till now [107]. The *α*-phase polymorph of Cl2-NDI was grown on n-octadecyl triethoxysilanemodified substrates resulting in ribbon-shaped crystals by drop casting via CHCl3 solution. In a study by He et al., the ribbon-shaped *β*-phase crystals were grown on various substrates like Si/SiO2 by sublimation in air ambient pressure [108]. The single-crystal transistors of *β*-phase crystals exhibited a maximum electron mobility of 3.5 cm2 V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup> .

In the *α*-phase crystal, Cl2-NDI molecules adopt a herringbone packing motif, where nearly half of each molecular skeleton overlaps with the adjacent molecule at a close π-stack distance of 3.27 Å. In contrast, molecules in the *β*-phase crystal exhibit a two-dimensional brick-wall packing arrangement with a π-stack distance of 3.29 and 3.32 Å. The selected area electron diffraction (SAED) studies revealed that single-crystal transistors for both the *α*-phase and *β*-phase crystals were measured along the π-π stacking direction. Compared to the *α*-phase crystal, the *β*-phase crystal possesses a weaker electronic coupling along the π-π stacking direction and a longer percolation pathway for electrons to cross the unit cell, which explains lower carrier mobility.

#### *3.3.7 Dihexyl-dibenzo[d, d']thieno[3, 2-b; 4,5-b']dithiophene*

In an investigation by He et al., dihexyl-dibenzo[d, d']thieno[3, 2-b; 4,5-b'] dithiophene (C6-DBTDT) was synthesized efficiently [47]; two polymorphs of C6-DBTDT were obtained by drop casting of solutions with different concentration in chlorobenzene or toluene. The platelet-like *α*-phase single crystals were prepared through drop casting from a high concentration chlorobenzene solution (5.0 mg/ mL). In contrast, the micro-ribbonlike *β*-phase single crystals were formed from a relatively diluted chlorobenzene solution (0.3 mg/mL). Single-crystal transistors were fabricated, where the *α*-phase and *β*-phase crystals exhibited hole mobilities of 8.5 and 18.9 cm2 V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup> , respectively.

It is generally believed that only the HOMO level contributes to hole charge carrier transport from one molecule to another adjacent molecule. The electronic couplings of adjacent molecules in the *α*-phase and *β*-phase crystals were calculated by the quantum chemical calculations. The results indicated that the electronic couplings of HOMO between adjacent molecules in *α*-phase crystal are larger than that in *β*-phase crystal. On the other hand, the electronic couplings of HOMO demonstrate that the *α*-phase crystals may facilitate charge transport, which is in opposition to the experimental results. The authors noted that the electronic couplings of (HOMO-1)s for *β*-phase could be much larger than that of *α*-phase, which is near to the HOMO level. It was thought that (HOMO-1) level plays an important role in the charge transporting behavior.

#### *3.3.8 2,8-Bis(butyl(methyl)amino)-indeno[1,2-b]fluorene-6,12-dione (BMA-IFD)*

In a recent investigation by Fan et al., a new molecule, indeno[1,2-b]fluorene-6,12-dione derivative, i.e., BMA-IFD, was designed and synthesized [55]. Two polymorphs of BMA-IFD were easily obtained by crystallization from solution. The ribbon-shaped *α*-phase crystal (**Figure 8a**) was obtained from chloroform (1 mg/ml), while the flake-shaped *β*-phase crystal (**Figure 8b**) was obtained from xylene solution (0.2 mg/ml). Single-crystal OFETs were fabricated and measured. The results showed that the β-phase polymorph exhibits hole mobility up to 1.26 cm2 V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup> , much higher than that of the α-phase crystals (0.21 cm2 V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup> ).

The α-phase and the *β*-phase crystals, whose structures were experimentally determined, were investigated by quantum chemical calculations to associate the

*Integrated Circuits/Microchips*

is up to 2.1 cm2

−86.8 × 10<sup>−</sup><sup>3</sup>

*diimide*

 V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup>

of crystal structures for the two polymorphs revealed smaller distances between neighboring molecules in the *α*-phase crystal which may facilitate charge transport.

Stevens et al. found two polymorphs of 7,14-bis((trimethylsilyl)ethynyl) dibenzo[b,def]-chrysene (TMS-DBC) using the physical vapor transport technology [58]. The first polymorph was obtained as red needles at low temperature, which was named as LT-phase (**Figure 7a**). The second polymorph was formed at high temperature as yellow plates and named HT-phase (**Figure 7b**). Further investigations found that the LT-phase can also be fabricated from solution and could not be converted into HT-phase by thermal annealing. Single-crystal OFETs of the two polymorphs were fabricated. The results revealed that the hole mobility of the HT-phase

, while that of the LT-phase is only 0.028 cm2

eV (**Figure 7b**). In contrast, the HT-phase possesses exhibits 2D charge

As shown in **Figure 7**, the LT-phase adopts one-dimensional (1D) slipped stacking, while the HT-phase exhibits two-dimensional (2D) brick-wall stacking. Quantum chemical calculations revealed that the LT-phase possesses 1D charge transport channel along the π-stacking direction with a transfer integral of

the t1 and t2 directions, respectively (**Figure 7e**). Though the HT-phase exhibits slightly smaller transfer integral (absolute value) than that of the LT-phase, its 2D charge transport channels benefit charge transfer, which allows charge carriers to take alternative pathways around defects or trap states. As a result, the HT-phase

*3.3.6 N,N′-bis-(heptafluorobutyl)-2,6-dichloro-1,4,5,8-naphthalene tetracarboxylic* 

Würthner et al. manufactured single-crystal transistors based on *α*-phase crystals of N,N′-bis-(heptafluorobutyl)-2,6-dichloro-1,4,5,8-naphthalene tetracar-

*(a) and (d) Crystal color, size, and shape of the LT red and HT yellow polymorphs of TMS-DBC. (b) and (c) Side and top views of the crystal packing in the red LT polymorph. (e) and (f) Side and top views of the crystal packing in the yellow HT polymorph. The directions corresponding to the largest calculated electronic* 

*couplings are indicated with arrows [58]. Copyright 2015, American Chemical Society.*

boxylic diimide (Cl2-NDI), which showed electron mobility up to 8.6 cm2

 V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup> .

eV along

 V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup> ,

and − 41 × 10<sup>−</sup><sup>3</sup>

*3.3.5 7,14-Bis((trimethylsilyl)ethynyl)-dibenzo[b,def ]-chrysene*

transfer channels with transfer integrals of −77.3 × 10<sup>−</sup><sup>3</sup>

facilitates higher mobility than the LT-phase.

**76**

**Figure 7.**

#### **Figure 8.**

*SAED and TEM images of (a) the α-phase and (b) the β-phase crystals (the scale bar is 5 μm). The transfer integrals of (c) the α-phase and (d) the β-phase crystals along the (001) directions. The molecules in panels e and f are colored differently only for clarity purposes [55]. Copyright 2018, American Chemical Society.*

charge transport properties with the molecular packing structures. The results show that a one-dimensional (1D) electron coupling between adjacent molecules is observed in the α-phase crystal (**Figure 8c**). In comparison, a two-dimensional (2D) electron coupling between adjacent molecules is found in the *β*-phase crystal (**Figure 8d**). Though the values of transfer integrals are close for the two polymorphs, the β-phase polymorph possesses a 2D charge transport network and therefore exhibits higher carrier mobility.

#### *3.3.9 Titanyl phthalocyanine*

Titanyl phthalocyanine (TiOPC) is a well-known organic semiconductor and photoconductor; however, it exhibits poor solubility in common solvents. In a recent study by Zhang et al., TiOPC crystals were synthesized by physical vapor transport (PVT) technique through a two-zone horizontal tube furnace [6]. Some sheet crystals were obtained at the temperature zone of about 210°C, while some ribbon crystals were grown at the temperature zone of about 180°C. The sheet and ribbon crystals belong to the *α*-phase and *β*-phase polymorphs, respectively. The measurements on single-crystal OFETs of the two polymorphs demonstrated that the *α*-phase crystals exhibit excellent charge transport property with mobility up to 26.8 cm<sup>2</sup> V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup> , while that of *β*-phase crystals are only 0.1 cm<sup>2</sup> V<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup> .

The crystal structures of the two polymorphs were determined, where the *α*-phase and *β*-phase crystals exhibit a 2D lamellar brick stone motif and an unusual 3D framework, respectively. The main difference in electronic coupling of the two polymorphs was a strong interlayer electronic couplings perpendicular to the current direction in the *β*-phase crystal. The strong interlayer electronic couplings may result in destructive interference effects that remarkably diminish the charge carrier mobility.

**79**

China

*Crystal Polymorph Control for High-Performance Organic Field-Effect Transistors*

This work is supported by the National Key R&D Program of China (2017YFA0204903), National Natural Science Foundation of China (NSFC. 51733004, 51525303, 221702085, 21673106, 21602093, 21572086, 1522203), 111 Project, and the Fundamental Research Funds for the Central Universities. The authors thank beam line BL14B1 (Shanghai Synchrotron Radiation Facility) for

\*

Chemical Engineering, Lanzhou University, Lanzhou, China

\*Address all correspondence to: haoli.zhang@lzu.edu.cn

provided the original work is properly cited.

1 State Key Laboratory of Applied Organic Chemistry, College of Chemistry and

2 School of Materials Science and Engineering, Nanchang University, Nanchang,

© 2020 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium,

Herein, the polymorphism in organic semiconductors is introduced, including the common strategies for polymorph control and investigations on OFETs from different polymorphs. Polymorphism is proved to be an excellent platform to directly correlate the molecular packing with charge transport for organic semiconductors; such investigations are very limited so far. A main challenge is to precisely tailor thermodynamic and kinetic factors of crystal nucleation and growth for large-area thin films or high-quality single crystals. Among the investigations on polymorphism, several polymorphs with outstanding charge transport performance have been obtained, demonstrating that altering the crystal polymorph structure of organic semiconductors is an efficient strategy to access high-performance OFETs. However, the majority of the high-mobility polymorphs are metastable. Consequently, getting insight into the relationship between molecular structure and crystal polymorph remains an important issue, which is essential for the rational design of molecular structures to further develop the desired crystal polymorphs with outstanding electrical characteristics.

*DOI: http://dx.doi.org/10.5772/intechopen.91905*

**4. Conclusion and outlook**

**Acknowledgements**

providing the beam time.

**Conflict of interest**

**Author details**

Zhi-Ping Fan1,2 and Hao-Li Zhang1

The authors declare no conflict of interest.

*Crystal Polymorph Control for High-Performance Organic Field-Effect Transistors DOI: http://dx.doi.org/10.5772/intechopen.91905*
