Tuning Semiconductor Performance of Nickel ... - ACS Publications

Jul 3, 2018 - significantly different stacking model in comparison to the ..... and all of its possible neighbors (m1−m14/m16) have been calculated ...
2 downloads 0 Views 2MB Size
Download PDF
Article Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX

pubs.acs.org/IC

Tuning Semiconductor Performance of Nickel Complexes through Crystal Transformation Yan-Fang Wu,‡,∥ Shuai Zhao,§,∥ Hong-Xu Na,‡ Pei-Yu Yang,‡ Haibing Xu,† Yuexing Zhang,*,† Yanli Chen,§ and Ming-Hua Zeng*,†,‡

Downloaded via UNIV OF LOUISIANA AT LAFAYETTE on September 25, 2018 at 14:38:12 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

Hubei Collaborative Innovation Center for Advanced Organic Chemical Materials, Ministry-of-Education Key Laboratory for the Synthesis and Application of Organic Functional Molecules, College of Chemistry and Chemical Engineering, Hubei University, Wuhan 430062, People’s Republic of China ‡ Key Laboratory for the Chemistry and Molecular Engineering of Medicinal Resources, Guangxi Normal University, Guilin 541004, People’s Republic of China § School of Science, China University of Petroleum (East China), Qingdao 266580, People’s Republic of China S Supporting Information *

ABSTRACT: Crystal transformation between two polymorphs (green, 1-G, and red, 1-R) of the square-planar nickel complex NiL2 (L = 2ethoxy-6-(N-methyliminomethyl)phenolate) and their tuning effect to semiconductor properties were studied both experimentally and theoretically. When 1-G is heated to 413 K, it converts to 1-R, whereas soaking 1-R in several kinds of solvents causes it to revert to 1-G. Crystallographic and PXRD studies reveal the dramatic changes in crystal dimensions due to the changes of packing models. Heating device made from 1-G (D-1-G(298)) at 413 K significantly increases the electrical conductivity from 6.55 × 10−4 S cm−1 for D-1-G(298) to 1.11 × 10−3 S cm−1 for D-1-G(413), showing significant crystal form dependence. Heat-treating D-1-G and D-1-R devices at different temperatures clearly reveals the reason for the conductivity tuning. Thus, the conductivity of NiL2-based devices could be well tuned through crystal transformation by heating or by soaking in solvent. Theoretical calculations clearly revealed the reason for such conductivity changes and also predicted that both polymorphs are good p-type semiconductors with hole mobilities of 1.63 × 10−2 (1-G) and 2.11 × 10−1 cm2 V−1 s−1 (1-R).

■

INTRODUCTION Polymorphism is an important phenomenon of molecular crystals, in which the disparity in crystal packing often results in subtle discrepancies in color, morphology, solubility, and in turn bulk properties (magnetic, dielectric, emission, or mechanical).1−4 Though the concurrent existence of polymorphs has long been recognized, the small number of studies have been restricted to pharmaceutical, inorganic ionic, and metallic compounds.5 Currently, growing interest has focused on metal−organic hybrid crystals and the way that polymorphism affects their properties.6 Among polymorphic crystals, one form can transform to another by various treatment methods such as heating, soaking in solvent, gas absorption, and so on.7−9 Such a crystal transformation is an important property applicable to many fields.10−14 Organic semiconductors have generated tremendous interest because of their potential use in organic field effect transistors (OFETs), light-emitting diodes (LEDs), and photovoltaic devices (PVs).15−17 In addition to π-conjugated organic oligomers, polymers, and macrocycle coordination complexes,18−21 semiconductors based on coordination complexes with small planar π-conjugated ligands have also aroused great © XXXX American Chemical Society

interest for the design of new semiconductor materials with superior performance.22 Among small ligands, conjugated Schiff bases have interesting optoelectronic properties and have been widely used as semiconductor materials. 23 Associated with the crystal transformation, the semiconducting properties of the polymorphs would be different. Consequently, studying the crystal transformation associated change of semiconductor performance would shed further light on extending the use of crystal transformation and developing new modification methods of semiconductor properties. Theoretical calculations have been proven to be useful tools to explain the polymorphism. For example, the supramolecular interactions and formation of two structural polymorphs from one building unit in a one-pot synthesis were studied with the help of DFT calculations.24 First-principles theoretical studies can play a pivotal role in semiconductor material design by enabling a fundamental understanding of the inherent charge transfer behavior in organic semiconductor materials, facilitating the determination of the corresponding material properties Received: July 3, 2018

A

DOI: 10.1021/acs.inorgchem.8b01841 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

symmetry.30 In the 1-R phase (Figure 1 and Figure S1 and Table S1), NiL2 molecules stack through π−π interactions and direct Ni···Ni interactions perpendicular to the molecular plane with an AB model. The B molecule turns over relative to A and rotates along the Ni···Ni axis by about 18°. This turning over and rotation would reduce the electronic interaction between NiL2 molecules in A and B orientations. The AB packing column forms a diamond net with a side length of 12.176 Å and included angle of 79° and through hydrophobic effects between alkyl groups with other A′B′ columns. The two columns shift 3.357 Å along the Ni···Ni axis and result in A···B′ and B···A′ stacking structures. However, in the 1-G phase, layers consist of a zigzag organization of the molecules along the b direction, with the square planes making an angle of ca. 78° with the layer’s plane. The zigzag layers pack through π−π interactions along the c axis with an Ni···Ni distance of 6.239 Å and through hydrophobic effects along the a axis.30 When 1-G is heated at 413 K for 1/2 h, it changes to 1-R (Figure 1 and Figure S2). Indeed, differential scanning calorimetry (DSC) on microcrystalline powders of 1-G (Figure S3) shows an endothermic peak at 413 K related to the 1-G → 1-R transformation. Variable-temperature powder X-ray diffraction (PXRD) patterns of 1-G (Figure S4) show that 1G is stable below 373 K. When the temperature is increased to 413 K, the pattern consists of diffraction peaks from both 1-G and 1-R. Once the temperature attains 453 K, only peaks for 1R are observed. In addition, the PXRD of heated 1-G as a function of time (Figure S5) shows the formation of 1-R in less than 10 min. By 2 h, only the pattern of 1-R is observed, indicating that a complete 1-G → 1-R transformation occurred at 413 K within 2 h. In turn, upon soaking 1-R crystals in solvents under ambient conditions, as shown in Figure S6, 1-R gradually transforms to 1-G. Indeed, DSC on microcrystalline powders of 1-R (Figure S3) shows an endothermic peak at 330 K, suggesting that 1-R could transform to 1-G at low temperature. Additionally, PXRD patterns of 1-R treated with different solvents, such as DMF, MeOH, MeCN, EtOH, and acetone (Figure S7), show that 1-R is completely transformed to 1-G within 36 h. Thus, these two polymorphs show a crystal phase transition. That is, 1-G completely converts into 1-R through heat stimulation at 413 K for 2 h while 1-R transforms into 1-G by immersion into solutions under ambient conditions for 36 h (Figure 1). The formation of single crystals is a very complex process involving many factors, among which the weak interaction between molecules is one of the determining components.24 To explain the formation of 1-G and 1-R and the transformation between them, complexation energies among all of the possible dimers in the two polymorphs were studied through DFT calculations at the B3LYP-D3/(6-311G*,SDD) level (Table 1 and Table S2). In 1-G, dimers formed from the center molecule and molecules on pathways 1 and 2 (D0_1 and D0_2) have the largest complexation energy (with a value of −27.06 kcal/mol) among all of the possible dimers. These results indicate that this kind of stacking with a plane to plane distance of 3.489 Å and center to center distance of 6.239 Å is the most favorable packing model and is the greatest driving force for forming a 1-G crystal. D0_7 and D0_8 have the second largest complexation energy of −8.63 kcal/mol, which is only one-third of that for D0_1 and D0_2. In comparison, the largest complexation energy in 1-R is greater than that in 1G, with a value of −35.57 kcal/mol for D05 and D06. However, the second largest complexation energy in 1-R is

and establishing connections among molecular/electronic structures of organic semiconductor materials, their properties, and the performance of the electronic/photovoltaic devices based on such organic semiconductor materials.25−28 Recently, a new type of red crystal (1-R) of the mononuclear complex NiL2 (L = 2-ethoxy-6-(N-methyliminomethyl)phenolate) was reported, and it was used to study the selfassembling process toward Ni 7 . 29 This crystal has a significantly different stacking model in comparison to the earlier reported green polymorph (1-G).30 Interestingly, we found that the two crystals can be transformed upon heating or upon soaking in solvent. In addition, the different packing models of these two crystals suggest different semiconductor properties. However, phase transitions and theoretical aspects, including understanding the nature of organic semiconductors and modifying their semiconductor performances, remain to be investigated. In the present work, the crystal transformation between 1-G and 1-R and their semiconductor properties are studied experimentally. Density functional theory calculations are performed to rationalize their transformation and their different conductivities.

■

RESULTS AND DISCUSSION The geometries of NiL2 molecules in 1-G and 1-R are very similar, having one Ni(II)2+ ion symmetrically chelated by two Schiff base ligands through the imine nitrogen and phenol oxygen atoms in a symmetry-imposed square-planar geometry (Figure 1). However, the molecular geometry in 1-G

Figure 1. Photos of crystals, the crystal transformation (1-G ↔ 1-R), and top and side views of 1-G (left) and 1-R (right) crystals. Conductivities of devices fabricated from the two crystals are also given.

somewhat deviates from planarity with a plane to plane distance of 0.53 Å between planes formed by the phenyl groups of the two ligands. In contrast, the two ligands are coplanar in 1-R. These results indicate that the molecular geometry is significantly influenced by packing interactions. The red polymorph (1-R) has orthorhombic Ibam symmetry,29 while the green polymorph (1-G) has monoclinic P21/c B

DOI: 10.1021/acs.inorgchem.8b01841 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Table 1. Hopping Pathway, Center Mass Distance, Transfer Integral for Holes and Electrons (t+ and t−, Respectively), Complexation Energy for Dimers Formed from the Center Molecule and Molecule on the Transfer Pathway (CE), and Mobility of the Two Polymorphs of NiL2 (1-G, Space Group P21/c; 1-R, Space Group Ibam) 1-G (P21/c) pathway 1, 2 3−6 7, 8 9−12 13, 14 mobility μ (cm2 V−1 s−1)

1-R (Ibam)

distance (Å)

t+ (meV)

t− (meV)

CE (kcal/mol)

pathway

distance (Å)

t+ (meV)

t− (meV)

CE (kcal/mol)

6.239 9.445 8.633 12.541 10.027

7.215 6.749 2.754 0.470 0.013

52.531 10.098 21.242 0.331 2.465

−27.065 −8.501 −8.626 −1.022 −3.587

1−4 5, 6 7−10 11−14 15, 16

12.631 3.357 12.176 15.819 15.459

18.774 16.547 1.755 0.063 0.036

22.054 12.947 5.104 0.007 0.004

−5.814 −35.573 −2.886 −0.461 −0.701

1.63 × 10−2 (holes) 7.98 × 10−7 (electrons)

2.11 × 10−1 (holes) 5.56 × 10−7 (electrons)

from 1-G and 1-R were heated at 383 and 413 K for 2 h, and the corresponding I−V curves and conductivities of the treated devices are shown in Figure S8. By control of the dispersion time by ultrasound, no 1-R to 1-G transformation was found, indicating that the fabricated device of 1-R this time is almost composed of pure 1-R crystals instead of the case shown in Figure S8. A conductivity of 1.20 × 10−3 S cm−1 is obtained for the D-1-R(298) device at room temperature, which is almost unchanged with device treatment temperature (1.12 × 10−3 and 1.42 × 10−3 S cm−1 for devices treated at 383 K (D-1R(383)) and 413 K (D-1-R(413)), respectively). These results indicate that if the possible 1-R to 1-G transformation during the device fabrication process is restricted, D-1-R has very high conductivity which is almost not influenced by the device treating temperature. Heating D-1-G(298) at a temperature of 383 K for 2 h hardly increases its conductivity (from 6.55 × 10−3 S cm−1 for D-1-G(298) to 6.67 × 10−3 S cm−1 for D-1-G(383)). However, the device after heating at 413 K for 2 h (D-1-G(413)) is found to have a very large conductivity of 1.11 × 10−3 S cm−1, comparable to that of D-1R. This is because heating at 383 K does not convert 1-G to 1R but only slightly improves the molecular packing order of the device, while heating at 413 K converts 1-G to 1-R and thus significantly improves the conductivity. To explain the different semiconductor properties of 1-G and 1-R and obtain the relationship among geometry, packing model and semiconductor performance, the charge transfer behaviors in the two crystals were theoretically simulated using the same method as described in our previous works.26−28 A comparison of the energy of the closed-shell system (S = 0) with that of the high-spin system (S = 1) shows that the closed-shell system is more stable than the high-spin system, in good agreement with the diamagnetic properties of NiL2 as well as experimental and theoretical findings of similar planar four-coordinated nickel(II) complexes.32,33 Consequently, the following calculations were performed for closed-shell systems. The calculated HOMO and LUMO energies of NiL2 are −5.04 and −1.34 eV, respectively, with the HOMO−LUMO gap being 3.70 eV. The large HOMO−LUMO gap reveals the high stability upon redox of NiL2. The vertical ionization energies (IEv) and vertical electron affinities (EAv) are calculated by the total energy difference between the optimized neutral molecule and the cation and anion at neutral molecular geometry, respectively. The IEv and EAv values of NiL2 are 6.32 and 0.04 eV, respectively. The charge injection barrier to the Au electrode (with a potential of 5.1 eV)26−28 for holes is thus significantly smaller than that for electrons, suggesting that NiL2 shows promise as a p-type organic semiconductor in

smaller than that in 1-G. As a complete result, the total complexation energy of 1-G is much larger than that of 1-R, indicating that the crystal energy of 1-G is greater than that of 1-R and proposing that 1-G crystallizes out more easily. This agrees well with the judgment from the respective volume per molecule values (V/Z, at both room temperature and 100 K; see Table S1): 1-G is more densely packed than 1-R, indicating that 1-R would have larger entropy in comparison to 1-G. In addition, DFT calculations show that 1-G has a lower energy than 1-R at room temperature. At relatively “low” temperatures, entropy has a limited role, so that the formation of 1-G from 1-R should be mainly driven by the lower enthalpy of 1-G in comparison to that of 1-R. However, the transformation from 1-G to 1-R through heating to 413 K indicates that the larger entropy of 1-R in comparison to that of 1-G would be the driving force for the transformation from 1-G to 1-R at high temperature. To demonstrate the potentials of 1-G and 1-R as electronic materials, semiconductor devices were fabricated and the corresponding electrical conductivities were measured.31 Electrical conductivities for devices fabricated by dropping ultrasound-dispersed crystal−MeOH liquids onto glass substrates with indium tin oxide (ITO) interdigitated electrodes (IDEs) were characterized through a direct current−voltage (I−V) measurement (Figure 2 and Figure S8). In the devices the densely packed model of the crystal was assumed to be retained and adhere to the ITO IDEs/glass substrate tightly (inset of Figure 2). To further understand the relation between crystal forms and electrical conductivity, the devices fabricated

Figure 2. I−V curves measured for D-1-G and D-1-R fabricated by methanol deposition onto ITO IDE/glass substrates at 298 K. The inset gives a picture of one D-1-G device. C

DOI: 10.1021/acs.inorgchem.8b01841 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry terms of charge injection from Au electrode. The EAv value of NiL2 (0.04 eV) is far from the suggested electronic affinity for an n-type semiconductor (between 2 and 3 eV), therefore suggesting that NiL2 can only work as a p-type semiconductor. When NiL2 is oxidized to a cation, the planar geometry is retained. In contrast, when it is reduced, the anion of NiL2 slightly deviates from planarity (Table S3). The average bond length change upon oxidation is 0.011 Å. However, upon reduction, the average deviation for the bond length of anion relative to that of the neutral form is 0.024 Å, much larger than that upon oxidization. These results indicate that the change in the bond length of NiL2 upon either oxidation or reduction is somewhat large in comparison to macrocycle π-conjugated molecules such as phthalocyanines.27 As a result, the reorganization energy for both hole and electron transfer would be large (vide infra). Nevertheless, the bond length variation upon reduction is larger than that upon oxidation, suggesting a much larger reorganization energy for electrons of NiL2 in comparison to the reorganization energy for holes. The calculated reorganization energies from adiabatic potential energy surfaces for hole-transport (λ+) and electron-transport (λ−) processes of NiL2 are 0.17 and 1.43 eV, respectively. The λ+ value of NiL2 is significantly smaller than the λ− value, in line with their different structure changes upon oxidation and reduction associated with the strong coupling between the geometric and electronic structures in the π-conjugated systems.26−28 Charge transfer integrals for holes and electrons (t+ and t−, respectively) between one randomly selected molecule (m0) and all of its possible neighbors (m1−m14/m16) have been calculated on the basis of the experimental crystal structures of 1-G and 1-R. As can be seen in Figure 3 and Figure S8 and Table 1 and Table S2, there are 14 and 16 possible direct contact dimers between m0 and the other molecules in 1-G and 1-R, respectively. Among the 14 dimers in the crystal of 1G, the largest transfer integrals for holes and electrons are obtained for pathways 1 and 2 with values of 7.22 and 52.53 meV, respectively, indicating the most favorable hole and electron transfer in these two transfer pathways. This kind of dimer can be considered as having π−π stacking with a plane to plane distance of 3.489 Å and in-plane transfers of 1.967 and 4.784 Å. It is worth noting that the electron transfer integral is generally much larger than the hole transfer integral due to the different orbital distributions of HOMO and LUMO (Figure 4). These results indicate that electron transfer would be more favorable in the packing model of 1-G from the viewpoint of the transfer integral. Due to the significantly different packing in 1-R in comparison to 1-G, charge transfer integrals in 1-R differ greatly from those in 1-G. Due to the sticklike packing model of 1-R, there is only an edge to edge packing model. Like 1-G, pathways 1−4 of 1-R show the largest transfer integrals for both holes and electrons with values of 18.77 and 22.05 meV, respectively. These dimers have packing modes with a plane to plane distance of 3.357 Å and in-plane transfers of 0 and 12.177 Å (Figure 3 and Table S2). Despite the larger in-plane transfer over the length of the NiL2 molecule in 1-R, the distance between NiL2 planes is slightly shorter than that in 1G, making the transfer integral for this kind of packing model in 1-R larger than that in 1-G. It is worth noting that charge transfer integrals of holes and electrons in 1-R are more balanced than in 1-G, indicating that 1-R may be a good ambipolar semiconductor in terms of charge transfer integral if

Figure 3. Hopping pathway in two different crystals of NiL2 (1-G, space group P21/c, top; 1-R, space group Ibam, bottom). The center molecule is displayed in ball and stick style, molecules on the same plane as the center molecule are displayed in capped stick style, and the other molecules are shown in wireframe style. Color scheme, nickel elements, green; O, red; N, blue; C, gray. H elements are omitted for clarity. Green lines connect the molecular centers, and the values given in green show the transfer distances corresponding to the hopping pathways, while black italic numbers denote the pathway numbers in Table 1.

the reorganization energy for electrons could be reduced to a value similar to that of the reorganization energy for holes. The total charge transfer mobilities in the crystals of 1-G and 1-R have been calculated according to the Marcus equation and Einstein relation, and the results are given in Table 1. For 1-G, the largest transfer integrals for holes and electrons are 7.22 and 52.53 meV, respectively. The largest transfer integral for holes is significantly smaller than that for electrons, indicating that electron transfer is more favorable than hole transfer from the viewpoint of electronic coupling. However, the electron reorganization energy (1.43 eV) is much larger than the hole reorganization energy (0.17 eV), which is unfavorable for charge transfer. As a total effect, the hole transfer mobility in 1-G is 2 × 104 times that of electron transfer mobility: 1.63 × 10−2 vs 7.98 × 10−7 cm2 V−1 s−1. These results indicate that both the reorganization energy and transfer integral could significantly influence the charge transfer mobility. In good agreement, in 1-R, the largest transfer integrals for holes and electrons are comparable: 18.77 vs 22.08 meV. As a result, the hole transfer mobility of 1-R is significantly larger (3 × 105 times) than electron transfer mobility: 2.11 × 10−1 vs 5.56 × 10−7 cm2 V−1 s−1. A comparison of the charge transfer mobilities of 1-R and 1-G D

DOI: 10.1021/acs.inorgchem.8b01841 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Å) at 293 K. Calculated diffraction patterns of the compounds were generated by Mercury. Device Fabrication and Conductivity Measurements. The crystals were put into MeOH and dispersed by ultrasound, and the dispersion liquid was carefully dropped onto glass substrates with indium tin oxide (ITO) interdigitated electrodes (IDEs). It is worth noting that since 1-R could transform to 1-G when it was soaked in different solvents, such as DMF, MeOH, MeCN, EtOH, and acetone, some 1-R would possibly transform to 1-G during the device fabrication process if the ultrasound dispersion time is not well controlled (Figure S8). The ITO substrate is composed of 10 pairs of interdigitated electrode arrays with the following dimensions: 125 mm electrode width, 75 μm spacing, 6000 μm overlapping length, and 20 nm electrode thickness. After complete evaporation of the solvents, the densely packed model of the crystal was assumed to be retained and adhere to the ITO IDEs/glass substrate tightly, leading to the electrical conductivity measurement in situ. The fundamental electrical and sensor measurements were performed using a Keysight B2912A precision source/measure unit with an incorporated direct current voltage supply. The electrometer was controlled by Quick IV measurement software. Current−voltage (I−V) curves were registered in the voltage range of −10 to +10 V. All experiments have been conducted at least twice to ensure reproducibility. Conductivity, σ, obtained by current−voltage (I−V) curves can be calculated by the equation

σ = dI /(2n − 1)LhV where d is the interelectrode spacing, I is the current, n is the number of electrode digits, L is the overlapping length of the electrodes, and h is the electrode thickness as the film thickness exceeds that of the ITO electrodes in the present case. Computational Details. The B3LYP functional with addition of the D3 version of Grimme’s dispersion with Becke−Johnson damping with SDD ECP for Ni and the 6-311G(d) basis set for other elements were used for all calculations using the Gaussian 09 package.34 The fully optimized geometries of NiL2 were obtained, and their harmonic-model vibrational frequencies were computed. The geometries of the oxidized and reduced species (radical cations and radical anions) were also calculated using the same methods as those employed for the neutral species. All of the electronic structure calculations were performed on the optimized geometries with the same methods. Reorganization energies for both holes and electrons were calculated directly from the adiabatic potential-energy surfaces (PES) of the neutral/cation and neutral/anion species. The transfer integrals for both holes and electrons in these two single crystals were calculated using the site-energy correction method.35 Charge carrier mobilities were then calculated using Marcus theory and employing the average method.25−28

Figure 4. HOMO and LUMO orbital maps of dimers for pathways 1 and 5 (D0_1 and D0_5) of 1-G and 1-R.

reveals that hole transfer mobility of 1-R is 13 times that of 1G due to the much larger hole transfer integral of the former in comparison to the latter while the electron transfer mobility of 1-R is slightly smaller than that of 1-G. In summary, both square-planar nickel complexes show good hole transfer properties as p-type semiconductors, with 1-R having a larger hole transfer mobility in comparison to that of 1-G. These results agree well with the results of electron conductivity measurements (Figure 2).

■

CONCLUSIONS The crystal transformation between 1-G and 1-R and their different semiconductor properties have been studied experimentally. Density functional theory calculations were performed to rationalize the transformation and different conductivities. 1-G and 1-R both have very good p-type semiconductor performance with simulated hole mobilities of 1.63 × 10−2 (1-G) and 2.11 × 10−1 cm2 V−1 s−1 (1-R,) and their conductivity could be tuned through crystal transformation from 6.55 × 10−4 S cm−1 (D-1-G(298)) to 1.42 × 10−3 S cm−1 (D-1-R(413)). This work thus gives a good example of a 3d transition-metal complex working as an effective p-type semiconductor and opens new doors for tuning semiconductor properties through crystal transformation.

■

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.8b01841. Single-crystal structures of 1-G and 1-R as well as their DSC data, pictures and PXRD patterns during the crystal transformation process, optimized geometry of the NiL2 molecule and its anion and cation, and the charge transfer pathway in 1-G and 1-R single crystals (PDF)

EXPERIMENTAL SECTION

■

Materials and Measurements. Compounds 1-G and 1-R were synthesized according to the method reported in ref 29. DSC analyses were performed using a Labsys evo TG-DSC/DTA under a constant flow of dry nitrogen gas at a rate of 5 °C/min. Powder X-ray diffraction (PXRD) spectra were recorded on either a D8 Advance (Bruker) or a Rigaku D/max-IIIA diffractometer (Cu Kα, λ = 1.54056

AUTHOR INFORMATION

Corresponding Authors

*E-mail for Y.Z.: [email protected]. *E-mail for M.-H.Z.: [email protected]. E

DOI: 10.1021/acs.inorgchem.8b01841 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry ORCID

tion Polymer Impacts the Electrical Conductivity. Inorg. Chem. 2018, 57, 8029−8032. (15) Wang, C.; Dong, H.; Jiang, L.; Hu, W. Organic semiconductor crystals. Chem. Soc. Rev. 2018, 47, 422−450. (16) Cui, Q.; Bazar, G. C. Narrow Band Gap Conjugated Polyelectrolytes. Acc. Chem. Res. 2018, 51, 202−211. (17) Sato, K.; Ichinoi, R.; Mizukami, R.; Serikawa, T.; Sasaki, Y.; Lutkenhaus, J.; Nishide, H.; Oyaizu, K. Diffusion-Cooperative Model for Charge Transport by Redox-Active Nonconjugated Polymers. J. Am. Chem. Soc. 2018, 140, 1049−1056. (18) Kong, X.; Zhang, X.; Gao, D.; Qi, D.; Chen, Y.; Jiang, J. Airstable ambipolar field-effect transistor based on a solution-processed octanaphthoxy-substituted tris(phthalocyaninato) europium semiconductor with high and balanced carrier mobilities. Chem. Sci. 2015, 6, 1967−1972. (19) Kan, J.; Chen, Y.; Qi, D.; Liu, Y.; Jiang, J. High-performance airstable ambipolar organic field-effect transistor based on tris(phthalocyaninato) europium(III). Adv. Mater. 2012, 24, 1755−1758. (20) Chen, Y.; Su, W.; Bai, M.; Jiang, J.; Li, X.; Liu, Y.; Wang, L.; Wang, S. High performance organic field-effect transistors based on amphiphilic tris(phthalocyaninato) rare earth triple-decker complexes. J. Am. Chem. Soc. 2005, 127, 15700−15701. (21) Zhang, Y.; Cai, X.; Bian, Y.; Jiang, J. Organic Semiconductors of Phthalocyanine Compounds for Field Effect Transistors (FETs). Struct. Bonding (Berlin, Ger.) 2010, 135, 275−321. (22) Beldon, P. J.; Henke, S.; Monserrat, B.; Tominaka, S.; Stock, N.; Cheetham, A. K. Transition metal coordination complexes of chrysazin. CrystEngComm 2016, 18, 5121−5129. (23) Isık, D.; Santato, C.; Barik, S.; Skene, W. G. Charge-Carrier Transport in Thin Films of π-Conjugated Thiopheno-Azomethines. Org. Electron. 2012, 13, 3022−3031. (24) Zhang, M.-Y.; Wang, Z.; Yang, T.; Zhang, Y.; Ma, X.-F.; Sun, Y.-C.; Ouyang, Z.-W.; Kurmoo, M.; Zeng, M.-H. Supramolecular Interactions Direct the Formation of Two Structural Polymorphs from One Building Unit in a One-Pot Synthesis. Chem. - Eur. J. 2016, 22, 13900−13907. (25) Wang, L.; Nan, G.; Yang, X.; Peng, Q.; Li, Q.; Shuai, Z. Computational methods for design of organic materials with high charge mobility. Chem. Soc. Rev. 2010, 39, 423−434. (26) Zhang, Y.; Cai, X.; Bian, Y.; Li, X.; Jiang, J. Heteroatom Substitution of Oligothienoacenes: From Good p-Type Semiconductors to Good Ambipolar Semiconductors for Organic FieldEffect Transistors. J. Phys. Chem. C 2008, 112, 5148−5159. (27) Zhong, A.; Bian, Y.; Zhang, Y. Semiconductor performance of phthalocyaninato Lead complex and its nonperipheral substituted derivatives for organic field effect transistors: density functional theory calculations. J. Phys. Chem. C 2010, 114, 3248−3255. (28) Zhang, Y.; Lahti, P. M.; Maroudas, D. Charge transfer properties of diphenyl substituted cyclopentadithiophene organic semiconductors: the role of fluorine and malononitrile substitutions and crystal ordering. Org. Electron. 2017, 50, 130−137. (29) Na, H.-X.; Yang, P.-Y.; Yin, Z.; Wang, Y.-H.; Chang, L.-X.; Si, R.; Kurmoo, M.; Zeng, M.-H. Stepwise Assembly of MII7 Clusters Revealed by Mass Spectrometry, EXAFS, and Crystallography. Chem. - Eur. J. 2016, 22, 18404−18411. (30) Kamenar, B.; Kaitner, B.; Stefanovic, A.; Waters, T. N. Structures of Three Bis(3-ethoxy-N-R-salieylideneaminato) nickel(ll) Complexes (R = H, Methyl, Ethyl). Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1990, 46, 1627−1631. (31) Chen, Y.; Li, D.; Yuan, N.; Gao, J.; Gu, R.; Lu, G.; Bouvet, M. Tuning the semiconducting nature of bis(phthalocyaninato) holmium complexes via peripheral substituents. J. Mater. Chem. 2012, 22, 22142−22149. (32) Ottenwaelder, X.; Aukauloo, A.; Journaux, Y.; Carrasco, R.; Cano, J.; Cervera, B.; Castro, I.; Curreli, S.; Munoz, M. C.; Rosello, A. L.; Soto, B.; Ruiz-Garcia, R. Synthesis, structure, spectroscopy and redox chemistry of square-planar nickel(II) complexes with tetradentate o-phenylenedioxamidates and related ligands. Dalton Trans 2005, 2516−2526.

Haibing Xu: 0000-0003-3909-414X Yuexing Zhang: 0000-0002-8510-5033 Yanli Chen: 0000-0003-3252-7889 Ming-Hua Zeng: 0000-0002-7227-7688 Author Contributions ∥

Y.-F.W. and S.Z. contributed equally.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the National Science Foundation for Distinguished Young Scholars of China (No. 21525101), the NSFC (Nos. 21571165, 21771192), the NSF of Hubei Province innovation group project (2017CFA006), the NSF of Shandong Province (ZR2017ZB0315), Hubei University, and the China University of Petroleum (East China). This work was also supported by the NSFGX (Grants 2014GXNSFFA118003 and 2017GXNSFDA198040).

■

REFERENCES

(1) Cruz-Cabeza, A. J.; Bernstein, J. Conformational Polymorphism. Chem. Rev. 2014, 114, 2170−2191. (2) Yu, L. Polymorphism in Molecular Solids: An Extraordinary System of Red, Orange, and Yellow Crystals. Acc. Chem. Res. 2010, 43, 1257−1266. (3) Tao, J.; Wei, R.-J.; Huang, R.-B.; Zheng, L.-S. Polymorphism in spin-crossover systems. Chem. Soc. Rev. 2012, 41, 703−737. (4) Zhang, G.; Lu, J.; Sabat, M.; Fraser, C. L. Polymorphism and Reversible Mechanochromic Luminescence for Solid-State Difluoroboron Avobenzone. J. Am. Chem. Soc. 2010, 132, 2160−2162. (5) Almarsson, O.; Zaworotko, M. J. Crystal engineering of the composition of pharmaceutical phases. Do pharmaceutical co-crystals represent a new path to improved medicines? Chem. Commun. 2004, 1889−1896. (6) Zhang, J.-P.; Huang, X.-C.; Chen, X.-M. Supramolecular isomerism in coordination polymers. Chem. Soc. Rev. 2009, 38, 2385−2396. (7) Kawano, M.; Fujita, M. Direct observation of crystalline-state guest exchange in coordination networks. Coord. Chem. Rev. 2007, 251, 2592−2605. (8) Yin, Z.; Zeng, M.-H. Recent advance in porous coordination polymer from the viewpoint of crystalline-state transformation. Sci. China: Chem. 2011, 54, 1371−1394. (9) Yang, T.; Kurmoo, M.; Zeng, M.-H. Recent Advances of SingleCrystal to Single-Crystal Transformation for Discrete Coordination Compounds. J. Indian Inst. Sci. 2017, 97, 299−309. (10) Zeng, M.-H.; Yin, Z.; Tan, Y.-X.; Zhang, W.-X.; He, Y.-P.; Kurmoo, M. Nanoporous cobalt(II) MOF exhibiting four magnetic ground states and changes in gas sorption upon post-synthetic modification. J. Am. Chem. Soc. 2014, 136, 4680−4688. (11) Zhang, M.; Yang, T.; Wang, Z.; Ma, X.-F.; Zhang, Y.; Greer, S. M.; Stoian, S. A.; Ouyang, Z.-W.; Nojiri, H.; Kurmoo, M.; Zeng, M.H. Chemical reaction within a compact non-porous crystal containing molecular clusters without the loss of crystallinity. Chem. Sci. 2017, 8, 5356−5361. (12) Hutchins, K. M.; Rupasinghe, T. P.; Ditzler, L. R.; Swenson, D. C.; Sander, J. R. G.; Baltrusaitis, J.; Tivanski, A. V.; MacGillivray, L. R. Nanocrystals of a Metal−Organic Complex Exhibit Remarkably High Conductivity that Increases in a Single-Crystal-to-Single-Crystal Transformation. J. Am. Chem. Soc. 2014, 136, 6778−6781. (13) Liao, P.-Q.; Huang, N.-Y.; Zhang, W.-X.; Zhang, J.-P.; Chen, X.-M. Controlling guest conformation for efficient purification of butadiene. Science 2017, 356, 1193−1196. (14) Dutta, B.; Dey, A.; Sinha, C.; Ray, P. P.; Mir, M. H. Photochemical Structural Transformation of a Linear 1D CoordinaF

DOI: 10.1021/acs.inorgchem.8b01841 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry (33) Carrasco, R.; Cano, J.; Ottenwaelder, X.; Aukauloo, A.; Journaux, Y.; Ruiz-Garcia, R. Molecular and electronic structure of square-planar nickel(II), nickel(III) and nickel(III) π-cation radical complexes with a tetradentate o-phenylenedioxamidate redox-active ligand. Dalton Trans 2005, 2527−2538. (34) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A., et al. Gaussian 09; Gaussian, Inc.: Wallingford, CT, 2009. (35) Valeev, E. F.; Coropceanu, V.; da Silva, D. A.; Salman, S.; Bredas, J. L. Effect of Electronic Polarization on Charge-Transport Parameters in Molecular Organic Semiconductors. J. Am. Chem. Soc. 2006, 128, 9882−9886.

G

DOI: 10.1021/acs.inorgchem.8b01841 Inorg. Chem. XXXX, XXX, XXX−XXX