Pentacoordinated Organotin(IV) Complexes as an Alternative in the Design of Highly Efficient Optoelectronic and Photovoltaic Devices: Synthesis and Photophysical Characterization

The synthesis of four pentacoordinated organotin(IV) complexes prepared in a one-pot reaction from 2-hydroxy-1-naphthaldehyde, 2-amino-3-hydroxypyridine and organotin oxides is reported. The complexes were characterized by UV-Vis, IR, MS, 1H, 13C and 119Sn NMR techniques. The compound based on 2,2-diphenyl-6-aza-1,3-dioxa-2-stannanaphtho[1,2-h]pyrido[3,2-d]cyclononene revealed the formation of a monomeric complex with a distorted five-coordinated molecular geometry intermediate between the trigonal bipyramidal and square pyramidal. In order to find possible applications in photovoltaic devices, hybrid films of organotin(IV) complexes embedded in poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) with graphene were deposited. The topographic and mechanical properties were examined. The film with the complex integrated into the cyclohexyl substituent has high plastic deformation, with a maximum stress of 1.69 × 107 Pa and a Knoop hardness of 0.061. The lowest values of 1.85 eV for the onset gap and 3.53 eV for the energy gap were obtained for the heterostructure having the complex with the phenyl substituent. Bulk heterojunction devices were fabricated; these devices showed ohmic behavior at low voltages and a space-charge-limited current (SCLC) conduction mechanism at higher voltages. A value of 0.02 A was found for the maximum carried current. The SCLC mechanism suggests hole mobility values of between 2.62 × 10−2 and 3.63 cm2/V.s and concentrations of thermally excited holes between 2.96 × 1018 and 4.38 × 1018 m−3.


Introduction
Thin-film organic photovoltaic technology has been the subject of considerable attention because of the advantages it provides, such as light devices and low preparation cost [1][2][3][4]. Within this field, small-molecule heterojunction solar cells receive more attention because of their clear molecular structure, molecular weight and controllable material purity [5][6][7]. In small-molecule solar cells, the use of fullerenes and phthalocyanines has been widely reported [8,9]; nevertheless, there are a large number of small molecules that are also worthy of study. In this regard, organotin(IV) complexes may be a novel option. Organotin(IV) derivatives have been given special consideration because of the stable bonds formed by tin with carbon atoms, as well as with heteroatoms [10][11][12]. Particular consideration has been given to organotin complexes derived from Schiff bases, since these bases provide a wide variety of molecular structural conformations as ligands [13][14][15][16][17]. Moreover, complexes with Schiff bases show thermo-and photostability, are easy to synthesize and as shown in Scheme 1. The reaction proceeded with yields of 57 to 97%. All complexes were isolated as red solids and were soluble in common organic solvents (chloroform, dichloromethane, dimethyl sulfoxide, methanol and ethanol). The synthesis of complexes 1a and 1c and their in vitro antibacterial activity have been previously described; however, the complexes were obtained in a two-step reaction, and the ligand was refluxed in methanol with dibutyl or diphenyl tin(IV) dichloride in the presence of Et3N as a base. On the other hand, neither the yields nor the complete NMR, mass spectrometry or UV-Vis characterization was reported [39]. Scheme 1. Synthetic strategy for the production of organotin(IV) complexes.
The IR spectra (Supplementary Materials, Figures S1-S8) evidenced the formation of organotin(IV) complexes by the absence of the vibrational stretching band ν(OH) as a result of the deprotonation of the phenolic and naphtholic hydroxyls and organotin(IV) complexation; the expected stretching band ν(C=N) corresponding to azomethine was observed from 1599 to 1602 cm −1 and exhibits a shift to lower wave numbers in comparison to the free ligand, Δδ = 50 cm −1 , indicative of the coordination of the azomethine nitrogen to the tin atom [40,41]. Additional evidence of the coordination of the donor atoms N and O from the ligand to the metal was the presence of the ν(Sn-N) and ν(Sn-O) bands found in the ranges 483-489 cm −1 and 548-550 cm −1 , respectively. For complex 1d, the stretching band ν(Si-C) at 827 cm −1 was observed (IR data are provided in Table S1).
The monomeric nature of complexes 1a-1d was revealed by mass spectrometry using the DART (direct analysis in real time) ion source ( Figures S9-S12). The spectra showed peaks at m/z 497,537, 549 and 557, corresponding to the molecular ions [M +. ] for 1a-1d, respectively. In addition, the expected isotopic peaks where 119 Sn is the most abundant isotope were observed in all cases. The 1 H NMR spectra (Figures S13-S16) for all complexes showed the signals for the expected structures. The absence of hydroxylic protons indicates deprotonation and coordination to the tin atom, in agreement with the FTIR observations. The aromatic region exhibits a single signal for the azomethine proton at 10.46-10.56 ppm and shows satellite signals due to the 3 J( 1 H-117/119 Sn) coupling, which supports Sn-N coordination; the values of the coupling constant were in the range of 49 to 65 Hz. The signals and multiplicities corresponding to the naphtholic and pyridine rings, as well as those corresponding to the organic fragments bonded to tin, were clearly identified; the complete assignment of the individual signals was confirmed by 2D NMR experiments. 13 C NMR for complexes 1a-1d showed (Figures S17-S20) the iminic carbon in the region of 156.5 to 157.8 ppm. Additional evidence of coordination was found in the aliphatic region; complexes 1a and 1b displayed four signals in ranges from 13.6 to 26.9 ppm and from 26.6 to 40.2 for the butyl and cyclohexyl groups bonded to tin. In the case of the 1d complex, the bis(trimethylsilyl)methyl group showed two single signals for methyl and methylene at 0.0 and 7.1 ppm; meanwhile, in the aromatic region for complex 1c from 128.0 to 136.4, signals for phenyl groups attached to tin were identified. HSQC and HMBC correlations permitted the identification of all carbon atoms ( Figures S21-S28). Satellite The IR spectra (Supplementary Materials, Figures S1-S8) evidenced the formation of organotin(IV) complexes by the absence of the vibrational stretching band ν(OH) as a result of the deprotonation of the phenolic and naphtholic hydroxyls and organotin(IV) complexation; the expected stretching band ν(C=N) corresponding to azomethine was observed from 1599 to 1602 cm −1 and exhibits a shift to lower wave numbers in comparison to the free ligand, ∆δ = 50 cm −1 , indicative of the coordination of the azomethine nitrogen to the tin atom [40,41]. Additional evidence of the coordination of the donor atoms N and O from the ligand to the metal was the presence of the ν(Sn-N) and ν(Sn-O) bands found in the ranges 483-489 cm −1 and 548-550 cm −1 , respectively. For complex 1d, the stretching band ν(Si-C) at 827 cm −1 was observed (IR data are provided in Table S1).
The monomeric nature of complexes 1a-1d was revealed by mass spectrometry using the DART (direct analysis in real time) ion source ( Figures S9-S12). The spectra showed peaks at m/z 497,537, 549 and 557, corresponding to the molecular ions [M +. ] for 1a-1d, respectively. In addition, the expected isotopic peaks where 119 Sn is the most abundant isotope were observed in all cases. The 1 H NMR spectra (Figures S13-S16) for all complexes showed the signals for the expected structures. The absence of hydroxylic protons indicates deprotonation and coordination to the tin atom, in agreement with the FTIR observations. The aromatic region exhibits a single signal for the azomethine proton at 10.46-10.56 ppm and shows satellite signals due to the 3 J( 1 H-117/119 Sn) coupling, which supports Sn-N coordination; the values of the coupling constant were in the range of 49 to 65 Hz. The signals and multiplicities corresponding to the naphtholic and pyridine rings, as well as those corresponding to the organic fragments bonded to tin, were clearly identified; the complete assignment of the individual signals was confirmed by 2D NMR experiments. 13 C NMR for complexes 1a-1d showed (Figures S17-S20) the iminic carbon in the region of 156.5 to 157.8 ppm. Additional evidence of coordination was found in the aliphatic region; complexes 1a and 1b displayed four signals in ranges from 13.6 to 26.9 ppm and from 26.6 to 40.2 for the butyl and cyclohexyl groups bonded to tin. In the case of the 1d complex, the bis(trimethylsilyl)methyl group showed two single signals for methyl and methylene at 0.0 and 7.1 ppm; meanwhile, in the aromatic region for complex 1c from 128.0 to 136.4, signals for phenyl groups attached to tin were identified. HSQC and HMBC correlations permitted the identification of all carbon atoms ( Figures S21-S28). Satellite signals were observed in most of the complexes, allowing us to calculate the coupling constants 1 J( 13 C-117/119 Sn), 2 J( 13 C-117-/119 Sn), and 3 J( 13 C-117/119 Sn). From the Holeček equations and the one-bond coupling constants [42,43], C-Sn-C bond angles with values from 120 • to 143 • were obtained. The 119 Sn NMR of organotin(IV) complexes was recorded in CDCl 3 ; the spectra ( Figures S29-S32) showed a single sharp signal in all cases as a result of the formation of discrete species in the solution. It is known that the chemical shift of 119 Sn NMR is shifted toward lower frequencies as the coordination number increases; however, the type of donor group on tin also has an effect. Dibutyl, bis(trimethylsilyl)methyl and dicyclo-hexyltin(IV) complexes exhibited less-negative chemical shifts (−191, −154 and −258 ppm) in comparison to diphenyltin(IV) at −332 ppm. In all cases, the values are indicative of a pentacoordinate environment, in accordance with the literature data [39,44,45].
Suitable crystals for single-crystal X-ray diffraction (SCXRD) were obtained only for complex 1c from dichloromethane. The molecular structure is illustrated in Figure 1. Crystal data are listed in Table 1. Selected bond lengths and angles are summarized in Table 2. Compound 1c crystallizes in the tetragonal space group I4 1 /a. The tin atom in the mononuclear complex 1c is coordinated by two carbon atoms of the phenyl groups, two oxygen atoms derived from the tridentate Schiff base ligand and one nitrogen atom of the imine group. The five-coordinate geometry defined by the C 2 NO 2 donor set is distorted; a measure of distortion away from the ideal trigonal-bipyramidal or square-pyramidal coordination geometries is given by the descriptor τ [46], for which these ideal geometries correspond to τ = 1.0 and 0.0, respectively. The value τ = 0.54 for compound 1c is indicative of an environment that is intermediate between these extremes. The tendency to prefer a square-pyramidal over a trigonal-bipyramidal geometry is most probably associated with packing effects and intermolecular contacts in the solid state. From the angle values (Table 2), the distortion from the trigonal-bipyramidal or square-pyramidal geometry of compound 1c can also be observed from the O(1)-Sn(2)-O(2) angle value [161.1(1) • ], which is smaller than the ideal 180 • . The major distortion in coordination geometry about the tin atom in compound 1c can be associated with the formation of five-membered (Sn1,O2,C16,C12,N1) and six-membered (Sn1,O1,C1,C10,C11,N1) chelate rings, resulting in tight N1-Sn1-O1 [82.78(8) • ] and N1-Sn1-O2 [78.45 (9) • ] chelate angles, which are shorter than 90 • . Thus, the ONO Schiff base behaves as a tridentate coordinating ligand. Fiveand six-membered rings result when the N-Sn coordination bond is formed. The annular tension of the formed cycles and the large covalent radius of the tin(IV) atom restrict the planarity of these complexes. The dihedral angles involving these chelate rings are in the range of 0.18(8) • to 6.70 (9) • , indicating the non-planarity of the ligand plane around the tin atom (see Table 2). dicative of a pentacoordinate environment, in accordance with the li [39,44,45].
Suitable crystals for single-crystal X-ray diffraction (SCXRD) were obta complex 1c from dichloromethane. The molecular structure is illustrated in F tal data are listed in Table 1. Selected bond lengths and angles are summariz Compound 1c crystallizes in the tetragonal space group I41/a. The tin atom nuclear complex 1c is coordinated by two carbon atoms of the phenyl group atoms derived from the tridentate Schiff base ligand and one nitrogen atom group. The five-coordinate geometry defined by the C2NO2 donor set is disto ure of distortion away from the ideal trigonal-bipyramidal or square-pyra nation geometries is given by the descriptor τ [46], for which these ideal ge respond to τ = 1.0 and 0.0, respectively. The value τ = 0.54 for compound 1 of an environment that is intermediate between these extremes. The tenden square-pyramidal over a trigonal-bipyramidal geometry is most probably as packing effects and intermolecular contacts in the solid state. From the angle 2), the distortion from the trigonal-bipyramidal or square-pyramidal geom pound 1c can also be observed from the O(1)-Sn(2)-O(2) angle value [161.1( smaller than the ideal 180°. The major distortion in coordination geometry atom in compound 1c can be associated with the formation of fiv (Sn1,O2,C16,C12,N1) and six-membered (Sn1,O1,C1,C10,C11,N1) chelate ri in tight N1-Sn1-O1 [82.78(8)°] and N1-Sn1-O2 [78.45(9)°] chelate angles, whi than 90°. Thus, the ONO Schiff base behaves as a tridentate coordinating liga six-membered rings result when the N-Sn coordination bond is formed. The sion of the formed cycles and the large covalent radius of the tin(IV) ato planarity of these complexes. The dihedral angles involving these chelate ri range of 0.18(8)° to 6.70(9)°, indicating the non-planarity of the ligand plan tin atom (see Table 2).     Table 2. Selected bond distances (Å) and angles (°) for compound 1c.

Bond Distances (Å)
In compound 1c, the Sn(1)-O(2) and Sn(1)-O(1) distances at the five-and six-membered rings are both 2.09(2) Å, close to the sum of the covalent radii of Sn-O atoms (2.10 Å), but smaller than that of Van der Waals radii for Sn and O atoms (3.75 Å). These values indicate that these complexes possess strong Sn-O bonds. The Sn(1)-N(1) bond distance is 2.15(2) Å and is equal to the sum of the covalent radii for Sn Table 2. Selected bond distances (Å) and angles ( • ) for compound 1c.

Bond Distances (Å)
In compound 1c, the Sn(1)-O(2) and Sn(1)-O(1) distances at the five-and six-membered rings are both 2.09(2) Å, close to the sum of the covalent radii of Sn-O atoms (2.10 Å), but smaller than that of Van der Waals radii for Sn and O atoms (3.75 Å). These values indicate that these complexes possess strong Sn-O bonds. The Sn(1)-N(1) bond distance is 2.15(2) Å and is equal to the sum of the covalent radii for Sn and N, suggesting a strong tin-nitrogen interaction. This value is comparable to those of analogous diorganotin(IV) complexes containing both five-and six-membered chelate rings, including a set of ONO donor atoms [47][48][49][50]. The Sn(1)-C(17) and Sn(1)-C(23) bond lengths are both 2.12(3) Å, and the bond angle C(17)-Sn(1)-C(23) is 128.5(1) • deviates from the ideal 180 • by 51.5 • . These values are in accordance with analogous diorganotin(IV) complexes containing phenyl groups attached to the tin atom with a set of ONO donor atoms [51,52]. The C-Sn-C angle decreases with the size increase of alkyl or aryl groups of diorganotin(IV) moieties.
The structural analysis of 1c also reveals the presence of C-H···N interactions between the hydrogen atom H26A i of the phenyl group attached to the tin atom and the nitrogen atom of the aromatic ring of the pyridine group, with a bond length of 2.722 Å (symmetry code: i 3/4−x, −1/4+y, −1/4+z), forming a 1D polymeric structure ( Figure 2). tin-nitrogen interaction. This value is comparable to those of analogous diorganotin(IV) complexes containing both five-and six-membered chelate rings, including a set of ONO donor atoms [47][48][49][50]. The Sn(1)-C(17) and Sn(1)-C(23) bond lengths are both 2.12(3) Å, and the bond angle C(17)-Sn(1)-C(23) is 128.5(1)° deviates from the ideal 180° by 51.5°. These values are in accordance with analogous diorganotin(IV) complexes containing phenyl groups attached to the tin atom with a set of ONO donor atoms [51,52]. The C-Sn-C angle decreases with the size increase of alkyl or aryl groups of diorganotin(IV) moieties.
The structural analysis of 1c also reveals the presence of C-H···N interactions between the hydrogen atom H26A i of the phenyl group attached to the tin atom and the nitrogen atom of the aromatic ring of the pyridine group, with a bond length of 2.722 Å (symmetry code: i 3/4−x, −1/4+y, −1/4+z), forming a 1D polymeric structure ( Figure 2). The molar conductance was measured in a methanol solution. The values were found in the range of 7.5-35.2 (ohm −1 cm 2 mol −1 ), indicating the non-electrolytic nature of these organotin(IV) compounds; thus, the donor atoms from the ligand are covalently bonded to the metallic center [53]. The UV-Vis absorption spectra of 1a-1d were measured in methanol solution as well ( Figure 3). All complexes showed analogous patterns. The two absorption bands in the regions 214-220 nm (εmax = 30,832-131,423 M −1 cm −1 ) and 237-256 nm (εmax = 17,567-602,014 M −1 cm −1 ) were assigned to the π-π* electronic transition of the aromatic ring; the bands in the ranges 345-371 nm (εmax = 4034-120,862 M −1 cm −1 ) and 481-488 (εmax = 13,274-325,429 M −1 cm −1 ) nm were attributed to π-π* (C=N) and the non-bonding electron pair n-π* (C=N) of the iminic nitrogen, as observed for similar organotin-Schiff-base complexes [44,54].
It is important to consider that, for applications in photovoltaics and organic electronics, the optical and electrical behaviors of the compounds in thin films must be tested, and the manufacturing method will depend mainly on the thermal stability of the complexes. Moreover, one must consider whether heating through Joule effects occurs and leads to device failure [55]. The thermal stability of 1a-1d complexes was determined by differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) over a temperature range from 25 to 550 °C under a nitrogen atmosphere. Figure 4 shows the thermal profiles obtained from the TG/DSC curves for 1a-1d complexes obtained under a nitrogen atmosphere at a heating rate of 10 °C⋅min −1 from 25 to 550 °C. Table 3 shows the key temperature points of the TG and DSC curves: the initial temperature decomposition (T0), the temperature at the peak maximum (Tmax) of every stage, weight loss due to decomposition and the final decomposition temperature (Tf). In general terms, the degradation profiles of organotin(IV) compounds 1a, 1b and 1d occur in two stages within a temperature range of 150 to 520 °C, with the maximum rate of weight loss at Tmax 305, 307 and The molar conductance was measured in a methanol solution. The values were found in the range of 7.5-35.2 (ohm −1 cm 2 mol −1 ), indicating the non-electrolytic nature of these organotin(IV) compounds; thus, the donor atoms from the ligand are covalently bonded to the metallic center [53]. The UV-Vis absorption spectra of 1a-1d were measured in methanol solution as well ( Figure 3). All complexes showed analogous patterns. The two absorption bands in the regions 214-220 nm (ε max = 30,832-131,423 M −1 cm −1 ) and 237-256 nm (ε max = 17,567-602,014 M −1 cm −1 ) were assigned to the π-π* electronic transition of the aromatic ring; the bands in the ranges 345-371 nm (ε max = 4034-120,862 M −1 cm −1 ) and 481-488 (ε max = 13,274-325,429 M −1 cm −1 ) nm were attributed to π-π* (C=N) and the non-bonding electron pair n-π* (C=N) of the iminic nitrogen, as observed for similar organotin-Schiff-base complexes [44,54].
It is important to consider that, for applications in photovoltaics and organic electronics, the optical and electrical behaviors of the compounds in thin films must be tested, and the manufacturing method will depend mainly on the thermal stability of the complexes. Moreover, one must consider whether heating through Joule effects occurs and leads to device failure [55]. The thermal stability of 1a-1d complexes was determined by differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) over a temperature range from 25 to 550 • C under a nitrogen atmosphere. Figure 4 shows the thermal profiles obtained from the TG/DSC curves for 1a-1d complexes obtained under a nitrogen atmosphere at a heating rate of 10 • C·min −1 from 25 to 550 • C. Table 3 shows the key temperature points of the TG and DSC curves: the initial temperature decomposition (T 0 ), the temperature at the peak maximum (T max ) of every stage, weight loss due to decomposition and the final decomposition temperature (T f ). In general terms, the degradation profiles of organotin(IV) compounds 1a, 1b and 1d occur in two stages within a temperature range of 150 to 520 • C, with the maximum rate of weight loss at T max 305, 307 and 332 • C, respectively. In contrast, for complex 1c, three stages of mass loss are identified, with T f at 451 • C. The thermal stability order is 1c > 1a > 1d > 1b; complex 1b has the lowest T 0 (143 • C) and the highest T f (517 • C) as a consequence of a longer degradation process. The DSC analysis shows endothermic peaks from 95 to 121 • C, attributed to the water of crystallization for 1a, 1c and 1d. [56] In the case of 1b, the endothermic peak at 208 • C could be associated with coordinated water. The peaks at 296, 318, 283 and 345 • C for 1a-1d correspond to the highest mass loss due to the ligand moiety and organic groups bonded to the tin atom through the mechanism LSnR2-R2SnO-SnO2 [57]. Interestingly, though this stage implies mass loss according to the TGA curves, the DSC curves show a marked exothermic character. The exothermic character of this stage, contrary to the endothermic character expected for the most prominent thermal degradation stage of these complexes, suggests that the evolved energy corresponds to secondary chemical recombination processes at the gas-gas interface, but not to the energy required for its degradation. This behavior shows the complexity of the process and reveals the concurrency of different competitive and sequential phenomena. However, the complexes show adequate thermal stability, which permits their deposition as thin films with the spin-coating technique. The study of these compounds in the solid state will determine their potential for use in photovoltaic devices.
The DSC analysis shows endothermic peaks from 95 to 121 °C, attributed crystallization for 1a, 1c and 1d. [56] In the case of 1b, the endothermic could be associated with coordinated water. The peaks at 296, 318, 283 an 1d correspond to the highest mass loss due to the ligand moiety and bonded to the tin atom through the mechanism LSnR2-R2SnO-SnO2 [57 though this stage implies mass loss according to the TGA curves, the DSC marked exothermic character. The exothermic character of this stage, con dothermic character expected for the most prominent thermal degradatio complexes, suggests that the evolved energy corresponds to secondary c bination processes at the gas-gas interface, but not to the energy required tion. This behavior shows the complexity of the process and reveals the different competitive and sequential phenomena. However, the compl quate thermal stability, which permits their deposition as thin films with technique. The study of these compounds in the solid state will determin for use in photovoltaic devices.

Topographical and Optical Characterization of Hybrid Film
After depositing the PEDOT:PSS-graphene-pentacoordinated organotin(IV) hybrid films by spin coating, AFM was performed in order to determine both the topography and homogeneity of the hybrid films. Film surface roughness is one of the most important parameters for optoelectronic devices, since roughness causes an increase in film resistance and a decrease in operational performance [58,59]. Table 4 shows the root-meansquare (RMS) roughness; the average roughness (Ra) and AFM images of the surface topography in a 5 × 5 μm area are shown in Figure 5. The highest RMS and Ra roughness occur according to the following trend: 1b > 1c > 1d > 1a. The different roughnesses and topographies in Figure 5 are related to the type of tin complex used. The complex with the cyclohexyl substituent (1b) corresponds to the film with the highest roughness, probably due to poor molecular stacking as a result of the geometry, structure and steric hindrance of the substituent. The roughness of this film is almost twice the roughness of the film with complex 1a, which has the aliphatic butyl radical. This suggests that the 1a complex particles are better embedded in the PEDOT:PSS matrix. The above also applies to the film with complex 1d (phenyl substituent), which has a roughness slightly larger than 1a. Finally, the film with complex 1c shows high roughness, which, although smaller than that of 1b, can affect charge transport in the hybrid film. It seems that the phenyl substituent also affects the incorporation of the complex into the polymeric matrix and its overlap with other organotin complex molecules. Regarding the mechanical properties of the hybrid films, Table 4 shows the unitary deformation (ε), the maximum stress (σmax) and the Knoop microhardness (HK), considering a maximum applied force of 990 N. The film with complex 1b is the one with the highest hardness and the highest mechanical resistance. These mechanical parameters are related to roughness: the highest hardness and mechan-

Topographical and Optical Characterization of Hybrid Film
After depositing the PEDOT:PSS-graphene-pentacoordinated organotin(IV) hybrid films by spin coating, AFM was performed in order to determine both the topography and homogeneity of the hybrid films. Film surface roughness is one of the most important parameters for optoelectronic devices, since roughness causes an increase in film resistance and a decrease in operational performance [58,59]. Table 4 shows the root-mean-square (RMS) roughness; the average roughness (R a ) and AFM images of the surface topography in a 5 × 5 µm area are shown in Figure 5. The highest RMS and Ra roughness occur according to the following trend: 1b > 1c > 1d > 1a. The different roughnesses and topographies in Figure 5 are related to the type of tin complex used. The complex with the cyclohexyl substituent (1b) corresponds to the film with the highest roughness, probably due to poor molecular stacking as a result of the geometry, structure and steric hindrance of the substituent. The roughness of this film is almost twice the roughness of the film with complex 1a, which has the aliphatic butyl radical. This suggests that the 1a complex particles are better embedded in the PEDOT:PSS matrix. The above also applies to the film with complex 1d (phenyl substituent), which has a roughness slightly larger than 1a. Finally, the film with complex 1c shows high roughness, which, although smaller than that of 1b, can affect charge transport in the hybrid film. It seems that the phenyl substituent also affects the incorporation of the complex into the polymeric matrix and its overlap with other organotin complex molecules. Regarding the mechanical properties of the hybrid films, Table 4 shows the unitary deformation (ε), the maximum stress (σ max ) and the Knoop microhardness (HK), considering a maximum applied force of 990 N. The film with complex 1b is the one with the highest hardness and the highest mechanical resistance. These mechanical parameters are related to roughness: the highest hardness and mechanical resistance are directly proportional to the roughness and follow the trend 1b > 1c > 1d > 1a. As for film deformation, the results shown in Table 4 indicate that it is not closely related to roughness; rather, deformation is apparently related to the structure of the organotin(IV) complex.   UV-Vis spectroscopy of the hybrid films was performed to supplement the results obtained in solution. If the use of organotin(IV) complexes in photovoltaic devices is required, it is necessary to understand their solid-state behavior. Regarding the films' transmittance, one aspect to notice in the spectrum in Figure 6a is the low transmittance of the four films. At a larger wavelength of λ > 580 nm, the films with complexes 1a and 1b do not exceed 25% transmittance, film 1d presents around 50%, and the film with complex 1c, which is the most transparent, has around 60%. These values are lower than those presented by hybrid films with PEDOT:PSS and graphene from heptacoordinated organotin(IV) complexes [60], although the film with compound 1c could be used as an elec- UV-Vis spectroscopy of the hybrid films was performed to supplement the results obtained in solution. If the use of organotin(IV) complexes in photovoltaic devices is required, it is necessary to understand their solid-state behavior. Regarding the films' transmittance, one aspect to notice in the spectrum in Figure 6a is the low transmittance of the four films. At a larger wavelength of λ > 580 nm, the films with complexes 1a and 1b do not exceed 25% transmittance, film 1d presents around 50%, and the film with complex 1c, which is the most transparent, has around 60%. These values are lower than those presented by hybrid films with PEDOT:PSS and graphene from heptacoordinated organotin(IV) complexes [60], although the film with compound 1c could be used as an electrode in photovoltaic devices. Additionally, in the spectrum, an absorption band is observed in the region between 430 and 570 nm. This band is attributed to the non-bonding electron pair π-π* (C=N) of the iminic nitrogen in the organotin complexes [44,54]. The presence of PEDOT:PSS and graphene slightly shifts this band toward red with respect to the complexes in solution. On the other hand, the absorption coefficient (α) for the hybrid films is experimentally obtained from [61]: where d is the thickness of the film (see Table 4), and T is the transmittance; reflectivity (R) is neglected due to its low level for each film (2%), which is caused by the presence of graphene in PEDOT:PSS [35]. Figure 6b shows α as a function of hv for the hybrid films. The presence of PEDOT:PSS and graphene considerably decreases film absorption, and the film with the 1d complex has the lowest absorption. It is important to consider that α is in the range 10 3 -10 4 cm −1 , confirming its potential as a transparent material for solar applications [62]. The absorption edge can be characterized in terms of the absorption edge parameters (i) the Urbach energy (EU) and (ii) the energy gap (Eg). The Urbach energy corresponds to the width of the band tail, which is related to localized states within the energy gap, possibly caused by structural defects [61]. The bulk heterojunction formed between the organotin complexes, PEDOT:PSS and graphene, including their interfaces, may be a source of structural defects. Moreover, the behavior of α in the low-energy range follows the exponential law given by Urbach's expression [61,63].

∝ exp ℎ
The EU values of hybrid films were calculated from the slope of the exponential edge and are shown in Table 5. As a reference, the value of EU is zero in a perfect semiconductor [62]. The highest EU belongs to the film with the 1a complex, and the lowest EU belongs to the film with the 1c complex. From these results, the film with the 1c complex shows the best semiconducting behavior with the smallest number of defects when compared with the other films. Apparently, the phenyl substituent favors electron delocalization, a better overlap among the complex molecules and a larger integration into the PEDOT:PSS-graphene matrix. While these EU values are larger than those for some inorganic semiconduc tor films, such as Bi2S3 (0.26 eV) and Bi2−xCrxS3 (0.35 eV) [62], they are nevertheless in the same range as those for inorganic semiconductor films, such as Er-doped ZnO (0.49-0.67 eV) [64]. They are also in the same range as those for hybrid films of the polymer:fullerene blend type, such as polyphenyl-C61-butyric acid methyl ester (0.42-0.48 eV) [61]. Unfor tunately, there is not much information available about the EU values for tin-complex thin The absorption edge can be characterized in terms of the absorption edge parameters: (i) the Urbach energy (E U ) and (ii) the energy gap (E g ). The Urbach energy corresponds to the width of the band tail, which is related to localized states within the energy gap, possibly caused by structural defects [61]. The bulk heterojunction formed between the organotin complexes, PEDOT:PSS and graphene, including their interfaces, may be a source of structural defects. Moreover, the behavior of α in the low-energy range follows the exponential law given by Urbach's expression [61,63].
The E U values of hybrid films were calculated from the slope of the exponential edge and are shown in Table 5. As a reference, the value of E U is zero in a perfect semiconductor [62]. The highest E U belongs to the film with the 1a complex, and the lowest E U belongs to the film with the 1c complex. From these results, the film with the 1c complex shows the best semiconducting behavior with the smallest number of defects when compared with the other films. Apparently, the phenyl substituent favors electron delocalization, a better overlap among the complex molecules and a larger integration into the PEDOT:PSS-graphene matrix. While these E U values are larger than those for some inorganic semiconductor films, such as Bi 2 S 3 (0.26 eV) and Bi 2−x Cr x S 3 (0.35 eV) [62], they are nevertheless in the same range as those for inorganic semiconductor films, such as Er-doped ZnO (0.49-0.67 eV) [64]. They are also in the same range as those for hybrid films of the polymer:fullerene blend type, such as polyphenyl-C61-butyric acid methyl ester (0.42-0.48 eV) [61]. Unfortunately, there is not much information available about the E U values for tin-complex thin films in their different coordination states. E g is the most relevant parameter in evaluating the potential of hybrid films for use in optoelectronic and photovoltaic devices as active layers. It was calculated through Tauc's model, which is based on the linear approximation to the energy axis according to the following relation [65,66]: where h is Planck's constant, and r is a number that characterizes the transition process, with r = 2 for indirect transitions in amorphous films. The frequency (ν) is experimentally obtained from: ν = c λ In this expression, c is the speed of light, and λ is the wavelength. The dependence of (αhν) r on hν was plotted, and E g was evaluated from the x-axis intercept at (αhν) 1/2 = 0 for amorphous structures. The values of E g calculated for the films are shown in Figure 6c and Table 5, and there are important differences among them. Considering that all of the films were deposited under the same conditions and stoichiometric amounts, the changes in their optical behavior can be attributed to the substituent bonded to the tin atom. Additionally, films 1c and 1d show two transitions; the first transition is at the onset gap E onset g , and the second one corresponds to E g [65,66]. The electronic transitions from π to π* explain E g and E onset g as a consequence of several factors, including defects, structural disorder and traps. The two transitions are smaller for the film with complex 1c, which has a phenyl substituent in its structure. The obtained values are of the same magnitude order as those for heptacoordinated tin(IV) complexes embedded in PEDOT:PSS and graphene (E onset g = 1.13-1.29 eV and Eg = 2.83-3.47 eV) [29,30,60]. They are also within the range for inorganic semiconductors. Considering the chemical shifts of 119 Sn NMR, the phenyl group is the one with the highest donor capacity, and this may be the cause of its lower band gap, in addition to the presence of the highly aromatic electrons of the organotin complex. It seems that 1c is the one among the four films with the greatest potential for use in optoelectronic and photovoltaic devices due to the fact that its E g and E U values are the smallest. Film 1c is also the one with the greatest transparency.

Device Fabrication and Electrical Characterization
In order to investigate the electrical properties of the hybrid films, simple devices with only one layer were fabricated, and their current-voltage (I-V) characteristics were measured under different lighting conditions and in darkness. In these devices, the PEDOT:PSSgraphene-organotin(IV) complex hybrid film was deposited over the anode, corresponding to the ITO film. The circuit closes with the silver cathode, which is deposited above the hybrid film (see Figure 7); the electrical behavior is shown in Figure 8. According to their I-V behavior, at low voltages, the devices do not undergo significant modification when subjected to radiation of different wavelengths. Even when enlarging the regions between −0.1 and +0.1 V in the graph, it is observed that there is no effect on device charge transport for the different radiation types. Voltages larger than 0.1 V are required for the ITO-injected holes and Ag-injected electrons to break the hybrid-film barrier, which is probably generated by the complex-PEDOT:PSS-graphene interfaces. In the low-voltage regions, ambipolar behavior can be observed, which changes with increasing voltage, especially in the devices with the 1b and 1d films. Another feature that can be observed in the I-V graphs is a maximum transported current of 0.02 A, which is the same for all of the devices, except for the one with the 1c film, which only carries half of that current. This is an important result, because the 1c film shows the best optical behavior in terms of parameters such as the band gap and Urbach energy, yet it carries the smallest current, which could be due to the energy difference between the work function of the device electrodes and the hybrid film.   As mentioned earlier, at low voltages, different lighting conditions do not significantly influence the ohmic behavior of the devices. This ohmic behavior can be described by the following equation [67]: is the hole mobility, p 0 is the thermally excited hole concentration, e is the electronic charge, V is the applied voltage and d is the film thickness. The ohmic behavior changes, with a slope change located at 0.44 V in the device with complex 1a and transitions in the I-V curve located at around 0.08 V in the other devices. These changes in electrical behavior indicate that the device has reached a space-charge-limited current (SCLC) regime, which is described by an exponential trap distribution [67,68]. By raising the voltage to increase the current, it is possible to create a condition in which the carriers do not migrate fast enough and concentrate in a region of the hybrid film PEDOT:PSS-graphene-organotin(IV) complex. Under these conditions, the device enters the SCLC regime, and its behavior is governed by the following equation [68]: where ε 0 is the vacuum permittivity (8.85 × 10 −14 CV −1 cm −1 ), ε is the dielectric constant of the material, k is Boltzmann's constant, T L is the temperature parameter characterizing the trap distribution and l is the slope of the ohmic zone, which can be obtained by dividing T L by the local temperature. A value of 3 can be assumed for the dielectric constant of the hybrid films, given that this value corresponds to the commonly accepted average value for organic semiconductors [68]. The electrical parameters µ and p 0 , evaluated in ordinary daylight, are shown in Table 6. According to previous studies [28], the organotin(IV) complex in such devices has a carrier mobility of approximately 10 −2 -10 −1 cm 2 /V . s. In this work, the obtained values of µ are quite large compared to similar materials and to heptacoordinated organotin complexes in PEDOT:PSS films [28,69]. They are also similar to PEDOT:PSS films doped with graphene [33], which can be attributed to the presence of graphene in the hybrid films. According to Pasha et al. [70], graphene promotes a screening effect between positively charged PEDOT and negatively charged PSS chains, reducing the coulombic interactions between the chains. At the nanometric level, the reduction in coulombic interactions between PEDOT and PSS chains facilitates phase separation. This allows electrical charges to flow more easily through areas with PEDOT, which is the conductive part of the PEDOT:PSS polymer. Additionally, at the microscopic level, graphene increases the π-π conjugation length in the PEDOT:PSS film, so charge transport increases [71]. From the I-V graphs obtained for each device, it is observed that, while graphene increases charge mobility in the hybrid film, it is the organotin complex, and specifically its substituent, that determines the type and form of electrical behavior in the devices. The electrical properties of the four films indicate the potential of pentacoordinated organotin(IV) complexes for use in photovoltaic devices.

Materials and Methods
All reagents and solvents were obtained from commercial suppliers (Merck KGaA, Darmstadt, Germany) and used without further purification. Cyclohexyl-and bis[(trimethylsilyl)methyl]tin oxides were prepared using the method previously described in the literature [72].

Physical Measurements
Melting points were measured with a Fischer-Johns MEL-TEMP II (Thermo Scientific, Waltham, MA, USA) apparatus and are not corrected. IR spectra were obtained with a Bruker Tensor 27 apparatus (BRUKER, Ettlingen, Germany) by means of the attenuated total reflection (ATR) technique. The NMR spectra of 1 H, 13 C and 119 Sn were acquired using the Bruker Avance III instrument (BRUKER, Rheinstetten, Germany) at 300 MHz, 400 MHz and HD 500 MHz in CDCl 3 . Direct analysis in real time (DART) mass spectra were recorded on a JEOL-JMS-T100LC instrument (JEOL, Tokyo, Japan). The X-ray diffraction study was performed with a BRUKER SMART APEX CCD diffractometer (BRUKER, Karlsruhe, Germany) at a wavelength λ (Mo-Kα) = 0.71073 Å (graphite monochromator) and temperature T = 298 K for complex 1c. The structure was solved using direct methods in the SHELXS program. All non-hydrogen atoms were anisotropically refined using a full-matrix, least-squares technique, and all hydrogen atoms were placed in idealized positions using the riding hydrogen model approximation. Structural solutions and refinements were performed using SHELXTL [73]. The molecular structure and packing were drawn with DI-AMOND [74]. Molar conductivity measurements were recorded with a Hanna 6484 system using anhydrous methanol as a solvent. The UV-Vis spectra were recorded using a Cary 50 Varian instrument with anhydrous methanol as a solvent at 2.04530 × 10 −5 M. The study of the thermal degradation was carried out using simultaneous TGA-DSC (Simultaneous Thermal Analyzer STA 449 F3 Jupiter, NETZSCH-Gerätebau GmbH -Selb, Germany). In each trial, a 4 mg constant-mass sample was placed in a 25 µL aluminum crucible and then heated from room temperature to 550 • C under a nitrogen atmosphere and using a total flow rate of 50 mL min −1 .

General
Procedure for the Synthesis of Diorganotin(IV) Compounds 1a-d 2-Hydroxy-1-naphthaldehyde and the corresponding diorganotin oxide were added in a stoichiometric ratio (1:1:1) to a solution of 2-amino-3-hydroxypyridine. The reaction mixture was refluxed for 30 h using a Dean-Stark trap in the case of complexes 1a and 1d, while the corresponding time for complexes 1b and 1c (containing the phenyl and cyclohexyl moieties) was 48 h. Afterward, the solvent was eliminated under reduced pressure to produce the desired complexes. All complexes were purified by methanol recrystallization.

Hybrid Film Fabrication and Characterization
The hybrid films were deposited by the spin-coating technique, and Smart Coater 200 equipment was used. The dispersion used for the manufacture of the films consisted of 6 mL of a graphene-PEDOT:PSS hybrid dispersion in dimethylformamide. Subsequently, a saturated dispersion was generated with the pentacoordinated organotin(IV) complex. The graphene-PEDOT:PSS-organotin(IV) complex mixture was dispersed with the G560 shaker of Scientific Industries Vortex-Genie. The dispersion was later deposited on the substrate, and the equipment was operated at a constant angular speed of 900 rpm for 20 s and then at an accelerated speed of 300 rpm/s and dried at 80 • C for 3 min. Thin films were deposited on glass, silicon wafers (c-Si) and indium tin oxide (In 2 O 3 ·(SnO 2 ) x )-coated glass (glass-ITO) substrates. Previously, the glass and glass-ITO substrates were sequentially washed in an ultrasonic bath with dichloromethane, methanol and acetone. The silicon substrate was washed with a p solution (10 mL HF, 15 mL HNO 3 and 300 mL H 2 O) to remove surface oxides. In order to obtain the quinoid form in the PEDOT:PSS polymer, the films were post-treated by exposure to isopropanol (IPA) vapor while heated at 40 • C for 10 min. For the hybrid films on Corning glass, the UV-Vis spectra were obtained in the 200-1100 nm wavelength range with a UV-Vis 300 Unicam spectrophotometer. Topographic and mechanical characteristics were investigated with an atomic force microscope (AFM) using a Ntegra platform. Finally, the devices were fabricated by using ITO as an anode and silver as a cathode: glass/ITO/organotin(IV) complex/Ag. For this evaluation, a programmable voltage source, a sensing station with lighting and a temperature-controller circuit from Next Robotix and an auto-ranging Keithley 4200-SCS-PK1 pico-ammeter were employed.

Conclusions
The synthesis of four pentacoordinated organotin(IV) complexes prepared by one-pot synthesis through the reaction of 2-hydroxy-1-naphthaldehyde, 2-amino-3-hydroxypyridine and organotin oxides was carried out. The organotin(IV) complexes were used to fabricate hybrid films with PEDOT:PSS and graphene, and the film with complex 1b (cyclohexyl substituent) had the best mechanical (σ max = 1.69 × 10 7 Pa, HK = 0.061) properties; however, the film with complex 1d ((trimethylsilyl)methyl substituent) had the best ohmic behavior, and the film with complex 1c (diphenyl substituent) had the smallest optical band gap (E onset g = 1.85 and E optical g = 3.53eV) and an Urbach energy of 0.48 eV. The organotin(IV) complex devices had a maximum electrical current of 0.02 A and carrier mobility of 10 −2 -10 −1 cm 2 /V . s. Graphene increases charge mobility in the hybrid film, and the substituent in the organotin complex determines the type and form of electrical behavior in the devices. The substituent in the organotin(IV) complexes also determines the mechanical and optical behaviors of the hybrid films.