SnSe 2 -Zn-Porphyrin Nanocomposite Thin Films for Threshold Methane Concentration Detection at Room Temperature

: Nanocomposite thin ﬁlms, sensitive to methane at the room temperature (25–30 ◦ C), have been prepared, starting from SnSe 2 powder and Zn(II)-5,10,15,20-tetrakis-(4-aminophenyl)-porphyrin (ZnTAPP) powder, that were fully characterized by XRD, UV-VIS, FT-IR, Nuclear Magnetic Resonance ( 1 H-NMR and 13 C-NMR), Atomic Force Microscopy (AFM), SEM and Electron Paramagnetic Resonance (EPR) techniques. Film deposition was made by drop casting from a suitable solvent for the two starting materials, after mixing them in an ultrasonic bath. The thickness of these ﬁlms were estimated from SEM images, and found to be around 1.3 µ m. These thin ﬁlms proved to be sensitive to a threshold methane (CH 4 ) concentration as low as 1000 ppm, at a room temperature of about 25 ◦ C, without the need for heating the sensing element. The nanocomposite material has a prompt and reproducible response to methane in the case of air, with 50% relative humidity (RH) as well. A comparison of the methane sensing performances of our new nanocomposite ﬁlm with that of other recently reported methane sensitive materials is provided. It is suitable for signaling gas presence before reaching the critical lower explosion limit concentration of methane at 50,000 ppm.


Introduction
Methane (CH 4 ) is well known as an odorless gas, and as flammable and combustible in the presence of oxygen. It is one of the components of natural gas and has multiple applications, such as mining, oil and gas operations or home safety. The lower explosion limit (LEL) of methane is 5.0 vol.%, which corresponds to 50,000 ppm.
The classical and versatile SnO 2 gas sensing material was initially developed by researchers from Japan several decades ago [1]. Generally, the gas sensing material is deposited onto planar interdigitated electrodes (IDEs), or onto some functionally equivalent structure, such as coils or meshes. The back side of the IDEs are provided, in most cases, with a continuous electrode for heating, to allow the setting of the working temperature of the sensor, against a given gas which we want to detect.
All of the above studies reveal that through a skillful chemical customization of the porphyrinsbased material in its synthesis stage, it is possible to make the sensor sensible towards specific analyte gases and, therefore, a precise tuning of the gas sensor's selectivity may be attained. Moreover, the good carrier mobility of the SnSe 2 nanocrystals, along with an appropriate metalloporphyrin, makes them attractive for gas sensing composites formation.
As of our knowledge, in spite of all advances so far, studies on the gas sensitivity of hybrid composites based on tin selenide and metalloporphyrin complexes have not yet been reported. In this sense, our paper aims to offer valuable insights on the subject by introducing a new methane-sensitive nanocomposite material, consisting of both SnSe 2 and Zn(II)-5,10,15,20-tetrakis-(4-aminophenyl)-porphyrin (ZnTAPP). The proposed composite was subject to thorough material investigations, as well as to extensive testing in controlled CH 4 atmospheres, and was found to be highly suitable for deployment in low-cost conductometric sensors.

Materials and Methods
In order to prepare the new methane-sensing nanocomposite material we have partly used commercially available high purity solvents, as well as materials synthesized in our laboratories, like the SnSe 2 ingot and the Zn(II)-porphyrin powder, as described below.

Snse 2 Preparation and X-ray Testing
The starting powder has been prepared from small pure Sn bar pieces and pure Se pellets, by melting the appropriate amount of both substances in a quartz ampoulle, which has been sealed under vacuum. The synthesis process took place in a rocking furnace, at a temperature around 700 • C, for two hours. The melt has been cooled down in the ampoulle, and the SnSe 2 powder has been obtained from the cooled-down ingot. The powder sample has been measured by X-ray diffraction (XRD) with the Cu Kα radiation on a Bruker D8 Advance diffractometer, in order to check the formation of the SnSe 2 compound. Figure 1 shows that our powder is polycrystalline SnSe 2 .
As of our knowledge, in spite of all advances so far, studies on the gas sensitivity of hybrid composites based on tin selenide and metalloporphyrin complexes have not yet been reported. In this sense, our paper aims to offer valuable insights on the subject by introducing a new methane-sensitive nanocomposite material, consisting of both SnSe2 and Zn(II)-5,10,15,20--tetrakis-(4-aminophenyl)-porphyrin (ZnTAPP). The proposed composite was subject to thorough material investigations, as well as to extensive testing in controlled CH4 atmospheres, and was found to be highly suitable for deployment in low-cost conductometric sensors.

Materials and Methods
In order to prepare the new methane-sensing nanocomposite material we have partly used commercially available high purity solvents, as well as materials synthesized in our laboratories, like the SnSe2 ingot and the Zn(II)-porphyrin powder, as described below.

Snse2 Preparation and X-ray Testing
The starting powder has been prepared from small pure Sn bar pieces and pure Se pellets, by melting the appropriate amount of both substances in a quartz ampoulle, which has been sealed under vacuum. The synthesis process took place in a rocking furnace, at a temperature around 700 °C, for two hours. The melt has been cooled down in the ampoulle, and the SnSe2 powder has been obtained from the cooled-down ingot. The powder sample has been measured by X-ray diffraction (XRD) with the CuKα radiation on a Bruker D8 Advance diffractometer, in order to check the formation of the SnSe2 compound. Figure 1 shows that our powder is polycrystalline SnSe2.

Zn-Porphyrin Synthesis
The synthesis of Zn(II)-5,10,15,20-meso-tetrakis-(p-amino-phenyl)-porphyrin, was done in three steps, involving the obtaining of 10,15,20-meso-tetrakis-(p-nitro-phenyl)-porphyrin, followed by the reduction in nitro-groups with SnC1 2 [63], and finally by metalation reaction with Zn salt [64]. All of the reagents used in this work have been provided by Sigma-Aldrich and Merck and were purum analiticum grade. A mixture containing 0.44g (2.8 mmol) p-nitrobenzaldehyde and 0.5 mL (5.33 mmol) acetic anhydride was added to 15 mL propionic acid, and brought to reflux under vigorous stirring in a nitrogen gas atmosphere. In the next step, 1 mL of propionic acid, containing 0.20 g (2.8 mmol) pyrrole, was added to the mixture and refluxed for 30 min under stirring. The tarry solution was slowly cooled and kept for 24 h.
The dark solid was separated by filtration, six times washed with 10 mL portions of H 2 O, and dried under vacuum. The powdery solid was mixed with 5 mL of pyridine, refluxed under stirring for 1 h, cooled to room temperature, and then stored at −4 • C overnight. The mixture was filtered and repeatedly washed with acetone to finally yield 150 mg (22%) of 5,10,15,20-meso-tetrakis--(p-nitro-phenyl)-porphyrin. A solution of 150 mg (0.18 mmol) 5,10,15,20-meso-tetrakis-(p-nitro-phenyl)-porphyrin in 10 mL concentrated HCl was bubbled with Ar for 1 h. A solution of 0.675 g (3 mmol) SnC1 2 · 2H 2 O in 2 mL of concentrated HCl, also bubbled with Ar, was added to the porphyrin solution.
The resulting mixture was stirred and heated in a water bath (75-80 • C) for 30 min. The hot-water bath was carefully replaced with a cold-water bath and then an ice-bath (-5 • C). The reaction mixture was then neutralized under Ar by slow addition of 10 mL of concentrated NH 4 OH, taking care to keep the exothermic reaction under control. The resultant basic solution was exposed to air, filtered, and the solid was vigorously stirred with 15 mL of 5% NaOH.
The obtained solid was again filtered, washed with water, dried, and then extracted using Soxhlet apparatus with 20 mL of CHC1 3 . The volume of the purple solution was reduced by rotary evaporation, and after the addition of 10 mL EtOH and the last evaporation, gave dark-green crystalline tetra-(p-amino-phenyl)-porphyrin, TAPP, yield about 75 mg (49%).
The structure of TAPP ( Figure 2) was confirmed by ( 1 H NMR), ( 13 C NMR), FT-IR and UV-VIS spectroscopy. A mixture containing 0.44g (2.8 mmol) p-nitrobenzaldehyde and 0.5 mL (5.33 mmol) acetic anhydride was added to 15 mL propionic acid, and brought to reflux under vigorous stirring in a nitrogen gas atmosphere. In the next step, 1 mL of propionic acid, containing 0.20 g (2.8 mmol) pyrrole, was added to the mixture and refluxed for 30 min under stirring. The tarry solution was slowly cooled and kept for 24 h.
The dark solid was separated by filtration, six times washed with 10 mL portions of H2O, and dried under vacuum. The powdery solid was mixed with 5 mL of pyridine, refluxed under stirring for 1 h, cooled to room temperature, and then stored at −4 °C overnight. The mixture was filtered and repeatedly washed with acetone to finally yield 150 mg (22%) of 5,10,15,20-meso-tetrakis--(p-nitro-phenyl)-porphyrin. The resulting mixture was stirred and heated in a water bath (75-80 °C) for 30 min. The hot-water bath was carefully replaced with a cold-water bath and then an ice-bath (-5 °C). The reaction mixture was then neutralized under Ar by slow addition of 10 mL of concentrated NH4OH, taking care to keep the exothermic reaction under control. The resultant basic solution was exposed to air, filtered, and the solid was vigorously stirred with 15 mL of 5% NaOH.
The obtained solid was again filtered, washed with water, dried, and then extracted using Soxhlet apparatus with 20 mL of CHC13. The volume of the purple solution was reduced by rotary evaporation, and after the addition of 10 mL EtOH and the last evaporation, gave dark-green crystalline tetra-(p-amino-phenyl)-porphyrin, TAPP, yield about 75 mg (49%).

Synthesis of ZnTAPP
A solution comprising 75 mg TAPP (0.11 mmol) dissolved in 70 mL CHCl 3 was brought to reflux and 156 mg (0.93 mmol) Zn(CH 3 COO) 2 dissolved in 3 mL of methanol were added under N 2 atmosphere and refluxed for 4 h at 65 • C. After completion of the reaction, monitored by UV-VIS spectroscopy, the solvent was removed, the excess of Zn salt was thoroughly washed with double distilled water and the product was purified by recrystallization in CHCl 3 (yield, 87%). The compound ZnTAPP, was also represented in Figure 2 and was characterized by FT-IR and UV-VIS spectroscopy, respectively.

Characterization of the Synthesized Tapp and Zntapp Porphyrins
The NMR spectra were performed on a Brucker 400 MHz spectrometer (Germany) and using dimethyl sulfoxide (DMSO)-d6 as solvent. Chemical shifts (δ) are revealed in ppm downfield from the internal standard tetramethylsilane. UV-VIS spectra were recorded on a JASCO UV-V-650 spectrometer (Pfungstadt, Germany), using standard 1 cm pass cells. Atomic force microscopy (AFM) images were performed on a Nanosurf ® EasyScan 2 Advanced Research AFM (Amsterdam, The Netherlands) at room temperature, with samples drop-casted from THF on silica plates in noncontact mode. FT-IR spectra were registered on a JASCO 430 apparatus as KBr pellets.
Electron paramagnetic resonance (EPR) investigations in the X (9.87 GHz) and Q (34.1 GHz) frequency bands were carried out on Bruker ELEXSYS-E580 and ELEXSYS-E500Q spectrometers, respectively. Two porphyrin samples in powder form were inserted into calibrated 2 mm i.d./3 mm o.d. pure silica tubes. The EPR spectra were recorded at room temperature (RT). The determination of the EPR parameters and lineshape simulation of the spectra were performed with the EasySpin v. 5.2.16 program [65].

UV-VIS Spectroscopy
From the UV-VIS spectroscopy (Figure 3), it can be observed that the Soret band of the Zn-metalated compound is bathochromically shifted to 433 nm, as compared to the one for the free base porphyrin, located at 424 nm. A splitting of its Soret band is also noted, the second peak being less intense and located at 394 nm, this is an indicative of a slight H-type process of aggregation. distilled water and the product was purified by recrystallization in CHCl3 (yield, 87%).
The compound ZnTAPP, was also represented in Figure 2 and was characterized by FT-IR and UV-VIS spectroscopy, respectively.

Characterization of The Synthesized Tapp and Zntapp Porphyrins
The NMR spectra were performed on a Brucker 400 MHz spectrometer (Germany) and using dimethyl sulfoxide (DMSO)-d6 as solvent. Chemical shifts (δ) are revealed in ppm downfield from the internal standard tetramethylsilane. UV-VIS spectra were recorded on a JASCO UV-V-650 spectrometer (Pfungstadt, Germany), using standard 1 cm pass cells. Atomic force microscopy (AFM) images were performed on a Nanosurf ® EasyScan 2 Advanced Research AFM (Amsterdam, The Netherlands) at room temperature, with samples drop-casted from THF on silica plates in noncontact mode. FT-IR spectra were registered on a JASCO 430 apparatus as KBr pellets.
Electron paramagnetic resonance (EPR) investigations in the X (9.87 GHz) and Q (34.1 GHz) frequency bands were carried out on Bruker ELEXSYS-E580 and ELEXSYS-E500Q spectrometers, respectively. Two porphyrin samples in powder form were inserted into calibrated 2 mm i.d./3 mm o.d. pure silica tubes. The EPR spectra were recorded at room temperature (RT). The determination of the EPR parameters and lineshape simulation of the spectra were performed with the EasySpin v. 5.2.16 program [65].

UV-VIS Spectroscopy
From the UV-VIS spectroscopy (Figure 3), it can be observed that the Soret band of the Zn-metalated compound is bathochromically shifted to 433 nm, as compared to the one for the free base porphyrin, located at 424 nm. A splitting of its Soret band is also noted, the second peak being less intense and located at 394 nm, this is an indicative of a slight H-type process of aggregation.
The Q bands of the ZnTAPP, located at 566nm and at 611 nm, manifest a hyperchromic effect in comparison with the Q bands of porphyrin-base, and are reduced from 4 in TAPP to only 2 in ZnTAPP, showing an increase in symmetry. The symmetry of the ZnTAPP is changed from D2h to D4h, the cleavage degree of the molecular orbital decreases and the degeneracy increases. As a consequence, the number of Q bands decreases. The Q bands of the ZnTAPP, located at 566nm and at 611 nm, manifest a hyperchromic effect in comparison with the Q bands of porphyrin-base, and are reduced from 4 in TAPP to only 2 in ZnTAPP, showing an increase in symmetry. The symmetry of the ZnTAPP is changed from D2h to D4h, the cleavage degree of the molecular orbital decreases and the degeneracy increases. As a consequence, the number of Q bands decreases.

FT-IR Spectroscopy
In the FT-IR spectra (Figure 4), all compounds revealed a large absorption band around 3300-3500 cm −1 for the stretching vibration of N-H bond, at 1583 cm −1 for ZnTAPP and located at 1599 cm −1 for TAPP, as typical bending vibrational band of the porphyrin group. Beside these frequencies, the absorption bands belonging to C=N, C=C appeared around 1705-1710 cm −1 and 1405-1465 cm −1 , respectively. In addition, the presence of vibrational absorption band at 996 cm −1 was assigned previously for Zn-N [66], and is a proof for the proper synthesis of the metalloporphyrin.

FT-IR Spectroscopy
In the FT-IR spectra (Figure 4), all compounds revealed a large absorption band around 3300-3500 cm −1 for the stretching vibration of N-H bond, at 1583 cm −1 for ZnTAPP and located at 1599 cm −1 for TAPP, as typical bending vibrational band of the porphyrin group. Beside these frequencies, the absorption bands belonging to C=N, C=C appeared around 1705-1710 cm −1 and 1405-1465 cm −1 , respectively. In addition, the presence of vibrational absorption band at 996 cm −1 was assigned previously for Zn-N [66], and is a proof for the proper synthesis of the metalloporphyrin.

EPR Spectroscopy
The EPR spectra of the two porphyrin powder samples measured in the Q-frequency band are displayed in Figure 5.
Similar features, such as the lines characteristic to isolated Cu(II) ions, present as native impurities, can be observed in both samples. Based on the spectra intensity, the concentration of the Cu(II) ions is less than 1 × 10 16 ions/mg. This Cu(II) spectrum is very similar with the spectra observed in other copper porphyrin complexes, where the Cu(II) ions are coordinated by four nitrogen atoms in a square planar configuration [67]. Indeed, as shown in the detailed view from Figure 6, the high-field part of the spectrum displays a complex structure due to the superhyperfine interaction with four nitrogen 14 N nuclei. A good simulation of the Cu(II) spectrum was obtained with the axial EPR parameters and a Gaussian line shape with peak-to-peak linewidth ∆B = 1.5 mT (see Figure 6).

EPR Spectroscopy
The EPR spectra of the two porphyrin powder samples measured in the Q-frequency band are displayed in Figure 5. Similar features, such as the lines characteristic to isolated Cu(II) ions, present as native impurities, can be observed in both samples. Based on the spectra intensity, the concentration of the Cu(II) ions is less than 1 × 10 16 ions/mg. This Cu(II) spectrum is very similar with the spectra observed in other copper porphyrin complexes, where the Cu(II) ions are coordinated by four nitrogen atoms in a square planar configuration [67]. Indeed, as shown in the detailed view from Figure 6, the high-field part of the spectrum displays a complex structure due to the superhyperfine interaction with four nitrogen 14 N nuclei. A good simulation of the Cu(II) spectrum was obtained with the axial EPR parameters g|| = 2.188 ± 0.001, g ┴ = 2.047 ± 0.001, A|| Cu = (197 ± 1) × 10 −4 cm −1 , A ┴ Cu = (30 ± 1) × 10 −4 cm −1 , A|| N = A ┴ N = (16 ± 1) × 10 −4 cm −1 and a Gaussian line shape with peak-to-peak linewidth ΔB = 1.5 mT (see Figure 6).    Similar features, such as the lines characteristic to isolated Cu(II) ions, present as native impurities, can be observed in both samples. Based on the spectra intensity, the concentration of the Cu(II) ions is less than 1 × 10 16 ions/mg. This Cu(II) spectrum is very similar with the spectra observed in other copper porphyrin complexes, where the Cu(II) ions are coordinated by four nitrogen atoms in a square planar configuration [67]. Indeed, as shown in the detailed view from Figure 6, the high-field part of the spectrum displays a complex structure due to the superhyperfine interaction with four nitrogen 14 N nuclei. A good simulation of the Cu(II) spectrum was obtained with the axial EPR parameters g|| = 2.188 ± 0.001, g ┴ = 2.047 ± 0.001, A|| Cu = (197 ± 1) × 10 −4 cm −1 , A ┴ Cu = (30 ± 1) × 10 −4 cm −1 , A|| N = A ┴ N = (16 ± 1) × 10 −4 cm −1 and a Gaussian line shape with peak-to-peak linewidth ΔB = 1.5 mT (see Figure 6).  The broad line observed around 450 mT in the Q-band ZnTAPP spectrum ( Figure 5) is very probably due to isolated Fe(III) ions present as native impurities in a very low concentration. This assignment is supported by the two lines observed in the X-band spectrum at the effective g-values 4.27 and 5.5, respectively ( Figure 5, inset), which could be ascribed to a Fe(III) complex with a S = 5/2, 3/2 spin admixed ground state [68].
A slightly asymmetric line can also be observed at g~2.0028 in TAPP and g~2.0030 in ZnTAPP, respectively ( Figure 5), with a peak-to-peak linewidth ∆B~1.0 mT, originating most probably from some radicals. The intensity of this line is four times larger in the ZnTAPP sample than in the TAPP sample. It is very possible that these stable radicals observed in the EPR spectra of the two samples, characterized by different g-values and intensities, are associated with different cations, such as: Cu 2+ or Fe 3+ , present as native impurities in the Sn and Zn salts used in metalation reaction for obtaining of TAPP porphyrin base and its Zn-metalloporphyrin, respectively.

Surface Morphology Characterization by AFM Microscopy
As already expected from the UV-VIS spectra that showed some H-type aggregation phenomena regarding the ZnTAPP, the AFM images (Figure 7a,b) proved a stronger aggregation in the case of metallated compounds. The defined architecture of triangular shape formed in TAPP (Figure 7a) is changed into a kvatarons-type structure [69], of larger aggregates (Figure 7b), obtained by hydrophobic interactions. assignment is supported by the two lines observed in the X-band spectrum at the effective g-values 4.27 and 5.5, respectively ( Figure 5, inset), which could be ascribed to a Fe(III) complex with a S = 5/2, 3/2 spin admixed ground state [68].
A slightly asymmetric line can also be observed at g ~ 2.0028 in TAPP and g ~ 2.0030 in ZnTAPP, respectively ( Figure 5), with a peak-to-peak linewidth ΔB ~ 1.0 mT, originating most probably from some radicals. The intensity of this line is four times larger in the ZnTAPP sample than in the TAPP sample. It is very possible that these stable radicals observed in the EPR spectra of the two samples, characterized by different g-values and intensities, are associated with different cations, such as: Cu 2+ or Fe 3+ , present as native impurities in the Sn and Zn salts used in metalation reaction for obtaining of TAPP porphyrin base and its Zn-metalloporphyrin, respectively.

Surface Morphology Characterization by AFM Microscopy
As already expected from the UV-VIS spectra that showed some H-type aggregation phenomena regarding the ZnTAPP, the AFM images (Figure 7a,b) proved a stronger aggregation in the case of metallated compounds. The defined architecture of triangular shape formed in TAPP (Figure 7a) is changed into a kvatarons-type structure [69], of larger aggregates (Figure 7b), obtained by hydrophobic interactions.

Nano-Ccomposite Thin Film Preparation
As starting materials we used SnSe2 and Zn(II)-5,10,15,20-tetrakis-(4-aminofenil)-porphyrin, both in powder form. They were dissolved in amine-and methyl-containing solvents [70], in concentrations of 10 −3 −10 −2 mol/L. The two obtained solutions were labelled with S for SnSe2 and with ZP for the ZnTAPP. We ultrasonically mixed the two solutions, and the resulting solution we have labelled with SZP. From this final solution we have drop casted, with an Eppendorf 200 analytical pipette, about 30 micro-liters onto an interdigited sensor support.
The samples, obtained as previously described, have been dried in an oven for 60-100 min, at 60-80 °C, resulting in a continuous and uniform thin grey layer, which covered well the interdigited electrodes ( Figure 8).

Nano-Ccomposite Thin Film Preparation
As starting materials we used SnSe 2 and Zn(II)-5,10,15,20-tetrakis-(4-aminofenil)-porphyrin, both in powder form. They were dissolved in amine-and methyl-containing solvents [70], in concentrations of 10 −3 −10 −2 mol/L. The two obtained solutions were labelled with S for SnSe 2 and with ZP for the ZnTAPP. We ultrasonically mixed the two solutions, and the resulting solution we have labelled with SZP. From this final solution we have drop casted, with an Eppendorf 200 analytical pipette, about 30 micro-liters onto an interdigited sensor support.
The samples, obtained as previously described, have been dried in an oven for 60-100 min, at 60-80 • C, resulting in a continuous and uniform thin grey layer, which covered well the interdigited electrodes ( Figure 8). 4.27 and 5.5, respectively ( Figure 5, inset), which could be ascribed to a Fe(III) complex with a S = 5/2, 3/2 spin admixed ground state [68].
A slightly asymmetric line can also be observed at g ~ 2.0028 in TAPP and g ~ 2.0030 in ZnTAPP, respectively ( Figure 5), with a peak-to-peak linewidth ΔB ~ 1.0 mT, originating most probably from some radicals. The intensity of this line is four times larger in the ZnTAPP sample than in the TAPP sample. It is very possible that these stable radicals observed in the EPR spectra of the two samples, characterized by different g-values and intensities, are associated with different cations, such as: Cu 2+ or Fe 3+ , present as native impurities in the Sn and Zn salts used in metalation reaction for obtaining of TAPP porphyrin base and its Zn-metalloporphyrin, respectively.

Surface Morphology Characterization by AFM Microscopy
As already expected from the UV-VIS spectra that showed some H-type aggregation phenomena regarding the ZnTAPP, the AFM images (Figure 7a,b) proved a stronger aggregation in the case of metallated compounds. The defined architecture of triangular shape formed in TAPP (Figure 7a) is changed into a kvatarons-type structure [69], of larger aggregates (Figure 7b), obtained by hydrophobic interactions.

Nano-Ccomposite Thin Film Preparation
As starting materials we used SnSe2 and Zn(II)-5,10,15,20-tetrakis-(4-aminofenil)-porphyrin, both in powder form. They were dissolved in amine-and methyl-containing solvents [70], in concentrations of 10 −3 −10 −2 mol/L. The two obtained solutions were labelled with S for SnSe2 and with ZP for the ZnTAPP. We ultrasonically mixed the two solutions, and the resulting solution we have labelled with SZP. From this final solution we have drop casted, with an Eppendorf 200 analytical pipette, about 30 micro-liters onto an interdigited sensor support.
The samples, obtained as previously described, have been dried in an oven for 60-100 min, at 60-80 °C, resulting in a continuous and uniform thin grey layer, which covered well the interdigited electrodes ( Figure 8). The surface morphology of the SnSe 2 and Zn-metalloporphyrin powder precursors, and the final gas sensing nanocomposite, have been studied by scanning electron microscopy (SEM), and are resented in Figure 9a-c. The film thickness was estimated from an SEM image taken in side view. It was found to be around 1.3 micrometers.
Let us observe that SnSe 2 shows a fragmented polycrystalline surface (Figure 9a), while the Zn-porphyrin features a spongeous-like network surface structure with smaller structural motifs ( Figure 9b). The surface arrangement of porphyrins induces modifications in the sensitivity, with respect to the non-aggregated porphyrins [71,72]. On Figure 9c one may see how the first two structures form our new nanocomposite, where the finer Zn-porphyrin covers the coarse SnSe 2 polycrystalline surface features. The merge of these two precursor structures with different rugosities proves to yield a nanocomposite with enhanced gas sensing ability for methane, compared to the very poor responses of the precursors tested individually.

IDEs.
The surface morphology of the SnSe2 and Zn-metalloporphyrin powder precursors, and the final gas sensing nanocomposite, have been studied by scanning electron microscopy (SEM), and are resented in Figure 9a-c. The film thickness was estimated from an SEM image taken in side view. It was found to be around 1.3 micrometers.
Let us observe that SnSe2 shows a fragmented polycrystalline surface (Figure 9a), while the Zn-porphyrin features a spongeous-like network surface structure with smaller structural motifs (Figure 9b). The surface arrangement of porphyrins induces modifications in the sensitivity, with respect to the non-aggregated porphyrins [71,72]. On Figure 9c one may see how the first two structures form our new nanocomposite, where the finer Zn-porphyrin covers the coarse SnSe2 polycrystalline surface features. The merge of these two precursor structures with different rugosities proves to yield a nanocomposite with enhanced gas sensing ability for methane, compared to the very poor responses of the precursors tested individually.

Testing The Sensor Material for Methane at Room Temperature
The resulting sensor was introduced in a designated testing chamber and kept completely isolated from the external environment for the whole duration of the test. Although, the temperature and pressure conditions inside the chamber can be strictly controlled and monitored, the SZP thin film sensor performed outstandingly at room temperatures of 23-28 °C.
The testing protocol implies applying an electrical current to the sensor and measuring the voltage variations, a simple but reliable method of deriving the behavior of its electrical resistance in the presence of an analyte gas. For this purpose, we used a source measure unit (SMU), a Keithley SMU 2450, connected directly to the gas chamber and, in the particular case of the SZP, it was determined that 1µ A is an optimal test current. The analyte gas -the methane -was introduced in the chamber in a pre-established concentration (1000 ppm) and at well set time intervals (10 min), through the designated Alicat Scientific mass flow controllers that regulate the gas flow to the test chamber. The testing protocol requires for the analyte to be alternated with a gas to which the sensor is insensitive, namely dry synthetic air, in order to prove the reproducibility of the sensor's response to methane. We maintained the synthetic air alternations for the same duration of 10 min. This cycle has been repeated for about ten times. The variation of the resistance during this test is given in Figure 10 below.
One can observe, on Figure 10, that the methane sensing film has a reproducible, prompt and systematic response to the presence of the methane: as 1000 ppm methane is added to the flowing dry air, the value of the electrical resistance of the film suddenly begins to drop sharply. As soon as the methane flux is stopped, the value of the electrical resistance begins to grow, again very sharply. The magnitude of the resistance's relative drop upon the presence of the methane, and the magnitude of the relative increase in the resistance as the methane flux is stopped is around 5%.

Testing The Sensor Material for Methane at Room Temperature
The resulting sensor was introduced in a designated testing chamber and kept completely isolated from the external environment for the whole duration of the test. Although, the temperature and pressure conditions inside the chamber can be strictly controlled and monitored, the SZP thin film sensor performed outstandingly at room temperatures of 23-28 • C.
The testing protocol implies applying an electrical current to the sensor and measuring the voltage variations, a simple but reliable method of deriving the behavior of its electrical resistance in the presence of an analyte gas. For this purpose, we used a source measure unit (SMU), a Keithley SMU 2450, connected directly to the gas chamber and, in the particular case of the SZP, it was determined that 1 µA is an optimal test current. The analyte gas-the methane-was introduced in the chamber in a pre-established concentration (1000 ppm) and at well set time intervals (10 min), through the designated Alicat Scientific mass flow controllers that regulate the gas flow to the test chamber. The testing protocol requires for the analyte to be alternated with a gas to which the sensor is insensitive, namely dry synthetic air, in order to prove the reproducibility of the sensor's response to methane. We maintained the synthetic air alternations for the same duration of 10 min. This cycle has been repeated for about ten times. The variation of the resistance during this test is given in Figure 10 below.
One can observe, on Figure 10, that the methane sensing film has a reproducible, prompt and systematic response to the presence of the methane: as 1000 ppm methane is added to the flowing dry air, the value of the electrical resistance of the film suddenly begins to drop sharply. As soon as the methane flux is stopped, the value of the electrical resistance begins to grow, again very sharply. The magnitude of the resistance's relative drop upon the presence of the methane, and the magnitude of the relative increase in the resistance as the methane flux is stopped is around 5%.
To simulate on-site real air composition, controlled 50% relative humidity conditions were attained using a controlled evaporation and mixing (CEM) unit, model Bronkhorst part of the gas sensors testing facility. The CEM-System consists of a thermal liquid flow controller, a mass flow controller for the carrier gas, and a temperature controlled mixing and evaporation device. The resident CEM is suitable for mixing liquid flows up to 30 g/h, resulting in saturated vapor flows of up to 4 L/min. The sensor's responses to 1000 ppm of CH 4 in controlled 50% relative humidity conditions are given in Figure 11.
We can notice from Figures 10 and 11, that the presence of 50% RH in the testing air results in a roughly 2.5 times lower response magnitude of the nanocomposite to methane, but the shape of the response preserves its periodicity, following the on and off switching of the methane promptly. This response is suitable as an input to an electronic signal processor and analyzer unit for various application purposes. To simulate on-site real air composition, controlled 50% relative humidity conditions were attained using a controlled evaporation and mixing (CEM) unit, model Bronkhorst part of the gas sensors testing facility. The CEM-System consists of a thermal liquid flow controller, a mass flow controller for the carrier gas, and a temperature controlled mixing and evaporation device. The resident CEM is suitable for mixing liquid flows up to 30 g/h, resulting in saturated vapor flows of up to 4 L/min. The sensor's responses to 1000 ppm of CH4 in controlled 50% relative humidity conditions are given in Figure 11. We can notice from Figures 10 and 11, that the presence of 50% RH in the testing air results in a roughly 2.5 times lower response magnitude of the nanocomposite to methane, but the shape of the response preserves its periodicity, following the on and off switching of the methane promptly. This response is suitable as an input to an electronic signal processor and analyzer unit for various application purposes.
This sensing material reacts to the presence or absence of methane through changing not only the value of the equivalent resistance, but also the slope's sign. For the later we have a very quick time response, of about 5-10 s, having a non-linear decrease. From Figure 11 we can see that the slope of the curve changes almost instantly from slightly positive to steeply negative, as methane contacts the sensing material at minute 30. As the sensing material's surface saturates, the steepness To simulate on-site real air composition, controlled 50% relative humidity conditions were attained using a controlled evaporation and mixing (CEM) unit, model Bronkhorst part of the gas sensors testing facility. The CEM-System consists of a thermal liquid flow controller, a mass flow controller for the carrier gas, and a temperature controlled mixing and evaporation device. The resident CEM is suitable for mixing liquid flows up to 30 g/h, resulting in saturated vapor flows of up to 4 L/min. The sensor's responses to 1000 ppm of CH4 in controlled 50% relative humidity conditions are given in Figure 11. We can notice from Figures 10 and 11, that the presence of 50% RH in the testing air results in a roughly 2.5 times lower response magnitude of the nanocomposite to methane, but the shape of the response preserves its periodicity, following the on and off switching of the methane promptly. This response is suitable as an input to an electronic signal processor and analyzer unit for various application purposes.
This sensing material reacts to the presence or absence of methane through changing not only the value of the equivalent resistance, but also the slope's sign. For the later we have a very quick time response, of about 5-10 s, having a non-linear decrease. From Figure 11 we can see that the slope of the curve changes almost instantly from slightly positive to steeply negative, as methane contacts the sensing material at minute 30. As the sensing material's surface saturates, the steepness This sensing material reacts to the presence or absence of methane through changing not only the value of the equivalent resistance, but also the slope's sign. For the later we have a very quick time response, of about 5-10 s, having a non-linear decrease. From Figure 11 we can see that the slope of the curve changes almost instantly from slightly positive to steeply negative, as methane contacts the sensing material at minute 30. As the sensing material's surface saturates, the steepness of the negative slope diminishes, but remains negative at 40 min. If we consider for time response, the time to reach 50% of the starting peak-to-valley value, then again, we can see about the same value. In our testing cycle, at minute 40 the methane is stopped, and the recovery process begins, and evolves until minute 50. Disregarding the signal's noise, we can observe an almost complete recovery of the signal to the starting value from minute 30. This way, one can estimate a time to full recovery of about 10 min. The time to 50% recovery is around 2.5-3 min only, since the recovery is also not linear.
After the recovery times, the response returns to almost the same value as in the absence of gas. This point forms the basis for the high degree of reversibility of the proposed device.
It is worth noting that this nanocomposite shows also a modest response to methane already from a concentration as low as 100 ppm, but the magnitude of the sensing material's response at 1000 ppm, presented above in Figures 10 and 11, is much better. At this stage it is suitable for signaling gas presence, before reaching the critical LEL concentration of methane at 50,000 ppm [73]. Industrial environments and mining sectors involving methane presence might benefit from the further development of such a threshold sensing material for early warning.

Discussion
SnSe 2 has a layered structure, consisting of triple-layer units, where the Sn layer sits between two Se layers [74]. In this way, the triple-layer units will face toward the neighboring next unit through Se atoms. Due to the lone pair electrons of the Se atoms, it seems plausible to admit that the electron density is higher in these Se-Se sheet-like interfaces, compared to the inner volume of the triple-layer unit, which includes the Sn sheet. We consider that this modulated electron density feature will support the electrical response of the SnSe 2 during the gas sensing process, through the available electrons in these interface volumes. Further on, we tried to enhance the electrical responsivity of SnSe 2 by making the nanocomposite material of SnSe 2 and ZnTAPP, knowing the remarkable properties of the Zn-porphyrins [75,76].
As a general chemical behavior, zinc porphyrins have a pentadentate coordination mode [36], which allowed only one more axial ligand to be complexed to the zinc atom (possessing a close d shell) in the center of porphyrin. Nevertheless, it was demonstrated in [36] that several zinc porphyrins can become six-coordinate, binding in solution two axial ligands (for example pyridine), if the resulting complex is suitably stabilized by chelation.
The influence of the central metal-especially its radius, electrostatic attraction and steric effects on the properties of porphyrins, as well as the influence of the used solvent were deeply investigated [77], taking into consideration that in solvated porphyrins the solvent molecules cause the distortion of the ring skeleton, and thus reducing the porphyrin's symmetry. So, even in solution, the ability of zinc porphyrins to form six-fold coordination complexes is extremely limited and an increased distortion in porphyrin conformation is not an indication of an enhanced interaction.
Thus, it is more likely that the zinc porphyrin prefer to link only one axial ligand, being more performant when the concentration of the gas is low, as revealed in this study.
Based on Pearson's principle [78], a strong interaction can be possible between two hard (i.e., weakly polarizable) or two soft (i.e., highly polarizable) Lewis' bases and acids. Thus, the volatile molecules that possess lone pairs of electrons, such as alcohols and amines, favor axial coordination with the zinc porphyrins, this effect being prevalent over hydrogen bonding formation [79]. As a consequence, in Zn-porphyrins the binding energy is increased by the weakly electron donating carbon atoms, such as methane. The exposure of the film to the CH 4 gas will lead to distortion of molecules, ending with returning to the monomeric state when the concentration of CH 4 is high [80].
In the presence of humidity, the sensing is more difficult, because water molecules compete with CH 4 gas for the coordination site. Taking into consideration the methodology used to prepare our sensor, the CH 4 analyte might interact with both the porphyrin, and the SnSe 2 semiconductor. Even so, we consider that the sensitivity of the sensor is mainly due to the interaction with the porphyrins, because it is known that the conductivity of porphyrins, surrounded by SnSe 2 semiconductors, is influenced by adsorbed molecules that also affect the extended π-π delocalization, decreasing it in this way [72].
The surface of the sensing nanocomposite layer, exposed to the analyzed gas, able to get in contact with its molecules, is mostly formed from Se atoms from the edges of the-Se-Sn-Se-layer stacks, or from the central Zn atoms of the Zn-porphyrin molecules, forming practically an electron-rich surface.
In our present understanding, the complex structure of our nanocomposite, and the results we have so far, makes it difficult to propose a comprehensive and plausible sensing mechanism. However, we think that the interplay between the layer-like structures of both the SnSe 2 and the Zn-prophyrin, the lone-pair electrons of the Se, and the electronegativity values of each kind of the interacting atoms have an important role in this sensing mechanism.
The gas sensing mechanism was previously described [27] as related to the charge transfer (decreasing in carrier concentration) and the flat band (reducing in the mobility) induced by gas adsorption. The electron-donating character of CH 4 is specifically indicated. We might further point out the fact that direct transfer (physisorption) of electrons from CH 4 occurs at low temperatures with less ability of charge transfer.
One major argument is that only chalcogenide atoms are associated with Lone-Pair (LP) states. The energy of LP states strongly depends on the chemical environment, so LP electrons close to electropositive atoms and donor electron effect molecules (such as CH 4 ) will have higher energy than those near electronegative atoms, having a withdrawing electron effect. So donor electron effect molecules, such as CH4, will raise some LP states into the gap, thus enlarging the valence band tail. On the other hand, it is well known that higher Se concentration determines LP band to become the valence band. This behavior of SnSe 2 is strongly potentiated by the ZnTAPP that, due to its four NH 2 substituent groups, is additionally increasing the donor electron effect on the calcogenide atoms.
Based on these, we can presume that methane sensing is dependent upon the above mentioned compositional disordering, meaning that conductivity is achieved in this case mainly by hopping between localized states in the extended band tails [84].

Conclusions
We obtained a new nanocomposite based on SnSe 2 and ZnTAPP as drop-casted thin films. These thin films proved to be sensitive to methane (CH 4 ) in concentrations as low as 1000 ppm, at a room temperature of about 25 • C, without the need of heating the sensing element. Since for methane there are known [85] the lower explosion limit (LEL) concentration of 5 vol%, corresponding to 50,000 ppm, and the upper explosion limit (UEL) of 15 vol%, corresponding to a concentration of 150,000 ppm, the nanocomposite material proposed by us here and in [26], allows for an early warning if methane might be present in a given monitored area. Table 1 presents data enabling the comparison of the methane sensing performance of our new nanocomposite with other methane sensitive materials. We have checked for the reproducibility of the sensor's response to methane in dry air with other sensors, made in the same way. The peak-to-valley amplitude, as well as the global tendency in response (i.e., the decrease in the presence of methane, increase without methane), was also the same.
Even if the magnitude of the resistive response of this new nanocomposite thin film is not as large as those of other classical sensors, the fact that it works at room temperature, without the need of a heating element, makes it a good choice to be considered for further sensor development. At this stage it is suitable for signaling gas presence before reaching the critical lower explosion limit (LEL) concentration threshold of methane at 50,000 ppm [84]. Industrial environments and mining sectors involving methane presence could benefit from the further development of such a threshold sensing material for early warning.

Conflicts of Interest:
The authors declare no conflict of interest with the research performed and reported here.