Organic materials are currently the main option for electronic and/or photonic devices, as they can impart multifunctionality, flexibility, transparency, and sustainability to emerging systems, such as portable / wearable optoelectronics. While functional organic materials include small molecules, oligomers, and polymers, transforming them in high quality thin films or multilayers [43] is indeed difficult. In the MAPLE method, a laser beam strikes a target, which is made of a frozen solution and when the laser beam hits the target, a plume of solvent and polymer is formed. The process takes place under controlled high vacuum or background gas, so the solvent is pumped away and just the desired material is collected on a substrate, gradually forming a thin film or a multilayer. Thioamides have been theoretically and experimentally shown to exhibit remarkable nonlinear optical properties. Since they could be cost-effectively synthesized and made into high-quality thin or multilayer films for optoelectronic applications, we considered them to be candidates for second harmonic generation (SHG).
Synthesis Of The Compounds
Thioamides α, β-unsaturated I − VII were synthesized using Pappalardo synthesis, a Claisen – Schmidt type condensation of thioacetamide with various aromatic substituted aldehydes in 10% ethanolic EtONa solution to obtain the desired target compounds, as presented in Fig. 2. The structure of the synthesized compounds is presented in in Fig. 3, and were subsequently confirmed by their 1H NMR and 13C NMR spectra.
The main features of the 1H-NMR spectra of the compounds I―VII are the signals corresponding to the doublet of doublets belonging to the vinyl group CH = CH, which appear in the range 7.01–7.90 ppm. The two thioamide protons are the most deshielded protons in the spectrum, and occur in the range of 9.37 − 9.87 ppm. Also, aromatic protons appear in the range of 7.16-8.83ppm, the most deshielded protons being those belonging to the pyridine nucleus of compound VII.
The 13C-NMR spectra confirm the structure of the thioacetamides I―VII. Also, the two vinyl carbons are highlighted, being the most deshielded in the spectrum, with chemical shifts of 196.35-196.4ppm.
The GAMESS program was used to optimize the molecular geometries of thioamides. The natural internal coordinates generated by GAMESS are determined by the geometry obtained by calculations. The energetic values of the HOMO and LUMO levels (Fig. 4) are at about the same value − 0.303 -316 Ha and − 0.006–0.015 Ha respectively, in the gas phase, for the compounds II, III, and V. Also, for these compounds, one may notice a high value of the hyperpolarizability βtot., as presented in Table 1 and depicted in Table 2.
Table 1. Chemical reactivity indices, total energy, dipole moment and first order hyperpolarizabilities βtot of the thioamides.
Properties/ thioamide
|
Etot (Kcal/Mol)
|
EHOMO (Ha)
|
ELUMO (Ha)
|
Egap (Ha)
|
DM (D)
|
btotal (esu)
|
I
|
-35783.093
|
-0.313
|
-0.012
|
0.301
|
4.62
|
10.432*10-30
|
II
|
-45996.891
|
-0.303
|
-0.006
|
0.297
|
4.467
|
26.308*10-30
|
III
|
-47416.464
|
-0.309
|
-0.015
|
0.294
|
4.951
|
17.081*10-30
|
IV
|
-36362.882
|
-0.308
|
-0.011
|
0.297
|
4.531
|
8.847*10-30
|
V
|
-42733.132
|
-0.316
|
-0.021
|
0.295
|
6.342
|
465.165*10-30
|
VI
|
-52648.601
|
-0.324
|
-0.036
|
0.288
|
9.436
|
10.537*10-30
|
VII
|
-36433.535
|
-0.324
|
-0.031
|
0.294
|
5.092
|
7.776*10-30
|
A low HOMO-LUMO gap is reflected in a molecule’s behaviour, i.e. making it very reactive and therefore less stable in general [44, 45]. The energy value Egap is the lowest for thioamide III, which possessed the smallest \({\beta }\)tot value 17.081*10− 30, of 0.294 Ha, but almost the same small Egap, of 0.295 Ha exhibits thioamide V, with the larger value of \({\beta }\)tot, of 465.165*10− 30. Therefore, it can be expected that it will exhibit the largest nonlinear optical effect. Also, for compound II, with an intermediate value of hyperbolizability \({\beta }\)tot, of 26.308*10− 30, we expect good nonlinear optical properties.
The low values of the hyperpolarizabilities of compounds I, IV, VI, and VII, are reflected in their low nonlinear optical properties, which have not been evidenced experimentally. A small energy difference Egap HOMO − LUMO supported by a very high hyperpolarizability βtot, as well as a distribution of the charge density on the HOMO − LUMO orbitals as small as possible, is the condition of the appearance of nonlinear optical properties in the case of studied thioamides, as it will further be discussed for compounds II, III and V, in the next section.
Thin Films Deposition By Maple
Laser techniques are already facilitating environmental and eco-friendly designs through to the useful processing of optic and electronic materials, for applications such as microelectronic devices, photovoltaic cells and photocatalytic materials, thermoelectric materials and devices, micro-/nano-systems for energy storage and conversion, field-enhanced spectroscopies, biomedical coatings and devices, etc [45–51]. What essentially happens in MAPLE is that a high-pressure gas is produced in the surface layer and, as a result of the pressure gradient, a supersonic jet of host molecules is ejected normal to the target surface. In practice, the MAPLE process is far more complicated than the idealized model discussed above and was previously and thoroughly discussed [52, 53].
Although the films have been grown at a laser fluence varied in the range of 0.1–1 J/cm2 (266 nm wavelength, 7 ns duration of the pulse, 10 Hz) at a 2 mm² laser spot, and 40.000 laser pulses, only the samples grown at ~ 0.4 J/cm² have been further considered for the optical and morphological investigations. This choice is based on literature, as it was previously observed that the optimal fluence range is generally found between 0.2 and 0.4 J/cm². Furthermore, the number of laser pulses (40.000) was chosen based on our expertise and literature data in order to reach a film thickness of ~ 180–200 nm (thickness variations are typically around ± 10%). All depositions took place in vacuum at roughly 10–4 mbar and with the reaction chamber being continuously pumped at constant volume flow rate in order to facilitate optimal thermal equilibrium of the laser-evaporated material. After visual inspection, the samples have been scanned by optical microscopy and profilometry, then analysed by means of spectroscopic-ellipsometry and atomic force microscopy, prior to any SHG investigations.
Spectroscopic-ellipsometry (Se) Investigations Of The Maple-grown Thin Films
Spectroscopic-ellipsometry is an optical technique for investigating the dielectric properties of thin films, i.e. to determine the complex refractive index or the dielectric function. This is an indirect investigation method, as the polarization change is quantified by the amplitude ratio, Ψ (psi), and the phase difference, Δ (delta).
Spectroscopic-ellipsometry measures the change of polarization upon reflection or transmission and compares it to a mathematical model. Here, the experimental data concerning the wavelength dependence of the amplitude ratio, Ψ (psi), and the phase difference, Δ (delta), was acquired in the 300–1700 nm spectral range. Subsequently, we determined the refractive index and extinction coefficients for the three compounds, as presented in Fig. 5 and Fig. 6. The samples were measured by using the Cauchy–Urbach formalism, i.e. approximated with a material with Cauchy dispersion and with Urbach absorption. The optical model used for fitting the experimental data consisted in a first approach of four layers: a Si bulk substrate, a native SiO2 layer (3 nm thick), the MAPLE grown thioamide thin film and a rough top layer. The rough layer is composed of the thioamide film and air in a 50:50 ratio, by using the Bruggeman effective medium (B-EMA) approximation, while the values for the bulk dielectric functions of Si and SiO2 substrates were taken from literature [54, 55]. Further details on the fitting procedure for optical absorbing thin films is described in detail elsewhere [56–58]. The results are quite similar for all samples, i.e. the refraction index (n) varies between 1.74 and 1.76, while the extinction coefficient (k) is ~ 3*10− 3. The thickness of the films was found to range between 170 and 180 nm (± 21 nm), and the roughness typically is below 10%.
Atomic Force Microscopy (Afm) Of The Male-grown Thin Films
The atomic force microscopy analysis of the thin films surface revealed they are rather smooth, continuous, with just a few droplets. Results are coherent with the SEM images (data not shown here). Different surface sizes and areas were scanned on the samples. Two 40 x 40 µm² and two 10 x 10 µm² AFM images are presented in Fig. 7, for (E)-3-(4-methoxyphenyl)prop-2-enethioamide (II) in (a.) and (b.), and for (E)-3-(naphthalen-1-yl)prop-2-enethioamide (III) in (c.) and (d.), respectively. AFM investigations are needed in order to evaluate the uniformity of the surface before any further optical (SE), and SHG investigations are envisaged. The MAPLE-grown thin films exhibit roughness typically below 10%, expressed as root mean square (RMS), at a thickness of ~ 175 nm, as determined by SE measurements (described in Section 3.4) and by profilometry investigations (data now shown here). The roughness evaluation was made over a series of three samples, on large area scans (up to 90 x 90µm²) with different scanned zones for each sample. It is the laser-frozen target interaction that directly determines the morphology of the films. Droplets and other types of fragments may be ejected from the target during MAPLE and collected on the thin film, and/or degradation of the thioamides molecular structure by releasing gas and particulates that influence the thin film growth mechanism due to the laser-induced thermal effects, as it was previously shown in literature. The resulting films will be used for subsequent optical and SHG determinations, as they have a low roughness.
Second Harmonic Generation (Shg) Behaviour Of The Maple-grown Thin Films
The SHG phenomena in thin films is proportional with their thickness, as a coherent optical process of radiation of dipoles that is dependent on the second term of the expansion of polarization. The frequency ω of the applied electric field, which radiates electric field of 2ω as well as 1ω, determines the oscillation of the dipoles. In some cases, almost 100% of the light energy can be converted to the SHG frequency. These cases typically involve intense pulsed laser beams passing through large enough films, and careful alignment to obtain phase matching. Figure 8 is a schematic of the SHG experimental setup.
As presented in Table 1 and depicted in Table 2, the molecular shape of (E)-3-(4-methoxyphenyl)prop-2-enethioamide, i.e. compound (II), is perfectly flat and could be considered as a classic push-pull molecule. A similar structure may be observed for (E)-3-(naphthalen-1-yl) prop-2-enethioamide, or compound (III). However, the molecular shape of (E)-3-(2-chlorophenyl) prop-2-enethioamide, or compound (V), is quite different: a twisted molecule, therefore a non-classic structure for a compound exhibiting NLO properties, as it will further be discussed.
The Ti:sapphire femtosecond laser pulses inducing the generation of the second harmonic were focused on the three compounds, i.e. (II), (III), and (V). The incident laser power was chosen in four tests: 100 mW, 150 mW, 200 mW or 250 mW. In the generation of second harmonic generation, is critical the phase matching of the radiated electric field of ω and 2ω. They have to be in phase, so as not to interfere destructively. The applied electric field is strong enough to generate the second order radiation, but not so strong so that the dominant term is the first order term. Thus, one could think of the SHG signal as a perturbation and thus those lights of two frequencies should be added, and should not cancel each other out. To conclude, the total output is sum, i.e. accurately locating each dipole in the appropriate position of the material.
The nonlinear behaviour in the spectrum of light emitted by the three chosen thioamides compounds, as thin films, under femtosecond laser beam irradiation (Fig. 9 and Fig. 10) is highlighted by the regular shape of the emission band. Compared to SHG measurements of other organic layers previously made by the same experimental set-up [45, 56–63] we can underline two main features of the thioamides layers: a stronger second harmonic generation signal and which, under a laser beam, would offer a greater resistance.
In order to separate bulk contributions, surface contributions, and buried interface contributions in thin film systems, several approaches can be used: (1) specific polarization configurations and azimuthal orientations can be selected; (2) substrates without a SHG response can be used; (3) the film thickness dependence of the SHG response can be used to distinguish between surface, interface, and bulk contributions; (4) the surface properties can be selectively modified, possibly resulting in a change in SHG response; (5) spectroscopic information can be used to reveal the origin of the SHG radiation, providing information not only on the microscopic origin but also on the macroscopic origin. Two types of substrates are typically used in SHG based on thin films: Si (100) and fused silica. The SHG response of these substrates is already discussed and compared in literature [64].
Therefore, the key roles played by the structure of the organic compounds, the crystalline state, and therefore the homogeneity of the analyzed sample, in order to have SHG capabilities, are central elements that we must take into account in the interpretation of our data. It is also worth mentioning that, for the thin films, the thickness and the surface morphology play an important role in the SHG intensity. Nevertheless, we have demonstrated efficient nonlinear wavelength conversion in three of the novel aromatic thioamides, revealing that they exhibit second harmonic generation capabilities. To conclude, an understanding of these effects combines the classical theory of light with the quantum nature of the energy levels in materials. Our work is a crucial step towards on-chip quantum wavelength conversion at the single-photon level. As future prospects, nanofabricated devices will be considered since they are promising candidates for applications at short wavelengths (UV-visible) where conventional approaches become challenging since much smaller poling periods are required.
Experimental section
E. Merck (Germany) and Alfa Aesar (Deutschland) were the two brands from which we sourced our laboratory grade chemicals. The NMR spectra were recorded on Bruker Advance Ultrashield Plus spectrometer operating at 300.18 MHz for 1H and 125 MHz for 13C [70]. The open tube capillary method was used to determine uncorrected melting points. A multi EA 4000 device from Analytik Jena was used to performed the elemental analysis. A Vertex 70-Bruker spectrophotometer, was used to acquire the Fourier transform infrared (FTIR) spectra, using samples in in KBr pellets. M + 1 peaks were determined on an Agilent 1100 series and an Agilent Ion Trap SL mass spectrometer (Santa Clara, CA, USA), operating at 70 eV. Thin layer chromatography (TLC) plates, (silica gel G) were used to confirm the purity of commercial reagents used, compounds synthesized and to monitor the reactions as well. The petroleum ether: toluene: acetic acid (5:4:1) and toluene: ethyl acetate: formic acid (5:4:1) were the two types of eluents used in thin layer chromatography (TLC) to monitor the reactions carried out. The spots were located under iodine vapours / UV light.
(E)-3-phenylprop-2-enethioamide (I)
1H-NMR (300 MHz, DMSO-d6): δ 9.41 (2H, d, J = 13 Hz, NH2); 7.42–7.62 (5H, m, Ar–H), 7.65 (1H, d, J = 8.0 Hz, Ar–CH), 7.01 (1H, d, J = 8.0 Hz, CH–CS). 13C-NMR (100 MHz, DMSO-d6): δ 196.36, 141.20, 134.76, 129.83, 129.09, 127.86, 126.72.
(E)-3-(4-methoxyphenyl)prop-2-enethioamide (II)
1H-NMR (300 MHz, DMSO-d6): δ 9.43 (2H, d, J = 13 Hz, NH2); 7.61 (2H, d, J = 8.4 Hz, Ar–H), 7.16 (2H, d, J = 8.4 Hz, Ar–H), 7.68 (1H, d, J = 8.0 Hz, Ar–CH), 7.02 (1H, d, J = 8.0 Hz, CH–CS), 3.85 (3H, s, OCH3). 13C-NMR (100 MHz, DMSO-d6): δ 196.36, 160.25, 141.34, 134.76, 129.09, 127.86, 126.72, 56.3.
(E)-3-(naphthalen-1-yl)prop-2-enethioamide (III)
1H-NMR (300 MHz, DMSO-d6): δ 9.41 (2H, d, J = 13 Hz, NH2); 7.42–7.94 (7H, m, Ar–H), 7.36 (1H, d, J = 8.0 Hz, Ar–CH), 6.94 (1H, d, J = 8.0 Hz, CH–CS). 13C-NMR (100 MHz, DMSO-d6): δ 196.35, 141.40, 134.78, 134.1, 131.5, 129.1, 129.83, 129.13, 128.7, 127.96, 127.1, 126.72, 126.6.
(E)-3-(furan-2-yl)prop-2-enethioamide (IV)
1H-NMR (300 MHz, DMSO-d6): δ 9.37 (2H, d, J = 13 Hz, NH2); 7.8 (1H, d, Ar–H), 7.68 (1H, d, J = 8.0 Hz, Ar–CH), 6.91 (1H, t, Ar-H), 6.81 (1H, d, J = 8.0 Hz, CH–CS), 6.62 (1H, d, J = 8.0 Hz, Ar–CH). 13C-NMR (100 MHz, DMSO-d6): δ 196.36, 152.10, 146.7, 132.3, 125.1, 112.8, 111.5.
(E)-3-(2-chlorophenyl)prop-2-enethioamide (V)
1H-NMR (300 MHz, DMSO-d6): δ 9.87 (2H, d, J = 13 Hz, NH2); 7.36–7.65 (4H, m, Ar–H), 7.85 (1H, d, J = 8.0 Hz, Ar–CH), 7.03 (1H, d, J = 8.0 Hz, CH–CS). 13C-NMR (100 MHz, DMSO-d6): δ 196.38, 141.50, 136.2, 134.8, 129.83, 129.09, 127.86, 127.3, 126.72.
(E)-3-(3-nitrophenyl)prop-2-enethioamide (VI)
1H-NMR (300 MHz, DMSO-d6): δ 9.85 (2H, d, J = 13 Hz, NH2); 7.71–8.52 (4H, m, Ar–H), 7.65 (1H, d, J = 8.0 Hz, Ar–CH), 7.05 (1H, d, J = 8.0 Hz, CH–CS). 13C-NMR (100 MHz, DMSO-d6): δ 196.4, 148.2, 141.4, 136.3, 132.8, 129.4, 129.46, 123.2, 121.3.
(E)-3-(pyridin-4-yl)prop-2-enethioamide (VII)
1H-NMR (300 MHz, DMSO-d6): δ 9.41 (2H, d, J = 13 Hz, NH2); 8.83 (2H, d, J = 8.4 Hz, Ar–H), 7.90 (2H, d, J = 8.4 Hz, Ar–H), 7.65 (1H, d, J = 8.0 Hz, Ar–CH), 7.01 (1H, d, J = 8.0 Hz, CH–CS). 13C-NMR (100 MHz, DMSO-d6): δ 196.36, 150.7, 145.7, 141.20, 134.76, 123.1.
Computational And Modeling Details
A quantum mechanical modeling was performed on each thioamide, using the GAMESS 2012 [44, 45], software in order to find theirs structural parameters. A computer cluster consisting of 96 cores and 12 nodes running Linux CeontOS was used to perform the modeling of the synthesized compounds. The results were visualized using wxMacMolPlt [65, 66]. Density functional theory (DFT), at the M11/ktzvp level was used to optimize the molecular geometries of the compounds. Tuhlar’s M11 [67] is a modern functional that provides better results then classical B3LYP. Also, the basis set used is a more recent one (Karlsruhe valence triple zeta basis with a set of single polarization), as introduced by professor Ahlrichs [68, 69].