Double-pass terahertz time-domain spectroscopy of 2D materials

Double-Pass Terahertz Time-Domain Spectroscopy of 2D Materials David Arcos,* Daniel Nu˜no, María C. Santos, and Núria Ferrer-Anglada* 1. Introduction Currently, the characterization of 2D materials, such as graphene and few-layer transition metal dichalcogenides (TMDCs), is a major focus in materials science and engineering, driven by their potential for diverse applications. These applications include high-performance energy storage, [1,2] gas sensing, [3–5] biomedical devices, [6,7] water purification, [8,9] and electronic devices, [10–13] among others. In the terahertz (THz) range, 2D materials are also highly promising, as they can be employed in both emission and detection stages, [14,15] enabling the development of optical modulators, [16] resonators, [17] and optical logic gates at THz frequencies. [18] Tungsten disulfide (WS 2 ) is a TMDC that consists of a layer of tungsten atoms sandwiched between two layers of sulfur atoms, held together by strong in-plane covalent bonds. Similar to graphite and its single-layer (SL) form, graphene, TMDC bulk crystals are composed of stacked SLs connected by weak out-of-plane van der Waals forces, which enable easy exfoliation of the 2D layers. [13] These materials, analogous to graphene, exhibit exceptional optical and electrical properties, ranging from insulators to metals or, as in the case of WS 2 , semiconductors with either direct or indirect bandgaps. [19] It is well known that the bandgap of WS 2 depends on its number of layers. In SL WS 2 , the gap is direct with a value of 2.03 eV, whereas in bulk WS 2 , the gap is indirect, with a value of 1.32 eV. [20] This direct gap in SL WS 2 has been shown to result in photoluminescence, making WS 2 a strong candidate for transparent, flexible, and efficient optoelectronic devices. [21] Regarding the analysis technique, THz time-domain spectroscopy (THz-TDS) is a nondestructive spectroscopic technique commonly used to explore the properties of materials within the THz frequency range, typically spanning from 0.1 to 10 THz. This technique utilizes short pulses of THz radiation to gather time-resolved information about the sample, which can then be transformed into frequency-domain data, providing spectral insights without damaging the samples. THz–TDS has previously been applied to WS 2[22,23] and other 2D materials such as MoS 2[24] and graphene on various substrates, [25–28] as well to heterostructures containing WS 2 , [29] to assess optical transmittance and surface conductivity in the THz frequency range. The primary objective of this work is to present an alternative configuration of the THz–TDS technique, based on a Michelson interferometer, with multiple potential configurations. We aim to demonstrate that this new setup can effectively obtain data from both substrates and 2D layers when operated on a single branch, accounting for the THz beam passing through the sample twice. Specifically, we use this setup to obtain the transmittance and surface conductivity of a WS 2 sample on fused quartz. D. Arcos Department of Engineering Universitat de Vic–Universitat Central de Catalunya (UVic–UCC) Carrer de la Laura 13, 08500 Vic, Spain E-mail: [email protected] D. Arcos, N. Ferrer-Anglada Department of Physics Universitat Politècnica de Catalunya (UPC) Campus Nord J. Girona 3, 08034 Barcelona, Spain E-mail: [email protected] D. Nu˜no, M. C. Santos Department of Signal Theory and Communications Universitat Politècnica de Catalunya (UPC) Campus Nord J. Girona 1, 08034 Barcelona, Spain The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/pssb.202400323. © 2024 The Author(s). physica status solidi (b) basic solid state physics published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. DOI: 10.1002/pssb.202400323 Terahertz time-domain spectroscopy (THz–TDS) is a nondestructive imaging and characterization technique. It is currently used in the field of material science to obtain the surface conductivity and transmittance of bulk and 2D materials in the range from hundreds of GHz up to few THz. In this research, an alternative setup of the THz–TDS technique is proposed, based on a Michelson interferometer, with a double pass through the sample using a mirror and a semitransparent wafer. A single-branch configuration is used to characterize a few-layer WS 2 sample on a fused quartz substrate. The objectives of the experiment are to demonstrate that the configuration is viable for obtaining the parameters of the sample and the substrate, to present the models and equations used, and to explain the advantages and limitations of the method compared to the transmission configuration. The optical transmittance and surface conductivity of WS 2 are obtained with the new configuration in the frequency range from 0.2 to 1.2 THz. Raman spectroscopy is used to analyze the sample quality before performing the measurements. RESEARCH ARTICLE www.pss-b.com Phys. Status Solidi B 2024, 2400323 2400323 (1 of 9) © 2024 The Author(s). physica status solidi (b) basic solid state physics published by Wiley-VCH GmbH 2. Sample Description and Raman In this study, we analyze a single few-layer WS 2 sample prepared at Sejong University in Seoul, Korea, by the research group of Prof. J. Eom. The sample was produced through chemical vapor deposition on a fused quartz substrate with a square surface area (20 mm per side) and a thickness of 1.00 5% mm. Selecting this specific substrate is crucial for ensuring THz transparency. Fused quartz (silica) was chosen due to its orientationindependent refractive index, unlike crystalline quartz, whose THz transparency varies with crystallinity. [30] The sample’s2D nature and quality were verified through Raman spectroscopy, using a 514 nm excitation laser line at 0.5 mW. The Raman spectrum obtained for WS 2 on fused quartz is shown in Figure 1. The most prominent peaks in the spectrum, located at 353 and 419 cm 1 , correspond to WS 2 bands. Bulk WS 2 is characterized by an E1 2g band at 351 cm 1 and an A 1g band at 420 cm 1 . As the number of layers decreases, the separation between these peaks also decreases: the E1 2g band undergoes a blue shift, while the A 1g band experiences a redshift, [31–33] as observed in our spectra. Additionally, after correcting the substrate effects and fitting the baseline, the E1 2g band at 353 cm 1 was deconvoluted into two Lorentzian components, corresponding to the 2LA(M) band at 349.4 cm 1 and the E1 2g ΓðÞband at 354.8 cm 1 . The intensity ratio I2LA=IA1G serves as an indicator of the number of WS 2 layers, ranging from 0.47 for bulk WS 2 to 2.2 for SL WS 2 . [31,33] In our spectra, the I2LA=IA1G ratio is 0.61. Based on the position of these three bands and the I2LA=IA1G ratio, and referencing the works of Loh et al. [31] and Berkdemir et al. [33] we conclude that our WS 2 sample contains at least 3–4 layers. 3. Experimental Section The experimental setup [34] was based on the TERA K8 commercial THz spectrophotometer from Menlo Systems, with a silicon wafer placed in the middle of the THz beam path, acting as a semi-mirror (see Figure 2). The spectrophotometer’s configuration was mainly based on a Michelson interferometer (see Figure 3), a technique that has been previously employed in THz–TDS for imaging [35] and characterization, [36] although, Figure 1. Raman spectra of WS 2 sample. Figure 2. THz–TDS schematic of the experimental setup. www.advancedsciencenews.com www.pss-b.com Phys. Status Solidi B 2024, 2400323 2400323 (2 of 9) © 2024 The Author(s). physica status solidi (b) basic solid state physics published by Wiley-VCH GmbH 15213951, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pssb.202400323 by Readcube (Labtiva Inc.), Wiley Online Library on [17/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License in our experiment, interferometry was not used to extract the results. Figure 4 illustrates 16 possible configurations of the interferometer, arranged in a matrix according to the devices included in each arm (or branch) of the interferometer. We designated branch A (see configuration X0) as the non-inverting path and branch B (see configuration 0X) as the inverting path, with inversion occurring due to the presence of the L2 lens. In each branch, the following four scenarios could occur. The signal could be discarded by redirecting the THz beam before it reached the branch’s mirror (X), so the contribution from that branch was not detected by the receiver. If the only element in the path was the mirror, an air reference (0) for the branch was obtained. Finally, a bare substrate (1) or a 2D sample on the substrate (2) could be introduced, in addition to the mirror, such that the main beam passed through the materials twice before reaching the receiver. The combination of these options for each branch produces the 16 configurations shown in Figure 4. It might appear that the configurations below the diagonal were redundant, except for the sign inversion, but the use of the lens improved the spatial resolution, making branch B more suitable for Figure 3. Photograph of the Michelson interferometer assembly. Figure 4. All the proposed configurations of the Michelson interferometer for measuring 2D material samples and substrates. Paths for branches A and B are indicated in configurations X0 and 0X, respectively. www.advancedsciencenews.com www.pss-b.com Phys. Status Solidi B 2024, 2400323 2400323 (3 of 9) © 2024 The Author(s). physica status solidi (b) basic solid state physics published by Wiley-VCH GmbH 15213951, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pssb.202400323 by Readcube (Labtiva Inc.), Wiley Online Library on [17/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License measurements of samples with small surface areas. In contrast, the diagonal configurations (00, 11 and 22) did not result in a fully canceled signal due to aberration caused by the lens. In addition, in the 1X and 2X configurations, if the samples were positioned near the focal plane of the lens, the contribution of the Gouy phase shift must be considered when extracting optical properties, particularly at low frequencies. [37,38] The XX configuration could be used to verify that canceling the signals from both branches results in the receiver detecting no signal, ensuring that there were no unwanted reflections in the setup. In the configurations presented in Figure 4, it was assumed that both the sample and the substrate were transparent. If the substrate was not transparent, the sample could be placed directly at the position of the M4 mirror for reflection measurements, without requiring additional adjustments. 4. Single-Branch Analysis As explained earlier, although the system is based on an interferometer, interferometry will not be used to extract the results. Instead, we will only use signals from one branch, corresponding to the first row and first column configurations in Figure 4. The main objective of this work is to demonstrate that the system can be used to characterize substrates and 2D samples. The optical beam path is essentially the same as that in the conventional transmission setup described in ref. [22], but in this setup, we include a mirror (M4) behind the sample that reflects the beam, causing it to pass through the sample again before being measured. It is important to note that the receiving antenna is not positioned behind the sample, as it would be in the transmission setup. The simplified ray diagram for signals passing through the reference (air) channel, S0ωðÞ, the channel with a single substrate, Ssub ωðÞ, and the channel containing both layers, S2D ωðÞ, is shown in Figure 5. Two additional paths, corresponding to the Fabry–Pérot’sfirst reflections, are represented by dotted lines. These beams are part of the Ssub ω ðÞ signal and occur in forward and backward directions. It must be noted that these signals arrive at exactly the same time for the truncation of the temporal signals, as explained later. Each signal can be expressed in the frequency domain (ω¼2πf) as the product of the Fourier transform of the pulse arriving at interface 1, which is common to all measurements, by the channel transfer function, which depends on the materials being characterized. From the ray diagram in Figure 5, transfer functions for the reference channel, H0ωðÞ, and for the signal passing through the substrate, Hsub ωðÞ, can be seen in Equation (1) and (2), respectively. These expressions have been derived similarly to the one-pass case in the conventional transmission setup. [39,40] H0ωðÞ¼p02dδþ2dsub ðÞ (1) Hsub ωðÞ¼p02dδ ðÞt0;sub psub dsub ðÞtsub;0X NFP m¼0 fp m sub "# 2 (2) where the NFP value in the summation represents the number of Fabry–Pérot reflections, fp sub, occurring within the substrate, as defined in Equation (3), and the propagation, transmission, and reflection coefficients are specified in Equation (4)–(6), respectively, and depend on the complex refractive indices of the materials at each interface. It is important to note that common propagation paths, such as those between interfaces 2 and 3 shown in Figure 5, are excluded from any transfer function. fp sub ωðÞ¼ ˜ nsub 1 ˜ nsub þ1  2 exp 2i ˜ nsubωdsub c (3) paω,dðÞ¼exp i ˜ naωd c  (4) ta;bωðÞ¼ 2˜ na ˜ naþ˜ nb (5) ra;bωðÞ¼ ˜ na˜ nb ˜ naþ˜ nb (6) Conversely, the transfer function for the signal passing through the complete sample, ˆ H2D ωðÞ, is presented in Equation (7). This expression assumes that the 2D material layer serves as a boundary condition on the substrate (dδ≪λ) with a surface conductivity σ2D, following the method proposed by Liam et al. in [40] for measuring 2D materials using a conventional transmission setup. ˆ H2D ωðÞ¼ ˆ t0;sub psub 2dsub ðÞtsub;0t0;sub ˆ tsub;0X NFP m¼0 fp m b ! 2 (7) fp bωðÞ¼ ˜ nsub 1 ˜ nsub þ12˜ nsubX1ðÞexp 2i ˜ nsubωdsub c  (8) X1¼1þ˜ nsub þσ2DZ0(9) Figure 5. Ray diagram of THz beams passing through the air, the bare substrate, and the entire sample. www.advancedsciencenews.com www.pss-b.com Phys. Status Solidi B 2024, 2400323 2400323 (4 of 9) © 2024 The Author(s). physica status solidi (b) basic solid state physics published by Wiley-VCH GmbH 15213951, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pssb.202400323 by Readcube (Labtiva Inc.), Wiley Online Library on [17/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License ˆ t0;sub ωðÞ¼ 2 1þ˜ nsub þσ2DZ0 (10) ˆ tsub;0ωðÞ¼ 2˜ nsub 1þ˜ nsub þσ2DZ0 (11) Here, fp bis defined in Equation (8) and (9); Z0¼120πΩrepresents the free-space impedance. The transmission coefficients ˆ t0;sub and ˆ tsub;0, which correspond to the air–sample interface (denoted as 1 in Figure 5), are specified in Equation (10) and (11), respectively. In other words, due to symmetry, the result for each channel essentially represents the square of the transfer function for a single pass through the sample. The common part before reaching the first surface is not included in the analysis, as it can be treated as part of the THz pulse waveform. Based on these transfer functions, new complex transmission coefficients are defined to characterize the substrate (see Equation 12). Once the complex refractive index is determined, the surface conductivity of the sample can be calculated (Equation 13). [40,41] Tsub ωðÞ¼ Ssub ωðÞ S0ωðÞ¼16 ˜ nsub ðÞ 2 ˜ nsub þ1ðÞ 4e2i ˜ nsub1 ðÞ ωdsub cX NFP m¼0 fp m sub ! 2 (12) ˆ T2D ωðÞ¼ S2D ωðÞ S0ωðÞ¼16X2˜ nsub ðÞ 2 ˜ nsub þ1ðÞ 2e2i ˜ nsub1 ðÞ ωdsub cX NFP m¼0 fp m b ! 2 (13) For analyzing each sample, reflection-free Fabry–Pérot models can be applied by selecting short time windows in the measurements. This approach eliminates the summation terms, reducing the transmission coefficient to the form shown in Equation (14). ˆ T2DðωÞFP ¼0¼16X2˜ nsub ðÞ 2 ˜ nsub þ1ðÞ 2e2i ˜ nsub1 ðÞ ωdsub c(14) Alternatively, one can select time windows that are sufficiently long to ensure that all reflections are effectively included. This allows the summations of the convergent geometric series to be replaced by their result, as shown in Equation (15). ˆ T2DðωÞFP!∞ ¼16X2˜ nsub ðÞ 2 ˜ nsub þ1ðÞ 2e2i ˜ nsub1 ðÞ ωdsub c1 1fp bωðÞ  2 (15) It should be noted that the Fabry–Pérot reflections between interfaces 2 and 3 in Figure 5 have not been considered in the analysis. This is because, on the one hand, we used dx>20dsub, and, on the other hand, the refractive index of the substrates used is less than 2, meaning that at least 10 reflections occur between the walls of the substrate before any are produced between the substrate and the mirror. [42] If the dxdistance is reduced to improve spatial resolution, a multilayer model must be used to account for these Fabry–Pérot reflections. The same applies if a much thicker substrate or one with a higher refractive index is used, as this limits the available time window before additional reflections are observed. However, since the reflections are more attenuated, shorter time windows can be employed. 5. Experimental Results In this section, the single-branch analysis is used to characterize a few-layer WS 2 sample on a fused quartz substrate. Preliminary measurements and calibrations using the 00 configuration reveal that the maximum amplitude of the pulses is reduced by more than 95%, suggesting that the delays in branches A and B are similar and that the beam distribution on each branch is close to 50%. The analysis continues by measuring the bulk materials in the time domain to verify that the new setup is suitable for extracting the substrate’s properties. Two types of fused quartz substrates from different suppliers are used: the first is the same fused quartz as the one supporting the WS 2 sample, with a thickness of 1.00 5% mm, and the second is a 20% thicker fused quartz (1.20 5% mm) with a rectangular surface area of 300 mm 2 .Figure 6 shows the time-domain waveforms for the two air references (obtained with X0 and 0X configurations), the bare substrate signals (obtained with X1 and 1X configurations) and the complete WS 2 /quartz sample signal (obtained with the 2X configuration). Figure 6 also includes the coordinates (time and amplitude) of the main peaks’maximums, which were used for preliminary thickness and refractive index estimations. All measurements in branch B have been multiplied by 1 to compensate for the inverting effect of lens L2 (see Figure 2). While the two reference signals are aligned in peak timing, they differ slightly in the shape; however, this discrepancy is not significant here since data extraction in each branch is done relative to its own reference. The thickness difference between the substrates is apparent in the relative delay of 1.05 ps, which increases to 3.4 ps for the first Fabry-Pérot reflections. Regarding amplitude reduction due to the substrates and sample, it is notable that the first Fabry–Pérot reflections are strongly attenuated. This suggests that the results obtained using the approximations in Equation (14) and (15) should be similar, making it unnecessary to use very long time windows to validate the approximation in Equation (15). As with the transmission setup, a refractive index estimate for each substrate can be obtained directly from the time-domain waveforms, considering the ray diagram of the beams (see Figure 5) and the reduced speed due to the substrate thickness, dsub. The first Fabry–Pérot reflection was also used to derive an additional equation for estimating the substrate thickness and confirming the consistency of the values. Equation (16)–(18) show the relationships between pulse times, refractive index, and substrate thickness. t0¼2dsub cþtref (16) t1¼2dsubnsub cþtref (17) t2¼4dsubnsub cþtref (18) where t0is the arrival time of the reference S0ωðÞpulse, t1is the arrival time of the first Ssub ωðÞsignal pulse, t2is the time of the www.advancedsciencenews.com www.pss-b.com Phys. Status Solidi B 2024, 2400323 2400323 (5 of 9) © 2024 The Author(s). physica status solidi (b) basic solid state physics published by Wiley-VCH GmbH 15213951, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pssb.202400323 by Readcube (Labtiva Inc.), Wiley Online Library on [17/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License first Fabry–Pérot reflection, and tref is an arbitrary delay due to air propagation, common to all measurements. Thickness and refractive index estimates for both substrates can then be obtained from the time differences, as shown in Equation (19) and (20). Δt10 ¼t1t0¼2dsub cnsub 1ðÞ (19) Δt21 ¼t2t1¼2dsubnsub c(20) From the data in Figure 6, we obtained that the first fused quartz sample has a thickness of dsub ≈1.022 mm and a refractive index of nsub ≈1.966. The thicker fused quartz shows a thickness of dsub2≈1.216 mm and a refractive index of nsub2≈1.943. Both refractive index values are consistent with literature [43] for this frequency range, and the thicknesses agree with the specifications provided by the manufacturers as well as with experimental measurements. Figure 7 presents the spectra corresponding to the signals in Figure 6. All spectra show strong molecular absorption peaks due to water vapor at 557, 752, and 988 GHz. [44,45] Between 100 and 300 GHz, the reference spectrum from branch B is clearly higher than that from branch A, but above 1 THz, the opposite trend is observed. In the intermediate region, the spectra are similar, though not identical. The spectra obtained for the sample and the bare substrates are consistent with the reference spectra for each respective branch. Compared to the spectra from the Figure 6. Time-domain electric field pulse of the transmitted THz wave through the two references (air), the bare substrates (fused quartz), and the WS 2 sample. Figure 7. Electric field amplitude spectra of the transmitted THz wave through the two references (air), the bare substrates (fused quartz), and the WS 2 sample. www.advancedsciencenews.com www.pss-b.com Phys. Status Solidi B 2024, 2400323 2400323 (6 of 9) © 2024 The Author(s). physica status solidi (b) basic solid state physics published by Wiley-VCH GmbH 15213951, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pssb.202400323 by Readcube (Labtiva Inc.), Wiley Online Library on [17/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License transmission configuration using the same antennas, [22] the bandwidth is slightly narrower. Above 1 THz, substrate signals become highly attenuated, resulting in a low signal-to-noise ratio (SNR). Figure 8 shows the transmittance for the substrates and the complete sample within the range of 0.2–1.2 THz. Notably, all transmittances in Figure 8 represent a single pass through the sample, calculated based on the fact that the transmission coefficients for this configuration are the square of those for a singlepass transmission configuration. The transmittance values and behavior align with those measured via the transmission method: [42] substrate transmittances decrease from ≈95% to 70%, and the WS 2 sample shows a nearly constant transmittance around 90% over the measured range. Although one of the substrates is 20% thicker, the transmittance remains nearly the same, indicating that the extinction coefficient κsub is close to zero, as seen in Figure 9. In other words, the amplitude reduction is mainly attributable to the transmission coefficients or reflections at the air–substrate interfaces rather than to propagation through the material itself. Figure 9 also shows that the refractive index of the substrate is nsub ¼1.960 0.009 from 300 GHz to 1 THz. Specifically, at 500 GHz, the measured value is nsub ¼1.962, which matches the value reported by M. Naftaly and R. E. Miles at this frequency. [46] The refractive index also agrees with the estimate derived from the time-domain analysis. Figure 8. Transmittance obtained from the amplitude spectra of the bare substrates (fused quartz), the WS 2 layer, and the entire sample in the range [0.2, 1.2] THz. Figure 10. Real part of the sheet conductivity of WS 2 obtained using single-branch analysis, compared to the values obtained from a transmission THz–TDS configuration for WS 2 and single-layer (SL) graphene in a previous study. [22] Figure 9. Complex refractive index of the fused quartz substrate in the range [0.2, 1.2] THz. www.advancedsciencenews.com www.pss-b.com Phys. Status Solidi B 2024, 2400323 2400323 (7 of 9) © 2024 The Author(s). physica status solidi (b) basic solid state physics published by Wiley-VCH GmbH 15213951, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pssb.202400323 by Readcube (Labtiva Inc.), Wiley Online Library on [17/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License The real part of the sheet conductivity of the WS 2 sample, obtained from Equation (15) (assuming all reflections are included), is shown in Figure 10. The WS 2 sheet conductivity remains stable between 2.2 and 2.6 mS sq1. For comparison, Figure 10 also includes sheet conductivity for an SL graphene sample and few-layer WS 2 sample on a similar substrate, produced using the same technique, and measured using a transmission setup in a previous study. [22] The sheet conductivity of WS 2 measured with this double-pass configuration is slightly higher but remains within the same order of magnitude as previous measurements with the transmission setup. The results are consistent and show, as expected, that the number of layers in the 2D material influences the conductivity of the samples. 6. Conclusion A set of configurations for THz–TDS based on a Michelson interferometer has been presented. The primary advantage of this approach lies in its versatility, enabling both interferometry as well as transmission and reflection measurements on the sample with minimal adjustments to the setup. With a sufficiently extended delay line, the single-branch configuration can, in fact, capture both transmitted and reflected pulses for semitransparent samples in a single measurement. When the signal from one branch is cancelled, measurements of 2D material samples, substrates, or air references can be conducted similarly to a transmission setup, but with a double pass over the sample. This double-pass configuration has been shown to allow transmittance and surface conductivity measurements in 2D materials, such as WS 2 on fused quartz. The bandwidth obtained with this technique is narrower compared to the transmission configuration, and the available time window for measurements is shorter due to the additional reflections involved. However, as the amplitudes of the Fabry–Pérot reflections are reduced, a shorter time window is also required. For characterizing samples with a large surface area, measurements can be taken from either branch. However, the branch with the lens (B) offers higher spatial resolution, while the branch without the lens (A) avoids the Gouy phase shift effect. The temporal signals allow for an estimation of the substrates’refractive indices, and it is even possible to obtain an approximate measurement of substrate thickness using the first Fabry–Pérot reflections. The equations provided for the double-pass configuration have achieved a high level of accuracy in obtaining the substrates’complex refractive index in the THz range. Regarding 2D material measurements, both the surface conductivity and transmittance obtained from the few-layer WS 2 sample with this configuration are consistent with values obtained in previous studies using a conventional transmission setup on similar samples, though the conductivity is higher than expected. The discrepancy in sheet conductivity could be attributed to differences in the number of layers, the grain size, or the reference substrate used, among other factors. Further investigation should be carried out to obtain more accurate values for the sheet conductivity. Finally, as this technique is based on a noncontact THz–TDS with low power pulses, the method is nondestructive. Acknowledgements This work has received funding from the CONFLOC grant (grant no. PID2022-1377540B), funded by MCIN/AEI/10.13039/501100011033/ FEDER, UE, from the EU’s Horizon Europe under Marie SklodowskaCurie grant agreement no. 101073265 (EWOC), and from the Catalan Government University and Research Aid Management Agency (AGAUR) through the Research Group grant 2021 SGR 01415. The authors would like to thank the Scientific and Technological Center at the Universitat de Barcelona (CCiT-UB) for providing Raman spectroscopy facilities and the Prof. J. Eom group from Sejong University, Korea, for providing the WS 2 sample. Conflict of Interest The authors declare no conflict of interest. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. Keywords 2D materials, surface conductivities, terahertz time-domain spectroscopies (THz-TDS), WS 2 Received: June 29, 2024 Revised: November 14, 2024 Published online: [1] N. Farooq, Z. Ur Rehman, M. I. Khan, S. Asghar, M. Saleem, R. Irshad, A. Sheikh, A. Shanableh, S. Manzoor, Z. U. Khan, New J. Chem. 2024,48, 8933. [2] X. Zhang, L. Hou, A. Ciesielski, P. Samorì, Adv. Energy Mater. 2016,6, 1600671. [3] K. K. Pawar, A. Kumar, A. Mirzaei, M. Kumar, H. W. Kim, S. S. Kim, Chemosphere 2024,352, 141234. [4] X. Tang, M. Debliquy, D. Lahem, Y. Yan, J.-P. Raskin, Sensors 2021,21, 1443. [5] J. H. Choi, J. Lee, M. Byeon, T. E. Hong, H. Park, C. Y. Lee, ACS Appl. Nano Mater. 2020,3, 2257. [6] S. K. Nandy, S. Das, S. Pandey, P. Kalita, M. K. Gupta, A. Kabra, P. Wadhwa, D. Kumar, Luminescence 2024,39, 4771. [7] S. Anju, P. V. Mohanan, Synth. Met. 2021,271, 116610. [8] K. Rana, H. Kaur, N. Singh, T. Sithole, S. S. Siwal, Next Mater. 2024,3, 100107. [9] H. Peng, R. Wang, L. Mei, Q. Zhang, T. Ying, Z. Qian, A. B. Farimani, D. Voiry, Z. Zeng, Matter 2023,6, 59. [10] E. Elahi, S. Nisar, M. Rabeel, M. A. Rehman, M. Ouladsamne, A. Irfan, M. Abubakr, J. Aziz, M. Asim, G. Dastgeer, Mater. Adv. 2024,5, 1226. [11] G. Iannaccone, F. Bonaccorso, L. Colombo, G. Fiori, Nat. Nanotechnol. 2018,13, 183. [12] M. Bozzi, L. Pierantoni, S. Bellucci, Radioengineering 2015,24, 661. [13] Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman, M. S. Strano, Nat. Nanotechnol. 2012,7, 699. [14] S. Gotra, V. Pandey, R. Yaduvanshi, Superlattices Microstruct. 2021, 150, 106754. [15] M. Wang, E. H. Yang, Nano-Struct. Nano-Objects 2018,15, 107. [16] Z. Sun, A. Martinez, F. Wang, Nat. Photonics 2016,10, 227. www.advancedsciencenews.com www.pss-b.com Phys. Status Solidi B 2024, 2400323 2400323 (8 of 9) © 2024 The Author(s). physica status solidi (b) basic solid state physics published by Wiley-VCH GmbH 15213951, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pssb.202400323 by Readcube (Labtiva Inc.), Wiley Online Library on [17/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License [17] Y. M. Bahk, K. H. Kim, K. J. Ahn, H. R. Park, Adv. Photonics Res. 2024, 5, 2300211. [18] W. Su, Z. Geng, IEEE Photonics J. 2018,10,1. [19] A. V. Kolobov, J. Tominaga, in Two-Dimensional Transition-Metal Dichalcogenides, Springer, Cham 2016. [20] J. Gusakova, X. Wang, L. L. Shiau, A. Krivosheeva, V. Shaposhnikov, V. Borisenko, V. Gusakov, and B. K. Tay, Phys. Status Solidi A 2017, 214, 1700218. [21] H. R. Gutiérrez, N. Perea-López, A. L. Elías, A. Berkdemir, B. Wang, R. Lv, F. López-Urías, V. H. Crespi, H. Terrones, M. Terrones, Nano Lett. 2013,13, 3447. [22] D. Arcos, D. Nu˜no, M. C. Santos, L. Ametller, N. Ferrer-Anglada, Phys. Status Solidi B 2020,257, 2000476. [23] J. Wang, J. Tian, T. Li, J. Zhou, in 2017 10th UK-Europe-China Workshop on Millimetre Waves and Terahertz Technologies (UCMMT), Liverpool, IEEE 2017, pp. 1–2. [24] D. Arcos, D. Gabriel, D. Dumcenkco, A. Kis, N. Ferrer-Anglada, Phys. Status Solidi B 2016,253, 2499. [25] S. Scarfe, W. Cui, A. Luican-Mayer, J. M. Ménard, Sci. Rep. 2021,11, 8729. [26] P. R. Whelan, Q. Shen, D. Luo, M. Wang, R. S. Ruoff, P. U. Jepsen, B. Zhou, Opt. Express 2020,28, 28819. [27] D. Gabriel, B. Sempere, C. Colominas, N. Ferrer-Anglada, Phys. Status Solidi B 2015,252, 2423. [28] J. D. Buron, D. H. Petersen, P. Bøggild, D. G. Cooke, M. Hilke, J. Sun, E. Whiteway, P. F. Nielsen, O. Hansen, A. Yurgens, P. U. Jepsen, Nano Lett. 2012,12, 5074. [29] D. Arcos, D. Nu˜no, M. F. Khan, J. Eom, L. Ametller, N. Ferrer-Anglada, Phys. Status Solidi B 2023,260, 2300202. [30] D. Grischkowsky, S. Keiding, M. Van Exter, C. Fattinger, J. Opt. Soc. Am. B 1990,7, 2006. [31] T. A. Loh, D. H. Chua, A. T. Wee, Sci. Rep. 2015,5, 18116. [32] F. Wang, I. A. Kinloch, D. Wolverson, R. Tenne, A. Zak, E. O’Connell, U. Bangert, R. J. Young, 2D Mater. 2016,4, 015007. [33] A. Berkdemir, H. R. Gutiérrez, A. R. Botello-Méndez, N. Perea-López, A. L. Elías, C. I. Chia, B. Wang, V. H. Crespi, F. López-Urías, J. C. Charlier, H. Terrones, Sci. Rep. 2013,3, 1755. [34] D. Nu˜no, PhD Thesis, Department of Signal Theory and Communications, UPC, Barcelona 2022. [35] J. L. Johnson, T. D. Dorney, D. M. Mittleman, IEEE J. Sel. Top. Quantum Electron 2001,7, 592. [36] S. Krishnamurthy, M. T. Reiten, S. A. Harmon, R. Cheville, Appl. Phys. Lett. 2001,79, 875. [37] P. Kužel, H. Nˇemec, F. Kadlec, and C. Kadlec, Opt. Express 2010,18, 15338. [38] S. Feng, H. G. Winful, Opt. Lett. 2001,26, 485. [39] L. Duvillaret, F. Garet, J. L. Coutaz, IEEE J. Sel. Top. Quant. Electron 1996,2, 739. [40] M. Liang, Z. Wu, L. Chen, L. Song, P. Ajayan, H. Xin, IEEE Trans. Microwave Theory Technol. 2011,59, 2719. [41] G. Grüner, C. Dahl, in Millimeter and Submillimeter Wave Spectroscopy of Solids, Springer, Cham 1998. [42] D. Arcos, PhD Thesis, Department of Physics, UPC, Barcelona 2022. [43] M. Naftaly, R. E. Miles, J. Non-Cryst. Solids 2005,351, 3341. [44] Y. Yang, A. Shutler, D. Grischkowsky, Opt. Express 2011,19, 8830. [45] Q. Jing, D. Liu, Int. J. Commun. Syst. 2018,31, 3718. [46] M. Naftaly, R. E. Miles, J. Appl. Phys. 2007,102,4. www.advancedsciencenews.com www.pss-b.com Phys. Status Solidi B 2024, 2400323 2400323 (9 of 9) © 2024 The Author(s). physica status solidi (b) basic solid state physics published by Wiley-VCH GmbH 15213951, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pssb.202400323 by Readcube (Labtiva Inc.), Wiley Online Library on [17/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License