High-Transmittance 2π Electrically Tunable Terahertz Phase Shifter with CMOS-Compatible Driving Voltage Enabled by Liquid Crystals

Round 1

Reviewer 1 Report

The paper was well witten and interesting. Please address the following minor comments.

Please explain why NWhs was used.

In Fig.2, Type-A to Type-C have two unit cells. Please explain the performance of the single unit cell.

Author Response

Authors’ reply to the reviewers’ comments:

High-transmittance 2pi electrically tunable terahertz phase shifter with CMOS-compatible driving voltage enabled by liquid crystals

Review 1:

We would like to thank the reviewer for your insightful comments. Author replied them and showed the response as follows. According to the suggestion, Author also modified the content of manuscript with underline and blue color.

1. Please expain why NWs was used.

(Answer) 
In the THz band, Author has showed that ITO NWhs exhibit superb transparency (~82%) in the previous work. Meanwhile, their DC mobility (~92 cm²V^-1s^-1) and conductivities (~ 245 W^-1cm^-1) are comparable to sputtered ITO thin films. Here, in order to achieve this transmittion-type THz phase shifter, we need to apply this kind of ITO NWs structure on our device as the electrode.

 

2.In Fig.2, Type-A to Type-C have two unit cells. Please explain the performance of the single unit cell. 

(Answer) 
In this work, we are willing to achieve one 2pi electrically tunable terahertz phase shifter. Basically, the thickness of LCs cell is proportional to the amount of the phase shift. In the other words, the single unit cell will be the "pi" electrically tunable terahertz phase shifter. The LC layer of one single unit cell was ~1.12 mm. That is, the total thickness of the LC material for two unit cells was approximately 2.24 mm.

Reviewer 2 Report

The transmittance performance is also critical for a phase shifter. Please include the transmittance versus driving voltage in Fig. 5.

Please include a table which compares the performance of the proposed work with prior works.

Author Response

We would like to thank the reviewer for your insightful comments. Author replied them and showed the response as follows. According to the suggestion, Author also modified the content of manuscript with underline and blue color.

1. The transmittance performance is also critical for a phase shifter. Please include the transmittance versus driving voltage in Fig. 5.

(Answer)

Thanks for this useful suggestion.

According to the [Y. Du, H. Tian, X. Cui, H. Wang, and Z.-X. Zhou, “Electrically tunable liquid crystal terahertz phase shifter driven by transparent polymer electrodes,” J. Materials Chemistry C (2016)], they showed the temporal waveforms of the THz pulse transmitted through the LC cell at various driving voltages in their Fig. 3. We can find the change of transmittance with voltages is very little. On the other hand, from our previous work [C.-P. Ku, C.-C. Shih, C.-J. Lin, R.-P. Pan, and C.-L. Pan, THz Optical Constants of the Liquid Crystal MDA-00-3461. Mol. Cryst. Liquid Cryst. 2011, vol. 541, no. 1, pp. 303-308.], we can find the extinction coefficients, k_e and k_o, are pretty close. In other words, the transmittance won’t have too much change with the driving voltages.

According to the suggestion, Author also modified the content of manuscript with underline and blue color.

2. Please include a table which compares the performance of the proposed work with prior works.

(Answer)

Thanks for this suggestion. We have added one table to compare the performance of the proposed work with previous work.

Reviewer 3 Report

The manuscript presents a THz tunable phase shifter based on the stacking of two planar liquid-crystal (LC) cells, whose electro-optic switching is enabled via ITO nanowhisker (NWh) electrodes. Although the device is rather simple and it has been demonstrated in numerous works in the literature, there are some innovative aspects, mainly the use of ITO-NWh transparent electrodes and the low driving voltages, which in principle suffice to guarantee publication. However, the following points must be addressed first:

 

[1] The reference list lacks some significant works in the field on “free-space” LC-THz phase shifters, e.g. [J. Appl. Phys. 121, 143106, 2017], [Opt. Commun. 431, 63, 2019], as well as resonant cavities, which permit the shrinkage of the LC cell thickness, e.g. [Nanotechnology 28, 124002, 2017]. Such works should be mentioned and a short discussion added in the introduction. 

 

[2] The device shows overall significant insertion losses, whose origin cannot be easily identified. In particular, these can stem from three main reasons: a) losses in the ITO-NWh, b) losses in LC and substrates, and c) Fabry-Perot reflections. 

 

The authors are invited to provide a THz-TDS measurement of a single substrate coated with NWh and preferably analyze the data by fitting the NWh permittivity with a Drude or Drude-Smith model (see e.g. [J. Appl. Phys. 121, 143106, 2017], [IEEE Photon. Technol. Lett. 30, 1579, 2018]), so that, apart from the permittivity, the DC conductivity of the electrodes is also evaluated (the latter could also be experimentally measured). 

 

Then, the transmittance/reflectance of the device can be calculated by assigning the proper complex permittivities at each layer (silica, NWh, LC). This can be done via full-wave simulations or also by using some matrix method, e.g. Berreman or, if not possible, at least the Jones matrix formulation. Such an analysis will quantify the amount of losses introduced by the NWh and LC, so that the practical applicability of the device can be assessed.

 

[3] The three configurations differ in terms of the voltage drop in the stacked LC layers. In type A the applied voltage drops fully across the LC cells, which is not the case for type B and C, where part of the applied voltage drops across the silica substrate, thus reducing the LC tunability. The authors are invited to quantify the effective voltage drop in the LC cells for types B and C in order to better interpret the results of Fig. 5. If an anisotropic model is not available, the LC low-frequency permittivity can be estimated as a voltage-dependent average between the ordinary and extraordinary one.

 

[4] Please provide information and/or results on the switching times of the device.

 

[5] Please provide some reference for Equations (1-3).

Author Response

We would like to thank the reviewer for your insightful comments. Author replied them and showed the response as follows. According to the suggestion, Author also modified the content of manuscript with underline and blue color. 

 1. The reference list lacks some significant works in the field on “free-space” LC-THz phase shifters, e.g. [J. Appl. Phys. 121, 143106, 2017], [Opt. Commun. 431, 63, 2019], as well as resonant cavities, which permit the shrinkage of the LC cell thickness, e.g. [Nanotechnology 28, 124002, 2017]. Such works should be mentioned and a short discussion added in the introduction.

 (Answer) Thanks for this great suggestion! We have added all the references you mentioned into the manuscript. 

 2. The device shows overall significant insertion losses, whose origin cannot be easily identified. In particular, these can stem from three main reasons: a) losses in the ITO-NWh, b) losses in LC and substrates, and c) Fabry-Perot reflections. The authors are invited to provide a THz-TDS measurement of a single substrate coated with NWh and preferably analyze the data by fitting the NWh permittivity with a Drude or Drude-Smith model (see e.g. [J. Appl. Phys. 121, 143106, 2017], [IEEE Photon. Technol. Lett. 30, 1579, 2018]), so that, apart from the permittivity, the DC conductivity of the electrodes is also evaluated (the latter could also be experimentally measured). Then, the transmittance/reflectance of the device can be calculated by assigning the proper complex permittivities at each layer (silica, NWh, LC). This can be done via full-wave simulations or also by using some matrix method, e.g. Berreman or, if not possible, at least the Jones matrix formulation. Such an analysis will quantify the amount of losses introduced by the NWh and LC, so that the practical applicability of the device can be assessed. The three configurations differ in terms of the voltage drop in the stacked LC layers. In type A the applied voltage drops fully across the LC cells, which is not the case for type B and C, where part of the applied voltage drops across the silica substrate, thus reducing the LC tunability. The authors are invited to quantify the effective voltage drop in the LC cells for types B and C in order to better interpret the results of Fig. 5. If an anisotropic model is not available, the LC low-frequency permittivity can be estimated as a voltage-dependent average between the ordinary and extraordinary one. 

(Answer) Thanks for this great comment. Authors totally agreed what the reviewer mentioned. Actually, in the previous work, we have studied several different optical and electrical properties of ITO nanostructures. In other words, we employed both transmission-type and reflection-type terahertz time-domain spectroscopies (THz-TDTS and THz-TDRS) to explore the far-infrared dielectric function of these samples. Their electrical properties, such as complex conductivities of plasma frequencies, carrier scattering times, were analyzed and found to be fitted well by the Drude-Smith model (Non-Drude behavior) over 0.1~1.4 THz. Please refer to [“Non-Drude Behavior in Indium-Tin-Oxide Nanowhiskers and Thin Films Investigated by Transmission and Reflection THz Time-Domain Spectroscopy,” IEEE Journal of Quantum Electronics 49, No. 8, pp. 677-690 (2013).] and [“Broadband terahertz conductivity and optical transmission of indium-tin-oxide (ITO) nanomaterials,” Opt. Express 21, No. 14, pp. 16670-16682 (2013).] In these two articles, authors provided whole the information of the permittivity, the DC conductivity of the electrodes. By using these ITO NWhs electrical parameters, the LC’s bend elastic constant, and tilt-angle information of ITO NWhs substrate, we can estimate the quantify the amount of losses between different layers, and the voltage drop in the stacked LC layers, respectively. According to the suggestion, Author also modified the content of manuscript with underline and blue color. 

 3. Please provide information and/or results on the switching times of the device. 

(Answer) Thanks for this great comment. We measured the response time properties of present ITO NWhs phase shifter. The 20 to 80 % rise time and fall time are around 52 sec and 240 sec, respectively. We have added the related content into the manuscript. 

4. Please provide some reference for Equations (1-3). 

(Answer) Thanks for this great comment. We have added the references for Equations (1-3).

Reviewer 4 Report

The issue of this paper is improvement of liquid crystal phase modulator operated with CMOS-compatible driven voltage. Previously, the authors reported liquid crystal phase modulators using ITO nanomaterial in Ref [5]. The authors redesigned and optimized liquid crystal phase modulators, operating as a quarter wave plate at 1 THz. Here the authors optimized them and demonstrated liquid crystal phase modulators. Required bias voltage for half wavelength operation at 1 THz is 2.6 V. I understand that the THz phase modulator is a critical component in the present THz technology, and I agree that this work is valuable for developers of the THz components. I agree that their experimental results are also reliable. Therefore, we recommend that this paper is suitable for the published in Applied Science.

 

Minor revision 

Is the citation of [14] in line 98 exact?. Check it again.

Author Response

We would like to thank the reviewer for your insightful comment. Author replied them and showed the response as follows.

1.     Is the citation of [14] in line 98 exact?  Check it again.

(Answer)

Thanks for this great comment. We have modified this error.

Round 2

Reviewer 3 Report

The authors have significantly improved the quality of the revised manuscript. A couple of minor points:

[1] Although the introduction has been updated, not all works suggested by the reviewer have been discussed.

[2] The authors provided a couple of references where they describe in detail the properties of the ITO nanowhisker electrodes in the THz spectrum. Still, it would be very helpful to have an estimate of the losses in each layer of the investigated LC cells, such that it becomes clear if and where improvements can be expected. 

Author Response

[1] Although the introduction has been updated, not all works suggested by the reviewer have been discussed.

(Answer)

Thanks for this great suggestion! Author has added those reference which Reviewer mentioned. Besides, Author also made one short discussion added in the introduction and modified the content of manuscript with underline and Red & blue color.

Short discussion (Modified content in the manuscript):

To meet the demands of applications in emerging technologies such as sub–terahertz (THz) radio–over–fiber communication link [1], numerous quasi–optic components, such as phase shifters [2–10], polarization converter [11-13],……………………………………………………………………………………

Besides, by applying twisted nematic (TN) LCs cell, the electrically tunable THz polarization converters based on PEDOT/PSS films [11], and subwavelength metallic gratings [12] have been demonstrated recently. Their main contribution is they show that LC cell with functional electrodes, such as highly transparent thin films and metallically periodic structures, will has lots of advantages in electrically tunable THz elements. The three most important advantages are low operating voltages, high transmittance, and simple structures.

[2] The authors provided a couple of references where they describe in detail the properties of the ITO nanowhisker electrodes in the THz spectrum. Still, it would be very helpful to have an estimate of the losses in each layer of the investigated LC cells, such that it becomes clear if and where improvements can be expected. 

(Answer)

Thanks for this great suggestion!

In this work, we compared the 2 pi THz phase shifters with three different structures, respectively. Here, we will analyze the one (Type C) with the highest averaged transmittance.

From the Fig. 1(a) and the Ref. 18, Ref. 19, Ref. 20, and Ref. 24, we can obtain whole the complex permittivities (& refractive indices) of ITO nanostructures, fused silica, and liquid crystals. By combining these information, Jones matrix formulation, and Fresnel equations, we can estimate the loss caused from different interfaces. The results of estimation show the most serious sources of loss come from the interfaces of Fused Silica/ITO NWhs (Transmission coefficient~49%) and LCs/ITO NWhs (Transmission coefficient~36%). However, for the interfaces of Air/Fused Silica and LCs/Fused Silica, the loss of field is slight and around 68%, and 94%, respectively.

Author also modified the content of manuscript with underline and Red color.

Author Response File: Author Response.docx