Bacterially synthesized tellurium nanostructures for broadband ultrafast nonlinear optical applications

Elementary tellurium is currently of great interest as an element with potential promise in nano-technology applications because of the recent discovery regarding its three two-dimensional phases and the existence of Weyl nodes around its Femi level. Here, we report on the unique nano-photonic properties of elemental tellurium particles [Te(0)], as harvest from a culture of a tellurium-oxyanion respiring bacteria. The bacterially-formed nano-crystals prove effective in the photonic applications tested compared to the chemically-formed nano-materials, suggesting a unique and environmentally friendly route of synthesis. Nonlinear optical measurements of this material reveal the strong saturable absorption and nonlinear optical extinctions induced by Mie scattering over broad temporal and wavelength ranges. In both cases, Te-nanoparticles exhibit superior optical nonlinearity compared to graphene. We demonstrate that biological tellurium can be used for a variety of photonic applications which include their proof-of-concept for employment as ultrafast mode-lockers and all-optical switches.

2) Why the nonlinear scattering properties of the Te nanocrystals in suspension are better than other materials such as single walled carbon nanotubes, or PcZn (t-Bu4PcZn)?
3) In solving Eq. (2) the beta is fixed as a constant; and it is extracted when the numerical solution is fit to the measurement. However, Fig. 3c-d show clearly that the beta is no longer a constant and it depends on the light irradiance. Therefore, it is meaningless. Eq. (2) is wrong in this case. 4) Why the thermo-optic coefficient of the Te composites are larger than the other material mentioned in the manuscript? 5) Can the authors elaborate the positive sign of the thermos-optic coefficient of the Te composites? Why is it?
Minor points: 1) On Page 6, the definition of Im (3) is proportional to NL which depends on the light irradiance by NL=0/(1+I/Isat). It is not correct! It should be "NL ≈ -0/Isat".

Reviewer #1 (Remarks to the Author):
This manuscript describes the potential use of bacterially grown Tellurium nanocrystal in laser devices. The work focuses mainly on saturable absorption, and to a lesser extent thermo-optic switching effects. The use of bio-grown materials for applications is quite interesting. Given the reported experiments, their potential for use is intriguing, but their real potential remains speculative as described below.

Reply:
We firstly wish to thank the reviewer for his/her useful suggestions. The reviewer's recognizing on the high quality and well presentation of our work is highly appreciated.
(1) The thermo-optic switching was tested in PMMA-Te composites. Indeed, the authors were able to demonstrate thermo-optic switching. However, their claims of speed are beyond speculative. Given the composite nature, the thermal conductivity will be limited by the polymer host. Indeed, examination of the switching transients indicate that the speed is limited by the thermal conductivity of the polymer in the millisecond range. The authors speculate that high speed switching could be demonstrated. However, they have not listed the thermal conductivity of Te to back this up, nor have they described how they might fabricate a sample that would demonstrate this speed. This would be much more interesting if they would have demonstrated high-speed switching.
Reply: Thanks for the reviewer for his/her suggestion in demonstration of Te as high-speed thermal-optic switching material. To back this up, we have conducted new comparative studies on the thermal-optical responses between a few well-known 2D materials (graphene and WS2) and tellurium (See Figure R1 below). To this end, we prepared a new bunch of samples with graphene, WS2 and biological/chemical tellurium hosted in PMMA. The optical switching speeds were investigated by measuring the rise/fall time of the modulated pulses. We found that in our experiments, the rising times have minor difference and are much shorter than the falling times. Therefore, the falling times are the bottle-neck of the response speed in this comparative study. Our measurements show both tellurium sample exhibit obvious less falling time than graphene and WS2.
Moreover, the modulated output from the tellurium samples have better signal-to-noise ratio than those from graphene and WS2. We believe such experiments demonstrate the tellurium can be better material than graphene and WS2 for thermal-optical switching. These new results are updated in the  (2) Regarding saturable absorption and applications, the material is indeed interesting as a potential high-performance mid-IR saturable absorber. However, experiments are performed in the near-IR, while the authors speculate that the small bandgap promises mid-IR performance.
Again, this has not been demonstrated. Its mid-IR performance may be limited because it is formed as a composite and the mid-IR absorption of the host might be a problem. Should the authors fabricate a sample and demonstrate superior mid-IR performance, the work would indeed be of keen interest to the community.

Reply:
We appreciate for this useful suggestion from the referee. As suggested by the reviewer, We thank the referee for his /her suggestion about the further demonstration of mid-IR NLO performance of tellurium, which helps to improve the quality of this work.

Figure R2
Mid-infrared open z-scans of Bio-Te and graphene PMMA films at 2.5 μm (a) and 2.8 μm (b) wavelengths, showing better saturable absorptive responses of Bio-Te than that of graphene.
(3) In summary, the use of bio-based Te materials for optoelectronics is quite interesting, and the preliminary experiments suggest that they might be. However, a number of claims are made regarding their potential that are not supported by data in these preliminary experiments.
Additionally, the authors do not suggest a path for producing the materials that would demonstrate these capabilities. The work could be of immediate interest and influence thinking in the field should the authors be able to demonstrate the performance on which they have speculated.
Reply: Following the suggestions by the reviewer, we have carried out additional experiments and demonstrated the capabilities of Te as the saturable absorber in mid-infrared as well as its superior performance in thermal-optical response than WS2 and graphene. We gratefully appreciate that the referee for his/her perspective suggestions in experiments that we can follow to make this work more complete and impactive. Although their results are interesting, there are a number of issues for the authors to consider for the improvement. The major issues are listed below: Reply: We thank the reviewer for his/her carefulness to point out our typos and improper expressions that save us from awkward.

Major issues:
1) Why the saturable absorption properties of the Te nanocrystals in suspension are superior to other materials such as graphene?
Reply: We thank the referee for his/her consideration on the superior saturable absorption of Te. In visible-NIR range, the better saturable absorption of Te may due to larger absorption crosssection than that of graphene in some wavelengths or longer saturable absorber recovery time [Keller, U., et al., IEEE J. Sel. Top. Quantum Electron. 2, 435-453 (1996)]. Imaginary part of dielectric permittivity of Te in visible and NIR range is also larger than Si, Ge, and even Au Considering that molar density of graphene is 3.75 that of Te, we will have their  ratio of 4.5 in favor of tellurium. In the mid-infrared, the graphene has two-photon absorption, which weaken the Such discussions are important and we thank the referee for reminding us. We emphasize this point in the revised manuscript (Line 21 Page 10 to Line 5 Page 11) As for the comparison with PcZn, from literature [Opt. Lett. 19, 625-627 (1994); Int. Rev. Phys.
Chem. (2012) 31(3), 319.] we know that in the optical limiting behavior of PcZn (also for other phthalocyanine solution) either two-photon absorption or reverse saturable absorption are prevailing.
These effects are determined solely by properties of the molecular electronic structure and represent a different physics to the nonlinear scattering processes we discussed above.

Calculation of /Cp for Te and graphite:
The linear absorptive coefficient α0 has a relationship α0 = N  ρNA/M, where the N is absorber particle density,  is absorptive cross-section, ρ is material density, NA is the Avogadro constant and M is the molar mass. From literature [CRC Handbook of Chemistry and Physics, 97th ed.], we know that the absorption coefficient α0 of graphite ~2.21×10 3 cm -1 and tellurium ~8.43×10 3 cm -1 at 1 eV photon energy. The molar thermal capacity Cp of Te and graphite is 25.73 J/mol/K and 8.517 J/mol/K respectively. The molar mass, M, of Te and graphite is 127 g/mol and 12 g/mol. The density ρ of Te and graphite is 6.3 g/cm 3 and 2.3g/cm 3 . With all the data above, the (/Cp)Te / (/Cp)graphite is calculated to be ~4.87.
3) In solving Eq. (2) the beta is fixed as a constant; and it is extracted when the numerical solution is fit to the measurement. However, Fig. 3c-d show clearly that the beta is no longer a constant and it depends on the light irradiance. Therefore, it is meaningless. Eq. (2) is wrong in this case.

Reply:
We thank the author for attentions on the correctness of Eq.(2). The Eq. 2, dI dz ⁄ = −(α + β I)I, as its form shows, takes account into second-order nonlinearity, i.e., nonlinearity with order higher than 3 won't be reflected by this equation. This equation has shown great fits to the experimental z-scan curves in Fig. 3a-b, implying the beta exacted from these fittings are reasonable for the corresponding laser pulse energy. The reason that the beta changes with the light irradiance is likely due to higher order nonlinearities, which didn't show in a single z-scan curve in  where we can see Te has obviously larger thermal expansion than other materials mentioned in the manuscript. Therefore, we believe the large thermal expansion coefficient may explain the strong thermo-optic effect of Te.
Literatures: Reply: Thank you for considering the sign of thermo-optic coefficient of Te composites and helping us to find this mistake. The sign should be negative and we correct this in revised manuscript. To confirm the value of thermo-optic coefficient, we have performed a series of temperature-dependent ellipsometry measurement on the Bio-Te-PMMA thin film on silicon substrate. The results are showed in the Figure R3 (a) below. The refractive indexes at 1.55 μm as a function of temperature are summarized in Figure R3(b), where the corresponding thermo-optic coefficient of Bio-Te-PMMA thin film is fit to be about -1.64×10 -4 K -1 .
To obtain the approximate thermo-optic coefficient of Bio-Te nanocrystal, we did the following estimation: The contribution is approximately proportional to the ratio of optical intensity distribution and in a thin film this is the ratio of material volume [K Wu et al., Photonic Research,