Graphene-based optical modulators have recently attracted much attention because of their characteristic ultrafast and broadband response. Their modulation depth (MD) and overall transmittance (OT), however, are often limited by optical loss arising from interband transitions. We report here an all-optical, all-fiber optical modulator with a Mach–Zehnder interferometer structure that has significantly higher MD and OT than graphene-based loss modulators. It is based on the idea of converting optically induced phase modulation in the graphene-cladded arm of the interferometer to intensity modulation at the output of the interferometer. The device has the potential to be integrable into a photonic system in real applications.
© 2016 Optical Society of America
Optical modulators play a crucial role in optical interconnects [1–3]. Graphene has generated exceptional interest for optical modulation because of its controllable carrier doping [4,5], broadband optical response , fast carrier relaxation time [7–9], and highly compatible integration . Liu et al. first demonstrated a graphene modulator by electrically tuning the Fermi level of a graphene film covering a silicon waveguide, but parasitic response limited the modulation bandwidth to 1 GHz . Since then, much effort has been devoted to circumventing the main issues of graphene-based optical modulation [12–34], including the relatively slow response time, low modulation depth (MD), and high insertion loss, originating from graphene properties and modulation schemes. Through optimizing the resistance-capacitance (RC) limit of the device, a modulator with 30 GHz bandwidth based on loss modulation in a critically coupled resonator was achieved . To overcome the “electrical bottleneck,” Li et al. demonstrated an all-optical modulator in a graphene-clad microfiber (GCM) with a bandwidth larger than 200 GHz . However, while high MD and overall transmittance (OT) from a modulator are often desired in real applications, the strong linear absorption in graphene unavoidably limits both MD and OT of a loss-based modulator to relatively low values.
Recently, Gan et al. reported a graphene phase shifter in a Mach–Zehnder interferometer (MZI) that, relying on an ohmic-heating-induced thermo-optic effect, could achieve a phase shift of . The thermal response time is unfortunately very slow , of the order of milliseconds. For ultrafast phase modulation, one needs to employ optical-field-induced refractive index change (optical Kerr effect) for phase modulation, which is known to be large in graphene in the nonlinear absorption regime . We demonstrate here an all-optical MZI modulator using a GCM as a phase modulator [25,37,38] in one arm of the MZI. Wave interference at the output from the MZI then converts the phase modulation from GCM to intensity modulation with high MD. Compared with graphene modulators relying on loss modulation , we obtained an increase of 4.6 times in MD with the MZI modulator.
Figure 1 shows a sketch of the experimental arrangement for two types of all-optical modulators with GCM, one based on loss modulation [Fig. 1(a)] and the other based on a MZI with phase modulation in one arm [Fig. 1(b); see also Fig. S1 in Supplement 1 for the experimental setup]. In the former case, the signal light was sent through the GCM tapered from a standard optical fiber, and the switching light was coupled into the GCM by a wavelength division multiplexing (WDM) device, as reported previously . The signal output, after going through an in-line fiber-optic long-pass filter used to block the switching light, was detected by an avalanche photodiode (APD). In the MZI case, the signal light was first split into two with a predetermined power ratio and sent into the two fiber arms of the MZI. The variable optical attenuator (VOA) in each arm was used to fine-tune the power ratio.
The GCM was fabricated by transferring and wrapping a chemical vapor deposition (CVD)-grown multilayer graphene sheet onto a free-standing microfiber ; typical optical and electron microscope images of a GCM are shown in Figs. 2(a) and 2(b), respectively. Here the multilayer graphene sheet was chosen to get a higher modulation depth and lower power consumption . The darkened area on the microfiber in Fig. 2(a) indicates the 20-μm-long graphene cladding, and the smooth surface with slight ripples in Fig. 2(b) shows a close view of the high-quality graphene cladding. Figure 2(c) displays the broadband linear transmittance of a GCM with a diameter of 1.1 μm and a graphene cladding length of 20 μm. The OT is around 10% from 1200 to 1700 nm; fluctuations in the spectrum are due to scattering loss from imperfections of the fiber coupler. Figure 2(d) gives the input-power-dependent transmission of a GCM (1 μm in diameter and 15 μm in graphene cladding length) measured with 1550 nm femtosecond laser pulses (83 MHz, 220 fs), exhibiting a differential transmittance of 20%, arising from nonlinear absorption of the GCM.
The all-fiber MZI in Fig. 1(b) had an optical path difference of between the two arms. Figure 2(e) shows the interference output of the MZI with the 1 μm GCM in one arm measured by a continuous wave (CW) diode laser. The interference spectrum has a free spectral range (FSR) of 340 pm and an extinction ratio of 20 dB around the 1550 nm wavelength.
Nonlinear absorption and the associated optical Kerr effect in graphene allow the GCM to function as an all-optical modulator to modulate the intensity and phase of the light going through it [16,36]. In our study, we used a CW laser beam (with linewidth) at 1550 nm as signal light, and 1064 nm nanosecond pulses (8 ns, 4.8 kHz) as the switching light. The latter consisted of one-strong–one-weak pulse pairs with peak intensity ratio of about 1.2, as shown in the top panel of Fig. 3(a). We first measured the intensity modulation on the signal light by sending the switching pulse pairs through the GCM-contained fiber [16,36]. The observed modulated output was presented in the middle panel of Fig. 3(a) with an input signal power of 200 μW. The OT was about 10%, and the MD was about 11.5% when the stronger switch pulse had a peak power of . The peak intensity ratio of the modulated signal pulse pair was about 1.7, which agrees well with expectations from the third-order nonlinear absorption behavior (i.e., approximately the third power of the intensity ratio of 1.2). It further confirms that loss modulation is induced by the saturable absorption effect of graphene.
To study the MZI modulation, we included the GCM-contained fiber as one arm in the MZI. The switching pulse pairs in the GCM-contained fiber modulated both the intensity and the phase of the signal light, but the phase modulation effect is much more significant on the output of the MZI. As seen in the bottom panel of Fig. 3(a), the MD for a signal power of 220 μW is already 16% when the peak power of the strong switching pulse is 0.75 W. Figure 3(b) plots the measured MD versus switching peak power at 1550 nm (at which the transmittance of the MZI is 3 dB below the maximum) in comparison with that obtained from loss modulation through the GCM-contained fiber described earlier. In both cases, MD increases with switching power, but the former is obviously much better; it increases to 52.5% if the switching peak power increases to 1.18 W, and is nearly 4.6 times better than the latter case (MD ). Also shown in Fig. 3(b) for comparison is the MD resulting from loss modulation in only the GCM-contained arm of the MZI. Because only half of the output signal from the MZI comes from the GCM arm, the contribution of MD from loss modulation in the GCM arm to the overall MD from the MZI appears insignificant. The MD of 52.5% measured here corresponds to a phase shift of about 0.18 π in the GCM arm. The phase shift should arise from the refractive index change induced in graphene by interband optical excitation of carriers . Since only half of the output signal of the MZI goes through the lossy GCM arm, the OT of the MZI modulator is also higher than the GCM-contained fiber modulator with the same signal output. For example, for an input signal power of 100 μW, the OT of the MZI is , while that of the GCM-contained fiber alone is 10%. Thus the MZI modulator has the advantages of both higher MD and higher OT than the GCM loss modulator.
Larger MD is clearly more desirable. Unfortunately, graphene is quite lossy, which limits the optimization of OT and MD simultaneously. Increasing MD or the phase shift in the GCM of the MZI requires increase of the graphene cladding length of the GCM, which, in turn, necessarily decreases the OT. For our MZI modulator, we have calculated the relation between MD and OT depicted in Fig. 3(c) (see Supplement 1 for more details). With a 3 μm long graphene cladding, the OT can be as high as 63% for the GCM-contained fiber modulator and 78% for the MZI modulator, but the corresponding MDs are only 2% and 11%, respectively. With a 45 μm graphene cladding length, the MDs can be as high as 39% and 99% in the two cases, but the corresponding OTs are only 0.1% and 0.2%, respectively. Considering possibility for practical applications, we selected in this work a graphene cladding length of 15 μm [points I and P in Fig. 3(c)] for the MZI modulator to achieve a moderate MD of 52.5% with an acceptable OT of 19%.
We notice in Fig. 3(a), with the expanded scale in Figs. 4(a)–4(c), that the pulse-modulated signal spikes consist of two parts—a nanosecond pulse resembling the profile of the switching pulse [Fig. 4(c)] and a long tail with a decay time of . The nanosecond pulse obviously comes from the refractive index change induced by optical excitation of carriers in graphene that has a picosecond time response . The long tail must arise from the thermally induced refractive index change by laser heating of graphene as mentioned earlier. The rise time of the tail is about 3 μs [Fig. 4(b)], which is 3 orders of magnitude shorter than what was reported in a previous study . This is because of the much shorter and thinner fiber used in our experiment. Also, it should be noted that this slow-response thermal effect may limit the ultrafast all-optical graphene modulation with high duty cycle due to switch-light-induced thermal accumulation.
In the above discussion, the switching light induces a larger OT or positive modulation on the output signal from the MZI modulator. However, it is also possible to have negative modulation (smaller OT) or positive and negative modulations separately for fast and slow responses, as presented in Fig. 4(e). Sketched in Fig. 4(f) is the dependence of the output from the MZI on the initial phase difference between the two arms of the MZI. For a given wavelength (see Supplement 1), if we set the operating point at P, then a positive phase shift in the GCM arm of the MZI should induce an increased output or positive modulation. On the other hand, if the operating point is set at N, a negative modulation should appear. At O near a valley, the larger fast phase shift yields positive modulation and the smaller slow phase shift yields negative modulation. This illustrates that the MZI modulator is much more flexible than a loss modulator.
In summary, we have demonstrated an all-optical, all-fiber, fast MZI modulator that has relatively high MD and OT. It operates through optically induced phase modulation in the GCM-contained arm and then wave interference at the output of the MZI. It has clear advantages over graphene-based loss modulators in terms of MD, OT, and versatility. The all-fiber configuration is readily integrable with standard optical fiber systems. The all-optical phase modulation scheme can also be applied to a photonic modulator by using other two-dimensional materials (e.g., ) and/or functional structures (e.g., polymer nanofibers  and optical microcavities ), which may pave the way to practical applications of graphene-like materials for optical modulation.
National Basic Research Program of China (2013CB328703, 2014CB921600); National Natural Science Foundation of China (NSFC) (61475140, 11374065); Fundamental Research Funds for the Central Universities (2016FZA5003); U.S. Department of Energy (DOE), Office of Science, Office of Basic Energy Sciences, Materials Sciences, and Engineering Division (DE-AC03-76SF00098).
We thank Yingxin Xu, Yipei Wang, and Zhilin Xu for their help with the experiment.
See Supplement 1 for supporting content.
1. G. T. Reed, G. Mashanovich, F. Y. Gardes, and D. J. Thomson, Nat. Photonics 4, 518 (2010). [CrossRef]
2. G. L. Li and P. K. L. Yu, J. Lightwave Technol. 21, 2010 (2003). [CrossRef]
3. K. Liu, C. R. Ye, S. Khan, and V. J. Sorger, Laser Photon. Rev. 9, 172 (2015). [CrossRef]
4. R. R. Nair, P. Blake, A. N. Grigorenko, K. S. Novoselov, T. J. Booth, T. Stauber, N. M. R. Peres, and A. K. Geim, Science 320, 1308 (2008). [CrossRef]
5. F. Wang, Y. B. Zhang, C. S. Tian, C. Girit, A. Zettl, M. Crommie, and Y. R. Shen, Science 320, 206 (2008). [CrossRef]
6. K. F. Mak, L. Ju, F. Wang, and T. F. Heinz, Solid State Commun. 152, 1341 (2012). [CrossRef]
7. J. M. Dawlaty, S. Shivaraman, M. Chandrashekhar, F. Rana, and M. G. Spencer, Appl. Phys. Lett. 92, 042116 (2008). [CrossRef]
8. S. Kumar, M. Anija, N. Kamaraju, K. S. Vasu, K. S. Subrahmanyam, A. K. Sood, and C. N. R. Rao, Appl. Phys. Lett. 95, 191911 (2009). [CrossRef]
9. M. Breusing, C. Ropers, and T. Elsaesser, Phys. Rev. Lett. 102, 086809 (2009). [CrossRef]
10. F. N. Xia, H. Wang, D. Xiao, M. Dubey, and A. Ramasubramaniam, Nat. Photonics 8, 899 (2014). [CrossRef]
11. M. Liu, X. B. Yin, E. Ulin-Avila, B. S. Geng, T. Zentgraf, L. Ju, F. Wang, and X. Zhang, Nature 474, 64 (2011). [CrossRef]
12. Z. P. Sun, A. Martinez, and F. Wang, Nat. Photonics 10, 227 (2016). [CrossRef]
13. M. Midrio, S. Boscolo, M. Moresco, M. Romagnoli, C. De Angelis, A. Locatelli, and A. D. Capobianco, Opt. Express 20, 23144 (2012). [CrossRef]
14. R. Hao, W. Du, H. S. Chen, X. F. Jin, L. Z. Yang, and E. P. Li, Appl. Phys. Lett. 103, 061116 (2013). [CrossRef]
15. Z. Liu, M. Feng, W.-S. Jiang, W. Xin, P. Wang, Q.-W. Sheng, Y.-G. Liu, D. N. Wang, W.-Y. Zhou, and J.-G. Tian, Laser Phys. Lett. 10, 065901 (2013). [CrossRef]
16. W. Li, B. G. Chen, C. Meng, W. Fang, Y. Xiao, X. Y. Li, Z. F. Hu, Y. X. Xu, L. M. Tong, H. Q. Wang, W. T. Liu, J. M. Bao, and Y. R. Shen, Nano Lett. 14, 955 (2014). [CrossRef]
17. M. Mohsin, D. Schall, M. Otto, A. Noculak, D. Neumaier, and H. Kurz, Opt. Express 22, 15292 (2014). [CrossRef]
18. C. Sturm, D. Tanese, H. S. Nguyen, H. Flayac, E. Galopin, A. Lemaitre, I. Sagnes, D. Solnyshkov, A. Amo, G. Malpuech, and J. Bloch, Nat. Commun. 5, 3278 (2014). [CrossRef]
19. L. Z. Yang, T. Hu, A. Shen, C. Y. Pei, B. Yang, T. G. Dai, H. Yu, Y. B. Li, X. Q. Jiang, and J. Y. Yang, Opt. Lett. 39, 1909 (2014). [CrossRef]
20. N. Youngblood, Y. Anugrah, R. Ma, S. J. Koester, and M. Li, Nano Lett. 14, 2741 (2014). [CrossRef]
21. S. Das, A. Salandrino, J. Z. Wu, and R. Q. Hui, Opt. Lett. 40, 1516 (2015). [CrossRef]
22. Y. H. Ding, X. L. Zhu, S. S. Xiao, H. Hu, L. H. Frandsen, N. A. Mortensen, and K. Yvind, Nano Lett. 15, 4393 (2015). [CrossRef]
23. X. T. Gan, C. Y. Zhao, Y. D. Wang, D. Mao, L. Fang, L. Han, and J. L. Zhao, Optica 2, 468 (2015). [CrossRef]
24. S. Y. Luo, Y. N. Wang, X. Tong, and Z. M. Wang, Nanoscale Res. Lett. 10, 1 (2015). [CrossRef]
25. M. Mohsin, D. Neumaier, D. Schall, M. Otto, C. Matheisen, A. L. Giesecke, A. A. Sagade, and H. Kurz, Sci. Rep. 5, 10967 (2015). [CrossRef]
26. C. T. Phare, Y. H. D. Lee, J. Cardenas, and M. Lipson, Nat. Photonics 9, 511 (2015). [CrossRef]
27. R. W. Yu, V. Pruneri, and F. J. G. de Abajo, ACS Photon. 2, 550 (2015). [CrossRef]
28. S. L. Yu, C. Meng, B. G. Chen, H. Q. Wang, X. Q. Wu, W. T. Liu, S. J. Zhang, Y. Liu, Y. K. Su, and L. M. Tong, Opt. Express 23, 10764 (2015). [CrossRef]
29. E. J. Lee, S. Y. Choi, H. Jeong, N. H. Park, W. Yim, M. H. Kim, J. K. Park, S. Son, S. Bae, S. J. Kim, K. Lee, Y. H. Ahn, K. J. Ahn, B. H. Hong, J. Y. Park, F. Rotermund, and D. I. Yeom, Nat. Commun. 6, 6851 (2015). [CrossRef]
30. C. Meng, S. L. Yu, H. Q. Wang, Y. Cao, L. M. Tong, W. T. Liu, and Y. R. Shen, Light. Sci. Appl. 4, e348 (2015). [CrossRef]
31. J. H. Chen, B. C. Zheng, G. H. Shao, S. J. Ge, F. Xu, and Y. Q. Lu, Light Sci. Appl. 4, e360 (2015). [CrossRef]
32. D. Ansell, I. P. Radko, Z. Han, F. J. Rodriguez, S. I. Bozhevolnyi, and A. N. Grigorenko, Nat. Commun. 6, 8846 (2015). [CrossRef]
33. L. H. Yu, J. J. Zheng, Y. Xu, D. X. Dai, and S. L. He, ACS Nano 8, 11386 (2014). [CrossRef]
34. Z. Shi, L. Gan, T. H. Xiao, H. L. Guo, and Z. Y. Li, ACS Photon. 2, 1513 (2015). [CrossRef]
35. M. K. Davis, M. J. F. Digonnet, and R. H. Pantell, J. Lightwave Technol. 16, 1013 (1998). [CrossRef]
36. H. Zhang, S. Virally, Q. L. Bao, L. K. Ping, S. Massar, N. Godbout, and P. Kockaert, Opt. Lett. 37, 1856 (2012). [CrossRef]
37. F. Zhou, R. Hao, X. F. Jin, X. M. Zhang, and E. P. Li, IEEE Photon. Technol. Lett. 26, 1867 (2014). [CrossRef]
38. B. C. Yao, Y. Wu, Z. G. Wang, Y. Cheng, Y. J. Rao, Y. Gong, Y. F. Chen, and Y. R. Li, Opt. Express 21, 29818 (2013). [CrossRef]
39. X. Wu, S. Yu, H. Yang, W. Li, X. Liu, and L. Tong, Carbon 96, 1114 (2016). [CrossRef]
40. F. Zhang, S. Han, Y. Q. Liu, Z. P. Wang, and X. G. Xu, Appl. Phys. Lett. 106, 091102 (2015). [CrossRef]
41. K. Wang, J. Wang, J. Fan, M. Lotya, A. O’Neill, D. Fox, Y. Feng, X. Zhang, B. Jiang, and Q. Zhao, ACS Nano 7, 9260 (2013). [CrossRef]
42. P. Wang, Y. Wang, and L. Tong, Light Sci. Appl. 2, e102 (2013). [CrossRef]