SEARCH

SEARCH BY CITATION

Keywords:

  • all-optical signal processing;
  • polymethine;
  • aromatic counterions;
  • third-order non-linear optics;
  • silicon-organic hybrid devices

Considerable effort has been devoted to enabling all-optical signal processing (AOSP) for future communications systems with significantly increased operational bandwidth.1 Suitable materials for AOSP must exhibit ultrafast third-order non-linear optical (NLO) response, as well as low linear and non-linear optical loss, to achieve very high speed optical switching or modulation (100 Gbit/s or beyond).1b While silicon is an attractive candidate material for AOSP due to its ease of chip-scale photonic integration, its third-order NLO response is accompanied by parasitic non-linear absorption losses and long-lived free carrier effects.1c, 2 Recently, several groups have demonstrated silicon-organic hybrid (SOH) devices for AOSP with bandwidths exceeding 100 Gbit/s by integrating organic materials with silicon slot waveguides,3, 4 effectively combining the large non-linearities of organic materials with the extraordinary modal field concentrations of silicon waveguides. Nonetheless, development of organic materials suitable for device integration which possess large non-linearities and small optical losses in the telecommunications region remains a challenge.

Polymethine dyes5 are an intriguing class of conjugated organic molecules for AOSP applications: the π-electrons along the conjugated backbone can be polarized easily, resulting in a large negative third-order molecular polarizability (γ)6 and the electronic absorption bands are usually very narrow with sharp low energy band edges, which can be useful in reducing linear absorption losses at the desired operational wavelengths. Furthermore, polymethine dyes terminated with selenopyrylium end groups were recently found to exhibit large magnitudes of Re(γ) at telecommunications wavelengths while also suppressing non-linear absorption such that excellent two-photon figures-of-merit (FOMs) (i.e., |Re(γ)/Im(γ)|) were achieved.7 If the molecular third-order non-linearities of polymethine chromophores can be translated into macroscopic non- linearities (e.g., χ(3)) in high-number-density chromophoric materials, such materials could enable high-performance AOSP devices. However, there are several major challenges that need to be overcome in order to effectively translate the large γ into a large χ(3). At high molecular number densities, the intrinsic optical properties of polymethine dyes can be deleteriously affected by intermolecular interactions resulting in, for example, aggregation8 and ion-pairing effects,9 both of which can lead to an increase in absorptive optical loss (linear and non-linear) at AOSP wavelengths. To mitigate the potential absorptive loss from these interactions, we have employed polymer guest-host approach to develop processable films with high chromophore number density. However, with this approach, phase separation of polymethine salts in a host polymer can also lead to undesirable scattering and optical loss.

Here, we report on the development of a miscible polymethine salt, polymer guest-host blend that exhibits low optical loss and large third-order optical non-linearities exceeding that of silicon and gallium arsenide at 1550 nm, and demonstrate its efficacy in an SOH device. To develop optical quality films suitable for integration into SOH devices, we doped an amorphous polycarbonate (APC), a highly processable host polymer with a sizable refractive index, with an anionic polymethine chromophore based upon the strong electron withdrawing moiety, 4,5,5-trimethyl-3-cyano-2(5H)-furanylidenepropane-dinitrile (TCF), as terminal groups. The TCF polymethine is notably resistant to aggregation-induced effects based on the similarity of linear absorption spectra of high chromophore density films to those of dilute chromophore solutions.10a As polymethine dyes are highly polarizable and ionic while the APC host polymer is neutral, the resulting guest-host films often suffer from significant phase separation due to lack of sufficient solvation energy of the TCF polymethine and its counterion in the polymer matrix.

One approach to improve the miscibility of the TCF polymethine salt and the host polymer is to modify the structure of the counterions so as to increase attractive intermolecular interactions with the polymer. Specifically, we reasoned that aryl-based cations with a somewhat delocalized positive charge would reduce the strength of the electrostatic TCF polymethine-cation interaction, thereby reducing the ion-pairing energy, and that an additional moderately polar substituent might improve the solvation of the cation via attractive dipole-dipole interactions with the carbonate groups of the polymer. To test the validity of this hypothesis, a series of aromatic counterions such as benzyl-triethyl ammonium chloride, benzyl-triphenyl phosphonium bromide, pentafluorobenzyl-triphenyl phosphonium bromide, and (4-diethylaminobenzyl)-triphenyl phosphonium iodine, were paired with TCF polymethines in the preparation of AJBC 1722-1725 salts.

The polymethines have been prepared by simple ion exchange reactions between the aromatic cation containing salts and the sodium salt of TCF based anionic polymethines, as shown in Scheme S1 in the Supporting Information, according to the method reported in literature.10b A polymethine with an aliphatic counterion (AJBC 1721) was also prepared as a model compound for comparison. The molecular structures of the polymethine salts are summarized in Figure 1. The linear optical properties of these polymethines in both solution and the solid-state were investigated, and their absorption maxima (λmax) are summarized in Table 1. Figure 2a shows their UV–vis–near-IR spectra in dilute dichloromethane (DCM) solutions. The spectra show characteristic features of long-chain polymethines: sharp bands in the near-IR with very large extinction coefficients (see Table 1) and weak vibronic shoulders. The different counterions seem to have minimal influence on the π[RIGHTWARDS ARROW]π* electronic transitions in DCM as evidenced by the similarity of their absorption properties. Thin guest-host films (thickness ≈ 100 nm) of AJBC 1721-1725 in APC were prepared by spin-coating solutions with the appropriate weight ratio of dye and polymer onto glass substrates. The absorption spectra of 50 wt% doped films are shown in Figure 2b. Overall, the film spectra exhibit qualitatively similar features to their solution spectra, albeit with increased band broadening, which suggests either inhomogeneous broadening, aggregation or modulation of the ground-state geometry9 is playing a role. In particular, the polymethine salt with the aliphatic cation (AJBC 1721) suffers greater bandwidth broadening compared to the other four polymethines with aromatic cations (AJBC 1722-1725). The film absorption spectrum of AJBC 1725, on the other hand, exhibits the narrowest linewidth. These observations are consistent with enhanced solubility for polymethine salts with bulky, polarizable aromatic counterions bearing a moderately polar group in the polycarbonate host polymer. Differential scanning calorimetry (DSC) measurements show that the polymethine salts with aromatic substituted cations AJBC 1722-1725 possess good thermal stability, with glass-transition temperatures, Tg, that are significantly higher (Supporting Information, Figure S1) than that of AJBC 1721 with a methyltrioctylammonium cation.

thumbnail image

Figure 1. The chemical structures of the TCF polymethine salts, AJBC 1721-1725.

Download figure to PowerPoint

thumbnail image

Figure 2. The absorption spectra of polymethines in: a) dilute DCM solutions (concentration of ≈2 × 10−6 M) and b) thin guest-host APC films doped with 50 wt% of chromophores.

Download figure to PowerPoint

Table 1. Linear optical and thermal properties of TCF polymethines.
 λmax,sola)λmax,fila)ϵmaxb)Tgc)
Dye[nm][nm][× 105 M−1 cm−1][°C]
  • a)

    a)Absorption maxima wavelengths of polymethines in dilute DCM solution or in thin films;

  • b)

    b)Molar extinction coefficients measured in dilute DCM solution. Error bars are estimated at ±2%.

  • c)

    c)Glass-transition temperatures (Tg) measured by DSC analyses.

AJBC 17219029383.156
AJBC 17229029213.5110
AJBC 17239029223.4106
AJBC 17249029222.9109
AJBC 17259029202.592

Thicker films (≈2 μm) of the guest-host polymers were prepared for morphological studies as well as characterization of their bulk linear and non-linear optical properties, by using the same process described above. The transmission optical microscopy images of these thick films under different magnifications are shown in Figure 3. Under 10× magnification (Figure 3d and 3g), the AJBC 1722 and 1725 films are noticeably more homogenous than that of AJBC 1721 (Figure 3a), which exhibits clear evidence of phase separation. At 50× magnification, evidence of phase separation can also be observed for the AJBC 1722 film (Figure 3e), while the AJBC 1725 film remains optically homogeneous even at a magnification of 100× (Figure 3i). These results seem to correlate well with the hypothesis that moderately polar, aromatic counterions with diffuse charge localization can be used to enhance the compatibility between polymethine salt guests and the APC host polymer.

thumbnail image

Figure 3. Optical transmission micrographs of thick guest-host APC films doped with 50 wt% of polymethine salt under different magnifications.

Download figure to PowerPoint

The refractive indices and linear optical losses of the films at 1550 nm were measured by the prism coupling method and the results are summarized in Table 2 (details given in the Supporting Information). The linear losses of the AJBC 1722 and AJBC 1724 films exceeded the detection limit of 20 dB/cm, while the mode-coupling profiles were found to be excessively broad (Figure S3). The large linear losses may be due to the weak polarity and polarizability of the counterion in AJBC 1722 and to the steric bulk of the fluorine substituents of the counterion in AJBC 1724, each adversely affecting solvation in polycarbonate host. On the other hand, the films containing AJBC 1723 and 1725 showed significantly lower losses of 9 dB/cm and 7.5 dB/cm, respectively, with reasonably narrow mode-coupling profiles (Figure S3). The reduced loss for AJBC 1723 and 1725 correlates with the enhanced optical quality observed for these films under optical microscopy.

Table 2. The linear and non-linear optical properties at 1550 nm of thick guest-host APC films doped with 50 wt% polymethine salt.
DyeIndexa)Lossb) [dB/cm]|χ(3)|c) [10−11 esu]|Reχ(3)/Imχ(3)|d)
  • a)

    a)The refractive index measured by prism coupling;

  • b)

    b)The linear optical loss measured by prism coupling. NA: not available, due to loss exceeding the deteciton limit;

  • c)

    c)The magnitude of the third-order susceptibility measured by femtosecond pulsed Z-scan; |χ(3)| = [(Reχ(3))2 + (Imχ(3))2]1/2, see Table S2 for Reχ(3) and Imχ(3) values. The errors were estimated to be ±10%;

  • d)

    d)Two-photon FOM measured by femtosecond pulsed Z-scan. The errors were estimated to be ±14%.

AJBC 17221.79NA4.53.4
AJBC 17231.799.03.82.7
AJBC 17241.76NA4.53.1
AJBC 17251.767.54.02.7

Third-order macroscopic non-linearities of composites containing AJBC 1722-1725 were measured at 1550 nm using the femtosecond pulsed Z-scan method7, 8 and the results are summarized in Table 2. First, it should be noted that third-order molecular polarizabilities, γ, were also determined in a similar manner for these molecules in solution (see Table S1). By employing a linear number density extrapolation to predict χ(3) values from γ values (see Table S1), it was found that these extrapolated values were in reasonable agreement with the values determined from the blended films suggesting that the chromophores can retain their polymethine-like characteristics after being processed into high number density solid films. This is encouraging particularly given the deleterious impacts that aggregation can have on the NLO response of polymethines in the solid-state.8 As a result, these polymethine/APC blend films exhibit large third-order susceptibilities11 that exceed those of organic materials previously used in SOH devices3, 4 and are approximately three times larger than the value for silicon.12 It should be noted that the film containing AJBC 1722 showed spot-to-spot variations in the measured |χ(3)| values (Table S2) indicating spatial inhomogeneity, likely a result of the aforementioned phase-separation in the film, an issue that was not encountered with the other films.

The macroscopic two-photon FOMs (i.e., |Reχ(3)/Imχ(3)|) for these polymethine/APC blend films are not sufficiently large for phase-based all-optical switching applications.7 In order to understand the origin of these reduced FOMs, non-degenerate two-photon absorption (ND-2PA) spectra of the films were measured to determine the positions of the two-photon resonances (see the Supporting Information for details). The lowest-lying 2PA bands correspond to vibronically allowed resonances, as often seen for other polymethines.7, 13 These bands are in close proximity to the excitation wavelength of 1550 nm which explains the reasonably large Imχ(3) values as well as the suppressed |Reχ(3)/Imχ(3)| ratios shown in Table 2. In fact, these Imχ(3) values correspond to 2PA coefficients (β) of ≈13 cm/GW,14 which exceed the value found for gallium arsenide, a semiconductor known to have a large non-linear absorptive response in the telecommunications region.12

Despite sub-optimal FOMs for phase-based switching, the large values of |χ(3)| and β are enabling for other AOSP applications, such as wavelength conversion via four-wave mixing (FWM)3, 15 and loss-based all-optical switching,16 respectively. To demonstrate the feasibility of using these materials for such applications, these TCF polymethine/APC blend films were applied as claddings on top of asymmetric slot silicon waveguides to create SOH devices. The calculated TE0 mode electric field distribution and a cross-sectional image of the silicon waveguide device obtained using SEM are shown in Figure 4. The processability imparted through the use of polymer guest-host blends allowed for the claddings to be applied via a facile spin-casting method. Wavelength conversion was performed using near-degenerate FWM in the SOH devices using a pulsed pump laser and a CW signal laser that were amplified and coupled into the SOH waveguide.17 In a typical measurement, the pump laser was centered at ≈1548 nm, with a repetition rate of 1 MHz and a 10 ns pulse width. The wavelength of the CW signal laser was ≈1550 nm. When coupled into the SOH waveguide, the typical peak powers for the pump and probe lasers were 1 W and 10 mW (CW), respectively. An optical spectrum analyzer (OSA) with a wavelength resolution of 60 pm was used to determine the conversion efficiency for generation of the idler at the output of the waveguide. Figure 5 shows a typical output spectrum in the FWM experiment. Simulation of the FWM conversion generated in the SOH (see the Supporting Information) also allows for the determination of the magnitude of the third-order susceptibility of the organic claddings. The |χ(3)| values obtained from the SOH waveguides were significantly larger than the |χ(3)| of silicon measured using the same testing system (≈1.1 × 10−11 esu, which is in good agreement with values reported in the literature12). The non-linearities extracted for 25 wt% guest-host blends of AJBC 1723 (2.0 × 10−11 esu) and AJBC 1725 (4.5 × 10−11 esu) were found to be in reasonably good agreement with the values extracted from the free-space femtosecond pulsed Z-scan technique (see Table S2).18 The strong non-linear absorption of these blends enabled observation of optical power modulation in an SOH device using a pump-probe geometry (see the Supporting Information for details). Furthermore, simulations of the loss modulation permitted extraction of the 2PA coefficients that were found to be in good agreement with the values discussed above.

thumbnail image

Figure 4. a) Calculated optical mode pattern (electric-field distribution |E|) for the TE0 mode of the asymmetric slotted waveguide (see the Supporting Information). b) SEM micrograph of the cross-section of a fabricated slotted waveguide waveguide with 100 nm and 310 nm width silicon rails, 180 nm slot width and 200 nm silicon rail thickness.

Download figure to PowerPoint

thumbnail image

Figure 5. Optical spectrum showing idler wavelength generated by near-degenerate four-wave mixing in silicon-organic hybrid slotted waveguide device coated with 25 wt% AJBC 1723 in amorphous polycarbonate. Note that the down-converted light was filtered out by the measurement system.

Download figure to PowerPoint

In summary, we have demonstrated a method to improve the optical quality of guest-host films of TCF-based polymethines at high number density in amorphous polycarbonate by modification of the structure of the aromatic cations of the polymethine salts. The third-order optical non-linearities of these guest-host blend films were investigated in detail and the |χ(3)| values were found to be three times larger than that of silicon at 1550 nm, while the β values exceeded that of gallium arsenide at the same wavelength. Integration of these highly non-linear materials in SOH waveguides has enabled optical wavelength conversion by FWM and optical power modulation via 2PA suggesting that such polymethine-based guest-host materials may be attractive candidates for AOSP applications.

Supporting Information

  1. Top of page
  2. Supporting Information
  3. Acknowledgements
  4. Supporting Information

Supporting Information is available from the Wiley Online Library or from the author.

Acknowledgements

  1. Top of page
  2. Supporting Information
  3. Acknowledgements
  4. Supporting Information

This work was supported by the NSF STC (DMR-0120967), the DARPA ZOE Program (W31P4Q-09-1-0012), and the AFOSR MURI (FA9550-10-1-0558). A.K.-Y.J. thanks the Boeing-Johnson Foundation for its support.

  • 1
    a) K. Nozaki1, T. Tanabe, A. Shinya, S. Matsuo, T. Sato, H. Taniyama, M. Notomi, Nat. Photonics 2010, 4, 477;b) A. Martínez, J. Blasco, P. Sanchis, J. Galán, J. García-Rupérez, E. Jordana, P. Gautier, Y. Lebour, S. Hernández, R. Spano, R. Guider, N. Daldosso, B. Garrido, J. M. Fedeli, L. Pavesi, J. Martí, Nano. Lett. 2010, 10, 1506;c) V. R. Almeida, C. A. Barrios, R. R. Panepucci, M. Lipson, Nature 2004, 431, 1081.
  • 2
    J. Leuthold, C. Koos, W. Freude, Nat. Photonics 2010, 4, 535.
  • 3
    C. Koos, P. Vorreau, T. Vallaitis, P. Dumon, W. Bogaerts, R. Baets, B. Esembeson, I. Biaggio, T. Michinobu, F. Diederich, W. Freude, J. Leuthold, Nat. Photonics 2009, 3, 216.
  • 4
    M. Hochberg, T. Baehr-Jones, G. Wang, M. Shearn, K. Harvard, J. Luo, B. Chen, Z. Shi, R. Lawson, P. Sullivan, A. K. Y. Jen, L. Dalton, A. Scherer, Nat. Mater. 2006, 5, 703.
  • 5
    A. Mishra, R. K. Behera, P. K. Behera, B. K. Mishra, G. B. Behera, Chem. Rev. 2000, 100, 1973.
  • 6
    a) S. R. Marder, J. W. Perry, G. Bourhill, C. B. Gorman, B. G. Tiemann, K. Mansour, Science 1993, 261, 186;b) S. R. Marder, C. B. Gorman, F. Myers, J. W. Perry, G. Bourhill, J. L. Brédas, B. M. Pierce, Science 1994, 265, 632;c) D. Lu, G. Chen, J. W. Perry, W. A. Goddard III, J. Am. Chem. Soc. 1994, 116, 10679;d) J. M. Hales, S. Zheng, S. Barlow, S. R. Marder, J. W. Perry, J. Am. Chem. Soc. 2006, 128, 11362.
  • 7
    J. M. Hales, J. Matichak, S. Barlow, S. Ohira, K. Yesudas, J.-L. Brédas, J. W. Perry, S. R. Marder, Science 2010, 327, 1485.
  • 8
    A. Scarpaci, A. Nantalaksakul, J. M. Hales, J. D. Matichak, S. Barlow, M. Rumi, J. W. Perry, S. R. Marder, Chem. Mater. 2012, 24, 1606.
  • 9
    P.-A. Bouit, C. Aronica, L. Toupet, B. L. Guennic, C. Andraud, O. Maury, J. Am. Chem. Soc. 2010, 132, 4328.
  • 10
    a) P.-A. Bouit, D. Rauh, S. Neugebauer, J. L. Delgado, E. Di Piazza, S. Rigaut, O. Maury, C. Andraud, V. Dyakonov, N. Martin, Org. Lett. 2009, 11, 4806;b) P.-A. Bouit, E. Di Piazza, S. Rigaut, B. Le Guennic, C. Aronica, L. Toupet, C. Andraud, O. Maury, Org. Lett. 2008, 10, 4159.
  • 11
    Conversion between the cgs units presented here and MKS units can be accomplished according to the following: χ(3) [m2 V−2, MKS] = (4π/9 × 10−8)χ(3) [esu or cm2 statVolt−2, cgs]. Accordingly, the values of |χ(3)| in Table 2 exhibit an average value of 5.8 × 10−19 m2 V−2. These third-order susceptibilities can also be converted to non-linear refractive indices (n2) according to the following equation: n2 = 3Reχ(3)/(4ϵ0cn02), where n0 is the refractive index of the medium. Consequently, the values of Reχ(3) in Table S2 can be converted to an average value of n2 = −5.0 × 10−17 m2 W−1.
  • 12
    M. Dinu, F. Quochi, H. Garcia, Appl. Phys. Lett. 2003, 82, 2954.
  • 13
    J. Fu, L. A. Padilha, D. J. Hagan, E. W. Van Stryland, J. Opt. Soc. Am. B 2007, 24, 56.
  • 14
    The values of Imχ(3) in Table S2 can be converted to two-photon absorption coefficients (β) according to the following equation: β = 3π(Imχ(3))/(ϵ0cn02λ), where n0 is the refractive index of the medium and λ is the excitation wavelength.
  • 15
    a) C. Koos, L. Jacome, C. Poulton, J. Leuthold, W. Freude, Opt. Express 2007, 15, 5976;b) T. Vallaitis, S. Bogatscher, L. Alloatti, P. Dumon, R. Baets, M. L. Scimeca, I. Biaggio, F. Diederich, C. Koos, W. Freude, J. Leuthold, Opt. Express 2009, 17, 17357;c) Q. Lin, O. J. Painter, G. P. Agrawal, Opt. Express 2007, 15, 16604.
  • 16
    a) B. C. Jacobs, J. D. Franson, Phys. Rev. A 2009, 79, 063830;b) Y. H. Wen, O. Kuzucu, T. Hou, M. Lipson, A. L. Gaeta, Opt. Lett. 2011, 36, 1413.
  • 17
    A. Spott, T. Baehr-Jones, R. Ding, Y. Liu, R. Bojko, T. O'Malley, A. Pomerene, C. Hill, W. Reinhardt, M. Hochberg, Opt. Express 2011, 19, 10950.
  • 18
    Guest-host blends with slightly lower doping wt% (25 wt% instead of 50 wt%) were chosen for the organic claddings due to larger linear losses found in the SOH waveguides than those measured in free-space (see Table 2). As no additional measures were taken to ensure optimal interfacial coupling between the cladding and guide, this discrepancy is likely due to additional scattering losses in the SOH geometry. Nonetheless, the optical losses of the 25 wt% AJBC 1723 and 1725 cladding films were found to be 17 and 10 dB/cm, respectively.

Supporting Information

  1. Top of page
  2. Supporting Information
  3. Acknowledgements
  4. Supporting Information

Detailed facts of importance to specialist readers are published as ”Supporting Information”. Such documents are peer-reviewed, but not copy-edited or typeset. They are made available as submitted by the authors.

FilenameFormatSizeDescription
adma_201202325_sm_suppl.pdf365Ksuppl

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.