Porphyrin–poly(arylene ether sulfone) covalently functionalized multi-walled carbon nanotubes: synthesis and enhanced broadband nonlinear optical properties

Abstract

A novel multi-walled carbon nanotube (MWNT) hybrid was synthesized by covalently attaching a porphyrinated poly(arylene ether sulfone) onto the surface of carbon nanotubes. The polymer functionalized nanotube exhibited excellent solubility in organic solvents, and the weight% of MWNTs in the resulting hybrid was found to be 36%. Considerable fluorescence quenching and shortened fluorescence lifetime were observed in the MWNT hybrid due to the photo-induced electron/energy transfer behavior from porphyrin moieties to MWNTs. Z-scan measurements demonstrated the improved broadband nonlinear optical performance of the MWNT hybrid at both 532 and 1064 nm in nanoseconds, which is attributed to the accumulation effect of nonlinear absorption and scattering, and a remarkable photo-induced electron/energy transfer process. Meanwhile, the MWNT hybrid can be homogeneously dispersed in a polymer matrix, further fabricated into a uniform film with high transparency and ultrafast optical limiting response under femtosecond laser pulses.

---

Introduction

Following the rapid development of nanotechnology in the past few decades, a variety of nanomaterials have emerged to reveal their potential in the design and fabrication of nanoscale photoelectronic devices.¹ As one of the most representative of carbon-based nanomaterials, carbon nanotubes (CNT) stand out for their potential application in the fields of composites,² optoelectronics,³ and sensors due to their unique properties.⁴ Of particular interest to our research are the CNT nanohybrids with superior nonlinear optical (NLO) properties, which motivates the development of optical limiting (OL) materials to protect human eyes and delicate sensors.⁵ Single-walled carbon nanotube (SWNT) and multi-walled carbon nanotube (MWNT) suspensions have been demonstrated to show superior optical limiting effects at 532 nm and 1064 nm in the nanosecond regime;^6–8 their broadband NLO properties primarily arise from strong nonlinear scattering owing to the formation of scattering centers, including thermally-induced solvent microbubbles and ionized carbon microplasmas.⁹ However, no single material has yet been considered as an ideal optical limiter capable of attenuating laser beams of various waveband (UV-vis-NIR) and pulse width (from nanosecond to femtosecond). To promote the OL effect and combine the characters of different materials, a promising approach is to functionalize carbon nanotubes with fullerene,¹⁰ organic chromophores,^11–13 conjugated polymers,^14–16 and nanoparticles with nonlinear optical absorption (NLA) properties.^17–19 As a result of the combination of different nonlinear optical mechanism and chemical properties of each component, these CNT-based nanohybrids exhibit enhanced optical limiting properties through a synergistic effect, as well as the improved solubility and compatibility of CNT in solvent and composite materials.^10–23

Unlike carbon nanotubes, porphyrins with extended π electrons have been the most intensively studied optical limiting materials for their strong reverse saturable absorption (RSA), high triplet yields, and long excited-state lifetimes.^24,25 Porphyrin functionalized CNT with a covalent linkage is expected to be a promising candidate for the application in photovoltaic and optical limiting devices by assembling donor–acceptor system, which involves an efficient energy or electron transfer process (PET/ET).^26–29 Liu et al. reported the enhanced optical limiting performance of three porphyrin covalently functionalized SWNTs hybrids in suspension at 532 nm, the combination of nonlinear scattering with RSA, and photoinduced electron or energy transfer from electron donor porphyrin moiety to acceptor SWNT (produce a charge-separated excited state) resulted in the superior optical limiting effect.²² Wang et al. prepared two types of MWNT based nanohybrids through 1,3-dipolar cycloaddition reactions, Z-scan showed that these MWNT–porphyrin nanohybrids exhibited enhanced NLO properties under both nanosecond and picosecond pulses at 532 nm due to a remarkable accumulation effect.²³ Although the association of nonlinear scattering of carbon nanotubes with the RSA of porphyrins has been verified to be a promising strategy to design ideal NLO materials, the NLO properties of these CNT–porphyrin nanohybrids were only investigated at 532 nm in liquid suspension system with unfavorable design and low stability, which seriously limit their practical application as optical devices.^22,23,28,29 In addition, this type of nanomaterials reported in the literature were almost constructed by the direct covalent attachment of a single tetraphenylporphyrin (TPP) to the surface of nanotubes, while the donor–acceptor ensembles could also be achieved by “polymer (carrying porphyrin units) wrapping” or electrostatic interactions,^30–32 However, their photophysical properties and NLO performance have never been revealed.

From the perspective of preparing diverse carbon nanotubes–porphyrin hybrids as broadband optical limiter and improving their solubility and compatibility with matrices, in this work, we present the synthesis of a novel copolymer functionalized multi-walled carbon nanotubes (MWNT-1) and demonstrate the electron communication between porphyrin units and carbon nanotubes. Z-scan measurements demonstrated the improved NLO performance of the resultant MWNT–polymer nanohybrids compared with each individual component, even better than that of the MWNT–zinc porphyrin hybrid counterpart (MWNT-2) at both 532 nm and 1064 nm. The polymer we used to functionalize the MWNTs is a novel designed poly(arylene ether sulfone) (PAES) incorporating zinc porphyrins in the backbones (ZnTNP–PAES). Poly(arylene ether)s represent a class of high performance polymers featuring excellent thermal, mechanical, nonflammable and film-forming properties, which can be tailor-made for diverse application.^33–35 Our previous work have suggested the superior optical limiting and thermal properties of the porphyrinated poly(arylene ether)s in both solution and solid films state.^36,37 Herein, we expect that the combination of the porphyrin containing poly(arylene ether sulfone) with CNTs would afford synergetic NLO properties via the π electron interaction of the two components, particularly improve the solubility and compatibility of the nanohybrid in composite materials.

Results and discussion

The synthetic procedures for poly(arylene ether sulfone)s incorporating zinc–porphyrins in the main chain are presented in Scheme 1. The structurally symmetrical 5,15-bis(4-hydroxyphenyl)-10,20-dinaphthylporphyrin (trans-DHTNP) with large nonlinear optical coefficient, was designed and served as a bisphenol monomer to copolymerize with bis(4-fluorophenyl)sulfone and (p-amino)phenylhydroquinone (APH) using a direct route, then the polymer was dissolved in DMAc to react with excess Zn(OAc)₂·2H₂O to generate the final polymer ZnTNP–PAES. Molecular weights of the resultant ZnTNP–PAES were determined to be 26 200 g mol⁻¹ (M_n) and 61 700 g mol⁻¹ (M_w) by GPC measurements. It is worth mentioning that the ZnTNP–PAES possess a T_g above 260 °C, higher than most of poly(arylene ether)s, this observation can be ascribed to the rigid macrocyclic porphyrin units and a large number of amino hydrogen-bond interactions, which significantly constrict the segmental motion. Fig. S1† shows the ¹H NMR spectrum of ZnTNP–PAES, the signals attributed to protons of porphyrin from 6.9 to 8.8 partially overlaps with those of aromatic protons in the backbones, while the characteristic resonance assigned to the internal pyrrole NH proton at −2.80 ppm is vanished because of the complexation with Zn²⁺. MWNT-2 was synthesized by grafting a 5-(4-hydroxyphenyl)-10,15,20-trinaphthylporphyrin zinc (OH-ZnTNP) on MWNT using a esterification reaction, and the structure are presented in Fig. S2.†

Scheme 1 The polymerization of ZnTNP–PAES and the synthetic procedure of MWNT-1.

The successful grafting of ZnTNP–PAES to MWNTs is confirmed by FTIR spectroscopy. As shown in Fig. 1a, the characteristic vibration for carboxylic acid group on the MWNT surface at 1713 cm⁻¹ is observed after the oxidation. The stretching vibration of aromatic sulfone at 1154 cm⁻¹, skeleton vibration of benzene rings at 1586 cm⁻¹ and 1482 cm⁻¹, and the stretching vibration of ether linkage at 1235 cm⁻¹ appear in ZnTNP–PAES. In comparison with the featureless spectrum of MWNTs, the spectrum of MWNT-1 clearly exhibits characteristic absorption bands of ZnTNP–PAES, indicative of considerable covalent attachment on the MWNT surface. The weak absorption at 3370 cm⁻¹ is ascribed to the stretching vibration of –NH₂ of ZnTNP–PAES, which is disappeared after grafting to MWNTs. The new bands at 1632 cm⁻¹ and 1520 cm⁻¹ corresponding to the formation of amide bonds can be detected for the MWNT-1 hybrids.

Fig. 1 (a) FTIR spectra of MWNT, MWNT-COOH, MWNT-1, and ZnTNP–PAES. (b) Raman spectra of MWNT, MWNT-1, and MWNT-2.

Raman spectroscopy (Fig. 1b) corroborates the surface functionalized MWNTs with excitation at 532 nm. The spectrum of pristine MWNTs typically exhibits two intense bands at 1341 cm⁻¹ (D-band) and at 1580 cm⁻¹ (G-band), the former is usually ascribed to sp³ hybridized carbon, crystal disorder, and chemical impurities, whereas the latter is generated by tangential vibration of sp² hybridized graphitic carbon.^38–40 The pristine MWNTs display a D-band more intense than the G-band due to the presence of impurities and amorphous carbon, with the relative intensity ratio (I_D/I_G) of 1.25. Following functionalization, no distinct shift can be discovered for the hybrids in the spectra, the relative intensity ratio of D-band to G-band reduce to 1.04 and 1.18 for MWNT-1 and MWNT-2, respectively. I_D/I_G is usually exploited as an efficient approach to evaluate the functionalization and defects in carbon nanomaterials. The decreased intensity ratio is consistent with some other results for covalently functionalized MWNTs,^41,42 where the functionalization does not lead to significant increase in the sp³ hybridized carbon atoms because the reaction sites primarily locate on the outer nanotube layer of MWNTs, however, the polymers or porphyrin molecules on the surface of MWNTs nanoscale layers contribute to an increase in the absorption intensity of the sp² hybridized carbon atoms. Furthermore, a large part of amorphous carbon and/or lattice distortions was removed during the purification in the first step.²³ Given the above information, the obvious reduction in the intensity ratio of D-band to G-band further supports the surface modification of the carbon nanotube hybrids.

Thermogravimetric analysis (TGA) is employed to investigate the thermal properties of the hybrids and evaluate the wt% (weight percent) of each component. Fig. 2 presents the TGA curves of ZnTNP–PAES, OH-ZnTNP, MWNTs, and MWNT hybrids under nitrogen. The pristine MWNTs are thermally stable over a wide temperature range and only show 5.2% weight loss at 800 °C due to the impurities trapped in the nanotubes. In contrast, MWNT-COOH exhibits approximately 18.7% weight loss on account of the defects caused by the acid treatment. The TGA curve of ZnTNP–PAES shows two degradation stage, the initial weight loss at about 160 °C is associated with the decomposition of amino pendants, and the second weight loss around 450 °C corresponds to the decomposition of polymer backbones. MWNT-1 presents a similar two step degradation pattern, with far more initial weight loss than MWNT-COOH above 100 °C, which may be assigned to the thermally unstable amide bond. The temperature range (450 °C to 650 °C) for the major weight loss slightly broadens owing to their covalent attachment. Assuming that the polymer residues wt% in ZnTNP–PAES at 700 °C remain the same as that in the hybrids, the ZnTNP–PAES in MWNT-1 hybrids can be approximately calculated to be 64%. Apparently, the MWNTs hybrids we prepared are thermally more stable than the graphene oxide hybrid, which suffered nearly 30% weight loss at 400 °C and 50% weight loss at 800 °C.⁴³

Fig. 2 TGA analysis of ZnTNP–PAES, OH-ZnTNP, and MWNT hybrids.

Fig. 3 shows the high-resolution N 1s X-ray photoelectron spectroscopy (XPS) of ZnTNP–PAES and MWNT-1 hybrids, together with spectral deconvolution. The N 1s XPS spectrum of ZnTNP–PAES can be fitted into three contributors corresponding to different species, N–H at 399.54 eV, C–N at 398.29 eV, and 400.55 eV originating from porphyrins. In contrast, the introduction of the polymers onto the surface of MWNTs leads to a decreased intensity of N–H relative to that observed for ZnTNP–PAES, and the peak assigned to C–N shows a slight red shift to 398.4 eV. Moreover, a new peak centering at 400.16 eV is attributed to the nitrogen atom of the resultant amide group. Thus the XPS spectroscopy data corroborates the successful synthesis of the MWNT-1 nanohybrids.

Fig. 3 N 1s spectra of (a) ZnTNP–PAES, (b) MWNT-1.

SEM and TEM measurements provide further insight into the morphology of these MWNT hybrids. The SEM images of MWNT-1 hybrids are shown in Fig. 4, pristine MWNTs exhibit a rather clean and smooth surface with the average tube diameter of 18–20 nm. In contrast, after functionalization the MWNT-1 hybrids are apparently shortened and irregularly coated with a large amount of ZnTNP–PAES with a diameter of ∼36 nm, almost doubles that of the unmodified MWNTs. TEM images (Fig. 5) also gives a typical smooth morphology of MWNTs, which entangle with each other in chaotic orientation, whereas the MWNT-2 hybrid appears to be stained with porphyrin chromophores at the tube ends and in some areas of the surface. Compared with MWNTs and MWNT-2, the nonuniform coverage of an amorphous organic layer can be distinctly observed in the TEM images of MWNT-1 hybrids, indicating the efficient grafting on the external surfaces of the nanotubes. A fraction of MWNTs aggregates to bundles via the covalent crosslinking of polymers, and the abundant polymers on the MWNT stalks reveal the tendency to form clusters (Fig. 5e), consistent with the MWNT composites modified with polyurethane–urea.⁴⁴

Fig. 4 SEM images of MWNT (a), MWNT-1 (b and c), and MWNT-2 (d).

Fig. 5 TEM images of MWNT (a and b), MWNT-2 (c and d), and MWNT-1 (e and f).

UV-vis absorption spectrum of ZnTNP–PAES in DMF (Fig. 6) exhibits an intense Soret-band at 428 nm, and two Q bands at 560 nm and 599 nm, and shows a profile identical to the absorption spectrum of porphyrin analogue OH-ZnTNP, whereas the spectrum of MWNT is characterized by the continuously decreasing absorption from 300 to 700 nm. In the case of MWNT-1, the Soret-band shows a red shift to 431 nm and a notable broadening relative to ZnTNP–PAES, this broadening absorption usually originates from the electronic interactions between porphyrin moieties and carbon nanotubes in the ground state as previously reported.^21–23 In contrast, the absorption of MWNT-2 shows a much weaker Soret-band, suggesting the smaller amounts of porphyrin chromophores in the hybrids system. Fluorescence spectra (Fig. 7) reveals the excited state interaction of the MWNT based hybrids in DMF, upon excitation of zinc porphyrins in the Soret-band, both ZnTNP–PAES and OH-ZnTNP exhibit strong fluorescence emission at 605 nm and 655 nm due to the S₁ to S₀ transition. However, substantial fluorescence quenching are observed in the hybrids since MWNTs have been considered as favorable electron acceptor and the porphyrin moiety serves as electron donor, resulting in the photo-induced electron/energy transfer (PET/ET) from porphyrin moieties to carbon nanotubes. The solution of MWNT-1 and MWNT-2 exhibits 58% and 87% quenching of fluorescence emissions relative to those of ZnTNP–PAES and OH-TNP, respectively (the concentrations of the porphyrin moieties in all samples were equivalent as determined by the intensity of the Soret-band with the background being taken into account). Similar to MWNT-2, the presence of the efficient electron transfer process in several porphyrin–MWNTs hybrid system constructed by an ester linkage have been proven.^22,27 However, the optical properties of the MWNT hybrids with covalent functionalization of porphyrinated polymers have scarcely been investigated, the flexible structural arrangement allows the porphyrin rings in ZnTNP–PAES to take a position facing the MWNTs surface, which may facilitate the energy transfer process. In addition, it can be deduced that MWNT-2 exhibits higher electron/energy transfer efficiency than MWNT-1, as evidenced by the magnitude of fluorescence quenching.

Fig. 6 UV-vis absorption spectra of ZnTNP–PAES, MWNT, MWNT-1, and MWNT-2. Inset: photograph of aqueous dispersions of (a) MWNT, (b) MWNT-2, (c) MWNT-1, (d) ZnTNP–PAES.

Fig. 7 Steady fluorescence emission spectra of ZnTNP–PAES and MWNT-1 (a), OH-ZnTNP, and MWNT-2 (b), transient fluorescence spectra of ZnTNP–PAES and MWNT-1 (c), OH-ZnTNP, and MWNT-2 (d), with the matching absorption of the Soret-band in DMF.

To further understand the excited state interaction of the MWNT hybrids, fluorescence lifetime measurements were conducted using transient fluorescence spectroscopy with 400 nm laser source. The fluorescence decay curves for ZnTNP–PAES and OH-ZnTNP in DMF are well fitted by a biexponential decay model (see Fig. 7c and d), with the lifetimes of 2.2 ns (45%) and 9.1 ns (55%) for ZnTNP–PAES, 2.6 ns (21%) and 10.3 ns (79%) for OH-ZnTNP, respectively. Compared with the intact ZnTNP–PAES, the fluorescence lifetimes (fitted using biexponential decay kinetics) of MWNT-1 hybrids was found to significantly decrease to 830 ps (37%) and 6.4 ns (63%), which further validates the rapid fluorescence emission quenching. Similarly, the fitting result of fluorescence lifetimes of MWNT-2 can be divided into a shorter component of 540 ps (48%) and a longer component of 7.4 ns (52%). Such fluorescence quenching and shortened lifetimes were also reported in porphyrin–graphene hybrids.^45,46 Accordingly, the conclusion can be drawn that a charge separation state or energy transfer scenario exists in these covalently functionalized MWNTs hybrids under photoexcitation.

The solubility of MWNTs hybrids are demonstrated in the digital photograph in Fig. 6, black sediment could be observed in the dispersion of pristine MWNT in several hours after sonication, while MWNT-1 and MWNT-2 in DMF solution remained stable for months. Absorption spectroscopy was employed to determine the solubility of the hybrids, the absorption spectra of MWNT-1 in DMF with different concentrations (5–25 mg L⁻¹) were measured. Fig. S4† and the inset display the plots of the absorption intensity at 431 nm versus the mass concentration. On the basis of Beer's law, the extinction coefficient of MWNT-1 is calculated to be 0.015 L mg⁻¹ cm⁻¹, with a correlation coefficient R of 0.998. Additionally, the absorbance at other wavelengths is also in accordance with Beer's law following a linear relationship, implicating that a stable and homogeneous suspension of MWNT-1 hybrid can be formed in solvents.

It has been demonstrated that the accumulation of different nonlinear mechanism by functionalizing carbon nanomaterials with nonlinear absorption chromophores or polymers will achieve a further synergistic effect.^10–18 Hence we performed Z-scan measurements to investigate the NLO properties of the MWNT hybrids at both 532 and 1064 nm with 6 ns laser pulse. The pulse energy was 150 μJ for 532 nm and 1800 μJ for 1064 nm, specific details are described in experimental section. MWNT dispersion was taken as a reference for comparison, and all samples had been set to have the same linear transmittance (LT) of 65% at 532 nm by adjusting the concentration. The β_eff was deduced using a curve fitting theory based on an intensity-dependent extinction coefficient from the Z-scan data.⁴⁷ As can be seen in Fig. 8a, the normalized transmittance and the corresponding scattering responses are plotted as functions of Z position, MWNT-1 performs much stronger NLO effect than the pristine MWNT and MWNT-2 on account of the effective combination of reverse saturable absorption (RSA) of porphyrin moieties and the nonlinear scattering (NLS) effect of CNTs.^21–23 The multiple mechanism can be reflected in Fig. S5 and S6.† Generally, the value of effective nonlinear extinction coefficient β_eff will decrease as the input fluence increases for RSA process due to the saturation of RSA, and it will keep invariable for two-photon absorption (TPA) process.⁴⁸ However, the β_eff value increases with the increasing pulse energy, implying the predominant role of NLS effect.

Fig. 8 Open-aperture Z-scan curves with normalized transmittance (open symbols) and scattering signal (solid symbols) for DMF dispersions of MWNT, MWNT-1, MWNT-2, and ZnTNP–PAES at (a) 532 nm and (b) 1064 nm. The solid line represents the theoretic fit.

Simultaneously, as confirmed by the fluorescence spectroscopy, a possible intramolecular photo-induced electron or energy transfer process also contributes to the enhanced NLO performance. For 532 nm laser pulse, regardless of the lowest normalized transmittance of MWNT-1 at focus, ZnTNP–PAES shows faster nonlinear optical response than MWNTs and the hybrids due to RSA at a lower input intensity, and the obtained β_eff value of ZnTNP–PAES is the largest of all the samples under the same condition (Table 1). However, it is clearly noted in Fig. 8b that ZnTNP–PAES shows no NLO signals at 1064 nm in spite of the very high pulse energy, while the curve of MWNT dispersion exhibits two symmetrical peaks on both sides of a valley at the focus, indicating a transition from saturable absorption (SA) to NLS as the energy density increases along the Z direction, and no significant enhancement can be found in the porphyrin functionalized MWNT-2 sample at 1064 nm. To our surprise, the open Z-scan curve of MWNT-1 hybrid exhibits a sharply decreased transmittance, accompanied by the much stronger scattering signals relative to those of MWNT and its counterpart MWNT-2. The effective β_eff value is about 5 times larger than that of MWNT dispersions by attaching the porphyrin polymers, which has no NLO performance at 1064 nm under our measurement conditions. These improvements suggest that the PET/ET behavior has a substantial impact on their NLO properties.

Table 1 NLO coefficients of MWNT hybrids and ZnTNP–PAES in DMF at 532 and 1064 nm

Sample	λ (nm)	I0 (μJ)	T (%)	βeff (cm GW^−1)
MWNT	532	150	65%	0.752
MWNT-1	532	150	65%	3.374
MWNT-2	532	150	65%	1.455
ZnTNP–PAES	532	150	65%	5.282
MWNT	1064	1800	70%	0.054
MWNT-1	1064	1800	71%	0.389
MWNT-2	1064	1800	73%	0.064
ZnTNP–PAES	1064	1800	90%	—

---

It has been proposed that the nonlinear scattering response of CNTs essentially shares the same process with that of carbon black suspension. The indispensable scattering centers arise from the growth of solvent microbubbles, induced by the thermal energy transfer from CNTs to the surrounding liquid. Another origin of the scattering centers is the thermal induced ionization of CNTs, followed by the formation of microplasma as scattering centers, which rapidly expand and strongly scatter light from the beam direction.^9,49 The former (solvent microbubbles) occurs at lower incident fluence, while the latter (CNTs ionization) occurs at higher incident fluence.^6,7 Generally, the centers can effectively scatter the incident beam when the size of centers grow to the magnitude of incident light wavelength. Therefore, in most cases the NLO performance of CNTs and CNT hybrids at 1064 nm are weaker than those at 532 nm, as the microbubbles and/or microplasma are demanded to expand to a larger size. From the nonlinear optical data listed in Table 1, it can be inferred that ZnTNP–PAES plays an important role on the enhanced NLO properties in the following aspects. First of all, the presence of ZnTNP–PAES substantially improves the solubility of MWNTs by forming a stable amide bond, leading to the uniform and homogenous dispersion. Secondly, the zinc porphyrins incorporated in poly(arylene ether sulfone)s preserve its intrinsic photophysical characteristics, hence the ZnTNP–PAES provide superior RSA behavior at 532 nm. Moreover, PET/ET process between zinc porphyrins and carbon nanotubes would further enhance the NLO properties, this synergic effect are corroborated by MWNT-1 and MWNT-2 at both wavelength in our system. It should be noted that larger magnitude of fluorescence quenching and faster lifetime decay for MWNT-2 hybrid has been detected (Fig. 7), implying the closer interaction between the two components and higher PET/ET efficiency in MWNT-2. However, it is worth mentioning that the contribution from PET/ET process may not be evaluated by simply comparing the quenching magnitude. In the case of MWNT-1, as a result of the covalent attachment and strong π–π interaction, the large amount of macrocyclic porphyrin units (reflected by the Soret-band absorption in UV-vis spectra) and benzene structure in the backbone facilitate ZnTNP–PAES to compactly wrap on the surface of MWNT, providing opportunities for abundant porphyrin units to involve in the PET/ET process from donor to acceptor. Therefore, one of the possible reason we proposed for the much more prominent NLO performance of MWNT-1, particularly at 1064 nm, is the larger amount of porphyrins involved in PET/ET process from donor to acceptor. Similar observation were also found by axially coordinated metal–porphyrins functionalized MWNTs.⁵⁰

The optical limiting responses of the samples in DMF are presented in Fig. 9, in which the normalized transmittance and scattering signals are plotted as a function of input fluence. The solvent exhibits no detectable OL performance. The OL curves reflect a similar trend with Z-scan measurement, that is, the transmittance decrease while the scattering signals increase strikingly with the input fluence, and MWNT-1 displays the most remarkable broadband limiting effects. The enhanced OL behavior and NLO performance were also found in graphene based hybrids.^51–55 Although it has been manifested that the combination of RSA and NLS, and the PET/ET process are responsible for the improved OL response, the exact interior mechanism of PET/ET and how this process improved the nonlinear optical effect are still obscure.^11–29 It is known that the nonlinear scattering stems from MWNT, and no scattering response is observed in ZnTNP–PAES. At the same linear transmittance of 65%, MWNT-1 hybrid samples should have a lower mass fraction of MWNT relative to pristine MWNT dispersion. Coincidently, in this study, MWNT-1 shows the largest scattering signals and β_eff value. Thus it may be the photo-induced energy transfer from porphyrins to carbon nanotubes that plays a decisive part by enhancing the nonlinear scattering efficiency at the near infrared region. Similar hypothesis was also proposed in a fluorene, thiophene, and benzothiadazole based conjugated polymer functionalized graphene sheet system.⁵⁶

Fig. 9 Optical limiting plots of normalized transmittance (solid symbols) and scattering response (open symbols) versus the incident pulse energy density for the samples at (a) 532 nm and (b) 1064 nm.

The excellent solubility of MWNT hybrids in solvents makes it highly feasible for fabrication of uniform films, therefore we prepared the MWNT-1 doped fluorinated poly(arylene ether sulfone)s (F-PAES) film with high transparency and homogeneity by casting the mixed solution onto a glass substrate. Unfortunately, the film could not stand the high power energy of nanosecond laser pulse, but they are stable under the irradiation of 340 fs pulse laser at 515 nm with the repetition of 100 Hz. Fig. S7† demonstrates the OL behavior of MWNT-1 both in solid F-PAES film and DMF (adjusted to have the same linear transmittance with the film sample of 72%), the normalized transmittance decreased to 0.88 and 0.92 for solution and film state, respectively. As a control sample, the neat F-PAES film with a high transparency of 94% was also measured under the same experiment, which gave negligible OL response as presented in Fig. S7.† Considering the femtosecond time scale and solid film state, NLS effect can be ruled out in the current measurement, the OL response might arise from two-photon absorption (TPA) process.

Conclusion

A novel poly(arylene ether sulfone) incorporating zinc porphyrins was designed and used to functionalize MWNTs. The obtained MWNT nanohybrid material was thoroughly characterized by FTIR, Raman, XPS, TGA, SEM, TEM, UV-vis absorption, and fluorescence spectroscopy, all the results clearly suggested that the ZnTNP–PAES was covalently grafted to MWNTs. Significant fluorescence quenching was observed for both ZnTNP–PAES and OH-ZnTNP modified MWTN hybrids, as a result of photo-induced electron or energy transfer process. The attachment of ZnTNP–PAES onto MWNTs greatly improved the solubility in organic solvents and the compatibility in polymer matrix. Open aperture Z-scan results revealed the considerably enhanced broadband optical nonlinearity and optical limiting performance of the ZnTNP–PAES functionalized MWNTs, due to the accumulation effect of NLS, RSA, and PET/ET from the excited porphyrin moieties to MWNTs. Furthermore, the nanohybrid embedded in polymer thin films exhibited good optical limiting properties for 340 fs pulse laser, which is expected to be further optimized for the composites fabrication of laser protection coatings and optoelectronic devices.

Experimental section

Materials

MWNTs with an average diameter of 10–20 nm and length of 10–30 μm were purchased from Chengdu Organic Chemicals Co. Ltd., Chinese Academy of Science. Anhydrous THF was distilled from sodium, N,N-dimethylformamide (DMF) and trimethylamine were dried and distilled from CaH₂ prior to use. Bis(4-fluorophenyl)sulfone (99%) was provided by TCI Shanghai development Co. Ltd., China. Other chemicals and reagents were used as received without further purification.

Synthesis of ZnTNP–PAES

Bisphenol monomer 5,15-bis(4-hydroxyphenyl)-10,20-dinaphthylporphyrin (trans-DHTNP) was synthesized by Alder's method.⁵⁷ 1-Napthaldehyde (7.809 g, 50 mmol), 4-hydroxybenzaldehyde (6.106 g, 50 mmol) were dissolved in 250 mL propionic acid, pyrrole (6.8 mL, 100 mmol) was added dropwise in 20 min with stirring at 130 °C. The mixture was refluxed for 40 min, most of propionic acid was distilled under reduced pressure. The black precipitate was filtered and washed with methane, and the crude product was purified by column chromatography (CHCl₃ : methane, 50 : 1, v : v as eluent). The third fraction was collected and evaporated to give bright purple crystalline 5,15-bis(4-hydroxyphenyl)-10,20-dinaphthylporphyrin (yield: 5.6%). Anal. calcd for (C₅₂H₃₄N₄O₂): C 83.63, H 4.59, N 7.50. Found: C 83.21, H 4.70, N 7.22%. ¹H NMR (500 MHz, DMSO-d₆, TMS), δ (ppm): 9.96 (s, 2H, OH), 8.9 (s, 2H, pyrrolic-β-CH). 8.79–8.82 (d, 2H, pyrrolic-β-CH), 8.51–8.54 (m, 2H, naphthyl CH), 8.46–8.50 (m, 4H, pyrrolic-β-CH), 8.20–8.32 (m, 4H, naphthyl CH), 8.01–8.03 (d, 4H, J = 7.1 Hz, ortho-C₆H₄OH), 7.48–7.53, 7.88–7.94 (m, 4H, naphthyl CH), 7.18–7.21 (d, 4H, J = 8.7 Hz, meta-C₆H₄OH), 6.92–7.15 (m, 4H, naphthyl CH), −2.58 (s, 2H, NH). MALDI-TOF MS: calculated for (C₅₂H₃₄N₄O₂): m/z [M + H]⁺ = 747.27; found: m/z = 748.5.

(p-Amino)phenylhydroquinone (APH) was synthesized following the procedure in our previous work.⁵⁸ Mp: 174 °C, m/z = 202. The synthetic procedure of ZnTNP–PAES was illustrated as Scheme 1, bis(4-fluorophenyl)sulfone (2.540 g, 10 mmol), trans-DHTNP (2.239 g, 3 mmol), APH (1.407 g, 7 mmol), and anhydrous K₂CO₃ (1.52 g, 11 mmol) were dissolved in tetramethylene sulfone (TMS 21 mL) and toluene (12 mL) in a three-necked flask under mechanical stirring in a nitrogen atmosphere. The mixture was refluxed at 130 °C to remove the resulting water through a Dean–Stark trap, then the reaction was heated to 180 °C for 6 h to complete the polymerization. The viscos solution was poured into deionized water to precipitate the dark purple polymer, which was washed with boiled water and ethanol to remove the solvent and monomer residues, the product was dried at 100 °C for 24 h. Using a similar method according to previous literature,⁵⁹ the precursor polymer (2.0 g) was dissolved in N,N-dimethylacetamide (DMAc) and stirred with excess Zn(OAc)₂·2H₂O at 80 °C for 4 h. Finally, the product was precipitated into deionized water, washed with water three times to remove the excess acetate, and isolated with filtration prior to being dried to give the final product ZnTNP–PAES (purple solid). Yield: 90%, M_n: 26.2 kg mol⁻¹, M_w: 61.7 kg mol⁻¹, PDI: 2.35, T_g (DSC): 260.1 °C.

Synthesis of ZnTNP–PAES functionalized MWNTs nanohybrid (MWNT-1)

The carboxyl-functionalized MWNTs (MWNT-COOH) were prepared according to the reported method.⁶⁰ 2 g MWNTs were purified and oxidized using a mixture (3 : 1) of H₂SO₄ (98%) and HNO₃ (60%) under sonication at 40 °C and then refluxed for 2 h, the mixture was diluted with deionized water, washed and filtered to obtain MWNT-COOH. 150 mg MWNT-COOH sample was treated with a large excess of thionyl chloride containing a catalytic amount of DMF at 70 °C for 24 h. After the removal of thionyl chloride under reduced pressure, the activated MWNT-COCl was washed with anhydrous tetrahydrofuran and filtered. To the suspension of MWNT-COCl in anhydrous DMF (50 mL) was added ZnTNP–PAES (400 mg) and trimethylamine (5 mL) under purified argon, the reaction was stirred at 100 °C for 72 h. The functionalized MWNTs were isolated by washing off the unreacted ZnTNP–PAES with excess DMF, and vacuum-filtered through a single layer nylon membrane (0.22 μm) several times until the filtrate being colorless, then a large amount of water was added to dissolve the generated trimethylamine salts during the reaction. Finally, the MWNT-1 hybrids was collected by filtration, and dried in vacuum at 80 °C for 24 h to give black solid (245 mg).

Synthesis of ZnTNP functionalized nanohybrid (MWNT-2)

500 mg 5-(4-hydroxyphenyl)-10,15,20-trinaphthylporphyrin zinc (OH-ZnTNP) was prepared according to our previous report.³⁷ OH-ZnTNP and 5 mL trimethylamine were added to 50 mL anhydrous DMF suspension of MWNT-COCl (200 mg). The reaction was continued at 100 °C for 72 h under the protection of pure argon. After cooling to room temperature, the suspension was filtered and washed extensively with DMF. Sonication and redispersion were repeated in chloroform to remove the absorbed porphyrins, then the hybrid was dispersed in water to remove the trimethylamine salts. The final product was collected by filtration and dried at 80 °C for 24 h to give MWNT-2 hybrids as black solid (206 mg).

Preparation of MWNT-1 hybrid/F-PAES film

Initially, MWNT-1 hybrid (3 mg) was dispersed in 2 mL DMAc under sonication, then the MWNT-1 dispersion was added to 8 mL DMAc solution of F-PAES (0.1 g mL⁻¹, see structure in Fig. S6†), after vigorous stirring for 12 h, the mixed solution was drop-cast on a glass substrate (10 cm × 10 cm), followed by heating at 110 °C in vacuum for 48 h.

Characterization

IR spectra were recorded on a Nicolet Impact 410 Fourier transform infrared spectrophotometer in transmittance. Mass spectra (MS) were measured on an AXIMA-CFR laser desorption ionization flying time spectrometer (COMPACT). Raman spectra were obtained at room temperature on a LabRAM HR Evolution (Horiba) Raman microscope, excited at 532 nm with Arion laser radiation. ¹H NMR were recorded on a Bruker 510 NMR spectrometer (500 MHz) in DMSO-d₆ with tetramethylsilane as a reference. Gel permeation chromatograms using polystyrene as a standard were performed on a high temperature chromatograph PL-GPC 220 with DMF as an eluent at a flow rate of 1.00 mL min⁻¹ at 80 °C. Differential scanning calorimetry (DSC) measurements were performed on a Mettler Toledo DSC 821e instruments at a heating rate of 20 °C min⁻¹ under a nitrogen atmosphere. Thermogravimetric analysis (TGA) was performed on a Netzch Sta 449 thermal analyser system at a heating rate of 10 °C min⁻¹ in nitrogen. Transmission electron microscopic (TEM) images were taken on a JEM-1200EX electron microscopic operated at an acceleration voltage of 100 kV. Scanning electron microscopic (SEM) images were collected on a FEI Nova NanoSEM 450 microscope. DMF solution of the MWNT hybrid materials were dropped onto the grids and silicon wafer and dried for TEM and SEM image, respectively. X-ray photoelectron spectra were recorded on PHI-1600 X-ray photoelectron spectrometer using Mg Kα (1253.6 eV) radiation as the radiation source. UV-vis absorption spectra and fluorescence emission spectra were recorded on UV2501-PC spectrophotometer and Shimadzu RF-3501 fluorescence spectrophotometer, respectively. Fluorescence lifetime measurements were performed on a FLS 920 steady-transient spectrophotometer with a time-correlated single-photon counting system using 400 nm laser source.

The nonlinear optical properties were measured using an open aperture Z-scan technique employing a Q-switched Nd:YAG laser of 6 ns pulses at a wavelength of 532 nm and 1064 nm with a repetition of 10 Hz. The laser beam was tightly focused with a 15 cm focus lens, and all samples were tested in 5 mm × 10 mm quartz cuvettes and moved along the axis of the incident beam (z direction). Meanwhile, another focusing lens at 45° to the incident beam was setup to collect the scattering signal. Three high-precision photo-detectors were used to monitor the reference, transmitted and scattering light, respectively. All samples were dissolved or dispersed in DMF, and were adjusted to have a linear transmittance of 65% at 532 nm. Optical limiting properties of the polymer films were investigated by I-scan system, which was performed on a mode-locked fiber laser using 340 fs pulses at 515 nm with the repetition rate of 100 Hz, the laser spot diameter was ∼20 μm. The optical arrangement is identical to the literature.⁶¹ Two high-accuracy photoelectric detectors were used to record the referenced and transmitted pulse energy, and an attenuator driven by a fine linear translation stage was employed to change the incident intensity (I₀), thus the dependence of transmission on I₀ can be obtained.

Acknowledgements

This work is financially supported by the Ph.D. Program Foundation of the Ministry of Education of China (No. 20120061110017), Jilin Science and Technology Development Plan Program of China (20150204001GX), and National Natural Science Foundation of China (No. 61308087 and 61522510). J. W. thanks the financial support from the External Cooperation Program of BIC, CAS (No. 181231KYSB20130007).

References

1. P. N. Prasad, Nanophotonics, WileyBlackwell, 2004.
2. P. M. Ajayyan, L. S. Schandler, C. Giannaris and A. Rubio, Adv. Mater., 2000, 12, 750–753.
3. P. Avouries, Acc. Chem. Res., 2002, 35, 1026–1034.
4. Y. Lin, S. Taylor, H. Li, K. A. S. Fernando, L. Qu, W. Wang, L. Gu, B. Zhou and Y. P. Sun, J. Mater. Chem., 2004, 14, 527–541.
5. L. W. Tutt and T. F. Boggess, Prog. Quantum Electron., 1993, 17, 299–338.
6. L. Vivien, P. Lancon, D. Riehl, F. Hache and E. Anglaret, Carbon, 2002, 40, 1789–1797.
7. J. E. Riggs, D. B. Walker, D. L. Carroll and Y. P. Sun, J. Phys. Chem. B, 2000, 104, 7071–7076.
8. P. Chen, X. Wu, X. Sun, J. Lin, W. Ji and K. L. Tan, Phys. Rev. Lett., 1999, 82, 2548–2551.
9. J. Wang, Y. Chen and W. J. Blau, J. Mater. Chem., 2009, 19, 7425–7443.
10. N. Mackiewicz, T. Bark, B. Cao, J. A. Delaire, D. Riehl, W. L. Ling, S. Foillard and E. Doris, Carbon, 2011, 49, 3998–4003.
11. N. He, Y. Chen, J. R. Bai, J. Wang, W. J. Blau and J. H. Zhu, J. Phys. Chem. C, 2009, 113, 13029–13035.
12. Z. Guo, F. Du, D. M. Ren, Y. S. Chen, J. Y. Zheng, Z. B. Liu and J. G. Tian, J. Mater. Chem., 2006, 16, 3021–3030.
13. K. Sanusi, E. K. Amuhaya and T. Nyokong, J. Phys. Chem. C, 2014, 118, 7057–7069.
14. L. J. Niu, P. P. Li, Y. Chen, J. Wang, J. J. Zhang, B. Zhang and W. J. Blau, J. Polym. Sci., Part A: Polym. Chem., 2011, 49, 101–109.
15. B. Zhang, J. Wang, Y. Chen, D. Fruchtl, B. Yu, X. D. Zhuang, N. He and W. J. Blau, J. Polym. Sci., Part A: Polym. Chem., 2010, 48, 3161–3168.
16. L. M. Zhang, S. D. Allen, C. Woelfle and F. J. Zhang, J. Phys. Chem. C, 2009, 113, 13979–13984.
17. H. X. Wu, D. D. Liu, H. Q. Zhang, C. Y. Wei, B. Zeng, J. L. Shi and S. P. Yang, Carbon, 2012, 50, 4847–4855.
18. B. Anand, S. A. Ntim, V. S. Muthukumar, S. S. S. Sai, R. Philip and S. Mitra, Carbon, 2011, 49, 4767–4773.
19. Y. Zhu, H. I. Elim, Y. L. Foo, T. Yu, Y. Liu, W. Ji, J. Y. Lee, Z. Shen, A. T. S. Wee, J. T. L. Thong and C. H. Sow, Adv. Mater., 2006, 18, 587–592.
20. D. Eder, Chem. Rev., 2010, 110, 1348–1385.
21. Y. H. Chen, V. S. Muthukumar, Y. B. Wang, C. Li, S. S. Krishnan, S. S. S. Sai, K. Venkataramaniah and S. Mitra, J. Mater. Chem., 2009, 19, 6568–6572.
22. Z. B. Liu, J. G. Tian, Z. Guo, D. M. Ren, F. Du, J. Y. Zheng and Y. S. Chen, Adv. Mater., 2008, 20, 511–515.
23. A. J. Wang, Y. Feng, L. L. Long, Y. L. Song, W. Yu, W. Zhao, M. P. Cifuentes, M. G. Humphrey and C. Zhang, Chem.–Eur. J., 2013, 19, 14159–14170.
24. M. O. Senge, M. Fazekas, E. G. A. Notaras, W. J. Blau, M. Zawadzka, O. B. Locos and E. M. N. Mhuircheartaigh, Adv. Mater., 2007, 19, 2737–2774.
25. M. Calvete, G. Y. Yang and M. Hanack, Synth. Met., 2004, 141, 231–243.
26. H. P. Li, R. B. Martin, B. A. Harruff, R. A. Carino, L. F. Allard and Y. P. Sun, Adv. Mater., 2004, 16, 896–900.
27. D. Baskaran, J. W. Mays, X. P. Zhang and M. S. Bratcher, J. Am. Chem. Soc., 2005, 127, 6916–6917.
28. Z. B. Liu, Z. Guo, X. L. Zhang, J. Y. Zheng and J. G. Tian, Carbon, 2013, 51, 419–426.
29. A. J. Wang, Y. Fang, W. Yu, L. L. Long, Y. L. Song, W. Zhao, M. P. Cifuentes, M. G. Humphrey and C. Zhang, Chem.–Asian J., 2014, 9, 639–648.
30. D. M. Guldi, H. Taieb, G. M. A. Rahman, N. Tagmatarchis and M. Prato, Adv. Mater., 2005, 17, 871–875.
31. K. G. Qu, H. X. Xu, C. Zhao, J. S. Ren and X. G. Qu, RSC Adv., 2011, 1, 632–639.
32. H. Ismail, B. Tiphaine, C. Renaud, M. Adina, F. Arianna, L. Jocelyne, D. Vincent, J. Bruno and C. Stephane, J. Am. Chem. Soc., 2014, 136, 6348–6354.
33. W. J. Cantwell and J. Morton, Composites, 1991, 22, 347–362.
34. G. Maier, Prog. Polym. Sci., 2001, 26, 3–65.
35. T. J. Peckham and S. Holdcroft, Adv. Mater., 2010, 22, 4667–4690.
36. Y. L. Du, S. L. Zhang, X. Jiang, K. Zhu, Z. Geng, Y. Fang, P. F. Huo, C. Liu, Y. L. Song and G. B. Wang, J. Polym. Sci., Part A: Polym. Chem., 2014, 52, 1282–1290.
37. Y. L. Du, K. Zhu, Y. Fang, S. L. Zhang, X. R. Zhang, Y. N. Lu, Y. C. Yang, Y. L. Song and G. B. Wang, RSC Adv., 2015, 5, 48311–48322.
38. B. Ballesteros, G. de la Torre, C. Ehli, G. M. A. Rahman, F. Agullo-Rueda, D. M. Guldi and T. Torres, J. Am. Chem. Soc., 2007, 129, 5061–5068.
39. A. C. Ferrari and J. Robertson, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61, 14095–14107.
40. L. G. Cancado, K. Takai, T. Enoki, M. Endo, Y. A. Kim, H. Mizusaki, N. L. Speziali, A. Joric and M. A. Pimenta, Carbon, 2008, 46, 272–275.
41. H. L. Wang, J. Liu and D. J. Qian, Synth. Met., 2012, 162, 881–887.
42. V. Georgakilas, A. Bourlinos, D. Gournis, T. Tsoufis, C. Trapalis, A. Mateo-Alonso and M. Prato, J. Am. Chem. Soc., 2008, 130, 8733–8740.
43. H. Xu, P. Wu, C. Liao, C. Lv and Z. Gu, Chem. Commun., 2014, 50, 8951–8954.
44. M. Xu, T. Zhang, B. Gu, J. L. Wu and Q. Chen, Macromolecules, 2006, 39, 3540–3545.
45. M. B. M. Krishna, N. Venkatramaiah, R. Venkatesan and D. N. Rao, J. Mater. Chem., 2012, 22, 3059–3068.
46. X. Y. Zhang, L. L. Hou, A. Cnossen, A. C. Coleman, O. Ivashenko, P. Rudolf, B. J. V. Wees, W. R. Browne and B. L. Feringa, Chem.–Eur. J., 2011, 17, 8957–8964.
47. M. Sheik-Bahae, A. A. Said, T. H. Wei, D. J. Hagan and E. W. Van Stryland, IEEE J. Quantum Electron., 1990, 26, 760–769.
48. A. Auger, W. J. Blau, P. M. Burnhamk, I. Chambrier, M. J. Cook, B. Isare, F. Nekelsona and S. M. O'Flaherty, J. Mater. Chem., 2003, 13, 1042–1047.
49. J. Wang and W. J. Blau, J. Phys. Chem. C, 2008, 112, 2298–2303.
50. A. J. Wang, J. B. Song, Z. P. Huang, Y. L. Song, W. Yu, H. L. Dong, W. P. Hu, W. P. Cifuentes, M. G. Humphrey, L. Zhang, J. D. Shao and C. Zhang, Nano Res., 2016, 9, 458–472.
51. Y. Xu, Z. Liu, X. Zhang, Y. Wang, J. Tian, Y. Huang, Y. Ma, X. Zhang and Y. Chen, Adv. Mater., 2009, 21, 1275–1279.
52. Z. Liu, Y. Xu, X. Zhang, X. Zhang, Y. Chen and J. Tian, J. Phys. Chem. B, 2009, 113, 9681–9686.
53. Z. Liu, Y. Wang, X. Zhang, Y. Xu, Y. Chen and J. Tian, Appl. Phys. Lett., 2009, 94, 021902.
54. X. Zhang, X. Zhao, Z. Liu, S. Shi, W. Zhou, J. Tian, Y. Xu and Y. Chen, J. Opt., 2011, 13, 075202.
55. X. Zhang, Z. Liu, X. Li, Q. Ma, X. Chen, J. Tian, Y. Xu and Y. Chen, Opt. Express, 2013, 21, 7511–7520.
56. A. Midya, V. Mamidala, J. X. Yang, P. K. L. Ang, Z. K. Chen, W. Ji and K. P. Loh, Small, 2010, 6, 2292–2300.
57. A. D. Alder, F. R. Longo, J. D. Finarelli, J. Goldmacher, J. Assour and L. Korakoff, J. Org. Chem., 1967, 32, 476.
58. B. J. Liu, W. Hu, C. H. Chen, Z. H. Jiang, W. J. Zhang, Z. W. Wu and T. Matsumoto, Polymer, 2004, 45, 3241–3247.
59. L. J. Twyman, A. Ellis and P. J. Gittins, Macromolecules, 2011, 44, 6365–6369.
60. J. Chen, A. H. Mark, H. Hu, Y. S. Chen, A. M. Rao, C. E. Peter and C. H. Robert, Science, 1998, 282, 95–98.
61. Y. X. Li, N. N. Dong, S. F. Zhang, X. Y. Zhang, Y. Y. Feng, K. P. Wang, L. Zhang and J. Wang, Laser Photonics Rev., 2015, 9, 427–434.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra17317a

---