A. UV-vis absorption spectrum

Typical UV-vis spectra of phthalocyanine feature two distinct absorption bands, namely Q-band and B-band. The Q-band corresponds to the π–π* transition of the monomer from the HOMO to the LUMO of the Pc 2-ring.^{Reference Snow and Jarvis27} The B-band belongs to the typical adsorption of MPc arising from the deeper levels to LUMO transition.^{Reference Chen, Zhong, Zhang, Li, Niu, Bin and Zhang28} As expected, aggregation is a common phenomenon for phthalocyanine compounds, which can be reflected from the shapes of their Q bands: monomer absorption (∼665 nm) and aggregate absorption (∼625 nm).^{Reference Zhan, Yang, Meng, Wei, Zhao and Liu29–Reference Luo, Gao, Cheng, Sun, Du, Xu and Wang32} Here, the aggregations tendency of the SHCuPc and normal HCuPc were significantly different from their synthesis method (Fig. 1). The dilute solution of the normal HCuPc showed a strong sharp peak at 650–700 nm, which was attributed to the copper Pc moiety; the shoulder peak appeared around 600–650 nm implied lower aggregate tendency. For the SHCuPc from the solvothermal synthetic route, the Q-band broadened and blue-shifted, and a new broad peak appeared at 500–550 nm, which aroused from interface aggregation of the Pc. The soret band at 250–350 nm corresponds to a metal-to-ligand charge transfer transition and is unaffected by aggregates, the shape of the split soret band of the SHCuPc showed less sharper and more broader compared with the normal HCuPc, indicating that highly ordered stacking form into spherical morphology happened during the solvothermal process.

FIG. 1.

UV-vis absorption spectrum of the synthesized SHCuPc and HCuPc in DMAc. (The concentration of the SHCuPc and HCuPc in DMAc was both 5 mg/L.)

B. FTIR spectrum

The FTIR spectrum of the SHCuPc was shown in Fig. 2, with absorption peaks at 1595, 1363, 1183, 948, 877, and 747 cm⁻¹, which were assigned to the phthalocyanine skeletal vibration. Absorption peaks observed around 1476, 1095, and 675 cm⁻¹ were assigned to the aromatic phenyl ring, C–N stretch vibrations, C–H in-plane bending vibrations, and C–H out-of-plane bending vibrations, respectively. No peak at 2230 cm⁻¹ was observed.^{Reference Gao, Cheng, Zhou, Li, Xu, Du and Zhang19} But for the IR spectrum of the HCuPc, the characteristic absorption at 2230 cm⁻¹ associating with the typical stretch vibrations of the C≡N in 1,2-bis(3,4-dicyanophenoxy)benzene showed that the reaction between the monomer and CuCl was not complete. This is also proved by the thermal analysis. From the DSC curves (Fig. 3), no obvious melt point appeared in the curve of the SHCuPc, otherwise, for the DSC curve of HCuPc, a melt point at 166 °C gave the evidence of 1,2-bis(3,4-dicyanophenoxy)benzene still existing, the other obvious exothermic peak at 301 °C, which was caused by the thermal cross-linking reaction of phthalonitrile, also confirmed the unreacted 1,2-bis(3,4-dicyanophenoxy)benzene. So from the FTIR and DSC data, it could reveal that, under certain temperature and pressure, the CN function group on the monomer 1,2-bis(3,4-dicyanophenoxy)benzene could react with CuCl more efficiently and completely via the ethylene glycol solvothermal synthetic route.

FIG. 2.

FTIR spectrum of SHCuPc and HCuPc.

FIG. 3.

DSC curves of the SHCuPc and HCuPc.

C. Morphology and structure of the hyperbranched CuPc microspheres

The morphologies of the as-synthesized hyperbranched CuPc microspheres under solvothermal treatment at 160 °C for 24 h were examined by SEM. Figure 4(a) was a low-magnification SEM image of the as-prepared product, from which numerous uniformly sized spheres with an average diameter of 500 nm could be clearly observed. These spheres were highly dispersed; moreover, no other morphologies could be detected, indicating a high yield of the product with the spherical morphology. The high-magnification SEM images of the individual spheres were also shown in Fig. 4(b), and little adhesion between the spheres was observed. This adhesion might happen during the cooperation of oriented aggregation and Ostwald ripening. Although oriented aggregation and Ostwald ripening were usually considered as two independent and competitive mechanisms, in this work, we assumed that the phthalocyanine macrocycle was rigid, the placement of rigid H-bonding moieties and their relative orientation allowed the design of a variety of specific architectures. Intramolecular H-bonding of the backbone helped to pre-organize the H-type aggregates in the macromolecules, and the H-bonding between the solvent EG and phthalocyanine helped to produce the spherical morphology. For the sample HCuPc, from Ref. Reference Guo, Yan, Kwon, Hayakawa, Kakimoto and Goodson26, without the solvothermal treatment, amorphous morphology shown in Fig. 4(c), the different morphologies of the SHCuPc and HCuPc lead to the different EM properties and discussed later.

FIG. 4.

(a) Low-magnification SEM image of the SHCuPC. (b) High-magnification SEM image of the SHCuPc. (c) Low-magnification SEM image of the HCuPC.

D. XRD analysis

It is well-known that CuPc molecules have a planar structure and stack easily to form molecular columns. The internal structures of self-organized SHCuPc and HCuPc were further investigated by the XRD technique. As can be seen from Fig. 5, for the normal HCuPc, a diffraction peak at 2θ = 9.5° was obtained and its d spacing was d = 0.923 nm, which corresponds to the β (001) plane,^{Reference Fox, Katz, Elshocht, Verbiest, Kauranen, Persoons, Thongpanchang, Krauss and Brus33,Reference Debe and Kan34} indicating that the b-axis of the CuPc molecules stacking structures. On the other hand, the diffraction peaks at 15.34 and 27.16 were also obtained. Their d spacings were d = 0.6669 and 0.4510, respectively. This indicates that the b-axis of the CuPc molecules was stacked in one-dimensional columnar aggregates. Compared with the HCuPc, SHCuPc showed an amorphous crystal structure, a highly dispersed peak at 2θ = 6.5° was obtained, and other diffraction peaks at 19.239, 26.538, 26.867, and 29.698 were obtained. Their d spacings were d = 0.6669, 0.5794, 0.4510, and 0.3331, respectively. It is understandable that van der Waals forces dominate the interaction between the molecules themselves. To minimize the intra-layer van der Waals potential, the CuPc molecules favor a close-packed configuration in the horizontal direction.^{Reference Guo, Yoshimoto and Itaya20,Reference Zhao, Jia, Wei, Pu and Liu22} At the same time, the face-to-face stacking of the molecules leads to the formation of molecular columns and the unaltered molecular orientation parallel to special morphology during growth.

FIG. 5.

(a) XRD patterns of the SHCuPc and HCuPc and (b) small angle XRD patterns of the SHCuPc and HCuPc.

E. Microwave EM and microwave absorbing properties

The microwave EM properties and microwave absorption behavior of SHCuPc/paraffin composites and HCuPc/paraffin composites were investigated in a frequency range of 2–18 GHz (Figs. 6 and 7). As can be seen from the EM data, two kinds of the CuPc from different synthetic methods gave similar peak forms not only for permittivity but also for permeability. Figure 7 shows the real part (ε′) and imaginary part (ε″) of the relative complex permittivity of the SHCuPc and HCuPc wax composites, where the ε′ values of the two samples retained an approximate constant of about 4.0 over 1–10 GHz and exhibited an abrupt decrease point at 12 GHz. Interestingly, the ε″ values of the two samples also achieved the max values at the frequency of 12 GHz. As the real and imaginary parts of the relative complex permittivity represent the energy storage ability and loss ability, respectively, ε′ and ε″ of both SHCuPc and HCuPc showed a sharp fluctuation throughout the whole frequency range, indicating that the two kinds of copper phthalocyanine possessed good dielectric loss. As for the permeability properties of two kinds of the copper phthalocyanine, the real part μ′ values of the samples exhibited an abrupt decrease from 2.5 to 0.0 at the frequency of 12 GHz, at the same time, the imaginary part μ″ values achieved the max values about 2.0. Figure 6 revealed that both ε′ and μ′ of two kinds of the samples exhibited frequency dispersion effect, which means dielectric loss and magnetic loss occurred simultaneously for the two samples, in favor of the EM wave absorption. It is interesting to find that broad peaks centered at 12 GHz are present in two samples in Fig. 6, indicating the occurrence of natural resonance in the prepared copper phthalocyanine from two synthetic methods.

FIG. 6.

(a) RL of SHCuPc/wax composites and (b) HCuPc/wax composites in the frequency range of 1–18 GHz.

FIG. 7.

(a) Permittivity spectra of the SHCuPc/wax composites; (b) permeability spectra of SHCuPc/wax composites; (c) permittivity spectra of the HCuPc/wax composites; (d) permeability spectra of SHCuPc/wax composites in the frequency range of 1–18 GHz.

According to the measured data of permittivity and permeability, the RL usually can be evaluated by the following equation:

(1)$${\rm{RL}} = 20\log |(Z_{{\rm{in}}} - Z_0 )/(Z_{{\rm{in}}} + Z_0 )\quad ,$$

where Z ₀ is the impedance of free space and Z _in is the input characteristic impedance, which can be expressed as

(2)$$Z_{{\rm{in}}} = Z_0 \sqrt {({\rm{\mu }}_r /{\rm{\varepsilon }}_r )} \tanh \left\{ {(2{\rm{\pi }}ft/c)\sqrt {{\rm{\mu }}_r } {\rm{\varepsilon }}_r } \right\}\quad ,$$

where c is the velocity of light and t is the thickness of an absorber. In this letter, the t values are in millimeter units. μ_r and ε_r are the measured relative complex permeability and permittivity, respectively. For microwave absorbing application 33, when the RL is less than −20 dB, we can say that Z _in and Z ₀ are matched. When the impedances are matched, the frequency and the absorber thickness are termed as “matching frequency” (f _m) and “matching thickness” (t _m), respectively. It can be seen that the two samples exhibited good ability of microwave absorption. For SHCuPc sample, with a matching thickness t _m = 5 mm, the maximum RL Rmax is approximately −50 dB at 16.5 GHz. At t _m = 5 mm, the bandwidth corresponding to RL below −10 dB (i.e., over 90% microwave absorption) is higher than 12 GHz. Otherwise, for HCuPc sample, with a matching thickness t _m = 2 mm, the maximum RL Rmax is approximately −40 dB at 14.9 GHz. At t _m = 2 mm, the bandwidth corresponding to RL below −10 dB is about 9 GHz. Both kinds of copper phthalocyanine showed excellent wideband microwave absorption properties, but the matching frequency and matching thickness are different; we assumed this difference to the morphology and aggregation properties of the samples.

Beside the dielectric loss or the magnetic loss, the EM wave can be absorbed via a so-called “geometrical effect,” which means that when the thickness of the absorber (t) satisfies the equation t = nλ_m/4 (n = 1, 3, 5…), the incident and reflected waves in the absorber are out of phase 180°. As a result, the reflected waves in the air-absorber interface are totally canceled. In the above equation, λ_m = λ_o/(|μ_r||ε_r|)^1/2 is the wave length in the absorber at some specific frequency, |μ_r| and |ε_r| are the moduli of μ_r and ε_r, respectively, and λ_o is the wave length in the free space (λ_o = c/f). Such a geometrical effect has also been reported by Yusoff et al.^{Reference Lippel, Wilson, Miller, Wöll and Chiang35,Reference Yusoff, Abdullah, Ahmad, Jusoh, Mansor and Hamid36} According to the EM wave theory, when an EM wave is incident on an absorber, Z ₁ should be equal to Z ₀ for zero reflection, where Z ₁ is the impedance in the absorber and Z ₀ is the impedance of free space. Z has a general expression as (i = 0 and 1 for the free space and the absorber, respectively). In the free space, both the relative permeability μ_r0 and the relative permittivity ε_r0 are equal to 1. Therefore, for zero reflection, the relative permeability and relative permittivity in the absorber should satisfy the relationship μ_r1 = ε_r1. However, for copper phthalocyanine obtained from the two synthetic methods, the permittivity is much larger than the permeability. The impedance matching condition μ_r1 = ε_r1 cannot be satisfied. Therefore, we think that the observed microwave absorbing performance is not only from dielectric loss and magnetic loss but also from geometrical effect.^{Reference Ma, Cao, Liu and Wang37} For the normal HCuPc, at f _m = 14.94 GHz, the ε_r1 = 3.298 − j2.336 and μ_r1 = 1.218 − j0.708, substituting these values into the equation t = λ_m/4, we obtain t = 2.1 mm, which is exactly equal to the matching thickness t _m found based on Fig. 7. So the above calculations justify our hypothesis that the excellent microwave absorbing properties observed at f = 14.94 GHz is due to the geometric effect. Considering that HCuPc are uniformly distributed in the composites as the absorbent, there are more chances to satisfy this equation. But for the SHCuPc, at f _m = 16.42 GHz, ε_r1 = 3.08 − j1.058 and μ_r1 = 2.571 − j0.772, substituting these values into the equation t = λ_m/4, we obtain t = 1.5 mm, however, for the SHCuPc, the impedance matching condition (μ_r1 = ε_r1) cannot be satisfied. Therefore, the RL showed in Fig. 6 might be attributed to the synergistic effects derived from the interaction between the natural resonance and special spherical morphology. Compared to the EM properties of the copper phthalocyanine from two synthetic methods, we believe that the EM property would be significantly influenced by the detailed morphology of the prepared samples.