Article pubs.acs.org/IC
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
Isomerization Reaction of mer- to fac-Tris(2-phenylpyridinatoN,C2′)Iridium(III) Monitored by Using Surface-Enhanced Raman Spectroscopy Bo-Han Wu, Min-Jie Huang, Cheng-Chang Lai, Chien-Hong Cheng, and I-Chia Chen* Department of Chemistry, National Tsing Hua University, Hsinchu, Taiwan 30013, Republic of China S Supporting Information *
ABSTRACT: We developed a new method by enclosing the complex tris(2-phenylpyridinato-N,C2′)Iridium(III), Ir(ppy)3 with surfactant cetyltrimethylammonium bromide (CATB), coated with a thin layer of silica then bonded to the surface of silver nanoparticle. These samples were used to acquire surface-enhanced Raman scattering (SERS) spectra. The thickness of silica layer was controlled to have efficient phosphorescence quenching and Raman enhancement by metal nanoparticle. The SERS spectra of fac- and mer-Ir(ppy)3, recorded at 633 nm excitation, display distinct ring breathing mode features because the total symmetric vibrational bands were enhanced. This provides a convenient means to differentiate these isomers with great sensitivity and to study their isomerization process. A direct conversion reaction of mer- to fac- isomerization is identified with time constant 3.1 min when mer was irradiated with Xe light. Via thermal activation, under moderate conditions (pH 5.5 and 343 K), we observed an intermediate particularly with new bands 320/662 cm−1 after heating for 17.5 h, and then those bands disappeared to form fac-Ir(ppy)3. On the basis of DFT calculations, the intermediate is proposed to contain octahedral N−N Ir(ppy)3−HO−silica structure; band at 320 cm−1 is assigned to iridium oxygen stretching mode νIr−O of this intermediate. Under acidic conditions, pH 1−2 catalyzed by silanol in silica, byproduct with band at 353 cm−1 was observed. According to the SERS bands and the calculation, this byproduct is assigned to be iridium(III) siloxide, and the new band is assigned to νIr−O.
■
acetic acid and silica gel at room temperature,10 but lacking a technique to monitor the conversion reaction in situ, the mechanism is unclear. Lai et al. reported the Raman spectra of several iridium complexes including fac-Ir(ppy)3.11 Combining the vibrational mode calculations using density functional theory (DFT), they were able to assign most of the vibrational bands and the metal related modes in the low-wavenumber region. On the basis of varied molecular symmetry, utilizing Raman spectroscopy appears to be a convenient means to differentiate these two stereoisomers. However, SERS has some intrinsic drawbacks. First, the measurements are usually made using the advantage of molecular affinity to the metal surface of nanoparticles or forming covalent bonds to gain sensitivity.12−14 For highly emissive molecules, the metal surface efficiently quenches excited state to avoid interference in attaining Raman scattering.15 Second, sample molecules tend to either reduce or oxidize during SERS measurements, because metal nanoparticles inject hot electrons/holes to nearby molecules after illuminating plasmonic wavelength light.16,17 If we try to use a conventional method to detect Ir(ppy)3 SERS signal, i.e., mixing metal nanoaprticles solution and Ir(ppy)3, we have no signal or wrong signals. Because Ir(ppy)3 has no moiety to
INTRODUCTION The iridium complexes with cyclometalating ligands emitting efficient phosphorescence are widely used in organic light-emitting devices (OLEDs). The photophysical character of iridiumcomplexes depends on ligands and coordination geometries.1,2 In tris-homoleptic iridium complexes, there exist two stereoisomers, facial ( fac) and meridional (mer), the structures of which are shown in Scheme 1, and each has varied photoluminescent properties.1,3−5 Yang et al. showed that the iridium complex, e.g., tris(2-phenylpyridinato-N,C2′)Iridium(III) Ir(ppy)3, has greater illumination efficiency in fac-form,3 but with carbene chelating, e.g., tris(1-phenyl-3-methylbenzimidazolin-2-ylidene-C,C2′)iridium(III) Ir(pmb)3, the mer-form showed a better OLED efficiency.2 The mer-form is a kinetics-controlled product. Conversion to fac-form can be both photoinduced and thermally activated. Karatsu et al.6 and Zheng et al.7 reported that the photoinduced conversion is an one-way geometrical isomerization. McDonald et al.8 and Tsuchiya et al.9 proposed the mechanisms of thermal and photochemical isomerization, respectively, mainly based on the NMR and circular dichroism measurements among others. In thermal isomerization, McDonald et al. proposed that the process is an alcohol-catalyzed reaction and can proceed at moderate temperatures.8 Deaton et al. found that mer-Ir(piq)2(ppy) can partially transform to fac-Ir(piq)2(ppy) in dichloromethane with © XXXX American Chemical Society
Received: January 15, 2018
A
DOI: 10.1021/acs.inorgchem.8b00113 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry Scheme 1. Preparation of SERS Samplesa
(a) Ir(ppy)3@CTAB: Ir(ppy)3 in CTAB micelle. (b) Ir(ppy)3@CTAB@SiO2: micelle covered with silica. (c) Ir(ppy)3@CTAB@SiO2−NH3+. (d) Ir(ppy)3@SiO2: CTAB is washed out to leave complex covered with silica. (e) Ir(ppy)3@CTAB@SiO2−NH3+ anchored on AgNP. (f) Ir(ppy)3@SiO2−NH3+. (g) Ir(ppy)3@SiO2−NH3+ anchored on AgNP.
a
Ir(ppy)3@SiO2−NH3+ and Ir(ppy)3@CTAB@SiO2−NH3+. A few milligrams of Ir(ppy)3 and 0.21 g of CTAB were charged in 10 mL of water, under sonication until the solution became clear to form ellipsoid micelle 25 × 5 nm2;23 then, these were mixed with 2.12 mL of 1 M ammonia. Next, 1 mL of TEOS was added dropwise and stirred for 20 min. Abundant acetone was added to the emulsion, which was centrifuged and washed with acetone until the supernatant contained no Ir(ppy)3. The precipitates were collected and redispersed by isopropanol to 20 mL. The surface of these silica nanoparticles was then modified with amine groups as 2 μL of APTES was added and stirred overnight to obtain Ir(ppy)3@CTAB@SiO2−NH3+, diameter ∼ 20 nm. We adapted an ionexchange method which removed surfactant CTAB by NH4NO3(alc).24 The precipitates Ir(ppy)3@CTAB@SiO2 were redispersed by 20 mL of ethanol and added 0.41 g of NH4NO3 for removing surfactant CTAB. After stirring for 1 h, samples were centrifuged and washed with ethanol one time and the precipitates redispersed in isopropanol to 20 mL. The surface of nanoparticles was then modified with amine groups as 2 μL of APTES was added and stirred overnight to obtain Ir(ppy)3@ SiO2−NH3+. Complex mer-Ir(ppy)3, enclosed in the silica pores because of its insolubility in solvent, isopropanol, prefers to stay inside the silica pores. Raman and SERS Measurements. The Raman and SERS spectra were recorded in a backscattering geometry. A He−Ne laser (Lasos) through a laser line filter, reflected by an edge filter (Semrock), served as the excitation light source 632.8 nm. A 10× objective (ZEISS, n.a. = 0.26) focused the laser beam onto sample, and the power at the sample region was set at ∼10 mW. The scattered signal passing through an edge filter, lens, an optical fiber, and a monochromator (iHR550, Horiba; length 0.55 m; grating 600 grooves/mm) was detected with a liquidnitrogen-cooled charge-coupled device (CCD, model Symphony II, Horiba). The spectral resolution was maintained at 3 cm−1. The integration period per scan was typically about 30 s and averaged for 5−10 scans for a spectrum. The 500 W xenon lamp was used as the light source for the photoinduced reactions. For thermally activation measurements, Ir(ppy)3@ SiO2−NH3+ (Ir(ppy)3@CTAB@SiO2) was charged (10 mL in a flask with round bottom) and heated at 343 K in oil bath. The pH value of solution was adjusted using 1 M HCl(aq). The ∼50 μL samples irradiated/heated for given time periods were mixed with 250 μL of AgNP to measure the SERS spectra. DFT Calculations. The structures and vibrational normal-mode analysis of fac-, mer-Ir(ppy)3, intermediate and product in thermal activation process were calculated using DFT method B3LYP with basis sets
covalently bind on metal surface to gain sufficient Raman enhancement. Besides, if we try to use ethanol as dispersion reagent, then iridium complexes tend to undergo redox reactions during SERS detection in solution and show the wrong signals (Figure S1). One way to avoid the hot electrons/holes during plasmonic excitation is to cover the metal nanoparticle with silica shell.18 Nevertheless, the silica surface is normally hydrophilic, and no SERS signal is detected. In this present work, for using SERS technique to detect iridium complexes in solution and to monitor their reaction, our strategy is covering the complex with thin silica shell then anchor to the surface of metal nanoparticles to obtain sufficient SERS enhancement. In this way, quenching in phosphorescence by metal surface remains efficient19 to avoid interference with the Raman signal. The structural information on iridium complexes as well as reaction intermediate can be detected. The method can be applied to any hydrophobic system which can be emissive. Here, we utilize this scheme to study the isomerization mechanism of mer- to fac-Ir(ppy)3.
■
EXPERIMENTAL SECTION
Materials. Silver nitrate and gold(III) chloride trihydrate (HAuCl4· 3H2O) were from Alfa Aesar. Sodium citrate, n-octyltrimethylammonium bromide (OTAB), cetyltrimethylammonium bromide (CTAB), tetraethylorthosilicate (TEOS), 2-propanol, ethanol, ammonium hydroxide solution (NH4OH), ammonium nitrate (NH4NO3), and (3-aminopropyl)-trimethoxysilane (APTES) were from Sigma-Aldrich. Hydrochloric acid (HCl) and acetone were from J. T. Baker. Sample fac-Ir(ppy)3 was purchased (Lumtec Technology) and used without purification, and mer-Ir(ppy)3 was synthesized following Thompson’s method.1 All chemicals were used as received. Pure deionized water (Milli-Q Millipore, 18.2 MΩ/cm) was used in all preparations. SERS Sample Preparation. SERS sample preparation is depicted in Scheme 1. The silver nanoparticles (diameter = 30 nm) covered with citrate were synthesized by citrate redox methods.20 Iridium complexes were embedded with surfactant CTAB (or OTAB). Then, ammonia and TEOS were added to form silica shell, Ir(ppy)3@CTAB@SiO2 (Ir(ppy)3@OTAB@ SiO2).21 Surface modification with APTES on silica shell surfaces provides a covalent attachment amine group (−NH2) to silver nanoparticles carboxylic groups (−COO−) of citrate.22 B
DOI: 10.1021/acs.inorgchem.8b00113 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
Figure 1. Raman of solid-phase and AgNP SERS spectra of fac- and mer-Ir(ppy)3. Raman spectrum of crystal form (a) fac-Ir(ppy)3 and (b) mer-Ir(ppy)3. SERS spectra of (c) fac-Ir(ppy)3@SiO2−NH3+ and (d) mer-Ir(ppy)3@SiO2−NH3+ on AgNPs. The asterisk and pound sign denote bands from silica and isopropanol, respectively (ν for stretching, γ for in-plane bending of ligands, Δ and Γ for ring−ring in-plane and out-of-plane twisting modes, respectively).
The bands centered at 761 and 816 cm−1 are signals from silica shell and solvent isopropanol, respectively. The broad band near 1450 cm−1 is attributed to silica, isopropanol, and Ir(ppy)3. The iridium complex gained more intensity because of electronic resonance Raman effect at 633 nm excitation especially for the mer form.29 In SERS spectra, the vibrational bands with totally symmetric modes are known to be better enhanced.11 We compare normal Raman signal with Ag SERS of fac-Ir(ppy)3 (Figure 1). The νIr−N has two bands at 236 (e) and 263 cm−1 (a). The a band is weak in normal Raman, but the intensity increases in Ag SERS. Bands at 641 cm−1 (e) and 671 cm−1 (a) are assigned to in-plane ring twisting modes. In normal Raman, the intensity in 641 cm−1 is greater, but in SERS, the a and e bands display similar intensity. The same explanation is applied for bands 1301/1310 and 1057/ 1064 cm−1 that are assigned to ring stretching modes mixed with C−C−H in-plane bending modes. For the ring breathing modes, in mer-Ir(ppy)3, SERS mode 1038 cm−1 (a) is much more intense than 1026 cm−1 (e), but in fac-Ir(ppy)3, the 1038 cm−1 (a) band is greatly enhanced to display approximate equal intensity. Consequently, this apparent variation in spectral features for the ring breathing modes 1026/1038 cm−1 provides a convenient means to distinguish these two isomers. Photoinduced Isomerization. We utilized the varied SERS features in isomers to monitor the reactions of the complex. The mer-Ir(ppy)3@CTAB@SiO2−NH3+ was illuminated by a 500 W xenon lamp, and we recorded the SERS spectra after varied irradiation periods. The solvent bands also serve as internal standards for position calibration and intensity normalization. The spectra are displayed in Figure 2. Because of greater extinction in absorption at 633 nm in mer, the mer SERS bands are more intense. After irradiation with Xe lamp, the mer bands at 651 and 1317 cm−1 decreased, and the fac bands at 641 and 1310 cm−1
LANL2DZ for iridium and 6-31g* for other atoms were employed.25−27 Gaussian 09 program package was used.28
■
RESULTS Raman and SERS Spectra. Figure 1 displays the Raman spectra of solid fac- and mer-Ir(ppy)3 and SERS spectra of Ir(ppy)3@SiO2−NH3+ absorbed onto silver nanoparticles. Lai et al. performed the detailed vibrational mode assignments.11 Briefly, metal−ligand bending and stretching and phenyl-pyridyl ring−ring modes lie in the wavelength region 150−350 cm−1, ring in-plane and out-of-plane twisting modes in 350−800 cm−1, ring breathing ∼1000 cm−1, and stretching modes within ring 1100−1600 cm−1. The bands of νIr−N and νIr−C lie at 236 (e symmetry)/263(a) and 278(a)/308(e) cm−1, respectively for fac-Ir(ppy)3; the Ir−ligand stretching modes split into 225, 243/266, and 278/309, 322 cm−1 in mer-Ir(ppy)3. In the wavenumber range of 650−1650 cm−1, minor variations in band positions and intensities are displayed for these two isomers. The SERS spectra for both isomers Ir(ppy)3@SiO2−NH3+ display similar band positions and bandwidth to those obtained in normal Raman scattering on solid crystals indicating no redox reactions or decomposition occurred. The calculated SERS enhancement factor solely via plasmonic effect is ∼104 (see the Supporting Information); this value agrees with the data in the study of Shanthil et al.12 The SERS spectra on Ir(ppy)3@CTAB@ SiO2−NH3+ are similar except for smaller scatter intensity and greater emission background. We estimated the thickness of silica shell as 2−3 nm (Figure S2). Without CTAB, complexes were closer to the silver surface so that the phosphorescence is significant quenched. Low in emission background, better signalto-noise ratio in Raman scattering of Ir(ppy)3@SiO2−NH3+ was attained. C
DOI: 10.1021/acs.inorgchem.8b00113 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
Figure 3. Mole fraction of fac- and mer-Ir(ppy)3 obtained in photoinduced reaction using the intensities of ring breathing bands (1026/1038 cm−1) in Ag SERS spectra. Circles are experimental data and lines are the best exponential fitting curves.
We then lowered pH in solution to increase the reaction rate for the first step to find the intermediate. At pH 6.5, silica has mostly SiO− group on surface. As pH is lowered, the amount of silanol SiOH group increases.30 At pH 5.5 and also at 343 K, after heating for 17.5 h, we observed the intermediate with spectral features similar to fac, as shown in Figure 4b. The solvent peak intensity can serve as an internal standard, so we normalized spectra with the band at 816 cm−1. Comparing Figure 4a,b, we found that the intermediate gained better SERS enhancement to have intensity exceeding fac. Hence, most bands appeared are attributed to the intermediate. The νIr−N (234/263 cm−1) and νIr−C (277/308 cm−1) are similar to those of the fac, except that one extra band at 320 cm−1 appeared. Tsai et al. assigned the νIr−O to be 304 cm−1 for FIrpic (bis[2-(4,6-difluorophenyl)pyridinato-C2,N](picolinato)iridium(III)).31 FIrpic has two fluorine electron withdrawing groups, weakening the Ir−O bonding. Hence, this band is tentatively assigned to νIr−O for the intermediate. For the in-plane twist of aromatic ring, 671/662/ 641 cm−1, the new band 662 cm−1 emerged. The ring breathing bands at 1030/1040 cm−1 and bands in the 1300−1600 cm−1 region are similar to those in fac. These bands disappeared after 21.5 h, and only fac-form was detected (Figure 4a). We also used surfactant OTAB with shorter chain to enclose mer-Ir(ppy)3 to obtain smaller micelles, and consequently smaller silica pores in the acid catalytic reaction. In these acidity reactions at room temperature, use of surfactant OTAB yielded the same Raman band positions but because of less distance from the silver nanoparticle surface than CTAB we obtained less emission background thus superior signal-to-noise ratio. After adding more HCl solution and reaching pH 1−2 at temperature 296 K, OTAB was replaced by H+ in the cavities, and the amount of silanol group was significantly increased. The SERS curve acquired right after adding the HCl solution on mer-Ir(ppy)3@OTAB@SiO2−NH3+ is shown in Figure 5. We observed no end product fac form, but new complex with distinct vibrational structure from mer was formed. The vibrational bands are blue-shift from those of mer-Ir(ppy)3 in the wavenumber range 900−1600 cm−1, for instance, 1038/1578/1600 cm−1 in mer and 1042/1582/1605 cm−1 in the new product. In the low frequency region, the νIr−N bands are moved to 247/259 cm−1, νIr−C to 296/315 cm−1 and a new band appeared at 353 cm−1. In this acid catalyzed silica sol−gel process, the silica pores tend
Figure 2. SERS curves recorded after illumination of Xe lamp on mer-Ir(ppy)3@CTAB@SiO2−NH3+ for 1, 2, 10, 17, 22, 37, and 55 min. The bottom curve is mer-Ir(ppy)3@CTAB@SiO2−NH3+ before radiation. The top curve is fac-Ir(ppy)3@CTAB@SiO2−NH3+. The asterisk and pound sign indicate the bands from silica and isopropanol, respectively.
increased with time. No new band appeared during this conversion process indicating no intermediate existing during photoinduced mer- to fac-isomerization. To obtain the intensity ratio of 1038 to 1026 cm−1 bands after correction over partial contribution from fac in 1038 cm−1 and individual enhancement factors, we calculated the molar fraction of both forms versus illumination time as shown in Figure 3. The decay of mer-form in Ir(ppy)3@CTAB@SiO2−NH3+ is rapid. From the decay and the corresponding rise on the fac form, we obtained a time constant (3.1 ± 0.6) min. In agreement with the results reported by Zheng et al.7 on using 19F NMR as a means to monitor the production and Karatsu et al.,6 the photoinduced reaction is a direct process. The controlled experiments were performed on fac-Ir(ppy)3@ CTAB@SiO2−NH3+ for irradiation up to 6 h. No reaction was observed in the SERS spectra, indicating the fac form is stable under irradiation of ultraviolet to visible light. Thermally Activated Isomerization. mer-Ir(ppy)3@ CTAB@SiO2−NH3+ was heated at 343 K for over 8 days, but no reaction occurred (Figure S7). As suggested by McDonald et al., the isomerization reaction can be catalyzed by hydroxyl molecules.8 Hence, we used silanol in silica to catalyze this reaction. Samples without surfactant CTAB mer-Ir(ppy)3@ SiO2−NH3+ were heated to 343 K, and SERS spectra were recorded at several time increments (Figure S8). The mer- peaks at 1317/651/322/309/243/226 cm−1 decreased over time and the fac- bands at 1312/641/306/234 cm−1 increased with a time constant of ∼18.5 h. At these moderate temperatures and pH ∼6.5, reaction is slow to observe sufficient amount of intermediate. D
DOI: 10.1021/acs.inorgchem.8b00113 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
Figure 4. SERS spectra of mer-Ir(ppy)3@SiO2−NH3+ at pH 5.5 and 343 K after heated for (a) 21.5 and (b) 17.5 h. The intensity is normalized to the solvent peak at 816 cm−1. (c) DFT Raman spectrum of Ir(ppy)3HOSiH3 (scaled by 0.97).
Figure 5. Ag SERS spectra of mer-Ir(ppy)3@OTAB@SiO2−NH3+ (a) with abundant HCl(aq) and (c) without HCl(aq). (b) DFT spectrum of Ir(ppy)2OSiH3 (scaled by 0.97).
to become smaller.30 In the end, the iridium complexes leaked out, and no SERS signal of iridium complex was detected after few hours. Moreover, the controlled reaction was carried out, and we found fac-Ir(ppy)3@OTAB@SiO2−NH3+ at pH 1−2 underwent no reaction (Figure S9).
calculated energy of oct N−N form is 0.183 eV lower than the oct C−N form. Hence, the N−N form is more feasible using the lowest energy pathway. The calculated Ir−O bond length is 2.325 Å for oct N−N which is relatively long and νIr−O is 313 cm−1. Figure 4c depicts the calculated Raman curve for the oct N−N form and the assignments. The agreement of the calculated with the observed spectra in the low wavenumber region is quite satisfactory. Some bond lengths and the calculated vibrational mode wavenumbers are listed in Tables 1 and 2, respectively. For oct N−N, νIr−N is calculated to be 240/257 cm−1, and νIr−C is 273/279/308 cm−1. νIr−N is blue-shifted from that of mer but the νIr−C of the dissociating ppy is red-shifted to 279 cm−1 with Ir−C bond length 2.210 Å whereas the others 1.993/2.079 Å. The new band at 662 cm−1 is assigned to the phenyl ring in-plane twist of the dissociating ppy. In the wavenumber range 800−1650 cm−1 (Figure 4b) the observed band positions 1005/ 1040/1310/1599 cm−1 assigned to pyridyl/phenyl ring breathing/ν(ring)+δ(CH)/pyridyl ring stretching modes are similar to
■
DISCUSSION For revealing the structures of the intermediate and the new product in acidity reactions, we employed DFT quantum chemical calculations and used the simplified structures for them. Intermediate. For the isomerization reaction, at pH 5.5 and 343 K, possibly the weak-bond Ir−N was dissociated, the iridium chelated to the oxygen of silanol, and then the hydrogen of the silanol moved toward the N site of the ppy ligand to form the intermediate. The model compound Ir(ppy)3HOSiH3, as displayed in Figure 6, is used in DFT calculation. Two isomers octahedral N−N and C−N exist for this model compound. Using method B3LYP/LAN2DZ(Ni), 6-31G*(H, C, N, O, Si), the E
DOI: 10.1021/acs.inorgchem.8b00113 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
and νIr−C at 296/312 cm−1 near the observed positions, as shown in Figure 5b. The calculated νIr−O at 350 cm−1 agrees well with the observed value. The bands for Ir−N and Ir−C are blue-shifted from those of fac- and mer-Ir(ppy)3; thus, a more rigid structure with shorter iridium-ligand bonds is for the byproduct. In high wavenumber region, the observed bands at 1042/1582/1605 cm−1 corresponding to phenyl ring breathing/ phenyl/pyridyl ring stretching are also slightly blue-shifted from those of fac- and mer-Ir(ppy)3. However, the calculated positions for these bands are similar for both isomers and the model compound. It is possible that Ir(III) siloxide was confined in the silica cavity leading to some restriction on ring vibrations. We propose that mer-Ir(ppy)3 under acidic condition can be protonated first on ppy then reacted with silanol. The protonated ppy (H2ppy+) was dissociated, and the Ir(III) siloxide was formed. Under the low pH condition, possibly reaction was too rapid to observe any intermediate. Because Ir(III) siloxide was relatively stable, it remained in the silica pore and was readily detected with SERS, but the fac product leaked out. Hence, under acidic conditions, this stable byproduct can be formed. Scheme 2 depicts the reaction mechanism for this conversion reaction. By normalizing the peak intensities of the breathing modes in these iridium complexes with the solvent peak at 816 cm−1 which served as internal standard, we can roughly infer that the SERS cross sections of Ir(III) siloxide, intermediate, and mer-Ir(ppy)3 are comparable and about 8 times greater than that of fac-Ir(ppy)3 in this system. Hence, at 17.5 h although the band intensities of the intermediate were dominant, the amount of fac was comparable with the intermediate.
Figure 6. Model compounds Ir(ppy)3HOSiH3 for the intermediate and Ir(ppy)2OSiH3 iridium(III) siloxide. tbp: trigonal bipyramidal, oct: octahedral, sp: square pyramidal. L, L′, and L′′ are labels for varied ppy ligand.
Table 1. DFT Bond Lengths of Optimized Geometries with B3LYP/LAN2DZ(Ni), 6-31G*(H, C, N, O, Si) for fac- and mer- Ir(ppy)3, Intermediate, and Ir(III) Siloxidea bond length (Å) Ir−N (L′) Ir−N (L) Ir−N (L′′) Ir−C (L′′) Ir−C (L) Ir−C (L′) Ir−O a
fac 2.191
2.026
mer
intermediate
Ir(III) siloxide
2.219 2.087 2.067 2.113 2.092 2.011
2.100 2.065
2.061 2.058
2.210 2.079 1.993 2.325
■
CONCLUSION In this study, we present the normal Raman and SERS data which provide applicable ways to identified fac- and mer-isomers of iridium complexes. As complex Ir(ppy)3 is dissolved in solvent, the strong phosphorescence provokes difficulty in monitoring isomerization reaction by Raman scattering. In addition, the iridium complexes tend to decompose due to reaction with nanoparticles at plasmonic excitation. We attempted to address this problem by designing a new sample preparation method. We covered the complex with nano thickness silica shell and allowed them to bond to the surface of nanoparticles to quench emission and to gain SERS enhancement. In the obtained spectra, the a symmetry vibrational bands tend to have better enhancement than the e symmetry. Hence, in SERS spectra, the intensities of ring breathing vibrational modes of these isomers differ significantly to provide a sensitive means to identify them and also to verify the assignments. In this way, we successfully monitored the photoinduced and thermally activated mer to fac isomerization reaction of Ir(ppy)3 in situ. The photoinduced isomerization is a direct conversion with time constant 3.1 min. At moderate temperatures, mer-Ir(ppy)3 underwent no thermal isomerization to fac-Ir(ppy)3 without hydroxyl group. By adjusting the pH to increase the amount of silanol groups in silica, the intermediate was produced with sufficient amount to be probed by silver SERS. On the basis of the theoretical simulation, we assigned the structure of the intermediate and the νIr−O mode to be 320 cm−1. At low pH and room temperature, mer-Ir(ppy)3 reacted to produce Ir(III) siloxide. Hence, Ir(III) siloxide is assigned to be a byproduct for the isomerization reaction. This present work offers a superior way to prepare SERS sample for hydrophobic molecules with low redox
2.017 1.995 2.051
Label for varied ligand ppy is shown in Figure 6.
those in fac-Ir(ppy)3. After heating over 21.5 h (Figure 4a), these intermediate bands diminished indicating that the mer-Ir(ppy)3 was catalyzed by SiO2 to form the intermediate with the silanol group bonded to iridium, and then the iridium oxygen bond dissociate to produce fac-Ir(ppy)3; the mechanism proposed is depicted in Scheme 2. Acidity Reactions. In acidity reactions at low pH, only two bands of each νIr−N and νIr−C appeared, and one new band at 353 cm−1 can be assigned to νIr−O (Figure 5a). McDonald et al. proposed the iridium alkoxide, Ir(C,N)2OR, to be the intermediate in thermal isomerization catalyzed by alcohol.8 In the present work the new product with two ppy ligands and Ir−O bonding is close to the proposed iridium alkoxide. We then used the simplified molecule Ir(ppy)2OSiH3, as shown in Figure 6 in the DFT calculations. There are geometric isomers for this model compound, trigonal bipyramidal (tbp) N−N and C−N, and square pyramidal (sp) cis and trans. According to the results of calculations, the structure of sp trans is unstable; the energies of these isomers lie in the order of tbp N−N (0) < tbp C−N (0.26 eV) < sp cis (1.59 eV). Assuming that the lowest energy path is the most probable, the vibrational analysis was performed based on isomer tbp N−N. The calculated results show that the Ir−C, Ir−N, and Ir−O bond lengths are 1.995/2.017, 2.058/2.061, and 2.051 Å, respectively, listed in Table 1. We obtained νIr−N at 244/254 cm−1 F
DOI: 10.1021/acs.inorgchem.8b00113 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
Table 2. Experimental and DFT Band Position (cm−1, x0.97) for fac- and mer-Ir(ppy)3, Intermediate, and Ir(III) Siloxide fac vibrational mode (cm−1)
mer
intermediate
Ir(III) siloxide
exp.
cal.
exp.
cal.
exp.
cal.
exp.
cal.
νIr−N
236
222 (E)
240 257
247 259
244 254
263 278 308
251 (A) 263 (A) 295 (E)
212 233 257 266 295 311
234 263
νIr−C
225 243 266 278 309 322
277
296 315 353
296 312 350
631 (A) 634 (E)
651
630 645
642 650
660 (A) 664 (E)
671
656 675
663 670
1005 1009
994 (A) 995 (E)
1005 1010
1010
998 1006
1030
1020 (E)
1026
1038 1581
1030 (A) 1593 (E)
1038 1578
1030 1042
1020 1032
1582
1593 1596
1599
1596 (A) 1605 (A) 1606 (E)
629 635 643 648 656 667 993 997 1004 1007 1019 1030 1588 1590 1597 1605 1610 1612
273 279 308 313 622 635 643 639 657 669 996 996 1004 988 1018 1030 1585 1591 1597 1604 1610 1611
1605
1611 1611
νIr−O ΔL′′ pyridyl ring ΔL pyridyl ring ΔL′ pyridyl ring ΔL′′ phenyl ring ΔL phenyl ring ΔL′ phenly ring pyridyl ring breathL′′ pyridyl ring breathL pyridyl ring breathL′ phenyl ring breathL′′ phenyl ring breathL phenyl ring breathL′ νphenyl ring,L′′ νphenyl ring,L νphenyl ring,L′ νpyridyl ring,L′′ νpyridyl ring,L νpyridyl ring,L′
641
671
1600
308 320 641
662 671 1005 1010 1030 1040 1580
1599
Scheme 2. Mechanism for Thermal Activated Silanol Catalyzed Reaction mer- to fac-Ir(ppy)3
■
potential. This provides a usage for sensitive SERS technique to wider variety of molecules and to monitor their reactions in situ.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
ASSOCIATED CONTENT
Chien-Hong Cheng: 0000-0003-3838-6845 I-Chia Chen: 0000-0002-5821-6416
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.8b00113. SEM and TEM images, experimental and calculated Raman spectra, SERS data, thermal isomerization monitoring, geometrical isomers, optimized geometries (PDF)
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We are grateful for National Tsing Hua University under project “Frontier Research Center on Fundamental and Applied G
DOI: 10.1021/acs.inorgchem.8b00113 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
Thiol-Protected Ag Nanoparticles and a TERS. J. Phys. Chem. C 2014, 118, 10329−10334. (17) Sharma, K.; Singh, G.; Singh, G.; Kumar, M.; Bhalla, V. Silver Nanoparticles: Facile Synthesis and Their Catalytic Application for the Degradation of Dyes. RSC Adv. 2015, 5, 25781−25788. (18) Li, J. F.; Huang, Y. F.; Ding, Y.; Yang, Z. L.; Li, S. B.; Zhou, X. S.; Fan, F. R.; Zhang, W.; Zhou, Z. Y.; Wu, D. Y.; Ren, B.; Wang, Z. L.; Tian, Z. Q. Shell-isolated Nanoparticle-enhanced Raman Spectroscopy. Nature 2010, 464, 392−395. (19) Lin, H.-H.; Chen, I. C. Study of the Interaction between Gold Nanoparticles and Rose Bengal Fluorophores with Silica Spacers by Time-Resolved Fluorescence Spectroscopy. J. Phys. Chem. C 2015, 119, 26663−26671. (20) Lee, P. C.; Meisel, D. Adsorption and Surface-enhanced Raman of Dyes on Silver and Gold Sols. J. Phys. Chem. 1982, 86, 3391−3395. (21) Aiello, D.; Talarico, A. M.; Teocoli, F.; Szerb, E. I.; Aiello, I.; Testa, F.; Ghedini, M. Self-incorporation of a Luminescent Neutral Iridium(III) Complex in Different Mesoporous Micelle-templated Silicas. New J. Chem. 2011, 35, 141−148. (22) Bhat, R. R.; Fischer, D. A.; Genzer, J. Fabricating Planar Nanoparticle Assemblies with Number Density Gradients. Langmuir 2002, 18, 5640−5643. (23) Aswal, V. K.; Goyal, P. S.; Thiyagarajan, P. Small-Angle NeutronScattering and Viscosity Studies of CTAB/NaSal Viscoelastic Micellar Solutions. J. Phys. Chem. B 1998, 102, 2469−2473. (24) Lang, N.; Tuel, A. A Fast and Efficient Ion-Exchange Procedure to Remove Surfactant Molecules from MCM-41 Materials. Chem. Mater. 2004, 16, 1961−1966. (25) Becke, A. D. Density-Functional Thermochemistry. III. The Role of Exact Exchange. J. Chem. Phys. 1993, 98, 5648−52. (26) Rassolov, V. A.; Ratner, M. A.; Pople, J. A.; Redfern, P. C.; Curtiss, L. A. 6-31G* Basis Set for Third-Row Atoms. J. Comput. Chem. 2001, 22, 976−984. (27) Hay, P. J.; Wadt, W. R. Ab Initio Effective Core Potentials for Molecular Calculations − Potentials for K to Au including the Outermost Core Orbitals. J. Chem. Phys. 1985, 82, 299−310. (28) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, revision A.01; Gaussian, Inc.: Wallingford, CT, 2009. (29) Strommen, D. P.; Nakamoto, K. Resonance Raman Spectroscopy. J. Chem. Educ. 1977, 54, 474. (30) Wu, S.-H.; Mou, C.-Y.; Lin, H.-P. Synthesis of Mesoporous Silica Nanoparticles. Chem. Soc. Rev. 2013, 42, 3862−3875. (31) Tsai, H.-R.; Lu, K.-Y.; Lai, S.-H.; Fan, C.-H.; Cheng, C.-H.; Chen, I. C. Metal−Ligand Bonding Strength of Fluoro-Substituted Cyclometalated Iridium(III) Complexes from Raman and Infrared Spectra. J. Phys. Chem. C 2011, 115, 17163−17174.
Sciences of Matter” and Ministry of Science and Technology of Republic of China for support of this research.
■
REFERENCES
(1) Tamayo, A. B.; Alleyne, B. D.; Djurovich, P. I.; Lamansky, S.; Tsyba, I.; Ho, N. N.; Bau, R.; Thompson, M. E. Synthesis and Characterization of Facial and Meridional Tris-cyclometalated Iridium(III) Complexes. J. Am. Chem. Soc. 2003, 125, 7377−7387. (2) Sajoto, T.; Djurovich, P. I.; Tamayo, A.; Yousufuddin, M.; Bau, R.; Thompson, M. E.; Holmes, R. J.; Forrest, S. R. Blue and Near-UV Phosphorescence from Iridium Complexes with Cyclometalated Pyrazolyl or N-Heterocyclic Carbene Ligands. Inorg. Chem. 2005, 44, 7992−8003. (3) Yang, C.-H.; Fang, K.-H.; Chen, C.-H.; Sun, I. W. High Efficiency mer-iridium Complexes for Organic Light-emitting Diodes. Chem. Commun. 2004, 19, 2232−2233. (4) Lee, J.; Chen, H.-F.; Batagoda, T.; Coburn, C.; Djurovich, P. I.; Thompson, M. E.; Forrest, S. R. Deep Blue Phosphorescent Organic Light-emitting Diodes with Very High Brightness and Efficiency. Nat. Mater. 2016, 15, 92−98. (5) Holmes, R. J.; Forrest, S. R.; Sajoto, T.; Tamayo, A.; Djurovich, P. I.; Thompson, M. E.; Brooks, J.; Tung, Y.-J.; D’Andrade, B. W.; Weaver, M. S.; Kwong, R. C.; Brown, J. J. Saturated Deep Blue Organic Electrophosphorescence Using a Fluorine-free Emitter. Appl. Phys. Lett. 2005, 87, 243507. (6) Karatsu, T.; Nakamura, T.; Yagai, S.; Kitamura, A.; Yamaguchi, K.; Matsushima, Y.; Iwata, T.; Hori, Y.; Hagiwara, T. Photochemical mer → fac One-way Isomerization of Phosphorescent Material. Studies by Time-resolved Spectroscopy for Tris[2-(4’,6’-difluorophenyl)pyridine]iridium(III) in Solution. Chem. Lett. 2003, 32, 886−887. (7) Zheng, Y.; Batsanov, A. S.; Edkins, R. M.; Beeby, A.; Bryce, M. R. Thermally Induced Defluorination during a mer to fac Transformation of a Blue-Green Phosphorescent Cyclometalated Iridium(III) Complex. Inorg. Chem. 2012, 51, 290−297. (8) McDonald, A. R.; Lutz, M.; von Chrzanowski, L. S.; van Klink, G. P. M.; Spek, A. L.; van Koten, G. Probing the mer- to fac-Isomerization of Tris-Cyclometallated Homoand Heteroleptic (C,N)3 Iridium(III) Complexes. Inorg. Chem. 2008, 47, 6681−6691. (9) Tsuchiya, K.; Ito, E.; Yagai, S.; Kitamura, A.; Karatsu, T. Chirality in the Photochemical mer → fac Geometrical Isomerization of Tris(1phenylpyrazolato, N,C2′)iridium(III). Eur. J. Inorg. Chem. 2009, 2009, 2104−2109. (10) Deaton, J. C.; Young, R. H.; Lenhard, J. R.; Rajeswaran, M.; Huo, S. Photophysical Properties of the Series fac- and mer-(1-Phenylisoquinolinato-N∧C2′)x(2-phenylpyridinato-N∧C2′)3-xIridium(III) (x = 1−3). Inorg. Chem. 2010, 49, 9151−9161. (11) Lai, S.-H.; Ling, J.-W.; Huang, Y.-M.; Huang, M.-J.; Cheng, C. H.; Chen, I. C. Characterization of Ir(ppy)3 and [Ir(ppy)2bpy]+ by Infrared, Raman Spectra and Surface-enhanced Raman Scattering. J. Raman Spectrosc. 2011, 42, 332−338. (12) Shanthil, M.; Thomas, R.; Swathi, R. S.; Thomas, K. G. J. Ag@ SiO2 Core−Shell Nanostructures: Distance-dependent Plasmon Coupling and SERS Investigation. J. Phys. Chem. Lett. 2012, 3, 1459− 1464. (13) Szafranski, C. A.; Tanner, W.; Laibinis, P. E.; Garrell, R. L. SurfaceEnhanced Raman Spectroscopy of Aromatic Thiols and Disulfides on Gold Electrodes. Langmuir 1998, 14, 3570−3579. (14) Ai, K.; Liu, Y.; Lu, L. Hydrogen-bonding Recognition-induced Color Change of Gold Nanoparticles for Visual Detection of Melamine in Raw Milk and Infant Formula. J. Am. Chem. Soc. 2009, 131, 9496− 9497. (15) Ameer, F. S.; Pittman, C. U., Jr.; Zhang, D. Quantification of Resonance Raman Enhancement Factors for Rhodamine 6G (R6G) in Water and on Gold and Silver Nanoparticles: Implications for SingleMolecule R6G SERS. J. Phys. Chem. C 2013, 117, 27096−27104. (16) Tanabe, I.; Egashira, M.; Suzuki, T.; Goto, T.; Ozaki, Y. Prevention of Photooxidation of Deoxymyoglobin and Reduced Cytochrome c during Enhanced Raman Measurements: SERRS with H
DOI: 10.1021/acs.inorgchem.8b00113 Inorg. Chem. XXXX, XXX, XXX−XXX
