Rethinking Sensitized Luminescence in ... - ACS Publications

Jul 29, 2016 - Department of Chemistry & Biochemistry, Florida Atlantic University, ... molecular triplet state was able to explain the observed lumin...
2 downloads 0 Views 2MB Size
Article pubs.acs.org/IC

Rethinking Sensitized Luminescence in Lanthanide Coordination Polymers and MOFs: Band Sensitization and Water Enhanced Eu Luminescence in [Ln(C15H9O5)3(H2O)3]n (Ln = Eu, Tb) Jeffrey D. Einkauf,† Tanya T. Kelley,† Benny C. Chan,‡ and Daniel T. de Lill*,† †

Department of Chemistry & Biochemistry, Florida Atlantic University, 777 Glades Road, Boca Raton, Florida 33431 United States Department of Chemistry, The College of New Jersey, 2000 Pennington Road, Ewing, New Jersey 08628, United States



S Supporting Information *

ABSTRACT: A coordination polymer [Ln(C15H9O9)3(H2O)3]n (1-Ln = Eu(III), Tb(III)) assembled from benzophenonedicarboxylate was synthesized and characterized. The organic component is shown to sensitize lanthanide-based emission in both compounds, with quantum yields of 36% (Eu) and 6% (Tb). Luminescence of lanthanide coordination polymers is currently described from a molecular approach. This methodology fails to explain the luminescence of this system. It was found that the band structure of the organic component rather than the molecular triplet state was able to explain the observed luminescence. Deuterated (Ln(C15H9O9)3(D2O)3) and dehydrated (Ln(C15H9O9)3) analogues were also studied. When bound H2O was replaced by D2O, lifetime and emission increased as expected. Upon dehydration, lifetimes increased again, but emission of 1-Eu unexpectedly decreased. This reduction is reasoned through an unprecedented enhancement effect of the compound’s luminescence by the OH/OD oscillators in the organic-to-Eu(III) energy transfer process.



INTRODUCTION

The unique spectroscopic properties of lanthanide (Ln) ions (particularly sharp emission profiles and long lifetimes) are often exploited in coordination polymers (CPs) and metal− organic frameworks for a myriad of applications from sensing to bioimaging.1−8 Due to the parity and often spin forbidden nature of these transitions, direct excitation of the Ln ion is inefficient. In coordination complexes, organic “antennae” ligands are used to populate the excited state of the Ln through an energy transfer process, resulting in sensitized emission commonly referred to as the antenna effect (Figure 1).9−11 To construct luminescent Ln complexes, design guidelines have been established, including ideal energy values of the ligand’s triplet state,12 necessity of efficient intersystem crossing from the ligand’s singlet to triplet state (ΔE > 5000 cm−1),13 and exclusion of high-energy oscillators in proximity to the ion.14 Such guidelines have been extended to CPs, but due to the unique nature of these higher-dimensional solids (as opposed to discrete molecular entities), a direct correlation is not always appropriate. With the vast number of potential applications for luminescent Ln CPs, a fundamental understanding of their spectroscopic properties is essential. In the antenna effect (Figure 1), an organic ligand absorbs incident radiation to populate 1S. In the presence of a heavy Ln ion, ISC to 3T is facilitated. If 3T is of appropriate energy to resonate with an excited state manifold of the Ln, ET can occur. Once populated, radiative emission (L) from the excited Ln can © XXXX American Chemical Society

Figure 1. Modified Jabłoński diagram depicting the antenna effect: A = absorption, F = fluorescence, P = phosphorescence, L = luminescence, NR = nonradiative decay, 1S = singlet state, 3T = triplet state, ISC = intersystem crossing, ET = energy transfer, BT = back transfer. Also depicted is NR deactivation of excited Ln ions by proximal high-energy oscillators (here, OH).

then ensue if the energy is not deactivated by any number of competing NR mechanisms.15−20 To evaluate the overall efficiency of this process, emission quantum yields are often determined. Together with excited state lifetimes and energy values of the ligand’s 1S and 3T, a simplified qualitative description of these processes is obtained relatively easily. Received: April 9, 2016

A

DOI: 10.1021/acs.inorgchem.6b00878 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 2. Crystal structure of 1 viewed down [010] (left; two of the 2D layers are side by side) and [100] (right; looking down onto one of the 2D layers). Orange polyhedra = Ln monomers, red spheres = oxygen, black lines = carbon. Hydrogen atoms have been omitted.



To assess NR deactivation for emissive electronic states, the energy gap law is often used and states that the energy gap between the emissive state and the highest level of the ground state of Ln ions can be quenched by the vibrational modes of high-energy oscillators (Figure 1).14,21,22 The energy gap between the excited and ground state is ∼12 000 cm−1 for Eu(III) and ∼15 000 cm−1 for Tb(III); OH oscillators, with vibrational energy of ∼3200−3500 cm−1, provide a mechanism for ET that results in the NR deactivation of energy from the excited Ln(III) ion.21 OD oscillators, with vibrational energy of ∼2500 cm−1, are less efficient in quenching Ln excited states.22 This feature of OD oscillators allows for the study of NR decay pathways in Ln systems. Three to four of these oscillator quanta (e.g., OH from aqua ligands) are sufficient to quench Eu(III) or Tb(III) emission. Extending these concepts to CPs has led to the development of luminescent materials that behave similarly though not identical to molecular complexes.23,24 CPs sit at the crossroad between discrete molecular species and extended inorganic network solids; at this interface between molecule and solid, the luminescent properties are a combination of both systems and behave differently than each isolated system alone. Essentially all discussions of sensitized luminescence of CPs reference the antenna effect, effectively comparing them directly to molecular complexes.15,16,25−38,17−19,39−44 In doing so, there is no regard to the infinite, network arrangement of CP structures, which results in fundamentally different behavior than with molecular species. While sensitized emission of Ln CPs is not radically different from that of complexes, adaptations to the antenna model may be necessary for CP systems. Three new isostructural Ln CPs, [Ln(C15H9O9)3(H2O)3]n (1-Ln = Eu(III), Gd(III), Tb(III)) have been synthesized from benzophenonedicarboxylate (BPDC, Figure 2) and evaluated according to the traditional antenna effect and to the treatment of the CP as a network solid. With comparison of both models, it was found that treating 1-Ln as a network solid with band structure was a more accurate description of its luminescence. Furthermore, analysis of H2O, D2O, and dehydrated analogues of 1-Eu showed that high-energy oscillators enhanced overall luminescence compared to the dehydrated system. With an ever-increasing number of potential applications, understanding the spectroscopic properties of CPs is essential.

EXPERIMENTAL SECTION

Synthesis of Title Compound. The title compound was synthesized by hydrothermal means. A mixture of Eu(NO3)3·5H2O (115 mg, 0.27 mmol) and BPDC (100 mg, 0.37 mmol) was added to DI water (5 mL) and placed in a 23 mL Teflon-lined stainless steel vessel. The vessel was heated at 200 °C for 24 h, 175 °C for 24 h, and then 150 °C for 24 h. The vessel was then removed from heat and left to cool to room temperature where yellow crystals were collected and washed twice with water and ethanol. The starting Ln nitrate salt was dehydrated, and water was substituted with deuterium oxide (5 mL); the reaction vessel was flushed with and sealed under nitrogen for synthesis of the deuterated analogues. The yellow crystals were collected and washed with deuterium oxide and dried under a stream of nitrogen gas. The unmodified sample was placed in a vacuum oven at 250 °C under reduced pressure of 5 in. Hg to remove coordinated aqua ligands to produce the dehydrated sample. The Tb(III) compound was synthesized and treated in an identical manner, though it resisted the formation of crystals suitable for single crystal Xray diffraction (XRD). A powder Gd(III) analogue was also synthesized according to this protocol for band studies. Powder XRD confirmed that the structures of the Tb(III) and Gd(III) compounds are the same as that of 1-Eu (see SI). Crystallography. A single crystal of Eu-BPDC was mounted on a glass fiber for data collection. Reflections were collected at 100 K on a Bruker AXS SMART diffractometer equipped with an APEXII CCD detector using Mo Kα radiation. Data were integrated using the Bruker program SAINT45 and corrected for absorption using the SADABS46 program. The structure of the compound was solved using the SIR9247 program and refined using SHELX-9748 within the WINGX49 software suite. Powder X-ray diffraction experiments were performed with an Olympus BTX II Benchtop XRD. PXRD patterns were compared to those calculated from single crystal data to confirm phase purity. Spectroscopic Measurements. Corrected excitation and emission spectra were collected on a powdered sample using a Horiba Jobin Yvon Spex Fluorolog-3 fluorimeter. The samples were placed into the solid-state sample holder with a quartz cover window and were irradiated at 45° with a 450 W continuous-wave xenon lamp. Quantum yields, lifetimes, and singlet/triplet state values were determined using a Perkin-Elmer LS55 fluorescence spectrometer. Quantum yields were determined at room temperature using the following equation

Φx =

1 − Rx I × ST × ΦST 1 − R ST Ix

(1)

where x is the sample measured and ST is a standard with known Φ. The standard, pyrene (Φ = 53% at 313 nm),50 along with the sample were mixed with a matrix material, PMMA (poly(methyl methacryB

DOI: 10.1021/acs.inorgchem.6b00878 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry late) average molecular weight ∼15 000 by GPC) to reduce the possibility of self-absorbance. The quantum yield of anthracene (Φ = 23%) was determined congruently against the pyrene references with all Ln samples to ensure validity of the quantum yields. Lifetime measurements were collected using the BioLight Fluorescence Application software and analyzed using Origin Pro 8.1. Singlet and triplet states of the organic were measured in 1:1 Gd:BPDC acetonitrile solutions (10−4 M) using the fluorimeter, as outlined by Crosby,51 where emission spectra were recorded at 77 K with time delays of 0−0.05 ms from 250 to 525 nm to distinguish between linker fluorescence and phosphorescence. Singlet and triplet states of 1-Gd were conducted under similar conditions, but at room temperature due to instrumental limitations. FTIR measurements were recorded on the 1-Ln samples on a Thermoscientific Nicolet iS5.

Tb(III) and Gd(III) compounds did not produce crystals suitable for single crystal XRD. The title structure is a twodimensional compound composed of EuO8 polyhedra exhibiting square antiprismatic coordination geometry. Surrounding each Eu(III) ion, there are three monodentate attachments from the BPDC linker’s carboxylate oxygen atoms (O2, O3, O7), a bidentate attachment (O12, O13) from a carboxylate group, and three bound water molecules (O16, O17, O18). The CP contains three crystallographically unique BPDC linkers. Each linker has one carboxylate group interacting with the Eu(III) center. Only one BPDC linker bridges Eu(III) monomers together (O2, O3) down the [001] direction. The second unique BPDC ion has a carboxylate group attached in a bidentate fashion to the Eu(III) center (O12, O13), while the third unique linker attaches in a monodentate fashion to the Eu(III) center from the carboxylate group (O7), leaving carbonyl oxygen atom (O8) uncoordinated. The carboxylate groups not involved in coordination with Eu(III) ions form one-dimensional channels with adjacent linkers through headto-tail hydrogen bonding in the same geometric plane that propagates down the [010] direction. Several hydrogen bonding interactions occur between carboxylate groups from neighboring linkers lining the onedimensional channel as well as from coordinated water ligands. Carboxylate oxygen atoms (O4, O9, O10) interact with oxygen atoms on carboxylate moieties across the channel on neighboring linkers (O5, O15, O14) with distances from 2.596 to 2.608 Å. The carbonyl oxygen atom (O8) interacts with a coordinated water molecule (O16) at a distance of 2.675 Å. One carboxylate oxygen atom (O13) forms a hydrogen bond with a coordinated water molecule at a distance of 2.708 Å. π−π interactions are also present within this compound with centroid distances ranging from 3.682 to 3.788 Å between neighboring linkers. The twisting of the linkers allows for a closer packing distance between aromatic rings. The BPDC linkers are all slightly twisted with torsion angles between adjacent aromatic rings of 42.44°, 43.14°, and 46.70°. This allows for π−π interactions with adjacent linkers, which assist in facilitating the stacking of the Eu monomers into chains that are propagated along [001]. The aromatic rings have interaction distances from 3.682 to 3.788 Å with neighboring linkers. Luminescence Spectroscopy of 1-Ln. 1-Eu and 1-Tb emission spectra (Figure 3) were recorded at 298 K on H2O,



RESULTS AND DISCUSSION Structural Description of 1. A Ln-CP (Ln = trivalent Eu, Gd, Tb) based on benzophenone-4,4′-dicarboxylic acid (BPDC) was synthesized and its structure determined (Figure 2, Table 1). The Eu(III) analogue is described here as the Table 1. Crystallographic Information for 1-Eu [Eu(C15H9O5)3(H2O)3]n fw cryst class space group a (Å) b (Å) c (Å) α (deg) β (deg) γ (deg) Z cell volume (Å3) density (mg m−3) μ (mm−1) temp (K) Rint R1a wR2a GOF total reflns a

R1 = ∑

|| Fo |−| Fc || ; |Fo|

wR2 =

2009.20 monoclinic P21/c 27.981(2) 11.2983(8) 11.8755(9) 90 98.9390(10) 90 4 3708.7(5) 1.799 1.781 100(2) 0.0916 0.0484 0.1232 1.000 8893

(

∑[w(Fo2 − Fc 2)2 ] 2 2

∑[w(Fo ) ]

1/2

)

.

Figure 3. Emission spectra of H2O (black), D2O (blue), and dehydrated (red) samples of 1-Eu (left) and 1-Tb (right). C

DOI: 10.1021/acs.inorgchem.6b00878 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Table 2. Photophysical Measurements of 1-Eu and 1-Tb Eu H2O quantum yield (%) lifetime (ms) BPDC linker

Tb

D2 O

35.5 ± 4.9 58.8 ± 7.1 0.265 ± 0.016 0.906 ± 0.023 1 S (molecule) = 30 580 cm−1 1 S (band) = 23 040 cm−1

dehydrated 26.9 ± 3.1 1.330 ± 0.039

H2O

D2O

dehydrated

6.0 ± 0.3 19.0 ± 1.5 0.280 ± 0.023 0.379 ± 0.030 3 T (molecule) = 23 420 cm−1 3 T (band) = 20 410 cm−1

26.1 ± 3.8 0.613 ± 0.039

Figure 4. Analysis of the ET mechanisms based on lowest-lying 1S/3T values (left; molecular BPDC triplet state shown for reference as a dashed line) and that of donor/acceptor spectral overlap (right; emission spectra at 0.0, 0.3, and 0.5 ms delay with decreasing singlet state energy; 1S/3T band levels noted * for reference on 0.3 ms delay spectrum).

band structure of transition-metal-based CPs is likely a result of hybridization between inorganic and organic bands. The situation is different in Ln CPs. The valence f-orbitals of Ln ions are contracted toward the nucleus, with filled 5s/5p orbitals radially extended beyond the 4f orbitals. This shields 4f orbitals from ligand field effects, resulting in nondirectional, largely ionic metal−organic bonding. Since Ln ions minimally interact with organic-based orbitals, band formation in Ln CPs is likely to be predominantly organic in nature.57 As such, the linkers behave like organic semiconductors in which the spectroscopic properties of the Ln centers in CPs can be treated as dopants within an organic semiconducting matrix.58 Thus, in Ln CPs the current treatment of luminescence being described using the antenna effect with molecular-based antennae as sensitizers is arguably not suitable. Instead, the band nature of the organic linker should be considered. Similar to how inorganic semiconductors sensitize Ln dopants, it is proposed that charge recombination between 1S/3T conduction bands (CB) and the valence band (VB) of the semiconducting organic material results in an ET to the Ln ion instead of radiative recombination. The precise nature of band transitions (direct versus indirect), charge recombination, and charge transport within this and other CP systems is not the focus of this current study, but will certainly need consideration to fully develop this model. To evaluate whether band or molecular sensitization is occurring within 1, time-delayed emission studies of 1-Gd were conducted to determine the energy of the 1S and 3T bands within this system. The 1S band lies at 23 040 cm−1 and the 3T band at 20 410 cm−1, significantly red-shifted from the 1S and

D2O, and dehydrated samples (equimolar concentrations collected during the same experiment after allowing the instrument to fully warm up) with an excitation wavelength of 375 nm. The profile of 1-Eu is characteristic of 5D0 to 7FJ transitions (J = 1−4) at 590, 616, 627, and 710 nm, respectively, of Eu with no discernible organic emission. Interestingly, the dehydrated sample displayed the lowest intensity emission and further manifested in the quantum yields (Table 2). Emission spectra of 1-Tb are characteristic of Tb’s 5 D4 to 7FJ (J = 6−3) transitions at 491, 545, 587, and 623 nm.52 Since BPDC is responsible for sensitizing Ln emission, its 1S and 3T state values were determined. The 3T of molecular BPDC (23 420 cm−1, Table 2) should be ideally suited for efficient ET to Tb(III)’s 5D4 manifold (ΔE = 2990 cm−1), yet ill suited for Eu(III). This should result in a higher quantum yield for 1-Tb than 1-Eu. Tb(III) complexes with a similar triplet state energy have been reported to display quantum yields of ∼45%, and reduce to ∼5% with Eu(III), though there are certainly exceptions based on the spectral distribution of the molecular triplet states.9 What is observed experimentally in 1, however, is the opposite (Table 2); BPDC is a good sensitizer for Eu(III), but a poor sensitizer for Tb(III). While the 3T of the organic has been widely used to describe Ln luminescence in coordination complexes, it may not be suitable for network solids such as CPs. From inorganic network solids to amorphous organic semiconductors, the “infinite” nature of these structures results in band formation. It has been demonstrated that transition-metal-based CPs and MOFs produce bands, though the precise nature of these bands is an ongoing area of research.53−56 Regardless of their nature, the D

DOI: 10.1021/acs.inorgchem.6b00878 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

the two bands in the ∼475−490 nm region. Using these values, the 5D2 manifold of Eu(III) lies within the singlet band, whereas 5D1 and 5D0 reside in the triplet band. This assessment would result in triplet band sensitization of both 5D1 and 5D0, but only the 5D2 manifold could potentially be sensitized by the singlet band. The 5D4 manifold of Tb(III) lies right at the interface between these two bands at ∼20 430 cm−1 (489 nm), but technically resides just inside the triplet band, though it is in the overlap region. On the basis of this, it could be argued that there is triplet sensitization of Tb(III)’s 5D4 manifold, though it would be difficult to conclusively distinguish since the 1S/3T bands do overlap in this region. Regardless of which method is used to deduce and describe the ET mechanism that is occurring (the “classic” antenna description based on discrete 1S/3T values or that based on spectral overlap), it is clear that sensitization by the band nature of the organic rather than its individual molecular structure is necessary when describing the luminescence of Ln CPs and MOFs. However, it should be stressed that the proposed “band model” is still considered to be the antenna effect, but sensitization occurs through a fundamentally different process involving the band nature of CPs and MOFs. Next, the high quantum yield of Eu(III) in such a highly hydrated CP was examined in more detail. Compounds 1-Eu and 1-Tb were synthesized in D2O to replace the high-energy OH oscillators with lower-energy OD. The presence of highenergy oscillators is commonly used to explain poor luminescence in Ln compounds, but their influence may not affect Ln CPs similarly due to structural attributes (e.g., extensive, strong hydrogen bonding network; lack of dynamic ligand exchange; etc.) Thus, it is feasible that their influence on Ln CP luminescence differs from that of solution-based complexes. As expected, luminescence in both systems improved in terms of both emission quantum yields and lifetimes upon substituting H2O with D2O (Table 2). Next, the aqua ligands were removed from the compounds. After verifying no structural change or crystallinity loss via powder XRD, luminescence was expected to improve even more. Being a measure of NR decay, lifetimes improved in both systems, indicating that the OH oscillator decay route had been removed.21 The quantum yield of 1-Tb increased as anticipated, but actually decreased in 1-Eu. These results were consistently reproducible among various samples and experimental trials, and give insight into the role of high-energy oscillators in sensitized Ln emission in CPs. Upon replacement of the OH oscillators with OD, the excited state lifetimes of the Eu and Tb CPs increased as expected. Using Horrock’s14 and Choppin’s59 equations (Table 3, eqs 2 and 3), the calculated number of bound aqua ligands matches within error those found in the crystal structure,

3

T state energies of the BPDC molecular antenna (Table 2). If an energy difference greater than 5000 cm−1 is needed for efficient ISC as is typically required in molecular systems,10 then ISC is not likely efficient within 1. Emission studies support this since 1S band emission is observed even after long detector delays (τ = 0.27 ms for the singlet band at 450 nm emission; 0.45 ms for the triplet band at 515 nm emission). This holds a specific consequence for 1 in that sensitization from the 1S band will be proposed to partially explain the luminescence within 1. In resonant ET as occurs in sensitized Ln systems, there is a slight disparity between the energy of the donor and acceptor, where the acceptor is lower in energy. Assuming this energy difference is comparable to Ln complexes,9 a difference of ∼2500−4000 cm−1 should resonate well with a given f-excited state manifold. With a 1S band of 23 040 cm−1, BPDC can resonate with 5D1 of 1-Eu (ΔE 1Sband−5D1 = 4020 cm−1), followed by internal conversion (IC) to 5D0 before emitting to the ground state (Figure 4). Similarly, the 3T band is able to resonate with 5D0 (energy difference of 3160 cm−1) with an overall quantum yield of 36% for 1-Eu. Thus, the combination of poor 1S/3T band ISC and ideally matched 1S/3T band energies for energy transfer to both 5D1/5D0 manifolds (respectively) is believed to result in the quantum yield for 1Eu, though the contribution of the 1S band is partially speculative at this point. For 1-Tb, the triplet band lies too low in energy to transfer energy to Tb(III)’s emitting 5D4 manifold. The 1S band, on the other hand, is ideally suited for ET to 5D4 (ΔE = 2610 cm−1). Complexes with similar 3T energy values have shown quantum yields of ∼40%. With a 6% quantum yield in 1-Tb, it appears that the ability of the 1S band to sensitize Ln emission is not as effective as an ideally suited 3T band, but the observed emission does further support the argument that some singlet band sensitization could be occurring in this system. In molecular systems, emission from 1S to the ground state is an allowed process, often happening faster than ET can occur to a Ln ion. Due to the heavy atom effect, ISC can compete with fluorescence to populate 3T of the organic component followed by ET to the Ln. In semiconductors, the situation is more complex. Aside from spin and parity selection rules, network solids also need to consider crystal momentum (k), whose selection rule states that a change in k between two states is forbidden. Thus, even though fluorescence from 1S of a molecule is allowed, analogous emission from a 1S band in the corresponding solid may not obey the momentum selection rule, resulting in a 1S band that may be long-lived enough for ET to a Ln ion. Although this pathway may be feasible in CPs, it may still be less effective than sensitization by the 3T band. As such, 1S sensitization of 5D4 combined with NR decay induced by the bound aqua ligands is believed to result in the 6% quantum yield of 1-Tb. These proposed ET mechanisms in 1-Eu and 1-Tb are based on a simplified qualitative assessment of the singlet/triplet band energies that is commonly used in the lanthanide literature, which assumes that the lowest-lying energy state is mostly involved in the ET process. Another analysis considers the entire spectral distribution of both the singlet and triplet bands where ET occurs when there is overlap between the donor’s (BPDC) emission spectrum and acceptor’s (Ln ion) absorption spectrum. Using this, the singlet band covers ∼20 620−23 810 cm−1 (420−485 nm), and the triplet band covers ∼15 380− 20 620 cm−1 (485−650 nm), though there is likely overlap of

Table 3. Bound Water Determination Results for the Title Compound qLn

1-Eu

1-Tb

2.81a 3.26b

3.91a,c

a Obtained using eq 2. bObtained using eq 3; there is currently no equivalent of eq 3 for Tb systems. cThe number of bound water molecules is within experimental error of the lifetime studies (approximately ±1/2 water molecule).

E

DOI: 10.1021/acs.inorgchem.6b00878 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

entire integrated Eu emission spectrum from 5D0−7F4 and the integrated magnetic-dipole emission band, 5D0 → 7F1. On the basis of the luminescence parameters of 1-Eu, the maximum quantum yield possible in this system once all OH/ OD oscillators are removed is 26% (Table 4). From κnonrad, assuming 100% ηsens, the maximum quantum yields possible in the hydrated system are 5%, and 17% in the deuterated. With significantly higher quantum yields in the OH/OD systems, it is deduced that these oscillators are somehow enhancing the sensitization process. When these oscillators are removed, 1-Eu luminescence is reduced. Contrary to 1-Eu, 1-Tb behaves as expected with maximum quantum yield for the dehydrated sample and lowest for the hydrated due to NR deactivation of the 5D4 manifold. From Table 4, it is seen that κrad values between the OH/ OD/dehydrated compounds are all nearly identical, and that there is a steady decrease in κnonrad as the OH oscillator is replaced by OD and then removed altogether. This is expected as the NR decay mechanism becomes progressively less efficient, but when the ηsens is calculated, the values seem implausible for the OH and OD systems. When dehydrated, BPDC transfers its energy to the Eu center, causing what should be the maximum quantum yield feasible within this system. However, eqs 4−7 assume a correlation between κnonrad and ηsens. This neglects enhancement of the sensitization process by the very oscillators that simultaneously contribute to κnonrad. Thus, ηsens values in the OH and OD systems appear nonsensical at well above 100%. However, these values seem to be the combination of BPDC’s intrinsic ηsens and the OH/OD oscillators’ ability to enhance luminescence.

indicating that these high-energy oscillators do indeed influence the f-excited state lifetimes within Ln-based CPs in a similar fashion to complexes. These equations were determined empirically from solution experiments, whereas in solid-state materials such as CPs the number of bound water molecules is precisely known from crystallographic data, thus providing empirical validation of these equations. This shows that these extended systems while different, can still be assessed using these equations derived from solution studies. qln = Aln (τH−21O − τD−21O)

(2)

qln = 1.05 × τH−21O − 0.70

(3)

In both complexes and CPs, such oscillators are detrimental to luminescence efficiency, as demonstrated by the increase of quantum yields upon OH to OD replacement in 1. Yet, for 1Eu, the high-energy OH/OD oscillators appear to have an overall enhancement effect on the luminescence that exceeds their concurrent NR quenching effects since the OH/OD systems have higher quantum yields than the dehydrated (Figure 3 and Table 2). It is postulated that the OH/OD oscillators somehow assist with the IC from 5D1 to 5D0 since their energy difference is on the order of these oscillators. This has been observed in other systems upon dehydration,38,44 and this could possibly explain such behavior. It is uncertain if this phenomenon is (a) unique to 1S sensitization, (b) unique to ET to 5D1 only, (c) possible through 3T band sensitization also, or (d) possible with manifolds higher than 5D1. This is currently under investigation to study the generality of this enhancement and to deduce the mechanism of enhancement. The emission spectra of the 1-Eu compounds can provide important information regarding radiative (κrad) and nonradiative (κnonrad) rates within Ln−organic systems, which offers insight into the luminescence of 1. The intrinsic quantum yield of Eu(III) (ΦEu) can be determined, as well as the efficiency of the organic material to sensitize Ln emission (ηsens), using the following equations (Table 4):44,60,61 ϕTOT = ϕEu ·ηSens

(4)

τobs τrad

(5)

⎛I ⎞ 1 = An3⎜ TOT ⎟ τrad ⎝ IMD ⎠

(6)

ϕEu =

κ nonrad = κobs − κ rad =

1 1 − τobs τrad



CONCLUSION A new Ln CP is reported that displays luminescent features that indicate a reevaluation of the current methods used to explain sensitized emission in CPs is warranted. The luminescence of 1-Eu and 1-Tb have been scrutinized by both the currently used “antenna model” and the proposed “band model”. In these systems, the band model describes the observed luminescence better. Instead of arising through ET from a molecular antenna, sensitization occurs from ET from the 1S/3T conduction bands and the valence band of the organic matrix. Details of this mechanism require further study, but data indicate that the band nature of the organic linker is better in describing luminescence of Ln CPs. The influence of high-energy oscillators (OH/OD) was also studied, and it was observed that they enhance the luminescence of 1-Eu with investigations into this mechanism ongoing.

(7)

The following abbreviations apply: ΦTOT is the quantum yield of the system; τobs and τrad are the experimental and radiative lifetimes, respectively; κobs/rad/nonrad are the observed/radiative/ nonradiative decay rates; A is the spontaneous emission probability of the 5D0 → 7F1 transition (14.65 s−1); n is the refractive index (∼1.5 for these solids); and ITOT/MD is the



S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.6b00878.

Table 4. Luminescence Parameters of the 1-Eu H2O, D2O, and Dehydrated Samples compd

ηsens

ΦEu

τrad

κrad

κnonrad

H2O D2 O dehydrated

697% 354% 102%

5.1% 16.6% 26.4%

0.0052 s 0.0055 s 0.0050 s

193.67 s−1 183.13 s−1 198.60 s−1

3579.9 s−1 920.6 s−1 553.5 s−1

ASSOCIATED CONTENT

Further experimental details, crystallographic data, thermogravimetric analysis, PXRD, excitation and emission spectra, DRUV−vis spectra, and infrared spectroscopy (PDF) Crystallographic information in CIF format (CCDC 1438661) (CIF) F

DOI: 10.1021/acs.inorgchem.6b00878 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry



(27) Zhou, Y.; Guo, Y.; Xu, S.; Zhang, L.; Ahmad, W.; Shi, Z. Inorg. Chem. 2013, 52, 6338−6345. (28) Gai, Y.; Jiang, F.; Chen, L.; Wu, M.; Su, K.; Pan, J.; Wan, X.; Hong, M. Cryst. Growth Des. 2014, 14, 1010−1017. (29) Matthes, P. R.; Nitsch, J.; Kuzmanoski, A.; Feldmann, C.; Steffen, A.; Marder, T. B.; Müller-Buschbaum, K. Chem. - Eur. J. 2013, 19, 17369−17378. (30) Che, G.-B.; Liu, S.-Y.; Zhang, Q.; Liu, C.-B.; Zhang, X.-J. J. Solid State Chem. 2015, 225, 378−382. (31) Zhang, H.-B.; Peng, Y.; Shan, X.-C.; Tian, C.-B.; Lin, P.; Du, S.W. Inorg. Chem. Commun. 2011, 14, 1165−1169. (32) Ma, X.; Li, X.; Cha, Y.-E.; Jin, L.-P. Cryst. Growth Des. 2012, 12, 5227−5232. (33) Monteiro, J. H. S. K.; Formiga, A. L. B.; Sigoli, F. A. J. Lumin. 2014, 154, 22−31. (34) Bo, Q.-B.; Zhang, H.-T.; Miao, J.-L.; Zhang, Z.-W. Eur. J. Inorg. Chem. 2013, 2013, 5631−5640. (35) Wang, Y.; Xue, Z.; Sun, Z.; Tan, C.; Wen, Y.; Hu, S.; Zhu, Q.; Sheng, T.; Wu, X. Dalton Trans. 2015, 44, 2217−2222. (36) Yang, Q.-Y.; Pan, M.; Wei, S.-C.; Hsu, C.-W.; Lehn, J.-M.; Su, C.-Y. CrystEngComm 2014, 16, 6469−6475. (37) Wei, X.-H.; Yang, L.-Y.; Liao, S.-Y.; Zhang, M.; Tian, J.-L.; Du, P.-Y.; Gu, W.; Liu, X. Dalton Trans. 2014, 43, 5793−5800. (38) Zhou, X.; Li, H.; Xiao, H.; Li, L.; Zhao, Q.; Yang, T.; Zuo, J.; Huang, W. Dalton Trans. 2013, 42, 5718−5723. (39) Luo, Y.; Zheng, Y.; Calvez, G.; Freslon, S.; Bernot, K.; Daiguebonne, C.; Roisnel, T.; Guillou, O. CrystEngComm 2013, 15, 706−720. (40) Zhang, H.; Zhou, L.; Wei, J.; Li, Z.; Lin, P.; Du, S. J. Mater. Chem. 2012, 22, 21210−21217. (41) Cepeda, J.; Balda, R.; Beobide, G.; Castillo, O.; Fernández, J.; Luque, A.; Pérez-Yáñez, S.; Román, P. Inorg. Chem. 2012, 51, 7875− 7888. (42) Bo, Q.-B.; Zhang, H.-T.; Wang, H.-Y.; Miao, J.-L.; Zhang, Z.-W. Chem. - Eur. J. 2014, 20, 3712−3723. (43) Sobieray, M.; Gode, J.; Seidel, C.; Poß, M.; Feldmann, C.; Ruschewitz, U. Dalton Trans. 2015, 44, 6249−6259. (44) Daiguebonne, C.; Kerbellec, N.; Guillou, O.; Bünzli, J.-C.; Gumy, F.; Catala, L.; Mallah, T.; Audebrand, N.; Gérault, Y.; Bernot, K.; Calvez, G. Inorg. Chem. 2008, 47, 3700−3708. (45) SAINT. SAINT, 5.053 ed.; Bruker Analytical X-Ray Systems, Inc.: Madison, WI, 1998. (46) Sheldrick, G. M. SADABS, 2.03 ed.; University of Gottingen: Gottingen, Germany, 2002. (47) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A. J. J. Appl. Crystallogr. 1993, 26, 343−350. (48) Sheldrick, G. M. SHELX97; Institüt für Anorganische Chemie der Universität: Göttingen, Germany, 1998. (49) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837−838. (50) Bril, A.; de Jager-Veenis, A. W. J. Res. Natl. Bur. Stand., Sect. A 1976, 80A, 401−407. (51) Crosby, G. A.; Whan, R. E.; Alire, R. M. J. Chem. Phys. 1961, 34, 743−748. (52) Binnemans, K.; Gorller-Walrand, C. Chem. Phys. Lett. 1995, 235, 163−174. (53) Alvaro, M.; Carbonell, E.; Ferrer, B.; Xamena, F. X. L. i.; Garcia, H. Chem. - Eur. J. 2007, 13, 5106−5112. (54) Silva, C. G.; Corma, A.; García, H. J. Mater. Chem. 2010, 20, 3141−3156. (55) Silva, C. G.; Corma, A.; García, H. J. Mater. Chem. 2010, 20, 3141−3156. (56) Xamena, F. X. L. i.; Corma, A.; Garcia, H. J. Phys. Chem. C 2006, 111, 80−85. (57) Aspinall, H. C. Chemistry of the f-Block Elements; CRC, 2001. (58) Liu, Y.; Luo, W.; Zhu, H.; Chen, X. J. Lumin. 2011, 131, 415− 422. (59) Choppin, G. R.; Peterman, D. R. Coord. Chem. Rev. 1998, 174, 283−299.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest. CCDC 1438661 can also be obtained free of charge from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac. uk/data_request/cif.



ACKNOWLEDGMENTS The authors would like to thank Florida Atlantic University for funding (in particular FAU’s office of Undergraduate Research and Inquiry) and the NSF (BCC; MRI Grant No. 0922931). The authors also thank Dr. Ana de Bettencourt-Dias of the University of Nevada, Reno, for spectrometer use.



REFERENCES

(1) de Bettencourt-Dias, A.; Barber, P. S.; Viswanathan, S.; de Lill, D. T.; Rollett, A.; Ling, G.; Altun, S. Inorg. Chem. 2010, 49, 8848−8861. (2) Bünzli, J.-C. G. Chem. Rev. 2010, 110, 2729−2755. (3) Cui, Y.; Yue, Y.; Qian, G.; Chen, B. Chem. Rev. 2012, 112, 1126− 1162. (4) Bünzli, J.-C. G. Acc. Chem. Res. 2006, 39, 53−61. (5) Bünzli, J.-C. G.; Piguet, C. Chem. Soc. Rev. 2005, 34, 1048−1077. (6) Cui, Y.; Chen, B.; Qian, G. Coord. Chem. Rev. 2014, 273−274, 76−86. (7) Li, B.; Wen, H.-M.; Cui, Y.; Qian, G.; Chen, B. Prog. Polym. Sci. 2015, 48, 40−84. (8) Chen, T.; Ma, S. Rev. Inorg. Chem. 2012, 32, 81−100. (9) Dexter, D. L. J. Chem. Phys. 1953, 21, 836−850. (10) Faulkner, S.; Pope, S. J. A. J. Am. Chem. Soc. 2003, 125, 10526− 10527. (11) Chandler, B. D.; Cramb, D. T.; Shimizu, G. K. H. J. Am. Chem. Soc. 2006, 128, 10403−10412. (12) Latva, M.; Takalo, H.; Mukkala, V.-M.; Matachescu, C.; Rodriguez-Ubis, J. C.; Kankare, J. J. Lumin. 1997, 75, 149−169. (13) Steemers, F. J.; Verboom, W.; Reinhoudt, D. N.; van der Tol, E. B.; Verhoeven, J. W. J. Am. Chem. Soc. 1995, 117, 9408−9414. (14) Horrocks, W. D. J.; Sudnick, D. R. J. Am. Chem. Soc. 1979, 101, 334−340. (15) Greig, N. E.; Einkauf, J. D.; Clark, J. M.; Corcoran, E. J.; Karram, J. P.; Kent, C. A.; Eugene, V. E.; Chan, B. C.; de Lill, D. T. J. Solid State Chem. 2015, 225, 402−409. (16) Ramirez, A. L.; Knope, K. E.; Kelley, T. T.; Greig, N. E.; Einkauf, J. D.; de Lill, D. T. Inorg. Chim. Acta 2012, 392, 46−51. (17) Zou, S.; Li, Q.; Du, S. RSC Adv. 2015, 5, 34936−34941. (18) Jia, L.; Hui, Y.-C.; Li, Z.; Sun, H.-L.; Wang, Z. CrystEngComm 2014, 16, 6483−6490. (19) Huang, Y.; Zhu, Q.; Sheng, T.; Hu, S.; Fu, R.; Shen, C.; Tan, C.; Wen, Y.; Bai, S.; Wu, X. CrystEngComm 2013, 15, 3560−3567. (20) Zhang, S.-R.; Du, D.-Y.; Tan, K.; Qin, J.-S.; Dong, H.-Q.; Li, S.L.; He, W.-W.; Lan, Y.-Q.; Shen, P.; Su, Z.-M. Chem. - Eur. J. 2013, 19, 11279−11286. (21) Stein, G.; Wurzberg, E. J. Chem. Phys. 1975, 62, 208−213. (22) Horrocks, W. D., Jr.; Sudnick, D. R. Acc. Chem. Res. 1981, 14, 384−392. (23) Allendorf, M. D.; Bauer, C. A.; Bhakta, R. K.; Houk, R. J. T. Chem. Soc. Rev. 2009, 38, 1330−1352. (24) Rocha, J.; Carlos, L. D.; Almeida Paz, F. A.; Ananias, D. Chem. Soc. Rev. 2011, 40, 926−940. (25) Haquin, V.; Etienne, M.; Daiguebonne, C.; Freslon, S.; Calvez, G.; Bernot, K.; Le Polles, L.; Ashbrook, S. E.; Mitchell, M. R.; Bünzli, J.-C.; Eliseeva, S. V.; Guillou, O. Eur. J. Inorg. Chem. 2013, 2013, 3464−3476. (26) Zhang, S.; Yang, Y.; Xia, Z.-Q.; Liu, X.-Y.; Yang, Q.; Wei, Q.; Xie, G.; Chen, S.-P.; Gao, S.-L. Inorg. Chem. 2014, 53, 10952−10963. G

DOI: 10.1021/acs.inorgchem.6b00878 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry (60) Shavaleev, N. M.; Eliseeva, S. V.; Scopelliti, R.; Bünzli, J.-C. G. Inorg. Chem. 2015, 54, 9166−9173. (61) Werts, M. H. V.; Jukes, R. T. F.; Verhoeven, J. W. Phys. Chem. Chem. Phys. 2002, 4, 1542−1548.

H

DOI: 10.1021/acs.inorgchem.6b00878 Inorg. Chem. XXXX, XXX, XXX−XXX