Deuteration of Perylene Enhances Photochemical Upconversion

Page 1 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

Deuteration of Perylene Enhances Photochemical Upconversion Efficiency Andrew Danos,† Rowan W. MacQueen,† Yuen Yap Cheng,† Miroslav Dvoˇr´ak,‡,k Tamim A. Darwish,¶ Dane R. McCamey,§ and Timothy W. Schmidt∗,† School of Chemistry, UNSW Sydney, NSW 2052, Australia, Department of Physical Electronics, Faculty of Nuclear Sciences and Physical Engineering, Czech Technical University in Prague, V Holesovickach 2, 180 00 Prague, Czech Republic, National Deuteration Facility, Bragg Institute, Australian Nuclear Science and Technology Organisation, Locked Bag 2001, Kirrawee DC, NSW 2232, Australia, and School of Physics, UNSW Sydney, NSW 2052, Australia E-mail: [email protected]

∗

To whom correspondence should be addressed School of Chemistry, UNSW Sydney, NSW 2052, Australia ‡ Department of Physical Electronics, Faculty of Nuclear Sciences and Physical Engineering, Czech Technical University in Prague, V Holesovickach 2, 180 00 Prague, Czech Republic ¶ National Deuteration Facility, Bragg Institute, Australian Nuclear Science and Technology Organisation, Locked Bag 2001, Kirrawee DC, NSW 2232, Australia § School of Physics, UNSW Sydney, NSW 2052, Australia k School of Chemistry, The University of Sydney, NSW 2006, Australia †

1 ACS Paragon Plus Environment


Abstract Photochemical upconversion via triplet-triplet annihilation is a promising technology for improving the efficiency of photovoltaic devices. Previous studies have shown that the efficiency of upconversion depends largely on two rate constants intrinsic to the emitting species. Here we report that one of these rate constants can be altered by deuteration, leading to enhanced upconversion efficiency. For perylene, deuteration decreases the first order decay rate constant by 16±9 % at 298 K, which increases the linear upconversion response by 45±21% in the low excitation regime. The effect is explained by a balance between the competing effects of changing the density of states and Frank-Condon factors relevant to intersystem crossing.

TOC Graphic (in final size).

H

decay rate (s–1)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

H H

H

H

H

H

H H

H H

H

D

D D

D

D

D

D

D D

D D

D

The global transition towards renewable energy is the most pressing environmental and moral imperative of the 21st century. Only direct solar collection has the capacity to satisfy the large and growing global energy demand in its own right, although it is more commonly used to complement other renewable energy sources. 1,2 At present, the two preeminent technologies for solar collection are photovoltaics (PV) and solar thermal. The widespread deployment of both technologies is currently limited by their higher cost of electricity production compared with traditional combustion of fossil fuels, although price parity is rapidly approaching in certain contexts. 3,4 These costs are largely attributed to fixed manufacture and installation, and so the overall cost of electricity generated can be reduced by improving operational 2 ACS Paragon Plus Environment

Page 2 of 21

Page 3 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(a.)

(b.)

TPTBPt N

perylene

N Pt

N

S1

S1

N

k2

2. ISC T1

T1

T1

T1

5. 3. TET

k1

1.

S0

4. TTA

S0

S0

S0

Figure 1: The mechanism of TTA-UC. (a.) 1. Photons are absorbed to take sensitizers from the ground to the excited singlet state. 2. Intersystem crossing (ISC) takes sensitizers to their lowest triplet state. 3. The triplet energy is transferred to the emitter molecule, which may undesirably decay with a rate constant k1 . (b.) 4. Emitter triplets annihilate to produce a higher energy singlet with rate constant k2 . 5. Higher energy photons are emitted. efficiency. The electrical conversion efficiency of PV devices, and consequently the cost of the electricity they produce, is limited by the optical properties of the absorbing material, usually a semiconductor. 5,6 Photons with energy below the bandgap of the absorbing material are unable to generate photocurrent, which for silicon represents 20% of incoming solar photons. 7 One way to access and harness the energy of these below-bandgap photons is photochemical upconversion (UC) via triplet triplet annihilation (TTA). 8–14 Significant progress has been made in recent years to improve PV applications of UC, 8,15–19 to develop integrated UC-PV devices, 20 and to better understand the underlying TTA process. 21–25 3 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 21

The process by which sub-bandgap photons can be reclaimed by UC is shown in Figure 1. Low energy photons are absorbed by a sensitizing species, which is chosen such that it undergoes efficient intersystem crossing (ISC) to the first excited triplet state (T1 ). Excitation is then transferred to an emitting species via triplet energy transfer (TET), a Dexter process which preserves multiplicity. 26–28 In this way, sustained excitation of the sensitizer population can generate a population of excited T1 emitters. When two excited emitter molecules undergo TTA, their total energy is redistributed to promote one molecule to the first excited singlet state (S1 ) and demote the other to the ground state (S0 ). The TTA process is spin allowed as it preserves the overall spin of the pair. The promoted emitter molecule rapidly fluoresces to produce a photon of higher energy than those initially absorbed by the sensitizers. These upconverted photons can be utilised by PV devices, adding to their photocurrent and improving their electrical conversion efficiency. 8,15,18,20 Previous studies have shown that the efficiency of UC depends on two rate constants intrinsic to the emitter molecule. 22,25 The kinetics of UC is governed by the rate equation: √ d[T] d IU C ∝ = kφ [S] − k1 [T] − k2 [T]2 dt dt

(1)

where IU C is the intensity of UC emission, [T] is the concentration of emitter molecules in the T1 state, [S] is the concentration of sensitizer molecules in their S0 state, kφ is the rate constant of excitation of sensitizer molecules, and k1 and k2 are the first and second-order decay rate constants for excited emitters, respectively. Under low excitation and steady state conditions (relevant to solar applications and where first-order decay dominates), the rate equation for IU C yields IU C

k2 kφ2 [S]2 ηΦP L ∝ 2k12

(2)

where η is the efficiency of the TTA process for two excited emitters to produce an excited singlet, and ΦP L is the fluorescence quantum yield of the emitter. 29 Equation 2 provides the roadmap for improving the PV enhancing capabilities of UC


Page 5 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


devices. Significant research effort has been made to identify and synthesize high performance emitters (those with high k2 , ΦP L , and η; and low k1 ), although a rational approach to improving each of these parameters simultaneously has not yet been developed. Previous studies of polycyclic aromatic hydrocarbons (PAHs) such as perylene in low temperature crystals or solids have shown that deuteration reduces k1 , as measured by the phosphorescence lifetime. 30–32 In research and in applications, small PAHs such as perylene make up the majority of UC emitters due to their chemical robustness and desirable UC parameters. 9,13,14

For this reason, confirming that deuteration causes an increase in UC performance by

reducing k1 is of general interest and wide applicability.

Experimental The hydrogen isotopologue of perylene was purchased from L. Light & Co. Ltd., while perylene-d12 with 98% isotopic purity was purchased from Sigma Aldrich. The sensitizer used for both emitters was Pt(II) meso-tetraphenyl tetrabenzoporphine (TPTBPt), purchased from Frontier Scientific. Upconverting solutions with 1 mM of either perylene isotopologue and 0.1 mM TPTBPt were prepared in toluene for the collection of action spectra. For UC kinetics, the concentrations were 2.5 mM of emitter and 0.25 mM of sensitizer. These solutions are referred to as P0 and P98 according to their degree of emitter deuteration. In order to ensure identical concentrations of sensitizer in comparative samples, the solutions were prepared by dispensing equal volumes from a common sensitizer stock solution. Each sample was rigorously deoxygenated by three freeze-pump-thaw cycles at 3×10−3 mbar and 77 K before measurement. Action spectra were collected using a lock-in pump-probe method described in Ref. 29 Perylene fluorescence was detected at 470 nm and repeated at several pump powers. The pump beam itself was passed through a 570 nm long-pass filter so as not to cause any direct excitation of the emitter. The experimental conditions were identical between P0 and P98



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

UC samples. Time-resolved photoluminescence spectra were collected in the same way as previous kinetic studies. 24 An Acton/Princeton iCCD spectrograph/camera was used to collect UC fluoresence following excitation of the sample by the 615 nm output of a TOPAS OPA pumped with a Clark MXR CPA 2210 femtosecond laser running at 1 kHz repetition rate. The electronic gating of the CCD was set to acquire in 1 µs intervals from 100 ns to 900.1 µs after the laser pulse, and to collect for 1000 shots at each time-point. Measurements were repeated at several pulse energies in the range of 150 nJ to 12 µJ using neutral density filters to attenuate the excitation.

Results and Discussion Action spectra The raw action spectra were processed by scaling the P0 traces up so that the direct fluorescence response peak at 445 nm was the same height for P0 and P98 for scans with the same pump power. Since the electronic properties of perylene are unaffected by deuteration (experimentally confirmed by their identical absorbance and fluorescence spectra, included in the Supporting Information), we expect this peak to be the same for both emitters, and attribute these small scaling factors to variability in detector sensitivity. The scaling factors are tabulated in the Supporting Information, with none larger than 4%. As the P0 traces were scaled up to the P98 traces, the effect of this correction is to make the reported relative efficiency enhancement a conservative underestimate. A linear baseline correction was applied across the UC response region of each spectrum (530 nm to 660 nm), and the response in this region then integrated to give a single number for direct comparison. A comparison between typical action spectra for P0 and P98 at a single pump power is shown in Figure 2, while Figure 3 compares the action spectra integrals at a range of pump powers. For the lock-in pump-probe action spectrum measurements, signal is proportional to the 6 ACS Paragon Plus Environment

Page 6 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(arb.)

Page 7 of 21

Figure 2: Action spectra of P0 and P98 samples taken under identical conditions. The large response from direct emitter excitation is seen at 450 nm, while the smaller UC response at 610 nm is enlarged in the inset.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3: Integrals of action spectra UC response region at different pump powers (spot size ≈12 mm2 ). Dashed lines show the fit of a model based on Equation 1. The efficiency of P98 under 160 mW pumping is about 7.5 %.


Page 8 of 21

Page 9 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


derivative of the UC intensity (from Equation 2) with respect to the probe beam intensity: dIU C d M∝ = dIpr dIpr

k2 kφ2 [S]2 ηΦP L 2k12

! (3)

where M is the action spectrum signal and Ipr is the intensity of the probe beam. An expression for the gradients in Figure 3 can then be derived by splitting kφ into excitation components due to the pump and probe beams, and evaluating the derivative: 2

dM d ∝ dIPu dIPu dIpr ∝

k2 (kφPu + kφpr )2 [S]2 ηΦP L 2k12

!

k2 [S]2 ηΦP L d2 (IPu σPu + Ipr σpr )2 2k12 dIPu dIpr

(4)

k2 [S]2 ηΦP L ∝ σPu σpr k12 where IPu is the pump intensity, with IPu Ipr , and σPu and σpr are the absorption crosssections of the sensitizer at the respective wavelengths of the pump and probe beams. For P0 and P98 samples measured under identical conditions, all of the experimental and many of the molecular parameters are assumed to be the same, and so the expression reduces to the equation of a straight line: dM κ = 2 dIPu k1

(5)

where κ is a combination of constants which are not expected to change under deuteration. Since this derivation is based on Equation 2, it too only applies in the low excitation regime. A similar analysis can be performed on Equation 1 to determine the shapes of the curves upon which the action spectra integrals lie, as shown in Figure 3. 33 In accordance with Equation 5, the experimental action spectra integrals grow linearly with pump power in the low excitation regime. Gradients were calculated using the data in Figure 3 over the range of low pump powers, up to 10 mW. These gradients correspond to the empirical relative efficiencies of the P0 and P98 UC solutions. Since the only difference



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

between these samples is the emitter species, we attribute the change in the efficiency to enhancement by deuteration. The P0 sample had a gradient of (30 ± 7) × 10−4 while the P98 sample had a gradient of (44 ± 1) × 10−4 (95% confidence intervals, arbitrary units), a 46% relative efficiency enhancement due to deuteration. Using Equation 5, the 46% increase in UC efficiency can be equated to a 17% decrease in the k1 parameter of P98. This figure is in close agreement with the fitted parameters used to generate the modelled lines in Figure 3 (derived from Equation 1), and also with direct measurement of k1 for P0 and P98 samples by photoluminescence kinetics, discussed below. We observe that the deuteration enhancement effect in steady state conditions diminished with increasing excitation power. The deuteration enhancement in the raw integrals falls to 18 % at 30 mW , while at 160 mW it is just 14 %. Extending the model curve indicates that deuteration should have no enhancement effect at very high powers. This is because when [T] is large, Equation 2 no longer applies, and instead the kinetics of UC are dominated by the second-order term k2 [T]2 . This situation corresponds to triplet pairs undergoing UC faster than first-order processes can quench excitation. In the high excitation limit, the UC response depends only on kφ [S] and ηΦP L , and is insensitive to k1 and therefore also deuteration. Nevertheless, in the low excitation conditions relevant to PV applications we expect deuteration enhancement to take effect. Experimentally, the regime change from low to high excitation is evident in the final three points of Figure 3. The model curve and data indicate that P98 enters the high power regime, where TTA is more efficient, at a lower threshold. Although determining the value of the threshold from the data is somewhat subjective, given the higher initial gradient of P98 and a common high power asymptote, mathematically P98 must enter the high power regime at a lower threshold, to allow P0 to “catch up”. This observation is also consistent with a decrease in k1 , as the threshold intensity can be shown to be proportional to (k1 )2 . 34

This lower threshold to higher efficiency operation makes the deuteration reduction in k1

particularly valuable for applications of UC devices.


Page 10 of 21

Page 11 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Photoluminescence kinetics Raw photoluminescence spectra were processed by integrating from 461 nm to 491 nm on each time-slice, where perylene UC fluorescence reabsorption by the sensitizer is minimal. The resulting traces were then normalised to the initial timepoint, and a square root applied to convert kinetics of IU C into kinetics of [T]. 24 A typical set of unprocessed spectra is included in the Supporting Information. The traces were then fit to the form: 1−β [T]t = A k1 t [T]0 e −β

(6)

where [T]t is the concentration of excited emitter triplets at time t, and A and β are fitting parameters. 24 The parameter β corresponds to the initial fraction of triplet decay which occurs through second-order processes, although this fraction decreases as time progresses and [T] diminishes. From β and the excitation pulse energy, k2 can also be compared between datasets:

k2 =

1 βk1 [T]0 1 − β

[T]0 = BEL

(7) (8)

where EL is the laser pulse energy, and B is a conversion factor (close to unity) between laser photons and excited triplets. 24 Since the P0 and P98 samples had the same [S], for the purposes of comparing k2 between samples we also assume they have the same value of B.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4: Comparison of kinetic rate constants for P0 and P98 samples. Error bars represent 95% confidence intervals.

The UC parameters found for P0 and P98, with k1 and k2 averaged across pulse energies, are summarised in Figure 4. Since upconversion is a collisional process, its kinetics (quantified by k2 ) are limited by the diffusion of emitter molecules and the intrinsic rate of TTA. 22 Deuteration does not change the geometry or chemical solvent interactions of perylene, and so its rate of diffusion should not change under deuteration. The TTA process is near-resonant and non-radiative, and so is insensitive to any molecular vibrations which deuteration might alter. Experimentally, despite large confidence intervals, the Bk2 values calculated for P0 and P98 support the assertion that k2 does not change for deuterated perylene. This result justifies the inclusion of k2 in κ in Equation 4, and the conclusion that deuteration enhancement of UC is a consequence of changing k1 . The values of k1 lie outside each other’s confidence intervals, indicating that deuteration indeed changes this molecular parameter. The change itself is a 16.7% decrease for P98, which is in close agreement with the ratios calculated using action spectra modelling and Equation 4.


Page 12 of 21

Page 13 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Discussion The dominant process which allows first-order decay of excited triplets, and therefore determines the size of k1 , is ISC from the T1 state to an excited vibrational level on the S0 manifold, followed by rapid thermalisation back to the ground state. 22,35 The spin-orbit coupling that allows the otherwise spin forbidden ISC process has been long studied in small PAHs, and its rate can be estimated by applying the Fermi golden rule. The coupling matrix element contains numerous molecular Frank-Condon factors (FCFs) for the vibrational states which can participate in the ISC transition. 36 Deuteration decreases the energies of vibrational modes. 37 Vibrational states on the S0 manifold isoenergetic with the T1 ground vibrational state will therefore become more numerous upon deuteration, but will also be combinations of higher vibrational overtones with lower FCFs. The overall effect of deuteration depends on the balance of additional final states acting to increase k1 , while the diminished strength of each individual transition acts to decrease k1 . The rate of ISC depends on the sum of these smaller but more numerous transition strengths, and the overall effect will be different in different molecules. Compared to studies of phosphoresence decay in similar small PAHs, we report considerably larger values of k1 with a smaller change under deuteration. 30 A key difference is that the UC measurements were performed in room temperature liquid solutions, while the phosphorescence studies were performed in solid solutions (solvent glasses or dopants in crystals) at 77 K or below. We attribute our larger k1 constants to be a result of thermally activated curve crossing to the S0 state from thermally populated higher vibrational levels on the T1 manifold. Such a mechanism adds to the low temperature rate of ISC by a similar absolute amount in both isotopologues, resulting in a smaller relative difference in k1 at higher temperatures. Interestingly, this indicates that temperature effects on UC may not be solely due to changing viscosities and collision rates in solvents, as previously reported. 38 However, a recent study has shown that in many aromatic molecules, there remains a persistent large difference in k1 between deuterated and non-deuterated compounds over a range of temper13 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

atures. 39 Direct study of UC at low temperatures is experimentally challenging as the TTA process relies heavily on molecular collisions in fluid solvents. It should be noted that a diminution of the isotopic effect could also be due to residual oxygen, despite our rigorous attempts to remove it. On the basis of these experiments and general theory we expect deuteration to enhance the performance of other small PAHs commonly used as UC emitters, such as rubrene and diphenylanthracene (DPA). As well as deuteration, other chemical substitutions such as fluorination or methylation can also affect molecular vibrations. However, these methods also change the electronic properties and geometry of the molecule, with unpredictable effects on other important emitter parameters (k2 ,ΦP L , η, and the energy of T1 ). Phosphorescence studies on selectively deuterated PAHs also demonstrate that changes in k1 can be attributed to single substitution sites in some molecules, while in other molecules the change is simply additive as the percentage of deuteration increases. 30,31 These concepts provide a framework for systematic improvement of the k1 parameter by judicious phonon engineering. Over a lengthy time-scale, H/D exchange may occur between deuterated molecules and non-deuterated solvents, which might diminish the enhancement over time. However, incorporation of deuterium into aromatic compounds, or the back exchange, requires forcing conditions (high temperatures and pressures, strongly nucleophilic or Lewis acidic additives) and/or the use of metal catalysts. We anticipate that the added cost of emitter deuteration (or other chemical substitution), when performed at scale, will be small compared to the overall cost of a UC enhancement device for PV applications, and small compared to the potential revenue from additional efficiency gains. 40 It has been known for nearly a decade that deuteration can significantly enhance the efficiency and high-voltage stabilities of OLEDs. 41–44 Further, it has been determined that TTA contributes up to 30 % of the luminescence of state-of-the-art OLEDS, and in some cases is dominant. 45–48 Indeed, this mechanism has been held to explain these devices exceeding


Page 14 of 21

Page 15 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the statistical limit of 25 % efficiency based on a 1:3 ratio of singlets to triplets formed by recombination of electrogenerated charge carriers. Given the results reported here, that deuteration enhances the efficiency of TTA, we hypothesise that the lengthening of triplet state lifetimes contributes to the observed deuteration-enhancement of OLED efficiency.

Conclusion Deuterated perylene was found to have 45% higher relative upconversion efficiency compared to the hydrogen isotopologue, and to enter the higher efficiency high power regime at a lower threshold. Action spectra, photoluminescence kinetics, and theoretical arguments all indicate that this enhancement is due to a 16% decrease in the molecular parameter k1 , which is controlled primarily by the rate of intersystem crossing. We expect this effect to be general for the range of small polycyclic aromatic hydrocarbons widely used for upconversion research and for applications in the low excitation regime.

Acknowledgement T.W.S and D.R.M. acknowledge the Australian Research council for Future Fellowships FT130100177 and FT130100214. M.D. and A.D acknowledge the Australian Renewable Energy Agency for funding of Post Doctoral Research Project (5 - F004) and scholarship (6S018) respectively. This work was supported by an Australian Research Council Discovery Project (DP110103300) and an Australian Institute of Nuclear Science and Engineering Research Award (ALNGRA13542). The authors gratefully acknowledge M. J. Y. Tayebjee for useful discussions.

Supporting Information Available Absorption, fluorescence emission, and fluorescence excitation spectra of perylenes, a table of scaling factors used in data processing, and a representative time resolved photoluminescence 15 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

spectrum. This material is available free of charge via the Internet at http://pubs.acs. org/.

References (1) Gr¨atzel, M. Recent Advances in Sensitized Mesoscopic Solar Cells. Acc. Chem. Res. 2009, 42, 1788–1798. (2) U.S. Department of Energy, International Energy Outlook 2015 ; Energy Information Administration, 2015. (3) Green, M. Third Generation Photovoltaics: Advanced Solar Energy Conversion; Physics and Astronomy Online Library; Springer, 2003. (4) Farrell, J.; Grimley, M. Public Rooftop Revolution; Insitute for Local Self-Reliance, 2015. (5) Shockley, W.; Queisser, H. J. Detailed Balance Limit of Efficiency of p-n Junction Solar Cells. J. Appl. Phys. 1961, 32, 510–519. (6) Hanna, M. C.; Nozik, A. J. Solar Conversion Efficiency of Photovoltaic and Photoelectrolysis Cells with Carrier Multiplication Absorbers. J. Appl. Phys. 2006, 100, 074510. (7) Hirst, L. C.; Ekins-Daukes, N. J. Fundamental Losses in Solar Cells. Prog. Photovolt: Res. Appl. 2011, 19, 286–293. (8) Schulze, T. F.; Schmidt, T. W. Photochemical Upconversion: Present Status and Prospects for its Application to Solar Energy Conversion. Energy Environ. Sci. 2015, 8, 103–125. (9) Singh-Rachford, T. N.; Castellano, F. N. Photon Upconversion Based on Sensitized Triplet-Triplet Annihilation. Coord. Chem. Rev. 2010, 254, 2560–2573.


Page 16 of 21

Page 17 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(10) Islangulov, R. R.; Kozlov, D. V.; Castellano, F. N. Low Power Upconversion Using MLCT Sensitizers. Chem. Commun. 2005, 3776–3778. (11) Baluschev, S.; Miteva, T.; Yakutkin, V.; Nelles, G.; Yasuda, A.; Wegner, G. UpConversion Fluorescence: Noncoherent Excitation by Sunlight. Phys. Rev. Lett. 2006, 97, 143903. (12) Baluschev, S.; Yakutkin, V.; Miteva, T.; Wegner, G.; Roberts, T.; Nelles, G.; Yasuda, A.; Chernov, S.; Aleshchenkov, S.; Cheprakov, A. A General Approach for NonCoherently Excited Annihilation Up-Conversion: Transforming the Solar-Spectrum. New J. Phys. 2008, 10, 013007. (13) Zhao, J. Z.; Ji, S. M.; Guo, H. M. Triplet-triplet Annihilation Based Upconversion: from Triplet Sensitizers and Triplet Scceptors to Upconversion Quantum Yields. RSC Adv. 2011, 1, 937–950. (14) Gray, V.; Dzebo, D.; Abrahamsson, M.; Albinsson, B.; Moth-Poulsen, K. Triplet-Triplet Annihilation Photon-Upconversion: Towards Solar Energy Applications. Phys. Chem. Chem. Phys. 2014, 16, 10345–10352. (15) Cheng, Y. Y.; F¨ uckel, B.; MacQueen, R. W.; Khoury, T.; Clady, R. G. C. R.; Schulze, T. F.; Ekins-Daukes, N. J.; Crossley, M. J.; Stannowski, B.; Lips, K. et al. Improving the Light-Harvesting of Amorphous Silicon Solar Cells with Photochemical Upconversion. Energy Environ. Sci. 2012, 5, 6953–6959. (16) Schulze, T. F.; Cheng, Y. Y.; F¨ uckel, B.; MacQueen, R. W.; Danos, A.; Davis, N. J. L. K.; Tayebjee, M. J. Y.; Khoury, T.; Clady, R. G. C. R.; Ekins-Daukes, N. J. et al. Photochemical Upconversion Enhanced Solar Cells: Effect of a Back Reflector. Aust. J. Chem. 2012, 65, 480–485. (17) Schulze, T. F.; Cheng, Y. Y.; Khoury, T.; Crossley, M. J.; Stannowski, B.; Lips, K.;



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Schmidt, T. W. Micro-Optical Design of Photochemical Upconverters for Thin-Film Solar Cells. J. Photon. Energ. 2013, 3, 034598–034598. (18) Schulze, T. F.; Czolk, J.; Cheng, Y.-Y.; Fuckel, B.; MacQueen, R. W.; Khoury, T.; Crossley, M. J.; Stannowski, B.; Lips, K.; Lemmer, U. et al. Efficiency Enhancement of Organic and Thin-Film Silicon Solar Cells with Photochemical Upconversion. J. Phys. Chem. C 2012, 116, 22794–22801. (19) Tayebjee, M. J. Y.; McCamey, D. R.; Schmidt, T. W. Beyond ShockleyQueisser: Molecular Approaches to High-Efficiency Photovoltaics. J. Phys. Chem. Lett. 2015, 6, 2367– 2378. (20) Nattestad, A.; Cheng, Y. Y.; MacQueen, R. W.; Schulze, T. F.; Thompson, F. W.; Mozer, A. J.; F¨ uckel, B.; Khoury, T.; Crossley, M. J.; Lips, K. et al. Dye-Sensitized Solar Cell with Integrated Triplet-Triplet Annihilation Upconversion System. J. Phys. Chem. Lett. 2013, 4, 2073–2078. (21) Baluschev, S.; Yu, F.; Miteva, T.; Ahl, S.; Yasuda, A.; Nelles, G.; Knoll, W.; Wegner, G. Metal-Enhanced Up-Conversion Fluorescence: Effective Triplet-Triplet Annihilation Near Silver Surface. Nano Lett. 2005, 5, 2482–2484. (22) Schmidt, T. W.; Castellano, F. N. Photochemical Upconversion: The Primacy of Kinetics. J. Phys. Chem. Lett. 2014, 5, 4062–4072. (23) Auckett, J. E.; Cheng, Y.; Khoury, T.; Clady, R. G. C. R.; Ekins-Daukes, N. J.; Crossley, M. J.; Schmidt, T. W. Efficient Up-Conversion by Triplet-Triplet Annihilation. J. Phys. Conf. Ser. 2009, 185, 012002. (24) Cheng, Y.; F¨ uckel, B.; Khoury, T.; Clady, R. G. C. R.; Tayebjee, M. J. Y.; EkinsDaukes, N. J.; Crossley, M. J.; Schmidt, T. W. Kinetic Analysis of Photochemical Upconversion by Triplet-Triplet Annihilation: Beyond Any Spin Statistical Limit. J. Phys. Chem. Lett. 2010, 1, 1795–1799. 18 ACS Paragon Plus Environment

Page 18 of 21

Page 19 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(25) Cheng, Y.; Khoury, T.; Clady, R. G. C. R.; Tayebjee, M. J. Y.; Ekins-Daukes, N. J.; Crossley, M. J.; Schmidt, T. W. On the Efficiency Limit of Triplet-Triplet Annihilation for Photochemical Upconversion. Phys. Chem. Chem. Phys. 2010, 12, 66–71. (26) Monguzzi, A.; Tubino, R.; Meinardi, F. Upconversion-Induced Delayed Fluorescence in Multicomponent Organic Systems: Role of Dexter Energy Transfer. Phys. Rev. B 2008, 77, 155122. (27) Fujisawa, J.; Ohba, Y.; Yamauchi, S. Direct Observation of Electron Spin Polarization Transfer in Triplet-Triplet Energy Transfer Between Porphyrins and Fullerene in Fluid Solution. Chem. Phys. Lett. 1998, 282, 181 – 186. (28) Dexter, D. L. A Theory of Sensitized Luminescence in Solids. J. Chem. Phys. 1953, 21, 836–850. (29) MacQueen, R. W.; Cheng, Y. Y.; Danos, A. N.; Lips, K.; Schmidt, T. W. Action Spectrum Experiment for the Measurement of Incoherent Photon Upconversion Efficiency Under Sun-Like Excitation. RSC Adv. 2014, 4, 52749–52756. (30) Siebrand, W. Radiationless Transitions in Polyatomic Molecules. II. Triplet Ground State Transitions in Aromatic Hydrocarbons. J. Chem. Phys. 1967, 47, 2411–2422. (31) Lin, S. H.; Bersohn, R. Effect of Partial Deuteration and Temperature on Triplet-State Lifetimes. J. Chem. Phys. 1968, 48, 2732–2736. (32) Hutchison, C. A.; Mangum, B. W. Effect of Deuterium Substitution on the Lifetime of the Phosphorescent Triplet State of Naphthalene. J. Chem. Phys. 1960, 32, 1261–1262. (33) Sripathy, K.; MacQueen, R. W.; Peterson, J. R.; Cheng, Y. Y.; Dvorak, M.; McCamey, D. R.; Treat, N. D.; Stingelin, N.; Schmidt, T. W. Highly Efficient Photochemical Upconversion in a Quasi-Solid Organogel. J. Mater. Chem. C 2015, 3, 616–622.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(34) Monguzzi, A.; Mezyk, J.; Scotognella, F.; Tubino, R.; Meinardi, F. UpconversionInduced Fluorescence in Multicomponent Systems: Steady-State Excitation Power Threshold. Phys. Rev. B 2008, 78, 195112. (35) Siebrand, W. Mechanisms of Intersystem Crossing in Aromatic Hydrocarbons. Chem. Phys. Lett. 1970, 6, 192 – 194. (36) Henry, B. R.; Siebrand, W. Spin-Orbit Coupling in Aromatic Hydrocarbons. Analysis of Nonradiative Transitions between Singlet and Triplet States in Benzene and Naphthalene. J. Chem. Phys. 1971, 54, 1072–1085. (37) Burland, D. M.; Robinson, G. W. Calculated Radiationless Transition Rates for Benzene and Deuterobenzene. J. Chem. Phys. 1969, 51, 4548–4559. (38) Monguzzi, A.; Bianchi, F.; Bianchi, A.; Mauri, M.; Simonutti, R.; Ruffo, R.; Tubino, R.; Meinardi, F. High Efficiency Up-Converting Single Phase Elastomers for Photon Managing Applications. Adv. Energ. Mater. 2013, 3, 680–686. (39) Hirata, S.; Totani, K.; Zhang, J.; Yamashita, T.; Kaji, H.; Marder, S. R.; Watanabe, T.; Adachi, C. Efficient Persistent Room Temperature Phosphorescence in Organic Amorphous Materials under Ambient Conditions. Advanced Functional Materials 2013, 23, 3386–3397. (40) Krause-Heuer, A. M.; Yepuri, N. R.; Darwish, T. A.; Holden, P. J. Mild Conditions for Deuteration of Primary and Secondary Arylamines for the Synthesis of Deuterated Optoelectronic Organic Molecules. Molecules 2014, 19, 18604. (41) Tong, C. C.; Hwang, K. C. Enhancement of OLED Efficiencies and High-Voltage Stabilities of Light-Emitting Materials by Deuteration. J. Phys. Chem. C 2007, 111, 3490–3494.


Page 20 of 21

Page 21 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(42) Tsuji, H.; Mitsui, C.; Nakamura, E. The Hydrogen/Deuterium Isotope Effect of the Host Material on the Lifetime of Organic Light-Emitting Diodes. Chem. Commun. 2014, 50, 14870–14872. (43) Wang, P.; Wang, F.-F.; Chen, Y.; Niu, Q.; Lu, L.; Wang, H.-M.; Gao, X.-C.; Wei, B.; Wu, H.-W.; Cai, X. et al. Synthesis of All-Deuterated Tris(2-Phenylpyridine)Iridium for Highly Stable Electrophosphorescence: The “Deuterium Effect”. J. Mater. Chem. C 2013, 1, 4821–4825. (44) Shao, M.; Keum, J.; Chen, J.; He, Y.; Chen, W.; Browning, J. F.; Jakowski, J.; Sumpter, B. G.; Ivanov, I. N.; Ma, Y.-Z. et al. The Isotopic Effects of Deuteration on Optoelectronic Properties of Conducting Polymers. Nature Comm. 2014, 5, 4180. (45) Sinha, S.; Monkman, A. P. Delayed Electroluminescence via TripletTriplet Annihilation in Light Emitting Diodes Based on Poly[2-Methoxy-5-(2’-Ethyl-Hexyloxy)-1,4Phenylene Vinylene]. Appl. Phys. Lett. 2003, 82, 4651–4653. (46) Kondakov, D. Y. Triplet–Triplet Annihilation in Highly Efficient Fluorescent Organic Light-Emitting Diodes: Current State and Future Outlook. Phil. Trans. Roy. Soc. Lond. A 2015, 373, 20140321. (47) Kondakov, D. Y.; Pawlik, T. D.; Hatwar, T. K.; Spindler, J. P. Triplet Annihilation Exceeding Spin Statistical Limit in Highly Efficient Fluorescent Organic Light-Emitting Diodes. J. Appl. Phys. 2009, 106, 124510. (48) Kondakov, D. Y. Characterization of Triplet-Triplet Annihilation in Organic LightEmitting Diodes Based on Anthracene Derivatives. Journal of Applied Physics 2007, 102, 114504.


Deuteration of Perylene Enhances Photochemical Upconversion

Recommend Documents