Triplet Excited State Energies and Phosphorescence Spectra of

Article pubs.acs.org/JPCB

Triplet Excited State Energies and Phosphorescence Spectra of (Bacterio)Chlorophylls Daniel A. Hartzler,† Dariusz M. Niedzwiedzki,‡ Donald A. Bryant,§,∥ Robert E. Blankenship,‡,⊥ Yulia Pushkar,† and Sergei Savikhin*,† †

Department of Physics, Purdue University, 525 Northwestern Avenue, West Lafayette, Indiana 47907, United States Photosynthetic Antenna Research Center (PARC), ⊥Departments of Chemistry and Biology, Washington University in St. Louis, One Brookings Drive, St. Louis, Missouri 63130, United States § Department of Biochemistry & Molecular Biology, The Pennsylvania State University, 108 Althouse Laboratory, University Park, Pennsylvania 16802, United States ∥ Department of Chemistry and Biochemistry, Montana State University, 103 Chemistry and Biochemistry Building, P.O. Box 173400, Bozeman, Montana 59717, United States

J. Phys. Chem. B 2014.118:7221-7232. Downloaded from pubs.acs.org by UNIV OF SUNDERLAND on 11/04/18. For personal use only.

‡

S Supporting Information *

ABSTRACT: (Bacterio)Chlorophyll ((B)Chl) molecules play a major role in photosynthetic light-harvesting proteins, and the knowledge of their triplet state energies is essential to understand the mechanisms of photodamage and photoprotection, as the triplet excitation energy of (B)Chl molecules can readily generate highly reactive singlet oxygen. The triplet state energies of 10 natural chlorophyll (Chl a, b, c2, d) and bacteriochlorophyll (BChl a, b, c, d, e, g) molecules and one bacteriopheophytin (BPheo g) have been directly determined via their phosphorescence spectra. Phosphorescence of four molecules (Chl c2, BChl e and g, BPheo g) was characterized for the first time. Additionally, the relative phosphorescence to fluorescence quantum yield for each molecule was determined. The measurements were performed at 77K using solvents providing a six-coordinate environment of the Mg2+ ion, which allows direct comparison of these (B)Chls. Density functional calculations of the triplet state energies show good correlation with the experimentally determined energies. The correlation determined computationally was used to predict the triplet energies of three additional (B)Chl molecules: Chl c1, Chl f, and BChl f.

T

photoprotection mechanisms used in natural pigment−protein complexes. Structurally (Figure 2), (B)Chl molecules are defined as cyclic tetrapyrroles with a fifth isocyclic ring and a central metal ion, usually Mg2+,10 though naturally occurring Zn2+ (B)Chls are known to exist.11,12 All molecules in this study are the Mg2+containing type. The rings are labeled A−E, and the carbons are numbered from 1 to 20, according to IUPAC-IUB notation.9 Transition dipole moments for the short wavelength (B or Soret bands) and long wavelength (Q bands) transitions are oriented along the axis connecting the nitrogen atoms of the A and C rings for the y-polarization (By and Qy) and the nitrogen atoms of the B and D rings for the x-polarization (Bx and Qx). The (B)Chls can be divided into three major subgroups according to the extent of the conjugated π system of the macrocycle: porphyrin type, chlorin type, and bacteriochlorin type.9,10,13 The porphyrin-type molecules contain a fully unsaturated macrocycle, meaning the conjugated π system

he sun is the main energy source of life on Earth, with nearly every organism depending on photosynthesis or its products for energy. An estimated 2.5 × 1024 J of solar energy reaches the surface of the Earth each year, and photosynthetic organisms capture about 2.5 × 1021 J.1 The main class of molecules involved in photosynthesis are the (Bacterio)Chlorophylls ((B)Chls),2 which are used both for sunlight energy capture in antenna complexes and charge transfer in reaction centers.3 Their abundance is so great they can be observed from space, causing the characteristic “red edge” enhanced spectrum of reflected solar radiation.4 In photosynthetic organisms, (B)Chls are bound to proteins that determine their function, prevent their degradation, and limit their ability to damage cellular constituents.3,5 Outside of these structures and under illumination, (B)Chls readily sensitize singlet oxygen (1O2) production with high quantum yield from the (B)Chl triplet excited state (Figure 1).6 This reactive oxygen species (ROS) rapidly degrades (B)Chls and surrounding biomolecules,7 resulting in damage to tissues or death of the organism.8 Knowledge of the (B)Chl triplet state energies is essential for developing appropriate photoprotection mechanisms in artificial systems as well as for elucidation of the © 2014 American Chemical Society

Received: January 16, 2014 Revised: May 7, 2014 Published: June 4, 2014 7221

dx.doi.org/10.1021/jp500539w | J. Phys. Chem. B 2014, 118, 7221−7232

The Journal of Physical Chemistry B

Article

carotenoid molecule with a triplet state lower in energy than the associated (B)Chl molecules is typically positioned within ∼4 Å of the (B)Chl molecules, allowing for rapid triplet energy transfer from (B)Chl to carotenoid. If the energy of the carotenoid triplet state is below that of the singlet oxygen, the carotenoid can safely return to the ground state, dissipating the energy as heat (Figure 1). The most direct method to measure the energy of the triplet excited state is by its phosphorescence spectrum. However, the optical transition from the T1 state of (B)Chl to the singlet ground state (phosphorescence) is spin-forbidden and has very low quantum yield, being 104 to 106 times lower than the fluorescence quantum yield (see Results). Due to the difficulty in detecting such signals, there are still several (B)Chls for which phosphorescence spectra have not been measured. In this work, we constructed a high-sensitivity, time-gated phosphorimeter and measured the phosphorescence spectra for 10 major (B)Chl species found in nature. Spectra for Chl c2, BChl e, and BChl g, as well as BPheo g are measured for the first time. Because all samples were measured under analogous conditions, the triplet state energies for these (B)Chl molecules can now be directly compared. The triplet state energies for these molecules were also determined by quantum chemical calculations and compared with experimental values. These computational methods were also used to predict the triplet state energies of three additional molecules: Chl c1, Chl f, and BChl f.

Figure 1. Simplified (B)Chl energy level diagram showing the formation of the (B)Chl triplet state (T1) from the excited singlet state (S1) followed by electronic energy transfer (EET) from the (B)Chl triplet state to a carotenoid molecule or an oxygen molecule. A − absorption, F − fluorescence, P − phosphorescence, IC − internal conversion, ISC − intersystem crossing.

extends over the entire macrocycle (excluding the C-131 and C132 carbon atoms of the E ring). The chlorin-type molecules contain a saturated bond between the C-17 and C-18 atoms, and the bacteriochlorin-type molecules contain saturated bonds between the C-7 and C-8 and between the C-17 and C-18 atoms, meaning that the conjugation does not extend across these carbon atoms.9The differences in the conjugated π systems among the subgroups have significant effects on the spectral properties of the molecules.13−15 In the case of symmetric porphyrin molecules, the symmetry of the π system results in a doubly degenerate lowest unoccupied molecular orbital (LUMO), which along with the near degeneracy of the two highest occupied molecular orbitals (HOMO and HOMO1) causes significant configuration interaction (CI) among the excited single-electron configurations. The result of this CI is a Q band with nearly zero transition dipole moment (for a symmetric metaloporphyrin) and a strong Soret band. Saturation of the C-17 to C-18 bond in chlorin-type molecules changes the symmetry of the π system, breaking the degeneracy of the LUMO and near-degeneracy of the HOMO and HOMO-1, which results in fully allowed Q band transitions and lowering the energy of the Qy band. Additional saturation of the C-7 to C-8 bond in bacteriochlorin-type molecules further redshifts Qy and increases its oscillator strength.13−15 In (B)Chls, the triplet state (T1) is populated from the first excited singlet (S1) state via intersystem crossing (ISC) (Figure 1), a spin-forbidden process that occurs with a high quantum yield of >50%.16,17 This high yield for a forbidden process comes from strong spin−orbit coupling (SOC) that causes mixing of the S1 and T1 states and from considerable overlap of the vibrational levels of S1 and T1.18−21 Once in the triplet excited state, a (B)Chl molecule can decay to the ground state through several pathways including IC, phosphorescence, and quenching by oxygen or other molecules.22 Monomeric (B)Chl molecules in an air-saturated solution will transfer their triplet energy to oxygen molecules within ∼1 μs, resulting in the formation of highly reactive singlet oxygen. The quantum yield of such energy transfer approaches 100%, because the natural lifetime of (B)Chl triplet state is ∼500 μs.10 To prevent singlet oxygen formation in photosynthetic proteins, a suitable

■

MATERIALS AND METHODS Isolation and Purification of (B)Chls. Chl a and b were obtained from Sigma-Aldrich Corporation and used with no further purification. The other chlorophyll pigments were extracted with 100% technical grade methanol from cells of the following bacterial or algal species: Chl c2 from Symbiodinium, Chl d from Acaryochloris marina, BChl a from Rhodobacter sphaeroides, BChl b from Blastochloris viridis, BChl c from Chlorobaculcum tepidum, BChl e from Pelodictyon phaeum, BChl g from Heliobacterium modesticaldum. Purification was done using an Agilent 1100 HPLC system equipped with a Zorbax Eclipse XDB-C18 reverse phase column (250 mm × 4.6 mm) with an isocratic flow rate of 1 mL/min and acetonitrile/ methanol/tetrahydrofuran (60:36:4, v/v/v) as a mobile phase. BChl d was extracted from the bchQRU mutant chlorosomes of Chlorobaculcum tepidum.23 In an anaerobic atmosphere, 30 μL of the isolated chlorosomes (optical density of ∼1000 cm−1) were added to 1 mL of hexane and 1 mL of aqueous buffer (1 M NaCl, 0.1 M KH2PO4, pH 7; both were deoxygenated). The sample was mixed vigorously using a vortex mixer and centrifuged for 10 min under 7200g until the phases were separated, with the upper layer being hexane and the lower layer being aqueous buffer. The carotenoids and quinones remained dissolved in the hexane phase and BChl d precipitated between the phases. The hexane and the aqueous phases were carefully removed and discarded. This step was repeated to obtain higher purity. Once more, 1 mL each of hexane and aqueous buffer are added, the mixture was then mixed and centrifuged until the solution separated into two phases. The hexane phase (top) was removed and discarded on this last repeat of this manipulation, and 1 mL of deoxygenated chloroform was added to the remaining solution, which was then vigorously mixed. Because BChl d dissolves in chloroform, another centrifugation separated the BChl d solution in chloroform and the aqueous buffer into two phases. The 7222



Article

Figure 2. Structures of the (Bacterio)Chlorophyll molecules studied, both experimentally and computationally. Numbering is based on IUPAC-IUB notation.9

The central magnesium ion (Mg2+) of a (B)Chl molecule can accept up to six ligands, four provided by the nitrogen atoms of the macrocycle and up to two more provided by the environment (e.g., solvents or protein residues). It has been shown that the coordination state of the central ion can influence the fluorescence and phosphorescence properties of the molecules with higher coordination states shifted to longer wavelengths.28,29 However, it is not possible in general to identify a specific Qy absorption or fluorescence maximum with a particular coordination state given that solvent effects have been shown to shift the Qy absorption maximum by up to 0.04 eV, independent of coordination state,30,31 while also causing a solvent-dependent fluorescence Stokes shift of 0.02−0.05 eV.32 Additionally, for a particular solvent, the direction of the Qy maximum shift can depend on the central metal ion, as seen with metal-substituted BChl molecules.33,34 However, data from Rätsep et al.35 suggest that in solvents where the (B)Chl coordination state is temperature-dependent, its absorption and fluorescence Qy maxima shift toward longer wavelength upon changing from 5- to 6-coordinate state, and that shift is larger

BChl d solution in chloroform is removed and dried under anaerobic conditions.24 Absorption spectra taken after purification show the final carotenoid to BChl d molar ratio can be at most 1:70, a reduction by a factor of approximately 5.4 compared to wild-type chlorosomes.25 In the frozen solvent used for phosphorescence measurements, molecular diffusion is precluded, and given the low concentration, it is highly unlikely any carotenoids were close enough to quench the BChl d triplet states. BPheo g was produced in large quantities during sample storage as a natural breakdown product of BChl g. It was identified as such by its absorption spectrum, with maxima at 530, 670, and 757 nm. Spectral interference in the short wavelength region with chlorin-type breakdown products26,27 (see the section, Sample Integrity, in Materials and Methods) prevented accurate identification of the BPheo g Soret peaks; however, a major peak was observed at 361 nm with a shoulder at 390 nm. These absorption peaks are consistent with BPheo g.26,27 7223



Article

Samples were excited by a 0.5−4 mJ, 5 ns laser pulse at a 10 Hz repetition rate (Ekspla NT 342B) into the Qy absorption band for Chl a (660 nm), Chl b (640 nm), Chl c2 (650 nm), Chl d (690 nm), BChl a (770 nm), BChl b (795 nm), BChl c (660 nm), BChl d (660 nm), BChl e (670 nm), and BPheo g (770 nm), whereas BChl g was excited in the Qx band (600 nm). This ensured dominant excitation of the molecules of interest. Note that quartz or fused silica cells are essential, as some glass cells were found to have trace Nd3+ impurities producing emission bands at ∼910, ∼1064, and ∼1330 nm, whereas some sapphire cells were found to have trace Cr3+ impurities, which produces emission at ∼695 nm. The laser pulses were focused onto the samples with a cylindrical lens (L5), and the sample emission was collected and focused onto a fast optical gate consisting of an optical chopper (Scitech, 300CD) closely flanked by slits, S1 and S2, to reduce scattered light and ensure fast gating (Figure 3). With a 10 cm diameter wheel rotating at 6000 rpm, this arrangement allowed the gate to open in approximately 20 μs. A 700 or 800 nm long wave pass (LWP) interference filter (Edmund Optics, 62−987 and 66−235 respectively) was placed in front of the monochromator (Oriel, MS257 with 77745 grating) to further reduce scattered laser light and fluorescence. Note that the 950 nm IR timing LED in the chopper head was replaced with a red 630 nm LED (RadioShack, model 276-020) to avoid spectral interference with the phosphorescence. Data acquisition was handled by an analog-to-digital converter (National Instruments, AT-MIO-16XE-10) and a software implemented boxcar integrator, which calculates the signal as the difference between two boxcar integration windows, a background window (before the optical gate opens), and a detector signal widow (after the optical gate opens). For each sample, 5−10 consecutive phosphorescence spectra were measured and averaged resulting in total integration times of 12−120 s for each wavelength point. All phosphorescence spectra (see Results) were corrected for the spectral sensitivity of the spectrometer, which was calibrated by use of a blackbody of known temperature (a tungsten lamp). A drawback of the germanium detector is its high cross-section for interaction with cosmicrays. These appear as large spikes in the detector output that occur randomly at a rate of approximately 0.1 s−1. These spikes were automatically excluded by the data acquisition software. Phosphorescence emission lifetimes were estimated by varying the electronic time delay (Berkeley Nucleonics, Model 7040) and measuring the emission intensity at different delay times, where delay time is defined as the time difference between the firing of the laser and the opening of the optical gate in Figure 3. The measured signal (S) as a function of delay time (t) was approximated using a single exponential:43

than the observed temperature-dependent shift in other (B)Chl and solvent systems.35 Although in nature (B)Chl molecules are known to exist in both the 5- and 6-coordinated state,36,37 in this study, care was taken to ensure that all molecules were in the 6-coordinate state because this proved to be the most straightforward to achieve as a pure state. Raman and fluorescence line narrowing experiments have shown that at room or low temperature (B)Chl molecules in pyridine, methanol (MeOH), and tetrahydrofuran (THF) are either entirely or predominantly 6-coordinated.31,38−40 For the spectroscopic measurements of the present study, all samples were dissolved in toluene/pyridine (4:1, v/v) with the exception of Chl c2, which was dissolved in MeOH/THF (4:1, v/v), because of its chemical instability in pyridine.10 Sample concentrations were approximately 20−100 μM during phosphorescence measurements. Spectroscopic Techniques. The phosphorescence quantum yield of (B)Chls are 104 to 106 times lower than the fluorescence quantum yield. Consequently, the red tail of fluorescence is 1−2 orders of magnitude stronger than the phosphorescence even at the peak of the phosphorescence emission band (Figure S1, Supporting Information). However, the fluorescence lifetime of the (B)Chls studied lie between 2 and 7 ns,10 whereas triplet state lifetimes are in the range of 100−900 μs at room temperature10 and become even longer at low temperatures.28,41,42 This allows effective gating of the phosphorescence in the time domain using a fast optical shutter.43−45 A time-gated phosphorescence spectrometer (Figure 3) was built using a sensitive, liquid nitrogen-cooled germanium

Figure 3. Block diagram of the home-built phosphorimeter. Sample in a cryostat is excited by a ∼5 ns laser pulse. Lenses L1 and L2 collect emission from the sample and focus it on the mechanical shutter consisting of slits S1, S2 and the chopper wheel. Light transmitted by the gate is collected (L3, L4) and filtered by a long wave pass filter (LWP) and monochromator (8 nm bandwidth) before detection by the cryogenically cooled germanium photodetector (Ge PD).

S(t ) = S0e t / τ

(1)

where S0 is a constant and τ is the emission lifetime. For each sample studied, the phosphorescence intensity was measured at three to four different delay times at several wavelengths flanking the phosphorescence emission maximum. These data were then fit globally for all wavelengths using a monoexponential function (eq 1). However, due to the small number of points used, the lifetimes determined by this method are not expected to be accurate beyond one significant digit. The ratios of the fluorescence and phosphorescence quantum yields were computed using the integrated photon flux of the fluorescence and phosphorescence spectra. Fluorescence spectra were measured using the described

photodetector (North Coast Scientific Corp., EO-817L) with a measured noise equivalent power (NEP) of 7 f W/(Hz)1/2. Samples were placed into quartz EPR tubes and loaded into a cryostat at 77K (Oxford Instruments, Optistat DN). Special care had to be taken to avoid condensation of liquid oxygen into the sample upon cooling, as its presence resulted in a strong singlet oxygen signal via direct laser excitation of the liquid oxygen.46 This was accomplished by warming the cryostat and sample to a temperature above the boiling point of liquid oxygen (90K) and purging the cryostat with nitrogen gas through a custom sample holder for 1 to 3 min. 7224



Article

Figure 4. Fluorescence (red) and phosphorescence (black) emission spectra. The left-hand scale is the fluorescence intensity normalized to one, and the right-hand scale is the phosphorescence intensity relative to the fluorescence intensity. The vertical dotted line corresponds to the emission maximum of singlet oxygen (1O2).

to the beam path until the prompt fluorescence signal is maximized. The fluorimeter was aligned in a similar way. Absorption spectra were measured with a UV−vis absorption spectrometer (Cary 300 Bio) at room temperature, because the samples had low optical quality at low temperature. Sample Integrity. To ensure sample integrity during experiments, room-temperature absorption spectra were recorded before and after the measurements. These indicated that the porphyrin and chlorin-type samples had not changed during measurements and the Qy absorption bands of the bacteriochlorin-type samples (BChl a, BChl b, and BChl g) had dropped in intensity by no more than 5%, with no apparent change in spectral shape. The drop in the bacteriochlorin Qy absorbance was accompanied by the appearance of chlorin-type byproducts that absorb in the 660−680 nm range.26,27,48 These byproducts do not spectrally overlap the bacteriochlorin Qy bands, thus selective excitation of only the bacteriochlorin-type molecules is possible, ensuring the small fraction of these products did not affect the shape of the phosphorescence spectra. This was also supported by phosphorescence spectra measured in the beginning and at the end of each experiment; no noticeable changes in shape and position of the phosphorescence band were observed. Computational Methods. Quantum chemical calculations were performed by Gaussian 0949 on eight cores of a Dell compute node (four 2.3 GHz 12-Core AMD Opteron 6176) operated as part of a cluster by the Purdue Rosen Center for Advanced Computing (RCAC). These calculations were performed using unrestricted DFT utilizing the B3LYP functional and 6-31G(d,p) basis in the vacuum state, that is, with no solvation model or explicit solvent molecules in the calculation.50 Ground state geometry optimization and vibrational frequency calculations were performed for both the singlet and triplet state of each of the 10 (B)Chl molecules measured in this study as well as for an additional three (B)Chls not experimentally measured. The final energy of the triplet state was taken as the difference between the predicted

phosphorescence spectrometer with a zero delay time. Calibrated neutral density filters were used to reduce the fluorescence signal intensity while all other parameters were kept the same as in the phosphorescence measurements. When measuring phosphorescence emission spectra, there was an approximately 20 μs minimum delay time (see spectrometer description, this section) due to the finite speed of the optical gating. This was corrected by using the measured lifetimes to project the phosphorescence emission intensity to its value at zero time delay. The high sample concentrations used to acquire the phosphorescence emission spectra caused significant reabsorption of the fluorescence, resulting in a truncated emission peak. To account for reabsorption of the fluorescence emission peak, fluorescence emission spectra were also recorded with a fluorimeter (Cary Eclipse) at 77K in the same solvents at lower sample concentrations of 1−5 μM (see Results). These spectra were then scaled to match the long wavelength vibrational bands of the fluorescence recorded with the phosphorescence spectrometer and used to estimate the relative quantum yield of phosphorescence (Figure S2, Supporting Information). Because the fluorescence does not overlap the long wavelength tail of the Qy absorption band, its shape is not affected by high sample concentration, thus this approach effectively corrects for the reabsorption.47 Emission maxima were determined by fitting a Gaussian function to the top 30% of peak amplitude (for example, 952− 991 nm for Chl a). This range was chosen to minimize the influence of the vibrational bands. The major source of uncertainty in fluorescence and phosphorescence emission maxima originates from misalignment of the sample with the optical beam path of the instrument (e.g., the straight line in Figure 3 connecting the sample with the monochromator through the center of lenses L1-L4). Offset from this line was noted to shift the emission peak by up to 10 nm, depending on the monochromator bandwidth. To counteract this, the phosphorimeter was aligned by reducing the bandwidth to approximately 3 nm and translating the sample perpendicular 7225



Article

Table 1. Experimental and Theoretical Values of Ten Types of (B)Chl and One BPheoa molecule porphyrin type Chl c2 chlorin type Chl a

λF (nm)b 645 640

680 677 672 675    Chl b 664 662 654 660   Chl d 706 705 699  BChl c 675 674   BChl d 667 663   BChl e 667  670 bacteriochlorin type BChl a 795 795 778 785 BChl b 825 822 821 BChl g 786 792 BPheo g 769 singlet O2

λP (nm)b

τP (μs)

ΦP / ΦF

927 (967)e

4000 833

1.6 × 10−4 

973 (1079)e 925 970 930 980 965 920 (1028)e 875 920 890 930 1068 (1132)e 978 981 1014 (1055)e 960 1040 960 (1065)e 920 960−980 953 912 (1060)e

2000 413 2500 1200 2400 1800  3000 556 5100 2800 4300 3100 1000 312 1050 1100 1000 474 1600 1100 1000 448 2100 1500 1000  461

2.6  9.9 2.6    1.2  2.6 2.0   5.3   ΦP 1.2    1.9    1.0  

1232 (1415)e 1214 1221 1256 (1415)e 1255 1169 (1336)e 1122 1272 1270

200 99   100 99  400 130 200  

1.2  ΦP ΦP 1.5   3.7  2.6  

× 10−5 × 10−5 × 10−5

× 10−4 × 10−4 × 10−4

× 10−6

= (1.5−3.5) × 10−5 × 10−5

× 10−5

× 10−4

× 10−6 ∼ 10−7 ∼ 10−7 × 10−6

× 10−6 × 10−5

coord. state

solventc

footnote ref

6 6

Me/THF THF (RT)

d f

6 6 [5]g [6]g [5]g 6 6 6 6 [5]g [6]g [5]g 6 6 6 [5]g [5]g 6 6 5 6 6 6 5 6 6 5 6

T/P P (RT) DE DE/PE DE P T/P (1%) T/P P (RT) DE DE/PE DE P T/P P (RT) DE water/Triton T/P P (RT) DE P T/P P (RT) DE P T/P T/P P (RT)

d f h h j j k d f h h j j d f l l d f m j d f m j d d f

6 6 5 6 6 6 6 6 6 N/A N/A N/A

T/P P (RT) triethylamine 2MeTHF/P (10%) T/P P (RT) T/P (16%) T/P P (RT) T/P  

d f k k d f k d f d d n

a Data include fluorescence and phosphorescence emission maxima (λF and λP respectively), the approximate phosphorescence emission lifetime (τP), the relative phosphorescence/fluorescence quantum yield (ΦP/ΦF), and the coordination state and solvent used during measurement. b Experimental uncertainties in peak wavelengths (Δλ) are defined as 1/2 the instrument bandwidth: ΔλF = ± 2.5 nm, Δλp = ± 4 nm. cSolvent temperature at 77K unless stated otherwise (RT → room temperature). T/P = toluene/pyridine (4:1 v/v). Me/THF = methanol/tetrahydrofuran (4:1 v/v). DE/PE = diethyl ether/petroleum ether (1:1 v/v). 2MeTHF/P (10%) = 2 methyl-tetrahydrofuran + 10% pyridine. Water/Triton = water + 2% Triton X100. DE = diethyl ether, P = pyridine; dThis work. eValues in brackets are calculated using eq 7 from ref 10 [ΔE(T1−S0)(cm−1) = 21258−1539 ln kint] and data from Table 1 of ref 10. fref 10. gCoordination state inferred from phosphorescence emission maximum. href 28. jref 51. k ref 29. lref 41. mref 42. nref 44.

are summarized in Table 1. The table also lists respective data obtained by other groups. Porphyrin-Type Molecule: Chl c2. Chl c2, the lone porphyrin-type Chl among the studied samples, exhibits the shortest wavelength fluorescence of any of the other Chls measured and the second shortest wavelength phosphorescence, just a few nanometers longer than Chl b (Table 1). The

total electronic and thermal free energies for the singlet and triplet ground states.

■

RESULTS Phosphorescence and fluorescence spectra recorded at 77K are shown in Figure 4, whereas the data on the fluorescence and phosphorescence emission maxima and relative quantum yields 7226



Article

relative quantum yield of Chl c2 phosphorescence is relatively high (10−4) for a (B)Chl and comparable to the upper range of chlorin-type Chls (Table 1). The phosphorescence spectrum of Chl c2 possesses a major vibronic band at 1029 nm with a relative intensity of 0.60 (relative intensity is defined as the ratio of emission intensity at the vibronic peak maximum to the intensity at the emission maximum listed in Table 1). This molecule was the only sample to be dissolved in a different solvent (MeOH/THF) due to its chemical instability in pyridine under illumination.10 It should be noted that this sample contained different phosphorescing species, likely a mixture of 5- and 6-coordinate molecules. Effort was made to excite the longest wavelength absorbing species (see the Discussion section). Chlorin-Type Molecules: Chl a, Chl b, Chl d, BChl c, BChl d, BChl e. These molecules fluoresce in the range of 664−706 nm and phosphoresce in the range of 918−1012 nm with Chl b emitting at the shortest wavelength and Chl d emitting at the longest wavelength (Table 1). Relative quantum yields spanned 2 orders of magnitude, 10−4 to 10−6, with Chl b having the highest and Chl d having the lowest relative quantum yield of this group (Table 1). The phosphorescence spectrum of Chl a possesses a major vibronic band clearly resolved at 1109 nm with a relative intensity of 0.30. The wavelength and relative intensities of the major vibronic bands for the other chlorin-type molecules are Chl b (1045 nm, 0.36), Chl d (1275 nm, 0.32), BChl c (1186 nm, 0.20), BChl d (1093 nm, 0.29), and BChl e (1042 nm, 0.52). The BChl e sample contained a mixture of 5- and 6coordination states. However, this sample allowed for predominate excitation of either the 5- or 6-coordinated states with proper choice of the excitation wavelength, allowing extraction of the phosphorescence emission maximum for both species (Figure 5 and Figure S2 in Supporting Information).

The 41 nm shift in the phosphorescence emission peak is consistent with the difference observed between the 5- and 6coordinate states of Chl a and Chl b. The difference between the fluorescence maxima could not be resolved with this method. Bacteriochlorin-Type Molecules: BChl a, BChl b, BChl g, BPheo g. These molecules exhibit the longest wavelength emission maxima of all the molecules studied in this work. The fluorescence emission maxima were in the range of 786−825 nm, and the phosphorescence maxima occurred in the range of 1170−1255 nm with BChl g emitting at the shortest wavelength and BChl b emitting at the longest wavelength in both cases (Table 1). The BChls of this group exhibit the lowest relative quantum yields of all molecules in this study, all three being in the order of 10−6 (Table 1). Due to their tendency to produce chlorin-type molecules, care had to be taken to avoid exciting these breakdown products (see the section Sample Integrity in Results). The phosphorescence spectrum of BChl g does not possess well-resolved vibronic bands. Instead, a broad feature is present that extends from approximately 1200−1500 nm and is best fit with two Gaussian components in addition to the main maximum at 1169 nm (amplitude of 0.86 and full width at half max (fwhm) of 45 nm), one vibronic component is centered at 1215 nm (amplitude of 0.15 and fwhm of 36 nm), whereas the second component is centered at 1296 nm (amplitude of 0.19 and a fwhm of 245 nm). The other bacteriochlorin-type molecules show evidence of similar broad vibronic features, in particular BChl b and BPheo g; however, the poor signal-tonoise of these spectra does not allow meaningful fits to be made. BPheo g was the molecule with the shortest emission wavelength and with the highest relative quantum yield of this class, which may be due to the absence of a coordinating metal ion. Singlet Oxygen Emission Spectrum. The singlet oxygen emission spectrum (Table 1, Figure 4) was measured by immersing an EPR tube into a transparent optical dewar (KGW-Isotherm) filled with liquid nitrogen and allowing atmospheric oxygen to condense into the tube. This liquid was excited directly with a laser pulse in the 600−700 nm range.46 Quantum Yield. The fluorescence quantum yields (ΦF) of many of the (B)Chl molecules used in this study have been measured at room temperature in various solvents. However, the fluorescence quantum yield depends on solvents52,53 and

Figure 5. Phosphorescence spectra of different species (5- and 6coordinated states) in BChl e sample excited at 640 nm (black) and 670 nm (red).

Table 2. Comparison of Triplet State Energies Predicted by DFT to the Measured Values molecule porphyrin type Chl c2 chlorin type Chl a Chl b Chl d BChl c BChl d BChl e bacteriochlorin type BChl a BChl b BChl g

experimental eV (nm)

calculation eV (nm)

1.338 (927)

1.283 (966)

1.274 1.348 1.161 1.223 1.292 1.301

1.221 1.385 1.136 1.224 1.241 1.289

(973) (920) (1068) (1014) (960) (953)

1.006 (1232) 0.987 (1256) 1.061 (1169) 7227

difference eV (nm) 0.055

(1016) (896) (1091) (1013) (999) (962)

0.054 −0.037 0.025 −0.001 0.051 0.012

0.809 (1532) 0.752 (1649) 0.826 (1501)

0.197 0.235 0.235



Article

temperature54,55 (e.g., ΦF of Chl a in ethanol increases by 64% between 300 K and 77K54). Because all the phosphorescence measurements in this study have been performed at 77K, the fluorescence quantum yield at this temperature should be used to compute the absolute phosphorescence quantum yield. Therefore, only the relative phosphorescence/fluorescence quantum yields at 77K are given in Table 1. DFT Calculations. The specific molecules used in the calculations were Chl a, Chl b, Chl c1, Chl c2, Chl d, Chl f, BChl a, BChl b, 8-ethyl-12-methyl-BChl c, 8-ethyl-12-methyl-BChl d, 8-ethyl-12-ethyl-BChl e, 8-ethyl-12-ethyl-BChl f, and BChl g (Figure 2). All molecules (except Chl c1 and Chl c2) possessed the same esterifying alcohol at R173, which consisted of a phytyl truncated at the fifth carbon atom and terminated with a methyl group (−CH3). Truncation at carbon number five produces the same esterifying alcohol (−C6H10-CH3) regardless of whether the starting point is a phytyl, farnesyl, or geranylgeranyl. The results of the DFT calculations were compared to the measured values for 10 (B)Chls (Table 2, Figure 6). These

Table 3. Triplet State Energies of Chl c1, Chl f, and BChl f Predicted by DFT molecule

DFT calculation eV (nm)

porphyrin type Chl c1 chlorin type Chl f BChl f

corrected valuea eV (nm)

1.34 (925) 1.15 (1078) 1.33 (932)

1.16 (1069) 1.34 (925)

a

The correction model derived in the Discussion section (eq 2) has been applied.

Table 4. Effect of Molecule Coordination State on the Emission Propertiesa

Chl a

Chl b

Chl d

BChl c

BChl d

BChl e

BChl a

Figure 6. Triplet state energies calculated by DFT plotted against experimentally measured values. Dotted line: least-squares fit (with slope constrained to one) for chlorin-type molecules. Dashed line: fit for bacteriochlorin-type molecules. The single data point for porphyrin-type molecules does not allow for a fit to be made, but it should be noted that it lies within the scatter for the chlorin type. The points for Chl c1 (porphyrin type) and Chl f and BChl f (both chlorin type) are plotted using the raw values predicted by DFT, no attempt at correction or experimental measurements have been made for these points.

coord.b state

fluor. (eV)

phos. (eV)

energy gap (eV)

footnote ref

5 6 Δd 5 6 Δ 5 6 Δ 5 6 Δ 5 6 Δ 5 6 Δ 5 6 Δ

1.85 1.84 0.01 1.90 1.88 0.02 1.78 1.76 0.02 -1.84 --1.86 --1.86 -1.60 1.56 0.04

1.34 1.28 0.06 1.42 1.35 0.07 1.27 1.16 0.11 1.29 1.22 0.07 1.35 1.29 0.06 1.35 1.30 0.05 1.02 1.01 0.01

0.51 0.56 −0.05 0.48 0.53 −0.05 0.51 0.60 −0.09 -0.62 --0.57 --0.55 -0.58 0.55 0.03

c c c c e f g f g f f f h f

a

A clear trend among the chlorin-type molecules is visible in the phosphorescence and S1 − T1 energy gap (see Figure 1) upon changing the state from 5- to 6-coordinated. bNote: different coordination states are achieved, in general, by use of different solvents.30,31 cref 28. dΔ denotes the difference between 5 and 6coordinated states; eref 41 fThis work. gref 42. href 29.

fluorescence is due to a change in coordination state or due to other solvent effects. Thus, the differences between triplet state energies for the same species measured in this paper and in the literature (and within the literature) are likely to stem from coordination number and solvent effects. In the course of this study it was observed that a diethyl ether/petroleum ether (1:1, v/v) solvent resulted in a uncontrolled mixture of 5- and 6-coordinated molecules, whereas a toluene/pyrinine (4:1, v/v) solvent nearly universally resulted in 6-coordinated molecules. The exceptions were Chl c2 and BChl e molecules, which were likely mixtures of coordination states. By exciting the Chl c2 and BChl e samples on the long wavelength side of the Qy absorption band, the 6coordinated states could be predominantly excited, diminishing the influence of the other states (Figure S3, Supporting Information). Note, Chl c2 was dissolved in MeOH/THF rather than toluene/pyridine, as described in Materials and Methods. Control of coordination state, either through choice of solvents or selective excitation, is important because mixtures

calculations produced excellent agreement with experiment with an average deviation of 0.03 eV from experimentally measured values for the porphyrin and chlorin-type molecules and an average deviation of 0.22 eV from experiment for bacteriochlorin-type molecules (BChl a, b, and g). The correlations observed between the experimental and calculated values (Figure 6) now allow for prediction, by use of a standard computational package, of the triplet states of molecules not directly measured. Triplet state energies for Chl c1, Chl f, and BChl f were estimated using this method (Table 3, Figure 6).

■

DISCUSSION Coordination State. Table 1 indicates that the fluorescence and phosphorescence emission maxima of the 5-coordinate state tend to be shifted toward shorter wavelengths compared to the 6-coordinate state (see Table 4). However, as discussed in Materials and Methods, it is unknown if the observed shift in 7228



Article

shown to shift the triplet state energies of several chlorin-type molecules by up to 0.11 eV (Table 4), it is possible that pigment−protein interactions could shift the triplet state energies of BChl a and BChl b below the energy of singlet oxygen, which would inhibit energy transfer from the BChl triplet state to O2 and thereby eliminate the need for carotenoids for photoprotection. This is possibly the reason why, for example, the Fenna−Matthews−Olson (FMO) complex contains a number of BChl a molecules but no carotenoids.59 Future work is planned to study the triplet state energy of this pigment−protein complex. As mentioned, the triplet excited state energy could also be affected by pigment−pigment interactions. For example, BChl c, BChl d, and BChl e are known to self-assemble into highly aggregated, semicrystalline structures in the chlorosomes of green sulfur bacteria.60 The triplet state energies of monomeric 6-coordinate BChl c, d, and e are 0.25, 0.32, and 0.33 eV, respectively, higher than the energy of singlet oxygen. Thus, these molecules, in their monomeric state, could readily sensitize the formation of singlet oxygen. However, in a study involving carotenoid-free mutants of BChl c containing chlorosomes, Kim et al.61 demonstrated the existence of a novel photoprotection mechanism that is not based on carotenoids in these highly aggregated antenna structures. In the aggregated state, these molecules experience strong pigment−pigment interactions resulting in a 60 to 80 nm red shift of the Qy absorption maximum by singlet exciton coupling.60 The close proximity of each pigment molecule to its neighbors60,62 also results in strong π−π interactions which may allow the formation of triplet excitons.61 It was shown that the formation of the triplet excitons in the BChl aggregates can be sufficient to lower the triplet excited state energy below that of singlet oxygen, resulting in the enhanced photoprotection observed in the BChl c aggregates of the carotenoid-less mutant.61 Evidence for the formation of triplet excitons in pigment aggregates has recently been demonstrated by Eaton et al.63 using a derivative of the artificial pigment perylene diimide (PDI). In the closely packed PDI aggregates, the triplet state energy was observed, by phosphorescence spectroscopy, to drop by 0.14 eV relative to the monomer.63 Future work is planned to study triplet exciton formation in artificially induced aggregates of BChl c, d, and e. Two known natural systems where coupling between BChl molecules has lowered the triplet states energies below that of singlet oxygen are the reaction center (RC) special pairs of Rhodobacter sphaeroides and Blastochloris viridis (formerly Rhodopseudomonas viridis64).43 These special pairs, which consist of two tightly coupled BChl molecules, show phosphorescence at 1318 nm (0.94 eV) and 1497 nm (0.83 eV) for the BChl a and BChl b containing species, R. sphaeroides and B. viridis, respectively.43 Thus, the triplet state energies of these BChl a and BChl b special pairs are significantly lowered with respect to the corresponding monomers and even below the level of singlet oxygen (0.975 eV). It is unknown if this offers any photoprotection to these organisms, however, it is known that B. viridis does not tolerate oxygen.65 Both organisms still require carotenoid protection for the pigments of their light-harvesting antenna and other RC pigments.66 Vibrational Levels of Fluorescence and Phosphorescence. Inspection of the emission spectra (fluorescence and phosphorescence) on an energy scale (Figure S4, Supporting Information) reveals that for the chlorin- and porphyrin-type

of these states can produce emission maxima intermediate to the different coordination states (see Figure S3 in Supporting Information). One side effect of the selective excitation procedure used in this study is a reduced signal-to-noise due to exciting the sample in a spectral region where its absorbance is intrinsically low (see Chl c2 and BChl e in Figure 4). The data of Table 4, in addition to clearly demonstrating that the local chemical environment can have a significant effect on the triplet state energy, suggest that an empirical relationship exists between the energies of the fluorescence and phosphorescence maxima. Both the fluorescence and phosphorescence bands shift to lower energies as the coordination number increases from 5 to 6 with the singlet excited state shifting ∼0.01 eV and the chlorin-type triplet excited state shifting 0.06 to 0.11 eV (Table 4). However, it is not as simple as a constant energy difference between the singlet excited and triplet excited states, because the fluorescence to phosphorescence energy difference is larger for 6-coodinate chlorin-type molecules compared to 5-coordinated molecules of the same kind (Table 4). As discussed earlier, it is unknown if the changes in the singlet state emissions (fluorescence) are due to differences in coordination state or other solvent effects. However, we can confidently say that the coordination state affects the triplet state energy, because our experiment shows that BChl e possessed two phosphorescence emitting species in the same solvent that are consistent with the 5- and 6coordinated states (Figure 5). The long wavelength form of BChl e possessed a phosphorescence emission peak at 953 nm, consistent with the value predicted by DFT (Table 2, Figure 6), whereas the phosphorescence emission peak of the short wavelength form was shifted by 41 nm (0.05 eV) to 912 nm, a difference consistent with the values reported in the literature for the 5- and 6-coordination states of other molecules (Table 4). Pigment Environment. It is known that the local pigment environment, whether it is a solvent or protein environment, can shift the singlet state energies of (B)Chl molecules as observed through absorption and fluorescence. The Qy absorption maximum of Chl a monomers in pyridine is 671 nm,10 whereas the Qy absorption maxima of Chl a bound in the cytochrome b6 f complex of spinach and Mastigocladus laminosus are 668 and 673 nm, respectively.56 In photosystem I (PS I), however, Chl a molecules have absorption maxima between about 670 and 720 nm.57 For BChl a, the Qy absorption maximum of monomers in pyridine is 781 nm;10 however, bound in the CmsA pigment−protein complex of Chloroflexus aurantiacus, the Qy absorption maximum is shifted to 795 nm.58 Both PS I and the CmsA complex show that a ∼0.01 to 0.03 eV shift in the absorption maximum is possible due to pigment− pigment and pigment−protein interactions. This shift is comparable to the fluorescence maxima shift (∼0.01 to 0.02 eV) between the 5- and 6-coordinated states from Table 4. Because the pigment−solvent interactions explored in Table 4 are known to shift the singlet state energies of (B)Chl molecules by approximately the same amount as pigment− pigment or pigment−protein interactions, it may stand to reason that the triplet state energies also experience a shift due to pigment−pigment or pigment−protein interactions on the order of ∼0.05 eV, similar to the shift observed in Table 4. The energies of the 6-coordinate BChl a and BChl b triplet states are only 0.03 and 0.01 eV higher than the energy of singlet oxygen (0.96 eV). Given that pigment−solvent interactions (via change of coordination state) have been 7229



Article

the upper range of the chlorin-type molecules. Variations within these subgroups depend on the location and type of functional groups attached to the macrocycle, with electron-donating or -withdrawing groups having a strong effect. The results show that it is energetically favorable for each of the molecules studied to sensitize the formation of singlet oxygen. However, the results also suggest it may be possible for pigment−pigment or pigment−protein interactions to lower the (B)Chl triplet state energy below that of singlet oxygen, making sensitization energetically unfavorable. This lowering of the triplet state energy could confer a novel type of photoprotection.

molecules, the locations of the major vibrational peaks are approximately the same, except for a shoulder on the low energy side of the main phosphorescence peak. The bacteriochlorin-type molecules show a much larger deviation for the higher vibrational modes, especially BChl b with a large (but poorly resolved) vibrational band in the phosphorescence that has no corresponding band in the fluorescence. However, the vibrational bands are much more intense with respect to the main band (the 0−0 transition) in phosphorescence compared to fluorescence (Figure 4, Figure S4). This indicates a different coupling strength between the vibrational modes and the electronic transition in triplet and singlet excited states. DFT Calculations. DFT calculations were performed with several simplifications such as truncated phytyl tails and in vacuum calculations with no axial ligands to Mg2+ ion. Nevertheless almost perfect agreement of calculated energy differences with the measurements for the porphyrin and chlorin types (all with deviations within 5% of the measured values) was found (Table 2, Figure 6). At the same time, it is not clear why DFT consistently overstabilizes the triplet states of bacteriochlorin-type molecules. The good linear correlation between the experimental values and DFT predicted values, however, allows one to propose correction curves for the DFT calculations (Figure 6). We grouped molecules into two classes according to their structures (chlorin and bacteriochlorin types) and assumed a constant offset between the DFT-calculated and experimental values. A linear least-squares fit with the slope constrained to one yielded the following corrections of the calculated energies, EDFT, and measured energies, Etriplet, for each class of molecule (error is defined as the standard deviation from the fit): Etriplet,chlorin = E DFT + 0.01± (0.03) eV

(2)

Etriplet,bacteriochlorin = E DFT + 0.22( ±0.02) eV

(3)

■

ASSOCIATED CONTENT

S Supporting Information *

Supporting Information regarding the relative intensities and vibrational features of the fluorescence and phosphorescence emissions are presented as well as the existence of, and correction methods used for, fluorescence reabsorption and the presence of multiple phosphorescence emitting species is discussed. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Spectroscopic and computational studies were supported by the Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences of the U.S. Department of Energy through Grant DE-FG02-09ER16084 to S.S. Work in the laboratory of D.A.B was funded by the Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences of the U.S. Department of Energy through Grant DEFG02-94ER20137. Computational studies were also supported by the DOE, Office of Basic Energy Sciences through Grant DE-FG02-12ER16340 to Y.P. Purification and collection of (B) Chls was a part of research performed in the Photosynthetic Antenna Research Center (PARC), an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Award Number DE-SC 0001035 to R.E.B. The authors would like to thank L. V. Slipchenko and B. Nebgen (Purdue University) for useful discussions regarding quantum chemical calculations.

While other, more sophisticated correction models may produce better results, the small number of data points precludes a more detailed analysis. The results of these calculations can be found in Table 3.

■

CONCLUSION The triplet state energies of 10 natural chlorophyll (Chl a, b, c2, d) and bacteriochlorophyll (BChl a, b, c, d, e, g) molecules and one bacteriopheophytin (BPheo g) have been directly determined under comparable conditions from their phosphorescence spectra. The triplet state energies of three additional moleculesChl c1, Chl f, and BChl fwere calculated using DFT. The major contribution to the spectral properties of a (B)Chl molecule comes from the macrocycle structure (i.e., porphyrin, chlorin, or bacteriochlorin type). Both the fluorescence and phosphorescence transitions (Figure 4, Table 1) of the chlorin- and bacteriochlorin-type molecules can be broadly grouped into their respective categories with the lowest energy fluorescence and phosphorescence transitions associated with the bacteriochlorin type, whereas the one porphyrin type aligns with the upper range of the chlorin-type molecules. Furthermore, both the relative quantum yield and the phosphorescence emission lifetime can also be grouped in this manner, with the chlorin-type molecules possessing longer phosphorescence emission lifetimes and higher relative quantum yields compared to the bacteriochlorin types (Table 1). The one porphyrin-type molecule studied, again, aligns with

■

REFERENCES

(1) Ke, B. Photosynthesis: An Overview. In Photosynthesis: Photobiochemistry and Photobiophysics; Amesz, J., Aro, E. M., Barber, J., Blankenship, R. E., Murata, N., Ort, D. R., Eds.; Kluwer Academic Publishers: Dordrecht, The Netherlands: 2001; Vol. 10, pp 1−46. (2) Scheer, H. An Overview of Chlorophylls and Bacteriochlorophylls: Biochemistry, Biophysics, Functions and Applications. In Chlorophylls and Bacteriochlorophylls; Grimm, B., Porra, R. J., Rüdiger, W., Scheer, H., Eds.; Springer: Dordrecht, The Netherlands, 2006; pp 1−26. (3) Melkozernov, A. N.; Blankenship, R. E. Photosynthetic Functions of Chlorophylls. In Chlorophylls and Bacteriochlorophylls; Grimm, B., Porra, R. J., Rüdiger, W., Scheer, H., Eds.; Springer: Dordrecht, The Netherlands, 2006; pp 397−412. 7230



Article

(4) Kiang, N. Y.; Siefert, J.; Govindjee; Blankenship, R. E. Spectral Signatures of Photosynthesis. I. Review of Earth Organisms. Astrobiology 2007, 7, 222−251. (5) Glaeser, J.; Klug, G. Photo-Oxidative Stress in Rhodobacter Sphaeroides: Protective Role of Carotenoids and Expression of Selected Genes. Microbiology 2005, 151, 1927−1938. (6) Wilkinson, F.; Helman, W. P.; Ross, A. B. Quantum Yields for the Photosensitized Formation of the Lowest Electronically Excited Singlet State of Molecular Oxygen in Solution. J. Phys. Chem. Ref. Data 1993, 22, 113−262. (7) Jen, J. J.; Mackinney, G. On the Photodecomposition of Chlorophyll in Vitro. I. Reaction Rates. Photochem. Photobiol. 1970, 11, 297−302. (8) Briviba, K.; Klotz, L. O.; Sies, H. Toxic And Signaling Effects of Photochemically or Chemically Generated Singlet Oxygen in Biological Systems. Biol. Chem. 1997, 378, 1259−1265. (9) Moss, G. P. Nomenclature of Tetrapyrroles. Pure Appl. Chem. 1987, 59, 779−832. (10) Niedzwiedzki, D. M.; Blankenship, R. E. Singlet and Triplet Excited State Properties of Natural Chlorophylls and Bacteriochlorophylls. Photosynth. Res. 2010, 106, 227−238. (11) Wakao, N.; Yokoi, N.; Isoyama, N.; Hiraishi, A.; Shimada, K.; Kobayashi, M.; Kise, H.; Iwaki, M.; Itoh, S.; Takaichi, S.; et al. Discovery of Natural Photosynthesis Using Zn-Containing Bacteriochlorophyll in an Aerobic Bacterium Acidiphilium rubrum. Plant Cell Physiol. 1996, 37, 889−893. (12) Tsukatani, Y.; Romberger, S. P.; Golbeck, J. H.; Bryant, D. A. Isolation and Characterization of Homodimeric Type-I Reaction Center Complex from Candidatus Chloracidobacterium thermophilum, an Aerobic Chlorophototroph. J. Biol. Chem. 2012, 287, 5720− 5732. (13) Gouterman, M. Spectra of Porphyrins. J. Mol. Spectrosc. 1961, 6, 138−163. (14) Gouterman, M.; Wagnière, G. H.; Snyder, L. C. Spectra of Porphyrins: Part II. Four Orbital Model. J. Mol. Spectrosc. 1963, 11, 108−127. (15) Weiss, C., Jr. The Pi Electron Structure and Absorption Spectra of Chlorophylls in Solution. J. Mol. Spectrosc. 1972, 44, 37−80. (16) Bowers, P. G.; Porter, G. Quantum Yields of Triplet Formation in Solutions of Chlorophyll. Proc. R. Soc. A 1967, 296, 435−441. (17) Borland, C. F.; McGarvey, D. J.; Truscottt, T. G.; Codgell, R. J.; Land, E. J. Photophysical Studies of Bacteriochlorophyll a and Bacteriopheophytin aSinglet Oxygen Generation. J. Photochem. Photobiol., B 1987, 1, 93−101. (18) Maggiora, G. M.; Ingraham, L. L. Chlorophyll Triplet States. Struct. Bonding (Berlin, Ger.) 1967, 2, 126−159. (19) Hameka, H. F. Spin-Orbit Interactions in Organic Molecules. In The Triplet State; Zahlan, A. B.; Androes, G. M.; Hutchison, C. A. Jr.; Hameka, H. F.; Robinson, G. W.; Heineken, F. W.; van der Waals, J. H., Eds.; Cambridge University Press: New York, 1967; pp 1−27. (20) Van der Waals, J. H.; de Groot, M. S. Magnetic Interactions Related to Phosphorescence. In The Triplet State; Zahlan, A. B.; Androes, G. M.; Hutchison, C. A. Jr.; Hameka, H. F.; Robinson, G. W.; Heineken, F. W.; van der Waals, J. H., Eds.; Cambridge University Press: New York, 1967; pp 101−132. (21) Marian, C. M. Spin-Orbit Coupling and Intersystem Crossing in Molecules. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2012, 2, 187−203. (22) Fujimori, E.; Livingston, R. Interactions of Chlorophyll in Its Triplet State With Oxygen, Carotene, Etc. Nature 1957, 180, 1036− 1038. (23) Gomez Maqueo Chew, A.; Frigaard, N. U.; Bryant, D. A. Bacteriochlorophyllide c C-8(2) and C-12(1) Methyltransferases Are Essential for Adaptation to Low Light in Chlorobaculum tepidum. J. Bacteriol. 2007, 189, 6176−6184. (24) Kim, H. Discovery of Novel Photoprotection Mechanisms in Photosynthesis Using Optical Spectroscopy and Biosensor Development. Ph.D. Thesis, Purdue University, West Lafayette, IN, 2007, 1− 121.

(25) Takaichi, S.; Oh-oka, H. Pigment Composition in the Reaction Center Complex from the Thermophilic Green Sulfur Bacterium, Chlorobium tepidum: Carotenoid Glucoside Esters, Menaquinone and Chlorophylls. Plant Cell Physiol. 1999, 40, 691−694. (26) Brockmann, H., Jr.; Lipinski, A. Bacteriochlorophyll, g. A New Bacteriochlorophyll from Heliobacterium chlorum. Arch. Microbiol. 1983, 136, 17−19. (27) Michalski, T. J.; Hunt, J. E.; Bowman, M. K.; Smith, U.; Bardeen, K.; Gest, H.; Norris, J. R.; Katz, J. J. Bacteriopheophytin, g. Properties and Some Speculations on a Possible Primary Role for Bacteriochlorophylls b and g in the Biosynthesis of Chlorophylls. Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 2570−2574. (28) Solov’ev, K. N.; Dvornikov, S. S.; Knyukshto, V. N.; Turkova, A. E. Phosphorence and State of Chlorophyll in Solutions. J. Appl. Spectrosc. 1983, 38, 73−79. (29) Takiff, L.; Boxer, S. G. Phosphorescence Spectra of Bacteriochlorophylls. J. Am. Chem. Soc. 1988, 110, 4425−4426. (30) Krawczyk, S. The Effects of Hydrogen Bonding and CoordinationInteraction in Visible Absorption and Vibrational Spectra of Chlorophyll a. Biochim. Biophys. Acta, Bioenerg. 1989, 976, 140−149. (31) Callahan, P. M.; Cotton, T. M. Assignment of Bacteriochlorophyll a Ligation State from Absorption and Resonance Raman Spectra. J. Am. Chem. Soc. 1987, 109, 7001−7007. (32) Szalay, L.; Tombácz, E.; Singhal, G. S. Effect of Solvent on the Absorption Spectra and Stokes’ Shift of Absorption and Fluorescence of Chlorophylls. Acta Phys. Acad. Sci. Hung. 1974, 35, 29−36. (33) Hartwich, G.; Fiedor, L.; Simonin, I.; Cmiel, E.; Schäfer, W.; Noy, D.; Scherz, A.; Scheer, H. Metal-Substituted Bacteriochlorophylls. 1. Preparation and Influence of Metal and Coordination on Spectra. J. Am. Chem. Soc. 1998, 120, 3675−3683. (34) Noy, D.; Yerushalmi, R.; Brumfeld, V.; Ashur, I.; Scheer, H.; Baldridge, K. K.; Scherz, A. Optical Absorption and Computational Studies of [Ni]-Bacteriochlorophyll-a. New Insight into Charge Distribution Between Metal and Ligands. J. Am. Chem. Soc. 2000, 122, 3937−3944. (35) Rätsep, M.; Linnanto, J.; Freiberg, A. Mirror Symmetry and Vibrational Structure in Optical Spectra of Chlorophyll a. J. Chem. Phys. 2009, 130, 194501. (36) Fiedor, L. Hexacoordination of Bacteriochlorophyll in Photosynthetic Antenna LH1. Biochemistry 2006, 45, 1910−1918. (37) Evans, T. A.; Katz, J. J. Evidence for 5- and 6-Coordinated Magnesium in Bacterio-Chlorophyll a from Visible Absorption Spectroscopy. Biochim. Biophys. Acta 1975, 396, 414−426. (38) Fujiwara, M.; Tasumi, M. Resonance Raman and Infrared Studies on Axial Coordination to Chlorophylls a and b in Vitro. J. Phys. Chem. 1986, 90, 250−255. (39) Koyama, Y.; Kakitani, Y.; Limantara, L.; Fujii, R. Effects of Axial Coordination, Electronic Excitation and Oxidation on Bond Orders in the Bacteriochlorin Macrocycle, and Generation of Radical Cation on Photo- Excitation of in Vitro and in Vivo Bacteriochlorophyll a Aggregates: Resonance Raman Studies. In Chlorophylls and Bacteriochlorophylls; Grimm, B., Porra, R. J., Rüdiger, W., Scheer, H., Eds.; Springer: Dordrecht, The Netherlands, 2006; pp 324−335. (40) Telfer, A.; Pascal, A. A.; Bordes, L.; Barber, J.; Robert, B. Fluorescence Line Narrowing Studies on Isolated Chlorophyll Molecules. J. Phys. Chem. B 2010, 114, 2255−2260. (41) Neverov, K. V.; Santabarbara, S.; Krasnovsky, A. A., Jr. Phosphorescence Study of Chlorophyll d Photophysics. Determination of the Energy and Lifetime of the Photo-Excited Triplet State. Evidence of Singlet Oxygen Photosensitization. Photosynth. Res. 2011, 108, 101−106. (42) Krasnovsky, A. A., Jr.; Cheng, P.; Blankenship, R. E.; Moore, T. A.; Gust, D. The Photophysics of Monomeric Bacteriochlorophylls c and d and Their Derivatives: Properties of the Triplet State and Singlet Oxygen Photogeneration and Quenching. Photochem. Photobiol. 1993, 57, 324−330. (43) Takiff, L.; Boxer, S. G. Phosphorescence from the Primary Electron Donor in Rhodobacter sphaeroides and Rhodopseudomonas 7231



Article

monas sulfoviridis to the Genus Blastochloris Gen. Nov. Int. J. Syst. Bacteriol. 1997, 47, 217−219. (65) Lang, F. S.; Oesterhelt, D. Microaerophilic Growth and Induction of the Photosynthetic Reaction Center in Rhodopseudomonas viridis. J. Bacteriol. 1989, 171, 2827−2834. (66) Ziegelhoffer, E. C.; Donohue, T. J. Bacterial Responses to Photo-Oxidative Stress. Nat. Rev. Microbiol. 2009, 7, 856−863.

viridis Reaction Centers. Biochim. Biophys. Acta, Bioenerg. 1988, 932, 325−334. (44) Krasnovsky, A. A., Jr. Photoluminescence of Singlet Oxygen in Pigment Solutions. Photochem. Photobiol. 1979, 29, 29−36. (45) Dvornikov, S. S.; Knyukshto, V. N.; Solovev, K. N.; Tsvirko, M. P. Phosphorescence of Chlorophyllis a and b and Their Pheophytins. Opt. Spectrosc. 1979, 46, 385−388. (46) Jockusch, S.; Turro, N. J.; Thompson, E. K.; Gouterman, M.; Callis, J. B.; Khalil, G. E. Singlet Molecular Oxygen by Direct Excitation. Photochem. Photobiol. Sci. 2008, 7, 235−239. (47) Penzkofer, A.; Leupacher, W. Fluorescence Behaviour of Highly Concentrated Rhodamine 6G Solutions. J. Lumin. 1987, 37, 61−72. (48) Steiner, R.; Cmiel, E.; Scheer, H. Chemistry of Bacteriochlorophyll b: Identification of Some (Photo)Oxidation Products. Zeitschrift für Naturforsch 1983, 38 c, 748−752. (49) 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.; et al. Gaussian 09, revision C.01; Gaussian, Inc.: Wallingford, CT, 2009. (50) Godbout, N.; Salahub, D. R.; Andzelm, J.; Wimmer, E. Optimization of Gaussian-Type Basis Sets For Local Spin Density Functional Calculations. Part I. Boron through Neon, Optimization Technique and Validation. Can. J. Chem. 1992, 70, 560−571. (51) Krasnovsky, A. A., Jr. Delayed Fluorescence and Phosphorescence of Plant Pigments. Photochem. Photobiol. 1982, 36, 733−741. (52) Hindman, J. C.; Kugel, R.; Wasielewski, M. R.; Katz, J. J. Coherent Stimulated Light Emission (Lasing) in Covalently Linked Chlorophyll Dimers. Proc. Natl. Acad. Sci. U.S.A. 1978, 75, 2076−2079. (53) Seely, G.; Connolly, J. Fluorescence of Photosynthetic Pigments in Vitro. In Light Emission by Plants and Bacteria; Govindjee; Amesz, J.; Fork, D. C., Eds.; Academic Press, Inc: Orlando, FL, 1986; pp 99−133. (54) Broyde, S. B.; Brody, S. S. Emission Spectra of Chlorophyll-a in Polar and Nonpolar Solvents. J. Chem. Phys. 1967, 46, 3334−3340. (55) Pospísil, P. Theoretical Simulation of Temperature Induced Increase of Quantum Yield of Minimum Chlorophyll Fluorescence ΦF(0). J. Theor. Biol. 1998, 193, 125−130. (56) Zhang, H.; Huang, D.; Cramer, W. A. Stoichiometrically Bound Beta -Carotene in the Cytochrome b6 f Complex of Oxygenic Photosynthesis Protects Against Oxygen Damage. J. Biol. Chem. 1999, 274, 1581−1587. (57) Savikhin, S.; Xu, W.; Soukoulis, V.; Chitnis, P. R.; Struve, W. S. Ultrafast Primary Processes in Photosystem I of the Cyanobacterium Synechocystis Sp. PCC 6803. Biophys. J. 1999, 76, 3278−3288. (58) Sakuragi, Y.; Frigaard, N.; Shimada, K.; Matsuura, K. Association of Bacteriochlorophyll a with the CsmA Protein in Chlorosomes of the Photosynthetic Green Filamentous Bacterium Chloroflexus auranticus. Biochim. Biophys. Acta 1999, 1413, 172−180. (59) Li, Y. F.; Zhou, W.; Blankenship, R. E.; Allen, J. P. Crystal Structure of the Bacteriochlorophyll a Protein from Chlorobium tepidum. J. Mol. Biol. 1997, 271, 456−471. (60) Orf, G. S.; Blankenship, R. E. Chlorosome Antenna Complexes from Green Photosynthetic Bacteria. Photosynth. Res. 2013, 116, 315− 331. (61) Kim, H.; Li, H.; Maresca, J. A.; Bryant, D. A.; Savikhin, S. Triplet Exciton Formation as a Novel Photoprotection Mechanism in Chlorosomes of Chlorobium tepidum. Biophys. J. 2007, 93, 192−201. (62) Ganapathy, S.; Oostergetel, G. T.; Wawrzyniak, P. K.; Reus, M.; Gomez Maqueo Chew, A.; Buda, F.; Boekema, E. J.; Bryant, D. A.; Holzwarth, A. R.; de Groot, H. J. M. Alternating syn-anti Bacteriochlorophylls Form Concentric Helical Nanotubes in Chlorosomes. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 8525−8530. (63) Eaton, S. W.; Shoer, L. E.; Karlen, S. D.; Dyar, S. M.; Margulies, E. A.; Veldkamp, B. S.; Ramanan, C.; Hartzler, D. A.; Savikhin, S.; Marks, T. J.; et al. Singlet Exciton Fission in Polycrystalline Thin Films of a Slip-Stacked Perylenediimide. J. Am. Chem. Soc. 2013, 135, 14701−14712. (64) Hiraishi, A. Transfer of the Bacteriochlorophyll b-Containing Phototrophic Bacteria Rhodopseudomonas viridis and Rhodopseudo 7232


Triplet Excited State Energies and Phosphorescence Spectra of

Recommend Documents