Two-Dimensional Electronic-Vibrational Spectroscopy of Chlorophyll a

Subscriber access provided by Chicago State University

Letter

Two-Dimensional Electronic-Vibrational Spectroscopy of Chlorophyll a and b Nicholas Henry Cohen Lewis, and Graham R. Fleming J. Phys. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acs.jpclett.6b00037 • Publication Date (Web): 19 Feb 2016 Downloaded from http://pubs.acs.org on February 20, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 19

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

Two-Dimensional Electronic-Vibrational Spectroscopy of Chlorophyll a and b Nicholas H. C. Lewis†,‡,¶ and Graham R. Fleming∗,†,‡,¶ †Department of Chemistry, University of California, Berkeley, California 94720, United States ‡Molecular Biophysics and Integrated Bioimaging Division, Lawrence Berkeley National Laboratory, Berkeley, California 94720, United States ¶Kavli Energy Nanoscience Institute at Berkeley, Berkeley, California 94720, United States E-mail: [email protected]

1

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

Abstract We present Two-Dimensional Electronic-Vibrational (2DEV) spectra of isolated chlorophyll a and b in deuterated ethanol. We excite the Q-band electronic transitions, and measure the effects on the carbonyl and C=C double bond stretch region of the infrared spectrum. With the aid of density functional theory calculations, we provide assignments for the major features of the spectrum. We show how the 2DEV spectra can be used to readily distinguish different solvation states of the chlorophyll, with features corresponding to the minority pentacoordinate magnesium (Mg) species being resolved along each dimension of the 2DEV spectra from the dominant hexacoordinate Mg species. These assignments represent a crucial first step towards the application of 2DEV spectroscopy to chlorophyll containing pigment-protein complexes.

Graphical TOC Entry

Keywords: Ultrafast Spectroscopy, 2D Electronic-VIbrational Spectroscopy

2


Page 2 of 19

Page 3 of 19

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


Chlorophyll serves a central role in the photosynthetic apparatus of most photosynthetic organisms, including plants. In higher plants, chlorophyll molecules (structure shown in Figure 1), in particular chlorophyll a (Chl a) and chlorophyll b (Chl b) serve as the major light absorbing molecules in photosynthetic pigment-protein complexes (PPCs). 1 The majority of Chl a and Chl b molecules can be found bound in the light-harvesting complex II (LHCII), which is the primary light harvesting PPC in plants. In this complex, the major role of the Chl a and b is to absorb light, and to transfer this energy through space until it reaches the reaction center, where the electronic excitation can drive charge separation, ultimately leading to the excitation energy being converted to chemical energy. A detailed understanding of the excitation energy transfer pathways in LHCII and related PPCs is essential for a complete and accurate understanding of the workings of the photosynthetic apparatus, and much progress has been made in recent years in mapping out the energetic landscape of these PPCs using multidimensional electronic spectroscopy. 2–11 There still remain important questions, however, about the details of the relationship between the energetic structures of these PPCs, which are readily accessible with electronic spectroscopies, and the spatial structures of the actual pigment layout in the protein. Recently, we have proposed a method by which two-dimensional electronic-vibrational spectroscopy (2DEV) could be used to make this link between the energetic and spatial domains, by using localized vibrational modes as a proxy for position within a complex system, such as a photosynthetic PPC. 12 This method critically relies on an ability to resolve and assign both electronic and vibrational features for each chromophore and electronic state of interest. A prerequisite for this to be possible is a detailed understanding of the spectral response of the relevant pigments in isolation. In this Letter, we present 2DEV spectra of isolated Chl a and Chl b dissolved in deuterated ethanol. By measuring the dependence of these spectra on the relative polarizations of the visible excitation and infrared (IR) probe laser pulses, in conjunction with density functional theory (DFT) calculations, we provide assignments for the major features of these spectra. We also provide evidence of features in the 2DEV

3



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 19

spectra that correspond to different solvation states of the Chl molecules, distinguishing the contributions of pentacoordinate and hexacoordinate magnesium (Mg) species. 32

71 31

b

21

R 5

3

a 2 1

m

4

A

6

9

N Mg

10

N

19

D

N 16

82

8

B

N

20

18

81

7

11

C

14

12

18

1

15

17

131

O

O

133

173

O

121

13

132

171

172

E

R = CH3 (Chl a), COH (Chl b)

O O 134

P1

P111

P71

P2 P5

P13

P9

P3 P31

P7 P4

P151

P6

P15

P11 P8

P10

P12

P14

P16

Figure 1: Structure of Chlorophyll molecule with IUPAC numbering and ring labels, and labelling of the distinct types of carbons on the porphyrin macrocycle (Ca and Cb on the pyrrole groups and Cm methine groups). The 2DEV spectroscopic technique provides a method for correlating a molecule’s electronic transitions, in the visible region of the electromagnetic spectrum, with its IR active vibrations. 13–15 The method is closely related to transient IR absorption pump-probe spectroscopy, whereby a molecule is electronically excited, and the resulting changes in the IR absorption is observed in a time resolved manner. Unlike transient IR absorption, however, 2DEV uses Fourier techniques to provide spectral resolution of the electronic excitation axis of the spectrum, in addition to the IR detection axis. This removes the trade-off between temporal resolution and frequency resolution and reveals new types of information concerning correlations in the system. 12,14,15 2DEV also makes it possible to more readily resolve and assign overlapping spectral features, by spreading the spectrum along an additional axis and 4


Page 5 of 19

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


thereby providing more information about their spectral structures. A detailed description of the experimental methods is provided in the Supporting Information. An additional technique that can aid in providing assignments for bands in complex spectra is the use of the polarization dependent response and the spectral anisotropy. In combination, the isotropic (magic angle) response Siso , the anisotropic response Saniso and the anisotropy parameter r = Saniso /Siso can be used to assign spectral features based on the relative angles between their transition dipole moments. 16 The anisotropy parameter in particular is very useful, as for a spectrally isolated oscillator it can be directly related to the angle θ between the transition dipole moments for the pumped and probed states, according to the well known relation r(θ) = 51 (3 cos2 θ − 1). Because of this property, it is also possible to use the dynamics of the anisotropy parameter to measure such quantities as the rotational correlation function, or energy transfer dynamics. 17,18 At a given waiting time, the anisotropy for a spectrally isolated transition is constrained to some frequency independent value in the region [−0.2, 0.4]. However, when spectral features of opposite sign interfere, it is possible for r to take on any value and demonstrate substantial spectral structure. 17 In this common scenario, r becomes much less useful as a quantitative measure of relative angles, but can still be used to identify features in complex spectra. The transient IR pump-probe spectra of the high energy C=O and C=C stretch region of Chl a and Chl b are shown in Figure 2. In each case there are two major positive features, which are due to vibrations on the electronic ground state. The stronger of these appears at 1660 cm−1 and is assignable to a C=O stretching motion. Based on a DFT harmonic frequency analysis this is most likely the antisymmetric combination of the C−131 ketone group and the C−133 ester group on ring E (see Figure 1 for labels). The weaker positive band appears around 1550 cm−1 , and is attributable to a C=C stretch mode localized on the macrocycle. This band has been previously assigned to a combination of Cb Cb and Ca Cm stretching motions. 19–21 Based on the DFT results, this mode is primarily localized on rings C and E, and can be considered as predominantly a pyrrole stretching mode on

5


Det. Freq. (cm -1 )

mΔT/T

10

b)

5 0 -5

5 0 -5

1550

-10

10

1650

250 fs

1600

Chl a

a)

1550 1600 1650 1700

-10

0

10

20

-5

-10 1550 1600 1650 1700

60

70

80

90

1700

10

1650

5 0

1600

d) Det. Freq. (cm -1 )

250 fs

0

50

mΔT/T

mΔT/T

5

40

-5

1550

Chl b *

30

Waiting Time (ps)

Frequency (cm-1)

c)

mΔT/T

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 6 of 19

1700


-10

0

10

Frequency (cm-1)

20

30

40

50

60

70

80

90

Waiting Time (ps)

Figure 2: Isotropic transient IR absorption spectra of Chl a (a, b) and Chl b (c, d). Panels a) and c) show the transient spectra at waiting time t2 = 250 fs, and panels b) and d) show the full time resolved spectra. The positive (yellow/red) features indicate vibrations on the electronic ground state, and the negative (blue) features indicate vibrations on an electronic excited state. The ∗ in panel c) indicates an artifact introduced by atmospheric water.

6


Page 7 of 19

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


ring C. In addition to these ground state vibrations, both the Chl a and the Chl b spectra show a strong negative band peaked near 1615 cm−1 , and showing a long broad shoulder to lower frequencies. This band is most readily assignable to the carbonyl stretching mode on an electronic excited state. The shoulder might be due to either a second mode centered near 1580 cm−1 , or to a positive ground electronic state feature interfering with the excited electronic state feature. The assignment of this shoulder will be made clear with the 2DEV spectra. In addition to these features common to both Chl species, the Chl b spectrum additionally shows a narrow negative feature at 1640 cm−1 . This is unique to the Chl b spectrum, and therefore is likely related to the formyl group at position C−71 on ring B, which distinguishes Chl b from Chl a (see Figure 1). Because it is related to an electronic excited state and is not readily paired with an electronic ground state feature, it is difficult to assign precisely. It may be either a C=O stretch mode, or a distortion induced in a pyrrole mode on ring B due to the nearby electron withdrawing group. A sharp positive feature is also observed in the Chl b spectrum at 1655 cm−1 , which is the result of an artifact caused by a depletion in the IR laser intensity by atmospheric H2 O being incompletely normalized from the final spectrum, and is not indicative of a Chl b vibrational band. This artifact does not significantly affect the 2DEV spectra due to the greater degree of signal processing necessary for the construction of 2D spectra. In making these assignments, we are proposing that all of the observed bands are results of primarily a 1 ← 0 vibrational transition, in which case the sign of the band is a direct indication of the electronic state on which this vibrational motion is occurring. It is also possible, in principle, to observe positive and negative features due to stimulated emission or induced absorption by excited vibrational states on either the ground or excited electronic states. These types of features would decay with the lifetime of the vibrational excited state, which would typically be on the order of ∼ 1 − 10 ps. All of the features in the current spectra decay on a substantially longer timescale, commensurate with the Qy lifetimes for isolated Chl a and Chl b. 22 Therefore we can rule out the possibility that the direct probing

7



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

of vibrationally excited states are contributing significantly to these spectra. The visible laser pulses used in these experiments had sufficient bandwidth to excite both the Qy and the Qx electronic transitions (see the Supporting Information), for both Chl a and Chl b. The pump-probe spectra inherently integrate over both these contributions, but by performing 2DEV measurements, these features can be resolved. The polarization sensitive 2DEV spectra of both Chl a and Chl b are shown in Figure 3 for a waiting time t2 = 250 fs. This time was chosen to be late enough to prevent any contribution from pulse overlap effects, yet early enough to avoid any substantial orientational relaxation which would complicate the anisotropy data, though rotational contributions are also substantially suppressed by performing the experiments with the sample at 77 K (the rotational lifetime in ethanol at 77 K is ∼ 100 ns according to the Debye relation). 23 All of the major spectral features observed at this early time are maintained to at least 250 ps. For the Chl a spectra in Figure 3 (a-c) there are clearly two major excitation bands, at around 14650 cm−1 and 15400 cm−1 . These structures are readily assignable from the electronic linear absorption. The intense anisotropic lower energy band is the Qy transition, while the weaker isotropic band is the x-polarized feature known to be a mixture between the Qx electronic transition and a vibronic excitation of the Qy band. 24 Due to substantial overlap between different spectral bands, the precise angles between pumped and probed transition dipole moments cannot be readily extracted, but the qualitative polarization dependence of these bands is generally consistent with the assignments we have provided for these vibrational bands and the computed transition dipole moments for these modes. For the Chl b spectra shown in Figure 3 (d-f) there is an intense anisotropic band at 15100 cm−1 which is due to the Qy transition. Strong signals corresponding to exciting the Qx transition are not observed for Chl b, but weak features located near 16600 cm−1 could be due to this electronic excitation. It is unclear why the Qx band is so weak in the Chl b spectra relative to those in the Chl a spectra, but this may be due to the lesser degree of vibronic mixing between Qx and Qy in Chl b, 24 or simply due to poorer overlap with the excitation laser

8


Page 8 of 19

Page 9 of 19

0.15

1700

a)

Chl b d)

1700

Chl a

0.2 0.15

-0.05

1660

0.1

1620

0.05 0 -0.05

1580

0

Det. Freq. (cm -1 )

1660 1620

0.05

1580

Det. Freq. (cm -1 )

0.1

-0.1

-0.15 × 10 4

0

1580

-0.01 -0.02

1.6

0.04

0.02 0.01 0 -0.01 -0.02

-0.04

1.45

1.65

× 10 4

1.5

1.55

1.6

1.65

Exc. Freq. (cm -1 ) 0.4

1700

c)

× 10 4

1540

1540

-0.04

1.55

1.7

-0.03

-0.03

1.5

1.65

0.03

Det. Freq. (cm -1 )

1620

0.01

Exc. Freq. (cm -1 )

1.6

1700

1700 1660

0.02

e)

0.03

1.45

1.55

Exc. Freq. (cm -1 ) 0.04

1.4

1.5

f)

1.7

× 10 4

0.4 0.3

1660 1620

0.1 0

1580

1580

0

Det. Freq. (cm -1 )

1660 1620

0.1

0.2

-0.1

1540

1540

-0.1 -0.2

1.4

1.45

1.5

1.55

1.6

Exc. Freq. (cm -1 )

1.65 × 10

Anisotropy r

Det. Freq. (cm -1 )

0.3 0.2

Anisotropic

Det. Freq. (cm -1 )

1.45

1.65

1660

1.6

1620

1.55

1580

1.5

-0.2

1700

1.45

Exc. Freq. (cm -1 )

b)

-0.15

1540

1540

-0.1

1.4

Isotropic

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


-0.2

1.45

4

1.5

1.55

1.6

1.65

Exc. Freq. (cm -1 )

1.7 × 10

4

Figure 3: Polarization sensitive 2DEV spectra of the carbonyl stretch region of Chl a (a, b, c) and Chl b (d, e, f) at waiting time t2 = 250 fs. Solid contours indicate positive values for the spectrum, whereas dashed contours indicate negative values. The isotropic response Siso is shown in panels a) and d), the anisotropic response Saniso in panels b) and e), and the anisotropy parameter r is panels c) and f). The anisotropy values are overlaid by the contour plot of the isotropic spectra for the same waiting time. The value of r can diverge in the vicinity of nodes in the isotropic spectrum, so the color bars in panels c) and f) are truncated to [−0.2, 0.4], the allowed range for an isolated transition.

9



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. It is important to note that while we certainly observe features in the 2DEV spectra which result from initially exciting the Qx band, at least for Chl a, this does not imply that the Qx electronic state has significant population at 250 fs. The lifetime of electronic relaxation from Qx to Qy has been calculated from non-adiabatic excited state molecular dynamics to be 128 fs for Chl a and 208 fs for Chl b. 25 In each case it is likely that the true rate is faster than observed in these calculations, due to the substantially overestimated Qx − Qy energy gap and the neglect of the vibronic mixing shown to exist between these states. 24 Therefore, by 250 fs, the majority of the population has already relaxed to the Qy state, and the observation of spectral bands at the Qx excitation energy are due to population in Qy which merely originated as population in Qx . The cross-correlation times of the visible and infrared laser pulses used in these experiments were measured to be ∼ 90 fs, comparable to the Qx lifetime, and therefore it is unlikely that any vibrational features attributable to population in the Qx state could be observed with confidence in these 2DEV spectra. For the isotropic response of both Chl a and Chl b, a significant shoulder on the low vibrational frequency side of the primary excited electronic state feature is observed. As mentioned previously this could reflect a second vibrational mode, but this interpretation is discredited by the anisotropic response shown in Figure 3 (b, e). For the anisotropic response of both molecules, there appears to be a single negative anisotropic feature. Indeed, by examining the anisotropy parameter r, shown in Figure 3 (c, f), a feature appears at around 1590 cm−1 with very large values for the anisotropy, around 0.4 for Chl b and reaching 0.6 in the Chl a spectrum. For a region where the isotropic spectrum has significant amplitude, this type of structure in the anisotropy strongly implies that there is a positive feature at this position in the spectrum, as the only way to obtain a value for r greater than 0.4 is when multiple features with opposite sign interfere. In this case, because the positive band is not sufficiently intense to fully cancel the strong negative feature, the value for r does not diverge, but in the case where the negative feature is anisotropic (as is here) with a sufficiently strong,

10


Page 10 of 19

Page 11 of 19

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


primarily isotropic positive feature, r can take on an arbitrarily large value. In this case, the positive feature is due to the well known band observed in the ground state IR absorption and resonance Raman of metallochlorins. 19–21 This vibrational mode is delocalized over the majority of the chlorin macrocycle, with a character of predominantly a Ca Cm stretch with some contribution of a Cb Cb stretch. The structure of the anisotropy does not allow us to determine with certainty whether the anisotropic negative feature is composed of a single broadened vibrational transition (presumably with C=O stretch character) or of two modes, with a lower energy transition related to the C=C mode. For both the Chl a and Chl b anisotropic spectra this feature has the qualitative appearance of a single band, which seems to suggest that it is likely to be predominantly due to a single broad transition. So far, we have provided assignments for most of the major features appearing in the transient IR absorption and 2DEV spectra. There are, however, several smaller features which are evident in the 2DEV spectra for which the assignments are less immediately obvious. These include weak positive bands that appear in both the Chl a and Chl b spectra at higher energies than the ground electronic state C=O stretch mode at 1660 cm−1 along both axes, as well as a substantial negative feature in the Chl b spectra displaced to higher energies along both axes from the dominant excited electronic state feature. The positive feature in the Chl a spectra appears at an excitation frequency of ∼ 14900 cm−1 and a detection frequency of 1680 cm−1 , while the corresponding feature in the Chl b spectra appears at an excitation frequency near ∼ 15350 cm−1 and a detection frequency of 1690 cm−1 . The negative feature appears at roughly the same excitation frequency, with a detection frequency near 1640 cm−1 . The assignment of these features is not immediately clear, as there is no known electronic excited state that lies between Qy and Qx , and they decay on the timescale of Qy relaxation and therefore cannot be attributed to vibrational excited states. We propose that these features are due to different coordination states of the central Mg. In nucleophilic solvents, like ethanol, it is possible for the Mg to exist in either a pentacoordinate or a hexacoordinate states, in which it has bound to either one or two

11



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

solvent molecules. It has been shown using resonance Raman and IR absorption that for Chl and Bacteriochlorophyll (BChl) these two solvation states can be distinguished using several vibrational bands in this high frequency region. 19,26,27 These studies show that for Chl a, Chl b and BChl a in methanol, the dominant species is the hexacoordinate state. For ethanol, both configurations are present, with Cotton and Van Duyne showing that Bchl a is predominantly in the hexacoordinate state while Fujiwara and Tasumi argue that this is reversed for Chl a and b, which are predominantly pentacoordinate. 26,27 In addition to these vibrational spectroscopies, it has been argued by J. J. Katz and co. that these solvation states can be distinguished by shifts in the electronic absorption spectra of the Q bands. 28,29 For Chl a dissolved in methanol, they show that in addition to the primary Qy absorption at 665 nm that is due to the hexacoordinate species, there is also a minor component located at 650 nm which they attribute to the pentacoordinate state. Using 2DEV we can correlate these shifts along the electronic and vibrational axes, and provide direct evidence that they are indeed due to the same species. In contrast to Fujiwara and Tasumi, we conclude that in ethanol the hexacoordinate state is the dominant species for both Chl a and Chl b. While not impossible, it would be surprising if such a dramatic change in the dominant solvation state of the Mg would arise due to the differences between methanol and ethanol, or between Chl and BChl. It is also possible that this has a temperature dependence, as Katz and co. have demonstrated that in n-propylether, BChl forms a mixture of penta- and hexacoordinate at 228 K, but becomes entirely hexacoordinate at 160 K. 29 Table 1: Chl S1 ← S0 excitation energies (cm−1 )

Chl a Chl b

No. 0 18346 18271

ethanol molecules 1Z 1E 2 18326 18213 18092 18159 18031 17769

To confirm our assignment of the blue-shifted bands as a minority population of pentaco12


Page 12 of 19

Page 13 of 19

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


ordinate Chl molecules, we performed a series of DFT and time-dependent DFT (TD-DFT) calculations on isolated Chl a and Chl b, on these species with a single ethanol molecule coordinated to the face of the molecule either E or Z with respect to the phytyl chain, and with the Mg coordinated to two ethanol molecules, one on each face. For each of the equilibrium ground state geometries for both Chl a and Chl b in the various solvation states, we performed TD-DFT calculations to determine estimates for the S1 ← S0 excitation energies. In each case, the calculation correctly identified S1 with the Qy transition. The results of these calculations are shown in Table 1. For Chl a, the energy difference between the hexacoordinate species and the Z pentacoordinate species is 234 cm−1 , and for the E pentacoordinate species is 121 cm−1 . For Chl b these differences are 390 cm−1 and 262 cm−1 , respectively. These values compare reasonably well to the observed energy differences of ∼ 250 cm−1 for Chl a and ∼ 340 cm−1 for Chl b. The TD-DFT calculations may not quantitatively predict the correct energy differences, but it does correctly predict the smaller energy difference for Chl a compared to Chl b. These values for the changes in electronic excitation energy for the various solvation states, together with the previously observed differences in the vibrational and electronic absorption spectra for the different states suggest that this is a plausible assignment for these spectral features, and that, at least at low temperature, the hexacoordinate species is the dominant solvation state for Chl a and Chl b in ethanol. In principle, it should be possible to use the 2DEV spectra to measure the exchange rate between the different coordination states of the Mg, in a similar manner as has been demonstrated with 2DIR spectroscopies. 30,31 We do not see evidence of a significant degree of exchange between these species, for either Chl a or Chl b, within 250 ps. Given that these experiments were performed at cryogenic temperatures (77 K), with any significant energetic barrier separating the different species, and with diffusion essentially eliminated, such a slow rate of exchange is reasonable. It would be interesting in future work to study the temperature dependence of the 2DEV spectra to determine the magnitude of this energy barrier between the pentacoordinate and hexacoordinate Chl species in varying solvents, as

13



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

well as to study how the ratio between these species changes at higher temperatures. To conclude, we have presented transient IR absorption spectra and 2DEV spectra of isolated Chl a and Chl b dissolved in deuterated ethanol and held at 77 K. We have provided assignments for the major spectral features that appear in the 1540 − 1700 cm−1 spectral region, including bands assignable to C=O and C=C stretch modes on the electronic ground state, and the C=O stretch on the Qy electronic excited state. For Chl a we observe strong features due to initial excitation of the Qx band, which either do not appear or are very weak in the Chl b spectra. The assignments were made by comparing the observed spectra with the well known IR and resonance Raman spectra of these Chl species, using DFT calculations, and by taking advantage of the polarization dependence of the spectra, which is capable of revealing spectral features which cannot be otherwise resolved. We have also observed features resolvable from the main Qy bands along both the electronic and the vibrational axes, which we assign to be the different solvation states of the Chl molecules, with either a pentacoordinate or a hexacoordinate Mg. This work demonstrates the potential usefulness of 2DEV as a technique for distinguishing molecular conformers, and serves as an important first step towards using 2DEV to study the electronic and vibrational properties of photosynthetic PPCs.

Acknowledgement The authors would like to thank Tom Oliver for helpful discussions. This work was supported by the Director, Office of Science, Office of Basic Energy Sciences, US Department of Energy under Contract DE-AC02-05CH11231, and the Division of Chemical Sciences, Geosciences and Biosciences Division, Office of Basic Energy Sciences through Grant DEAC03-76F000098 (at Lawrence Berkeley National Laboratory and University of California, Berkeley). The DFT calculations were performed using the computation resources at the College of Chemistry Molecular Graphics facility, which is funded by NSF under Contract

14


Page 14 of 19

Page 15 of 19

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


CHE-0840505.

Supporting Information Available Additional information about the experimental methods, sample preparation and the electronic structure calculations are provided in the Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org/.

References (1) Blankenship, R. E. Molecular Mechanisms of Photosynthesis; Blackwell Publishing, 2002. (2) Calhoun, T. R.; Ginsberg, N. S.; Schlau-Cohen, G. S.; Cheng, Y.-C.; Ballottari, M.; Bassi, R.; Fleming, G. R. Quantum Coherence Enabled Determination of the Energy Landscape in Light-Harvesting Complex II. The Journal of Physical Chemistry B 2009, 113, 16291–16295. (3) Schlau-Cohen, G. S.; Calhoun, T. R.; Ginsberg, N. S.; Read, E. L.; Ballottari, M.; Bassi, R.; van Grondelle, R.; Fleming, G. R. Pathways of Energy Flow in LHCII from Two-Dimensional Electronic Spectroscopy. The Journal of Physical Chemistry B 2009, 113, 15352–15363. (4) Schlau-Cohen, G. S.; Calhoun, T. R.; Ginsberg, N. S.; Ballottari, M.; Bassi, R.; Fleming, G. R. Spectroscopic Elucidation of Uncoupled Transition Energies in the Major Photosynthetic Light-Harvesting Complex, LHCII. Proceedings of the National Academy of Sciences 2010, 107, 13276–13281. (5) Hayes, D.; Engel, G. S. Extracting the Excitonic Hamiltonian of the Fenna-Matthews-

15



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

Olson Complex Using Three-Dimensional Third-Order Electronic Spectroscopy. Biophysical Journal 2011, 100, 2043–2052. (6) Anna, J. M.; Ostroumov, E. E.; Maghlaoui, K.; Barber, J.; Scholes, G. D. TwoDimensional Electronic Spectroscopy Reveals Ultrafast Downhill Energy Transfer in Photosystem I Trimers of the Cyanobacterium Thermosynechococcus Elongatus. The Journal of Physical Chemistry Letters 2012, 3, 3677–3684. (7) Ostroumov, E. E.; Mulvaney, R. M.; Anna, J. M.; Cogdell, R. J.; Scholes, G. D. Energy Transfer Pathways in Light-Harvesting Complexes of Purple Bacteria as Revealed by Global Kinetic Analysis of Two-Dimensional Transient Spectra. The Journal of Physical Chemistry B 2013, 117, 11349–11362. (8) Fidler, A. F.; Singh, V. P.; Long, P. D.; Dahlberg, P. D.; Engel, G. S. Probing Energy Transfer Events in the Light Harvesting Complex 2 (LH2) of Rhodobacter Sphaeroides with Two-Dimensional Spectroscopy. The Journal of Chemical Physics 2013, 139, 155101. (9) Dostál, J.; Vácha, F.; Pšenčík, J.; Zigmantas, D. 2D Electronic Spectroscopy Reveals Excitonic Structure in the Baseplate of a Chlorosome. The Journal of Physical Chemistry Letters 2014, 5, 1743–1747. (10) Butkus, V.; Gelzinis, A.; Augulis, R.; Gall, A.; Büchel, C.; Robert, B.; Zigmantas, D.; Valkunas, L.; Abramavicius, D. Coherence and Population Dynamics of Chlorophyll Excitations in FCP Complex: Two-Dimensional Spectroscopy Study. Journal of Chemical Physics 2015, 142, 212414. (11) Duan, H.-G.; Stevens, A. L.; Nalbach, P.; Thorwart, M.; Prokhorenko, V. I.; Miller, R. J. D. Two-Dimensional Electronic Spectroscopy of Light-Harvesting Complex II at Ambient Temperature: A Joint Experimental and Theoretical Study. The Journal of Physical Chemistry B 2015, 150827072234000. 16


Page 16 of 19

Page 17 of 19

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


(12) Lewis, N. H. C.; Dong, H.; Oliver, T. A. A.; Fleming, G. R. A Method for the Direct Measurement of Electronic Site Populations in a Molecular Aggregate using TwoDimensional Electronic-Vibrational Spectroscopy. Journal of Chemical Physics 2015, 143, 124203. (13) Oliver, T. A. A.; Lewis, N. H. C.; Fleming, G. R. Correlating the Motion of Electrons and Nuclei with Two-Dimensional Electronic–Vibrational Spectroscopy. Proc. Natl. Acad. Sci. U.S.A. 2014, 111, 10061–10066, also see corrections on Page 16628. (14) Dong, H.; Lewis, N. H. C.; Oliver, T. A. A.; Fleming, G. R. Determining the Static Electronic and Vibrational Energy Correlations via Two-Dimensional Electronic-Vibrational Spectroscopy. Journal of Chemical Physics 2015, 142, 174201. (15) Lewis, N. H. C.; Dong, H.; Oliver, T. A. A.; Fleming, G. R. Measuring Correlated Electronic and Vibrational Spectral Dynamics using Line Shapes in Two-Dimensional Electronic-Vibrational Spectroscopy. Journal of Chemical Physics 2015, 142, 174202. (16) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.; Springer, 2006. (17) Jonas, D. M.; Lang, M. J.; Nagasawa, Y.; Joo, T.; Fleming, G. R. Pump-Probe Polarization Anisotropy Study of Femtosecond Energy Transfer within the Photosynthetic Reaction Center of RhodobactersphaeroidesR26. The Journal of Physical Chemistry 1996, 100, 12660–12673. (18) Lin, Y.-S.; Pieniazek, P. A.; Yang, M.; Skinner, J. L. On the Calculation of Rotational Anisotropy Decay, as Measured by Ultrafast Polarization-Resolved Vibrational PumpProbe Experiments. The Journal of Chemical Physics 2010, 132, 174505. (19) Lutz, M. Resonance Raman Studies in Photosynthesis. Advances in Infrared and Raman Spectroscopy 1984, 11, 211–300.

17



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

(20) Hildebrandt, P.; Spiro, T. G. Surface-Enhanced Resonance Raman Spectroscopy of Copper Chlorophyllin on Silver and Gold Colloids. The Journal of Physical Chemistry 1988, 92, 3355–3360. (21) Fonda, H. N.; Oertling, W. A.; Salehi, A.; Chang, C. K.; Babcock, G. T. Normal-Mode Characteristics of Chlorophyll Models. Vibrational Analysis of Metallooctaethylchlorins and their Selectively Deuterated Analogs. Journal of the American Chemical Society 1990, 112, 9497–9507. (22) G S Singhal, E. R. Measurement of the Fluorescence Lifetime of Chlorophyll a In Vivo. Biophysical Journal 1969, 9, 586–591. (23) Mourou, G.; Malley, M. M. Molecular orientational relaxation times in liquids. Optics Communications 1975, 13, 412–417. (24) Reimers, J. R.; Cai, Z.-L.; Kobayashi, R.; Rätsep, M.; Freiberg, A.; Krausz, E. Assignment of the Q-Bands of the Chlorophylls: Coherence Loss via Qx - Qy Mixing. Scientific Reports 2013, 3, 2761. (25) Bricker, W. P.; Shenai, P. M.; Ghosh, A.; Liu, Z.; Enriquez, M. G. M.; Lambrev, P. H.; Tan, H.-S.; Lo, C. S.; Tretiak, S.; Fernandez-Alberti, S. et al. Non-Radiative Relaxation of Photoexcited Chlorophylls: Theoretical and Experimental Study. Scientific Reports 2015, 5, 13625. (26) Cotton, T. M.; Van Duyne, R. P. Characterization of Bacteriochlorophyll Interactions in vitro by Resonance Raman Spectroscopy. Journal of the American Chemical Society 1981, 103, 6020–6026. (27) Fujiwara, M.; Tasumi, M. Resonance Raman and Infrared Studies on Axial Coordination to Chlorophylls a and b in vitro. The Journal of Physical Chemistry 1986, 90, 250–255.

18


Page 18 of 19

Page 19 of 19

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


(28) Cotton, T. M.; Trifunac, A. D.; Ballschmiter, K.; Katz, J. J. State of Chlorophyll a in vitro and in vivo from Electronic Transition Spectra, and the Nature of Antenna Chlorophyll. Biochimica et Biophysica Acta (BBA) - Bioenergetics 1974, 368, 181–198. (29) Evans, T. A.; Katz, J. J. Evidence for 5- and 6-Coordinated Magnesium in Bacteriochlorophyll a from Visible Absorption Spectroscopy. Biochimica et Biophysica Acta (BBA) - Bioenergetics 1975, 396, 414–426. (30) Woutersen, S.; Mu, Y.; Stock, G.; Hamm, P. Hydrogen-Bond Lifetime Measured by Time-Resolved 2D-IR Spectroscopy: N-Methylacetamide in Methanol. Chemical Physics 2001, 266, 137–147. (31) Zheng, J.; Kwak, K.; Asbury, J.; Chen, X.; Piletic, I. R.; Fayer, M. D. Ultrafast Dynamics of Solute-Solvent Complexation Observed at Thermal Equilibrium in Real Time. Science 2005, 309, 1338–1343.

19


Two-Dimensional Electronic-Vibrational Spectroscopy of Chlorophyll a

Recommend Documents