Gas-Phase Ion Fluorescence Spectroscopy of Tailor-made Rhodamine Homo- and Heterodyads: Quenching of Electronic Communication by π-Conjugated Linkers

Article abstract of DOI:10.1002/anie.202008314

While many key photophysical features are understood for electronic communication between chromophores in neutral compounds, there is limited information on the effect of charges in practically relevant ionic chromo/fluorophores. Here we have chosen posit

Full text of DOI:10.1002/anie.202008314

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Gas-Phase Ion Fluorescence Spectroscopy of Tailor-made
Rhodamine Homo- and Heterodyads: Quenching of Electronic
Communication by π-Conjugated Linkers
Authors: Steen Brondsted Nielsen, Anne Ugleholdt Petersen, Christina
Kjær, Cecilie Jensen, and Mogens Brøndsted Nielsen
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202008314
Link to VoR: https://doi.org/10.1002/anie.202008314

10.1002/anie.202008314
Angewandte Chemie International Edition
RESEARCH ARTICLE
Gas-Phase Ion Fluorescence Spectroscopy of Tailor-made
Rhodamine Homo- and Heterodyads: Quenching of Electronic
Communication by -Conjugated Linkers
Anne Ugleholdt Petersen,[a] Christina Kjær,[b] Cecilie Jensen,[a] Mogens Brøndsted Nielsen,*[a] and
Steen Brøndsted Nielsen*[b]
[a]
[b]
Dr. A. U. Petersen, C. Jensen, Prof. Dr. M. Brøndsted Nielsen
Department of Chemistry
University of Copenhagen
E-mail: mbn@chem.ku.dk
Dr. C. Kjær, Prof. Dr. S. Brøndsted Nielsen
Department of Physics and Astronomy
Aarhus University
E-mail: sbn@phys.au.dk
Supporting information for this article is given via a link at the end of the document.
Abstract: While many key photophysical features are understood for
electronic communication between chromophores in neutral
compounds, there is limited information on the effect of charges in
practically relevant ionic chromo/fluorophores. Here we have chosen
positively charged rhodamines and prepared a selection of homo- and
heterodimers with alkyl or -conjugated, acetylenic bridges.
Protonated molecules were transferred as isolated ions to gas phase
where there is no solvent screening of charges, and fluorescence
spectra were measured with a custom-made ion-trap setup. Our work
situated in a network.[4] The efficiency of energy transfer strongly
depends on the separation between two chromophores and is
inversely proportional to the power of six. At the so-called Förster
distance R0, the efficiency is 50%. Two important factors that
determine R0 are the reorientation factor and the spectral overlap
of the donor-emission spectrum and the acceptor-absorption
spectrum. The former depends on the orientation of the two
transition-dipole moments relative to each other and to the
interfluorophore distance vector.
reveals
strong
polarization
of
the -spacer
(induced
In most textbooks the interaction between two
chromophores (or fluorophores) is typically described with the
assumption that both chromophores are neutral, and that they do
not perturb each other in the ground state (see e.g., [3]). Indeed,
the two examples mentioned above involve neutral chlorophylls
and DNA bases where such assumptions are reasonable.
However, many dyes used as labels in FRET experiments on
biomacromolecules are ionic, e.g., positively charged rhodamine
and cyanine dyes and negatively charged fluorescein ones.[4] One
ionic dye provides an electric field at the other dye when the two
are in close proximity, which causes an induced dipole moment
and vice versa. Hence two, say, cationic dyes actually
communicate in the ground state even though there is no overlap
of their electronic wavefunctions. Photoexcitation of one dye
increases its polarizability and as a result leads to an increase in
the average distance between the positive charge distributions of
the two dyes. The resultant Stark shift of energy levels is
evidenced from redshifted emission spectra as reported in
previous gas-phase work on homodimers of rhodamine 575
cations (R575+)[5] and heterodimers of R575+ and rhodamine 640
cations (R640+)[6]; the two dyes were bridged by either methylene
linkers or peptides.
dipole/quadrupole) when it experiences the electric field from one /
two dyes. Hence, -spacers provide efficient shielding of charges by
reducing the Coulomb interaction, whereas two dye cations polarize
each other when connected by an alkyl. The screening influences the
Förster Resonance Energy Transfer efficiency that relies on the
dipole-dipole interaction.
Introduction
The photophysics of many biological systems is governed by
electronic interactions between two or more chomophores (light
absorbers). One important example is photosynthesis where
multiple chlorophyll molecules within photosynthetic proteins are
responsible for light harvesting and efficient energy transfer to the
reaction center where charge separation occurs.[1] Another is the
nontrivial quantum physics of stacked bases in DNA and Watson-
Crick base pairs, which determines the excited-state dynamics
and fate after UV excitation.[2]
When two chromophores are in close proximity, exciton
coupling of the two individual excited-state wavefunctions occurs,
which implies that the new stationary states are not associated
with only one of the chromophores but with both; in other words,
the excitation energy is spatially delocalized over both
chromophores.[3] At large separations the coupling is weak, and
we can to a good approximation consider the excitation being
associated with only one chromophore. Förster Resonance
Energy Transfer (FRET) is then responsible for a downhill
movement of the excitation energy between chromophores
It is clear that the internal Coulomb repulsion is largest when
the dyad systems are isolated in vacuo where there is no solvent
screening of the two charges. Gas-phase experiments on simple
model systems therefore provide the most direct information on
such charge-charge interactions. Unfortunately, however, such
experiments suffer from a thin cloud (low density) of ionic
fluorophores, and specialized apparatus is needed to obtain
reasonable signal-to-noise ratios.[7]
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202008314
Angewandte Chemie International Edition
RESEARCH ARTICLE
In Aarhus we use the home-built LUNA (LUminescence
iNstrument in Aarhus) setup where ions produced by electrospray
ionization are stored in the center of a cylindrical Paul trap.[8] Mass
selection is followed by photoexcitation and collection of emitted
photons. One of the end electrodes is a mesh grid to allow as
many photons as possible to exit the trap and be detected. The
photon collection efficiency of our setup is about 5%. While one
obvious requirement for gas-phase ion fluorescence
spectroscopy is that the ions are fluorescent, the technique
benefits from no limitation on the size of the molecular ion. This is
in contrast to photodissociation-action spectroscopy where
dissociation on the instrumental time scale is necessary, i.e.,
microseconds to tens of milliseconds dependent on the setup
used, and dissociation yields can significantly depend on
excitation wavelengths.[9]
Monomers - carboxylic acids
H
N
H
O
N
N
O
N
O
O
OH
OH
R575+
RB+
RB+
NH2
Monomers
R575+
M1+
- amide linkages
m5+
RB+
R575+
NH2
RB+
m1+
TMS
TMS
m2+
M2+
In this work we have synthesized a multitude of homo- and
heterodimers of rhodamine 575 (R575) and rhodamine B (RB)
connected by alkyl or -conjugated, linear bridges that represent
new unexplored dyad systems. Our synthesis work is based on
development of suitable acetylenic building blocks that were
subjected to metal-catalyzed coupling reactions under various
conditions. The compounds isolated are on the closed lactam
forms (R575 / RB) but upon protonation, they undergo ring-
opening to form the cationic and fluorescent amide forms (R575+
/ RB+) as shown in Scheme 1. In heterodimers, R575+ is the donor
and RB+ the acceptor. The -spacer is linked to the lactam/amide
nitrogen atom of the dye via a methylene group and is therefore
not in conjugation with the chromophore unit.
A summary of ions that were subjected to gas-phase
fluorescence experiments is given in Figure 1; these include also
relevant monomer derivatives for comparison. Our results reveal
differences in the communication between two ionic dyes with the
nature of the linker, and that a -linker may to some extent act as
an electrostatic shield between the two dyes.
RB+
d12+
Homodimers
- amide linkages
RB+
RB+
RB+
d22+
RB+
RB+
RB+
RB+
d32+
d42+
R575+
RB+
R575+
R575+
RB+
d52+
D62+
R575+
D22+
Heterodimers
RB+
R575+
- amide linkages
F22+
RB+
R575+
RB+
R575+
F12+
F42+
Figure 1. Summary of rhodamine monomers (m / M), homodimers (d / D), and
heterodimers (FRET systems, F) subject to studies in this work. For the
definition of amide linkages, see Scheme 1.
It should be noted that photoexcited electron transfer from
one dye to another, often mediated by a -conjugated bridge,[10]
is not energetically feasible for our gas-phase dyads as the
detachment energy of a cationic donor is too high compared with
the recombination energy of the cationic acceptor (energy defect
Results and Discussion
First we present results from the synthesis part, followed by those
from gas-phase ion fluorescence spectroscopy. Finally, we
provide an overall discussion based on computed geometric
structures of simple model dimers.
of about
6
eV[6]). The only possible processes after
photoexcitation are therefore light emission from either R575+ or
RB+, potentially after FRET in the latter case.
Synthesis
Rhodamine B dimers d2-5 were made starting from the monomer
1 according to Scheme 2. The known starting material 1 was
made in analogy with a general literature procedure[5], by
converting R575 perchlorate to its acid chloride (using thionyl
chloride) that was then treated with propargylamine (see
Supporting Information; SI; yields of 65-73%). It has previously
been made from the acid chloride generated using POCl3,[11] or
H
N
H
N
H
N
H
N
O
O
H
O
N
R
R
N
H
O
R575 (lactam)
R575+ (amide)
by
using
O-benzotriazole-N,N,N',N'-tetramethyl-uronium
hexafluorophosphate (HBTU) and triethylamine to attach
propagylamine.[12] The other starting material 2 was prepared by
treating the acid chloride generated from RB chloride with p-
iodoaniline (yield of 52%); compound 2 has previously been made
with the use of toluenesulfonylchloride and DMAP.[13] An oxidative
coupling of m4 and trimethylsilylacetylene under Hay conditions
(CuCl, TMEDA, and air) gave m2 in a yield of 26%, while by using
Pd(PPh3)2Cl2 and CuI an improved yield of 47% was obtained.
Monomers m2 and m4 were now explored as precursors for
the homodimers d2 and d3. An oxidative dimerization of m4,
N
O
N
R
N
O
N
H
O
N
R
N
H
O
RB (lactam)
RB+ (amide)
Scheme 1. Equilibria between closed and open, protonated rhodamine forms.
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202008314
Angewandte Chemie International Edition
RESEARCH ARTICLE
using conditions of Haley and co-workers[14] gave the dimer d2 in
Monomer 3 was readily achieved from R575 perchlorate
according to Scheme 3 in quantitative yield, without need for
purification. For synthesis of 3, a much better yield was obtained
via the acid chloride route, while using the formerly reported
method by refluxing R575 perchlorate and propargylamine in
ethanol gave a yield of 40-68%;[16] the advantage of proceeding
via the acid chloride is that this method also works for compounds
not soluble in ethanol. The dimer D2 was prepared in quantitative
yield by an oxidative coupling of 3 using Pd2dba3, dppe, CuI, and
I2. Monomer M2 could under these conditions be obtained in a
yield of 37% (from 3 and trimethylsilylacetylene), but changing to
the Pd(PPh3)2Cl2 and CuI catalyst system gave a higher yield of
95%.
The heterodimer F4 was achieved in a Sonogashira
coupling between 3 and 4, with the conditions described earlier,
affording the product in 35% yield. It was possible to form F2 by
coupling terminal alkynes 1 and 3, but homodimers d2 and D2
were also formed, and these could in our hands not be separated
from F2. To circumvent this problem, monomer 5 was first
prepared by coupling N-Boc-propargylamine with 3. After Boc-
deprotection, the amine 5 was reacted with the acid chloride of
R575, furnishing the product F2 in a yield of 34%. The dimers d1
and F1 can easily be made from m1 by treating it with the acid
chloride of the rhodamine monomers and Et3N in acetonitrile.
good yield. We note that using these conditions for
a
heterocoupling between m4 and trimethylsilylacetylene only gave
m2 in a yield of 4%. Desilylation of m2 using K2CO3 in MeOH
followed by an oxidative homocoupling gave d3 in a yield of 11%.
The dimers d4 and d5 (as well as monomer m3[11c] and m6) were
synthesized by Sonogashira coupling reactions. It was found that
this reaction works well when performed in a 1:1 mixture of THF
and Et3N, with the catalyst system of Pd(PPh3)2Cl2 and CuI.
Another specification was that the reaction was done under
sonication at 30 °C, which allowed for fairly short reaction times
(and previously shown advantageously for Sonogashira
couplings[15]). Using these conditions and the coupling partners
1,4-diiodobenzene and m5, the dimers d4 and d5 were obtained
in yields of 63% and 13%, respectively.
TMS
, a)
TMS
RB
RB
m2 (47%)
b), a)
n = 2
c)
N
O
N
n
n = 1
RB
d2 (n = 1; 59%)
d3 (n = 2; 11%)
N
I
I
O
, d)
Gas-Phase Fluorescence Studies
1
To measure dispersed fluorescence spectra of isolated and
mass-selected gaseous ions we use the LUNA setup that in a
unique way combines an electrospray ion source with a cylindrical
Paul trap (fluorescence cell for mass-selected ions) and a light
detection part (detailed in [8]). Briefly, ions made by electrospray
ionization of samples dissolved in methanol and acetic acid (10:1)
were accumulated in an octopole pretrap that was emptied at a
repetition frequency of 20 Hz. Ion bunches were transferred to the
Paul trap that was filled with helium buffer gas. Following mass
selection, ions were irradiated by a nanosecond laser pulse
delivered by a 20-Hz pulsed EKSPLA laser (Nd:YAG laser in
combination with an optical parametric oscillator (OPO)). The
emitted photons were collected and detected with a spectrometer
equipped with a CCD camera (Andor Technology). This sequence
was repeated 100 times with no emptying of ion trap in between
each injection of ion bunches. A background spectrum with no
ions in the trap was then recorded to correct for background
photons. A notch filter was placed before the entrance to the
spectrometer to reduce the number of scattered photons from the
excitation pulse. The spectrometer slit width was 1250 m.
Band maxima for all ions are summarized in Table 1, and all
spectra are included in the SI. In the following we discuss the
spectra in detail.
RB
RB
d4 (63%)
I
RB
, d)
2
RB
RB
d5 (13%)
Scheme 2. Synthesis of RB homodimers (lactam forms): a) Pd(PPh3)2Cl2, CuI,
Et3N/PhMe, air, rt. b) K2CO3, MeOH/THF; c) Pd2dba3, dppe, CuI, I2, DIPA/THF,
air, rt. d) Pd(PPh3)2Cl2, CuI, Et3N/THF, 30 °C, ))).
b)
H
N
H
N
O
R575
R575
D2 (quantitative)
N
I
RB
O
4
R575
RB
3 (quantitative)
c)
F4 (35%)
H2N
a)
H
N
H
N
O
H2N
RB
5
OH
R575
RB
d)
O
ClO4
F2 (34%)
R575
e)
d)
RB
F1 (82%)
RB
d1 (47%)
NH2
R575
RB
RB
m1
Scheme 3. Synthesis of R575 homodimers (lactam forms) and R575/RB
heterodimers (lactam forms). a) i. SOCl2, CH2Cl2, Δ, 2h, ii. Et3N, MeCN, 0 °C. b)
Pd2dba3, dppe, CuI, I2, DIPA/THF, air, rt. c) Pd(PPh3)2Cl2, CuI, Et3N/THF,
30 °C, ))). d) i. R575 perchlorate, SOCl2, Δ; ii. Et3N, MeCN, 0 °C. e) i. RB chloride,
SOCl2, Δ, ii. Et3N, MeCN, 0 °C.
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202008314
Angewandte Chemie International Edition
RESEARCH ARTICLE
Table 1. Summary of emission band maxima from gas-phase studies. All values
are in nm (eV).
R575+
M1+
M2+
D22+
D62+ [a]
508±2
508±4
515±3
519±4
528±5
(2.44±0.01)
(2.44±0.02)
(2.41±0.01)
(2.39±0.02)
(2.35±0.02)
RB+
m1+
m2+
m5+
543±3
540±3
551±3
549±3
(2.28±0.01)
(2.30±0.01)
(2.25±0.01)
(2.26±0.01)
d12+
d22+
d32+
d42+
d52+
557±4
549±4
548±3
548±3
552±5
(2.23±0.02)
(2.26±0.02)
(2.26±0.01)
(2.26±0.01)
(2.25±0.02)
F12+
F22+
F42+
556±4
551±5
516±3
(2.23±0.02)
(2.25±0.02)
(2.40±0.01)
[a] Taken from ref. [5].
Figure 2. Gas-phase fluorescence spectra of (a) R575 monomers and (b) R575
dimers. In (c) spectra of a R575 monomer and the corresponding dimer are
presented for comparison. The dashed lines indicate band maxima of R575+
(furthest to the blue) and of M2+ (furthest to the red). All spectra were recorded
with an excitation wavelength of 487 nm.
R575 monomers and dimers: Fluorescence spectra of
M1+ and M2+ are shown together with that of R575+ in Figure 2a.
As reported earlier, replacement of the carboxylic acid group in
R575+ with an amide functionality that terminates in a primary
amine (M1+) has no effect.[5] However, a clear redshift occurs
when the amide substituent contains a diyne unit (M2+). The diyne
is not in conjugation with the amide, and we therefore ascribe the
redshift to polarization of the -electronic cloud in the electric field
from the xanthene cation to give a favorable charge induced-
dipole interaction (Figure 3). This interaction increases in strength
after photoexcitation of the xanthene on account of the in general
higher polarizability of an excited state than of a ground state [17]
;
the positive charge distribution in the xanthene moves closer
towards the diyne that as a result gets even more polarized.
Now, a homodimer with the diyne bridge (D22+) displays a
spectrum that is similar to that of M2+ except that it is broadened
towards the red (Figure 2b,c). The broadening is likely the result
of multiple structures present in the ion cloud, and that each
structure is somewhat flexible. It is noteworthy that the band
maximum for a homodimer where the bridge contains six
methylene units (D62+, spectrum included in Figure 2b), i.e., same
number of carbon atoms in bridge as in D22+, is significantly
redshifted. The red end of the two D22+ and D62+ spectra is,
however, similar. The redshifted emission from D62+ relative to
R575+ was earlier ascribed to a Stark effect as the internal
Coulomb repulsion between the two cationic dyes is reduced after
photoexcitation[5]; again ascribed to the higher polarizability in an
excited state. Interestingly, the -spacer seems to counteract this
effect. Indeed, in a simple picture each dye will polarize the -
spacer such that there is an excess of negative charge close to
each dye and a positive charge in the center of the spacer (Figure
3). The result is a -spacer with a quadrupole moment, which
provides an electric field at each dye that opposes that from the
positive charge. The -spacer shields one dye from the other,
thereby lowering the effect of internal Coulomb repulsion.
Figure 3. Top: Symmetric dye (in red) with one positive charge and a linker (in
grey) that is composed of methylene units. The charge is evenly distributed over
the dye but is on the figure shown as a point charge in the center. The charge
distribution becomes asymmetric when two cationic dyes are tethered together
due to the internal Coulomb repulsion and the high polarizability of the dyes.
Bottom: A -linker (in black) is polarized in the field from one cationic dye; the
electric field due to the induced dipole moment of the-linker acts back and
changes the charge distribution in the dye, i.e., the positive charge moves on
average towards the -linker. When a second dye is attached, a quadrupole
moment is induced in the-spacer, which effectively shields the two positive
charges.
RB monomers and dimers: Figure 4 shows spectra of
monomers and dimers. A -electron cloud in the form of a diyne
(m2+) or phenylethynyl (m5+) causes a significantly redshifted
spectrum compared to that of the parent monomer species (RB+).
As before, we ascribe this to a polarization effect. The spectrum
of m1+ is slightly to the blue of RB+, which may be due to some
interaction between the xanthene and the primary amine.
Spectra of homodimers with -spacers are surprisingly
similar to that of monomers that contain a -appendage. There is
little dependence on the -spacer; the spectra of d22+ (diyne),
d32+ (tetrayne), and d42+ (diethynylbenzene unit) are almost
identical while that of d52+ (phenylethynyl) is slightly broadened to
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202008314
Angewandte Chemie International Edition
RESEARCH ARTICLE
the red. In contrast, the spectrum of d12+ where two methylene
units bridge the two fluorophores is significantly redshifted as
earlier reported also for homodimers of R575+.[5] Taken together,
these results indicate that a -spacer shields the two positive
charges in homodimers, and therefore that the spacer length is of
less importance than is the case for methylene units (Figure 3).
distances and less efficient energy transfer; we will return to
possible geometries later on.
Figure 5. Gas-phase fluorescence spectra of (a) F12+ recorded using two
different excitation wavelengths (487 nm and 513 nm). The 487-nm spectrum is
also presented in (b) together with those of R575+ and RB+ dye monomers (487-
nm and 513-nm excitation, respectively) and a d12+ RB dimer (513-nm
excitation).
Figure 4. Gas-phase fluorescence spectra of (a) RB monomers and (b) RB
dimers. Spectra of a RB monomer and the corresponding dimer are shown in
(c) for comparison. The dashed lines indicate absorption band maxima of RB+
(furthest to the blue) and of m2+ (furthest to the red). All spectra were recorded
with an excitation wavelength of 513 nm.
Heterodimers: Our work includes three heterodimers with
the R575+/RB+ FRET pair. In the first (F12+) the two dyes are
bridged by two methylene groups. Two spectra obtained after
photoexcitation at either 487 nm or 513 nm are shown in Figure
5. Except for a small shoulder to the blue (515 – 525 nm) for 487-
nm excitation, the spectra are quite similar. It should be noted,
however, that the notch filter for 513-nm excitation prevents the
transmission of photons to the spectrometer in a small region
around 513 nm (514  9 nm). In any case, fluorescence occurs
dominantly from the RB+ dye, either due to direct excitation of the
dye in the heterodimer or after FRET. The emission spectrum is
significantly redshifted compared to that of RB+ monomers and is
almost identical to that of d12+ (RB+ homodimer with two
methylene units). As discussed above and in previous work on a
heterodimer of R575+ and R640+,[6] this is simply due to the
electric field that one cationic dye senses from the other. The fact
that the spectra of F12+ and d12+ are nearly identical excludes the
possibility of exciton coupling as the energy of an exciton state
formed from a linear combination of two RB+ excited states is
clearly different to that from a linear combination of R575+ and
RB+ excited states. Also, even though the alkyl spacer is small,
the separation between the two xanthenes is too large (11 Å or
more, vide infra) for exciton coupling really to be significant. The
small shoulder towards the blue observed in the 487-nm
excitation spectrum of F12+ is indicative of some minor emission
from R575+. This may originate from structures where the relative
dipole orientation factor is small, which results in short Förster
Figure 6. Gas-phase fluorescence spectra of (a) F22+ recorded using four
different excitation wavelengths (445 nm, 487 nm, 513 nm, and 532 nm). The
hole in the 532-nm spectrum is due to blocking by the notch filter (533  9 nm).
The 487-nm spectrum is also presented in (b) together with those of M2+ and
m2+ dye monomers (487-nm and 513-nm excitation, respectively) and the d22+
RB dimer (513 nm). A sum of the m2+ spectrum and 0.1 times the M2+ spectrum
is shown in (c) together with the F22+ spectrum.
In heterodimer F22+, the two dyes are bridged by a spacer
that contains a diyne. Spectra recorded at excitation wavelengths
of 445 nm, 487 nm, 513 nm, and 532 nm are shown in Figure 6a.
At higher wavelengths, the RB+ absorption increases relative to
that of R575+. The notch filter prevents emitted photons in a
region around the excitation wavelength to be transmitted to the
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202008314
Angewandte Chemie International Edition
RESEARCH ARTICLE
spectrometer, which implies that only the part of the spectrum 9
nm above the excitation wavelength is valid for 513 nm and 532
nm. However, as no dependence on excitation wavelength is
seen, we conclude that any structural change before FRET is
limited. Indeed, if excitation of R575+ had led to a geometry
change of the dimer, we would expect a Stokes shift for the
emission from RB+ after FRET.
From Figure 6b, it is clearly evident that the emission
spectrum of RB+ in F22+ is the same as those of m2+ (monomer
RB+ with diyne linkage) and d22+ (homodimer of RB+ with diyne
-spacer). The spectrum of R575+ monomers with diyne linkage
(M2+) is also included in the figure for comparison. Indeed, a linear
combination of the M2+ (10%) and m2+ (100%) normalized
monomer spectra can reasonably well reproduce the spectrum of
F22+. To obtain a FRET efficiency we need to consider the
absorption cross sections of the two dyes as well as the internal
conversion rates.
Gas-phase absorption spectra of RB+ and R575+ parent
species were earlier reported by Jockusch and co-workers.[18] The
absorption by RB+ at 487 nm is about 30% that at maximum
absorption at 531 nm. In the case of R575+, the maximum
absorption is at 495 nm, and at 487 nm the absorption is about
95%. Assuming equal absorption cross sections, the absorption
by R575+ is about three times larger than that by RB+ at 487 nm.
This assumption is somewhat justified from similar fluorescence
signals from the two monomer dyes when they are photoexcited
close to their maximum absorption. In the heterodimer both dyes
have about 10 nm redshifted emission compared to R575+ and
RB+ parent monomers. If we assume the absorption spectra are
also redshifted by 10 nm, the absorption by R575+ is about five
times larger than by RB+ in the heterodimer. The internal
conversion rates of both dyes are likely low as the dyes are both
strongly fluorescent. If we assume them to be equal for R575+ and
RB+ in the heterodimer, we estimate based on the absorption
cross sections and the F22+ spectrum, a FRET efficiency of about
89%.
Finally, spectra of the third heterodimer, F42+
(diethynylbenzene spacer unit), recorded with three different
excitation wavelengths are shown in Figure 7a. Again, the spectra
look similar at the red end, and we conclude that the fluorescence
from RB+ is independent of whether the excited state is formed
after direct photon absorption or after FRET, indicative of no or
limited geometry change when R575+ is initially photoexcited. The
most striking result for F42+ is that emission occurs dominantly
from R575+. Panel b in the figure reveals that the emission band
maximum is the same as that of M2+ (R575+ with diyne), again
demonstrating that the charge from another dye has little effect in
the presence of a -spacer. The panel also includes the m5+ (RB+
with phenylethynyl) spectrum. A linear combination of the two
model monomer spectra (0.25  m5+ + M2+) are seen in panel c to
fully reproduce the F42+ spectrum. After correction for the
absorption of 487-nm photons directly by RB+ in the heterodimer,
we estimate the FRET efficiency to be only about 4%.
Figure 7. Gas-phase fluorescence spectra of (a) F42+ recorded using three
different excitation wavelengths (487 nm, 513 nm, and 532 nm). Two of the
excitation wavelengths (513 nm, 532 nm) are within the fluorescence band. The
holes in the spectra are due to blocking by the notch filter. The 487-nm spectrum
is also presented in (b) together with those of M2+ and m5+ dye monomers (487-
nm and 513-nm excitation, respectively). A sum of the M2+ spectrum and 0.25
times the m5+ spectrum is shown in (c) together with the F42+ spectrum.
Computational Studies
To explore structural heterogeneity, we have done density
functional theory (DFT) calculations on
a simple model
homodimer (Figure 8, top) using the Gaussian program
package[19]. We chose the rhodamine core structure and two
methylene units as bridge to limit the computational time. Seven
different geometries were found by optimization at the B3LYP/6-
31G level and are shown in Figure 8; there are of course more
possible conformers but we believe that those obtained serve as
good representatives. Also vibrational frequencies were
calculated to verify that all structures are local minima and not
transition states. Based on the frequencies the internal energy is
estimated to 0.7 eV at 298 K. The structures were subjected to
single-point calculations at the B3LYP/6-311++G(2d,p)
(SCF=Tight) level. The highest-energy isomer (VII) is 0.55 eV
higher in energy than the lowest-energy one (I).
A summary of relative energies, distances between the
fluorophore centers (R), and relative dipole orientation factors (2)
are given in Table 2 (coordinates and details on how to calculate
 are given in the SI). The interfluorophore distance varies
between 11.0 Å (conformer II) and 12.7 Å (III). The lowest value
of 2 is 0.04 (V) and the highest 0.95 (I). Indeed, the two lowest-
energy structures have significantly different2 values (0.95 and
0.16) and R-values (12.5 Å and 11.0 Å). These two structures will
give different fluorescence spectra as the charge-induced
polarization of the two fluorophores is different, which accounts
for broader spectra seen for dimers than for monomers. Only
Stark-shifted emission is seen for dimers as the monomers only
carry one positive charge. In addition, the higher-energy
structures will contribute to this broadening, in particular if
equilibrium is not established under the experimental conditions.
While the ions undergo thermalizing collisions with the room-
temperature helium-buffer gas in the trap, some RF-heating is
The RET seen for F42+ is surprisingly small considering the
short interfluorophore separation (23 Å, vide infra), which
indicates that Through Bond Energy Transfer (TBET) is not active.
TBET normally accounts for faster RET than predicted by FRET
and occurs via a -linker (see e.g. a discussion of TBET in [10a]).
However, a CH2 group at each end of the linker in F42+ breaks the
direct conjugation with the dye; i.e., hyperconjugation is
necessary to mediate -orbital overlap.
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202008314
Angewandte Chemie International Edition
RESEARCH ARTICLE
unavoidable (due to energetic collisions between high-velocity
ions and helium). The large variation in 2 also suggests that
structures are present for which FRET is less likely to occur even
though R is short. These could explain the small emission seen
from R575+ after photoexcitation of F12+ (Figure 5).
Calculated structures of homodimers modeling d22+/ D22+/
F22+ and d42+ / F42+ are shown in Figure 9, again using the
simplified rhodamine core structure for each dye; results are
summarized in Table 3. Here the linear, rigid -linker limits the
number of possible structures. For F22+ models, we found three
structures extremely close in energy (basically isoenergetic), and
where the two dyes are oriented with nearly parallel or
perpendicular transition dipole moments providing high and low
2 values, respectively (Table 3 and Figure 9). The former
orientation favors FRET in contrast to the latter. In the case of
F42+ models, we locate two low-lying structures that both have
nearly parallel transition dipole moments (Table 3 and Figure 9).
A low-lying transition state connects these “anti” (I’’) and “syn”
(II’’) structures from a rotation around the -spacer-(CH2) single
bond. The transition state is 0.014 eV higher in energy than the
“anti” form, and at room temperature there is therefore basically
free rotation around the single bond. Importantly, the structures
found close in energy for F22+ and F42+ models and the low-lying
transition state for the F42+ model indicate that the two dyes to a
good approximation can rotate freely relative to each other, along
an almost linear rod, within the excited-state lifetime, and
therefore that a value of 2/3 is a reasonable estimate for 2 for
both heterodimers. Likewise, Winters et al.[20] found based on DFT
calculations that the ground state of a butadiyne-linked porphyrin
dimer has a broad distribution of dihedral angles at room
temperature due to a very low barrier for rotation.
Figure 8. Optimized geometries of the simplified dimer structure shown on top.
The transition dipole moments and the interfluorophore distance vector are
indicated for I (top).
Table 2. Relative conformer energies, interfluorophore distances (R), and
relative dipole orientation factors (2) for homodimers shown in Figure 8.
Conformer
E (in eV)
0
R (in Å)
12.5
11.0
12.7
12.3
11.4
11.9
11.2
2
I
0.95
0.16
0.50
0.84
0.04
0.17
0.49
II
0.049
0.33
0.33
0.35
0.40
0.55
III
IV
V
VI
VII
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202008314
Angewandte Chemie International Edition
RESEARCH ARTICLE
with the phenylethynyl unit. This picture supports our assumption
from earlier that the energy transfer is not mediated by the -
spacer in heterodimers.
Table 3. Relative conformer energies, interfluorophore distances (R), and
relative dipole orientation factors (2) for acetylenic scaffolds shown in Figure 9.
Conformer
E (in eV)
0
R (in Å)
18.4
2
I’
II’
0.008
0.85
0.91
0.95
0.79
0.061
0.0066
0.012
0
18.1
III’
I’’
18.2
22.6
II’’
TS’’
0.0014
0.014
22.7
22.7
Figure 10. Model rhodamine with ethynylphenyl appendage (structure shown in
box). MP2/6-311++G(2d,p) // B3LYP/6-31G calculated orbitals of HOMO–1,
HOMO, and LUMO.
Figure 9. Optimized geometries of simplified dimers with acetylenic bridges;
same rhodamine units as those shown in Figure 8, top. Left: Butadiyn bridge.
Right: Diethynylbenzene bridge.
It is worth to mention that while a polarity model does provide
a simple explanation for our results, the molecules with alkyl
spacers have rather complex structures (Figure 8) rendering a
direct comparison of linkers somewhat difficult. Indeed, as our
calculations show, an alkyl spacer allows for more conformational
freedom. In contrast, the molecules with -spacers are nearly
linear (Figure 9), which is optimum for polarization of the -
electrons in the triple bonds. For comparison, buta-1,3-diyne has
a slightly larger boiling point than butane, which is partly ascribed
to the polarization of the -electrons in the former molecule and
as a result greater dispersion forces. It would be interesting to
better establish the connection between geometric structure and
Stark-shifted emission. This would require conformer selection,
for example done by ion-mobility spectrometry. Another approach
would be cold-ion spectroscopy. At low temperatures, it is
possible to address each conformer selectively due to specific
absorption lines. Our instrumental work is going in this direction
with the current construction of LUNA2, a setup operating at 77 K,
and hopefully also a cryogenically cooled setup in a foreseeable
future.
Based on the assumption of 2 being 2/3 for both F22+ and
F42+, we can provide an estimate of the Förster distance simply
from calculated interfluorophore distances. In F22+ the energy-
transfer efficiency is more than 50%, which implies that R(F22+) <
R0 while in F42+ the efficiency is less than 50% and R(F42+) > R0.
As R(F22+) = 18 Å and R(F42+) = 23 Å, an estimate for R0 midway
between the two R values is 20.5  2.5 Å. This value is
significantly less than that of 67 Å reported by Jockusch and co-
workers[21] for the R575+/R640+ FRET pair tethered to a peptide,
even though the spectral overlap is larger for R575+/RB+ than for
R575+/R640+. We note that R0 is proportional to the spectral
6
overlap and to 2, and that 2 was assumed to be 2/3 for the
R575+/R640+ FRET pair. This significant difference in R0 reflects
the efficient screening provided by a -spacer as such screening
lowers the dipole-dipole interaction between the two dyes and as
a result the Förster distance.
Finally, we calculated the molecular orbitals of the rhodamine
model with
a
phenylethynyl appendage at the MP2/6-
311++G(2d,p) // B3LYP/6-31G level of theory (Figure 10). It is
evident that the HOMO is located on the phenylethynyl unit with
no overlap with the xanthene dye. Likewise, the HOMO-1 and the
LUMO are on the xanthene dye, and there is no -orbital overlap
Conclusion
8
This article is protected by copyright. All rights reserved.

10.1002/anie.202008314
Angewandte Chemie International Edition
RESEARCH ARTICLE
A. Wyer, S. Brøndsted Nielsen), Springer, Berlin/Heidelberg, 2003,
Chapter 3.
In this work, we have reported the synthesis of new dyads where
two charged dyes (Rhodamine 575 or Rhodamine B) are bridged
by almost linear linkers (rods). The preparations were based on
acetylenic coupling reactions, and oxidative alkyne-alkyne
dimerizations were finely tuned to reach optimum conditions for
the various rhodamine scaffolds.
Gas-phase fluorescence studies on the cationic dyes reveal
strong polarization of the -linker when it experiences the electric
field from one or two dyes. As a result, -spacers provide efficient
shielding of charges, i.e., they reduce the Coulomb interaction
between two dye cations. This was evidenced from almost
identical fluorescence spectra of homodimers and corresponding
monomer derivatives. In contrast, two dye cations repel each
other strongly when connected by an alkyl bridge, which results
in red-shifted emission spectra due to the Stark effect.
In the case of heterodimers, a -spacer acts as a shield
against the dipole-dipole interaction that governs energy transfer
from donor to acceptor. Hence the energy-transfer efficiency is
much reduced compared to that for corresponding systems with
alkyl spacers. Indeed, from DFT-calculated structures, we provide
an estimated Förster distance of only 20.5  2.5 Å for the
Rhodamine 575 / Rhodamine B FRET pair tethered together with
a -spacer despite a high spectral overlap of donor emission and
acceptor absorption. This screening effect may have direct
implications on the energy transfer when fluorophores are
integrated in biological systems where aromatic units potentially
could act as -shields.
[10] a) B. Albinsson, M. P. Eng, K. Pettersson, M. U. Winters, Phys. Chem.
Chem. Phys. 2007, 9, 5847-5864; b) P. P. Lainé, S. Campagna, F.
Loiseau, Coord. Chem. Rev. 2008, 252, 2552-2571; c) M. Natali, S.
Campagna, F. Scandola, Chem. Soc. Rev. 2014, 43, 4005-4018; d) Y.
Hou, X. Zhang, K. Chen, D. Liu, Z. Wang, Q. Liu, J. Zhao, A. Barbon, J.
Mater. Chem. C 2019, 7, 12048-12074.
[11] a) M. J. Jou, X. Chen, K. M. K. Swamy, H. N. Kim, H.-J. Kim, S. Lee, J.
Yoon, Chem. Commun. 2009, 46, 7218-7220; b) O. A. Egorova, H. Seo,
A. Chatterjee, K. H. Ahn. Org. Lett. 2010, 12, 401-403; c) X. Cao, W. Lin,
Y. Ding, Chem. Eur. J. 2011, 17, 9066-9069.
[12] a) N. B. Yapici, S. Jockusch, A. Moscatelli, S. R. Mandalapu, Y. Itagaki,
D. K. Bates, S. Wiseman, K. M. Gibson, N. J. Turro, L. Bi, Org. Lett. 2012,
14, 50-53; b) N. B. Yapici, S. R. Mandalapu, T-L. Chew, S. Khuon, L. Bi,
Bioorg. Med. Chem. Lett. 2012, 22, 2440-2443; c) N. B. Yapici, Y. Bi, P.
Li, X. Chen, X. Yan, S. R. Mandalapu, M. Faucett, S. Jockusch, J. Ju, K.
M. Gibson, W. J. Pavan, L. Bi, Sci. Rep. 2015, 5, 8576.
[13] M. K. Lee, P. Rai, J. Williams, R. J. Twieg, W. E. Moerner, J. Am. Chem.
Soc. 2014, 136, 14003-14006.
[14] E. L. Spitler, J. M. Monson, M. M. Haley, J. Org. Chem. 2008, 73, 2211-
2223.
[15] a) K. Qvortrup, A. S. Andersson, J.-P. Mayer, A. S. Jepsen, M. Brøndsted
Nielsen, Synlett 2004, 2818-2820; b) A. S. Andersson, K. Qvortrup, E. R.
Torbensen, J.-P. Mayer, J.-P. Gisselbrecht, C. Boudon, M. Gross, A.
Kadziola, K. Kilså, M. Brøndsted Nielsen, Eur. J. Org. Chem. 2005, 3660-
3671; c) A. R. Gholap, K. Venkatasan, R. Pasricha, T. Daniel, R. J. Lahoti,
K. V. Srinivasan, J. Org. Chem. 2005, 70, 4869-4872; d) S. S. Palimkar,
P. H. Kumar, R. L. Lahoti, K. V. Srinivasan, Tetrahedron 2006, 62, 5109-
5115; e) N. Fu, Y. Zhang, D. Yang, B. Chen, X. Wu, Catalysis Comm.
2008, 9, 976-979.
[16] a) W. Lin, X. Cao, Y. Ding, L. Yuan, L. Long, Chem. Commun. 2010, 46,
3529-3531; b) R. Kagit, M. Yildirim, O. Ozay, S. Yesilot, H. Ozay, Inorg.
Chem. 2014, 53, 2144-2151; c) J.-T. Hou, K. Li, K.-K. Yu, M.-Z. Ao, X.
Wang, X.-Q Yu, Analyst 2013, 138, 6632-6638.
Acknowledgements
[17] P. Suppan, N. Ghoneim, Solvatochromism, The Royal Society of
Chemistry, Cambridge, UK, 1997, Chapter 3.
SBN and MBN acknowledge support from the Independent
Research Fund Denmark | Natural Sciences (9040-00041B).
[18] a) M. W. Forbes, R. A. Jockusch, J. Am. Soc. Mass Spectrom. 2011, 22,
93-109; b) S. K. Sagoo, R. A. Jockusch, J. Photochem. Photobiol. A:
Chem. 2011, 220, 173-178.
[19] Gaussian 03, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T.
Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi,
V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.
Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M.
Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C.
Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J.
Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma,
G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wallingford CT,
2004.
Keywords: Alkynes • Dyads • Cross-coupling • Fluorescence •
gas-phase FRET
[1]
[2]
Y. C. Cheng,G. R.Fleming, Annu. Rev.Phys.Chem. 2009, 60, 241-262.
C. T. Middleton, K. de la Harpe, C. Su, Y. K. Law, C. E. Crespo-
Hernandez, B. Kohler, Annu. Rev.Phys.Chem. 2009, 60, 217-239.
C. R. Cantor, P. R. Schimmel, Biophysical Chemistry, Part II, Techniques
for the Study of Biological Structure and Function, W. H. Freeman and
Company, New York, 1980, p. 392f.
[3]
[4]
[5]
J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 2nd edn,
Kluwer Academic/Plenum Publishers, New York, 1999.
C. Kjær, H. Lissau, N. K. G. Salinas, A. Ø. Madsen, M. H. Stockett, F. E.
Storm, T. H. Hansen, J. U. Andersen, B. W. Laursen, K. V. Mikkelsen, M.
Brøndsted Nielsen, S. Brøndsted Nielsen, ChemPhysChem 2019, 20,
533-537.
[6]
[7]
C. Kjær, Y. Zhao, M. H. Stockett, L. Chen, K. Hansen, S. Brøndsted
Nielsen, Phys. Chem. Chem. Phys. 2020, 22, 11095-111000.
a) J. T. Khoury, S. E. Rodriguez-Cruz, J. H. Parks, J. Am. Soc. Mass
Spectrom. 2002, 13, 696-708; b) V. Frankevich, X. W. Guan, M. Dashtiev,
R. Zenobi, Eur. J. Mass Spectrom. 2005, 11, 475-482; c) Q.-Z. Bian, M.
W. Forbes, F. O. Talbot, R. A. Jockusch, Phys. Chem. Chem. Phys. 2010,
12, 2590-2598; d) M. Kordel, S. Schooss, C. Neiss, L. Walter, M. M.
Kappes, J. Phys. Chem. A 2010, 114, 5509-5514; e) K. Honma, Phys.
Chem. Chem. Phys. 2018, 20, 26859-26869.
[20] M. U. Winters, J. Kärnbratt, M. Eng, C. J. Wilson, H. L. Anderson, B.
Albinsson, J. Phys. Chem. C 2007, 111, 7192-7199.
[21] F. O. Talbot, A. Rullo, H. Yao, R. A. Jockusch, J. Am. Chem. Soc. 2010,
132, 16156-16164.
[8]
[9]
a) M. H. Stockett, J. Houmøller, K. Støchkel, A. Svendsen, S. Brøndsted
Nielsen, Rev. Sci. Instrum. 2016, 87, 053103; b) M. H. Stockett, J.
Houmøller, S. Brøndsted Nielsen, J. Chem. Phys. 2016, 145, 104303.
a) S. M. J. Wellman, R. A. Jockusch, J. Phys. Chem. A 2015, 119, 6333-
6338; b) J. A. Wyer in Photophysics of Ionic Biochromophores (Eds.: J.
9
This article is protected by copyright. All rights reserved.

10.1002/anie.202008314
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
RET
RET 89%
RET 4%
HN
O
N
O
O
O
HN
N
NH
HN
A communication barrier: A -spacer blocks the electronic communication between two ionic rhodamine dyes in a dyad isolated in
vacuo as it screens for the electric field that each dye senses from the other. As a result energy transfer based on the dipole-dipole
interaction is less efficient than that for systems with alkyl spacers. This conclusion is reached from gas-phase ion fluorescence
spectroscopy of monomers and dimers prepared by acetylenic coupling reactions.
Institute and/or researcher Twitter usernames: Mogens Brøndsted @mogens_br
10
This article is protected by copyright. All rights reserved.