WorksPhysics and Astronomy
Witnessing Molecular Conformation via Optical Cycling Centers
Abstract
Molecular conformation plays a central role in determining molecular properties and dynamics. Here, we report the production and conformer-resolved spectroscopy of cold calcium 3-hydroxyphenoxide radical (CaOCH-3-OH) in a cryogenic buffer-gas cell. Internal rotation of the hydroxyl group gives rise to distinct - and -conformers. The CaO moiety acts as an optical cycling center (OCC), enabling the and transitions of both conformers to be resolved by laser-induced fluorescence spectroscopy. Dispersed fluorescence measurements reveal diagonal vibrational branching fractions of 0.91-0.93, indicating favorable optical cycling characteristics. Following excitation of off-diagonal vibronic transitions near the calculated torsional barrier in the state, the emission remains predominantly conformer specific, indicating that conformational identity is largely preserved over the excited-state fluorescence lifetime. These results demonstrate that a metal-centered OCC can serve as an intrinsic, site-specific optical reporter of molecular conformation, providing a foundation for repetitive optical readout and studies of conformational dynamics at the single-molecule level.
Introduction
Molecular conformers are distinct spatial arrangements of a molecule with identical atomic connectivity. Their structural differences influence chemical reactivity and molecular recognition, leading to distinct spectroscopic signatures.[1, 2] Conformers have therefore been widely investigated in spectroscopy,[3, 4, 5] reaction dynamics [6, 7, 8, 9], and intramolecular tunneling.[10, 11, 12] Nevertheless, the selective identification, preparation, and control of individual conformers remain challenging because small torsional barriers can allow rapid interconversion at ambient temperature. Low-temperature techniques, such as cryogenic matrix isolation [10, 11, 12], supersonic beam expansion[5, 13], buffer-gas cooling [4, 14, 15], and sympathetic cooling of ions [6, 16], can suppress thermal interconversion, enabling conformer-resolved spectroscopic characterization and even selective preparation. Despite these advances, many existing methods rely on destructive detection schemes, such as photoionization [5] or photodissociation [8]. While infrared [10, 11] and microwave spectroscopy [17, 18] can probe conformers non-destructively, their relatively small absorption cross-sections limit the sensitive, repetitive readout required for quantum state manipulation at the single-molecule level. A bright, intrinsic optical readout has yet to be realized for conformational isomers.
Optical cycling centers (OCCs) provide a promising route toward such an optical interface. An OCC can often be realized from an alkaline-earth metal atom, such as Ca or Sr, attached to the molecular framework either directly or through an electronegative atom, as in CaO– and SrO–.[19, 20, 21] When incorporated into molecules, the unpaired radical electron remains strongly localized on the metal center, giving rise to atomic-like electronic structures. The resulting minimal change in molecular geometry upon excitation leads to highly diagonal Franck–Condon factors (FCFs) and nearly closed optical cycling transitions. These properties enable repeated photon scattering and have advanced laser cooling and trapping of diatomic molecules, such as SrF [22], CaF [23], BaF [24], AlF [25], and small polyatomic molecules including SrOH [26], CaOH [27], and CaOCH [28]. The functionalization of OCC moiety on large polyatomic molecules has been proposed theoretically [19, 20, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38] and demonstrated spectroscopically for a series of alkoxides,[39, 40] aromatic compounds [41, 42, 43, 44, 45], and adamantane-based derivatives,[46] thereby extending the potential of optical cycling to organic systems and opening new frontiers in precision measurement and quantum information science.[47, 48, 49]
The optical signature of an OCC serves not only as a handle for quantum control but also as a sensitive probe of its local electronic environment. In aromatic systems of MOPh-X (M = Ca or Sr, Ph = phenyl, X = substituent), ring substituents modulate the metal-centered transition frequencies and FCFs through inductive and resonance effects, allowing, in principle, an OCC to report detailed chemical information about its ligands.[33, 34, 43, 44, 45, 46] However, a fundamentally different and as-yet-unexplored question remains: can an OCC distinguish between two conformations of the same molecule, where the perturbation to the metal center’s electronic environment arises solely from a change in the spatial orientation of a functional group, rather than from a change in chemical identity?
Here, we address this question using calcium 3-hydroxyphenoxide (CaOPh-3-OH) as a model system. As shown in Figure 1a, the precursor resorcinol (HOPh-3-OH) reacts with Ca atoms to form CaOPh-3-OH (See SI for experimental details). The hydroxyl substituent at the meta position gives rise to anti and syn conformers, defined by its orientation relative to the Ca-O axis and separated by a calculated torsional barrier of 0.15 eV. Using laser-induced fluorescence (LIF) excitation spectroscopy, we demonstrate that the OCC unambiguously distinguishes the two conformers through their distinct and transitions, while dispersed LIF (DLIF) spectroscopy further reveals favorable vibrational branching fractions (VBFs) for both conformers. The resolved conformer-specific transitions and favorable VBFs support the development of a bright, intrinsic, and potentially repetitive optical readout of conformational states, an important step toward quantum state manipulation of individual molecules. We also use off-diagonal excitation to access vibrational levels with energies approaching the torsional barrier in the state and use conformer-specific fluorescence to search for vibrationally induced interconversion. These measurements assess the capability of a metal-centered OCC to distinguish conformational states while retaining favorable optical cycling properties and to serve as an optical probe of remote structural dynamics.

Figure 1. (a) The reaction of resorcinol with metastable excited Ca atoms can produce single–OCC molecules with two conformations of CaOPh-3-syn-OH and CaOPh-3-anti-OH, and bi–OCC molecule CaOPh-3-OCa. (b) Excitation scan of the reaction products shows four peaks that are respectively assigned to syn-B (i.e., of CaOPh-3-syn-OH), anti-B, anti-A and syn-A. (c) Calculated energy levels of , and levels of both conformers and the corresponding natural transition orbitals of CaOPh-3-OH. anti-X has a lower energy of 11 meV than syn-X. Calculated transition energies are given.
Results and Discussion
Conformer-resolved excitation spectroscopy and OCC-based conformational sensing. The reaction of metastable excited Ca atoms with resorcinol in a cryogenic buffer gas cell can produce three species: the anti and syn conformers of the single–OCC molecule (CaOPh-3-OH), and a bi–OCC product (CaOPh-3-OCa), as illustrated in Figure 1a. The LIF excitation spectrum of the reaction products in the cryogenic cell shows four well-resolved peaks in the wavelength range of 608-618 nm (Figure 1b).
To assign these features, we performed electronic structure calculations for both conformers of the single–OCC species, considering the two lowest electronically excited states, and (See SI for computational details). The results, summarized schematically in Figure 1c, provide two complementary pieces of evidence for the assignment. First, the anti conformer is calculated to lie 11 meV below the syn conformer in ground-state energy. Its stronger LIF signals are qualitatively consistent with a larger population. However, thermal equilibrium at 20 K would strongly suppress the syn population. The clear observation of the syn conformer therefore indicates a kinetically trapped, nonequilibrium distribution during molecular formation and buffer-gas cooling. Second, the calculated excitation energies increase in the order of syn-A (1.918 eV) < anti-A (1.923 eV) < anti-B (1.935 eV) < syn-B (1.937 eV), placing the two anti conformer transitions between those of the syn conformer. Taken together, these results readily assign the four peaks in Figure 1b, in order of increasing excitation energy, to syn-A, anti-A, anti-B and syn-B, respectively. The measured – splittings are ≈ 15 meV and 21 meV for the two conformers. anti and syn conformers, respectively, in good agreement with the corresponding calculated values of 12 meV and 19 meV. For both conformers, the natural transition orbitals (NTOs) involved in these transitions are predominantly localized on the Ca atom, with −type character in the ground state and −type character in the two excited states, consistent with previous studies. [33, 34, 35, 41, 42, 43, 44, 45] No transitions attributable to the bi−OCC species are observed in the excitation spectrum. This absence is likely due to a low formation yield under our experimental conditions, as formation requires either secondary reactions, involving metastable metal atoms, or three-body collisions. [37, 50]
These results establish the OCC as a local optical reporter capable of distinguishing conformers separated by a small interconversion barrier based on their resolved electronic transitions. Additional calculations on naphthalene- and pyrene-based analogues indicate that the calculated separation between the anti-A and syn-A transition energies decreases as the rotating OH group is positioned farther from the OCC (Figure S1), suggesting that the conformational sensitivity of the OCC is predominantly local.

Figure S1. Computed TDDFT vertical excitation energies for a series of increasingly large conjugated molecules. The blue and red points correspond to the excitation energy of the state for the syn- and anti- conformers, accordingly.
Conformer-resolved vibrational branching and optical-cycling prospects. Having established conformer-resolved optical signatures, we next characterized the vibrational branching of each conformer to
assess their optical cycling properties. The excitation laser wavelengths were tuned to each of the four peaks in Figure 1b, and DLIF spectra were recorded for radiative decay from the and states of both conformers (Figure 2). For display, the spectra are plotted as wavenumber shifts relative to the excitation frequency and normalized to their respective 0-0 peaks. All four spectra are dominated by sharp diagonal 0-0 peaks, with only weak off-diagonal vibrational features. The vibrational bands were assigned by comparing the observed shifts with calculated harmonic and vibrational-perturbation-theory (VPT) anharmonic frequencies [45, 51] and FCFs (Table S1–Table S2) at the PBE0/def2-TZVPPD level of theory. Vibrational branching fractions were determined from the integrated areas obtained by fitting the bands with Voigt profiles.[43, 44, 45, 46]

Figure 2. DLIF spectra of CaOPh-3-OH species. (a) anti , (b) anti , (c) syn , and (d) syn . The vertical orange lines indicate the theoretical frequencies and FCFs in VPT anharmonic calculations. The identified vibrational decay channels are labeled as , where or denotes the upper electronic state, denotes the ith vibrational mode populated in the state, and the superscript 0 indicates that the decay originates from the vibrational ground level of the upper state. The peaks labeled with Ca represent the Ca atomic lines at 610 nm (), 612 nm () and 616 nm (), respectively. They are produced by laser ablation. The peaks labeled with + indicate the contamination from CaOH, while those with * are unknown contamination. (e) and (f) show calculated normal mode displacement vectors involved in the observed Fermi resonance coupling for the anti and syn conformers, respectively. The corresponding harmonic and VPT anharmonic frequencies are given.
| Modes | Symm. | anti | syn | ||
| Harm freq. | Anharm. freq. | Harm freq. | Anharm. freq. | ||
| A′ | 54 | 54 | 55 | 49 | |
| A′ | 57 | 64 | 58 | 49 | |
| A″ | 231 | 228 | 231 | 216 | |
| A″ | 247 | 236 | 248 | 255 | |
| A′ | 300 | 304 | 297 | 299 | |
| A″ | 357 | 335 | 356 | 322 | |
| A′ | 359 | 361 | 360 | 358 | |
| A″ | 471 | 457 | 468 | 462 | |
| A′ | 500 | 499 | 504 | 502 | |
| A′ | 535 | 532 | 534 | 529 | |
| A″ | 635 | 628 | 635 | 627 | |
| A″ | 650 | 637 | 649 | 635 | |
| A″ | 705 | 683 | 705 | 843 | |
| A″ | 771 | 762 | 779 | 738 | |
| A′ | 807 | 797 | 808 | 583 | |
| A″ | 858 | 841 | 857 | 238 | |
| A″ | 880 | 860 | 875 | 813 | |
| A′ | 980 | 966 | 988 | 1179 | |
| A′ | 1001 | 985 | 1008 | 952 | |
| A′ | 1016 | 1004 | 1016 | 1047 | |
| A′ | 1106 | 1088 | 1098 | 1081 | |
| A′ | 1171 | 1155 | 1168 | 1166 | |
| A′ | 1180 | 1168 | 1191 | 1207 | |
| A′ | 1248 | 1236 | 1231 | 1188 | |
| A′ | 1316 | 1290 | 1325 | 1262 | |
| A′ | 1364 | 1334 | 1362 | 1422 | |
| A′ | 1378 | 1355 | 1381 | 1334 | |
| A′ | 1519 | 1483 | 1500 | 1476 | |
| A′ | 1532 | 1506 | 1544 | 1616 | |
| A′ | 1640 | 1601 | 1653 | 1609 | |
| A′ | 1677 | 1638 | 1667 | 1621 | |
| A′ | 3174 | 3036 | 3171 | 3034 | |
| A′ | 3191 | 3057 | 3184 | 3048 | |
| A′ | 3208 | 3069 | 3205 | 3064 | |
| A′ | 3211 | 3079 | 3222 | 3070 | |
| A′ | 3887 | 3702 | 3887 | 3696 |
Table S1. Calculated harmonic and anharmonic VPT frequencies of both conformers in the ground state . All vibrational modes are ordered with increasing frequency. All units are in .
| anti A–X | anti B–X | ||||||||
| Final states | Harm. freq. | Anharm. freq. | Harm. FCF | Anharm. FCF | Final states | Harm. freq. | Anharm. freq. | Harm. FCF | Anharm. FCF |
| syn A–X | syn B–X | ||||||||
| Final states | Harm. freq. | Anharm. freq. | Harm. FCF | Anharm. FCF | Final states | Harm. freq. | Anharm. freq. | Harm. FCF | Anharm. FCF |
Table S2. Calculated harmonic and anharmonic (VPT) frequencies and Franck-Condon factors (FCF ) of vibrational decays from both and transitions for the anti and syn conformers.
Figure 2a presents the dispersed spectrum following excitation of the origin band of the anti conformer at 615.57 nm. The most prominent off-diagonal features appear near cm, and a magnified view shows two resolved peaks. Based on the calculated vibrational frequencies and FCFs (Table S1–Table S2), the peak at cm is assigned to the transition, where is an in-plane Ca-O/ring stretching mode (Figure 2e). The neighboring peak at cm is assigned to the combination band of two out-of-plane bending modes (Figure 2e). It gains intensity through Fermi resonance mixing with [45], as predicted by VPT calculations, although the calculation gives the opposite relative intensity ordering. A weaker band at is assigned to another in-plane Ca-O/O-H scissoring mode (Figure S2), as predicted by the calculations. Two additional weak bands at and , which are not predicted to carry appreciable intensity by either harmonic or VPT calculations, are tentatively assigned to the fundamental modes and (Figure S2), respectively, based on their calculated frequencies. Features labeled "Ca" originate from atomic Ca emissions.

Figure S2. Vibrational displacements of the lowest seven fundamental vibrational modes for both conformers. The calculated harmonic frequencies and symmetries are given.
The DLIF spectrum of the anti transition is shown in Figure 2b. Three prominent peaks within of the origin are attributed to either atomic Ca emissions or the origin band () [43, 52]. The inset shows two weak peaks near , which are assigned to (Figure 2c) at and combination band at , respectively. As in the spectrum in Figure 2a, the combination band gains
set shows two weak peaks near , which are assigned to (Figure 2c) at and combination band at , respectively. As in the spectrum in Figure 2a, the combination band gains intensity due to Fermi resonance coupling with . Two additional weak features at and are assigned to the overtone of in-plane rocking mode and the fundamental in-plane scissoring mode (Figure S2), respectively.
The spectra of the syn conformer (Figure 2c-Figure 2d) are slightly more congested because of overlapping emissions from atomic Ca and CaOH ("+"). After excluding these contaminant features, the remaining vibrational features closely resemble those of the anti conformer. In the spectrum, the two bands near are assigned to the same Fermi dyad (Figure 2f). Unlike in the anti conformer, the dominant lies closer to the origin and is more intense, consistent with VPT calculations. The calculated Hamiltonians and eigenstate compositions (Table S3 and Table S4) show that the exchange of dominant characters results from the reversed deperturbed energy ordering, while the similar coupling strengths and mixing coefficients indicate essentially equivalent Fermi resonance mixing in both conformers. Two weak features near correspond to and the combination band, as predicted by the VPT calculations (Table S2). The feature at , assigned to the fundamental mode (Figure S2), overlaps substantially with the CaOH contamination. The spectrum (Figure 2d) similarly exhibits the characteristic bands associated with , and . An additional feature at is tentatively assigned to the overtone . Features marked with asterisk do not match any of the calculated frequencies and likely come from unidentified contaminants.
| Basis state | anti | syn | ||
| 298.013 | 6.871 | 294.212 | −6.717 | |
| 6.871 | 294.417 | −6.717 | 298.240 |
Table S3. Effective VPT Hamiltonians for the Fermi dyad in the state. The Hamiltonians are given in the basis . The diagonal elements are the deperturbed zeroth-order energies, and the off-diagonal elements are the Fermi coupling matrix elements. All values are in . The sign of the off-diagonal coupling depends on the phase convention of the basis states.
| Conformer | Component | Energy () | (%) | (%) |
| anti | Lower | 289.113 | 37 | 63 |
| Upper | 303.317 | 63 | 37 | |
| syn | Lower | 289.214 | 64 | 36 |
| Upper | 303.238 | 36 | 64 |
Table S4. Energies and zeroth-order compositions of the mixed eigenstates obtained by diagonalizing the VPT Hamiltonians in Table S3. The dominant contributions are shown in bold.
Overall, both conformers exhibit vibrational branching dominated by the diagonal 0-0 transition. After excluding the identified contaminant contributions, the fitted areas of the observed peaks yield 0-0 branching fractions of 0.91-0.93, with the remaining intensity distributed among a small number of vibrational channels (Table S5). These approximate values do not include systematic uncertainties including wavelength-dependent instrument response, or unresolved and unobserved decay channels.[43, 44, 45, 46] Nevertheless, they are comparable to those reported for other calcium phenoxides derivatives [43, 44, 45, 46] and support conformer-selective fluorescence detection and promising optical cycling.
| anti | ||||
| Final states | Exp. freq. (cm) | Exp. VBF (uncer.) | VPT freq. (cm) | VPT FCF |
| () | ||||
| () | ||||
| () | ||||
| () | ||||
| () | ||||
| () | ||||
| anti | ||||
| Final states | Exp. freq. (cm) | Exp. VBF (uncer.) | VPT freq. (cm) | VPT FCF |
| () | ||||
| () | ||||
| () | ||||
| () | ||||
| () | ||||
| syn | ||||
| Final states | Exp. freq. (cm) | Exp. VBF (uncer.) | VPT freq. (cm) | VPT FCF |
| () | ||||
| () | ||||
| () | ||||
| () | ||||
| () | ||||
| () | ||||
| syn | ||||
| Final states | Exp. freq. (cm) | Exp. VBF (uncer.) | VPT freq. (cm) | VPT FCF |
| () | ||||
| () | ||||
| () | ||||
| () | ||||
| () |
Table S5. Experimental vibrational branching fractions (VBFs) and calculated VPT frequencies and Franck–Condon factors for the observed vibrational decays in DLIF spectra in Figure 2. The uncertainties of the experimental frequencies are within cm. VBF values in parentheses are uncertainties propagated from the fitted peak areas and do not include systematic contributions from the wavelength-dependent instrument response or unresolved and unobserved decay channels, as discussed in Ref. [43, 44, 45, 46].
Off-diagonal excitation scan and search for conformational switching. The favorable diagonal FCFs of the two conformers suggest efficient optical cycling, but also raise the question of whether vibrational excitation in the upper electronic state can induce conformational interconversion. [10, 11, 12, 13, 17] Both conformers possess a low-frequency out-of-plane OH rocking mode associated with the torsional coordinate (Figure 3a). The calculated torsional barrier in the state is 0.15 eV relative to the anti conformer (Figure 3b). Interconversion may occur through tunneling below the barrier or through over-barrier motion when sufficient energy is deposited into, or efficiently coupled to, the torsional coordinate. [11, 12, 17] The OH group can rotate from anti to syn by passing on either side of the molecular plane. Interference between these two paths may affect whether torsional states near or above the barrier are localized around one conformer or spread across both. To probe the possible interconversion, we performed off-diagonal excitation scans (Figure 3c) by tuning the laser wavelength to vibrationally excited levels of the and states () while monitoring fluorescence within a spectral window around the corresponding 0-0 transition. Because the frequencies of corresponding vibrational modes are similar in the ground (Table S1) and excited electronic states (Table S6), the vibrationally diagonal emission () occurs close to the 0-0 transition and falls within this detection window. In this scheme, if conformational switching occurs, excitation of one conformer would produce emission signature of the other conformer. Figure 3d shows excitation spectra from 606 to 570 nm, extending to above the syn-A0 and anti-A0 origins and beyond the calculated torsional barrier in the state near 572 nm. More than thirty vibrational resonances are observed across the (upward) and (downward) spectra of both conformers. These resonances can be broadly categorized into three types: conformer-specific features, like and ; partially overlapped features with contributions to both conformers, such as and ; and fully overlapped peaks with distinct intensity ratios, exemplified by . At higher excitation energies, a larger fraction of features contain contributions from both conformers, reflecting the greater vibrational density of states and resulting spectral congestion.[47, 52] In addition, the overall fluorescence signal decreases rapidly at excitation wavelengths below 580 nm, consistent with the calculated very small FCFs for transitions to higher-lying vibrational levels (Table S2).

Figure 3. (a) Vibrational displacements of torsional mode of both conformers. Both modes involve the OH flipping responsible for the conformational switching. The calculated harmonic frequencies are given. (b) Schematic periodic potentials along the OH torsional coordinate in the and states. The endpoints at and represent the same geometry, giving rotational paths between anti and syn. The calculated rotation barrier of 0.15 eV is given for the state of the anti conformer. The colored curves illustrate vibrational states schematically and are not calculated torsional wavefunctions. The vertical double-headed arrows indicate conformer-specific excitation and fluorescence; the horizontal arrow indicates possible excited-state interconversion. (c) Excitation and detection scheme for the off-diagonal excitation scans. The excitation wavelength is scanned across vibrationally excited levels, while fluorescence is monitored at a fixed wavelength near the corresponding vibrationally diagonal emission. (d) Full excitation spectra of the (upward) and (downward) of both conformers. The emission fluorescence is monitored on respective diagonal decays at 615.2 nm (anti ), 616.1 nm (syn ), 610.7 nm (anti ) and 609.6 nm (syn ). The dash lines indicate the excitation wavelengths of the diagonal 0-0 transitions. The shaded region at 572 nm represents the calculated torsional barrier in the state.
| Mode | anti-A Sym. | anti-A Freq. | anti-B Sym. | anti-B Freq. | syn-A Sym. | syn-A Freq. | syn-B Sym. | syn-B Freq. |
| 53 | 56 | 53 | 56 | |||||
| 60 | 56 | 60 | 57 | |||||
| 231 | 231 | 232 | 231 | |||||
| 249 | 246 | 250 | 247 | |||||
| 298 | 298 | 296 | 295 | |||||
| 359 | 358 | 358 | 355 | |||||
| 359 | 360 | 360 | 360 | |||||
| 471 | 470 | 469 | 468 | |||||
| 500 | 501 | 503 | 504 | |||||
| 535 | 535 | 535 | 535 | |||||
| 637 | 635 | 637 | 635 | |||||
| 649 | 648 | 648 | 647 | |||||
| 705 | 703 | 705 | 703 | |||||
| 772 | 772 | 780 | 780 | |||||
| 807 | 807 | 808 | 807 | |||||
| 860 | 859 | 859 | 856 | |||||
| 882 | 878 | 878 | 874 | |||||
| 981 | 981 | 991 | 989 | |||||
| 1002 | 1001 | 1009 | 1008 | |||||
| 1016 | 1016 | 1016 | 1016 | |||||
| 1106 | 1107 | 1098 | 1099 | |||||
| 1170 | 1170 | 1168 | 1168 | |||||
| 1180 | 1180 | 1191 | 1191 | |||||
| 1246 | 1247 | 1229 | 1229 | |||||
| 1316 | 1316 | 1325 | 1325 | |||||
| 1362 | 1363 | 1363 | 1362 | |||||
| 1374 | 1375 | 1376 | 1377 | |||||
| 1519 | 1519 | 1499 | 1500 | |||||
| 1530 | 1530 | 1542 | 1542 | |||||
| 1642 | 1642 | 1655 | 1654 | |||||
| 1676 | 1676 | 1666 | 1666 | |||||
| 3175 | 3176 | 3171 | 3171 | |||||
| 3192 | 3192 | 3185 | 3185 | |||||
| 3208 | 3208 | 3205 | 3205 | |||||
| 3210 | 3210 | 3222 | 3223 | |||||
| 3886 | 3886 | 3886 | 3887 |
Table S6. Calculated harmonic vibrational frequencies and symmetries of the anti and syn conformers in the and states. All vibrational modes are numbered in order of increasing harmonic frequency. Frequencies are in cm.
The vibrational resonances were tentatively assigned by comparing their excitation-energy offsets with calculated harmonic frequencies and combination energies (Table S6). The excitation scans contain several candidate Fermi dyads. For example, peaks , located 292/309 above syn-A0, and peaks , located 291/307 above anti-A0, are tentatively assigned to the lower and upper members of the Fermi dyad in the state. The individual vibrational labels remain tentative in the absence of excited-state VPT calculations. All assignments are summarized in Table S7.
| Observed peaks in Figure 3d | Excitation wavelength (nm) | (cm) relative to anti-A0 | anti-B0 | syn-A0 | syn-B0 | Assignments | DLIF spectra |
| 605.65 | 266 | 146 | 292 | 120 | , | ||
| 605.05 | 282 | 163 | 309 | 136 | , | Fig. 4 | |
| 604.75 | 291 | 171 | 317 | 144 | , | Fig. S3 | |
| 604.15 | 307 | 187 | 333 | 161 | , | Fig. 4 | |
| 603.35 | 329 | 209 | 355 | 183 | , | Fig. S3 | |
| 602.75 | 346 | 226 | 372 | 199 | , | Fig. S3 | |
| 602.45 | 354 | 234 | 380 | 207 | , | Fig. S3 | |
| 601.15 | 390 | 270 | 416 | 243 | , | Fig. S3 | |
| 600.35 | 412 | 292 | 438 | 265 | , ; , | Fig. S3 | |
| 599.85 | 426 | 306 | 452 | 279 | , | Fig. S3 | |
| 599.45 | 437 | 317 | 463 | 290 | , | ||
| 598.85 | 454 | 334 | 480 | 307 | , | Fig. S3 | |
| 598.25 | 470 | 351 | 497 | 324 | , | ||
| 598.05 | 476 | 356 | 502 | 329 | , | Fig. S3 | |
| 597.55 | 490 | 370 | 516 | 343 | , | ||
| 597.45 | 493 | 373 | 519 | 346 | , | ||
| 597.05 | 504 | 384 | 530 | 357 | , | Fig. S3 | |
| 595.95 | 535 | 415 | 561 | 388 | , | ||
| 595.15 | 557 | 438 | 584 | 411 | , ; , | Fig. S3 | |
| 594.65 | 572 | 452 | 598 | 425 | , | ||
| 594.25 | 583 | 463 | 609 | 436 | , | ||
| 593.85 | 594 | 475 | 621 | 448 | , | ||
| 593.55 | 603 | 483 | 629 | 456 | , ; , | Fig. 4 | |
| 592.65 | 628 | 509 | 655 | 482 | , | Fig. 4 | |
| 592.05 | 645 | 526 | 672 | 499 | , | ||
| 590.05 | 703 | 583 | 729 | 556 | , | ||
| 589.55 | 717 | 597 | 743 | 571 | , | ||
| 588.85 | 737 | 618 | 763 | 591 | , | ||
| 588.45 | 749 | 629 | 775 | 602 | , | ||
| 587.95 | 763 | 644 | 789 | 617 | , | ||
| 587.55 | 775 | 655 | 801 | 628 | , | ||
| 587.05 | 789 | 670 | 816 | 643 | , | ||
| 586.65 | 801 | 681 | 827 | 654 | , | ||
| 585.25 | 842 | 722 | 868 | 695 | , | ||
| 584.85 | 853 | 734 | 880 | 707 | , | ||
| 584.25 | 871 | 751 | 897 | 724 | , | ||
| 582.95 | 909 | 789 | 935 | 763 | , ; , | Fig. 4 | |
| 582.05 | 936 | 816 | 962 | 789 | , | Fig. 4 | |
| 580.85 | 971 | 851 | 997 | 825 | , ; , | Fig. S3 | |
| 577.35 | 1075 | 956 | 1102 | 929 | , | Fig. S3 | |
| 575.95 | 1118 | 998 | 1144 | 971 | , ; , | Fig. S3 | |
| 575.15 | 1142 | 1022 | 1168 | 995 | , ; , | Fig. S3 |
Table S7. Assignments of the off-diagonal vibrational resonances observed in the and states of both conformers in Figure 3d. Excitation energies are reported relative to the corresponding vibrational origins. Tentative assignments are made by comparing the measured energy offsets, , with calculated harmonic frequencies (Table S6) and FCFs (Table S2). Bold values indicate the energy offsets associated with the dominant conformer-state contributions and used for the assignments. The final column identifies the DLIF spectra recorded at selected excitation wavelengths.
To identify the contributing conformers through their characteristic vibrationally diagonal emission, DLIF spectra were recorded for twenty-one resonances selected to cover both electronic states and conformers, a broad range of excitation energies, and both isolated and spectrally overlapping features. Six representative spectra are shown in Figure 4, while the remaining spectra are provided in Figure S3. Figure 4a and Figure 4b show DLIF spectra following excitations of the upper members (604.15 nm) and (605.05 nm), respectively. Both spectra exhibit dominant emissions near the respective anti-A0 and syn-A0, confirming their associations with the corresponding conformers. A weaker anti-A0 feature in Figure 4b arises from coincident excitation of the overlapping resonance (Figure 3d), which is assigned to the anti-A, based on its frequency matching and DLIF spectrum in Figure S3. In Figure 4c, the excitation-energy offset of 628 and the dominant anti-A0 emission support the assignment to anti-A, . Figure 4d is more complex because the excitation wavelength overlaps resonances and in Figure 3d. Its dominant syn-A0 emission is assigned to of the syn-A state, while the weaker feature near the anti-A0/B0 is attributed to excitation of the anti-A, level. The dominant anti-B0 and syn-B0 emissions in Figure 4e and Figure 4f, respectively, are attributed to decays from vibrational level of the corresponding states.

Figure 4. Six representative DLIF spectra recorded following excitations of peaks (a) , (b) , (c) , (d) , (e) , and (f) in Figure 3d. Each spectrum is dominated by vibrationally diagonal emission associated with one of the four conformer-specific electronic origins: anti-, syn-, anti-, or syn-. The excitation energy offsets relative to the corresponding electronic origins and the corresponding assignments are indicated in each figure. Assignments of the excitation resonances are obtained by comparing those offsets with the calculated vibrational frequencies (Table S6).

Figure S3. The remaining fifteen DLIF spectra showing the excitations of selected vibrational resonances in Figure 3d. Each spectrum is dominated by vibrationally diagonal emission associated with one of the four conformer-specific electronic origins: anti-A0, syn-A0, anti-B0, or syn-B0. The excitation energy offsets relative to the corresponding electronic origins and the corresponding assignments are indicated in each figure. Assignments of the excitation resonances are obtained by comparing those offsets with the calculated vibrational frequencies (Table S6). Asterisks denote experimental artifacts.
For all resonances examined by DLIF, the emission remains dominated by the conformer associated with the assigned excitation transition, and no unambiguous evidence for conformational interconversion is observed within the excited-state fluorescence lifetime. This result is consistent with the metal-localized character of the OCC transition (Figure 1c), resulting in weak coupling to the remote OH torsional coordinate and negligible calculated FCF for the torsional mode (Table S2). Consequently, the vibrational levels accessed in the upper electronic states may contain insufficient torsional character to promote measurable switching, even when their total energies approach the calculated barrier. Nevertheless, weak interconversion through sub-barrier tunneling or vibrational coupling to the torsional coordinate cannot be excluded.
Conclusion and Outlook
In summary, we report the conformer-resolved optical spectroscopy of cold CaOPh-3-OH radicals. LIF excitation spectroscopy resolves four conformer-specific electronic origins, assigned to the – and – transitions of the anti and syn conformers. These distinct transitions enable conformer-selective excitation and spectroscopic identification of each conformer. This finding establishes that an OCC can distinguish conformers that differ solely by the spatial orientation of a remote hydroxyl group. DLIF measurements further yield highly diagonal vibrational branching fractions of 0.91–0.93 for all four transitions, indicating that favorable optical cycling properties are retained in both conformers. Off-diagonal excitation scans reveal an extensive manifold of excited vibrational levels in the and states, extending to energies near the calculated torsional barrier in the state. For all levels examined by DLIF, however, the emission remains dominated by the conformer associated with the initially excited transition, and no conclusive evidence for conformational interconversion within the excited-state lifetime is observed.
Together with the recent proposal to use an OCC to probe photoisomerization in azobenzene derivatives,[38] our findings suggest a broader role for OCCs as local optical reporters of molecular structure and dynamics. The sensitivity of the OCC transition to a remote structural change, combined with its favorable vibrational branching, could enable repeated optical readout of molecular conformation. This capability could facilitate studies of photochemical transformations,[38] non-adiabatic couplings[52, 53] and anharmonic vibrational coupling and energy redistribution.[45, 47] Ultimately, molecules incorporating OCCs could provide a platform for the independent control and optical readout of remote structural dynamics in individual molecules.
Acknowledgements – This work was supported by the AFOSR (grant no. FA9550-20-1-0323), the NSF (grant no. OMA-2016245, PHY-2207985, PHY-2110421 and DGE-2034835), NSF Center for Chemical Innovation Phase I (grant no. CHE-2221453). This research is funded in part by the Gordon and Betty Moore Foundation. Computational resources were provided by XSEDE and UCLA IDRE shared cluster hoffman2. This work also used Bridges-2 at the Pittsburgh Supercomputing Center through allocation CHE230091 from the Advanced Cyberinfrastructure Coordination Ecosystem: Services & Support (ACCESS) program, which is supported by U.S. National Science Foundation grants 2138259, 2138286, 2138307, 2137603, and 2138296.
Supplementary Information
Experimental methods
The experiments were conducted within a cryogenic buffer-gas cell operated maintained at 20-25 K.[44, 45, 46] CaOPh-3-OH radicals were produced by reacting Ca atoms with resorcinol (Sigma Aldrich). An Nd:YAG laser (Minilite) operating at 1064 nm with a pulse energy of approximately 6 mJ and a repetition rate of 10 Hz, was used to ablate calcium metal pellet. To minimize variations in molecular production, the focused ablation spot was continuously swept across the target using a movable mirror. The ablated Ca atoms reacted with resorcinol vapor delivered from a heated reservoir through a heated gas line. The resulting molecules were collisionally cooled by neon buffer gas with a number density of approximately . In the excitation region, a tunable pulsed dye laser (PDL, LiopStar-E), operating at 10 Hz with a linewidth of at 620 nm, was utilized to excite the electronic transitions of CaOPh-3-OH. The resulting fluorescence, with a lifetime of approximately 20-30 ns,[41, 43] was collected using a lens system and directed into a monochromator (McPherson Model 2035) equipped with a 1200 lines/mm grating. All excitation scans were recorded using a photomultiplier tube (PMT), with photon counts accumulated by an SRS SR430 multichannel scaler, whereas the DLIF spectra were recorded using an ICCD camera (Andor iStar 320T).
To locate the conformer-specific electronic origins, the PDL wavelength was scanned while the monochromator grating was tuned synchronously to track the corresponding diagonal fluorescence. For off-diagonal excitation scans, the PDL wavelength was scanned while the monochromator grating was fixed near the corresponding electronic origin. Because corresponding vibrational frequencies are similar in the ground and excited electronic states, vibrationally diagonal emission from the excited vibrational levels occur near this origin. DLIF spectra were acquired by fixing the PDL wavelength at a selected excitation resonance and recording the wavelength-dispersed fluorescence with the monochromator and ICCD camera. The PDL wavelength was scanned in increments of 0.1 nm. Signals were accumulated over 100 and 200 laser shots at each excitation wavelength for the electronic-origin and off-diagonal scans, respectively. The origin-excited and off-diagonal DLIF spectra were accumulated over 20,000 and 1,000 laser shots, respectively. Vibrational branching fractions were extracted following the procedures described in Refs. [44, 46] No correction was applied for the wavelength-dependent instrument response because its variation over the relevant spectral range was estimated to change the diagonal branching fractions by less than 1%.
Theoretical methods
Harmonic Franck–Condon factors (FCFs) were computed using the following scheme. The ground-state geometry was first optimized with the PBE0 DFT functional and the def2-TZVPPD basis set.[54, 55] The D3 dispersion correction was applied in all DFT calculations.[56] Harmonic frequencies were then evaluated at the equilibrium geometry, along with the third and fourth derivatives of the potential required for the subsequent anharmonic treatment. Excited-state geometries and harmonic frequencies were obtained for the target excited states using TD-DFT. The ground- and excited-state Hessians were then used to perform a harmonic FCF calculation with the ezFCF software.[57] All DFT calculations were carried out in the Gaussian16 software package.[58]
To account for anharmonic effects such as Fermi resonance, the harmonic states were corrected using the VPT2 Python code.[51] Intensity borrowing was included by multiplying the harmonic FCF by the expansion coefficient of the state participating in the resonance. Our approach followed that established in Ref. [45].
References
- [1]P. J. Canfield, I. M. Blake, Z.-L. Cai, I. J. Luck, E. Krausz, R. Kobayashi, J. R. Reimers, and M. J. Crossley, Nat. Chem. 10, 615 (2018).
- [2]B. Kuhlmann and P. Bradley, Nat. Rev. Mol. Cell Biol. 20, 681 (2019).
- [3]M. S. de Vries and P. Hobza, Annu. Rev. Phys. Chem. 58, 585 (2007).
- [4]T. R. Rizzo, J. A. Stearns, and O. V. Boyarkin, Inter. Rev. Phys. Chem. 28, 481 (2009).
- [5]S. M. Park and C. H. Kwon, J. Phys. Chem. A 125, 9251 (2021).
- [6]Y.-P. Chang, K. Długołęcki, J. Küpper, D. Rösch, D. Wild, and S. Willitsch, Science 342, 98 (2013).
- [7]E. Champenois, D. Sanchez, J. Yang, J. Figueira Nunes, A. Attar, M. Centurion, R. Forbes, M. Gühr, K. Hegazy, F. Ji, et al., Science 374, 178 (2021).
- [8]S. T. Park, S. K. Kim, and M. S. Kim, Nature 415, 306 (2002).
- [9]M. H. Kim, L. Shen, H. Tao, T. J. Martinez, and A. G. Suits, Science 315, 1561 (2007).
- [10]A. Mardyukov, H. Quanz, and P. R. Schreiner, Nat. Chem. 9, 71 (2017).
- [11]R. Fausto, G. C. Odilz, and C. M. Nunes, Chem. Soc. Rev. 51, 2853 (2022).
- [12]M. Pettersson, E. M. Maçôas, L. Khriatchchev, R. Fausto, and M. Räsänen, J. Am. Chem. Soc. 125, 4058 (2003).
- [13]B. C. Dian, J. R. Clarkson, and T. S. Zwier, Science 303, 1169 (2004).
- [14]L. Satterthwaite, G. Koumarianou, P. B. Carroll, R. J. Sedlik Jr, I. Wang, M. C. McCarthy, and D. Patterson, J. Phys. Chem. A 127, 4096 (2023).
- [15]Q. Borengasser, T. Hager, A. Kanaherarachchi, D. Troya, and B. M. Broderick, J. Phys. Chem. Lett. 14, 6550 (2023).
- [16]L. Xu, J. Toscano, and S. Willitsch, Phys. Rev. Lett. 132, 083001 (2024).
- [17]B. C. Dian, G. A. Brown, K. O. Douglass, and B. H. Pate, Science 320, 924 (2008).
- [18]A. O. Hernandez-Castillo, C. Abeysekera, B. M. Hays, I. Kleiner, H. W. L. Nguyen, and T. S. Zwier, J. Mol. Spectrosc. 337, 51 (2017).
- [19]T. A. Isaev and R. Berger, Phys. Rev. Lett. 116, 063006 (2016).
- [20]I. Kozyryev, L. Baum, K. Matsuda, and J. M. Doyle, ChemPhysChem 17, 3641 (2016).
- [21]H. Zhou, T. Khvorost, A. N. Alexandrovа, and J. R. Caram, Acc. Chem. Res. 58, 1181 (2025).
- [22]E. S. Shuman, J. F. Barry, and D. DeMille, Nature 467, 820 (2010).
- [23]L. Anderegg, B. L. Augenbraun, E. Chae, B. Hemmerling, N. R. Hutzler, A. Ravi, A. Collopy, J. Ye, W. Ketterle, and J. M. Doyle, Phys. Rev. Lett. 119, 103201 (2017).
- [24]Z. Zeng, S. Deng, S. Yang, and B. Yan, Phys. Rev. Lett. 133, 143404 (2024).
- [25]J. Padilla-Castillo, J. Cai, P. Agarwal, P. Kukreja, R. Thomas, B. G. Sartakov, S. Truppe, G. Meijer, and S. Wright, Phys. Rev. Lett. 135, 243401 (2025).
- [26]Z. D. Lasner, A. Frenett, H. Sawaoka, L. Anderegg, B. Augenbraun, H. Lampson, M. Li, A. Lunstad, J. Mango, A. Nasir, et al., Phys. Rev. Lett. 134, 083401 (2025).
- [27]N. B. Vilas, C. Hallas, L. Anderegg, P. Robichaud, A. Winnicki, D. Mitra, and J. M. Doyle, Nature 606, 70 (2022).
- [28]D. Mitra, N. B. Vilas, C. Hallas, L. Anderegg, B. L. Augenbraun, L. Baum, C. Miller, S. Raval, and J. M. Doyle, Science 369, 1366 (2020).
- [29]B. L. Augenbraun, J. M. Doyle, T. Zelenivsky, and I. Kozyryev, Phys. Rev. X 10, 031022 (2020).
- [30]M. V. Ivanov, S. Gualandri, and A. I. Krylov, J. Phys. Chem. Lett. 11, 1297 (2020).
- [31]M. V. Ivanov, F. H. Bangerter, P. Wójcik, and A. I. Krylov, J. Phys. Chem. Lett. 11, 6670 (2020).
- [32]J. Kłos and S. Kotochigova, Phys. Rev. Res. 2, 013384 (2020).
- [33]C. E. Dickerson, H. Guo, A. J. Shin, B. L. Augenbraun, J. R. Caram, W. C. Campbell, and A. N. Alexandrovа, Phys. Rev. Lett. 126, 123002 (2021).
- [34]C. E. Dickerson, H. Guo, G.-Z. Zhu, E. R. Hudson, J. R. Caram, W. C. Campbell, and A. N. Alexandrovа, J. Phys. Chem. Lett. 12, 3989 (2021).
- [35]C. E. Dickerson, C. Chang, H. Guo, and A. N. Alexandrova, J. Phys. Chem. A 126, 9644 (2022).
- [36]P. Yu, A. Lopez, W. A. Goddard, and N. R. Hutzler, Phys. Chem. Chem. Phys. 25, 154 (2023).
- [37]T. Khvorost, P. Wójcik, C. Chang, M. Calvillo, C. Dickerson, G. Lao, E. R. Hudson, A. I. Krylov, and A. N. Alexandrova, J. Phys. Chem. Lett. 15, 5665 (2024).
- [38]P. Wójcik, T. Khvorost, G. Lao, G.-Z. Zhu, A. Macias, J. R. Caram, W. C. Campbell, M. A. García-Garibay, E. R. Hudson, A. N. Alexandrova, and A. I. Krylov, J. Phys. Chem. A 129, 1929–1940 (2024).
- [39]A. C. Paul, K. Sharma, H. Telfah, T. A. Miller, and J. Liu, J. Chem. Phys. 155 (2021).
- [40]H. Telfah, K. Sharma, A. C. Paul, S. S. Riyadh, T. A. Miller, and J. Liu, Phys. Chem. Chem. Phys. 24, 8479 (2022).
- [41]D. Mitra, Z. D. Lasner, G.-Z. Zhu, C. E. Dickerson, B. L. Augenbraun, A. D. Bailey, A. N. Alexandrova, W. C. Campbell, J. R. Caram, E. R. Hudson, and J. M. Doyle, J. Phys. Chem. Lett. 13, 7029 (2022).
- [42]B. L. Augenbraun, S. Burchevsky, A. Winnicki, and J. M. Doyle, J. Phys. Chem. Lett. 13, 10771 (2022).
- [43]G.-Z. Zhu, D. Mitra, B. L. Augenbraun, C. E. Dickerson, M. J. Frim, G. Lao, Z. D. Lasner, A. N. Alexandrova, W. C. Campbell, J. R. Caram, J. M. Doyle, and E. R. Hudson, Nat. Chem. 14, 995 (2022).
- [44]G. Lao, G.-Z. Zhu, C. E. Dickerson, B. L. Augenbraun, A. N. Alexandrova, J. R. Caram, E. R. Hudson, and W. C. Campbell, J. Phys. Chem. Lett. 13, 11029 (2022).
- [45]G.-Z. Zhu, G. Lao, C. E. Dickerson, J. R. Caram, W. C. Campbell, A. N. Alexandrova, and E. R. Hudson, J. Phys. Chem. Lett. 15, 590 (2024).
- [46]G. Lao, T. Khvorost, A. Macias, H. W. T. Morgan, R. H. Lavroff, R. Choi, H. Zhou, D. Usvyat, G.-Z. Zhu, M. A. García-Garibay, A. N. Alexandrova, E. R. Hudson, and W. C. Campbell, Nat. Chem. 18, 84–91 (2026).
- [47]B. L. Augenbraun, L. Anderegg, C. Hallas, Z. D. Lasner, N. B. Vilas, and J. M. Doyle, Adv. At. Mol. Opt. Phys. 72, 89 (2023).
- [48]S. L. Cornish, M. R. Tarbutt, and K. R. Hazzard, Nat. Phys. 20, 230 (2024).
- [49]D. DeMille, N. R. Hutzler, A. M. Rey, and T. Zelevinsky, Nat. Phys. 20, 741 (2024).
- [50]E. Brudner, T. Gur, N. Genossar-Dan, P. B. Changala, M. C. McCarthy, J. F. Stanton, and J. H. Baraban, J. Phys. Chem. A 129, 6385 (2025).
- [51]M. A. Boyer and A. B. McCoy, J. Chem. Phys. 157, 164113 (2022).
- [52]H. Zhou, P. Wójcik, G.-Z. Zhu, G. Lao, T. Khvorost, J. R. Caram, W. C. Campbell, A. N. Alexandrova, A. I. Krylov, and E. R. Hudson, Phys. Rev. Res. 7, 043180 (2025).
- [53]P. Wójcik, H. Zhou, T. Khvorost, G.-Z. Zhu, G. Lao, J. R. Caram, A. N. Alexandrova, E. R. Hudson, W. C. Campbell, and A. I. Krylov, J. Chem. Phys. 164, 014306 (2026).
- [54]C. Adamo and V. Barone, J. Chem. Phys. 110, 6158 (1999).
- [55]A. Hellweg and D. Rappoport, Phys. Chem. Chem. Phys. 17, 1010 (2015).
- [56]S. Grimme, J. Antony, S. Ehrlich, and H. Krieg, J. Chem. Phys. 132, 154104 (2010).
- [57]S. Gozem and A. I. Krylov, WIREs Comput. Mol. Sci. 12, e1546 (2021).
- [58]M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsujii, X. Li, M. Caricato, A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendel, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman, and D. J. Fox, “Gaussian 16 Revision C.01,” (2016), gaussian Inc. Wallingford CT.