LO-Chirality: Single-Particle Handedness Mapping

• LO-Chirality is defined as the quantification and spatial mapping of chiral nano-objects using the dissymmetry factor in circularly polarized luminescence.
• It employs photon-count-limited spectroscopy with stringent calibration and artifact correction to achieve single-particle resolution of handedness.
• Machine learning algorithms enable over 99% accurate handedness classification and spatial mapping, uncovering local chiral amplification phenomena.

Local‐Optical Chirality (LO-Chirality) refers to the ability to distinguish, quantify, and spatially map the handedness (chirality) of individual chiral nano-objects at the single-particle level through their interaction with circularly polarized electromagnetic fields. In practice, LO-Chirality quantifies the local dissymmetry of photon emission from a chiral emitter, operationally captured via the dissymmetry factor in circularly polarized luminescence (CPL). Advances in photon-count-limited spectroscopy combined with machine-learning–assisted analysis now enable robust, quantitative determination and mapping of LO-Chirality in single nanocrystals, molecules, or aggregates. This framework establishes new experimental and theoretical standards for probing chiral symmetry breaking, chiral material growth, and local manifestations of chirality at the mesoscopic scale.

1. Theoretical Framework: Optical Chirality and Luminescence Dissymmetry

LO-Chirality in the optical context is rooted in the asymmetry of electromagnetic field interactions with chiral matter. For a chiral emitter, the key observable is the emission circular intensity differential (ECID), defined as

$$\mathrm{ECID}(\lambda) \equiv I_L(\lambda) - I_R(\lambda),$$

where $I_L$ and $I_R$ denote the detected intensities of left‐ and right‐circularly polarized emission, respectively. To normalize by total signal strength, the dissymmetry factor (also called $g_\mathrm{lum}$ or glum) is introduced:

$$g_\mathrm{lum}(\lambda) \equiv \frac{2\left[I_L(\lambda) - I_R(\lambda)\right]}{I_L(\lambda)+I_R(\lambda)}, \qquad -2 \leq g_\mathrm{lum} \leq 2.$$

In most physical realizations, $|g_\mathrm{lum}| \ll 1$; values for Eu$^{3+}$ transitions in chiral nanocrystals can reach $0.15$–$0.4$ on individual lines.

Electromagnetic theory also allows the definition of a local optical chirality density,

$$C(\mathbf{r}, t) = -\frac{\varepsilon_0 \omega}{2} \mathrm{Im}\left[\mathbf{E}^*(\mathbf{r}, \omega) \cdot \mathbf{B}(\mathbf{r}, \omega)\right],$$

which, for a pure circularly polarized plane wave, distinguishes sign and magnitude for left‐ vs right‐circular polarization. The experimentally observable ECID is therefore directly proportional to the difference in local optical chirality density as sampled by the emitter (Vinegrad et al., 2018).

2. Experimental Methodology: Photon-Count–Limited CPL Spectroscopy

A robust LO-Chirality measurement at the single-nanocrystal level requires precise photon-count–limited CPL spectroscopy, with stringent calibration against artifacts. The canonical workflow comprises:

1. Excitation: A supercontinuum source, spectrally filtered and converted to circular polarization via a quarter-wave plate, excites nanocrystals on a mapped substrate. This ensures uniform response independent of nanocrystal orientation.
2. Reference and Beam Splitting: A non-polarizing beamsplitter provides a real-time reference arm to monitor and correct for laser-power drifts.
3. Sample Scanning/Collection: A high-NA objective focuses light on isolated nanocrystals, whose locations are pre-mapped by TEM and confirmed optically.
4. Emission Polarization Analysis: Emitted light, after blocking residual excitation wavelength, passes through a waveplate and tunable linear polarizer. Rotating the polarizer selectively transmits LCP or RCP emission for sequential measurement.
5. Spectral Detection: Collected luminescence is fiber-coupled into an imaging spectrograph and detected on a cooled EM-CCD for maximal photon collection efficiency.

Crucially, systematic biases are quantified by calibrating with an ensemble known to be racemic ($I_L$0) and scaling detected intensities accordingly. Detection sensitivity is fundamentally limited by photon shot noise: with total counts $I_L$1, the statistical uncertainty in $I_L$2 is $I_L$3, permitting confident detection of dissymmetry factors down to $I_L$4 in single 100-second measurements. Signal-to-noise is optimized for high-glum transitions and integration times are extended as needed for lower-dissymmetry lines (Vinegrad et al., 2018).

3. Machine Learning–Enabled Handedness Assignment and Spatial Mapping

Instead of solely relying on $I_L$5 subtraction, single-polarization emission spectra (either LCP or RCP) were found to contain rich handedness-specific fingerprints, notably in line ratios and spectral features. Applying a support vector machine (SVM) classifier trained on normalized spectral feature ratios, researchers achieved greater than 99% accuracy in distinguishing “D” vs “L” handedness nanocrystals—validated by cross-examination and relabeling of outlier samples. Core normalized features include:

• Height ratio $I_L$6 nm / $I_L$7 nm
• Area ratio $I_L$8 nm / $I_L$9 nm
• Height ratio $I_R$0 nm / $I_R$1 nm
• Height ratio $I_R$2 nm / $I_R$3 nm

After classifier training, deployment on a racemic nanocrystal ensemble yielded a chirality map: the spatial coordinates of each nanocrystal (measured via TEM) were color-coded by predicted handedness, resolving both the global $I_R$4 ratio and sub-micrometer-scale spatial distributions. Notably, clustering of like-handed nanocrystals revealed local chiral amplification or symmetry-breaking phenomena, only observable due to single-particle spatial resolution (Vinegrad et al., 2018).

Performance metrics for the classifier after one relabeled outlier:

• Accuracy = 1.00
• Precision$I_R$5 = 1.00, Recall$I_R$6 = 1.00
• Same for L

In a mapped ensemble of 162 nanocrystals, the result $I_R$7 was consistent with the 1:1 expectation for a racemic mixture but allowed the detection of non-statistical clustering at the local scale.

4. Sensitivity, Generalization, and Extensions

The approach requires as few as $I_R$8–$I_R$9 collected photons for robust LO-Chirality determination at the single-particle level. This enables extension to single molecules or aggregates, provided their $g_\mathrm{lum}$0 and total emission prior to photobleaching exceeds $g_\mathrm{lum}$1 detected photons. Key strategies for improved sensitivity include:

• Employing photoelastic modulators (PEMs) for rapid $g_\mathrm{lum}$2 switching and lock-in detection, which mitigates slow drift errors.
• Increasing solid angle collection (higher-NA objectives), custom achromatic retarders for wider spectral coverage, and improved optical isolation.
• Combining with super-resolution methods (e.g., STED, PALM/STORM) to spatially resolve intra-entity chirality distributions, relevant for biological aggregates.

The ultimate detection limit is governed by shot noise ($g_\mathrm{lum}$3) and residual systematic polarization biases ($g_\mathrm{lum}$4) in the collection optics, necessitating regular calibration (Vinegrad et al., 2018).

5. Physical Significance: Symmetry Breaking, Amplification, and Probing Mechanisms

LO-Chirality mapping provides a tool of exceptional sensitivity and spatial resolution for fundamental studies of chiral symmetry breaking in material growth and physicochemical processes. The ability to assign and spatially map the handedness of individual entities enables direct investigation of questions such as:

• Whether chiral amplification occurs locally, resulting in non-random clustering of like-handed enantiomers.
• The existence and scale of spontaneous symmetry breaking in ensembles designed to be racemic, informing the understanding of chiral phase transitions and growth mechanisms.
• The effect of synthesis, environmental, or external-field perturbations on the distribution and spatial organization of chirality.

LO-Chirality quantification thus underpins experimental access to symmetry-breaking dynamics in nanomaterials and could, by extension, inform the design of chiral photonic, catalytic, or quantum devices at the mesoscopic scale.

6. Comparative Perspective and Outlook

LO-Chirality, defined and operationalized via photon-count–limited CPL spectroscopy and machine-learning correlative microscopy, realizes a local, quantitative, and spatially resolved definition of chirality absent in traditional ensemble-averaged approaches. It circumvents artifacts endemic to circular dichroism or differential scattering microscopy and provides a formal link to local optical chirality densities from field theory.

Given the minimal photonic budget required, and the compatibility with advanced imaging and computational analysis schemes, LO-Chirality measurement is poised to become a standard in nanoscience and materials research. It offers a unique probe of symmetry-breaking, growth phenomena, and spatial heterogeneity in systems ranging from single nanocrystals to complex biological assemblies. Prospective improvements in optical components, detection schemes, and data-driven analysis promise further gains in sensitivity and throughput.

The methodology and framework established for LO-Chirality are expected to have substantial impact on the understanding and exploitation of chirality in nanoscale systems, materials design, and fundamental studies of spontaneous symmetry breaking in condensed matter and molecular science (Vinegrad et al., 2018).