Graphene Liquid Cell Technology

• Graphene liquid cells are sealed nanofluidic enclosures using atomically thin graphene membranes to confine liquids or hydrated specimens.
• They achieve zeptoliter volume control and sub-nanometer spatial resolution via precise fabrication and integrated spacer layers.
• The technology enables in situ electron and photon spectroscopy, unveiling dynamic nanoscale phenomena and chemical interfacial interactions.

A graphene liquid cell is a hermetically sealed micro- or nanofluidic enclosure in which liquids or hydrated specimens are confined between two atomically thin graphene membranes or a graphene membrane and a substrate. Owing to the exceptional electron transparency, mechanical strength, molecular impermeability, and conductivity of graphene, these devices enable in situ imaging, spectroscopy, and nanoscale manipulation of fluids, nanoparticles, and biomolecular assemblies under native, hydrated, or electrolytic conditions while maintaining compatibility with high vacuum instrumentation such as transmission electron microscopy (TEM), scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS), and near-field infrared microscopy. The platform provides sub-nanometer to nanometer spatial resolution, precise volume control (reaching the zeptoliter scale), and unique opportunities to study physical, chemical, and biological processes at solid–liquid, gas–liquid, and electrolyte interfaces.

1. Design and Fabrication of Graphene Liquid Cells

Graphene liquid cells (GLCs) are fabricated by encapsulating a thin layer of liquid or hydrated specimen between two graphene membranes or between graphene and a supporting substrate. Two widely used geometries are: (1) parallel sheets of monolayer or few-layer graphene trapping the liquid, often with a spacer layer (e.g., hBN) to control pocket thickness, and (2) a single-sheet configuration where graphene covers a droplet on a flat substrate, forming a "nano-sandwich" (Shin et al., 2014, Kelly et al., 2017, Khatib et al., 2015, Arble et al., 2021).

Fabrication protocols typically involve:

• Synthesis of monolayer or few-layer graphene via CVD on Cu or Ni foils, followed by PMMA-assisted wet transfer onto the target substrate or grid.
• Cleaning steps (acetone, isopropanol washes, annealing) to minimize polymer residues and improve adhesion and transparency.
• For precision volume control, lithographically patterned hBN or SiN spacers of defined thickness and lateral dimensions are sandwiched between graphene sheets, forming cylinders or wells ranging from 100 nm to several microns in diameter and 10–100 nm depth, yielding volumes down to the zeptoliter regime (Kelly et al., 2017).
• In nanopipette-type cells, a sharp metal tip is coated with graphene, trapping liquid volumes <100 zL at the apex, with volume regulated by the encapsulation speed and the number of encapsulation cycles (Qiu et al., 2020).
• Graphene transfer methods (PMMA, dry stamping) and sealing (van der Waals adhesion, capillary forces, or external frames) are optimized to ensure molecular leak-tightness and mechanical robustness, even under UHV or pressure differentials >1 atm (Kraus et al., 2014, Stoll et al., 2012, Guo et al., 2016).

Table: Key Parameters in GLC Architecture

Parameter	Typical Values/Range	Notes
Graphene thickness	0.34 nm/ML, 1–3 ML standard	Balances transparency and mechanical strength
Spacer layer (hBN)	10–100 nm	Defines cell thickness and liquid volume
Well diameter	100 nm – 1.5 µm (patterned); tens–hundreds nm (pockets)	Controls volume, spatial resolution
Encapsulated volume	1–200 zL (patterned wells); <100 zL (nanopipette)	Zeptoliter precision possible

2. Physical Principles Enabling In Situ Imaging and Spectroscopy

Graphene GLCs exploit several critical material properties:

• Electron/Photon Transparency: Monolayer graphene transmits >50% of low-energy electrons (>300 eV) and is near-transparent in the visible and mid-IR range, permitting direct use in TEM, SEM, XPS, X-ray fluorescence, and IR nanospectroscopy. The Beer–Lambert law for electron attenuation through thickness $d$ is $T(E, d) = \exp(-d/\lambda(E) \cos\theta)$, with $\lambda(E)$ as the inelastic mean free path (Kraus et al., 2014, Stoll et al., 2012).
• Mechanics and Impermeability: Young’s modulus $\sim$1 TPa and atomic thickness allow stable spanning of micron-scale apertures under up to bars of differential pressure (Shin et al., 2014, Kraus et al., 2014). Atomic-scale defect density and van der Waals adhesion confer molecular impermeability to both liquids and gases.
• Electrical Conductivity: Sheet resistance $\sim$200 $\Omega$/sq prevents charging under electron or X-ray irradiation, essential for imaging and spectroscopic stability (Yang et al., 2015, Stoll et al., 2012).
• Chemical Inertness: Graphene is inert in a wide range of electrolytes and can be used with aqueous, organic, and even toxic/reactive liquids (Kraus et al., 2014, Guo et al., 2016).
• Compatibility with Advanced Excitation Modes: The platform allows for high-current, low-energy probing (1–200 keV), and the ultrathin window minimizes beam-induced scattering and background attenuation.

3. Imaging, Spectroscopy, and Analytical Performance

GLCs have demonstrated unprecedented analytical resolution in multiple modalities:

• TEM/STEM: Sub-nanometer resolution (<0.1 nm lattice imaging) and single-nanometer elemental mapping have been achieved with engineered van der Waals GLCs (Kelly et al., 2017). Liquid thickness is precisely controlled by hBN spacers, confirmed by EELS (low-loss, oxygen K-edge). Mapping of ~1.5 nm Fe shells coating 6–8 nm Au cores in liquid was accomplished with spatial FWHM ≈1 nm.
• SEM/EDX: Single-layer graphene enables <5 nm spatial resolution in liquid at 10 keV, with robust elemental analysis (70% transmitted Au Lα intensity in liquid vs dry reference) (Yang et al., 2015, Stoll et al., 2012). SE escape from nanometer-depths allows real-time imaging of Brownian motion, aggregation, and dynamic interactions with ~30 ms time resolution.
• Photoelectron Spectroscopy (XPS/PEEM): GLCs support in situ XPS/PEEM of liquids at ambient or elevated pressure. Electron attenuation lengths of 1–2 nm (for monolayer and few-layer graphene) keep transmission above 20–60% for O 1s at 400–900 eV (Kraus et al., 2014, Guo et al., 2016, Nemšák et al., 2018). Full-field imaging reveals chemical changes, electrical double-layer structure, and interfacial reactions under applied potential.
• Infrared Near-field Nanospectroscopy: Graphene enables s-SNOM and nano-FTIR of hydrated biomolecules, with spatial resolution ~20–30 nm. Amide I/II bands of proteins and nanoconfined water bending modes are resolved with spectral features preserved compared to dry-state controls, demonstrating minimal perturbation of electronic/vibrational structure by the graphene interface (Khatib et al., 2015).
• X-ray Fluorescence and Synchrotron-based Techniques: 2–4 layer graphene cells facilitate combined SEM, FTIR, and XRF imaging of hydrated cells, with integrated hydrogel pads to prolong hydration; elemental O mapping confirms preservation over hours to days (Arble et al., 2021).

4. Nanofluidic Phenomena and Direct Observation of Dynamic Processes

GLCs have enabled real-time, atomic-resolution observation of gas–liquid and solid–liquid interfacial phenomena previously inaccessible:

• Nanobubble Stability and Dynamics: Direct imaging in UHV-TEM revealed a critical nanobubble radius $R_{\mathrm{crit}}\approx4$–6 nm below which bubbles collapse within 1 min and above which stability persists >10 min. Contact angle measurements showed $\theta_c\approx71^\circ$ independent of size. Merging dynamics demonstrated both Ostwald ripening and coalescence pathways, with size-dependent criteria for each (Ostwald for $R'/R<0.7$, coalescence for $R'/R\gtrsim0.8$) (Shin et al., 2014).
• Gas Transport Mechanisms: GLCs reveal a direct gas diffusion mode through ultrathin water films, bypassing dissolution into the liquid and contrasting with classical three-step gas transport (condensation, diffusion, evaporation). This manifests as instantaneous gas transfer events during bubble coalescence, driven by pressure differences $\Delta P$ up to hundreds of MPa in <2 nm water films.
• Brownian and Correlated Motion: Tracking of sub-nanometer metallic clusters (e.g., W in EGLCs) enabled extraction of diffusion coefficients two to six orders of magnitude lower than in bulk solution, reflecting nanoconfinement and substrate interactions (Kelly et al., 2017). Pre-coalescence correlated motion and size-filtered diffusivity analyses are directly accessible due to high spatial and temporal resolution.

5. Applications Across Analytical and Physical Sciences

GLCs form a versatile methodological platform spanning multiple research domains:

• Electrochemistry: Multiplexed microchannel GLCs enable direct PEEM/XPEEM and potentiostatic studies of electrified liquid–solid interfaces, mapping chemical state and ion adsorption at nanoscale resolution under dynamic bias (Guo et al., 2020, Guo et al., 2016, Nemšák et al., 2018). The lateral isolation of chambers allows combinatorial and parallel screening within a single device.
• Biological and Soft Matter Science: Addressable graphene encapsulation supports multi-modal imaging and spectroscopy (SEM, FTIR, XRF) of live or fixed cells, viruses, and biomimetic gels in hydrated state, without cryo-fixation or chemical dehydration (Arble et al., 2021, Khatib et al., 2015). Hydration lifetime can be prolonged by integrated hydrogel pads, maintaining viable samples over extended durations in both vacuum and ambient conditions.
• Nanoanalysis and Manipulation: Graphene nanopipettes (GNPs) deliver and confine zeptoliter-scale liquid volumes for atom probe tomography (APT), achieving 3D chemical mapping of solvated single molecules at sub-nanometer resolution (Qiu et al., 2020). This enables single-molecule delivery and quantitative intracellular sampling (nanobiopsy) and addresses critical challenges in analytical nanochemistry.
• Catalysis, Environmental Science, Energy Storage: In situ tracking of nucleation and growth (e.g., in electrocatalysis or crystallization), operando observation of redox and ion intercalation phenomena, and mapping of gas evolution during battery operation are possible at spatial and temporal scales inaccessible to conventional liquid cells (Shin et al., 2014, Kelly et al., 2017, Guo et al., 2016).
• Spectro-Microscopy of Complex, Reactive, or Toxic Fluids: The molecular impermeability and robustness of graphene windows allow for safe encapsulation and high-fidelity spectroscopic analysis of toxic, radioactive, or highly reactive solutions, minimizing contamination or system damage risk (Kraus et al., 2014).

6. Performance Limitations, Practical Challenges, and Optimization Strategies

Despite advances, the operation of GLCs is subject to constraints:

• Radiation Damage and Bubble Formation: Under high-dose electron or X-ray exposure, radiolytic decomposition (water → H₂, O₂, radicals) and gas bubble formation beneath graphene can lead to delamination or rupture, with typical endurance thresholds at $\sim10^4$ e⁻/nm² (20 keV) (Stoll et al., 2012, Kraus et al., 2014). This limits continuous imaging and imposes the need for low-dose, fast-scan, or intermittent acquisition strategies.
• Membrane Integrity and Leakage: Defects (tears, wrinkles, grain boundaries) and PMMA residue from transfer can compromise seal quality, necessitating careful transfer protocols, use of thicker (3–5 ML) or multilayer (hBN, composite) membranes, and, where possible, dry transfer approaches (Stoll et al., 2012, Arble et al., 2021, Kelly et al., 2017).
• Hydration Control: For prolonged experiments in vacuum or high-vacuum compatible modalities, water retention is enhanced by integrated hydrogel pads, multilayer membrane stacks, and microfluidic channels (Arble et al., 2021). However, diffusional water loss remains a limiting factor for the thinnest cells, especially under sustained vacuum.
• Analytical Throughput and Statistical Reproducibility: In arrayed devices, localized membrane failure is non-catastrophic—release from isolated wells does not compromise the full array (Kelly et al., 2017, Guo et al., 2016), enabling statistically meaningful, combinatorial studies.
• Chemical Reactivity: While graphene is inert in most aqueous and organic media, radiolytic and photonic products (e.g., hydroxyl, peroxide) can attack defects; employing alternative 2D materials (h-BN, MoS₂) is a prospective optimization for selective applications (Kraus et al., 2014, Arble et al., 2021).

7. Outlook and Technological Prospects

GLCs are being extended to more sophisticated architectures and analytical methodologies:

• Integration with Microelectronic and Microfluidic Systems: Patterned electrodes, heating elements, and channels within hBN spacers or polymer frames for electrochemical or flow-through studies are being developed (Kelly et al., 2017, Arble et al., 2021, Nemšák et al., 2018).
• Multimodal and Time-Resolved Measurements: The combination of in situ IR, optical, and X-ray spectroscopy with high-resolution electron microscopy, as well as pump–probe and ultrafast stimulation, are actively explored using GLCs (Guo et al., 2016, Arble et al., 2021).
• Correlative Single Particle/Sample Analysis: GNPs make possible correlative sequential measurements (e.g., cryo-TEM followed by APT) on the same trapped single molecule or biological vesicle (Qiu et al., 2020).
• Combinatorial and High-Throughput Approaches: Microchannel and arrayed GLCs support statistical and machine learning–driven analyses of large sample populations, with prospects for high-throughput materials screening and data mining (Guo et al., 2016, Nemšák et al., 2018).

GLCs thus bridge the longstanding "pressure gap" between atomic-resolution electron or photon microscopies and true liquid-phase or ambient-pressure environments, enabling the direct visualization, quantification, and manipulation of physical, chemical, and biological processes at the nanometric length scale with unprecedented fidelity and versatility (Shin et al., 2014, Kelly et al., 2017, Yang et al., 2015, Stoll et al., 2012, Kraus et al., 2014, Guo et al., 2020, Arble et al., 2021, Guo et al., 2016, Nemšák et al., 2018, Qiu et al., 2020, Khatib et al., 2015).