1. Introduction
Nanofluidics is a field that involves the study and application of fluids confined to nanometer-scale geometries. Although the term “nanofluidics” is not entirely new, the research field itself is still in its early stages. Issues related to nanoscale fluidic phenomena have occasionally been addressed in the contexts of colloid science, membrane science, and chemical engineering since the mid-20th century [
1], [
2], [
3], [
4], [
5], [
6], [
7], [
8], [
9], [
10], [
11]. However, the field remained somewhat dormant until recently owing to the lack of specialized research tools. Recent advancements have seen the field of nanofluidics, particularly with the introduction of experimental tools such as nanopores [
3], [
12], [
13], [
14], [
15], [
16], [
17], [
18], [
19], [
20], [
21], [
22], [
23], [
24], nanotubes [
25], [
26], [
27], [
28], [
29], [
30], [
31], nanopipettes [
4], [
32], [
33], [
34], [
35], nanowires [
36], [
37], [
38], [
39], nanogaps [
40], [
41], [
42], [
43], nanoporous polymer membranes [
44], [
45], [
46], [
47], [
48], [
49], [
50], [
51], two-dimensional layered materials [
52], [
53], [
54], [
55], and more recently chip-based nanofluidic devices (hereinafter referred to as “nanofluidic devices”) [
9], [
56], [
57], [
58], [
59], [
60], [
61], [
62], [
63], [
64].
Nanofluidic devices are compact, solid-state, planar, and transparent devices featuring precisely defined nanochannel structures, commonly fabricated using advanced nanolithography technologies [
5], [
10], [
61], [
62], [
63], [
64], [
65], [
66], [
67], [
68], [
69], [
70], [
71], [
72]. Owing to these characteristics, nanofluidic devices offer the greatest potential for precise and accurate manipulation within nanostructures. The application of nanofluidic devices has led to the discovery of new phenomena and unusual effects confined to the nanoscale in fluids. These include altered liquid properties of water, such as lower electric permittivity [
73], [
74], [
75], [
76], higher proton mobility [
77], [
78], [
79], increased viscosity compared to bulk scales [
74], [
80], and nonlinear transport behaviors, such as ion current rectification [
80], [
81], [
82] and concentration polarization [
18], [
83], [
84], [
85], [
86]. These phenomena and effects, which are not observed in bulk or at the microscale, become prominent at the nanoscale, unveiling new opportunities for insights and novel fluid applications. Hence, while nanofluidics is often viewed as an evolution of the well-established field of microfluidics, it is clearly not merely an extension of microfluidics, and should be recognized as a new, separate field. In addition, the small, planar, and transparent nature of nanofluidic devices facilitates their integration with various microscopes, analytical instruments, and cutting-edge technologies [
67], [
84], [
85], [
86], [
87], [
88], [
89], [
90], [
91], [
92], [
93]. This versatility positions nanofluidic devices for broad application in numerous fields beyond fluidic mechanics, including chemistry, biology, physics, diagnostics, medicine, pharmacy, energy, drug discovery, chemical engineering, materials science, and information sciences [
89], [
94], [
95], [
96], [
97], [
98], [
99], [
100], [
101], [
102], [
103], [
104], [
105], [
106], [
107], [
108], [
109].
In combination with other experimental tools that are commonly used in laboratories, nanofluidic devices offer accessible, efficient, and powerful platforms for the generation, control, observation, and monitoring of nanofluidic flows, phenomena, and effects. Liquids or solutions can be easily introduced into nanochannels through inlets and outlets on chips, employing mechanisms like hydrodynamics [
6], [
92], [
110], electrokinetics [
61], [
110], and capillary action [
111], [
112], [
113]. Consequently, pressure-driven flows (PDF), electroosmotic flows (EOF), and capillary filling flows (CFF) become the primary methods used in order to generate nanofluidic flows in nanofluidic devices. Correspondingly, the velocities and directions of these flows can be controlled by regulating the electric potential difference (i.e., voltage) [
114], [
115], [
116], [
117], [
118], pressure difference [
62], [
117], [
61], and surface wettability of the nanochannels [
119], [
120], [
121], [
122], [
123], [
124], [
125].
Because of the planar, transparent, and mechanically robust nature of nanofluidic devices, the behavior and intricacies of nanofluidic flows and transport phenomena within nanochannels can be directly observed and measured using various optical microscopy systems. Additionally, some transport phenomena and effects that are challenging to observe directly with current microscopy technologies can be indirectly characterized by the real-time monitoring of certain changes in the physical or chemical properties between the inlets and outlets on chips. These include parameters such as the ionic conductivity [
76], [
126], [
127], electrical resistance [
128], [
129], [
130], [
76], tunneling current [
43], [
130], [
131], [
132], [
133], fluidic pressure drop [
134], [
135], [
136], and concentration [
137], [
138]. Accordingly, research on nanofluidic devices has significantly advanced the understanding of nanoscale transport, including the discovery of new phenomena and unusual effects.
While the current understanding of nanoscale transport has revealed unique mechanisms for transporting the mass of ions and small molecules at the nanoscale, as extensively reported in many excellent reviews [
24], [
44], [
139], [
140], [
141], [
142], [
143], [
144], there is now a growing trend in the use of nanofluidic devices for manipulating ultrasmall nanoscale fluids and nanometric objects. In this article, we have defined this emerging mechanism as “nanofluidic manipulation.” We propose that nanofluidic manipulation offers broad and potent approaches for handling nanoscale matter, whether non-biological or biological, at the single nanometric entity or single molecule level, and is applicable across a wide range of disciplines (
Fig. 1).
The rationale for this is threefold. First, the dimensions of the nanochannels are comparable to the length scales of many non-biological and biological nanometric objects of interest, providing ideal nanospaces for confining individual entities and molecules, thereby facilitating their manipulation. Secondly, nanochannels can effectively limit Brownian motion, which is a significant challenge in manipulating single nanometric objects in solution. Third, the unique physicochemical properties of nanochannels, such as extremely high surface-area-to-volume ratios, very short diffusion lengths, and minimal heat transfer distances, pave the way for envisioning future manipulation mechanisms and techniques with unparalleled performances that are currently beyond our reach.
In recent years, the nascent field of nanofluidic manipulation has seen several preliminary explorations, laying the groundwork for various potential approaches to manipulate ultrasmall nanoscale fluids and single nanometric objects, as illustrated in
Fig. 1. These exploratory directions encompass a range of topics: the manipulation of sub-picoliter—that is less than 1 pL—fluids; the formation and operation of nanoscale interfaces; the handling of single one-dimensional (1D) nanometric objects, such as DNA; the manipulation of single quasi-zero-dimensional nanometric objects, including non-biological nanoparticles, such as nanoparticles, biological nanoparticles, such as assemblies of biomacromolecules like bacteria, vesicles, viruses, and individual biomacromolecules like proteins; and individual small molecules. These directions are categorized based on the dimensions and sizes of the individual nanometric objects, highlighting the degree of difficulty in nanofluidic manipulation. As the dimensions and sizes of nanometric objects decrease, the challenges in manipulating them increase significantly because of the heightened impact of Brownian motion and the need for highly precise control. This article aims to delve into these emerging directions of nanofluidic manipulation, focusing on specific developments and trends, rather than offering an exhaustive summary of the entire field. We present the current progress in this domain, highlighting notable studies that align with the aforementioned directions. This article also discusses the challenges encountered in these studies and considers future opportunities for advancing the field of nanofluidic manipulation.
2. Basic nanofluidic manipulation
Current technical challenges in nanofluidic manipulation revolve around achieving precise control and manipulation of fluids within nanoscale channels, often at volumes as small as sub-pLs, confined within the closed spaces of nanoscale dimensions. A solid understanding of fundamental nanofluidic manipulation techniques is crucial because of the unique challenges presented by ultrasmall volumes and confined spaces in these processes. Therefore, it is imperative to establish methodologies specifically tailored to sub-pL fluids and nanoscale interfaces in order to address this critical challenge. This innovation in nanofluidic manipulation not only opens doors to precision in handling sub-pLs, but also contributes to transformative innovations in scientific, engineering, and industrial applications.
2.1. Sub-pL fluids
The ability to control fluids, whether at the macroscopic or microscopic level, has had a profound impact in various fields, including biotechnology, chemistry, and materials science [
5], [
86], [
144], [
1], [
57], [
145]. Despite the development of sophisticated methodologies for controlling small amounts of fluids, regulating fluids at sub-pL scales remains a significant challenge due to the ultrasmall size (10-1000 nm scale) of nanochannels. Currently, some establishments have emerged that aim to control fluids at the sub-pL level by introducing fluidic components such as valves within tiny nanofluidic channels [
56], [
86], [
1], [
57], [
145], [
146], [
59]. One of the examples to control fluids at the sub-pL level by Kazoe et al. [
146] introduced a glass nanovalve where nanoscale glass deformation is induced by an applied force. The mechanism was simple and valve functions were successfully verified.
Furukawa et al. [
145] proposed a novel nanovalve utilizing nanobubbles and light, known as a nanobubble valve. A design of a nanofluidic device with a cross-sectional view of a nanochannel connected to a chromium pad is proposed. The nanobubble valve mechanism involves heat generation and a phase transition (evaporation) induced by the adsorption of a laser beam [
145], [
147].
The application of MHz-order vibrations has been extended to nanoscale manipulations. A novel surface acoustic wave (SAW)-induced pumping method has been reported for manipulating nanoslit channel-confined fluids, suspended nanoparticles and molecules [
148]. Recently, Zhang et al. [
149] demonstrated active 200 femtoliter (1 fL = 10
−15 L) fluid droplet manipulation using MHz-order SAWs within high-aspect ratio nanoslit channels, as shown in
Fig. 2(a). This method involves forming traps as locally widened regions along the channel, enabling propulsion, splitting, mixing, and merging of individual fluid droplets [
149]. Deng et al. [
147] presented a new device with an electrokinetic trapping (EKT)-based nanofluidic preconcentrator integrated with oscillating (acoustic wave) bubble valves to trap concentrated antigens and immunobeads during an immunoassay.
Another innovative approach involves a well-tailored thermoresponsive polymer that can achieve transition between hydrated and dehydrated states, resulting in swollen and shrunken states, respectively. Xu et al. [
57] developed an active nanovalve in which a thermoresponsive material was applied to fL nanofluidic channels, and the change in the polymer brush length served as a soft matter-regulated active nanovalve. The valving performance was evaluated in order to demonstrate both closed and open states, as shown in
Fig. 2(b).
2.2. Nanoscale interfaces
This section explores the intricate realm of nanofluidic manipulation, specifically focusing on the utilization of nanoscale interfaces to control the fluid at nanoscale. Controlling multiphase fluids at immiscible liquid-liquid or gas-liquid interfaces is crucial for various chemical operations, including reactions, extraction, purification, and pre-concentration [
106], [
122], [
150], [
151], [
152]. Owing to their importance, we thereby delve into the current progress and challenges that lie ahead in the realm of nanofluidic manipulation, with a specific focus on nanoscale interfaces, particularly in the context of gas-liquid and liquid-liquid interfaces.
Previously, the formation of attoliter (1 aL)-fL volume droplets in nano/microchannels has proven valuable for synthesizing functional nanoparticles and facilitating drug delivery [
106], [
153]. The generation of ultrasmall, monodispersed droplets with defined diameters through flow focusing, facilitated by continuous phase and interface instabilities, has been successful. However, achieving parallel multiphase flows poses challenges in balancing the fluid and Laplace pressures derived from surface tension. In nanospaces where the Laplace pressure can reach 100-1000 kPa, multiphase fluid manipulations are particularly difficult. Kazoe et al. [
66] addressed this challenge by developing a method to form aqueous-organic parallel two-phase flows in a nanochannel. This relies on nanoscale partial surface modification, utilizing a partially hydrophobic nanochannel to stabilize a parallel two-phase flow with a high Laplace pressure. Using this method, a water-dodecane parallel two-phase flow was successfully formed on the partial surface of nanochannels modified with octadecyltrichlorosilane (ODS) [
66], [
106], [
154].
Droplets or plug flows, instrumental in volume control and compartmentalization for highly efficient reactions, involve a delicate interplay between viscous forces, surface tension, fluid pressure, and Laplace pressure at the interfaces [
59], [
60], [
150], [
155], [
156], [
157], [
158]. Takagi et al. [
159] developed nanofluidic analytical pretreatment methods that exploit nanochannels to downsize the chemical unit operations to fL-pL volumes. In
Fig. 3(a), a schematic of the fL-droplet shooter is presented, showing time-series images of a droplet shooting at a frequency of 13 Hz. The device configuration comprises a nanochannel that delivers a liquid sample along with the nanochannels, facilitating the initial gas-phase flow focus and resulting in the formation of a gas-liquid-gas parallel flow [
159].
An intriguing application of nanoscale interfaces in nanochannels involves the manipulation of gas-liquid interfaces (GLIs). These unique interfaces serve as distinct regions allowing molecules to transition from one state to another and have recently been downscaled to the nanoscale through the utilization of nanofluidics [
122], [
89]. Kawagishi et al. [
122] fabricated controllable nanoscale GLIs in fL-order nanofluidic channels by combining precise local surface control, tailored physicochemical effects, and optimized nanofluidic operation. This approach involves the formation of nanoscale GLIs induced by local phase changes in hydrophilic/hydrophobic nanopatterned nanofluidic channels, resulting in uniform, stable, and arrayable nanoscale GLIs.
Fig. 3(b) illustrates a conceptual schematic of the fabrication of nanoscale GLIs in hydrophilic/hydrophobic nanopatterns.
Exploring the fundamentals of nanofluidic manipulation serves as a cornerstone for the advancement in this evolving field. The significance of refining nanofluidic manipulation techniques lies in the precision required to handle nanometric objects, unlocking progress across diverse domains including chemistry, biology, chemical and biomedical engineering, materials, mechanical engineering, and various industrial applications. However, achieving accuracy, precision, and high throughput in manipulation tasks, whether on the scale of single nanometric entities or molecules, is a formidable challenge because of the extremely small dimensions involved. Addressing this critical challenge is imperative in order to realize the full potential of nanofluidic manipulation and its transformative impact on scientific, engineering, and industrial applications. Therefore, the manipulation of single 1D nanometric objects—objects with 1D in the nanometer-sized range, such as single DNA strands—and single quasi-zero-dimensional nanometric objects, or objects with all three axis dimensions in the nanometer-sized range, such as single viruses or liposomes, is of paramount importance. The ability to control these single 1D or quasi-zero-dimensional nanometric objects not only opens avenues for innovative breakthroughs, but also propels advancements across diverse scientific and technological domains. This positions nanofluidic manipulation as a key driver of progress in diverse fields, thereby shaping the landscape of future advancements.
3. Single 1D nanometric objects
Nanochannels have emerged as a promising tool for manipulating single 1D nanometric objects, such as DNA molecules, presenting advantages over traditional techniques, such as optical and magnetic tweezers, which require tethering to solid surfaces or beads [
6], [
47], [
110], [
160]. Although the nanofluidic manipulation of single DNA molecules has opened up exciting avenues for research and applications, several challenges remain. These include improving throughput, scalability, and user-friendliness while minimizing the risk of DNA damage during manipulation. Nanochannels facilitate extensive confinement and linear stretching of single long DNA chains without the requirement for tethering [
47], [
110], [
160], [
161]. This not only enables direct visualization, probing, mapping, and barcoding, but also holds potential for future DNA sequencing across the entire length scale of the DNA molecule [
47], [
132], [
160].
Various techniques have been employed to manipulate single DNA molecules [
162], [
163], [
136], [
164], [
165], [
166], [
167], [
168], [
169]. A notable example is the knot factory that guides the production of DNA knots and serves as a model for knot formation [
136], [
162], [
164], [
167]. Amin et al. [
136] introduced an approach with a knot factory-on-a-chip, which leverages a low-Reynolds-number flow to compress single DNA molecules against the slit barriers in the nanochannels, as shown in
Fig. 4(a). Initially, the DNA chain extends, and upon compression, the flow is released, allowing the DNA molecules to relax and reveal knots as sharply localized regions of high intensity along the extended molecule. This innovative system not only efficiently forms and detects knots, but also aids in developing models to quantify conditions that favor knot production. Furthermore, the speed of DNA translocation can be reduced by manipulating the charge distribution, thereby enhancing the interactions between the DNA and the nanochannel wall. For instance, Liu et al. [
170] demonstrated that electrically tuning the wall surface charge using an external gate bias significantly slowed DNA translocation through a nanochannel. Their experiments utilized a nanofluidic field-effect transistor with an integrated gate electrode, enabling the electrical modulation of the surface charge of the dielectric alumina wall. A positive gate bias reduces the DNA translocation speed by one order of magnitude because of the enhanced electrostatic attraction between DNA and the capillary surface. This device has great potential for use in high-resolution DNA sequencing. Another important technique involves the use of nanoslits with tunable control over confinement, enabling the study of DNA translocation through these tightly confined spaces [
161]. Furthermore, Robin et al. [
171], [
172] reported pioneering experiments demonstrating the application of these nanoslits in neuromorphic computing. By harnessing the unique properties of confined DNA translocations, their work highlighted the potential of nanofluidic manipulation in the development of advanced computing systems that mimic neural architectures.
In a recent study, Yu et al. [
169] presented a strategy for single-stranded DNA manipulation based on dynamic and tunable confinement, as can be seen in
Fig. 4(b). By leveraging the control of a pneumatic microvalve and deformation of nanoslits made of an elastomeric material, they dynamically formed and continuously adjusted a uniform triangular nanochannel array. These size-tunable nanochannels, effective down to 20 nm in size and up to submillimeters in length, are well suited for linearizing megabase pair (Mbp)-long DNA molecules. Ström et al. [
166] demonstrated DNA sorting with a throughput of one to five orders of magnitude greater than other microfluidic techniques. Their device, with a small footprint (23 mm × 0.5 mm) and micrometer-range feature sizes, was considerably easier to fabricate than parallelized nano-array-based approaches, as shown in
Fig. 4(c). High-throughput long DNA sorting in a microscale deterministic lateral displacement (DLD) array at ultra-high flow velocity is presented, showing sorting of 166 kilobase pair (kbp) (green) and 48.5 kbp (red).
Understanding how DNA interacts with proteins is crucial for elucidating biological processes and this has led to the use of nanofluidic devices in single-DNA studies [
43], [
47], [
160], [
163], [
165], [
168], [
173]. Öz et al. [
174] demonstrated the use of a nanochannel enabling single-molecule studies of DNA-protein interactions without anchoring the molecule to a foreign object. In a more recent study, Sharma et al. [
163] manipulated DNA-analyte interactions by inventing a nanofluidic device with a reaction chamber in the middle of every nanochannel, as illustrated in
Fig. 4(d). These reaction chambers, which are larger than the nanochannels entering and exiting them, allow single DNA molecules to be entropically confined within each chamber [
163].
4. Single quasi-zero-dimensional nanometric objects
The precise and accurate control of single quasi-zero-dimensional nanometric objects holds significant promise in both the non-biological and biological realms. However, challenges have emerged owing to the unfavorable scaling of most interaction mechanisms at small scales. By navigating through the intricacies of single quasi-zero-dimensional nanometric objects, preliminary research has sought to unlock pioneering opportunities to address critical challenges in transporting, trapping, capturing, and manipulating these entities, which is an essential step toward fully realizing their potential.
This section delves into the current progress in nanofluidic manipulation of single non-biological and biological entities, shedding light on the challenges and innovative approaches that shape this dynamic field. Achieving precise and accurate control of these single quasi-zero-dimensional nanometric objects, whether non-biological or biological entities, will pave the way for broader applications in areas such as electronics, catalysis, material design, sensing, drug discovery, diagnostics, and targeted therapies. This advancement has pushed the boundaries of what is achievable at the nanoscale level, showcasing the transformative potential of nanofluidic manipulations in diverse technological domains.
4.1. Non-biological
This section explores the current progress in the nanofluidic manipulation of single non-biological nanoparticles, emphasizing the challenges and innovative approaches that shape this dynamic field. Controlling the motion and determining the properties of nanoscopic objects in liquids are crucial. However, challenges arise because of the unfavorable scaling of most interaction mechanisms at small-length scales. The control and transport of nanoscale objects in fluids present unique difficulties, necessitating innovative approaches to overcome the inherent scaling challenges.
A diverse array of approaches have been developed in order to address the elusive manipulation and characterization of single nanoscale objects. Non-biological nanoparticles, including polymeric nanoparticles (e.g., polystyrene and silicon) [
175], [
176], [
177], quantum dots [
178], [
179], and metallic nanoparticles (e.g., Au and platinum) [
180], [
181], [
182], [
183], possess unique properties that render them invaluable for various technological applications. Their distinctive characteristics make them central to the ongoing progress in nanofluidic manipulation, opening new possibilities in electronics, catalysis, and material design. Nanofluidic trapping has emerged as a prominent technique that offers precise control over the trapping strength and manipulation of particle positions. Various trapping methodologies have been introduced to achieve sustained particle confinement under ambient conditions for prolonged durations, such as anti-Brownian electrokinetics (ABEL) [
184], convex-lens-induced confinement (CLIC) [
185], [
186], and geometry-induced electrostatic (GIE) [
187] trapping. Recently, a notable example of nanofluidic trapping was reported by Levin et al. [
181], [
182], who demonstrated the trapping of shape-selected colloidal Au nanocrystals in an array of parallel nanofluidic channels, as shown in
Fig. 5(a). This platform ensures identical reaction conditions for each particle, enabling parallel quantitative activity monitoring (
Fig. 5(a)). This platform allows identical reaction conditions for each particle, enabling parallel quantitative activity monitoring [
181], [
182]. This approach highlights the potential of nanofluidic systems to facilitate highly controlled and reproducible experiments essential for understanding the nuanced behavior of nanoparticles under specific conditions. Altenburger et al. [
183] introduced nanofluidic reactors designed for single-particle catalysis studies by integrating plasmonic imaging and spectroscopy. Their innovative approach ensured uniform reaction conditions for individual particles, allowing for highly parallelized studies involving tens of single nanoparticles within the same experimental setup. This methodology not only enhances the efficiency of catalysis research, but also provides deeper insights into the catalytic behaviors of individual nanoparticles, which are often masked in bulk studies.
More recently, Hӧller et al. [
176] presented transporting individual nano-objects involves guiding them through a nanochannel network, confining them in electrokinetic nanovalves, as illustrated in
Fig. 5(b). This collaborative effect combines an applied alternating current (AC) electric field and rationally engineered nanotopography, enabling the precise control and tracking of fluorescent nano-objects. Their work was extended to sub-100 nm conjugated polymer nanoparticles [
176]. Moreover, Eberle et al. [
179] introduced switchable electrokinetic nanovalving for guiding, confining, releasing, and sorting nano-objects. This innovative technique allows on-demand motion control of various particles, including single quantum dots, even in high-ionic-strength environments. Previously, Magnasco [
188] and Prost et al. [
189] proposed that such particle transport could be achieved with artificial Brownian motors (BMs) based on an asymmetric energy landscape and non-equilibrium fluctuations. Skaug et al. [
190] employed a rocking Brownian motor with a diameter of 60 nm for the directed transport of Au spheres. They shaped a nanofluidic geometry slit to create asymmetric potentials and achieved directed transport through an oscillating electric field. Fernandez-Cuesta et al. [
178] integrated a 35 nm gap plasmonic bowtie antenna with a 30 nm × 30 nm nanochannel. The channel ran through the antenna gap and delivered the analyte directly to the hotspot. They demonstrated how the antenna probes into zeptoliter (1 zL = 10
−21 L) volumes inside a nanochannel. The nanochannel was integrated with a plasmonic bowtie nanoantenna (
Fig. 5(c)); the nanochannel had dimensions similar to the antenna gap and ran perfectly aligned and leveled with it.
4.2. Biological
The manipulation and examination of individual biological entities represents a transformative capability with significant implications for advancing our understanding of fundamental biological processes [
1], [
6], [
11], [
91], [
110]. At its core, nanofluidic manipulation involves the precise control of these entities in dimensions of the order of nanometers. The term “single biological entities” encompasses both assemblies of biomacromolecules and individual biomacromolecules. Assemblies of biomacromolecules, such as bacteria [
191], [
192], viruses [
193], [
194], [
195], [
196], [
197], [
198], [
199], liposomes [
107], [
198], [
199], [
200], [
201], [
202], and extracellular vesicles (EVs) [
71], [
203], [
204], [
205], [
206], [
207], [
208], [
209], play critical roles in cellular functions. Understanding their dynamics at the nanoscale level is pivotal for advancing our knowledge of biological processes. Similarly, individual biomacromolecules such as proteins [
110], [
210], [
211] exhibit distinct behaviors and functions that contribute to the overall functionality of living systems. Therefore, the manipulation and examination of these biological entities offer insights into their behaviors, interactions, and potential applications across diverse fields. Despite their immense potential, challenges persist in the field of the nanofluidic manipulation of single biological entities. Issues such as transporting, trapping, and capturing, particularly for biological entities within the 1-1000 nm range, are among the current challenges. Recent advancements in nanofluidic manipulation techniques, highlighted in the literature, showcase progress in addressing these challenges. Exploring the assemblies of biomacromolecules and individual biomacromolecules at the single nanometric object level contributes to our understanding of cellular mechanisms and it also holds promise for applications in fields such as drug discovery, diagnostics, and targeted therapies. Despite these challenges, advancements in nanofluidic manipulation techniques continue to drive the progress in this area of research.
4.2.1. Assemblies of biomacromolecules
Single bacteria, such as
Escherichia coli (
E. coli), with dimensions in the micrometer range, require nanofluidic structures at the hundred-nanometer scale for manipulation [
191], [
192]. Hong et al. [
192] introduced a nanofluidic device capable of actively loading more than 200 single bacterial cells into a nanochannel array, as shown in
Fig. 6(a). This setup not only quantifies antibiotic permeability but also serves as a transport pathway for cell-cell signaling [
192]. Kk et al. [
191] developed a nanofluidic device for optical DNA mapping of ten different samples simultaneously by automating the image acquisition process.
Single viruses falling within tens to hundreds of nanometers have also been successfully manipulated using nanofluidic devices [
110], [
193], [
195], [
199]. Some of the most important studies were on viral size using a nanofluidic manipulation platform with other optical [
193], [
198], [
199] and electrical [
110], [
195], [
197], [
212] detection, in which a voltage readout was used to detect a single virus when it passed through a nanochannel. For example, Zhou et al. [
195] and Zhang et al. [
212] have been continuously developing a sophisticated approach using electrical detection called repulsive pulse sensing, which detects the change in current caused by a viral capsid entering from the wide region of a nanochannel to the narrow region of the channel. By increasing the number of narrow regions to obtain multiple signals per capsid, the device improved size detection precision. In addition, nanostructures that were close to the capsid size could distinguish the size difference of a single unit of capsid protein [
195], [
197], [
212]. More recently, Eberle et al. [
179] demonstrated a switchable electrokinetic nanovalving able to confine and guide single nanoobjects, including macromolecules, with sizes down to around 10 nm, in a device. The nanovalves are based on spatiotemporal tailoring of the potential energy landscape of nanoobjects using an electric field modulated collaboratively by wall topography and embedded electrodes in a nanochannel system. They combined nanovalves to isolate single entities from an ensemble, and demonstrated their guidance, confinement, release, and sorting. They demonstrated on-demand motion control of single adenoviruses and another biomolecule suspended in electrolytes with a broad range of ionic strengths up to the biological level, as shown in
Fig. 6(b) [
179].
Single liposomes, spherical vesicles composed of lipid bilayers, are crucial in various scientific disciplines. Nanofluidic manipulation allows for the controlled transport and manipulation of fluids at the nanoscale, offering a powerful toolkit for investigating and exploiting liposome properties [
110], [
201], [
202], [
204]. Recent progress, highlighted by Marie et al. [
200], involved the use of nanofluidic traps as fL reactors to study the fusion of a few hundred liposomes (
Fig. 6(c)). The trap works by combining diffusioosmotic flow in a funnel-shaped nanochannel with diffusiophoretic particle transport. Two liposome samples with complementary DNA probes labelled with either Förster resonance energy transfer (FRET) donors or acceptors were introduced sequentially into the nanochannels and trapped [
200].
For single EV manipulation, nanofluidic devices can provide nanostructures that can be used to not only isolate EVs secreted from cells but also effectively localize EVs [
71], [
205], [
206], [
207], [
208], [
209]. EVs are typically tens to hundreds of nanometers in size. Therefore, nanostructures are also designed to have at least one dimension, similar to that of EVs, for effective manipulation. Owing to the high heterogeneity of EVs (size and functional differences) [
71], [
205], [
206], [
207], [
208], [
209], [
213], it is important to characterize this heterogeneity to understand their heterogeneity. For example, size sorting of EVs obtained from human epithelial breast cancer cells was performed using a nanofluidic device with nanopillars. This nanofluidic device sorted EOF-driven vesicles using nanopillars, which induced the displacement of EVs in the direction perpendicular to the EOF based on EVs size [
209]. Another example used a specific nanostructure called “attoliter-in-femto-Array (aifA)” for isolating the single EVs directly at normal concentration [
206]. More recently, Hong et al. [
71] addressed these challenges by using geometry-induced electrohydrodynamic tweezers (GET). GET allows massive parallel trapping of single nanosized objects such as EVs within seconds near plasmonic hotspots without causing photothermal damage. The scalability of GET, coupled with its ability to provide instantaneous plasmonic trapping with single-particle resolution, paves the way for high-throughput plasmon-enhanced single-particle spectroscopy.
4.2.2. Individual biomacromolecules
Single proteins can also be manipulated using nanofluidic manipulation techniques, providing an opportunity to explore their structural and functional characteristics in unparalleled detail [
20], [
102], [
175], [
210], [
211]. Recently, Špačková et al. [
214] presented nanofluidic scattering microscopy (NSM), allowing real-time label-free imaging of single protein in solution. This approach enables the simultaneous determination of molecular weight and hydrodynamic radius without surface immobilization. The nanofluidic devices used in the experiments featured a series of nanochannels with tailored cross-sectional dimensions (
Fig. 7(a)) connected to the macroscopic inlets and outlets through two microchannels with cross-sectional dimensions of 50 μm × 1.5 μm. Liquid sample transport and imaging rely on pressurizing the inlets, controlling diffusion during measurements, and meticulous data collection [
214].
Further advancements in this field have been demonstrated by Gordon [
215], and Mathew and Gordon [
216], who utilized nanoaperture optical tweezers for the manipulation, sensing, and spectroscopy of single proteins. Their innovative use of a double nanohole in a metal film (
Fig. 7(b)) enabled the trapping of single proteins with low laser power, highlighting the potential of this technique for studying dynamic biological processes [
215]. In their study, Hong et al. [
175] introduced opto-thermo-electrohydrodynamic tweezers capable of trapping and dynamically manipulating nanometer-scale objects several micrometers away from a high-intensity laser focus. This novel approach expands the possibilities for noninvasive manipulation of proteins, facilitating detailed studies of their behavior and interactions.
The electrical properties of individual proteins have emerged as a critical area of research in bioelectronics. Electron tunneling and quantum mechanical tunneling (QMT) probes are powerful tools for this purpose. Jiang et al. [
217] developed a generalizable and straightforward protocol for fabricating electron-tunneling probes with a gap of less than 5 nm, suitable for measuring the conductance of single proteins with enhanced stability and precision. These probes enable detailed investigation of the electrical properties of proteins, which is essential for understanding their roles in biological systems and their potential applications in nanotechnology.
Techniques such as nanocavity diffusional sizing (NDS) provide opportunities to manipulate single protein assemblies in solution without the need for surface immobilization for analysis and detection. Vanderpoorten et al. [
218] fabricated nanofluidic devices with nanocavity confinement functionalities, enabling single-molecule studies on prolonged observational timescales. Jacquat et al. [
219] introduced the NDS approach, which extracts nanoparticle sizes from measurements of particle residence times within fluidic nanocavities using single-molecule confocal microscopy. The main concept and experimental realization of the NDS approach involve placing the observation volume of a confocal microscope within the trapping cavity of a nanofluidic device (
Fig. 7(c)) filled with an aqueous solution containing single proteins of interest. Extracting the sizes of single proteins involves recording the time trajectories of the single proteins diffusing into and out of the observation volume, with residence time distributions providing insights into the hydrodynamic radius.
In the quest for stable trapping methods, Svirelis et al. [
211] presented a concept that utilizes nanoscale chambers with macromolecular gates in solid-state materials. Responsive polymer brushes with gates that can be collapsed on demand by electrical control over the local temperature enable the noninvasive and tether-free trapping of proteins. This innovative technology allows the confinement of a large number of proteins inside nanoscale chambers, reaching concentrations up to 60 g∙L
−1. The long trapping time (at least 1 h) under physiological conditions, rapid liquid exchange, and ligand access rendered this platform highly versatile. More recently, Yang et al. [
210] introduced a nanofluidic aptamer nanoarray (NANa) that enabled the immunologically specific capture of single proteins at normal concentrations (
Fig. 7(d)). This method combines nano-in-nano integration [
63], [
220] and aptamer technologies in square nanochannels, facilitating the stochastic capture of single proteins, such as platelet-derived growth factor composed of two B subunits (PDGF-BB), at normal concentrations, following a Poisson distribution [
210].
5. Single small molecule
Single-molecule studies have significantly reshaped our understanding of molecular behavior, providing insights beyond the reach of traditional ensemble techniques. The emergence of nanofluidic manipulation has ushered in an era of unparalleled control over individual molecules, facilitating meticulous manipulation and analysis at the nanoscale. However, this precise control encounters challenges when dealing with ultrasmall single molecules (less than 10 nm), which pose difficulties in regard to manipulation owing to their diminutive dimensions. Achieving sensitivity and resolution capable of handling molecules of diverse sizes and properties is imperative for precise and accurate manipulation. In addition, maintaining the molecular stability during nanofluidic manipulation is a critical concern, necessitating strategies to counter thermal diffusion in solution, commonly known as Brownian motion [
11], [
1], [
89], [
175], [
221]. Despite these challenges, nanofluidic manipulation of single molecules is at the forefront of scientific discovery, thereby providing unprecedented insights into molecular dynamics. To date, the field demands ongoing collaboration and innovation to overcome existing challenges. One of the few achievements in the manipulation of single-molecule solutions is highlighted. Lesoine et al. [
221] demonstrated this remarkable feature by utilizing a 600 nm × 400 nm cross-sectional channel. The intentional design with a cross section smaller than the confocal diameter enabled the detection-limited 1D motion of single molecules within all-silica nanochannels, each with a cross-sectional diameter of less than 100 nm.
Recently, Ghosh et al. [
222] demonstrated the use of all-silica nanochannels for multiplexed dynamic single-molecule detection of sub-3 nm-sized objects diffusing within a nanometric-confined volume (
Fig. 8(a)). Their innovative approach of utilizing an all-silica nanofluidic environment facilitated the electrokinetic handling of individual molecules while resolving molecular shot noise. This high-throughput nanochannel fabrication, combined with two-focus fluorescence correlation spectroscopy (2fFCS) for 1D confined detection of fast-moving single organic molecules (such as single Alexa Fluor 647 molecules), provides a fundamental understanding of molecular shot noise. This study lays the groundwork for experiments that require the precise physical manipulation of molecular dynamics, enhancing our ability to study fast molecular interactions and behaviors [
222].
In another significant advancement, Zhan et al. [
223] achieved optical trapping and release of single molecules (∼2 nm, such as single OPE3-SMe) in a solution using two coupled plasmonic nanotips controlled by a mechanically controllable break-junction setup (
Fig. 8(b)). This setup leverages a strong electric field in the gap under illumination, thereby enabling the trapping of individual molecules. The tunneling current through the nanogap distinguishes the trapping states of the single target molecule. The surface plasmon-based trapping effect demonstrated by this setup is universal for other molecules, suggesting broad applications for selective single-molecule trapping and manipulation in various chemical and biological contexts.
Kawagishi et al. [
95] introduced a flexible glass-based hybrid nanofluidic device designed to precisely control the blocking, opening, and directional flow of individual small molecules (such as a single Cy3 molecule) within nanochannels (
Fig. 8(c)). This innovative device allows for the real-time tracking of regulated single small molecules in nanofluidic environments, providing valuable insights into the dynamic behaviors of single small molecules under diverse nanofluidic conditions with varying spatial restrictions. This development enhances our understanding of nanoscale molecular dynamics, offering potential applications in molecular diagnostics and nanoscale fluidic control.
6. Challenges and future perspectives
These studies have marked early strides in nanofluidic manipulation, focusing on regulating, valving, interfacing, controlling, trapping, capturing, sorting, and conducting various processes at the nanoscale. These processes involve single DNA molecules, nanoparticles, vesicles, viruses, proteins, and even small molecules. Despite the emerging progress, the field holds opportunities for innovation and enhancement of these capabilities. Addressing several critical issues is essential. First, confining single small molecules requires more precise and accurate nanostructure fabrication owing to their ultrasmall dimensions. Future advancements in nanostructures and nanogeometries are pivotal to meet this requirement. Second, real-time and high-speed manipulation requires detection systems with high spatiotemporal resolution for the immediate and precise handling of entities. Third, achieving high throughput in nanofluidic manipulation requires innovative approaches to device design and integration with automated systems. Finally, further coupling of nanofluidic devices with other techniques, such as optical, magnetic, acoustic, and thermal fields, is crucial for the more efficient manipulation of single nanoscale objects.
6.1. Precision and accuracy
Nanofluidic manipulation stands at the forefront of scientific innovation, addressing the urgent demand for precision and accuracy in handling fluids at the nanoscale and manipulating single nanometric objects. The intricate dimensions of nanoscale objects pose distinct challenges, prompting the need for innovative strategies to achieve meticulous control and manipulation. The current state of nanofluidic manipulation reflects progress in nanostructure design, fabrication techniques, and fluidic control handling. However, its capabilities in terms of precision and accuracy remain challenging.
To unleash the full potential of nanofluidic manipulation, a paradigm shift towards an elevated level of precision and accuracy is imperative. ① Advancements in nanostructure design, fabrication methodologies, and fluidic control technologies must strive for a level of precision and accuracy that transcends the existing limitations, enabling the handling of single 1D and quasi-zero-dimensional nanometric objects with unparalleled precision and accuracy. Envisioning nanofluidic devices with meticulously designed geometries that perfectly accommodate single nanometric objects opens new avenues for precise transportation, separation, isolation, trapping, and organization. In the future advancements in nanostructure design and fabrication, exploring technologies like nano-in-nano integration [
63], [
220] and angstrom-scale fabrication beckon [
15], [
68]. These pursuits focus on refining and innovating nanostructures, paving the way for unprecedented precision and accuracy in manipulation. In addition to refining and innovating nanostructures, manipulating individual nanometric objects in nanofluidic channels through chemical approaches holds significant potential [
139], [
224], [
225], [
226]. The surface charge can be controlled using various methods, including chemical surface modification and application of external electric fields. By altering the surface charge in designated areas of a nanochannel, the surface charge density can be modulated, thereby influencing the interaction between the nanochannel surface and single nanometric objects. This capability is fundamental for controlling single nanometric objects and enabling functions such as valving and capturing. By carefully designing the charge distribution, precise control over single nanometric objects, which is essential for effective nanofluidic manipulation, can be achieved. ② Innovative nanochannel technologies play a crucial role in overcoming the challenges posed by ultrasmall dimensions. Surface modification [
1], [
11], [
20], [
154] and site-specific nanopatterning [
1], [
6], [
69], [
72] have emerged as innovative approaches for nanochannel engineering. These technologies not only enhance fluidic control handling, but also enable improved manipulation, thereby contributing to the quest for enhanced precision and accuracy. Achieving elevated levels of precision and accuracy in nanofluidic manipulation has immense potential for transformative impacts. The pursuit of increased precision and accuracy is pivotal in regard to unlocking capabilities across scientific, medical, and technological applications.
6.2. Real-time and high-speed
The imperative for real-time and high-speed nanofluidic manipulation arises from the need to achieve instantaneous and precise control over single nanometric objects or molecules, thereby opening up avenues for rapid responses, dynamic interventions, and immediate observations in scientific, medical, and technological applications. Despite significant advances in technology, the current status faces challenges in achieving optimal efficiency, particularly for high-speed and sensitive cameras operating at the speeds necessary for real-time interventions.
In order to fully harness the potential for real-time and high-speed nanofluidic manipulation, thereby ① A high speed and sensitive camera is necessary. With the capability of the camera, the timescales of single nanometric object events were investigated, enabling real-time observation and control of dynamic processes at unprecedented speeds. ② Integrating the new high-speed and sensitive camera into super-resolution microscopy systems, such as stochastic optical reconstruction microscopy (STORM) [
212], [
213], is pivotal for achieving high temporal spatial resolution. This integration enhances the precision of real-time high-speed observations and provides detailed insights into nanoscale events. Achieving this requires advancements in real-time and high-speed microscopy to handle the rapid timescales associated with single nanometric object events, thereby providing detailed insights into the dynamic processes at the nanoscale in real time.
6.3. High throughput
Manipulation of single nanoscale objects with high throughput is a significant challenge in nanofluidic manipulation, demanding innovative device design and automated systems. Collaboration between engineers and nanofluidic techniques is pivotal for the development of scalable systems capable of simultaneously handling multiple large-scale nanometric objects. The current status of high-throughput nanofluidic manipulation has led to progress in device design and automation. However, challenges persist in the parallelization of the analysis of multiple nanometric objects. The need for seamless, simultaneous manipulation and analysis of nanometric objects necessitates a paradigm shift in current methodologies. ① Future prospects include parallel analysis techniques that enable the simultaneous manipulation and analysis of multiple single-nanometric objects. This approach is crucial for conducting efficient large-scale studies. ② The integration of automated systems is essential for achieving a high throughput. Automation streamlines processes, reducing the time and labor involved in manipulating and analyzing single nanometric objects. ③ The integration of machine learning and data science enhances data-driven approaches and optimizes the outcomes during high-throughput nanofluidic manipulation.
Addressing the challenges of high-throughput manipulation and arraying of single nanometric objects is essential for unlocking transformative impacts in science and technology. Collaborative efforts between engineers and nanofluidic experts, coupled with advancements in automation and parallel analysis techniques, are the key to realizing the full potential of high-throughput nanofluidic manipulations.
6.4. Diverse coupling with external fields
Nanofluidic manipulation can benefit significantly from coupling with various external fields such as optical, magnetic, acoustic, and thermal fields. However, challenges remain in optimizing these interactions for diverse applications. In future, a deeper understanding of the synergies between nanofluidic devices and external fields will lead to the development of hybrid systems that exploit the unique advantages of each coupling method. Future prospects include, but are not limited to, ① improved abilities of nanofluidic devices to couple with optical, magnetic, acoustic, and thermal fields for precise, flexible and noninvasive manipulation of single nanometric objects; and ② hybrid coupling strategies that combine multiple external fields synergistically for unprecedented control over nanometric objects.
Addressing these challenges and embracing interdisciplinary collaboration will drive the future of nanofluidic manipulation. The convergence of expertise in physics, chemistry, materials science, and engineering is essential for overcoming hurdles and unlocking the full potential of nanofluidic manipulation for applications in medicine, materials synthesis, and other fields.
6.5. Innovating future
Continuous advancements in nanofluidic manipulation promise to reshape the landscape of science, engineering, and industry through pioneering innovations (
Fig. 9). This ongoing evolution not only deepens our understanding of these fields, but also contributes significantly to addressing global-scale challenges. Precision and accuracy in controlling single nanometric objects are key to revolutionizing processes such as sampling, analysis, and synthesis. In biology, manipulating individual nanometric objects, such as single DNA strands, unlocks possibilities in regard to DNA sequencing. Technological advancements driven by nanofluidic manipulation are equally significant. The real-time, high-speed manipulation of nanoscale objects introduces new dimensions to sensing and actuator technologies. These improvements are essential for developing advanced applications, such as molecular robotics and efficient batteries for electric vehicles. Along with high-throughput manipulation and diverse coupling with external fields, industries can witness significant improvements in molecular robotics through advanced nanofluidic manipulation techniques. Moreover, the integration of nanofluidic manipulation into computing and information processing promises faster and more energy-efficient technologies, promoting advancements in artificial intelligence and quantum computing.
In an industrial context, nanofluidic manipulation is poised to revolutionize manufacturing processes, enabling the production of nanoscale materials with enhanced properties. This fosters innovation in industry and contributes to sustainable practices. The societal impact of nanofluidic manipulation is profound, influencing aspects of human life ranging from healthcare to environmental sustainability. By facilitating breakthroughs in medical treatments, clean energy technologies, and efficient manufacturing, nanofluidic manipulation has the potential to address pressing global challenges and improve overall quality of life. In future, the interdisciplinary nature of nanofluidic manipulation will ensure its relevance in shaping a more technologically advanced, sustainable, and interconnected world.
As we move forward, manipulating single nanometric ions (less than ∼1 nm) within nanofluidic channels is crucial, but challenging owing to their extremely small dimensions. Addressing this challenge is essential for achieving the precise and accurate control of these single nanometric ions at the nanoscale within nanofluidic channels or devices. Additionally, the surface charge can be manipulated through various methods, including surface modification and the application of external fields. Single nanometric ions can be controlled by altering the surface charge density at a specific position within the nanochannels, thereby enabling functions such as transport, valving, and capture. By addressing the challenges associated with manipulating single nanometric ions and leveraging techniques to modulate the surface charge, significant progress can be made in nanofluidic technology, leading to breakthroughs in numerous scientific and industrial fields.
7. Conclusions
In summary, this article has focused on the burgeoning frontiers of nanofluidic manipulation, delving into specific developments and trends rather than providing an exhaustive field-wide summary. Our exploration has traversed the current progress in this domain, highlighting noteworthy works aligned with the identified directions. In addition, we have engaged in discussions surrounding the challenges inherent in these explorations, contemplating the future prospects that could propel nanofluidic manipulation to new heights. Recent advancements in the development of nanofluidic devices signal promising directions for the manipulation of single nanometric objects, both biological and non-biological, at the individual object level. Nanofluidic devices demonstrate the ability to exert precise control over nanoscale structures, enabling a range of functions, including regulation, valving, interfacing, processing, trapping, and capturing. Despite these capabilities, many challenges remain, primarily owing to the ultrasmall scale of manipulation. However, the importance of studying single nanometric objects continues to drive the growth of nanofluidic manipulation. This field aims to overcome these challenges, paving the way for future breakthroughs that promise exceptional spatial and temporal resolutions, high throughput, and effective integration with external fields. The significance of nanofluidic manipulation is beyond the scope of this focused discussion. It resonates as a topic of broad interest given its potential to revolutionize various scientific and industrial areas. The evolving landscape of nanofluidic manipulation holds the key to unlocking unprecedented levels of precision in the study and control of single nanometric objects. As research continues to push boundaries and address challenges, the transformative impact of nanofluidic manipulation is expected to redefine the possibilities of scientific exploration and technological innovation.
Acknowledgments
This work was partially supported by Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research (JSPS KAKENHI) (JP21H04640, JP23KF0274, JP24KF0160, JP19KK0129, JP20H00497, JP18H01848, JP16K13653, JP26706010, and JP26630403), Ministry of Education, Culture, Sports, Science and Technology Grants-in-Aid for Scientific Research (MEXT KAKENHI) (JP21H05231, JP19H04678, JP17H05468, and JP26107714), Japan Science and Technology Agency Precursory Research for Embryonic Science and Technology (JST PRESTO) (JPMJPR18H5), Core Research for Evolutionary Science and Technology (CREST) (JPMJCR18H2), and the Asahi Glass Foundation, Japan. The authors would like to thank Dr Qun Ma for providing valuable comments. Also, the authors acknowledge the support received from the Japan Society for the Promotion of Science (JSPS) (P23374) and from the Development and Promotion of the Science and Technology Talents Project of Thailand (DPST532133).
Compliance with ethics guidelines
Nattapong Chantipmanee and Yan Xu declare that they have no conflict of interest or financial conflicts to disclose.
PDF
(5058KB)
