To reduce unwanted variation in the passage speed of DNA through solid-state nanopores, we demonstrate nanoscale pre-confinement of translocating molecules using an ultra-thin nanoporous silicon nitride membrane separated from a single sensing nanopore by a nanoscale cavity. We present comprehensive experimental and simulation results demonstrating that the presence of an integrated nanofilter within nanoscale distances of the sensing pore eliminates the dependence of molecular passage time distributions on pore size, revealing a global minimum in the coefficient of variation of the passage time. These results provide experimental verification that the inter- and intra-molecular passage time variation depends on the conformational entropy of each molecule prior to translocation. Furthermore, we show that the observed consistently narrower passage time distributions enables a more reliable DNA length separation independent of pore size and stability. We also demonstrate that the composite nanofilter/nanopore devices can be configured to suppress the frequency of folded translocations, ensuring single-file passage of captured DNA molecules. By greatly increasing the rate at which usable data can be collected, these unique attributes will offer significant practical advantages to many solid-state nanopore-based sensing schemes, including sequencing, genomic mapping, and barcoded target detection.

Elucidating the kinetics of DNA passage through a solid-state nanopore is a fertile field of research, and mechanisms for controlling capture, passage, and trapping of biopolymers are likely to find numerous technological applications. Here we present a nanofiltered nanopore device, which forms an entropic cage for DNA following first passage through the nanopore, trapping the translocated DNA and permitting recapture for subsequent reanalysis and investigation of kinetics of passage under confinement. We characterize the trapping properties of this nanodevice by driving individual DNA polymers into the nanoscale gap separating the nanofilter and the pore, forming an entropic cage similar to a "two pores in series" device, leaving polymers to diffuse in the cage for various time lengths, and attempting to recapture the same molecule. We show that the cage results in effectively permanent trapping when the radius of gyration of the target polymer is significantly larger than the radii of the pores in the nanofilter. We also compare translocation dynamics as a function of translocation direction in order to study the effects of confinement on DNA just prior to translocation, providing further insight into the nanopore translocation process. This nanofiltered nanopore device realizes simple fabrication of a femtoliter nanoreactor in which to study fundamental biophysics and biomolecular reactions on the single-molecule level. The device provides an electrically-permeable single-molecule trap with a higher entropic barrier to escape than previous attempts to fabricate similar structures.

The neural network method of solving differential equations is used to approximate the electric potential and corresponding electric field in the slit-well microfluidic device. The device's geometry is nonconvex, making this a challenging problem to solve using the neural network method. To validate the method, the neural network solutions are compared to a reference solution obtained using the finite-element method. Additional metrics are presented that measure how well the neural networks recover important physical invariants that are not explicitly enforced during training: spatial symmetries and conservation of electric flux. Finally, as an application-specific test of validity, neural network electric fields are incorporated into particle simulations. Conveniently, the same loss functional used to train the neural networks also seems to provide a reliable estimator of the networks' true errors, as measured by any of the metrics considered here. In all metrics, deep neural networks significantly outperform shallow neural networks, even when normalized by computational cost. Altogether, the results suggest that the neural network method can reliably produce solutions of acceptable accuracy for use in subsequent physical computations, such as particle simulations.

The translocation of polymers through nanopores with large internal cavities bounded by two narrow pores is studied via Langevin dynamics simulations. The total translocation time is found to be a nonmonotonic function of polymer length, reaching a minimum at intermediate length, with both shorter and longer polymers taking longer to translocate. The location of the minimum is shown to shift with the magnitude of the applied force, indicating that the pore can be dynamically tuned to favor different polymer lengths. A simple model balancing the effects of entropic trapping within the cavity against the driving force is shown to agree well with simulations. Beyond the nonmonotonicity, detailed analysis of translocation uncovers rich dynamics in which factors such as going to a high force regime and the emergence of a tail for long polymers dramatically change the behavior of the system. These results suggest that nanopores with internal cavities can be used for applications such as selective extraction of polymers by length and filtering of polymer solutions, extending the uses of nanopores within emerging nanofluidic technologies.