Separation and concentration of nanoscale species play an important role in various fields such as biotechnology, nanotechnology and environmental science. Inevitably, the separation efficiency strongly affects the quality of downstream detections or productions. For biotechnology and diagnostic applications, conventional separation techniques such as centrifugation, chromatography, filtration, and electrophoresis have been well established and the related instruments and reagents are readily available commercially. However, other factors such as cost, processing time, bulky instruments, infrastructure, and well trained technicians limit their applications in resource-limited settings. Consequently, innovations in materials science that can separate bionanoparticles efficiently and do not require complex setups, reagents or external fields are highly demanded. This work focuses on developing new materials for the affinity separation of bio-nanoparticles such as viruses or macromolecules from a complex mixture, such as whole blood. To enhance the interaction between target nanoparticles and the capture bed, methods to produce porous matrices with a uniform pore size matching the dimension of targets are studied. Furthermore, regarding viral separation from whole blood, macroporous materials are further patterned into microarrays to allow multiscale separation. Considering the needs in resource-limited settings, these materials are integrated with microfluidic technologies to reduce the volume of samples and reagents, simplify operating processes, and enable the use of inexpensive and portable components. Beyond the application of viral separation as demonstrated in the work, the fundamental study of macroporous material formation and transport in these materials also shed light to the separation of many other nanospecies in multiscale materials.Specifically, two macroporous materials, based on template synthesis, are created in this work. The first type employs porous anodic aluminum oxide (AAO) films as the template to create hexagonal arrays of nanoposts. However, pore sizes and interpore distances (cell size) of ordered porous AAO films are limited by the conventional fabrication process. Moreover, the process usually yields defective pore morphologies and large pore and cell size distributions. To overcome these limitations, a patterning method using nanobead indentation on aluminum substrate prior to anodization is evaluated to control the growth of AAO. Together with controlled anodizing voltages and electrolytic concentrations, AAO pore and cell sizes are shown to be tunable and controllable with narrow size distributions within submicron range. A high degree of order of AAO pore arrangement is also demonstrated. In addition, overall anodization becomes more time-efficient and stable at high anodizing voltages. Secondly, a three-dimensional (3D) assembly of microbeads is used as a template to fabricate a spherical pore network with small interconnected openings. After depositing and drying a suspension containing both micro- and nanobeads, the microbeads assemble into a 3D close-packed structure while the nanobeads fill the interstitial space. When the nanobeads are melted and microbeads are removed, a spherical pore matrix then form with small interconnected openings. Such the opening size is in submicron range can be adjusted depending on the size of microbead. The advantages of the two macroporous materials are not only controllable and tunable pore size, but also high surface-to-volume ratio due to the nanoscale features. With a ratio on the order of ~1 µm-1, the porous materials provide a significantly large binding surface. Computational and experimental results reveal that porous materials with a pore size matching the nanoparticle size are suitable for their capture. Separation of human immunodeficiency virus (HIV) is used as a model and capture yields of ~99 % and ~80 % are achieved in the nanopost structure and spherical pore network, respectively, after treated with a functional chemistry. Hence, the properties of these two macroporous materials are suitable as a size-exclusion and affinity separation for viral particles.To further explore multiscale separation, i.e. capturing viruses from whole blood, micropatterned arrays of macroporous materials have been designed. In this design, a microscale gap allows the passage of microparticles such as blood cells, and the nanoscale pores promote permeation for affinity capture of bionanoparticles. Consequently, particles with a size difference of 3-4 orders of magnitude can be separated in a simple flow-through process. Computational analyses are employed to study the effect of micropattern shape and layout. A half-ring pattern is shown to reduce flow resistance and promote fluid permeation compared to a circular pattern. In the experiment, the micropatterned porous arrays yield around 4 times higher viral capture from whole blood compared with a micropatterned solid array. The micropatterned porous devices are capable of handling a large volume of fluid sample without clogging by cells. Therefore they can be used for nanoparticle concentration. Our study also indicates that the layout of micropatterns can be adjusted to improve the capture yield. For example, an increase in pattern radius, or a decrease in gap distance between each post and in width of half ring will enhance fluid permeation in the porous structure. When combined with downstream detection, these materials integrated into microfluidic platforms can be created as point-of-care diagnostics, as well as other applications for particle separation and analysis.