Large Molecule Capture

Nanopore Optical Interferometry

Silicon Kinetic's patented technology, called Nanopore Optical Interferometry (nPOI) makes use of chemically functionalized nano-porous silicon as a heterogeneous substrate for biochemical binding. Like porous silica beads in HPLC, the nano-porous silicon allows for much more protein and analyte interactions in a flow cell than does any competitive, planar surface, even those which use chemical techniques to immobilize more than a monolayer of protein. White light interferometry, a time-tested technique often used for thin film characterizations, is used to interrogate the binding events in real time.

Nanoporous silicon

Silicon forms the heart of almost all semiconductor manufacture. Given its abundance and encyclopedic knowledge base, it is perhaps a surprise that, until now, silicon has not been widely implemented as a biosensor substrate.

When a semiconductor wafer, like one used for integrated circuit manufacture, is anodized in the presence of hydrofluoric acid (HF), a nanoporous structure is formed. For the purpose here, this layer may be thought of as a series of parallel pores oriented normally to the surface of the wafer as shown in Figure 1, though other structures are known in the literature. Within this basic structure, the diameter and pore depth can be exquisitely controlled on the nanoscale. Typical for nPOI, pore diameters between 10–150 nm and depths between 0.4–2.1 μm are created. would need to have a diameter of at least five times this value in order to freely allow the protein to enter and exit. But, pores cannot be made arbitrarily large as the refractive index probe technique puts requirements on the largest possible size.

Within that range, there is however an optimum. As biomolecules are to be interrogated, pores need to be wide enough for them to enter and exit freely. Taking IgG as a reasonably large protein, its molecular weight of 150 kDa corresponds to a diameter of 10 nm. Pores would need to have a diameter of at least five times this value in order to freely allow the protein to enter and exit. But, pores cannot be made arbitrarily large as the refractive index probe technique puts requirements on the largest possible size. More detailed information is available from our application notes.

Porous Silicon

Figure 1 Scanning electron microscope (SEM) images of porous silicon. Surface views show average pore sizes of 10 and 80 nm, while three profile views show varying depths of etching, 0.4, 0.7 & 2.1 µm.


Figure 2 Sketch of the label free binding experiment. Flow configuration shown in A, and optical read out is sketeched in B.

White-light interferometry

Since the pores of poSi are smaller than the wavelength of white light and have a different refractive index than the buffer solution, the buffer-poSi interface at the top of the chip is partially reflecting. Further down the chip, the interface between the poSi and bulk silicon is likewise reflecting, as well as plane-parallel to the top surface. Such a structure forms a Fabry-Perot interferometer and can be used to interrogate the refractive index inside the pores. Reflection of white light, from the visible through near infrared, upon the thin porous surface creates an optical fringe pattern which is collected in real time. The basic data for the system then, is a reflection spectrum of the poSi biochip, which changes as a function of time.

The measured reflection spectrum, also called the interferogram or sometimes fringe pattern, contains the information about binding through its (OPD) parameter. OPD is a linear function of both depth of the pores and npoSi which increases as biomolecules bind to the surface of the pores as described below. OPD, plotted as a function of time, tracks the amount of biomolecule bound as a function of time.

In Figure 2A, analytes are shown in the flow cell. They are flowed across the porous region and enter the pores through difusion. In B, the white light interferometry is shown. The buffer-porous region interface is partially reflecting, as is the porous-region, bulk silicon region. Light reflects off of each of these region, and since they are parallel, they interfere. More detailed information is available from our application notes, and various applications examples, are presented. Quotes may be requested on line.

Capture Capacity

The typical sensorgrams one obtains using biomolecular interaction analysis (see e.g. protein-small molecule interactions) are sensitive in the sense that the technique detects down to the ppb range (assuming a tight binder ligang) and can measure molecules that are as small as 100 Da. Yet, the data is not necessarily information-rich. That is, one only knows which analyte corresponds to the signal, if one already knows what went into the system.

The key difference between SKi Pro and competitive technologies, is that the nano-porous silicon used by the SKi Sensor chips, has a much higher capture capacity than does other chips, which allows for applications like on-line affinity capture-mass spectrometry (AC-MS).

To quantify the advantage, one simply compares the internal surface are of a cyliner, of let's say 100 nm diameter. In the typical heterogeneouse, label-free binding experiment, any 100 nm circle 0.008 µm2 of useable surface area, while the 100 nm pore has 0.63 µm2 of useable surface area. Therefore, with similarly sized flow cells – necessary for similar fluidic performance – one can obtain 80× the amount of analyte captured on SKi Pro as competitive techniques. Typically one can capture up to 100 ng of protein or 3 ng of small molecule in 2 µL of eluent, well within the detectivity of down stream mass spectrometery.

Surface Area Comparison

Figure 3 Using pores, instead of planar surfaces, allows for much higher capture capacity.