A nanofluidic biosensor using surface-enhanced Raman scattering (SERS) was developed to detect the β-amyloid (Aβ) protein, one of the biomarkers of Alzheimer’s disease (AD). Recent studies have indicated that investigating changes in relative concentrations of structure specific Aβ oligomers in cerebral spinal fluid (CSF) during the progression of AD could be important indicators for diagnosing AD pre-mortem. However, there is no definitive pre-mortem diagnosis of AD thus far because of the lack of technology available for sensitive Aβ detection. Hence, the development of a system for detecting the structure specific Aβ oligomers, along with the concentrations of these oligomers in CSF, would be useful in the investigation of the molecular mechanisms of Aβ cytotoxicity associated with AD. In this thesis, a nanofluidic trapping device trapping system for detecting biomolecules at sub-picomolar concentrations was developed for using SERS. The device, with a microchannel leading to a nanochannel, carries out dual functions: encouraging sizedependent trapping of gold nanoparticles (60nm) at the entrance of the nanochannel as well as restricting the target molecules between the gaps created by the aggregated nanoparticles. Initially, the trapping capability of the nanofluidic device was tested using fluorescent polystyrene and gold nanoparticles. UV-vis absorption spectroscopy was used to characterize the gold nanoparticle clusters at the entrance to the nanochannel. The device established controlled, reproducible, SERS active sites within the interstices of gold nanoparticle clusters and shifted the plasmon resonance to the near infrared, in resonance with incident laser light. Two strongly Raman active molecules, adenine and Congo red, were used to test the feasibility of the SERS nanofluidic device as a platform for the detection of multiple analytes. The results showed that strong SERS signals were obtained from the nanoparticle clusters at the nanochannel entrance. Once the feasibility of the approach was determined with strong Raman molecules, Aβ was detected using this nanofluidic SERS platform. Distinct surface-enhanced Raman spectra of Aβ was observed in different conformational states as a function of concentration and structure (monomer versus oligomer form) due to Aβ refolding from α-helical to a predominantly β-pleated sheet form. The sensor was also shown to potentially distinguish Aβ from insulin and albumin, confounder proteins in cerebral spinal fluid. Thus, a novel platform was developed to detect picomoler levels of Aβ with the ultimate goal of facilitating the diagnosis and understanding of Alzheimer’s disease by means of detecting structure specific oligomers of Aβ.