Optimization of 3-aminopropyltriethoxysilane functionalization on silicon nitride surface for biomolecule immobilization.

The 3-aminopropyltriethoxysilane (APTES) is a common method for biomolecule immobilization on silicon and silicon derivatives such as silicon nitride (Si3N4). However, there are many parameters which impact the efficiency of APTES modification such as APTES concentration and reaction time. Thus, various APTES concentrations (0.1%, 0.5%, 1%, 2%, 5%, and 10%) under different reaction times (15, 30, 60 and 120 min) were compared to achieve the optimal APTES modification condition which produced a thin and stable APTES layer on Si3N4 surface. The modified surfaces were characterized by contact angle (CA) measurement, Fourier transform infrared (FTIR) spectroscopy and spectroscopic ellipsometry to determine the wetting property, chemical bonding composition and surface thickness, respectively. In addition, biotin was used as a model to determine the effectiveness of APTES modification condition by coupling with glutaraldehyde (GA). The Alexa Flour 488 conjugated streptavidin was performed to visualize the presence of biotin using fluorescence microscopy due to the specifically binding between biotin and streptavidin. The atomic force microscopy (AFM) was utilized to determine the surface topology which was an indicator to demonstrate the agglomeration of APTES molecule. Moreover, ion sensitive field effect transistor (ISFET) was employed as a biosensor model to demonstrate the effect between surface thickness and sensitivity of biosensor. The results show that the APTES thickness is directly correlated to the APTES concentration and reaction time. Since the importance parameter for ISFET measurement is the distance between biomolecule and sensing membrane of ISFET, the thicker APTES layer negatively impacts the sensitivity of ISFET based biosensor because of the ion shielding effect. Therefore, these results would be valuable information for development of Si3N4 biosensor, especially ISFET based biosensor.