Nuclear magnetic resonance (NMR) spectroscopy is one of the most powerful methods available for determining the three dimensional structure of complex chemical and biochemical molecules. The information derived from NMR has proved essential in determining primary, secondary and tertiary structures of proteins, as well as investigating enzymatic mechanisms and the binding sites of therapeutic drugs. It is also used extensively in chemical synthesis, being able to determine structure and stereochemistry of the products of intermediate steps in a multi-stage synthesis. In spite of its great potential, the application of NMR techniques to the nanoliter volume samples often encountered in microanalytical chemical analysis and studies of biological systems at the cellular has been limited. However, recent advances in the design of solenoidal NMR microcoils by our group (Olson et al., 1995) provide a basis for applying NMR methods in a variety of new applications. For example, we have begun to investigate chemical and biological applications in NMR studies of the mass limited quantities encountered in combinatorial chemistry (Gordon et al., 1994), in biological product analysis from single cells entrapped in gel microdroplets (Weaver et al., 1991) and as detectors in nanoliter volume microelectrophoresis systems (Fan and Harrison, 1994; Jacobson et al., 1994). The major drawback in such new applications of NMR techniques is the intrinsic low sensitivity of NMR compared with other analytical methods. The problem is often compounded by the very small mass of sample which is available. This can be due to: (i) the quantity which can either be synthesized or isolated, (ii) limited solubility in suitable solvents, or (iii) degradation over the long data acquisition times. Almost every commercial high resolution NMR spectrometer uses a small RF coil which fits around a 5 mm fused-silica tube giving an effective cell volume of approximately 0.5 ml. It has long been recognized that this radiofre quency (RF) coil can be reduced in size, but the advantages of this reduction have not been demonstrated until recently. Our group showed that by using solenoidal coils of diameters 300-400 microns, improvements in S/N of over 100 can be achieved over a conventional 5 mm coil for mass limited samples. Further improvements in the SNR can potentially be made by reducing the coil diameter yet further. However, fabrication of solenoids at such dimensions becomes highly problematic. Switching to lithographic fabrication techniques, however, opens up a new window of opportunity. The major problems are efficient interfacing of the microscopic planar coils with the rest of the spectrometer. In our recent NSF SGER grant (NSF BIR 93-19399, "Monolithic Gallium Arsenide Receiver for NMR Microscopy") we successfully designed and constructed a hybrid version of the proposed RF coil/preamplifer system at 300 MHz for 1H-NMR application studies at 7.05 T. The results obtained using this hybrid prototype in NMR spectroscopy experiments agree well with theoretical predictions and demonstrate the feasibility of using active monolithic detectors for improved detection performance in cellular studies. In addition, we fabricated a 500 MHz integrated circuit NMR detector for further improvement in SNR. This design can be easily extended to higher frequencies. The goal of the proposed research is to build a family of monolithic GaAs NMR receivers that will interface with current (250, 300, 500 MHz) and emerging (750, 1000 MHz) NMR microscopy systems. We will extend this design to a broadband multistage amplifier configuration that will operate over a frequency range 100 - 500 MHz, eliminating the need for tuning and matching of the receive coil. We propose to design these detector systems using an integrative approach that involves circuit simulation (Microwave Design Software, HP), device fabrication (Center for Compound Semiconductor Microelectronics, UIUC), electrical characterization (S-parameter and noise measur ements), and NMR evaluation (at each of the target magnetic field strengths).
