IEEE JOURNAL OF SOLID-STATE CIRCUITS, VOL. 44, NO. 5, MAY 2009 1629 CMOS RF Biosensor Utilizing Nuclear Magnetic Resonance Nan Sun, Student Member, IEEE, Yong Liu, Member, IEEE, Hakho Lee, Member, IEEE, Ralph Weissleder, and Donhee Ham, Member, IEEE Abstract—We report on a CMOS RF transceiver designed for detection of biological objects such as cancer marker proteins. Its main function is to manipulate and monitor RF dynamics of protons in water via nuclear magnetic resonance (NMR). Target objects alter the proton dynamics, which is the basis for our biosensing. The RF transceiver has a measured receiver noise figure of 0.7 dB. This high sensitivity enabled construction of an entire NMR system around the RF transceiver in a 2-kg portable platform, which is 60 times lighter, 40 times smaller, yet 60 times more mass sensitive than a state-of-the-art commercial benchtop system. Sensing 20 fmol and 1.4 ng of avidin (protein) in a 5 L sample volume, our system represents a circuit designer’s ap- proach to pursue low-cost diagnostics in a portable platform. Index Terms—Biosensor, CMOS integrated circuit, low noise amplifier, nuclear magnetic resonance, RF transceiver. I. INTRODUCTION S ILICON radio frequency (RF) integrated circuits (ICs) have been at the center stage of various wireless chip developments over the past years. Here we report on a silicon RF IC designed for a different application, that is, sensing biological objects. In a disease development, biomolecules characteristic to the disease, such as viruses and cancer marker proteins, emerge. The ability to sense these biomolecules would facilitate disease detection. Researchers from many areas of science and engi- neering are developing a variety of biosensors, aiming at in- creased sensitivity or low-cost diagnostics [1]. Our RF biosensor is a “circuit designer’s way” to pursue low-cost diagnostics in a portable platform. Fig. 1 shows our RF biosensor, central to whose operation is the silicon RF transceiver IC. The underpinning physical phenomenon of our sensing method is nuclear magnetic reso- nance (NMR), the resonant interaction between RF magnetic fields and protons in water, which is altered by the presence of target biological objects. This NMR-based biosensing was de- veloped in 2002 [2] and has been used within a state-of-the-art commercial benchtop NMR system [3] which, however, is bulky and heavy (120 kg). The main contribution of our work is to drastically shrink the entire NMR system by devel- Manuscript received July 28, 2008; revised December 08, 2008. Current ver- sion published May 01, 2009. N. Sun and D. Ham are with the School of Engineering and Applied Sciences, Harvard University, Cambridge, MA 02138 USA (e-mail: nansun@seas.har- vard.edu; donhee@seas.harvard.edu). Y. Liu was with Harvard University, Cambridge, MA 02138 USA, and is now with the IBM T. J. Watson Research Center, Yorktown Heights, NY 10598 USA. H. Lee and R. Weissleder are with the Center for Systems Biology, Massa- chusetts General Hospital, Harvard Medical School, Boston, MA 02114 USA. Digital Object Identifier 10.1109/JSSC.2009.2017007 Fig. 1. CMOS RF biosensor (this work) utilizing NMR. The entire system weighs 2 kg, where the commercial magnet dominates the weight. Prior to this work, we built an intermediate miniature NMR biosensor [4] where we used the same magnet and in-house fabricated microcoil with discrete electronics. oping the RF transceiver IC, hence enabling the NMR-based biosensing in the portable form of Fig. 1. Occupying only 2.5 liters and weighing only 2 kg, our system is 60 times lighter, 40 times smaller, yet 60 times more mass sensitive than the state-of-the-art benchtop NMR system: our system is actually the smallest complete NMR system ever built. As a biosensor, it detects 20 fmol, 1.4 ng of avidin (protein) in a 5 L sample volume. This detection threshold can be even further improved, for it is currently limited not by our transceiver sensitivity, but by the specific bioassay used [4], [5]. Our system might offer a new way to pursue low-cost diagnostics in a portable platform. After part of this work was briefly reported in the IEEE ISSCC [7], new key experiments ensued to complete this work. The goal of the present paper is to describe this entire work (design of, and experimentation with, the RF IC) in Sections IV–VII. We add two perspective sections, Sections III and VIII, to explain how our RF IC facilitated the dramatic miniaturization, and how this work is differentiated from existing miniaturization efforts. All of this would be best understood with some familiarity with NMR, so we start by reviewing the NMR basics. II. REVIEW OF NUCLEAR MAGNETIC RESONANCE Although NMR is a well-established subject [6], we provide this review to quickly introduce the basic concepts of NMR rel- 0018-9200/$25.00 © 2009 IEEE