Delivered volume was also highly reproducible over this large volume range. Should smaller volumes, e.g. less than 100 μl, be required then the internal diameter of the tubing contained within the peristaltic pump could be reduced to improve accuracy. Over the past few years a number of different solutions have been designed to address reproducibility in delivery of hyperpolarized substrate. A system by Bowen and find more Hilty [8] was designed for in vitro use to rapidly (1200 ms)

inject hyperpolarized dissolute into a high resolution NMR spectrometer. Specifically their system used high pressure, >40 bar, to ensure that an aqueous solution reliably filled a 5 mm NMR tube without air bubbles – a common issue due to the high viscosity and surface tension of water. Due

to its high operating pressure their design would not be readily applicable to in vivo use without stepping down the pressure. A computer controlled in vivo injector was described by Comment et al [9], further improved in [10], that addressed the issue of bubble formation by allowing the chase gas (used to assist transfer of the sample from the polarizer to the injector) to exit through vents. A hydraulically driven plunger then sealed the vent holes as the sample was injected into the animal. DAPT An in-line optical sensor halted the injection if a bubble was detected within the injection cannula. The presence of a vent hole affects the accuracy of such a system, as there would be some variability in the amount of liquid injected into the animal as these vents were sealed. Hydraulic based systems also have some inaccuracy due to friction in actuating the hydraulic cylinder(s). In our described system, the possibility of injecting an air bubble was minimized by having a continuous fluid path from the cannula to the RV. The outlet pipe click here of the RV to the pump was also always submerged. The ingress of hyperpolarized substrate passed down the side of the RV wall to smoothly fill the RV and a vacuum pump removed excess gas. In practice, no bubbles

were found to have formed within the RV and so this was not regarded as a safety issue. However, an optical bubble detection system, as described [9], could be added and operated with the flow diversion system described here to prevent accidental injection of air into the animal. The design of the RV would permit other quality control systems, similar to those used in a clinical DNP polarizer [11], e.g. volume, temperature, free radical concentration sensors, to be added. Although not included on the current injector, an electrical or chemical heating system would prevent administration of relatively cold substrate to the animal. This would be due to the reduced temperature of the hyperpolarized substrate as it passes through the cannula to the animal while in the room temperature magnet bore (14 °C). Injection of cool substrate has been observed by us to cause an approximate 0.