10/246,284, filed September 17,2002. U. S. application serial No. 10/246,284 is a continuation-in-part of U. S. patent application No. 10/155,474, filed May 24,2002, which is a continuation-in-part of US patent application serial No. 09/942,884 filed August 29,2001.

US patent application serial No. 09/942,884 claims the benefit of U. S. Provisional application No. 60/298,147 filed June 13,2001. The entire disclosure of each of the above-identified US applications is incorporated by reference herein for all purposes.

BACKGROUND The invention relates to systems in which liquids flow at low rates, in particular at rates of less than about 100 microliters/minute. There is increasing interest in such systems, because they reduce sample sizes, reduce waste, and improve compatibility with other systems, for example High Performance Liquid Chromatography ("HPLC") systems.

Conventional pumping systems, which operate at relatively high flow rates, e. g. at least 0.1 ml/min, are described in for example J. Chrom. Sci, 12,425, 432 (1974) by Mcnair et al. , Rev. Sci. Instrum. 62,1642-1646 (1991) by LeBlanc, and U. S. Patent No. 5,777, 213 (Tsukazaki). Such systems do not provide precise control at low flow rates, and respond slowly to adjustment of the flow rate. When such systems are used to produce low flow rates, the conventional procedure is to split off a small proportion of a much larger liquid flow (see for example Journal of Chromatography A, 856,117-113 (1999) by Vissers).

SUMMARY OF THE INVENTION The present invention provides methods, systems and apparatus for delivering liquids at low flow rates, for example in the range of about 1 nanoliter/minute to about 100 microliters/minute, and, if desired, varying the flow rate in a controlled manner.

In a first aspect, this invention provides a method of supplying a liquid to a liquid outlet at a flow rate of less than about 100 microliters/minute, the method comprising 1) applying pressure from a first pressure source to a first liquid, thus causing the first liquid to flow through a first conduit; 2) detecting a first measured rate at which the first liquid is flowing through the first conduit ; 3) comparing the first measured rate with a first desired rate of flow of the first liquid through the first conduit; and 4) using information obtained in step 3 to adjust the pressure applied to the first liquid by the first pressure source to adjust the rate at which the first liquid flows through the first conduit towards the first desired rate of flow; the method preferably having at least one of the following characteristics A) the method includes 5) applying pressure from a second pressure source to a second liquid, thus causing the second liquid to flow through a second conduit; 6) detecting a second measured rate at which the second liquid is flowing through the second conduit; 7) comparing the second measured rate with a second desired rate of flow of the second liquid through the second conduit; 8) using information obtained in step 7 to adjust the pressure applied to the second liquid by the second pressure source to adjust the rate at which the second liquid flows through the second conduit towards the second desired rate of flow; 9) mixing the first liquid from the first conduit with the second liquid from the second conduit; and 10) supplying the mixture obtained in step 9 to the liquid outlet; B) the first pressure source comprises a first pneumatic-to-hydraulic booster, and step 4 is carried out by a controller, e. g. a servo-loop, connected to the first booster; C) the first pressure source comprises a first pneumatic pressure source, a first pneumatic-to-hydraulic booster, and a first pressure modulator which is located between the first pneumatic pressure supply and the first booster and which controls the amounts of pneumatic pressure supplied to the first booster; step 2 is carried out by a first flowmeter which measures the rate at which the first liquid flows from the first booster to the liquid outlet; and step 4 is carried out by a controller, e. g. a first servo-loop, which instructs the first pressure modulator; and D) the first pressure source comprises a first pneumatic pressure source, a first pneumatic-to-hydraulic booster, a first pressure modulator which is located between the first pneumatic pressure supply and the first booster and which controls the pneumatic pressure supplied to the first booster, and a first pressure sensor which is located between the first pneumatic pressure supply and the first booster; step 2 is carried out by a first flowmeter which measures the rate at which the first liquid flows from the first booster to the liquid outlet; and steps 3 and 4 are carried out by a combination of a first inner servo-loop which communicates with the first pressure sensor and the first pressure modulator, and a first outer servo-loop which communicates with the first flowmeter and the first inner servo-loop, the first outer servo-loop comparing the first measured flowrate to the first desired flowrate and outputting a first pressure set point to the first inner servo-loop, and the first inner servo-loop instructing the first pressure modulator to adjust the first pneumatic pressure supply.

In a second aspect, the invention provides systems and apparatus suitable for use in the method of the first aspect of the invention, this term being used to include liquid-filled apparatus which is in use, apparatus which must be filled with one or more liquids before it can be used, and novel components which must be assembled with other components before such apparatus is obtained.

BRIEF DESCRIPTION OF THE DRAWINGS The invention is illustrated in the accompanying schematic drawings, in which Figure 1 shows a system of the invention whose liquid outlet is connected, through an injection valve, to a separation column; Figure 2-5 show different systems of the invention using different methods to control the flow of the liquid; and Figure 6 shows the variation over time of the flow of two different liquids through the liquid outlet of a system of the invention.

DETAILED DESCRIPTION THE INVENTION In the Summary of the Invention above and in the Detailed Description of the Invention, the Examples, and the claims below, and in the accompanying drawings, reference is made to particular features of the invention. It is to be understood that the disclosure of the invention in this specification includes all appropriate combinations of such particular features. For example, where a particular feature is disclosed in the context of a particular aspect or embodiment of the invention, or a particular Figure, or a particular claim, that feature can also be used, to the extent appropriate, in combination with and/or in the context of other particular aspects and embodiments of the invention, and in the invention generally.

The term"comprises", and grammatical equivalents thereof, are used herein to mean that other components, ingredients, steps etc. are optionally present in addition to the component (s), ingredient (s), step (s) specifically listed after the term"comprises". The term "at least"followed by a number is used herein to denote the start of a range beginning with that number (which may be a range having an upper limit or no upper limit, depending on the variable being defined). For example"at least 1"means 1 or more than 1, and"at least 80%" means 80% or more than 80%. The term"at most"followed by a number is used herein to denote the end of a range ending with that number (which may be a range having 1 or 0 as its lower limit, or a range having no lower limit, depending upon the variable being defined).

For example, "at most 4"means 4 or less than 4, and"at most 40%"means 40% or less than 40 %. When, in this specification, a range is given as" (a first number) to (a second number)" or" (a first number)- (a second number)", this means a range whose lower limit is the first number and whose upper limit is the second number. Where reference is made herein to "first"and"second"components, e. g. first and second conduits, this is generally done for identification purposes; unless the context requires otherwise, the first and second components can be the same or different, and reference to a first component does not mean that a second component is necessarily present (though it may be present).

Mixing Two or More Different Liquids.

In preferred embodiments, the method of the invention has characteristic A set out above, i. e. the flow of two (or more) different liquids is independently controlled, and the two liquids are mixed before the mixture is discharged through the liquid outlet. In this embodiment, the techniques used in the various steps taken to control the flow of each liquid can be the same or different. However, it is often convenient to use the same techniques for both liquids, with the operation of those techniques being adapted to the particular flow rates needed for the different liquids. The flow rates of the two (or more) liquids can be is the same or different, and one or more of them can be constant or can vary in a controlled manner. In some embodiments, for example when the mixture of liquids is used in gradient liquid chromatography, the desired flow rates of the two (or more) liquids vary in a controlled manner as a function of time, and often the sum of the first and second desired flow rates remain substantially constant.

The liquids can be mixed by diffusion or by passive or active mixing devices. This preferred embodiment provides a predetermined flow rate of a mixed liquid of predetermined composition and permits programmed variation in the proportions of the liquids in the mixture. The programmed variation in liquid composition can for example be in the form of a series of step changes, and/or a continuous ramp, i. e. a gradient of constant or varying slope, or any of the other forms known in the separation arts. Flow controllers and servo loops can be combined to provide more complicated liquid composition variations.

Delay Volumes The delay volume of the system, i. e. the volume of the liquid or mixture of liquids after the desired conditions (e. g. a mixture containing the liquids in the desired ratio) have been established and before the liquid is discharged from the liquid outlet, is preferably small, so that desired changes in the discharged liquid take place with little delay. The delay volume may be for example less than 1 microliter or 100nL.

Detecting Liquid Flow Rates Any flow meter operating efficiently at the desired flow rates can be used to determine the rates at which the liquid is flowing through the conduit. Preferably the flowmeter provides a continuous signal over the whole range of desired flow rates, including liquid flow in both directions. Preferably the signal bandwidth of the flowmeter, i. e. the frequency corresponding to the minimum time between meaningful readings, is faster than 1 Hertz, particularly faster than 10 Hertz. Known flow meters include those disclosed in J.

MEMS, 6,119-125 (1997), by Enoksson et al.; thermal mass flow meters; thermal heat tracers as disclosed in US Patent No. 6,386, 050; optical flow meters, for example those disclosed in Appl. Opt. , 33,6073-6077 (1994) by Carvalho et al.

A preferred flowmeter comprises a capillary tube through which the liquid flows, and one or more pressure sensors which measure the pressure drop across the capillary tube, either directly or by measuring the pressures at the ends of the tube and subtracting one measurement from the other. A capillary tube has a high hydraulic resistance, but a very low liquid volume, allowing rapid response. The capillary tube can be part of the conduit through which the liquid flows. The pressure sensor can be a pressure transducer, preferably one having a volume of at most 5 microliters. Preferably, the length and diameter of the capillary tube are such that the pressure drop across it is at least one of (a) 50 psi (3.5 kg/cm2) and (b) at least 5% of the pressure applied to the liquid by the pressure source. The diameter of the capillary tube can be for example 35 to 65 microns, for example about 50 microns. Flow rates can be calculated in known manner from knowledge of the dimensions of the capillary tube and the properties of the liquid. For example, when the liquid is water, a pressure drop of about 450 psi (31.5 kg/em') through a capillary tube having a length of 10 cm and an internal diameter of 10 microns, indicates a flow rate of about 500 nL/min.

Pressures, Pressure Sources and Flow Rates.

Any pressure source can be used to drive the liquid or liquids in the system of the invention. Suitable pressure sources include electrokinetic pumps (e. g. as disclosed in US Patent No. 5,942, 093), electrokinetic flow controllers (e. g. as disclosed in U. S. Patent application serial Nos. 09/942,884 and 10/155,474), mechanically activated pumps, pneumatically activated pumps, electropneumatic pumps with or without a hydraulic amplifier, and combinations thereof. Many current designs of positive displacement pumps, such as lead-screw driven pumps, do not have a sufficiently consistent output to provide desired precision at the low flow rates used in the present invention, but they may be useful in active flow rate feedback in future designs.

Some embodiments of the invention make use of pressure sources comprising a pneumatic-to-hydraulic booster in liquid connection with the liquid supply. Any known booster can be used, including a liquid head that uses a dynamic seal on a moving solid rod that displaces liquid, the rod being coupled to the shaft of a conventional pneumatic piston.

The gain of the booster is typically greater than 1, but may also be equal to 1 (direct transfer of pressure with no amplification) or less than 1 for lower pressure applications.

There may be a check valve between the liquid supply and the booster so that liquid cannot flow from the booster to the liquid supply. There may be a pressure modulator between the primary power source, e. g. a pneumatic pressure source, and the booster. Any known pressure modulator can be used, including an electro-pneumatic controller in which an input current or voltage produces a command signal to one or more actuators within the controller. The actuator generally acts to increase or decrease the amount of air flow through the electro-pneumatic controller in order to maintain an output pressure proportional to the command signal. The system can include at least one servo-loop which is located between the flowmeter measuring the flow rate of the liquid and the pressure modulator, which compares the measured and desired flow rates, and which instructs the pressure modulator to adjust the pneumatic pressure supply so that the liquid flows at the desired rate.

Preferably, any change in the rates at which the first liquid flows through the first conduit results only from a change in the pressure applied to the first liquid by the first pressure source, and does not involve any mechanical change of the system. This feature, in combination with the fact that the measured flow rate is used to control the pressure applied to the liquid, reduces variability introduced by factors (if present all) such as check valve leakage, pump seal leakage, flexing and creep of mechanical seals, thermal expansion of components, and compression of liquids. When there are two (or more) liquids, the pressure sources controlling the liquid flows can be the same or different. Preferably the pressure source is continuously variable, is capable of providing flow rates in the range of 1 nL/minute to 10 Ill/minute or 100 Ill/minute into, for example, back pressures from about 1 atmosphere up to 5000 psi (350 kg/cm2) or higher, e. g. 10000 psi (700 kg/cm2), and has a response time of seconds or less, e. g. less than 1 second, thus allowing rapid changes in flow rates. In electrokinetic pumps and flow controllers, the hydraulic resistance is high, but the compressible volume is very low, resulting in very rapid changes. Pneumatic booster pressure supplies have larger volumes, but the pistons have very low resistance to volumetric changes, once again allowing very rapid changes.

Comparing Measured and Desired Liquid Flow Rates, and controlling the Pressure Source The measured and desired liquid flow rates can be compared in any operable way, and the results used to control the pressure source (s) in any operable way. Preferably, the liquid flow rates are adjusted to compensate for volumetric changes of the mixture which will affect the flow rate. For example, the system can make use of a controller which computes the physical properties of the liquid upon which the volume depends, such as composition, temperature and pressure. For example, the composition and mixing ratio of both liquids can be input to the controller; the flowmeter can measure the pressure; and a thermocouple in communication with the controller can take the temperature measurement. Alternatively, the system can be temperature-controlled and the temperature communicated to the controller.

In some embodiments of the. invention, at least one servo-loop controller (abbreviated herein to servo-loop) is used to control the liquid flows. The servo-loop can be of any type known in the art for example a PID loop, and can be constructed, for example, using discrete analog circuits, discrete digital circuits, dedicated microprocessors or a computer. In one class of such embodiments, the pressure source comprises a pneumatic pressure source, a pneumatic-to-hydraulic booster, and a pressure modulator which is located between the pneumatic pressure supply and the booster and which controls the amounts of pneumatic pressure supplied to the booster; the detection of the measured flow rate is carried out by a first flowmeter which measures the rate at which the liquid flows from the booster to the liquid outlet; and the comparison of the measured and desired rates of flow, and the use of the information obtained from that comparison to adjust the pressure applied to the liquid by the pressure source, are carried out by a servo-loop. In another class of such embodiments, the pressure source comprises a pneumatic pressure source, a pneumatic-to-hydraulic booster, a pressure modulator which is located between the pneumatic pressure supply and thebooster and which controls the pneumatic pressure supplied to the booster, and a pressure sensor which is located between the pneumatic pressure supply and booster; the detection of the measured rate of liquid flow is carried out by a first flowmeter which measures the rate at which the liquid flows from the first booster to the liquid outlet; the comparison of the measured and desired flow rates is carried out by a combination of an inner servo-loop which communicates with the pressure sensor and the first pressure modulator, and an outer servo- loop which communicates with the flowmeter and the inner servo-loop, the outer servo-loop comparing the measured and desired flowrates and outputting a first pressure set point to the first inner servo-loop, and the inner servo-loop instructing the pressure modulator to adjust the pneumatic pressure supply.

Response Times It is desirable that the system should have a rapid response time, i. e. should respond rapidly to pressure changes initiated by changes in the desired flow rate. The tenn"response time"is used herein to denote the time taken to reach a flow rate which is within 5% of the desired flow rate. Preferably, the system has a response time of less than 1 second, particularly less than 0.6 second, even when there is a substantial change in the desired flow rate, for example a change from a first desired flowrate to a second desired flow rate which is from 0.3 to 3 times, preferably from 0.2 to 5 times, the first desired flow rate.

The time response of the system can be understood in terms of the hydraulic resistances and capacitances. As with an electronic circuit, the product of these two gives a characteristic time constant. The hydraulic capacitance in the disclosed systems is dominated by the volumes and compressibility of the liquid, but also includes contributions from sources such as the deflection of a diaphragm in a pressure transducer.

The capillary flow meter described previously has a reasonably high hydraulic resistance, but a very low liquid volume, allowing rapid response. The compressible liquid volume (leading to capacitance) in the systems shown in the Figures is preferably on the order of 5 microliters and is due to the pressure transducer mounting.

Further Systems The liquid outlet of the systems of the invention can be connected to any further liquid flow system requiring a low-volume liquid supply. Such further systems include for example gradient HPLC systems, mass spectrometer systems, flow injection systems, analysis systems, drug delivery systems, and chemical reactor systems. Many suitable further systems include include columns with diameters from 50 urn to 1 mm (often referred to as capillary columns). The output of the column can be supplied to a detector. The detector can be for example a laser-induced fluorescence detector, an optical absorption detector, a refractive index or electrochemical detector, a mass spectrometer, or NMR spectrometer, or any other detector known in the HPLC arts. The liquid flow rate in capillary systems typically ranges up from nanoliters per minute ("nL/min") to 100 microliters per minute ("q, l/min") and may be less thanlO microliters per minute. Precise control of the flow rate in such systems is important, for example so that analyte retention times (and, therefore, analyte identification) can be reliably predicted, and so that false analyte signals can be avoided.

Flow control is particularly important for gradient separations, in which the liquid composition is varied during the course of the separation. In gradient HPLC, the liquid outputs from two (or more) sources are combined to provide a desired flow rate of known and varying composition.

Two or more systems of the invention can be run in parallel from common sources of liquids to perform multiple operations, for example separations, in parallel.

The Drawings Figure 1 shows a system of the invention comprising two variable pressure liquid supplies 12, each supplying a liquid 39 (the liquids from the two supplies being different), and a flowmeter 14 for each liquid 39. After passing through the flow meters, the liquids are mixed together, and the mixture exits through a liquid outlet 16. Each of the flow meters sends signals to a controller 18, which adjusts the pressures of the liquid supplies 12 so that the mixture contains desired proportions of the liquids 39 and flows out of the liquid outlet 16 at a desired flow rate less than 100 microliters/minute, for example less than 10 microliters/minute. The liquid outlet 16 is connected through an injection valve 20 to an HPLC separation column 22.

Figure 2 shows a system on the invention comprising a pneumatic pressure supply 32 which is connected through a pressure modulator 44 to a pneumatic-to-hydraulic booster 26.

The booster 26 comprises two coupled pistons 28 and 30 and a cylinder 34 containing liquid 39 which is forced from the booster by a pressure which is controlled by varying the pneumatic pressure applied to the piston 28. The gain of the booster is proportional to the ratio of the first piston area to the second piston area. The cylinder 34 can be refilled by withdrawing the second piston 30 and pulling fresh liquid from a liquid supply 38 through a liquid inlet 37 and a check valve 36. The flow rate of the liquid is measured by a flowmeter 14 and is input to a servo-loop controller 40. A desired flow rate is input to the servo-loop controller 40 through a set-point input 42. The servo-loop controller 40 then compares the measured flow rate to the desired flow rate and, if the measured flow rate does not equal the desired flow rate, instructs the pressure modulator 44 to adjust the pneumatic pressure to the first piston 28 of the booster 26 to achieve the desired liquid flow rate.

The liquid source 38, the check valve 36, the pressure supply 32, the pressure modulator 44, and the pneumatic-to-hydraulic booster 26 can make up one of the variable pressure liquid supplies 12 of Fig. 1. The servo-loop controller 40 and set-point input 42 can make up the controller 18 of Fig. 1.

Figure 3 shows a system similar to that shown in Figure 2, but employing two nested servo loops as the controller. An outer servo-loop 40a compares the measured flow rate to the desired flow rate then outputs a pressure setpoint 43 to an inner servo-loop 40b. A gas pressure sensor 46 measures the pneumatic pressure applied to the first piston 28 of the booster 26. The measured pneumatic pressure is input to the inner servo-loop 40b controller as well as the pressure setpoint 43 from the outer servo-loop controller 40a. If the measured flow rate is not the desired flow rate, the inner servo-loop 40b instructs the pressure modulator 44 to adjust the pneumatic pressure applied to the first piston 28 of the booster 26.

Figure 4 shows a system similar to that shown in Figure 3 in which one of the inner servo-loop 40b inputs is the measured liquid pressure from a liquid pressure sensor 48 located between the booster 26 and the flowmeter 14.

Figure 5 shows a system similar to that shown in Figure 3 in which the flowmeter 14 comprises a capillary tube 50 and first and second pressure sensors 48a and 48b, located at either end of the tube. The flow rate is calculated using the measured pressure difference across the known flow conductance of the tube 50. In another arrangement, the first pressure sensor 48a of the flowmeter 14 can provide input to the inner servo-loop 40b.

Wherein a controlled mixture of liquids is require, two (or more) systems of the invention, each of which may be as described in one of Figures 2-5 and which may be the same or different, can supply different liquids which are mixed before they reach the liquid outlet. The systems can, if appropriate, share one or more components. For example, when two systems as shown in Figure 5 are used, they can make use of a shared pressure sensor 48b.

Example 1 A system as shown in Figure 5 was prepared, and the setpoint was set to a constant value of 2700 nL/min over 50 minutes. In this system, the pressure modulator 44 was an electro-pneumatic controller Control Air 900-EHD, the pneumatic-to-hydraulic booster 26 was Haskel MS-36, and the pump liquid was deionized water. Each of the pressure sensors 48a and 48b was an Entran EPX transducer. The transducers were at either end of a 10 cm length of a 10 urn ID capillary tube 50. The pressure drop across the capillary tube was about 100 psi (7 kilograms per square centimeter). The flow rate was calculated by a microprocessor. The resulting metered flow of liquid was accurate within 0.02 % of its setpoint over a period of 50 minutes, i. e. a flow rate accuracy of 0.56 nL/min RMS.

Example 2 A system was prepared using two systems as shown in Figure 5 connected to the same liquid outlet, and sharing a pressure sensor 48b. The flow profile was a 20 minute constant gradient delivery of the first and second liquids, which were water and acetonitrile, varying between 100 and 300 nL/min, with a 1 minute period of constant flow at the beginning and end of the test. Thus, the two liquids were mixed in concentrations ranging from 25% to 75% over the duration of the test, and at a total constant flow rate of 400 nL/min into the column.

The measured flow rate from each controller indicated accuracy within 0.28 nL/min RMS of 'the setpoint over the full range of the test conditions.