Sample Presence Detector for Automatic Sample Injector

This invention relates to chromatography and more particularly to a sample injection apparatus and process which enables rapid injection of sequential chromatographic samples with minimal overfill.

Existing automatic sample injector systems utilize rate of transfer and time duration to control movement of samples from storage coils into the metering loop on the sample injector valve. These systems require the application of a large excess of sample to insure filling of the metering loop on the injector valve.

Other sample injector systems employ detection schemes to control the transfer of samples into the metering loop of an injector valve. Some of these employ photoelectric detectors which, while being serviceable in systems utilizing non-polar solvents, are sensitive to entrained air. Others also employ conductivity cells, but these conductivity cells are located in the plumbing external to the valve, requiring that the necessary overfill be extended to include an extra amount of sample which must fill that volume of the system beyond the valve to the conductivity cell detector.

The disclosed detection cell apparatus in combination with the flushing and sample injection process herein disclosed is an improvement upon the sample injection valve disclosed in Ogle et al. U.S. Pat. No. 4,444,066, the disclosure of that patent constituting prior art.

A conductivity detector connected in intimate proximity to a sample injection valve enables rapid sample injection into the chromatographic column with minimum sample overfill required. The injection valve includes a housing or stator having an interior moving member or rotor. The valve rotor includes passageways. The passageways rotate into and out of registry with complementary valve conduits connected to wells within the valve housing or stator. These passageways include a sample trapping passageway, an exit passageway, and a chromatographic column passageway. The rotor is placed in a first flush and sample injecting position with respect to the stator. In this first position and during a flush cycle, the chromatographic column connects across the valve rotor at the chromatographic column passageway between two of the inlet and outlet conduits for passage of buffering solution. The remaining two conduits on the inlet side and the remaining two conduits on the outlet side are connected by the respective sample trapping passageway and exit passageway to a circuit independent of the chromatographic column. Flushing all conduits with deionized water occurs in a direction towards and through the sample storage. For the sample transfer, rinse solution is cut off, the circuit is maintained filled with deionized water and a sample holding mechanism with a full sample is indexed into the circuit. Flow is reversed through the exit passageway and valve sample trapping passageway under suction from a peristaltic pump. Sample transfer occurs and continues until the leading edge of the sample begins to leave the sample trapping passageway in the valve rotor. A dielectrically insulated conductive conduit, typically a stainless steel capillary. penetrates into the one of the entrance wells from the sample trapping passageway. Upon entrance of the leading edge of the sample to the entrance well, a circuit is completed between the injection valve housing or stator and the insulated conductive conduit. The completed circuit through an amplified signal halts pumping, trapping the leading edge of the sample at the sample trapping passageway within the valve rotor. The rotor then indexes to the insertion position, and the sample is sequentially injected through the recently flushed path into the column for conventional chromatographic analysis.

An object of this invention is to disclose in combination with an injection valve utilized in chromatographic circuit, a detector for trapping a sample with minimum overfill. A valve rotor moves a sample trapping passageway into registry with a chromatographic injection circuit. The sample trapping passageway discharges during sample injection to a conduit and well penetrated by a conductive conduit, typically a stainless steel capillary. When sample is injected through the sample trapping channel, the leading edge of the sample passes into the well, completes a signal generating electric circuit between the capillary and valve, and stops injection pumping. Thereafter, the valve rotor indexes within the valve stator to an injection position in series with the chromatographic column, the sample is injected, and conventional analysis follows.

An advantage of the apparatus and process is that minimal overfill of the sample is required. Specimen is conserved.

A further advantage of apparatus and process is that run time between sequential samples is reduced.

An additional object of this invention is to incorporate the sample trapping passageway into a prior art circuit which is readily flushed by rinse solution of previous sample. According to this aspect of the invention, after sample injection, the sample trapping passageway indexes into a rinse circuit. A deionizing solution washes the sample trapping passageway and all conduits through which the sample passes. There results the ability of the sample trapping passageway and associated injection circuit to receive sequential specimens for analysis.

FIG. 1A is an overall circuit schematic of the invention illustrating the chromatographic circuitry;

FIG. 1B is a schematic of a rotary table for dispensing sequential samples;

FIG. 1C is a sectional schematic through rotary table of FIG. 1B illustrating a circuit for drying the sample holder;

FIG. 2 is a partial schematic similar to FIG. 1A illustrating the flushing cycle of this invention;

FIG. 3 is a partial schematic similar to FIG. 1A illustrating the sample transfer of this invention;

FIG. 4 is a partial schematic similar to FIG. 1A illustrating sample injection; and,

FIG. 5 is a schematic section of the injection valve illustrating the particular cell construction that enables the determination of the leading edge of the sample passing from the sample trapping passageway of the rotor into a well of the valve stator.

The detector described herein constitutes a direct improvement of the sample injector system described in Ogle et al. U S. Pat. No. 4,444,066 issued 4/24/84 to the assignee herein. The operation of that system, as modified by the addition of the sample presence detector of FIG. 5 is described in FIGS. 1A, 1B, 1C, 2, 3 and 4. FIG. 5 is a sectional view of the valve, taken from the earlier disclosure, with the improved detector disclosed herein in place.

Referring to FIG. 1A, the complete sample injection system comprises a rotary storage table 11 having capacity to store 48 separate sample storage coils, positioning each sample storage coil in sequence over two operating stations--A and B--shown in the inset of FIG. 1B. Upon demand, the sample is drawn from its storage coil 12 into metering loop 13 by the operation of peristaltic pump 14. The routing of fluid through the injector valve V. into the metering loop 13, and subsequently into the chromatographic stream is controlled by the rotation of the injector shear-face rotor 15 which contains the passageways shown.

For the purposes of the schematics of FIGS. 1A, 2, 3 and 4, only the shear face rotor 15 will be illustrated. Reference to the stator construction is set forth in FIG. 5 and to prior art Ogle et al. U.S. Pat. No. 4,444,066.

FIG. 1A shows this valve rotor 15 in position to direct column effluent from the column pump 16, through the valve bypass passageway 17 in the shear-face rotor 15 onto the chromatographic column 18. In this valve position, the sample metering loop is connected to the sample table and to the peristaltic pump. The transfer of the sample, cleaning and drying of the storage coil, and injection of the sample is accomplished by a series of operations in a programmed schedule. These operations are described in the following discussion. For purposes of clarity, only those elements of FIG. 1A that pertain to the operation being described will be included in subsequent FIGS. 1C, 3, 4 and 5.

Referring to FIG. 2 and since a clean system is critical to the operation of the detector the Flush Cycle will be described first. This operation presumes that the sample which was contained in the storage coil on the table has already been injected. Flush proceeds as follows:

Solenoid operated flush valve 19 is opened, permitting deionized water stored under nitrogen pressure in reservoir 20 to flow in the direction shown--through the sample metering loop 13, through the sample storage coil 12 and out to waste. At the same time, the transfer pump 14 runs, ensuring that the pump is thoroughly primed with deionized water. The Flush Cycle is a timed operation that is under analyzer process control. By the end of this cycle, the system is completely filled with deionized water. During this cycle the column pump is connected to the column by the injector valve bypass passage 17 as shown in FIG. 1. Column 18 has buffering solution forced through the column by pump 16 through passageway 17 (See FIG. 1A)

Referring to FIG. 1C. the sample storage table is indexed one position. This causes the storage coil 12 that has just been flushed to advance from station A to station B. At this time, solenoid valve 21 opens, admitting nitrogen to the sample storage coil. The deionized water in the storage coil is forced out of the coil to drain and the coil is blown dry.

Also, as the sample table was indexed the storage coil for the next sample to be injected was moved into position at station A. Coil Dry is a timed operation under analyzer process control. During this period, the column pump remains connected to the chromatographic column via the bypass passage 17 as shown in FIG. 1A.

Referring to FIG. 3, the storage coil containing the sample to be injected is now at station A of the storage table (see FIG. 1B). The entire system, from the table interface at station A through the sample metering loop 13 and the sample transfer pump 14 is filled with deionized water. The sample in the storage coil 12 is separated from the table interface by a section of air in the vertical portion of the capillary in the storage coil above the table interface. This slug of air is the result of the prior art coil loading procedure, and is purposely left in place to prevent the sample from siphoning out of the storage loop while it awaits its turn to be transferred to the sample metering loop. Under process control, the transfer pump 14 is turned on causing the flush liquid, followed by the air slug and the sample to move in the direction shown toward valve V. The air slug, followed by the leading edge of the sample enters the injector valve, passes through the metering loop, and out past the injector valve shear-face rotor 15 where it moves into the conductivity cell chamber 27 just beyond the valve shear face. The presence of the leading edge of the sample is detected by a sudden increase in conductivity of the fluid in the cell. This change is converted by amplifier 22 to a logic level signal that is used by the control system 23 to stop the transfer pump motor 24.

The flush liquid and the air slug are invisible to the micro-conductivity cell. The unique construction of the conductive cell permits the advancing sample to be immediately recognized within the injector valve.

As will be seen in FIG. 5, the internal volume of the injector valve beyond the shear-face is quite small. Therefor, the amount of overfill required to insure a full metering loop is small--approximately five microliters.

During the cycle, the column pump remains connected to the chromatographic column via the bypass passage 17 of the injector valve as shown in FIG. 1A.

With the sample in place in the injector loop, the shear-face rotor 15 of the valve V is now rotated to the position shown in FIG. 4. The bypass passage 17 is rotated out of contact with the injector valve ports, and the sample metering loop is now connected to these ports. The injector valve remains in this position only long enough for the sample to be swept out of the metering loop and on to the column. The shear-face rotor 15 then rotates back to the position shown in FIG. 1A through 3. The control program then repeats the system Flush and Dry cycles described above, in preparation for transferring and injecting the next sample.

FIG. 5 shows the construction of the conductivity cell 27 and the manner by which it has been located within the injector valve, just beyond the shearface. The reader will understand that this is an improvement upon that valve claimed and described in Ogle et al. U.S. Pat. No. 4,444,066. Stainless steel (hypodermic) capillary 25 enclosed within insulating capillary 26 is inserted into the injector valve effluent port and secured with a conventional capillary nut and gland assembly. Valve V is made of conducting material and electrically communicated to ground.

The conducting capillary 25 does not reach to the end of its insulating sheath and capillary 26. Hence, a micro-volume conductivity cell 27 is formed within the injector valve body. The valve body, being of conductive metal, forms one electrode of the cell connected to circuit ground, while the hypodermic capillary 25 forms the opposite electrode of the cell. This electrode is connected by a conventional female electrical socket connector 28 to the electrical wire 29 leading to the input of buffer amplifier 22. Conventional nonconducting capillary 30 is forced over the hypodermic capillary to complete the effluent path from the injector valve to the sample transfer pump.

The manner in which the cell is formed within the injector valve body is a function of the construction of the capillary assembly that replaces that which normally connects to the valve. This assembly, and its ensuing advantages, could be applied to any of several general purpose sample injector valves now on the market. Its application is restricted to the detection of conductive fluids.

It will be understood that we prefer a stainless steel capillary. It will be understood that other conductive members penetrating the valve well could also be used.

Further, we have chosen to improve the injection valve of Ogle et al. in U.S. Pat. No. 4,444,066. This valve includes a rotor and stator construction. It will be apparent that the valve construction utilized by this invention can be altered. For example, linearly sliding valve members can as well be used.

Prior art has employed time and rate of fluid flow to locate the limits of the sample aliquot as it is transferred into the metering loop. In order to assure that these limits safely extended beyond the shear faces of the injector valve, a comparatively large overfill (50 microliters) of sample was required. Using the detector and transfer system described here, the required overfill is reduced to five microliters. The two tables below demonstrate this improvement for the three metering loop volumes that are standard.

In the high-sensitivity, small-volume applications which are of particular interest, the use of the sample presence detector has dramatically reduced the required overfill. This can become a critical consideration in those cases where sample availability is limited.