Anal.Chem. 1997, 69, 1174-1178
`
`Generating Electrospray from Microchip Devices
`Using Electroosmotic Pumping
`R. S. Ramsey and J. M. Ramsey*
`
`ChemicalandAnalyticalSciencesDivision,OakRidgeNationalLaboratory,OakRidge,Tennessee37831-6142
`
`A method of generating electrospray from solutions emerg-
`ing from small channels etched on planar substrates is
`described. The fluids are delivered using electroosmoti-
`cally induced pressures and are sprayed electrostatically
`from the terminus of a channel by applying an electrical
`potential of sufficient amplitude to generate the electro-
`spray between the microchip and a conductor spaced from
`the channel terminus. No major modification of the
`microchip is required other than to expose a channel
`opening. The principles that regulate the fluid delivery
`are described and demonstrated. A spectrum for a test
`compound, tetrabutylammonium iodide, that was con-
`tinuously electrophoresed was obtained by coupling the
`microchip to an ion trap mass spectrometer.
`
`instruments that incorporate liquid-
`Miniaturized chemical
`phase separations such as electrophoresis and electrochromatog-
`raphy are being increasingly recognized as a convenient means
`of manipulating and analyzing small quantities of material.1-15
`These “micro” devices are fabricated on glass or quartz substrates
`using standard micromachining techniques such as photolithog-
`raphy, wet chemical etching, and thin-film deposition .16 The
`products are planar devices with micrometer-sized channels
`through which materials are manipulated using electrokinetic
`forces. Structures with various interconnecting channels may be
`easily fabricated, allowing separations and reactions to be per-
`formed on picoliter volume samples at relatively high speed.
`Perhaps the greatest promise of these devices lies in the ability
`
`(1) Manz, A.; Harrison, D. J.; Verpoorte, E.; Widmer, H. M. Adv. Chromatogr.
`1993, 33, 1.
`(2) Manz, A.; Harrison, D. J.; Verpoorte, E. M. J.; Fettinger, J. C.; Paulus, A.;
`Lu¨di, H.; Widmer, H. M. J. Chromatogr. 1992, 593, 253.
`(3) Harrison, D. J.; Fluri, K.; Seiler, K.; Fan, Z.; Effenhauser, C. S.; Manz, A.
`Science 1993, 261, 895.
`(4) Jacobson, S. C.; Hergenro¨der, R.; Koutny, L. B.; Ramsey, J. M. Anal. Chem.
`1994, 66, 1114.
`(5) Jacobson, S. C.; Ramsey, J. M. Electrophoresis 1995, 16, 481.
`(6) Jacobson, S. C.; Moore, A. W.; Ramsey, J. M. Anal. Chem. 1995, 67, 2059.
`(7) Burggraf, N.; Manz, A.; Effenhauser, C. S.; Verpoorte, E.; de Rooij, N. F.;
`Widmer, H. M. J. High Resolut. Chromatogr. 1993, 16, 594.
`(8) Raymond, D. E.; Manz, A.; Widmer, H. M. Anal. Chem. 1994, 66, 2858.
`(9) Effenhauser, C. S.; Paulus, A.; Manz, A.; Widmer, H. M. Anal. Chem. 1994,
`66, 2949.
`(10) Woolley, A. T.; Mathies, R. A. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 11348.
`(11) Woolley, A. T.; Mathies, R. A. SPIE Proc. 1995, 2398, 36.
`(12) Jacobson, S. C.; Hergenro¨der, R.; Koutny, L. B.; Ramsey, J. M. Anal. Chem.
`1994, 66, 2369.
`(13) Moore, A. W., Jr.; Jacobson, S. C.; Ramsey, J. M. Anal. Chem. 1995, 67,
`4184.
`(14) Jacobson, S. C.; Hergenro¨der, R.; Moore, A. W., Jr.; Ramsey, J. M. Anal.
`Chem. 1994, 66, 4127.
`(15) Jacobson, S C.; Ramsey, J. M. Anal. Chem. 1996, 68, 720.
`(16) Ko, W. H.; Suminto, J. T. In Sensors; Gopel, W., Hasse, J., Zemmel, J. N.,
`Eds.; VCH: Weinhein, Germany, 1989; Vol. 1, pp 107-168.
`
`to design and construct systems with the necessary geometry to
`perform diverse and complex chemical and analytical functions
`on a given sample.14,15,17 Reduced sample and reagent consump-
`tion and increased precision and reproducibility, relative to bench-
`scale instruments, are other advantages that have been cited.4,17-19
`Analyte detection remains an important issue for the microchip
`devices, however, given the small sample volumes typically
`analyzed. Most analyses have, in fact, been conducted using laser-
`induced fluorescence (LIF), where small sample volumes are
`easily probed and high sensitivity in terms of low mass detection
`limits may be obtained. Single-molecule detection has been
`accomplished “on-microchip” using LIF.20 To extend the ap-
`plicability of these miniaturized devices to species that are not
`fluorescent (or are easily converted to fluorescent species), it is
`desirable to have other modes of detection available. The
`informational output would be significantly improved by coupling
`these devices, for example, with mass spectrometry (MS) where
`both molecular weight and structural data for analytes may be
`acquired. We report here a method for delivering solutions to a
`channel opening and for generating electrospray from solutions
`emerging from the channel opening, enabling these microdevices
`to be directly interfaced with the macroworld of mass spectrom-
`eters.
`
`EXPERIMENTAL SECTION
`The microchips were fabricated using standard procedures as
`previously described.6 Figure 1 shows a schematic diagram of a
`planar glass microchip used to generate electrospray. The
`drawing is not to scale. The relative channel lengths for the
`experiments discussed below are 33 mm from the uppermost
`reservoir to the side-arm intersection, 22 mm for the side arm,
`and 1 mm from the side-arm intersection to the channel opening.
`The channels were machined into the surface of the glass
`substrate using photolithographic patterning and wet chemical
`etching. Channel dimensions were (cid:24)10 (cid:237)m deep and 60 (cid:237)m wide
`at half-depth. A glass cover plate was then direct bonded over
`the open channel structure to form a closed network. Small fluid
`reservoirs attached with epoxy resin where the channels exit from
`underneath the cover plate allow fluidic communication and
`electrical contact with the channels. The channel opening from
`where the electrospray is generated was created by scoring and
`breaking the fabricated microchip. The exit end on some
`
`(17) Jacobson, S. C.; Koutny, L. B.; Hergenro¨der, R.; Moore, A. W., Jr.; Ramsey,
`J. M. Anal. Chem. 1994, 66, 3472.
`(18) Seiler, K.; Harrison, D. J.; Manz, A. Anal. Chem. 1993, 65, 1481.
`(19) Jacobson, S. C.; Hergenro¨der, R.; Koutny, L. B.; Warmack, R. J.; Ramsey, J.
`M., Anal. Chem. 1994, 66, 1107.
`(20) Fister, J. C.; Davis, L. M.; Jacobson, S. C.; Ramsey, J. M. High Sensitivity
`Detection on Microchips. Presented at Laser Applications to Chemical
`Analysis, Orlando, FL, March 18, 1996.
`
`1174 AnalyticalChemistry,Vol.69,No.6,March15,1997
`
`S0003-2700(96)01067-0 CCC: $14.00 © 1997 American Chemical Society
`
`Downloaded via UNIV OF VIRGINIA on November 6, 2018 at 18:26:57 (UTC).
`
`See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
`
`Agilent Exhibit 1268
`Page 1 of 5
`
`

`

`DISCUSSION
`As a solution-based technique, a variety of electrospray ioniza-
`tion sources have evolved to accommodate different methods of
`liquid sample introduction including electrically driven flow.23-26
`Successful coupling of capillary electrophoresis (CE) with MS
`requires that the interface supply the conditions necessary to
`establish a stable electrospray and the electrical bias necessary
`to drive the electrophoresis. In addition, the efficiency of the CE
`separation should not be impaired by the interface.
`Intercon-
`nected channel structures that allow fluidic and electrical contact
`with a CE channel are easily fabricated in planar structures and
`allow flexible designs for accomplishing electrospray CE. The
`planar substrates onto which the capillary channels are etched,
`however, differ significantly from other electrospray ion sources
`where fluids are sprayed from needles or capillary tubes.26-28 To
`determine whether solutions emerging from a channel could be
`electrosprayed without modifying the exit end of the microchip,
`some initial experiments were performed using positive pressure
`to force fluid through the channel exit.
`In general, the low
`currents generated in electrospray are insufficient to induce the
`electroosmotic flow necessary to supply the fluid to the channel
`opening to sustain an electrospray. Pressures of a few psig were
`therefore applied to the uppermost reservoir of a simple cross-
`microchip (similar in structure to that illustrated in Figure 1 but
`without the side arm) while a potential of 3-6 kV, relative to a
`grounded electrode, was also applied to the reservoir. The
`electrode (a flat aluminum disk) was placed (cid:24)3-5 mm from the
`exit of the microchip. Parts a and b of Figure 2 are respectively
`photomicrographs showing a bead of liquid as it emerges from
`the channel and its deformation into a Taylor cone by the electric
`field between the droplet and the ground plate as a potential is
`applied. A fine spray can also be seen exiting the tip of the Taylor
`cone. The current measured by a picoameter at the target
`electrode ranged upward from 20 nA, depending upon the spacing
`from the microchip, the composition of the solution, and the
`applied potential. Pure water and mixtures of water and methanol
`were electrosprayed in this manner.
`An alternative method of transporting fluid to the channel
`opening is to use electroosmotically induced pressures. The
`velocity, v, of a fluid through a channel when driven by electroos-
`mosis is given by eq 1, where E is the electric field strength, œ is
`
`v ) (cid:15)o(cid:15)œE/4(cid:240)Ł
`
`(1)
`
`the zeta potential across the Stern layer at the channel-fluid
`interface, (cid:15) is the dielectric constant of the fluid, (cid:15)(cid:239) is the
`permitivity of free space, and Ł is the viscosity of the fluid at the
`interface. The steady-state velocity of the electroosmotically
`induced fluid flow does not depend on the cross-sectional dimen-
`sions or geometry of the channel when all dimensions are much
`
`(23) Niessen, W. M. A.; Tjaden, U. R.; van der Greef, J. J. Chromatogr. 1993,
`636, 3.
`(24) Cai, J.; Henion, J. J. Chromatogr., A 1995, 703, 667.
`(25) Smith, R. D.; Wahl, J. H.; Goodlett, D. R.; Hofstadler, S. A. Anal. Chem.
`1993, 65, 574A.
`(26) Covey, T. In Biochemical and Biotechnological Applications of Electrospray
`Ionization Mass Spectrometry; Snyder, P. A., Ed.; ACS Symposium Series
`619; American Chemical Society: Washington DC, 1995; pp 21-59.
`(27) Whitehouse, C. M.; Dreyer, R. N.; Yamashita, M.; Fenn, J. B. Anal. Chem.
`1985, 57, 675.
`(28) Fenn, J. B.;Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Science
`1989, 246, 64.
`
`AnalyticalChemistry,Vol.69,No.6,March15,1997 1175
`
`Figure 1. Schematic diagram of a microchip used to electroos-
`motically pump fluids and to generate electrospray.
`
`microchips were polished to allow a clear microscopic view of
`the opening and liquid as it emerged from the channel. The
`polishing did not noticeably affect microchip performance. Elec-
`trical potentials were applied to the solutions in the top and side-
`arm reservoirs through platinum wires connected to separate
`power supplies (Spellman CZE1000R).
`A linear polyacrylamide coating was used to eliminate or
`reduce electroosmotic flow in the side-arm channel.21 All channels
`were initially rinsed with 0.1 N sodium hydroxide and water. The
`side-arm channel was then filled and flushed with a 1% (v/v)
`solution of 3-(trimethoxysilyl)propyl methacrylate (Aldrich) in
`methanol for a minimum of 30 min by applying vacuum to the
`exit port. The channel was then rinsed with methanol and flushed
`for 30 min with a 3% (w/v) aqueous acrylamide solution containing
`0.1% (w/v) potassium persulfate (Bio-Rad Laboratories) and 0.1%
`(v/v) N,N,N¢,N¢- tetramethylethylenediamine (Bio-Rad Laborato-
`ries). Excess acrylamide solution was removed and the channel
`extensively rinsed with water. Reservoirs other than the side arm
`were filled with methanol during treatment with methacrylate and
`with water during treatment with acrylamide.
`The microchips were imaged using laser-induced fluorescence
`to monitor fluid flow. The beam from an argon ion laser (514.5
`nm, 100 mW) was expanded to a 4 mm spot on the surface of the
`microchip using a lens, and the fluorescence signal was collected
`with an optical microscope following spectral filtering (550 nm
`cut-on) and measured with the CCD. Rhodamine B and 6G dye
`were used for imaging.
`Mass spectrometry was performed on a Finnigan MAT ITMS
`(San Jose, CA) modified for ES ionization.22 The microchip was
`secured with adhesive to a flat plate which was bolted to an x-y-z
`translational stage. The chip extended (cid:24)1 cm past the edge of
`the plate, and the channel opening was carefully positioned 3-5
`mm from the inlet orifice on the front aperture plate of the mass
`spectrometer. Focusing conditions were optimized to maximize
`the intensity of the analyte ions. Ions were injected for 100 ms
`and held for an additional 300 ms in the ion trap to promote
`desolvation prior to mass analysis.
`
`(21) Hjerten, S. J. J. Chromatogr. 1985, 347, 191.
`(22) Van Berkel, G. J.; Glish, G. L.; McLuckey, S. A. Anal. Chem. 1990, 62,
`1284.
`
`Agilent Exhibit 1268
`Page 2 of 5
`
`

`

`Figure 3. Diagram of a structure, with intersection I1 and three
`channels, C1, C2, and C3, connecting the respective ports, P1, P2,
`and P3, to illustrate electroosmotically induced pumping principles.
`
`the channel axis,
`
`E(z) ) - dV
`dz
`
`)I
`
`F
`A(z)
`
`(4)
`
`where V is the electric potential (assumed to be decreasing with
`increasing z). Inserting (1) and (4) into (2) gives,
`F ) I((cid:15)o(cid:15)œF/4(cid:240)Ł)
`
`(5)
`
`the
`is a constant along the channel,
`Since electric current
`electroosmotically induced flow rate is independent of channel
`cross section, assuming that the channel interface and fluid are
`i.e., the material constants (cid:15), œ, F, and Ł are
`homogeneous;
`constant. Changes in channel cross section result in modifications
`of the electric field strength and thus fluid velocity to maintain
`constant flow rate. For an interconnected channel structure, the
`electroosmotic flow will follow the current path as that path is
`the direction of the potential gradient or electric field.
`The above analysis indicates for a structure such as shown in
`Figure 3 that when an electric potential is applied between ports
`1 and 2, no flow will pass from port 1 to port 3. The channels in
`Figure 3 are indicated by Ci, where i is the label defining the
`port, and the intersection is labeled as I1. For microchip channel
`structures considered here, the Reynolds numbers are much less
`than unity. Thus the viscous forces far exceed the inertial forces
`and there is no significant pressure generated at the intersection
`by the momentum of the fluid flow in channel 1. The above
`equations assume that the electroosmotic pumping is everywhere
`equivalent. Reduction of the electroosmotic fluid flow in C2
`relative to C1, however, will generate an excess pressure in
`channel 1, allowing pumping of fluid from port 1 to ports 2 and 3.
`A number of methods may be used to reduce the electroosmotic
`flow in C2 relative to C1 including reducing the dimensions below
`that of the Stern layer, reducing the œ, or increasing the surface
`viscosity, within channel C2. Assuming that the electroosmotic
`
`Figure 2. Photomicrographs of (a, top) a water droplet (volume (cid:24)12
`nL) forced through a channel opening by positive pressure and (b,
`bottom) the Taylor cone and electrospray generated at the opening
`applying a 3 kVpotential between the microchip and a target electrode
`spaced a few millimeters from the channel opening.
`
`greater than the Stern layer thickness. When an electric potential
`is applied across a channel of complex geometry and axially
`varying cross-sectional area, the current and electroosmotically
`induced volumetric flow rate are constant along the channel axis,
`assuming homogenous interfacial and fluid conditions. This
`implies that there will be no pressure gradients induced by
`electroosmotic pumping under homogenous conditions. This can
`be confirmed from the following equations.
`The flow rate, F, is given by
`
`F ) vA(z)
`
`(2)
`
`where A(z) is the cross-sectional area of the flow channel as a
`function of the axial distance z. Ohm’s law for the channel can
`be written as,
`
`V ) IR(z) ) IFs
`
`L dz
`A(z)
`0
`
`(3)
`
`where I is the current, R the resistance, L the channel length,
`and F the resistivity of the buffer solution. The electric field
`strength, E, is given by the gradient of the electric potential along
`
`1176 AnalyticalChemistry,Vol.69,No.6,March15,1997
`
`Agilent Exhibit 1268
`Page 3 of 5
`
`

`

`flow velocity has been reduced to zero in this channel, then the
`pressure generated at the intersection I1 can be calculated using
`eq 1 and the equation for pressure-induced flow in channels.29 At
`neutral pH and field strengths of (cid:24)350 V/cm as typically used in
`these experiments, the linear velocity of the electroosmotic flow
`is (cid:24)2.5 mm/s, which translates to a flow rate of 1.5 nL/s, given
`channels of the dimensions discussed here. Approximately 95%
`of this flow exits the channel for the microchip described in Figure
`1. Flow rates at these levels are sufficient to sustain a stable
`electrospray.30-33
`These pumping principles were demonstrated using a micro-
`chip of the design illustrated in Figure 1. A fluorescence image
`of the side-arm channel intersection with homogeneous channel
`conditions, i.e., all channel walls have native glass surfaces and
`uniform œ, is shown in Figure 4a. A positive potential relative to
`the side-arm reservoir was applied to the uppermost reservoir
`connected to the central separation channel. At the intersection,
`a small amount of fluid can be seen to extend past the intersection,
`but the bulk flow is toward the side-arm reservoir. There is no
`flow of fluid below the intersection. The spatial distribution of
`fluorescence is stationary with time. The narrow bright line seen
`in the image that extends below the intersection is an artifact from
`specularly reflected light from a channel facet. Figure 4b shows
`a fluorescence image taken of the same microchip intersection
`under the same experimental conditions except that the side-arm
`channel and the channel between the intersection and channel
`opening have been surface modified with linear polyacrylamide.
`The linear polyacrylamide at the walls greatly increases the surface
`viscosity and thus reduces the electroosmotic flow, as indicated
`by eqs 1 and 2.
`It is clear in Figure 4b that the dye solution
`propagates below the intersection indicating the induced pressure
`generated at the intersection by the spatially inhomogeneous
`surface viscosity. The images in Figure 4 were taken using
`Rhodamine B (100 (cid:237)M in water), a zwitterionic dye that is near
`neutral charge at neutral pH. Similar images were obtained using
`the cationic dye Rhodamine 6G. The splitting at the side-arm
`intersection for ions must also include electrophoretic forces,
`features of which are presently under study.
`This electroosmotically driven fluid flow has been used to
`supply the fluid for electrospray from the microchip. Figure 5
`shows a photomicrograph of electrospray generated from a 60%
`water/40% methanol solution. A voltage of 6 kV was applied to
`the uppermost reservoir and 4 kV was applied to the side-arm
`reservoir, providing (cid:24)1.2 kV potential drop for electroosmotic
`pumping within the uncoated channel segment and (cid:24)4.8 kV
`potential for electrospray formation. The microchip was positioned
`3-5 mm from the target plate which was held at ground potential.
`Alternatively, of course, the side arm may be held at ground
`potential and the counter electrode at the elevated voltages
`necessary to establish the electrospray. The target is evident on
`right-hand side of the photograph. As shown in Figure 2b with
`the pressure-induced flow, Taylor cone formation is clearly evident
`with an emanating electrospray stream. An argon ion laser was
`used to illuminate the electrospray in the image shown in Figure
`
`(29) Batchelor, G. K. An Introduction to Fluid Mechanics; University Press: New
`York, 1967.
`(30) Wilm, M.; Mann, M. Anal. Chem. 1996, 68, 1.
`(31) Wahl, J. H.; Goodlett, D. R.; Udeseth, H. R.; Smith, R. D. Anal. Chem. 1992,
`64, 3194.
`(32) Valaskovic, G. A.; Kelleher, N. L.; McLafferty, F. W. Science 1996, 273,
`1199.
`(33) Ramsey, R. S.; McLuckey, S. A. J. Microcolumn Sep. 1995, 7, 461.
`
`Figure 4. CCD images of 100 (cid:237)M Rhodamine B in water obtained
`at the intersection of a side-arm channel with a main channel on a
`microchip, applying a positive potential at the top channel relative to
`the side arm and using (a) all native glass surfaces and (b) a linear
`polyacrylamide surface coated side-arm channel.
`
`5. Supplemental information on the electrospray may also be
`viewed by accessing a video clip. Features that should be noted
`are the narrowly focused plume and the stability of the spray.
`A mass spectrum obtained from a 10 (cid:237)M solution of tetrabu-
`tylammonium iodide (molecular weight 369) electrosprayed from
`the microchip using electroosmotic pumping to deliver the analyte
`is shown in Figure 6. Voltage conditions on the microchip were
`the same as described above, and the sample in 60% water/40%
`methanol was continuously electrophoresed from the main chan-
`nel to the exit where it was electrosprayed. The spectrum was
`averaged over 2 s to simulate peak widths that may be obtained
`by microchip electrophoresis at low field strengths. Assuming a
`flow rate of (cid:24)1.5 nL/s, only 30 fmol of sample was consumed
`during this acquisition period. The tetrabutylammonium ion (M
`- I)+ at mass/charge 242 is clearly evident as are some solvent
`
`AnalyticalChemistry,Vol.69,No.6,March15,1997 1177
`
`Agilent Exhibit 1268
`Page 4 of 5
`
`

`

`which the analyte is sprayed. This approach is less involved than
`applying a conductive coating to the exit end to establish electrical
`sheathless CE-MS
`contact,
`a
`strategy
`employed
`in
`interfaces.24,25,31-34 As such, the interface is not dependent upon
`the longevity or durability of such a coating, factors that have been
`a consideration in the sheathless interfaces.34 The linear poly-
`acrylamide coating used to modify the surface of the side channel
`to pump solutions to the exit is covalently bound to the glass
`surface and is expected to be relatively stable when buffers at
`moderate pH are used.35 Other treatments that eliminate or
`reduce electroosmotic flow that may be more inert under acidic
`or basic conditions would, of course, be appropriate for surface
`modification as well. We were also able to electrospray solutions
`from openings that were irregular in shape as a result of polishing
`the electrospray end as well as from more uniform trapezoidal
`geometries of the channels themselves, indicating that the exact
`profile at these micrometer-scale dimensions are not critical. This
`may be because the liquid droplet as it emerges from the channel
`forms the tip itself from which the electrospray is established.
`Methods for spatially confining the droplet prior to the onset of
`electrospray may lower the onset potential. No particular differ-
`ences where noted, however, when the exit end of the microchip
`was silanized to increase hydrophobicity. To preserve column
`efficiency as the fluid profile moves from electroosmotic flow in
`the primary channel to laminar flow below the side-arm intersec-
`tion, the distance between the side-arm channel and electrospray
`exit should be minimized. This can be easily accomplished by
`breaking the microchip in the proper location or machining the
`side channel close to the exit. Overall, the interface has been
`shown to provide a stable electrospray and is simple and adaptable
`in design. For example, we are presently examining a microchip
`format that provides bulk fluid for electrospray when coated
`separation channels that have insufficient electroosmotic flow to
`sustain an electrospray are used. With the demonstration that
`microchips can be easily interfaced with mass spectrometry, the
`advantages afforded by the devices,
`including the ease of
`manipulating small sample volumes, the ease of
`integrating
`complex chemical reactions and separations, the high degree of
`automation and precision, and the potential for rapid screening
`resulting from construction of massively parallel analytical sys-
`tems, become more readily available for MS applications.
`
`ACKNOWLEDGMENT
`This research was sponsored by the Department of Energy
`Office of Research Development. Oak Ridge National Laboratory
`is managed by Lockheed Martin Energy Research Corp. for the
`U.S. Department of Energy under Contract DE-AC05-96OR22464.
`Technical discussions and assistance by S. C. Jacobson and S. A.
`McLuckey are greatfully acknowledged.
`
`review October 17, 1996.
`Received for
`December 16, 1996.X
`
`Accepted
`
`AC9610671
`
`X Abstract published in Advance ACS Abstracts, February 15, 1997.
`
`Figure 5. Photomicrograph of the Taylor cone and electrospray
`generated from a 60% water/40% methanol solution that was
`electroosmotically pumped using a coated side-arm microchip.
`
`Figure 6. Mass spectrum obtained from electroosmotically pumped
`10 (cid:237)M tetrabutylammonium iodide (in 60% water/40% methanol),
`averaging the data over a 2 s period and consuming (cid:24)30 fmol of
`analyte. Signal intensity is shown in the upper right-hand corner.
`
`cluster ions. Other major ions, which are not identified, may be
`due to fragmentation of the analyte. A heated interface, which
`would promote solvent evaporation for the electrospray droplets,
`would be expected to provide more intense (M - I)+ ions. The
`variation in total ion counts scanning over an m/z range from 50
`to 650 for a 4.5 min interval was low (5.9% RSD for a total of 405
`scans), indicating the overall stability of the electrospray source.
`
`CONCLUSIONS
`The microchip electrospray interface developed is simple in
`that it does not require any major modification of the microchip
`itself such as the attachment or machining of a tip at the channel
`exit. Electrical connections to generate the ES and drive an
`electrophoretic separation in microchip electrophoresis applica-
`tions are made via solutions that contact the primary channel from
`
`(34) Kriger, M. S.; Cook, K. D.; Ramsey, R. S. Anal. Chem. 1995, 67, 385.
`(35) Huang, M.; Vorkink, W. P.; Lee, M. L. J. Microcolumn Sep. 1992, 4, 233.
`
`1178 AnalyticalChemistry,Vol.69,No.6,March15,1997
`
`Agilent Exhibit 1268
`Page 5 of 5
`
`