Miniaturization and Geometry Optimization of a Polymer-Based

Anal. Chem. 2007, 79, 8076-8082

Miniaturization and Geometry Optimization of a Polymer-Based Rectilinear Ion Trap Miriam Fico,† Meng Yu,‡ Zheng Ouyang,† R. Graham Cooks,† and William J. Chappell*,‡

Department of Chemistry and School of Computer & Electrical Engineering, Purdue University, West Lafayette, Indiana 47907

The fabrication, operation, and characterization of a polymer-based rectilinear ion trap mass analyzer is discussed. A novel, fast prototyping technique, stereolithography (SLA)-based fabrication, traditionally reserved for end use production parts and to fabricate master molds for rubber products, is applied here as a tool to create precise, arbitrary geometries. Taking full advantage of the SLA methodology, an open corner, polymer-based ion trap has been fabricated and tested. The use of a custom resin, Nanoform 15120 (DSM Somos, New Castle, DE), has resulted in a polymer with high heat deflection temperature and greater structural stability at higher temperatures and lower capacitance. The mass analyzer was mounted in a polymer holder and tested in a custom vacuum system using modified LCQ Duo (Thermo Fisher Corp.) electronics. The resolution, mass/charge range, and MS/ MS capabilities were examined using electrospray ionization as well as atmospheric pressure chemical ionization. In the course of this study, three traps of different sizes were fabricated, beginning with a “full size” device measuring 10 × 8 × 50 mm. The next two traps were scaled down by linear factors of a half and a third. SLA is shown to allow fabrication of light, small rectilinear ion traps, which are less expensive and have the same performance as traditional machined devices of the same size. In addition, smaller traps can be built just as easily, and they show unit mass resolution to mass 300, tandem mass spectrometry capabilities, and low power consumption.

pole7) has rapidly become an active area of instrumentation science. Ion trap mass analyzers have particular advantages when miniaturized, because the required power drops rapidly with size while this type of analyzer can be operated at relatively high pressure. The reduced vacuum demands arise from reduction in the ion’s mean free path, which is affected by a number of factors including size, rf frequency, and rf amplitude. Additionally, the decrease in the dimensions of the ion trap leads to a reduction in the operating voltages, ensuring that operation at higher pressures will not lead to electrical discharge. Because of these and related considerations, miniaturized ion trap mass spectrometer systems on the 10-kg scale have been introduced commercially.8 A commercial line of portable gas chromatography/mass spectrometer systems based on cylindrical ion trap (CIT) (Griffin Analytical Technologies, L.L.C, West Lafayette, IN) and quadrupole (Constellation Technology, Largo, FL) technology have been developed. MKS Instruments, Inc. (Wilmington, MA) has released a CIT-based system for leak detection. Additional companies have also developed portable mass spectrometer systems based on quadrupole (Microsaic Systems, Surrey, United Kingdom) and toroidal (Torion Technologies, Inc., Pleasant Grove, UT) ion trap technologies. The rectilinear ion trap9 (RIT) geometry has been developed to take advantage of two features of earlier types of mass analyzers: the large storage capacity of the linear quadrupole ion trap10 and the overall design simplicity of the cylindrical ion trap.11 The simplicity achieved in the design of the device (four12 or six13 planar electrodes) makes the rectilinear ion trap an excellent

Development of fieldable instrumentation is of growing interest in analytical chemistry as an increasing number of applications from airport security to environmental and food safety monitoring requires more rapid and accurate analysis. As an analytical tool, mass spectrometry generates both highly specific and sensitive results in a short period of time; however, the transfer of the technology from the laboratory to the field has been limited by the complex and delicate nature of the instrumentation and by the power required for operation. In response to this, research on miniature mass analyzers of almost every kind (quadrupole ion trap,1,2 time-of-flight,3,4 magnetic sector,6 and linear quadru-

(2) Kornienko, O.; Reilly, P. T. A.; Whitten, W. B.; Ramsey, J. M. Rapid Commun. Mass Spectrom. 1999, 13, 50-53. (3) Berkout, V.; Cotter, R. J.; Segers, D. J. J. Am. Soc. Mass Spectrom. 2001, 12, 641-647. (4) Cotter, R.; Fancher, C.; Cornish, T. J. Mass Spectrom. 1999, 34, 13681372. (5) Prieto, M. C.; Kovtoun, V. V.; Cotter, R. J. J. Mass Spectrom. 2002, 37, 1158-1162. (6) Diaz, J.; Giese, C.; Gentry, W. J. Am. Soc. Mass Spectrom. 2001, 12, 619632. (7) Orient, O.; Chutjian, A.; Garkanian, V. Rev. Sci. Instrum. 1997, 68, 13931397. (8) Keil, A. D.; Hernandez, H. H.; Noll, R. J.; Fico, M.; Gao, L.; Ouyang, Z.; Cooks, R. G. Anal. Chem. submitted. (9) Ouyang, Z.; Cooks, R. G. U.S. Patent 6,838,666, 2005. (10) Douglas, D. J.; Frank, A. J.; Mao, D. Mass Spectrom. Rev. 2005, 24, 1-29. (11) Langmuir, D. B. U.S. Patent 3,065,640, 1962. (12) Song, Y.; Wu, G.; Song, Q.; Cooks, R. G.; Ouyang, Z.; Plass, W. R. J. Am. Soc. Mass Spectrom. 2006, 17, 631-639. (13) Ouyang, Z.; Wu, G.; Song, Y.; Li, H.; Plass, W. R.; Cooks, R. G. Anal. Chem. 2004, 76, 4595-4605.

* To whom correspondence should be addressed. Telephone: 765-494-6225. Fax: 765-494-6440. E-mail: [email protected]. † Department of Chemistry. ‡ School of Computer & Electrical Engineering. (1) Badman, E.; Cooks, R. G. J. Mass Spectrom. 2000, 35, 659-671.

8076 Analytical Chemistry, Vol. 79, No. 21, November 1, 2007

10.1021/ac0711384 CCC: $37.00

© 2007 American Chemical Society Published on Web 10/04/2007

candidate for miniaturization. Different methods of fabrication have been investigated to create smaller high-precision ion traps, including semiconductor fabrication processes on silicon wafers 14-16 and low-temperature cofired ceramic17 fabrication. The stereolithography (SLA) methodology allows access to arbitrary geometries, opening the possibility of generating complex threedimensional ion trap structures. A polymer-based prototype ion trap has been operated successfully as a mass analyzer;18 however, the thermal sensitivity of the polymer used was problematic and optimal performance was not obtained. Preliminary characterization of this earlier instrument did not include the important tandem mass spectrometry experiments needed in a fieldable instrument in which complex mixtures are to be characterized without chromatographic separation. The availability of a resin with uniquely favorable properties (Nanoform 15120, DSM Somos, New Castle, DE), allowed an overall redesign of the original polymer-based ion trap geometry, including key issues related to the heating and cumulative charge build-up which were of most concern. The original design of the polymer-based ion trap featured closed corners that left a significant amount of exposed insulator inside the trap, allowing for charge accumulation. In addition, strong rf fringing fields present in the corners of the trap were found to lead to heating and deformation of the polymer, resulting in failures in trap operation. In this study, RITs9 are reduced to one-half and one-third of the original size13 to test the capability of using SLA in the fabrication of small, high precision mass analyzers. In addition, the closed corners are removed and the use of a superior resin minimizes heating and deformation problems. Both analytical and operational comparisons are made between the three sizes of polymer device. An important question is the viability of a scaleddown device in meeting the power reduction requirements of a portable instrument while simultaneously satisfying the analytical performance imperatives. EXPERIMENTAL Stereolithography Fabrication The stereolithographic fabrication process used to produce ion traps began with a three-dimensional drawing of the ion trap created in a computer aided design program such as AutoCAD 2002.19 Two-dimensional cross sections, approximately 50 µm thick, were obtained by slicing the three-dimensional structure in planes orthogonal to its z-axis. These cross sections were input as features of interest then a UV laser rastered the corresponding pattern across the surface of a photosensitive resin, realizing the structure in a layer-by-layer process. Subsequent metallization of the electrodes of the resulting 3-D structure was accomplished by first inserting a mask within the structure, selectively shielding portions of the trap from the sprayed silver ink (Heraeus Inc.). (14) Pau, S.; Pai, C. S.; Low, Y. L.; Moxom, J.; Reilly, P. T. A.; Whitten, W. B.; Ramsey, J. M. Phys. Rev. Lett. 2006, 96, 120801-120804. (15) Geear, M.; Syms, R.; Wright, S.; Holmes, A. J. Microelectromech. Syst. 2005, 14, 1156-1166. (16) Cruz, D.; Chang, J. P.; Fico, M.; Guymon, A. J.; Austin, D. E.; Blain, M. G. Rev. Sci. Instrum. 2007, 78, 015107-015109. (17) Chaudhary, A.; van Amerom, F. H. W.; Short, R. T.; Bhansali, S. Int. J. Mass Spectrom. 2006, 251, 32-39. (18) Yu, M.; Fico, M.; Kothari, S.; Ouyang, Z.; Chappell, W. J. IEEE Sens. J. 2006, 6, 1429-1434. (19) Autodesk, I., 2002 ed.; Autodesk, Inc: San Rafael, CA, 2001.

The process was completed by copper electroplating to yield individual conductive electrodes. The selection of the type of polymer used as the skeleton of the trap has important consequences for its ultimate performance. When a polymer is exposed to a microwave field, it will absorb energy from the field and this is translated into an increase in temperature of the material. Upon exposure to a strong enough rf field, the polymer itself may begin to deform and even melt as a result of the extreme temperatures. One of the characteristics used to describe polymers is the heat deflection temperature (HDT), which is the highest temperature that the polymer structure may withstand before deforming under a specified load.20 In these studies, Nanoform 15120 (DSM Somos, New Castle, DE) polymer was used as the basis for the device and after undergoing both the UV and thermal post cure processes, its HDT is 269 °C at 0.455 MPa. It is expected, given the high HDT value of this material, that it will be able to withstand the increases in temperature as a result of rf heating without sacrificing geometrical precision. In comparison, a material used previously, Accura SI 10 (3D Systems, Valencia, CA), has an HDT of 56 °C at 0.455 MPa with the standard 90-minute UV cure process. RIT Mass Analyzer Electrode Geometry The geometry of the RIT electrodes used in these experiments was based on that previously reported13 but modified as described above and carefully optimized. Three different traps of incrementally smaller sizes were constructed such that the geometrical proportions between the x and y electrode dimensions were maintained. The full-size device (Figure 1) has dimensions of x0 ) 5 mm, y0 ) 4 mm, z0 ) 50 mm (referred to as 5 × 4 mm for the rest of the paper). Note that x0 and y0 are dimensions measured from the center line of the trap, but z0 is the full length. The alignment and placement of the four electrodes are constrained by the stereolithography (SLA) process as the full-size trap was fabricated in a monolithic fashion with built-in mounting structures on either end (Figure 1a). Two polycarbonate holders were press fitted onto either end of the piece to mount the custom-machined stainless steel end-cap electrodes, to make electrical connections to the RIT electrodes and to allow mounting of the trap itself along the ion optical rail. Two additional traps were tested as a part of this study, their dimensions in relation to the full-size device were half sized (x0 ) 2.5 mm, y0 ) 2 mm, z0 ) 25 mm (referred to as 2.5 × 2 mm for the rest of the paper)) and third sized (x0 ) 1.66 mm, y0 ) 1.33 mm, z0 ) 16.66 mm (referred to as 1.66 × 1.33 mm). In order to achieve uniform coverage over the smaller devices during the metallization step, a two-piece assembly was constructed (Figure 2). The electrodes were inserted into the corresponding spaces of the end supports, thus ensuring alignment within the RIT. In the same manner as for the full-size trap, two polycarbonate holders were used to mount the end-cap electrodes and make electrical connections. While this is assembled in two pieces, the fact that the features are laser-defined allows for precise definition of the ion traps. Electrical Characteristics. Electrical characteristics (Table 1), including capacitance and breakdown potential, were determined for each trap prior to operation. The capacitance of the (20) ASTMD648-98c; ASTM International: West Conshohocken, PA, 2000; Vol. 08.01.

Analytical Chemistry, Vol. 79, No. 21, November 1, 2007

8077

Table 1. Electronic Characteristics of the RITs Studied

Figure 1. (a) Three-dimensional Inventor Professional 11 (Autodesk, Inc.) drawing of the x0 ) 5 mm, y0 ) 4 mm, z0 ) 50 mm polymer-based RIT. Each x electrode features an ejection slit, which is 29 mm long and 1 mm wide. (b) Two-dimensional cross section of the same trap, defining the x0 and y0 dimensions.

RIT size (x0 mm, y0 mm)

capacitance (pF)

operating frequency (MHz)

breakdown potential (kV)

stainless steel (5, 4) polymer (5, 4) polymer (2.5, 2) polymer (1.66, 1.33)

29

0.730

2.0

6.5

0.932

2.0

5.7

0.974

2.0

4.1

0.962

2.0

For each trap, the breakdown potential was measured once the trap was mounted in the instrument and the manifold reached a base pressure of at least 1 × 10 - 5 Torr. A high-voltage supply was connected to the y electrodes, while the x electrodes and the end-cap electrodes were grounded to the manifold. The applied dc voltage on the y electrodes was raised in 100-V increments while the current passing from the electrodes to ground was monitored. This step was repeated until one of two conditions is met: (1) the current passing to ground was greater than or equal to 15 µA or (2) the applied potential was greater than 2.0 kV. These conditions were used as safeguards to protect the electronics, the 2.0-kV value since it represents the maximum voltage required for the desired mass range in these experiments. The amplification of the rf voltage was achieved by creating an LC tank circuit from the inductance of the secondary winding of the rf transformer and the capacitance of the trap. According to (1), the resonant frequency of the circuit will vary with the

f)

1 x 2π LC

(1)

inductance and capacitance in the usual fashion. Throughout the experiments, the coil inductance remains constant; therefore, only the capacitance of the trap affects the resonant frequency. The governing fundamentals for this mass analyzer are described by the Mathieu parameters. The mass analysis eq 2 can be written21 as follows:

m ) z q Figure 2. (a) Two halves of the x0 ) 2.5 mm, y0 ) 2 mm, z0 ) 25 mm ion trap prior to final assembly with a penny for scale. The alignment hole is highlighted to depict where the electrodes are inserted. (b) Final assembly, clearly indicating where the nonslit electrodes are inserted and aligned.

trap was measured while the RIT was at atmospheric pressure, mounted in its polycarbonate holder with all electrical connections made. A LCR meter (4192A LF impedance analyzer, HewlettPackard) was connected to the trap to measure the capacitance (at 1 MHz, the nominal operating frequency of RITs13) between the x and y electrodes. The end caps were held at ground potential relative to the meter. The breakdown potential limits the maximum potential that may be applied between the electrodes. In the previously reported monolithic polymer-based ion trap system, the breakdown potential of 0.8 kV limited the mass range to approximately m/z 200 at an operating frequency of 1.20 MHz. 8078


8V0-P 2 2 ejectΩ (x0

+ y02)

(2)

where m/z is the mass to charge ratio of the ion of interest, V is the zero to peak value of the rf voltage, qeject is the value of the dimensionless Mathieu parameter qz when the ion is ejected, Ω is the angular frequency of the rf, and x0 + y0 are the dimensions of the RIT. Of considerable interest is the inverse relationship between the dimensions of the trap and the applied voltage, indicating that as the size decreases so does the power required for operation, in the case where all other parameters remain the same. Instrument Description. The RIT analyzer being tested is mounted in a single polycarbonate holder, designed (Autodesk Inventor Professional 11, Autodesk, Inc.) and machined in-house. (21) March, R. E.; Todd, J. F. J. Practical Aspects of Ion Trap Mass Spectrometry; CRC Press: Boca Raton, FL. 1995.

Figure 3. Two-dimensional representation of the ion optics and mass analyzer configuration, indicating the pressure regimes in each vacuum stage and typical operating voltages.

The end support for each of the three fabricated RITs was designed to be press fitted into the polycarbonate holder, thus ensuring alignment between the traps and the ion beam path as well as making electrical connections to the RIT electrodes. The RIT was mounted in line with the rest of the ion optics via four mounting screws, which affixed the polycarbonate mount to the octapole holder (Figure 3). Helium buffer gas was introduced directly into the trap via a through hole in the polycarbonate holder. The pumping system, vacuum manifold, and atmospheric interface of a triple-quadrupole model TSQ 7000 (Thermo Fisher Corp.) were modified and utilized together with an octapole ion guide from a TSQ 700 (Thermo Fisher Corp.) to complete the remaining elements of the mass spectrometer system. A single DeTech 397 (Scientific Instrument Services, Ringoes, NJ) conversion dynode/electron multiplier combination was used to monitor the ejected ions. Balanced, two-phase rf voltages were applied to the x and y electrodes of the RIT via a specially constructed coil. Modified LCQ Duo electronics and ion trap control language (Thermo Fisher Corp.) were used to operate the instrument. The vacuum manifold design included an atmospheric interface, which allowed access to a number of ionization techniques; specifically, electrospray ionization (ESI) and atmospheric pressure chemical ionization (APCI) are two examples that were explored in this work. Preliminary experiments were conducted using ESI, which offers access to a number of high-mass, nonvolatile analytes,22 allowing for the characterization of the mass range and resolution of the trap. The electrospray source was home-built using fused-silica capillary (ChromTech, Apple Valley, MN) for the inner (analyte solution) and outer (nitrogen sheath gas) capillaries, which were held in place with a 1/16-in. Swagelok tee (Indiana Fluid Systems, Indianapolis, IN). Flow of sample solution into the capillary was controlled by a syringe pump (PHD 2000, Harvard Apparatus, Inc., Holliston, MA) at a rate of 5.00-10.0 µL/min. Typically, a bias of 5.5 kV was applied directly to the needle using the ionization source high-voltage supply from the LCQ Duo. The nitrogen sheath gas pressure was regulated by a standard gas tank regulator to a typical pressure of 80 psi. The entire assembly was mounted on a ring stand with miniature finger clamps and then aligned with the instrument’s inlet capillary. (22) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Mass Spectrom. Rev. 1990, 9, 37-70.

Table 2. Operating Parameters of Ion Traps Tested

RIT size (x0 mm, y0 mm) stainless steel (5, 4) polymer (5, 4) polymer (2.5 2) polymer (1.66, 1.33)

rf V0-P at m/z 150

resonance ejection voltage (V0 -P)

end-cap dc bias (V)

optimal scan rate (amu/ charge‚s)

120

5.7

100

16,665

220

3.5

100

5,555

70

0.63

100

11,111

25

0.31

40

11,111

The APCI source was a simple corona discharge needle fabricated from a short length of 0.07-mm tungsten wire spot welded onto the end of stainless steel wire. The same high-voltage source used for the ESI experiments was then connected to the end of the wire, typically a bias of 3.5 kV was applied. RESULTS AND DISCUSSION Ion Trap Evaluation. After the electrical characterization tests were completed, the polymer-based RIT was mounted into the manifold and its performance established. A test solution of the amino acid arginine (Sigma Aldrich Corp., Milwaukee, WI) was prepared in a 49:49:2 methanol/water/acetic acid mixture and allowed to flow into the source at a rate of 5 µL/min. The sample was ionized by ESI. A common set of ion optical parameters was uploaded to the instrument control boards for testing all three polymer traps and the 5 × 4 mm stainless steel system.The only remaining variables are the operating voltages on the trap itself Table 2. Once a mass spectrum is acquired, the voltage scan rate is calibrated using the characteristic23 protonated cluster peaks of arginine. A typical spectrum of arginine acquired from the 5 × 4 mm polymer-based trap is depicted in Figure 4. Figures of merit, such as mass range and resolution, are easily determined using the arginine clusters as points of comparison between the traps tested. Atmospheric Pressure Chemical Ionization Data. APCI is a widely used atmospheric pressure ionization technique characterized by its simplicity and speed, as well as a wide range of (23) Zhang, D.; Wu, L.; Koch, K. J.; Cooks, R. G. Eur. Mass Spectrom 1999, 5, 353.


8079

Figure 4. ESI mass spectrum of the amino acid, arginine, under typical operating conditions. The series of protonated clusters was used for mass calibration and comparison between the differently sized traps.

Figure 5. APCI mass spectrum of the CW simulant, methyl salicylate, under typical operating conditions.

applications.24 Typical APCI systems use an electrical discharge to form reagent ions from the reagent gas, which then ionize the analyte through ion/molecule reactions.25 In this experiment, protonated water clusters are usually formed and they in turn transfer a proton to the analyte molecule during chemical ionization. Several compounds were successfully analyzed in this manner, including the chemical warfare agent stimulant, methyl salicylate. Figure 5 depicts the APCI mass spectrum recorded when the saturated headspace vapor was analyzed using the 2.5 × 2 mm trap. The m/z 153 peak corresponds to the protonated molecule, while the peak at m/z 121 corresponds to the fragment generated by the characteristic loss of methanol. Resolution. For each of the tested ion traps, the maximum resolution was determined by optimizing the trapping and ejection parameters; the amplitude of the resonance ejection voltage was found to exert the greatest influence on peak shape. For all three traps, the protonated cluster ions of arginine, m/z 175 and 349, were used to compare the resolution. Figure 6 depicts regions of the mass spectra including these two ions from each of the three polymer-based RITs. All three traps were operated using a resonance ejection q value set at qz ) 0.83, whereas the amplitude of the resonance ejection signal was optimized to achieve maximum resolution for each device. The resolution (R) for the 5 × 4 mm trap was found to be 438 at m/z 175 and 696 at m/z 349 (determined by the full width half-maximum (fwhm) method), where the axial ejection voltage was 3.5 V0-P. By comparison, for the 2.5 × 2 mm trap, R ) 110 at m/z 175 and R ) 142 at m/z 349, where the axial ejection voltage was 0.63 V0-P. The 1.66 × 1.33 mm trap exhibited similar characteristics with R ) 116 at m/z (24) Abian, J. J. Mass Spectrom. 1999, 34, 157-168. (25) Horning, E. C.; Horning, M. G.; Carroll, D. I.; Dzidic, I.; Stillwell, R. N. Anal. Chem. 1973, 45, 936-943.

8080


175 and R ) 148 at m/z 349, and the axial ejection voltage was 0.31 V0-P. MS/MS Data. RITs can be used to take full advantage of the intrinsic increase in specificity realized through tandem mass spectrometry, as has been previously shown.26 This technique generates richer data sets and allows the user to confirm the identity of an analyte in the presence of a matrix27 as well as to determine or confirm molecular structural characteristics.28 Experiments were undertaken to demonstrate this capability on the 5 × 4 mm polymer-based RIT. The trapping conditions remain the same as described above, while a notched waveform29 was used to isolate the ion of interest. The sequential study of the fragmentation of the protonated reserpine ion is depicted in Figure 7. A full spectrum of a 5 mM solution of reserpine in the previously described methanol/water solution was acquired using ESI. Then, the parent ion was isolated (Figure 7b) at qz ) 0.83 and subsequently activated using a resonance excitation signal applied to the x electrodes. The product ion spectrum (Figure 7c) was recorded after collision-induced dissociation with He buffer gas with an efficiency of ∼76%. Here efficiency30 is defined as the ratio of the total fragment intensity to the parent ion intensity prior to fragmentation. Performance of Miniature Traps. As already noted, a decrease in the size of an ion trap mass analyzer will lead to a reduction in the power consumption of the components used to control the device. However, this characteristic is only observed (26) Song, Q.; Kothari, S.; Senko, M. A.; Schwartz, J. C.; Amy, J. W.; Stafford, G. C.; Cooks, R. G.; Ouyang, Z. Anal. Chem. 2006, 78, 718-725. (27) Kondrat, R. W.; Cooks, R. G. Anal. Chem. 1978, 50, A81. (28) deHoffmann, E. J. Mass Spectrom. 1996, 31, 129-137. (29) Louris, J. N.; Taylor, D. M. U.S. Patent 5324939, 1993. (30) Yost, R. A.; Enke, C. G.; McGilvery, D. C.; Smith, D.; Morrison, J. D. Int. J. Mass Spectrom. Ion Phys. 1979, 30, 127-136.

Figure 6. Partial mass spectra of the m/z 175 and 349 protonated clusters of arginine compared across the three different traps. (a) The 5 × 4 mm RIT has resolution of R ) 438 (fwhm definition) for m/z 175 and R ) 696 for m/z 349. (b) The 2.5 × 2 mm RIT has resolution of R ) 110 for m/z 175 and R ) 142 for m/z 349. (c) The 1.66 × 1.33 mm RIT has resolution of R ) 116 for m/z 175 and R ) 138 for m/z 349.

Figure 7. Tandem mass spectrometry experiments performed using the 5 × 4 mm polymer-based ion trap. (a) Full mass spectrum of protonated reserpine (m/z 609) recorded in ESI. (b) Isolation of m/z 609. (c) Product ion spectrum recorded after collision-induced dissociation of the parent species with background He gas at an efficiency of 76%.

when all other parameters listed in eq 2 remain constant. In the series of experiments discussed here, this condition was approximately satisfied because the operating frequencies for the three devices are close in value to each other. The parameters described in Table 2 clearly demonstrate reduced power con-

sumption as evidenced by the correlative decrease in rf levels and trap size. Each of the three traps is operated at nominally the same frequency; therefore, the power consumption is proportional to the square of the voltage. Therefore, each of the smaller traps requires only a fraction of the rf power consumed by its full-size Analytical Chemistry, Vol. 79, No. 21, November 1, 2007

8081

counterpart (the 2.5 × 2 mm uses 10% and the 1.66 × 1.33 mm uses 1%). From the Mathieu parameters, it is expected that when the size of the device is reduced, the operating frequency and scan rate must be adjusted accordingly in order to maintain adequate resolution. However, for the applications to which these traps would be most useful, unit resolution is adequate. Therefore, a reduction in power consumption is much more advantageous than maintaining the resolution normally obtained from bulky bench top instruments, at least for applications to small molecules. Comparison of Performance of Polymer and Stainless Steel Ion Traps. The figures of merit of the 5 × 4 mm stainless steel RIT and the 5 × 4 mm polymer-based ion trap were compared to judge their capacity to perform mass analysis over a wide mass range and to perform tandem mass spectrometry. There were no significant functional differences between the two devices, aside from the differences in capacitances. The maximum operating voltages, which directly translate to the mass range of the analyzer, were identical for both the polymer and stainless steel models. Both devices were used to successfully analyze ions created by APCI and ESI, showing a versatility that should allow their application to other atmospheric pressure ionization sources. Intrinsically, SLA allows for greater design complexity and reduced fabrication times than computer numerical controlled (CNC) machining, which requires extensive setup and process planning.31 On the other hand, the fabrication precision of CNC machining and SLA are roughly comparable although some very high precision conventional machining has been reported. The elimination of the assembly step required after CNC machining ensures a simpler, monolithic design, which provides increased interelectrode alignment accuracy. (31) Frank, M.; Joshi, S. B.; Wysk, R. A. J. Chin. Inst. Ind. Eng. 2003, 12, 240246.

8082


CONCLUSION The SLA technique has been successfully applied to the fabrication of ion trap mass analyzers, a process commonly achieved by traditional metal-machining techniques. Problems with thermal and dimensional instability of the polymer were avoided using a new high-quality resin circumventing earlier problems with heating and charge buildup. The implications of the success of this project lead in the direction of incorporation of other SLA fabricated pieces into the mass spectrometer, to allow for further reductions in weight and access to arbitrary geometries. The savings realized in the form of reduced power consumption and capability for fast prototyping are noteworthy. The half- and third-size traps both demonstrated significantly lower operating voltages (at nominally the same operating frequency) in comparison to their full-size counterpart. As a conduit toward a miniaturized mass spectrometer, clearly, the results of this study indicate that the polymer-based devices represent a viable alternative to other techniques. Additionally, as a fast prototyping technique, many iterations of a design may be tested within the time it would take to machine a comparable device. The SLA process appears to have promise for bridging the gap between full-size and miniaturized ion trap components. ACKNOWLEDGMENT This work was supported by the National Science Foundation (0528948-ECS). M.F. acknowledges fellowship support from the Department of Education Graduate Assistantships in Areas of National Need (GAANN) program.

Received for review May 30, 2007. Accepted July 26, 2007. AC0711384

Miniaturization and Geometry Optimization of a Polymer-Based

Recommend Documents