Home

Lab-on-a-Chip Diagnostic Biosensor

Contents

1. run5 folder Data 4 8 LEAC COUPLING LT110401 2 4 8 WG4 Run5 LEAR in Lab runs title Run 5 LEAR in Lab run get_run_data run5 folder nch_scan cal run5 title rung folder Data 4 8 LEAC COUPLING LT110401 2 4 8 WG4 Run6 LEAR in Lab recouple run6 title Run 6 LEAR in Lab recouple run 2 get_run_data run6o folder nch_scan cal run6 title run folder Data 4 8 LEAC COUPLING LT110401 2 4 8 WG4 Run7 LEAR in Lab play with polarizer run title Run 7 LEAR in Lab play with polarizer run 3 get_run_data run7 folder nch_scan cal run7 title run8 folder Data 4 8 LEAC COUPLING LT110401 2 4 8 WG4 Run8 rung title Run 8 run 4 get_run_data run8 folder nch_scan cal run8 title run9 folder Data 4 8 LEAC COUPLING LT110401 2 4 8 WG4 Run9 run9 title Run9 run 5 get_run_data run9 folder nch_scan cal run9 title case 1 sfull run collected on 4 9 ch_scan 1 31 nch_ scan length ch_scan full run Zolder Data 4 9 LEAC 4 9 m full_run 123 Page full_run_title Full Run run case 2 get_run_data full_ run _folder nch_scan cal full run title Sother data collected on 4 9 1 ch_scan nch_scan first _folde first title run less _scatte less _scatte run 2 get_run_data less__ case 3 sdata after 1 ch_scan nch_ scan nl folder nl title n n2_ folder n2_ title n 2 run3 folder n3_
2. 1lxn array of structures each containing data for one run ields in run structure are deil cell array of dark currents from dark1CSV grouped into 4x number of samples per channel matricies different cycles dezi cell array of dark currents from dark2CSv Les cell array of light currents from light1CSV pes cell array of the photocurrents obtained First matrix is hotocurrents obtained when subtracting the last run of dark1CSV from the irst run of light1CSV Second matrix is photocurrents obtained when ubtracting the first run of dark2CSV from the last run of light1CSV v corresponding voltages for c field an ho BWNHEE HSK CO DECHE run 3 lv 2 refers to the second set of light voltages found in he file light1Csv H nput variable data to plot determines which data is plotted see switch AP oP d d X d X d d AP dp d oP P d CP AP dp d a case block below determines whether to plot data as well as organize it If data is splotted the results are saved in the same folder as the data run_plots 0 122 Page sfrequently used variables fcount 1 sfigure count cal gain 0 10313 gain in V uA cal offset 3 9941 szeo offset in V colors vector b g r k m e Scolors vector for plotting scollect and orgainze data switch data_to_plot case 0 data collected on 4 8 runs 5 9 ch_scan 18 31 nch_scan length ch_scan
3. 23 Page upward displacement of the syringe plunger The pump used for active flow had a range of 5 200 um s Equation six was derived using the ideal gas law and syringe geometry to relate the change in syringe pressure to the pump s speed setting x which can then be used with equation five to relate the pump s speed setting to volumetric flow of the fluid _ nRT syringe aD Ax 4 6 Analysis of the syringe pump velocity and volumetric flow rate using the above equations generates an approximation that follows a linear relationship The fluid flow rate can overcome viscous and frictional resistances at the same rate as the pressure difference is pulling at the lowest possible syringe plunger speed which was 5 um second Even at this lowest setting some backpressure builds up as was confirmed experimentally When the syringe pump was stopped fluid continued to flow because there was still an unequilibrated systemic pressure difference C Fluid Flow and Leaking Using both large fluorescing particles and a solution of polyclonal IgG antibody fluorescently labeled with Alexa fluor 488 fluid flow was confirmed and channel leaking was investigated The fluid flow imaging was one of the first tests conducted to confirm that the channels created by the new mask design were functional under capillary flow Figure IV 5 is a still shot removed from a video of moving fluorescent particles in PDMS channels bound to a glass sli
4. ceeeeees 68 Table of Figures Appendix Figure A 1 Diagram showing the interdisciplinary team dynamic of the LEAC prolect 74 Figure A 2 Gantt chart of Spring Semester 75 Figure A 3 PDMS Mask design drawn in AS 77 Figure A 5 LEAC Chip SCHEMAtIC ese 56s oss ches 04 oa cb oooh ca chou cb ca dea ob Saad a oa beach cach de ckcych a dEd cenchens 78 Figure A 6 Simplified 2D diagram of adjustments made to configuration 1 necessary to measure WG 1 4 5 6 left and a photograph of the probe station in configuration 1 with the sample holder shown on the bottom right fiber launch on the bottom left and probe ca 79 Figure A 7 Block diagram of configuration 2 right and photograph of the probe station in configuration 2 1 EE 80 Figure A 8 Block diagram of configuration 3 81 Figure A 9 Coordinate definition for alignment instructions with the lucky testing dinosaur Lenny EE 81 Figure A 10 Probe card as received from Alpha Probe ss ssssssssssssssssssssssssessrerssrrssressrtnstersrrrseressrensres 85 Figure A 11 Image showing the probe card probes being cleaned and the resulting marks 87 Figure A 12 Circuit schematic for the multiplexing and amplification functions cccceeeeteeees 90 Figure A 13 Top traces Of DCH een 91 Figure A 14 Bottom traces Of DCH 91 Figure A 15 BOM of MUX AMP Circuit c ccccscccesscccssscesssecesseecessecessseceeseecesssecssaeeesssecesseeeeseeessa
5. Microfluidics The microfluidics mask design provided fully functional fluid flow for both passive and active methods although finding a solution to why fluid will not flow passively on a chip remains to be investigated As described previously there were complications in achieving a permanent bond with plasma oxidation on the most recent working chips for reasons not completely understood It is possible that obtaining a better understanding of the differences in surface chemistry between the old and new chips would explain this phenomenon In the future having a new mask design with either 0 8 mm reservoirs or with the reservoirs completely removed would be convenient to have as different fluid flow methods require different sized punches For example active flow using 63 Page PEAK tubing requires a 0 8 mm punch on each end of the channel while passive flow requires a 1 5 mm inlet reservoir and a 1 0 mm outlet reservoir Another potential area for future work in the design and integration of microfluidics lies in simulations It would be helpful to use a program like Fluent to simulate the flow patterns In addition in conjunction with the apparatus setup it would be useful to gain more insight into the temperature changes incurred by lengthy exposure to the microscope light during positioning and how long it subsequently takes for the device to return to room temperature In the distant future it will be necessary to devise a reconfi
6. direction by sliding the platen over the sample If necessary it is possible to achieve control to around 50 microns by partially locking or firmly holding the platen and tapping it gently into place Fine adjustments can then be made using the XY capability of the sample holder At this stage just get it close in the X direction and very close in the Y direction but give lots of clearance to the probe card until it is Y adjusted clear of PDMS It is common to not see the last one or two probes when the probe card is well above the sample surface which is in focus however one can usually judge where the probe tips are 83 Page 6 Make contact between probe card and sample Using the Z movement capability of the platen provided by the when on the front right of the probe station bring the probe card near the surface of the sample At first do this while watching with the naked eye from the level of the sample Once the probe card is roughly above the sample start watching through the microscope view on the computer and also watching the edge sensor indicators on the front of the circuit box The edge sensor lights will turn off when they make contact with the surface of the sample so this is an absolute indication of the status however with experience it is possible to see when contact is made by watching the motion of the probes through the microscope If one of the edge sensors indicates contact before the other then there is a dread
7. A LEAC Chip Layout The layout of the current LEAC chip design was designed by Rongjin Yan The design was chosen to allow for significant processing error in the cleanroom as well as to provide for a variety of waveguide and buried detector sizes and structures Having multiple waveguides and detectors types allowed for experimentation with numerous variables without having to order and process different mask designs each time The mask used to fabricate the chips was a 3 by 3 grid consisting of 9 unique designs as pictured in Figure III 1 gt CS CH gt q CS a oo e Ca o a p Ca o gt Oo Wa MAR lt DC ID OO 4 ii a0rgaaen Figure III 1 LEAC chip mask designs The bottom left sector of the mask contains the nine designs needed to create the buried detectors A zoomed in image of the buried detectors is shown below in Figure III 2 14 Page Figure II 2 Comparison of the single versus segmented buried detectors The two solid structures on the left produce a single buried detector structure beneath the waveguide of silicon once the fabrication is completed It is these single detectors that were used in the experiments conducted on the LEAC chips These detectors were selected because their layout prevents the waveguide from having a vertical s
8. Probe Card Specification and Implementation The LEAC chip has aluminum pads for contact as opposed to gold because high contact resistance is not a problem with the device therefore we chose to use industry standard tungsten probes However because tungsten rhenium probes are less fibrous than tungsten probes they are less prone to collecting oxides from the aluminum pads Tungsten rhenium probes could maintain ideal probing characteristics without having to be cleaned for long periods and they would also have more contact than tungsten only probes If a new probe card is ever required using tungsten rhenium probes should be considered An unforeseen problem was encountered when the probe card was integrated into the testing system Adding the PDMS microfluidics introduced an approximately 2 mm tall obstacle to the center of the chip In order to avoid the possibility of any part of the probe card contacting the PDMS we increased the height of the epoxy ring holding our probes from the standard 80 mils to 250 mils The left half of the epoxy ring that would be interior to the LEAC chip during probing was also removed as is illustrated in Figure 6 4 Another possibility that was initially considered was cutting the PCB of the probe card completely off halfway through the hole accommodating probes This option was discarded when the vendor informed us that this might damage the structural integrity of the card and possibly warp H These would be
9. SI Sp 69 SE FOE ot data obtained if run plots 1 r each run plots photocurrents photocurrents normalized by dark currents phtocurrents with error bar not yet implemented dark currents light currents noise of dark currents noise of light currents deltas between tests of dark currents deltas between tests of light currents run_plots run_count 1 length run Stake only last quarter of all currents found for count l length run run_count dcl dark currents count mean run run_count dcl count 2 end for count 1 length run run_count dc2 dark currents counttlength run run_count dcl mean run run_count dc2 count 2 plo o end for count 1l length run run_count 1c light_currents count mean run run_count lc count 2 end plot photocurrents from 2 seperate tests figure fcount fcount fcount 1 t ch_scan mean run run_count pc 1 2 ch_scan mean run run_count pc 2 2 legend Test 1 Test 2 title strcat Photocurrents from run run_count run_title xlabel Detector Number ylabel Currents uA axis 17 31 0 0241 Sused to get good comparison axis between tests on 4 8 125 saveas gcf strcat run run_count folder photo fig saveas gcf strcat run run_count folder photo jpg close Page splot normalized photocurrents figure fcount fcount fcount 1 plot ch_scan mean run run_count pc 1 2 mean ru
10. Ze d d fl Signal Vv 2 015 e 2 005 H H 100 200 300 500 600 700 Time s Figure A 25 Run 1 of the sucrose test Stems showing events during test 112 Page The blue stems shows the times when the tubing was switched from one solution to another There is an uncharacterized delay of when the tube is switched between solutions and when the new fluid actually reaches the LEAC chip This is something that needs to be determined in future experiments Run 2 had similarly confusing results e WG5 e ABCommon Pad used e 3V bias on common pad e Pump Speed of 40 e John noticed that the pump just stopped for some reason at some unknown time and John had to restart Unknown where dead time is Run 2 on LT110401 1 PDMS Sucrose experiment T T T TT T T T T T Sucrose Lei 1 44 4 Air 1 42 re ke 4 ay 1 4 Pump Start 0 ES 5 Ir al a Recouple Laser gt lt 1 14 4 tar L 1 li li L li 100 200 300 400 500 600 700 800 900 Time s Figure A 26 Run 2 of the sucrose test Stems showing events during test It can be seen in Figure A 26 that there is an obvious change in the photodetector current at about 550 seconds Itis not determined however whether this was caused from the change in solutions as there had already been multiple exchanges beforehand It is also of note that the pump appeared to stop midway so
11. configurations only allow access to two Shown below is a block diagram of the 4 set ups as well as a photograph of the hardware in place in this configuration Figure A 5 Simplified 2D diagram of adjustments made to configuration 1 necessary to measure WG 1 4 5 6 left and a photograph of the probe station in configuration I with the sample holder shown on the bottom right fiber launch on the bottom left and probe ca Configuration 2 Access WG 3 and 7 This configuration only allows access to two waveguides To convert to this configuration from configuration 1 leave the probe card in place and rotate the aluminum plate 90 counter clockwise Then move the sample holder back into a similar position as it was in configuration 1 These adjustments effectively move the fiber launch 90 degrees counter clockwise while leaving everything else the same The adjustment is done in this way because the fiber launch is bolted securely down while the sample holder is still free to 79 Page move A block diagram and photograph of configuration 2 are shown at the top of the next page Figure A 6 Block diagram of configuration 2 right and photograph of the probe station in configuration 2 left Configuration 3 Access WG 2 and 6 These two waveguides are the most precarious to access with the probe station Configuration three is essentially a mirror image of configuration 2 in terms of the sample holder and fiber launch while t
12. critical detection method made possible by the binding of secondary antibodies to the primary ones An antibody derived from the same type of host will bind to another antibody from that same type of host Thus the antigen antibody reactions can be fluoresced using a secondary antibody reaction like that shown in Figure VIII 1 For the purposes of our testing two distinct antibody antigen pairs were primarily used antigen ESAT 6 and its corresponding antibody a ESAT 6 and antigen AG 85 and its corresponding antibody a AG 85 Both of these are tuberculosis protein fragments The antibody a ESAT 6 is rabbit derived while antibody a AG 85 is mouse derived A solution of fluorescently labeled goat a rabbit rabbit derived and goat a mouse mouse derived secondary antibodies was created as a common detection solution for both antibody types B Chip Surface Treatment In order to make biomolecular testing possible on a LEAC chip or glass slide surface a chemical treatment process was applied This procedure was implemented on all glass slide silicon nitride and LEAC chip tests involving biomolecules 50 Page The first step of this process is a thorough cleaning of the surface to be treated Making sure there are no dust particles on the surface is imperative to the functionality of the treatment The next step silanizes the surfaces and is carried out using EDS N 2 aminoethyl 3 aminopropyl trimethoxysilane whose purchasing info ca
13. currents errors k title strcat Photocurrents from run run_count run_title dp d legend Errors found taking into account drift between light and dark measurements errors found without taking into account drift xlabel Detector Number ylabel Currents uA o tests on 4 8 saveas gcf strcat run run_count folder photo error fig saveas gcf strcat run run_count folder photo error jpg close splot light currents figure fcount fcount fcount 1 hold on for count 1l length run run_count 1c plot ch_scan mean run run_count lcf count 2 colors vector count end title strcat Light Currents from run run_count run_title xlabel Detector Number ylabeli saveas gcf strcat run run_count folder light fig saveas gcf strcat run run_count folder light jpg close Currents uA 127 Page axis LI7 32 E SI Sused to get good comparison axis between splot dark currents from both sets figure fcount fcount fcount 1 hold on for count l length run run_count dcl plot ch_scan mean run run_count dcl count 2 colors vector count end for count l length run run_count dc2 plot ch_scan mean run run_count dc2 count 2 colors vector count length run run_count dcl end title strceat Dark Currents from run run_count run_ title xlabel Detector Number ylabel Currents uA saveas gcf strcat run run
14. dark currents baseline mean run run_count dcl 1 2 for count 1 length run run_count dcl plot ch_scan mean run run_count dcl count 2 baseline colors vector count end for count l length run run_count dc2 plot ch_scan mean run run_count dc2 count 2 baseline colors vector countt length run run_count dcl end title strcat Deltas of Light Currents and Dark Currents from run run_count run_title Dark Current Deltas are Solid Light Current Deltas are Dashed xlabel Detector Number ylabel Currents uA saveas gcf streat run run_count folder deltas fig saveas gcf strcat run run_count folder deltas jpg close end end end function data_array break Intro cycles data nch_scan Sbreaks a data matrix which contains time series data for different channels and different cycles through channels into its different cycles Stores each seperate cycle as a subsequent entry in the cell array sdata_array Assumes different cycles are recorded in the rows following sthe previous cycle data_array cell 0 for count 1 size data 1 nch_scan 129 Page data_array end 1 data nch_scan count 1 1l nch_scan count end end function run data get un data folder nch scan cal run_ title SOrganizes the data from each run Method of organization is specified in Sswitch case loop in main function of this m file sdetermine data files da
15. described in step 7 You are now ready to take data on a LEAC sample using the probe station Closedown Procedure Once the experiment has concluded the system should be separated in the order that it came together First tubing should be removed from the PDMS The pump side will generally retain enough of a vacuum to instantly suck up fluid from that side however someone should be ready to blot fluids as the inlet side is removed from the PDMS Second the fiber should be moved back a distance that prevents the tip from being damaged Third translate the probe station upwards a generous distance to ensure clearance Unlock the platen and slide it back to resting position Last gently remove the sample from the sample holder using tweezers and a razor blade Be sure to clean the surface of the glass slide and check for residue on the underside of the sample Appendix E Probe Card Guide A Probe card overview and considerations for specifications and ordering This appendix is intended to serve as an introduction to probe cards and the terminology associated with using them Section VI outlines the requirements and selection of the probe card used with the LEAC chip Probe cards are commonly used to manually and automatically probe test wafers in industry The probe card is essentially a printed circuit board PCB with a circular hole which accommodates probes The PCB part of the probe card generally has traces from these probes to metal
16. desen 77 B Procedure for fabrication Of an SU 8 mod 77 Appendix D Probe Station Usage Instructions ssssssssssseesssreesssrersssrersssreeessrrenssseresssrresssrenssrrees 78 LEAC chip layout and Probe Station Configurations to access w veguldes 78 Configuration 1 Access WG 1 8 4 5 s sesssssssssssessssrirsssrersssrreessrrensssrensssrensssreenssrernsssrennssreenssseensse 79 Configuration 2 Access WG 3 and ie 79 Configuration 3 Access WG 2 and 80 Alignment Proceduren eebe AE 81 Appendix E Probe Card Guide 85 A Probe card overview and considerations for specifications and ordering ccssccceeesseeeeees 85 B Commonly used probe card parameters terminology and abbreviatons 88 4 Page Appendix F Multiplexing Amplification Circuit Schematic and BOM sss sssnssennssenseessresseneseeesseesseee 90 Appendix G LabVIEW Manual For Use e 93 A Front Panel avewinniteaicdiniinnnnnnnnnncnnnninnnnnnnnnnnncnnininnnnnncinnnnn 93 B AAT LL Ay tal Fe EES eg 96 Cx ee E Le 96 Appendix H EDS SMCC Surface Treatment Protocol cccsccccsssssccesssseccesssseeeessseeeesssnseeessssneeeeees 99 Appendix E Biomolecular Assay Procedure ccssccccsssssccesssssecesessssecesssusecesessueeesssuseeesssaseesessaeeess 102 EY 01 424 010 a0 eRe pee ee er ppp pe 102 Process ANC PrOCECUIe sccsserccssssssserccsserscsssccessccsssnccsssccessncsesscscssacsessnasessuecesaesesseeseseuccesanscons 102 Appendix J Budg
17. due to other variables as well Figure IX 8 illustrates such thermal drift In the test to which figure 8 corresponds dark currents were collected sequentially on each photodetector and then repeated Specifically dark currents were collected starting at photodetector 1 and running through photodetector 32 in each case The blue trace corresponds to the first test taken the green trace the second test taken the red trace the third test taken and the black trace the final test taken 144 Currents uA ln J 20 22 24 26 28 30 32 i Detector Number Figure IX 8 Sequential measurements of dark currents on LEAC chip demonstrating thermal drift 61 Page Readily apparent in Figure IX 8 is a decreasing trend in dark currents with respect to time It was determined that this phenomenon can likely be attributed to the temperature of the LEAC chip decreasing over the duration of the test due to cooling of the LEAC chip which is initially heated from use of the microscope light for alignment and laser coupling In the future users should attempt to minimize drift due to temperature by waiting lon enough for the chip to cool back down to room temperature after the system is setup Also to reduce the effects of testing variables such as temperature between when light and dark current measurements are taken to obtain a particular photocurrent some technique which allows for measuring of the dark and light currents closer te
18. equal to at least half the number of total detector pads available on the chip Other waveguides can then be probed by rotating the LEAC chip or probe card relative to each other Figure VI 2 illustrated this convenient geometry Because our probe station does not have the precise planarity and alignment found with probing stations in industry we decided to order a probe card with edge sensor capabilities to allow for determination of planarity discrepancy between the probe card and a LEAC chip These edge sensors are a combination of two probes which are normally connected together When one of them makes contact with a surface they yield an open circuit By hooking up LEDs to these edge sensors we were able to determine when contact occurs Since our probes are in a linear configuration we chose to use two edge sensors one on each side of the probe card which would enable easy detection of planarity errors for the full range of the probed area The position of the edge sensors on our probe card can be seen in Figure VI 2 and an edge sensor probe configuration can be seen in Figure VI 4 We chose to use non insulated edge sensors because they were considerably cheaper 15 compared to 35 for insulated sensors Because there are places on the LEAC chip where they could contact without causing interference in conducting regions this choice has no negative implications on the proper workings of the device 42 Page C Problems Encountered With
19. in several other ways 1 Microscope positioning To allow the microscope to stay in place above the sample and simultaneously stabilize the tubing guides were fabricated from small aluminum piping They were cut to length and bent to configure and attach to the single objective lens of the microscope The aluminum piping then tapers inwards guiding the tubing into the inlet and outlet holes of the PDMS This design allows the microscope to remain in position above the interface between the fiber and waveguide thereby enabling the necessary monitoring of the quality of laser coupling throughout the setup sequence and during data acquisition Figure V 7 depicts the new mechanical fixtures added to the objective lens as tubing guides Figure V 7 Mechanical fixtures on microscope objective One setback of this setup is that it required greater lengths of tubing to account for the more distant positioning of the pump Longer tubing results in more time to clear the system of air once the pump movement is initiated and reduces the sample capacity of the syringe For a single sample measurement this would not present a formidable concern but for continuous switching of many samples this might need to be addressed 36 Page VI 1 Improved sample mounting The problem of sample stability discussed above was not fully addressed until late in the design process Once the team had more experience using the probe station it was found that
20. interface and signal amplification can be found in the following section Although the new probe card did require a more complex probe station to precisely align a LEAC chip sample with the probes once this was implemented acquiring large sets of data was facilitated and much more efficient 39 Page B Probe Card Design Specifications and Selection Our probe card is essentially a configuration of precisely positioned probes which can be brought down into contact with the LEAC chip simultaneously allowing for probing of many different signal paths rapidly Figure VI 2 through Figure VI 4 are photographs and microscope images of the probe card Although there are only 40 probes total a 72 probe probe card was ordered because this is the smallest size which would accommodate the required number of probes in the desired fashion These probes are epoxied into an aluminum ring which is connected to a PCB The PCB has dimensions 4 500 x 9 375 x 0 062 with a circular hole on one side to accommodate the epoxy ring and probes As seen in Figure VI 4 there are 32 probes in the central block of probes and 4 on the outside that are components of the two edge sensors The rest of the PCB contains signal traces connecting to the probes which end in edge fingers on the other side of the PCB Thus the probes are electrically connected to by connecting a card edge connector to that side of the PCB ue D EI Di H Di a V EI DI az an
21. makes part of it lower with respect to the plane of the probe card and sample However part of it will still need to be at the same height as the sample to allow coupling of the laser into the waveguide This would allow the fiber launch to sit under the probe card safely while retaining the necessary function LabVIEW Programming LabVIEW is a powerful tool that can interface with hardware devices automate experiments and capture data There are many things that can be done in the future to help with the program moving forward However the best approach would probably be to start the program over from scratch The program used was based on LabVIEW 8 5 which is a fairly old version Unfortunately this version has caused the program to no longer open correctly when used with any version of LabVIEW 2011 After getting a feel for the sort of experiments and forward progress that wants to be made the biggest choice that will be made is what kind of data the group wants captured through the program There are a few key components in which the user interface needs to give the user meaningful real time data the program needs to save data in a useful format and the code should be combined with MATLAB to automate data processing Perhaps it should be looked into not using LabVIEW at all and interfacing to the devices with MATLAB An addition that needs to be added to the LabVIEW control interface is a manual real time selection of channels As of n
22. manufacturer that it might crack and break It was decided that Teflon tubing was to be tried first The Teflon tubing however was not very flexible and any attempt to bend it caused a large force to be placed in the PDMS This is unacceptable since we need a very good seal on the PDMS It was then decided to switch to some PEEK tubing that was available in the lab from Upchurch Scientific part 1569 Even though it required a resizing of the entry hole in the PDMS to firmly accept the tubing it no longer was pushing the PDMS around Strain relief was added in addition that helped minimize any additional strain The next key problem was flow through the system as the PDMS did not have a very strong bond Too much force from the fluid might rupture the seal and end any experiment Through experimentation it was found that applying a vacuum pressure at the outlet caused less force on the system then applying positive force at the inlet and would be less likely to rupture the PDMS glass seal To create this vacuum pressure a syringe pump was added to the outlet which could be controlled to adjust how much force was placed on the system Sucrose dilutions were made by initially weighing out one gram of sugar in five milliliters of water Subsequent 1 5 serial dilutions were then made to obtain different levels of sucrose concentrations These dilutions correspond to a RI change from DI water of 1 10 4 1 104 5 and 1 10 7 Testing of the solutions woul
23. noise it would be a perfect linear fit Another section of code which was used to analyze data is shown in section FFT Code This code uses MATLAB s built in FFT function to perform a Discrete Fourier Transform of a data vector Essentially this converts the input data vector to the frequency domain The output frequency domain data can then be plotted to see which frequency components are present in the data For example a strong 60 Hz peak indicates interference from power lines for other circuitry which operate at 60 Hz 115 Page Based on output spectra we encountered using the code of section FFT Code it was noticed that there was strong interference from power signals in the data To eliminate this a digital low pass filter was created After testing various filters a 10 Hz cutoff frequency low pass filter was decided upon This filter is finite impulse response has a default order of 1000 and a default roll off factor of 5 Hz The code used to implement this filter is shown in FIR Filter Code It is often useful to plot measured dark currents light currents and photocurrents vs channel number to see trends in the data It is also sometimes useful to plot time series data of these variables for individual channels to see if some channels are exceptionally noisy while others are relatively clean Code used to accomplish this is contained in Currents versus Photodetectors Upon discovering that dark currents were drifting f
24. setting Probe Material Overtravel Scrub mark Taper Tip angle Tip Diameter Tip Length Tungsten Probes Tungsten rhenium probes 89 Page One thousandth of an inch A measurement of the maximum degree to which the probes are not in the same plane The material used in the construction of probes The vertical distance probes are supposed to be driven beyond initial contact to ensure a good electrical connection The mark a probe tip makes when applying overtravel The length of probe for which it transitions from its bigger size at the epoxy ring to its smaller size at the probe tip The angle between the beam of a probe and its tip The diameter of the tip of a probe The length of the tip of a probe The most commonly used type of probes Provides good structural integrity but gets oxide crystals stuck in the probes leading to increased contact resistance over time Similar to tungsten probes except they do not get oxide crystals stuck in them as readily Appendix F Multiplexing Amplification Circuit Schematic and BOM fe e ss Drawing Ke PE wu E S LC Figure A 11 Circuit schematic for the multiplexing and amplification functions 90 Page TI ZZ EZ E NOISI A3Y H XAW wt lLINDHYID dWY me ST caps orto o o eg o Ic3 pi o e o c3 Qoo si T Pboan za TE o o ar Bg fA Cd o _7 00 0lo0 90000 Wl eooo
25. there was no fluid moving for an unknown period of time during this experiment 113 Page It is obvious that this initial data is not enough to make any claims about the performance of the system It does however reveal more tests that can be done to further the understanding of the system 1 In addition to recording when the tube is switched between solutions also record when it appears the new fluid has entered the LEAC chip region 2 Rotate the solutions much more slowly to remove ambiguity about which solution is causing which change With these changes perhaps more data can be collected that can further the understanding of the LEAC chip References 1 J Hanson 1994 Refractometry Analyzing Results Online Available http www2 ups edu faculty hanson labtechniques refractometry interpret htm 2 Reichert Team n d Sucrose Solution Percent by Weight Concentration Brix Value Versus Refractive Index at 20 Celsius and 589nm Wavelength per ICUMSA Online Available http www reichertai com files applications 1039637372 PDF 3 S M Monnerat et al 2006 Concentration profiles and effective diffusion coefficients of sucrose and water in osmo dehydrated apples Online Available http www ibilce unesp br departamentos eng cida Monnerat pdf 114 Page Appendix L MATLAB Code Used to Analyze Data After data was exported from the LabVIEW program used to collect it it was frequently desired to per
26. this feature Probe Card Requirements A Benefits of a Probe Card To improve the original probing system a probe card specified and ordered to enable interfacing with the LEAC chip more rapid and efficient than was previously possible Purchase of the probe card required an investigation of critical probe card parameters and system requirements Readers 37 Page looking for more general information on probe cards and how to use the probe card for this system should consult Appendix E Being able to electrically interface with LEAC chips is of vital importance in testing If the LEAC chip is commercialized it will likely come in a package with external leads which can be easily soldered to a PCB or otherwise connected to external circuitry While the LEAC chip is still in its early developmental and testing phases it is cheaper and simpler to be able to electrically interface with LEAC chips without having to place them in such packages This also allows for the addition of other features to the top of a LEAC chip such as microfluidic channels made of PDMS long after the chip has been fabricated The means for electrically interfacing with a LEAC chip without having to place it inside a package is to have metallic pads on a LEAC chip which connect to the desired signal paths These pads can then be externally probed by metallic probes such as the ones pictured in Figure VI 1 Figure VI 1 Original chip probing system The
27. this idea had merit the probe card vendor advised that this feature would significantly increase the cost an accommodation the project budget could not make Thus the circuit was kept external and extra precautions were taken to ensure effective shielding for the probe card s electronic interface Finally the ordering and time it took to physically receive the probe card was a lengthier process than initially anticipated This played a significant role in preventing the project from reaching the original dates of stated goals We had planned on receiving the probe card by the 20th of October but did not actually receive it until the 22nd of November The major contributing factors to this delay were difficulties in determining exact probe card configuration and difficulty in communicating with the vendor exactly what was needed for this application The realization that the probes could potentially hit the PDMS was a problem realized later than it should have been In the future such problems could be avoided by making detailed drawings of how the setup should interact with all desired system components well in advance 44 Page VI Signal Amplification and Switching Circuit A Signal Amplification and Channel Selection An automatic amplification and switching circuit was created for the device that will allow for the realization of real time measurements on the LEAC chip A block diagram of the developed system is shown in Figure VII 1 b
28. title n 3 T De case 4 Sdata colle 1 ch_scan nch_scan nl_ folder nl title n n2_ folder n2_ title n 2 run3_ folder n3_ title n 3 otherwise r r 8 31 length ch_scan r First TC folder r eitle scat pat 2327 leng terning th ch_scan Data 4 11 1 Run 1 get_run_data runl_ fol Data 4 11 1 Run 21 gt get_run_data run2 folder nch_scan cal run2 ti Run 3 get_run_data run3 folder nch_scan cal run3 ti cted on 4 8 232 length ch_scan Data 4 8 L Ran 1 3 Data 4 8 L Run Zi Data 4 8 L Run 2 JData 4 9 1 Data 4 9 LI Less Scatter EAC 4 9 m first get_run_data first_folder nch_scan cal first_ title BAC 4 9 m less scatter ter folder nch_scan cal less scatter title LT110401 2 After Patterning Runl lder nch_scan cal runl title LT110401 2 After Patterning Run2 tle Data 4 11 LT110401 2 After Patterning Run3 tle runs 1 3 EAC COUPLING LT110401 2 4 8 WG4 Run1 get_run_data runl folder nch_scan cal runl title EAC COUPLING LT110401 2 4 8 WG4 Run2 get_run_data run2 folder nch_scan cal run2 title EAC COUPLING LT110401 2 4 8 WG4 Run3 get_run_data run3 folder nch_scan cal run3 title disp invalid input variable return 124 Page end Spl SLO SI b2 3 34 SIs 36
29. to align and the PDMS inside of the metal pads and introduce binding within 1 5 minutes of the removal of the chip and PDMS from the plasma chamber Alignment by hand and with the naked eye was attempted on non working chips and although close alignment was achieved some pads remained covered by PDMS To mitigate this problem a simple alignment system was designed by combing xyz translational stages for fine tune positioning of the chip a glass slide holder to which the PDMS would be reversibly attached and upside down and a magnifying lens to better visualize the alignment An image of this mechanical apparatus is depicted in Figure IV 7 below Figure IV 7 PDMS chip aligner Figure 4 8 shows the mask designs of the chip and channels as well as a piece of PDMS permanently bound to a working chip using the alignment device pictured in Figure IV 7 26 Page Figure IV 8 Chip mask design left PDMS mask design with cutting guide center and actual PDMS aligned to chip and permanently bound from plasma oxidation right V Probe Station A Introduction to Mechanical Probe Stations Because there had been no previous design work on the LEAC chip device there was a significant amount of mechanical fixtures that needed to be designed built and implemented to bring together the various components of the system in a more user friendly and timely fashion This aspect of the project was challenging for several reason and required continua
30. voltages data to currents dark currents polyval linear fit dark voltages splot dark currents plot_currents dark_currents Dark plot_all_ channels Fs sthe function plot currents is contained at the end of this m file sreturn from function if light voltages data was not provided if nargin lt 5 Eet end Sconvert light voltages data to currents light currents polyval linear fit light voltages splot light currents plot_currents light currents Light plot_all_ channels Fs sdetermine photocurrents photocurrents light currents dark_currents splot photocurrents plot_currents photocurrents Photo plot_all_ channels Fs end 120 Page function plot currents currents data Cvpe plot all channelz Fs time l1 size currents 2 Fs Stime vector for currents data plot time averaged currents Vs Channel figure plot mean currents 2 plot_title strcat data_type Currents Average Value Vs Channel title plot title xlabel Detector Number ylabel Current uA SDetermine filtered data for each channel currents filtered zeros size currents for count 1 size currents 1 currents filtered count leac_firrcos filter currents count Fs end Sif plotting all channels is enabled plot dark currents for all channels if plot_all_ channels for count 1 size currents 1 figure ccd currents count SCurrent Channel Data plot time ccd b time leac_fir
31. with the channels on the corner of the side to be printed on flipping it over and marking the widening of the channels with a permanent marker on the back side The slide was then flipped back over and the PDMS taken off of the surface to be printed on After printing the antigens the microfluidic channels were then used to print the antibody solution Below Figure VIII 4 shows a section of the results from this test taken on an on campus fluorescent microscope Figure VIII 4 Combined fluorescent images of assay using hand printing and microfluidic channels showing ESAT 6 binding with itself lower right and a AG 85 crosslink binding with ESAT 6 upper left Figure VIII 4 shows the binding and crosslinking expected from the earlier assay Both of these spots were printed with antigen ESAT 6 The lower channel was then filled with a ESAT 6 which shows a very bright spot in red right at the widening of the channel where the waveguide would be The upper channel was filled with a AG 85 which shows a lighter spot in teal in the channel widening representing the crosslink binding Both of these are the results expected in a working system and prove the successful integration of the microfluidic channels with the biomolecule pin printing technique 53 Page The hand printing technique was very successful on a glass slide however in order to mark the points to print on the surface must be transparent which poses a problem for LEAC c
32. 4 T 4 F4 Current uA FHP pepe tet ett E EE EE E EE E EE E EE E 0 5 10 15 20 25 30 Detector Number Figure IX 6 Dark currents light currents and photocurrents from a LEAC chip Figure IX 7 shows photocurrents from a specific instance of coupling the laser from a laser diode into one of the waveguides on a specific LEAC chip The results are not necessarily typical of what is usually obtained as more typical results include significant noise This could possibly be attributed to either poor chips being fabricated recently or our circuitry not being as sensitive to photocurrents as that which has been used in the past In particular the lock in amplifier which has been used in the past to measure photocurrents may be more accurate than this system s approach for measuring all the dark currents on each photodetector then measuring the light currents when this cycling is finished A description of how the lock in amplifier works can be found in section Il it may useful to implement such a mechanism or at the least a test for comparison in the future 60 Page Laser Coupled Photocurrents from a LEAC Chip Currents uA e 10 15 20 25 30 Detector Number Figure IX 7 Photocurrents from a LEAC chip test Probably the most important thing discovered from conducting LEAC chip characterization tests is that the dark currents drift greatly over time The dark currents will drift due to local temperature changes and may drift
33. ER a ae G NRE Figure VI 3 New probe card ordered to specifications 40 Page Figure VI 4 Top view of the probe card showing probes and epoxy ring configuration left and a microscopic view of the probes in contact with LEAC chip pads and an edge sensor shown on the right hand side right The main motivation behind ordering a probe card for testing the LEAC chip was to make the process of measuring photocurrents across an entire waveguide much quicker Another benefit of having a probe card is that it allows different photodetectors on the LEAC chip to be probed sequentially without changing any other component of the LEAC chip setup For example the temperature of the air in the room surrounding the LEAC chip might change due to sporadic air conditioning or currents during the time required to realign an individual probe to a different pad This would undesirably change the testing conditions from photodetector to photodetector The requirements for the probe card were relatively standard this lab on a chip device is not have a high frequency high power high temperature low temperature or high probe density system We do have a low current system with total measured currents in the range of microamperes and signals of interest as low as nanoamperes but with nominal leakage currents on probe cards in the pA V range this was not a problem that needed accounted for For these reasons we chose to use the increasingly standa
34. If your experiment is not dependent on your surface a glass slide is recommended The surface must also be surface treated The recommended surface treatment consists of sulfo SMCC and EDS This allows for a thiol group to be on the surface which proteins will bind well to Tuning Plasma Chamber 102 Page Every once in awhile the plasma chamber will need to be tuned it is a good precaution to start with this step on the day you intend to run the assay This is to make sure the plasma is running correctly 1 Close the valve to the gas chamber a This is the green handle located on the left side of the machine 2 Plug the machine in press the AC On button then press the Start button a Switching the dial to PRES allows you to see the pressure i Pump the pressure down until the gauge shows a pressure of 750 mTORR Figure A 18 Plasma Chamber Cleaning the Surface A clean surface to bind biomolecules to is essential 1 Place the chip or slide on a kimwipe or hold it tightly in pair of thin tipped tweezers over a large beaker SOmL or larger a Kimwipes can be found in various places throughout the lab and either size will work for this step b Tweezers can be found in a beaker on the countertop 2 For all rinsing steps If rinsing over a kimwipe allow the waste solutions to dry and evaporate before throwing the kimwipe away If rinsing over a beaker dispose of the waste 103 Page solutio
35. LEAC chip currently has 39 different photodetectors which can be probed for each one of the eight on chip waveguides The geometry of this can be seen from the LEAC chip mask in Figure VI 2 This constitutes 39 different signal pads and 2 ground pads for each waveguide To facilitate more rapid measurement of photocurrent through these photodetectors a probe card was ordered and implemented this year 38 Page cn wm Y Hunt Figure VI 2 Mask design of the LEAC chip with color coding where red indicates signal pads green indicates ground pads and purple indicates the approximate locations of edge sensor touchdown Prior to integration of the probe card making measurements with the LEAC chip entailed probing individual pads with individual probes Figure VI 1 shows a setup for accomplishing this and measuring the current across a photodetector for a specified voltage with each iteration After the implementation of the probe card probing the first or last 34 photodetectors connected to any one waveguide on the LEAC chip became a much rapid process that involved no adjustments one the system was in place All that is now required for measurements is to align the probes of the card and make contact with the metal pads once and then use a system of multiplexers amplifiers and LabVIEW software to systematically measure the currents running through each individual photodetector at a given applied voltage More detail on the circuitry LabVIEW
36. Lab on a Chip Diagnostic Biosensor Final Report Spring Semester 2011 By Andrew Armstrong John Blatt Justin Grantham Dan Higley Kelli Luginbuhl Kelly Nienburg Prepared to partially fulfill the requirements for ECE402 Department of Electrical and Computer Engineering Colorado State University Fort Collins Colorado 80523 Project Advisors Dr Kevin Lear Dr Dave Kisker Abstract One third of the world s entire population is infected with some form of tuberculosis TB and only 60 of all new cases are detected The current methods for diagnosing tuberculosis are limited by decoupled diagnosis and treatment as well as expense and the expertise needed to administer the tests These limitations have provided a niche for a new inexpensive and portable technology that can provide rapid and improved point of care diagnosis The device developed and detailed in this paper uses label free technology to detect an antigen antibody interaction This device makes use of a semiconductor chip and waveguides wherein an antibody that binds to an antigen on the waveguide s surface will modulate the evanescent field causing a decrease in photocurrent to be detected by buried detectors The initial state of the device was far from practical use Microfluidics using active flow were designed and integrated to provide the chip with real time detection capabilities The original manual two probe measurement system was improved by incorporating a prob
37. S stamps are aligned with each other when they are put onto the chip The aligner makes this accurate but practice and perhaps improvement in this technique would improve alignment 64 Page Another and potentially more detrimental problem is that of the inability to permanently bond PDMS to surface with biomolecules already attached It was found that in order to use the syringe pump and tubing system a permanent bind between the PDMS and the LEAC chip was important The tubing system will pull PDMS that is not permanently bound off of the LEAC chip and for real time data it is imperative that the second dimension PDMS channels stay attached However to get a permanent bind between PDMS and a chip both surfaces need to be placed in the plasma chamber which is impossible with the previously printed first dimension of antigens The plasma treatment would ruin the first dimension patterning Unless a way to avoid a need for permanent binding within the system is discovered this problem may be cause for finding a new form of microfluidic channels that does not involve PDMS stamps Probe Station Future work on the probe station should make the system as a whole more user friendly and easier to align and manipulate The recent addition of translation stages under the fiber launch and probe card attempted to meet that need but are lacking in some ways The translation stage under the probe card is probably a good solution for that particular
38. SE WEEK NO IMPR INVERTS CIRCUIT eo d l EE aT CH Ae oi VDD VSS GND EN Figure A 12 Top traces of PCB war slo 39 N e S ado oopoons SCH oo WHO oe No oo Nez OO soc GIO O max oo E oo O OO o gt oo oo H o ole o d 4 ojo a ojo Q ojo E z tia VDD VSS GND EN Figure A 13 Bottom traces of PCB 91 Page Name Component i Package H Value H Manuf H Manuf Part Ne Distrib Distrib Part Ney Qty H C1 C C2 C C3 C CONN1 2540 6002 3MMH 40 CONN2 2520 6002 3MMH 20 IC1 LF412_custom DIL IC2 LF412_custom DIL IC3 LF412_custom DIL PL 2WP DSC R1 R 0 125W 5 M R 0 400 150 R2 R 0 125W 5 M R 0 400 150 R3 R 0 125W 5 M R 0 400 R4 R 0 125W 5 M R 0 400 150 F 150 R5 R 0 125W 5 M R 0 400 150 150 200 R6 R 0 125W 5 M R 0 400 R7 R 0 25W 5 MCR 0 500 100K R8 R 0 25W 5 MCR 0 500 R9 R 0 25W 5 MCR 0 500 1k R10 R 0 25W 5 MCR 0 500 1k R11 R 0 25W 5 MCR 0 500 47k R12 R 0 25W 5 MCR 0 500 100 1k R 0 25W 5 MCR 0 500 R 0 25W 5 MCR 0 500 1k R 0 25W 5 MCR 0 500 1k U1 DG508AAK_JG DIP 16D21 7 DG508AAK PCB Artist Bill of Materials is provided for reference only and must be verified by the user Figure A 14 BOM of MUX AMP Circuit 92 Page Appendix G LabVIEW Manual For Use LabVIEW Manual for LEAC Data Recording and Control VI v1 0 LEAC Data Recording and Control is the
39. _count folder dark fig saveas gcf strcat run run_count folder dark jpg close splot noise of light currents figure fcount fcount fcount 1 hold on for count l length run run_count 1lc plot ch_scan std run run_count lc count 0 2 colors vector count end title strceat Light Currents Noise from run run_count run title xlabel Detector Number ylabeli saveas gcf strcat run run_count folder light_ noise fig saveas gcf strcat run run_count folder light noise jpg close Currents uA splot noise of dark currents figure fcount fcount fcount 1 hold on for count l length run run_count dcl plot ch_scan std run run_count dcl count 0 2 colors vector count end for count l length run run_count dc2 plot ch_scan std run run_count dc2 count 0 2 colors vector count end title strceat Dark Currents Noise from run run_count run_ title xlabel Detector Number ylabeli saveas gcf strcat run run_count folder dark noise fig saveas gcf strcat run run_count folder dark noise jpg close Currents uA 128 Page splot deltas of light currents and dark currents figure fcount fcount fcount 1 hold on splot deltas of light currents baseline mean run run_count lc 1 2 for count l length run run_count 1c plot ch_scan mean run run_count lc count 2 baseline strcat colors vector count end splot deltas of
40. ained from the buried detector The amplifier will then lock in to the modulated current s frequency and phase induced by the laser giving a direct readout of the photocurrent no matter how large the dark current may be This phenomena is described by equations two and three Vpsa Vsig VL sin w t due sin wrt Berl 2 Vpsa Kalen sig Berl after LP Filter 3 In the equation above Vpsa is the output of the lock in amplifier Vsig is the amplitude of the input signal V is the amplitude of the reference signal Osi and w are the phase and frequency of the input signal and Dr and w are the phase and frequency of the reference signal It can be seen that the output of the lock in amplifier will produce a DC voltage corresponding to the component of the input and reference signals that are of the same frequency and phase Due to time and resource constraints this method was not implemented by the senior design group this year However it is highly recommended that this technique be employed for any future work done with the current circuit 13 Page IL LEAC Chip Fabrication The LEAC chips used in all experiments were manufactured in Colorado State University s Electrical and Computer Engineering cleanroom by both graduate and undergraduate students Standard cleanroom procedures and processing were used Some of the more critical steps are described below and the full fabrication flow can be found in Appendix B
41. an antibody l V Curves of LEAC LT110401 2 Before and After Annealing 1 min at 400 C 0 10 T T T T T T T 10 T 10 Taa E EEEE Q 5 6 Vi 6 10 M g WG4 Light 10 L WG4 Dark WG4 Light After Anneal H WG4 Dark After Anneal 10 L fi L fi fi fi fi 10 50 15 10 5 5 10 15 20 0 Voltage V Figure IX 4 Typical IV curve of a LEAC chip before and after annealing Y axis is the magnitude of the current 58 Page Another test that was conducted to confirm proper working of LEAC chips involved turning a light source near the LEAC chip off in the middle of a time series collecting data from a single photodetector on the chip Figure IX 6 shows a plot of this including measurements of dark currents light currents and photocurrents across a single waveguide on a single LEAC chip The results of this procedure are not typical there is often much less uniformity from detector to detector than is apparent in this plot Microscope Light Switching Experiment On to Off Microscope Light Switched from On to Off Voltage V 0 100 200 300 400 500 Time s Figure IX 5 Time series of data from photodetector when the microscope was switched from on to off in the middle of the test 59 Page Dark Currents Light Currents and Photocurrents for a LEAC Chip 10 dark currents obt currents Photocurrents Eee EE d ae Ge tette rt Tt e a billed
42. ary recruitment of advisers from the chemical engineering and the mechanical engineering departments could provide invaluable advice and support 70 Page XI 1 2 3 4 5 6 7 8 9 10 11 Bibliography Centers for Disease Control and Prevention 2010 Tuberculosis TB Online Available http www cdc gov tb statistics default htm Mayo Clinic Staff 2011 Tests and diagnosis Online Available http www mayoclinic com health tuberculosis DS00372 DSECTION tests and diagnosis R Yan et al Waveguide biosensor with integrated detector array for tuberculosis testing Dept Elect Engr Colorado State University Fort Collins 2005 J N Stern et al Molecular signatures distinguishing active from latent tuberculosis in peripheral blood mononuclear cells after in vitro antigenic stimulation with purified protein derivative of tuberculin PPD or Candida Department of Cancer Immunology and AIDS Dana Farber Cancer Institute Boston 2009 M Frahm et al Discriminating between latent and active tuberculosis with multiple biomarker responses Center for AIDS Research Duke University Medical Center Durham NC Mar 2011 N Chegou Host Markers in Quantiferon Supernatants Differentiate Active TB from Latent B Infection in 2nd Global Symposium on Global IGRAs 2009 No Copyright Smithsonian 1997 Cost Per Wafer Online Available http smithsonianchips si edu ic
43. assay fluorescent image right cropped and labeled fluorescent image of TB antigen antibody assay 52 Figure VIII 4 Combined fluorescent images of assay using hand printing and microfluidic channels showing ESAT 6 binding with itself lower right and a AG 85 crosslink binding with ESAT 6 upper Figure VIII 5 Fluorescent image of ESAT 6 and a ESAT 6 binding in channel widening from assay using PDMS printing with microfluidic Channels ccccsssccccsssssccesssssececsssseeeessssecesesssseesesseeeesseaaes 54 Figure IX 1 Current versus Voltage gain plot of MUX AMP chreunt 56 6 Page Figure IX 2 Drift test of MUX AMP CirCUit ccccccsscccesscessecesssecesseecesseceessecesseecesuecesssecesseeeesseeesaeees 56 Figure IX 3 RC response Of MUX AMP rout 57 Figure IX 4 Typical IV curve of a LEAC chip before and after annealing Y axis is the magnitude of CHE CUIFENT ase ot el en 58 Figure IX 5 Time series of data from photodetector when the microscope was switched from on to off in the middle of the test 59 Figure IX 6 Dark currents light currents and photocurrents from a LEAC Chip cccsssscceeesseeees 60 Figure IX 7 Photocurrents from a LEAC Chip test 61 Figure IX 8 Sequential measurements of dark currents on LEAC chip demonstrating thermal drift 61 Figure IX 9 Magnitude squared of filtered and unfiltered data 63 Figure X 1 Schematic of location of PCB that needs to be modified to invert Circuit
44. ate file making the data available for manipulation and analysis The data on the front panel displays basic standard deviation math for instant response to the user This feature is useful as it allows the user to immediately detect unresponsive pads or unexpected values before the entire experiment has been conducted The LabVIEW program also contains the ability to 48 Page record data based on a reference pad checking back to it after each pad reading This feature of the program allows for monitoring of drift which is invaluable to the user if any corrective action is needed to account for changes in environment such as light or temperature which can skew the data E Obstacles in Data Collection and Control While creating the LabVIEW program a few problems arose The first was the speed of recording data points It was unknown whether a value of up towards 8000 data points per second would disrupt the system or still be able to display the desired outputs With experimentation it was found that not only could the computer handle this output in a stable and long term experiment but that it also provided enough data to employ filters with the capabilities of applying and removing large amounts of noise The second problem encountered was how to display the data to the user while maintaining a clean and easily understandable front panel Too much data at once could be overwhelming while too little would be worthless A simple compromise
45. be vast In addition because complementary metal oxide semiconductor CMOS technology is no longer cutting edge the cost of mass producing chips like the one used in this project would be quite low around 2 50 making this a feasible method for improving point of care diagnosis in third world countries Thus the goals of this project were vast as no senior design work had been done previously on this project The goals can be summarized into the following categories but are by no means exhaustive and all goals were selected with the aim of either making the device more attractive to the medical industry or more automated and easier to use for future experimentation 1 LEAC Chip Fabrication To optimize fabrication techniques to obtain a successfully repeatable procedure in producing chips with consistent behavior 2 Microfluidics Integration To design a master mold for channels confirm fluid flow and integrate the channels with the chip for the realization of real time testing and rapid sample delivery 3 Development of Probe System To build a mechanical structure for aligning all elements of the testing apparatus accommodating for changes to the probing system and the incorporation of microfluidics on the surface of the chip 4 LabVIEW Automation of Data Collection Develop LabVIEW code for multi channel testing and a more user friendly data acquisition and review This involves the software interfacing with the hardware and automa
46. be working great and probably does not need any revisions done to it until other parts of the system are developed more One modification that may want to be made is the inversion of the amplification circuit In order to achieve this the jumper must be placed on the board and pin 1 of IC2 needs to be lifted up from the board See Figure X 1 This will result in increasing voltages for increasing currents which may make data collection more straight forward The PCB design solved most of the problems encountered early on in the year The schematic is located in Appendix F as well as in electronic form in the lab on the senior design CD Any future revisions of the probe card should be made by modifying the current schematic and PCB design in Advanced Circuits PCB artist 67 Page Optional Jumper toa change inversion of circuit Figure X 1 Schematic of location of PCB that needs to be modified to invert circuit A future revision of the circuit might want to attempt to place the amplifier portion of the circuit before the multiplexing This would remove the large time constant issue of the amplifier having to change to the different channels signal levels as well as eliminate any possible MUX noise from being amplified Lastly it would be beneficial to incorporate the lock in amplifier set up as discussed in Section Il This would allow for the direct measurement of the photocurrent and may remove a lot of confusion associated with changi
47. channels and data collection In accomplishing this goal a few key features were defined to better define the end vision of the device These key features include having both a friendly and intuitive user interface that would require minimal training to use to be able to record hours of data and be left overnight without crashing and to be able to interactively return data to the user in real time to facilitate the analysis of data in future experiments Due to the nature of LabVIEW the ability to maintain stability over a lengthy experiment was easily incorporated with all data being accounted for in a series of sources and sinks This feature removed extraneous and unorganized data that uses up memory and that the program is unable to access thus mitigating many previous problems With the differentiation between the front panel and the wiring diagram a more user friendly interface was simplified from a complex juggling act to 47 Page Version 1 0 LEAC Data Recording and Control VI 4 23 2011 Start Channel Initial Setup Variables Run Variables e Fixed Variables Loop Control A1 Analog Read Channel S DeviceType To set the program into F Data Stream Time s End Channel E an infinite loop Tih g A7 5 iUSBDA BIR J a En Bi S switch up To stop flip Save to File f icelndex switch down while Data Points Per Second Stabilize Time ms de Ra 4 Spring_11 Testing lvm A E SM 128 F 2000 On U tfil KEE Velo Ref
48. currents and photocurrents under different conditions and how these variables experience drift over time Dark currents are the currents through a photodetector when light incident on the photodetector is as low as possible The photocurrent is defined as the dark current subtracted from the current measured or light current when a specific light source is on The basic function of the photodetectors inside the LEAC chip are to sense changes in the intensity of light incident upon them To ensure that the current through these photodetectors does change in the expected way we conducted tests involving the generation of IV curves for different photodetectors on a single LEAC chip with the microscope light above them either on or off Figure IX 4 shows a typical result of such a test It is apparent that the current through the photodetector when the microscope light is on is generally higher for a specific voltage bias then the current when the microscope light is off for the same voltage bias as would be expected Also it is apparent from this graph that annealing the chip contributed to making the observed currents more consistent Finally Figure DA shows that the photocurrent to dark current ratio is much higher for the negative biasing direction of the photodetectors which indicates that this is an easier region to operate in when trying to sense changes in photocurrents due to external changes in index of refraction such as in the presence of
49. d be repeatable for successful detection on a LEAC chip Some other features of this step in the process that will pose a 54 Page problem in the future is that once surface treated and printed with antigens the LEAC chip will not be able to be plasma treated without destroying the biomolecules As passive flow was never achieved on a chip and active flow required the permanent bonding to attain a good enough seal for flow there is much to address for future experiments IX System Characterization A Circuit Characterization Draft Tests were conducted to characterize the amplification and switching circuitry used in conjunction with the LEAC chip These tests verified that the gain of the circuit was consistent across different channels under different input currents and over sufficiently long periods of time Shown in Figure IX 1 are plots of the input current versus output voltage from the circuit before and after conducting tests that took about an hour These plots show a gain that is nearly the same in both measurements before and after the hour of testing Note that the voltage offset drifted more significantly than the gain but it was determined that this was likely due to a change in output of the power supply which is used to generate the voltage offset This discrepancy could be prevented in the future by using a different power supply or using a resistive divider to drop the voltage input of the offset stage to an acceptable valu
50. d component would take hours to locate on the unorganized circuit Figure VIl 2 shows the circuit before and after modifications were made Figure VII 2 Multiplexing and Amplification Circuit before left and after redesign and PCB order right The figure above shows the circuit after the redesign had been performed The PCB layout was designed in PCB Artist available freely through Advanced Circuits of Denver CO This PCB version of the board allowed for much easier maintenance and provided more consistent performance than the previous circuit For a complete schematic and bill of materials BOM of the circuit the reader can refer to Appendix F The interface card shown in Figure VII 3 was also built and provided an better interface between the probe card and switching circuit 46 Page Figure VII 3 Interface cable to connect the probe card to the switching circuit Finally to allow for the optimal control of the circuit an enclosure was modified to house all of the input and output components of the Hytek DAQ device This house can display edge sensor information with LED lights and it has the ability to select which side of the probe card that the potential is applied to C Software Overview The goal in implementing software in the LEAC project was to fully automate data collection Previously the signal pads had to be changed tediously with manual probes This new system allows for the complete automation of switching
51. d start with the lowest concentration of sucrose and continue going up in sucrose concentration Between each sucrose solution DI water would be run for a period of time in an attempt to cleanse the system 111 Page In this planned experiment it was also needed to have an idea of the magnitude of diffusion coefficients to get the difference from the ideal and what the plug flow actually would be doing Taking numbers from 3 it was found that diffusion should have little effect on the fluid flow and not cause any problems during the experiment To confirm this theory a quick test was performed using different colored food dyes to see how much they interacted with each other In the course of the experiment it was found that a small air bubble formed while switching between the different feeds which provided an effective block for preventing mixing It was also believed that temperature effects from the microscope light might interfere with the RI of the solutions and will need to be investigated further The initial attempt at performing a sucrose RI test produced very odd results Figure A 25 shows the amplifier output versus time for the following conditions using chip LT110401 1 e AB Common pad used on circuit e 3 V bias on common pad e Pump Speed of 60 e Total Stream Time 12min e 8 104 3 g mL sucrose solution Run 1 Stems at actions T T T Pump Start DI Sucrose T DI DI Sucrose Pump Stop Recouple Laser 2 025 y
52. de The video was taken on a fluorescent microscope Figure IV 5 Still shot from video of moving fluorescent particles 24 Page As the integration of the device progressed permanent binding became an increasingly critical component of the design Without permanent binding more significant leaking occurred out of the channels which would could short circuit the probe pads or make measurements indistinguishable from one another on the actual chip In addition once it was established that active methods were needed to achieve flow on the chip a permanent bond was necessary to give the setup enough rigidity Before a better setup was introduced to relieve strain on the PEAK tubing a bend or an adjustment of the tubing was enough torque to pull the PDMS off of the chip A successful irreversible bond between the chip and PDMS was first achieved on non working chips by plasma treating both the chip and PDMS with 18 seconds of plasma oxidation at 20 watts with the plasma machine located in Scott Lynn s laboratory During plasma oxidation oxygen radicals attack the surfaces rendering them more hydrophilic The plasma oxidation adds silanol groups SiOH to the PDMS and when attached to the chip the bonds eventually relax to their normal state resulting in a permanent bond Figure IV 6 shows a fully fluorescing solution of labeled antibodies running through the channels with a permanent bond on both a glass slide and a chip with no evid
53. e A 10 Image showing the probe card probes being cleaned and the resulting marks Typical epoxy ring probes have a beam and taper which is where the probe comes out of the epoxy ring and begins to taper towards its tip size At the junction of the taper of the probe and the probe tip the probe is bent so it points downward at the point to be contacted The part of the probe which is pointed downward is called the probe tip An example probe beam and taper lengths is a few hundred mils and a probe tip length of around ten mils Probe tips can be increased in length 16 with the consequence that they become more fragile Probes should be sized to be around 30 40 of the size of the bond pad to which they are to connect Edge sensors are sets of two or three probes which form a closed circuit between two probes when they are not in contact with the device being probed and form an open circuit when they are in contact with the device to be probed Edge sensors are commonly used in industry to automatically 87 Page probe test wafers the machine doing the probing can tell when contact is made For people that are experienced with using probe cards edge sensors will likely not be necessary but for this project it was found that having edge sensors helped tremendously in determining when contact was made and determining if there were any planarity errors between the probes Edge sensors can be designed to have an isolated tip 3 pr
54. e by hand cutting with a razor blade The channels were designed to run parallel to each other but to cross the waveguide perpendicularly This feature of the mask enables the device to potentially detect multiple antigen antibody complexes on the same waveguide nearly simultaneously Figure IV 1 shows the final mask designs Of the 32 individual pieces there are eight each of four different designs The four designs vary only in spacing and number of channels Different designs were selected for fluorescent assay imaging versus chip use and having multiples of each design made for more efficient production 19 Page Figure IV 1 Submitted mask design left and single zoomed in view of channel design right For integration with a working LEAC chip the bottom right design in Figure IV 1 was used With only a single syringe pump the designs with extra channels were not necessary This design was picked over the other 4 channel design because the outer channels were positioned closer to the edge and consequently the expanded width crosses the waveguide closer to where the laser was coupled This proved beneficial because the power and photocurrents have larger and more quantitatively measureable values at the start of the waveguide than at the end The power decreases as light continues to propagate down the waveguide All channels measure 100 um in width and expand to 570 um at the widest part where they traverse the waveguides The expansio
55. e card and a LabVIEW controlled multiplexing and amplification circuit to automate the process The circuit outputs a voltage proportional to the current that can easily be read and logged by the computer The probe station was completely redesigned to accommodate for these new features Thus the device has been brought to a stage ready for real time experimentation and fully automated data acquisition Because the chip was designed with multiple waveguides the future vision of this project is its use in inexpensively testing for several diseases simultaneously without the need of a lab or trained professional Because the chip uses trailing edge fabrication technology mass production should put the cost of a single chip under 2 50 making this device useful for many applications Future work on the device will include experimentation to determine the chip s limit of detection to successfully detect two different antigen antibody complexes to quantify sensitivity and selectivity to scale down the supporting electronics to a single chip and to fully automate apparatus setup 2 Page Table of Contents I Introduction and Motivation 9 TES Backeteoupgd ee 11 A Local Evanescent Array Coupled LEAC Chip Theory ccssscccssssseceessssseeeesssseeesssseeeeesssneeeeeees 11 B Modulation of the Laser sea eea aen shade aaken Gans ESAE EE SEEN REE EES erer Aa ier ESS 13 UL LEAC Chip Fabricatie tesan anana aa Zedi egdegdgeg ate
56. e cd CEICM SECTION2 pdf Stanford Research Systems n d About Lock In Amplifiers Online Available www thinksrs com Transene n d Silicon Dioxide SiO2 Etchants Online Available http www transene com sio2 html silox R Yan et al Waveguide biosensor with integrated detector array for tuberculosis testing Dept Elect Engr Colorado State University Fort Collins 2005 Silicon Kinetics 2009 Kinetic Characterization of Antibody Antigen Interactions Online Available http www siliconkinetics com pages pdf 20files Antibody 20Characterization p df 71 Page 12 13 14 15 16 17 18 Food and Agriculture Organization for the United Nations n d POLYDIMETHYLSILOXANE Online Available http www fao org ag agn jecfaadditives specs Monograph1 Additive 315 pdf Lynn Jr N S 2009 Dissertation Study and Design of minimally instrumented microfluidic unit operations for a portable biosensor mixing pumping and reaction Fort Collins Colorado State University Micromanipulator 1992 September 4 Basics of Low Current Probing Retrieved April 30 2011 from Micromanipulator Corporation Web site http www micromanipulator com assets application_notes BasicsOfLowCurrentP robing PD Accuprobe 2003 Probe Card Tutorial Online Available http www accuprobe com Downloads Probe 20 20Tutorial pdf Nott Private Communication October November 2010 Dr Nicholas Scott Lynn Co
57. e the PDMS and sample both of which were detrimental to experimental results and the longevity of the equipment The following provides a breakdown of the major design concerns and troubleshooting solutions that were used to resolve the issues 1 Pump design Must implement a pump system allowing active flow through the micro fluidic channels As discussed in detail in section four permanent plasma bonding between the sample and the PDMS was explored with varying degrees of success In addition an improvised syringe pump was implemented to create a vacuum capable of pulling fluid through a channel This was desirable as the vacuum tended to improve the bond between PDMS and the chip whereas a pushing motion would have tended to disrupt the bond 2 Pump Positioning The tubing must be interfaced to the chip in a way that minimizes the risk of breaking the seal or translating the sample In this stage of the design the microscope would be moved at an intermediate step in the alignment procedure to allow the pump system to move into position directly above the probe card PEEK tubing would then be carefully inserted through the hole in the probe card while avoiding touching the probes and then held in place to avoid moving the PDMS or the sample This apparatus is shown in Figure V 6 below 34 Page Inlet Tube Holder Figure V 6 Pump positioned above probe card with tubes running through the epoxy ring and into the sample The pi
58. e while at the same time having greater stability from its larger voltage output However it is important to be cognizant that the resistors of this resistive divider could potentially add noise to the output of the circuit However because the voltage offset stage is the last stage of the circuit anyway this noise should be much less than the noise already apparent 55 Page Input Current vs LabVIEW Read Output Voltage 3 T T T Ch 1 Before Testing Ch 11 Before Testing Linear Least Squares Fit Before Testing Voltage 0 26475 V uA Current 0 25723 V 257 Ch 1 After Testing Linear Least Squares Fit After Testing Voltage 0 26481 V uA Current 0 25194 V T T T T T T Voltage V an Wie 4 0 L L L L L L L L L 4 Curent uA Figure IX 1 Current versus Voltage gain plot of MUX AMP circuit A gain and voltage offset test such as the one illustrated in Figure IX 1 should be conducted anytime a change to the circuit has been made as circuitry changes will likely affect these parameters It is necessary to know the gain and voltage offset in order to be able to convert the output voltage obtained back to an input current for data analysis Another test was conducted to see how much the output of the circuit drifted over time given a constant input A plot of the resulting time series from this test is shown in Figure IX 2 The plot indicates that the output of the circuit drifted less t
59. ece of tubing not connected to the syringe pump is held in place by a secured magnetic clamp E Problems Encountered With the Second Probe Station Configuration This design was successful in the sense that it resolved the issue of providing active fluid flow while preventing the tubes from moving the sample and thus damaging the probes or metal pads However a problem not realized until an experiment was conducted was the alignment of the fiber to the waveguide Alignment of the fiber must be extremely accurate to allow adequate coupling into the waveguide and it is this factor that dictates that the fiber should be the last thing aligned as all other steps can move the sample enough to disrupt laser coupling Also due to the sensitivity of the adjustment it is impossible to manage without the use of a microscope This called for another drastic revision of the probe station and alignment protocol because to allow the microscope to be in place throughout the experimental procedure required a different design of the pump s location setup and tubing strain relief F Final Probe Station Configuration The final state of the probe station was a product of the developing design specifications as well as the increasing experience of team members assembling the apparatus The final design 35 Page addressed the problems detailed above and provided for the necessary microscope placement while also improving the function of the probe station
60. ecent advances in techniques and technology has become around the same price as the blade probe card technology The epoxy ring probe card topology has now overtaken the blade probe card topology in terms of popularity An epoxy ring probe card can be used for most applications except those which deal with high frequencies gt 3 GHz or very low leakage currents lt 1 pA V gt The most commonly used materials to make probes out of are tungsten tungsten rhenium and beryllium copper Tungsten is the most commonly used and cheapest of these materials It provides good probe stability and probe life for most applications Oxide crystals can become trapped in Tungsten probes periodically due to their fibrous nature P This means tungsten probes will require occasional cleaning to maintain acceptable contact resistance Cleaning was performed once on the probe card by carefully sliding it backwards while in contact to a fine grit polishing paper the exact grit was not recorded as a scrap piece was used this process should not be very sensitive to the grit see Figure A 10 Tungsten rhenium probes are very similar to tungsten probes except they do not have the fibrous nature of tungsten and thus trap oxide crystals within them less readily Beryllium copper probes are used when low contact resistance or high current is required These probes wear faster because beryllium copper is softer than tungsten or tungsten rhenium 86 Page Figur
61. ed by the Hytek There is a caveat that 128 data points must be taken every time the program runs and to ensure that it does not crash a hard limit of 1 second and 128 data points per second was implemented Removing these safeguards works against the idea of making the program extremely easy to use Data Points Per Second determines how many data points will be taken during each second with how many seconds having being chosen from Data Stream Time The maximum is eight thousand data points per second Multiplied together Data Stream Time and Data Points Per Second will give how many data points in total each pad will have recorded per run The minimum amount of data points taken per run of the program is 128 data points as dictated by Hytek Automation s program As a safety measure the minimum has been set 94 Page to 1 second Data Stream Time with 128 Data Points Per Second This will ensure that at the bare minimum of only one channel being run once the program will not crash e Stabilize time is the time in milliseconds between when the Hytek switches digital output channels and when it starts to record data Experiments have shown that 95 of noise associated with switching is gone after 2000ms e The reference channel is a toggle that after each pad is read will then switch to the reference channel and take a data sample identical to other pads in the run In the middle right box are the Fixed Variables as pictured in Figu
62. ed planarity problem This can be caused by the sag of the probe card along its length old tape making the sample sit incorrectly flexing of multiple tape layers if they are stacked and other unknown causes To remedy this eliminate tape residues and stacking of tape and use paper shims under the probe card when necessary Most planarity issues can be solved by over traveling This is translating downward beyond the point the point when the first edge sensor makes contact Over travel distance of 2 3 mil 002 003 are typical and well within the tolerances of the probes The deepest over travel ever performed with the probe card was at least 4 5 mil 0045 and did not damage the probe card however this should not be necessary Once contact is made with the sample the edge detectors should be taken out of the circuit via a switch on the front of the circuit box Lastly it should be noted that the use of the platen control wheel to make contact with the sample will cause the probes to shift slightly to the left upon relaxing force on the wheel This is due to the probe station s use of a chain to convert the wheel s motion into the platen s vertical motion When released the chain relaxes back to the closest link Laser Coupling Using the fiber launch motion capability and the microscope align the fiber to the waveguide with the microscope light on Turn the microscope light off and turn on the laser diode driver The laser diode shou
63. eering Biomedical Engineering Mechanical Engineering Figure A l1 Diagram showing the interdisciplinary team dynamic of the LEAC project e Weekly meetings with the entire team and advisors This provided all team members the chance to request additional resources and information as well as report progress e Weekly meetings with only the undergraduates in order to become further involved in each others disciplines and projects and learn how to integrate our pieces together e Developed and maintained a Gantt chart shown on the following page which enabled us to better visualize the different pieces of the project and their relationships and identify the critical path e Rotated leaders throughout the year so that each team member gained the experience of leading group meetings managing the project deadlines 74 Page Figure A 2 Gantt chart of Spring Semester 75 Page Appendix B LEAC Chip Fabrication Flow This Appendix intentionally left blank 76 Page Appendix C Microfluidics Fabrication Procedures Channel Width 100 um Channel Depth 35 40 um j 1 0 mm a Outlet Reservoir A Inlet Reservoir EE LEAC Chip _ J PDMS Mask Design e J Kelli Luginbuhl 19 5 mm 2 22 2011 Figure A 3 PDMS Mask design drawn in LASI This procedure intentionally left out 77 Page Appendix D Probe Station Usage Instructions LEAC chip layout and Probe Station Con
64. ees 92 Figure A 16 Front Panel of LEAC Data Recording and Control vi 0 93 Figure A 17 Sample section of coded wires from the LabVIEW program 96 Figure A 18 Flow Chart of LabVIEW code sssssssssssesssssisssssensssreesssrerrssrersssreressseresssernsssreesssreenssneens 98 Figure A KH Eve ul TEE 103 Figure A 20 Inside Plasma Chamber cccssssccesssssccesssseceessssecesssseeesssnsseeesssssecesesuseesesssseeesssaaes 105 Figure A 21 Use of micropipettes ccccccsssssccessssseceesssseceessssseeesssssecesesseeeeseeuseeeeesseeesesssseeesseaaes 106 7 Page Figure A 22 Knocking oft PDMS solution cccccssscceessseccesssseeessssecesessseecesseuseeesesaeeesssssseeesesaaes 107 Figure A 23 Stirrer apparatus with various beakers cccccccesssccesssssecesesseceessssseeeesssseeeesssseesessaaes 107 Figure A 24 Budget Table ere 109 Figure A 25 RI of Sucrose Diluted in Water 110 Figure A 26 Run 1 of the sucrose test Stems showing events during Test 112 Figure A 27 Run 2 of the sucrose test Stems showing events during Test 113 8 Page I Introduction and Motivation With the extent of advancements made in both the pharmaceutical and diagnostic testing industries it is surprising that there are still 1 7 million tuberculosis related deaths per year worldwide and only 60 detection of all TB infected individuals This disparity between infection and its detection can be largely attributed
65. elow VOLTAGE OUT LEAC CHIP x500 million TIA 64 Channel Multiplexer LASER Figure VII 1 Block diagram of the multiplexing amplification and software automated control system The probe card s 32 different channels are fed into a multiplexing circuit that is digitally controlled by a Hytek data acquisition device which is in turn controlled by LabVIEW The output of the multiplexers are then fed into a multi stage transimpedance amplification system that converts the nanoampere sized currents to voltages that can be directly read into the computer and logged for subsequent data analysis B Multiplexing and Amplification Circuit The multiplexing circuit was originally designed by Matt Duwe a graduate student at CSU Although it was a helpful start there were many components missing that were needed in order to utilize the 45 Page circuit for the new LEAC chip system The first thing piece that was modified in the circuit was the addition of a level shifting stage In order for the Hytek device to be able to read the signals leaving the amplifier the voltage must first be input into a range of 0 4 096 volts Since the zero point of both the circuit and the chip tends to move vacillate the level shifter provides the user with an effective way to precisely adjust the starting point for data acquisition Secondly the circuit was difficult to maintain Solder joints repeatedly broke off and the damage
66. ence of leaking Figure IV 6 Fluorescing capillary fluid flow through permanently bound PDMS on glass slide left and fluorescing active fluid flow through permanently bound PDMS on LEAC chip right Difficulties were encountered once working chips were finished with the commercial etchant The plasma oxidation techniques that worked on older chips glass slides and unaltered SiN pieces did not work to permanently bind the new chips to PDMS Several measures were taken to eliminate all surface treatment variables and it was concluded that the surface chemistry of the chip must have been altered by the new fabrication process to make the originally successful plasma oxidation 25 Page technique invalid It is also possible that an organic scum was present on the surface although extensive cleaning of the chips with sonication and high powered plasma oxidation were attempted This issue is still in the process of being resolved and will need to be reinvestigated once additional chips are fabricated for testing Another problem encountered with the permanent bonding technique was that the working plasma treatment was conducted in a different lab where there were no mask aligners like those in the cleanroom This presented an obstacle because a relatively high level of precision was needed to successfully integrate PDMS channels with a chip without covering the metal pads needed for probing To ensure the best possible binding it was desired
67. ence of note was whether the middle area of the mask would be exposed during etching This variable will be further investigated in future experiments for optimizing fabrication flow and end products B Problems Encountered During Fabrication A number of problems were encountered during fabrication in the past year An attempt will be made in this section to briefly go over as many of these problems as possible and perhaps offer potential paths to try for future fabrication efforts The first problem encountered during the fabrication of the LEAC chip this year was the lack of availability of the desired wafer type A silicon on insulator SOI chip within the means of the project budget was sought for several months before finding a suitable solution An example of an SOI chip is pictured in Figure III 4 SiO2 Insulator Figure HI 2 Cross sectional view of an SOI wafer SOI was desired because it would most likely be single crystal silicon This would be advantageous for the LEAC chip for its lack of grain boundaries in the silicon layer which makes for a much better 17 Page photoconductor This would in turn increase the amount of current induced from the evanescent field likely improving the chip s ability to detect bimolecular films at a nanometer level Unfortunately the SOI wafers finally received yielded significant thickness variation between the device layers This caused the buried detector thicknesses to cha
68. epare surfaces for silanization 1 Thoroughly rinse substrate with a stream of methanol followed by a rinse with DI water Quickly place the substrate into a glass container containing a 1 1 0 3 0 7 ratio of acetone methanol water ethanol Make sure all substrates are covered with the solution and sufficiently separated from one another 2 Sonicate the solution for a minimum time of 10 min 3 Using a squirt bottle thoroughly rinse each substrate with methanol followed by a longer rinse with DI water Dry the sample with dry N2 gas 4 Place all substrates in a plasma chamber expose the substrates to 100 W plasma 900 mTorr for a minimum time of 5 min 5 Store the samples in a isolated environment preferably a closed glass petrie dish Surface silanization 99 Page This should optimally be carried out in a glass container because the silanization procedure will irreversibly transform glass from a hydroxyl terminated surface to a amine terminated surface A suitable container should be reused for all silanization procedures A glass petrie dish works well here 1 Prepare a 40 mL solution of EDS If other EDS solution volumes are required proportionally scale individual volumes a 37 mL anhydrous methanol b 2 mL DI H2O c 0 4 mL glacial acetic acid d 0 4 mL EDS 2 Pour the solution over all substrates making sure that all substrates are fully immersed in solution Leave substrates immersed for a minimum time of 20 min Longe
69. er The other primary goal of this senior design team was to integrate microfluidics onto the LEAC chip to give the device real time data collection capabilities in responding to a change in refractive index when fluid or a layer is present on the surface of the waveguide Successfully accomplishing this systematic integration required intensive revision of the probe station The probe card and microfluidics encompassed the goals that were the driving force for to the evolutionary design process of the probe station B First Probe Station Configuration Without Microfluidics The following entails the original specifications for the probe station as well as the steps taken to meet those specifications 1 Linearity Linearly align the LEAC sample contacts to the probes of the probe card The strategy implemented to ensure linearity was to fix the probe card in a position that was parallel to the edge of the probe station and then move the sample with respect to it using a rotation stage mounted to an XY manipulation stage XY motion enabled movement of the sample into a roughly accurate position under the card The rotation was then employed to linearly align contacts to all of the 32 probes Figure V 2 shows these physical configurations of the probe station 29 Page Rotation H Stage em 299 A H Fiber Launch sample 2 piezo controlied Figure V 2 Photograph of the sample holder and fiber launch system with translationa
70. er to provide the user real time output from the data It analyzes the data and presents it after all data from a pad is taken It will frequently show the Current DIO Channel to be one more then the data shown from the graph 95 Page B Wiring Diagram Figure A 16 Sample section of coded wires from the LabVIEW program C Code Sections Without a local copy of the software mentioned above LabVIEW will be unable to load the LEAC Data Recorder and Control VI All data blocks below are labeled onto the wiring diagram to help bring attention to where each is located Part A Variables This is where all main variables are kept to keep them out of the code loops so they don t get lost or overwritten This also allows local variables to be placed multiple times within the program and provide better control Part B Start Up Code This code was adapted from the Hytek proprietary software including creating a session with the iUSBDAQ The right half of this is dedicated to doing once per program run code which includes deleting a file and making a new one if the Start New File switch is turned to the Off position on the front panel It also creates a header on the lvm including time and date and some column headers for ease of reading Part C Once Per Run Code 96 Page This code is executed only once per run It controls which pads will be read from the variables Start Channel and End Channel It will also put the curren
71. ers cut down in size Because 32 pieces are made at once having a more even surface yielded more uniform features among batches The SU 8 mold created from soft lithography was then used to create a reusable master mold out of polydimethylsiloxane PDMS PDMS is a nontoxic optically transparent polymer that was used in all microfluidics fabrication for the device PDMS has a refractive index around 1 4 a feature of note for any baseline measurements of the chips The chemical formula of PDMS is shown in Figure IV 2 below where the bracketed n term represents the repeated monomer unit with typical values between 90 and 410 CH3 CH CH a 0 GH s CH3 CH CH3 Figure IV 2 Molecular structure of PDMS B Fluid Flow The channels designed for the lab on a chip device were compatible for either passive or active flow and both types were used for experiments The advantages to capillary flow are its simplicity independence from bulky pumping setups and small sample requirements Capillary flow is introduced to the microfluidic system by adding a small volume of fluid often less than 3 uL to the larger inlet reservoir with a pipette The channel fills spontaneously by capillary force and differential capillary pressure and fluid evaporation serve to sustain fluid flows with relatively 21 Page constant rates for about 30 minutes Figure IV 3 depicts the effect of passive flow and the creation eoni 13 of evaporating
72. es remove the PDMS stamp and clean surface a Using two sets of tweezers hold the chip into a PBS with tween solution and knock off the PDMS with the second set of tweezers 106 Page Figure A 21 Knocking off PDMS solution b Run the chip through five beakers on the stirrer holding in each for 3 to 5 seconds a Rinse in order of front left corner to back right corner going through the front two first The order of the solutions to rinse through on the stirrer is PBS with tween PBS DI water DI water DI water Figure A 22 Stirrer apparatus with various beakers c Immediately dry the chip with nitrogen gas 8 Cover in between steps to avoid dust falling on the surface BSA Blocking 107 Page 1 Place the chip in a small beaker and pipette enough BSA in PBS solution 45 mg mL to cover the whole surface 1mL is usually enough 2 Repeat steps 6 8 of First Dimension Patterning procedure a Note This step does not use a PDMS stamp Second Dimension of Patterning Bind Antibodies 1 Using the second dimension PDMS stamp that corresponds with the stamp used in the first dimension follow all steps 1 8 of the First Dimension Patterning procedure using the new PDMS and the antibody solutions instead of the antigen solutions Third Dimension of Pattering Detection Solution 1 Follow the BSA Blocking procedure replacing the BSA in PBS solution with the Detection Solution containing t
73. esessseesessseeseees 45 B Multiplexing and Amplification Circuit ccccccccssssccesssscceesssseecessseecesesseeeessnsseeeesssseeessssneeess 45 Cr Software EE 47 D Capabilities of LabVIEW Program 48 E Obstacles in Data Collection and Control 49 VII Biomolecular Testing cscccccsssscccssssssceesssssececsssseeesesuesessssssecesesssesesessseeseseuseeeeseaseeessssaseess 49 A Antigen Antibody Reacthons 49 3 Page Be Chip Surface Treatrenk geet egeteret gedet 50 C Fl rescent Assay Jrmagibg gereegelt e EA en 51 Di Printing le TE 52 IX System Characterization sccssicceccesess cess geen 55 A Circuit Characterization DEabt eseu ash evevaateaninivssieeeieseenndesueneatanates 55 Be LEAC Chip Characterization isciccicivesseicicasteasteesiticiveesteiest deet ieee gegen 57 X Conclusions and Recommendations for Future Work cescceeeccceeseeeeeeeeeeeeeeeeeeceeeeeseneeesaes 63 lte vele 63 Biomolecular dl 64 ARON TESS Dee EE EE 65 LabVIEW Programming orisni E E E aE aa o E Ea EE E TEE E R 66 Multiplexing Amplification Circuit 67 AROLE Ot E 68 Team Management 69 Da MERER E 1 o Lo 2418 ah AAE denen eegene eegene es 71 XLI Te ale Ee EE EC 73 Appendix A Project Management 74 Appendix B LEAC Chip Fabrication How 76 Appendix C Microfluidics Fabrication Procedures sssssssssssnssseersssresssrenesssrensssreesssreesssrrensssrens 77 A Complete schematic with dimensions of the PDMS mask
74. et Summary and Bill of Material 109 Appendix K Refractive Index Characterization ccccssccccsssssecesssssecceessseceessnseeessssseeessssseesesssaeeess 110 Appendix L MATLAB Code Used to Analyze Data 115 Gain Plot Analysis Code 116 AE E 117 al Rel hol reeche eeh eebe eeh eeben eebe 117 Currents Versus Dhotodetectors ttnt 118 Dritt Plo terea a ETE RAAE E E A ATEA AAR E E 122 Table of Figures Figure Il 1 3 D Structure of the LEAC chip cccsssscccsssssccessssseeessssseeeesesseeesssseeessssuseeesssaseeessssneeess 12 Figure llt LEACChip mask A SISNS seirer EE EEEE EEEE 14 Figure IIIl 2 Comparison of the single versus segmented buried detectors 15 Figure IlIl 3 Zoomed in view of the most used metal mask desen 16 Figure IIl 4 Cross sectional view of an SOI water 17 Figure III 5 Via etching procedure for exposing pads for probing csssccccsssssceesssssceessssseeessssseeess 18 Figure IV 1 Submitted mask design left and single zoomed in view of channel design right 20 Figure IV 2 Molecular structure Of PDMS ssssssssssnssesseesssensssrenrssreerssrrerssreeessrrnsssernsssrensssrrenssseens 21 Figure IV 3 Capillary flow through microfluidic Channels c scccssesssssssssessesessesecetsecstsecetsecersncetenceene 22 Figure IV 4 Active fluid flow apparatus Conflguration 23 Figure IV 5 Still shot from video of moving fluorescent particles ccsccccsessseccesssseeessssseeeesssseee
75. evolution of programs designed for the specific purpose of controlling the Hytek and multiplexer and obtaining data from the LEAC chip This data is displayed in a real time format to the user while saving raw data for further analysis This program was designed using LabVIEW v8 5 with a Hytek Automation iUSBDAQ U120816 including proprietary software available from Hytek Automation s website Version 1 0 LEAC Data Recording and Control VI 4 23 2011 Initial Setup Variables Run Variables E Sucia Fixed Variables Loop Control val Analog Read Channel mene T 2 eech DeviceType To set the program into E ata Stream Time s e nni z Jiusepaq AQ an infinite loop flip A 1 7 21 switch up To stop flip Save to File eee Ee Time ms Devicelndex switch down while ai oll er on mize Time Ms E ing 4 _Spring_11 Testing lvm A A J 0 Urabe SE 2000 On U tfil R E n Use current file Rehence Channel No Ges H Off Start new file CSV Filepath 4 _Spring_11 Testing csv Current DIO Channels Reference DIO Chan O00000 6999959 Bright green is high state dark green is low Read in binary left most button is DIO 1 Channel Std Deviation fo fo Mean _ Variance fo jo 1 1 1 1 1 5000 10000 15000 20000 25000 30000 Data Point EI E Figure A 15 Front Panel of LEAC Data Recording and Control v1 0 A Front Panel 93 Page This program was designed to be quickly and easily learned by havi
76. figurations to access waveguides The layout of the LEAC chip and the probe card dictate the way that the probe station must be configured to make a measurement on a given waveguide Figure 1 Schematic of the LEAC chip fo ell i a scil 3 mE EO i Figure A 4 LEAC chip schematic The figure above shows the details of the LEAC chip mask layout Each of the eight octets has the same details as those given for Waveguide 2 in the figure The two common pads on each waveguide are the pads to which a potential is applied during measurement The signal pads are connected to via other probes on the probe card and are switched through to collect data The probe card must be brought into contact with the appropriate pads and then the optical fiber must be moved into position to couple the laser into the waveguide The alignment procedure is complicated by the fact that the center of the sample is likely to be covered by PDMS micro fluidic channels This dictates that the probes of the probe card must always approach their target pads from the exterior of the chip 78 Page Due to the symmetry of the chip if a given configuration allows access to a given waveguide the sample can simply be rotated 180 using the sample mount rotation stage allowing access to a second waveguide Configuration 1 Access WG 1 8 4 5 This configuration is the most flexible as it allows access to four waveguides where other
77. form more sophisticated data analysis operations on it This was mostly done using MATLAB A description of the MATLAB code used in analysis as well as some of the actual code used follows The descriptions which follow are simply basic descriptions of what each code section s function is For information on how to use this code or how exactly it works the comments section of the code or the code itself should be consulted One thing which is usually good to do when collecting testing data on the LEAC chip is to obtain a calibration curve to convert the output voltages back to input currents To do this tests were conducted in which a high value resistance 1 Mohm was jumpered across a ground probe and a signal probe on the PCB part of the probe card A Keithley power supply was used to bias this resistor with a voltage for which the current falls in the range measurable by the amplification and switching circuit LabVIEW was then used to take ten second averages of the voltage output of the amplification and switching circuit This in addition to the current output reading from the Keithley power supply were written down Finally a Matlab script such as the one in Gain Plot Analysis Code was used to determine the parameters to convert the output voltage back to an input current This script also gives an indication of how well the data fits a linear fit by plotting it If the amplification and switching circuitry were perfect and there was minimal
78. guration of the probe station to accommodate fluid flow through several channels for sample delivery to multiple sections of a waveguide Finally in conjunction with work on biomolecules and antigen antibody bonding calculations and experiments should be conducted in an attempt to determine approximate diffusion and kinetics properties of tuberculosis molecules between the microfluidic channels and the waveguide surface Biomolecular Printing Biomolecular printing still needs to be improved for use in the LEAC chip system Due to the inability to use the electronic printer the PMDS punch printing method has the greatest likelihood for success in this system This method however still offers problems that need to be addressed for full system integration to be possible The accuracy and repeatability of the alignment of the two printing step is crucial for this method to allow for reliable data to be taken from the LEAC chip A more consistent method for creating the first dimension printing pattern would increase accuracy greatly A mask design and mold for this piece of PDMS with the reservoirs measured out to exactly correspond with the widening in the channels would be a large improvement on the previous method of marking the areas with a permanent marker This would allow for precise punches for the first dimension printing that correspond directly with the second dimension channels It would still be very important to ensure that the two PDM
79. han a few millivolts over the course of the entire test Circuit Drift Test Raw Data 10Hz Cutoff Frequency Low Pass Filtered Data 0 24 0 235 o eu 0 23 gt 0 225 0 1 2 3 4 5 6 Time ms x 10 Figure IX 2 Drift test of MUX AMP circuit 56 Page Tests were also conducted on the circuit to determine its time constant in switching between channels An oscilloscope trace of such a test is viewable in Figure IX 3 In this plot the trace which changes very quickly is the signal from the computer interface card to the switching channels The trace that changes less rapidly is the output of the circuit Note that the time constant has changed since conducting this test because the low pass filtering that was originally installed at the output of the circuit has been removed Nevertheless it is important to know the circuit s time constant in order to know how long one should wait between switching channels to start collecting data Thus it is beneficial to perform this type of test anytime a change to the circuit has been made as well Tests to determine the voltage offset gain and time constant of the circuit should be conducted anytime the circuit has been altered Krrrrrr OT Figure IX 3 RC response of MUX AMP circuit B LEAC Chip Characterization Several tests were conducted with the aim of characterizing unique working LEAC chips in terms of 57 Page their dark
80. he probe card remains fixed This presents clearance issues for nearly every piece of hardware in the system One would need to guarantee clearance between the fiber and the probes of the probe card Also the card would need to be extended far enough off the platen so that the platen did not contact the fiber launch There would also need to be enough room to be able to manipulate the controls of the fiber launch These adaptations have not yet been made and as such there may be less than obvious problems that appear during implementation Below a block diagram of this configuration is shown accessing WG 2 and 6 80 Page Probe Card Probe Card Figure A 7 Block diagram of configuration 3 Alignment Procedure To align the probe card to the sample and then the fiber to the waveguide to take a measurement assuming you have followed one of the configurations above follow these steps Keep in mind some of these steps will take practice to get right on the first time and others will require iteration even after significant experience The figure below defines X Y and Z to be used in these instructions Figure A 8 Coordinate definition for alignment instructions with the lucky testing dinosaur Lenny 81 Page 82 Page 1 2 3 4 5 Pre Align probe card If the probe card has recently been moved or adjusted ensure that the probe array is still roughly perpendicular to the edge of the platen by visually inspecti
81. he secondary antibodies Note If you will not fluorescently image the surface that day leave it covered in the dark and in a refrigerator 108 Page Appendix J Budget Summary and Bill of Materials Funds available to us this semester were 300 00 of senior reimbursement expense available through the ECE department as well as our graduate mentor s research budget if we could justify the expense as needed in furthering the research project The table below lists our expenses Item Company Cost MUX AMP PCB Advanced Circuit 48 80 Proto board Banana Plugs Mountain States 9 98 Electronics Edge Connectors Digikey 7 10 LEDs Switches Radioshack 11 79 Probe Card Alpha Probes 305 00 Foam Core Board for Michaels 7 00 Banners 6 0 537 62 SE 79 00 139 00 REENEN 52035 ee a mee A E Figure A 23 Budget Table Student expenditures will be applied for reimbursement and the remainder of our 300 00 will be used to pay back some of the expenditures from the research account 109 Page Appendix K Refractive Index Characterization The goal in creating the sucrose experiment was to accomplish the first real time detection of a changing fluid moving through the LEAC chip to better grasp how sensitive the detection was and to determine how much mixing occurred within the tubes while pumping solutions Using the LEAC system to detect the change in refractive index RI of a sucrose solution would allow u
82. here VM stands for Voltage Measurement and C stands for Channel Number The voltages in the dark voltages matrix correspond to voltages taken with as little light as possible incident upon the LEAC chip AP d AP AP AP d X AP DP CP light voltages A data matrix of the same format as dark voltages except o scorresponding to voltages taken when the light source which is being used o Sto test with is on microscope light coupled laser etc swhether to plot time series data for each channel splot all channels 0 do plot options has string allt an 1t splot all channels D don t do plot options does not have string Tall Sin it plot_all_ channels isempty strfind options all 1 generate linear least squares fit 119 Page linear fit polyfit gain_voltages gain_currents 1 sdetermine correlation coefficient of data R corrcoef gain_voltages gain currents splot gain results figure plot gain _voltages gain currents gain voltages polyval linear fit gain voltag es k least squares title strcat Linear Least Squares Fits Current num2str linear fit 1 uA V Voltage num2str linear fit 2 uA legend Obtained Gain Data least_squares title title Output Voltage Vs Input Current For Gain Data xlabel Voltage V ylabel Current uA sreturn from function if no data besides gain data was provided if nargin lt 4 return end sconvert dark
83. hips or even silicon nitride PDMS printing was tested on a silicon nitride slide to work around this problem This type of printing could be more applicable on a LEAC chip without a transparent surface A featureless piece of PDMS cut to the same size as the piece with channels was punched with a 0 8mm tool to have holes that would align with the widening of the channels that would overlay the waveguide on an actual chip This punched piece was used to print the antigens on the silicon nitride slide by simply pipetting different antigens into the holes and allowing for an impropriate incubation period Figure VIII 5 below shows one image from the results of this test Figure VIII 5 Fluorescent image of ESAT 6 and a ESAT 6 binding in channel widening from assay using PDMS printing with microfluidic channels Figure VIII 5 shows positive results however not all channels showed expected results This is likely due to the difficulty in aligning the channels with the printed spots from the PDMS punches and also because the silicon nitride that was used had not been surface treated causing lower affinity of the printed antigens to the surface This PDMS punch printing method is the only method used that is feasible for obtaining results on a LEAC chip however alignment of the two different printing steps needed is an obstacle that needs to be worked out more thoroughly in the future Alignment of the two printing steps needs to be done perfectly an
84. inge with a luer lock connection A small modification was made during fabrication 0 8 mm punches were used on either end to provide a tight seal around the tubing as opposed to the 1 5 mm inlet and 1 0 mm outlet punch sizes typically used for capillary flow The pump works by pulling up on the syringe plunger thereby creating a 22 Page vacuum which causes fluid to flow from the sample container through the channels and up into the syringe Throughout the development of the system configuration larger syringes were found to allow for more sustained periods of testing Figure IV 4 Active fluid flow apparatus configuration For both types of flow the movement of fluid can be modeled by equation four which relates volumetric flow Q to a pressure difference and a resistance term K Ab Q e 4 The resistance term K represents the viscous and frictional resistances within the channel to laminar flow neglecting the expansions at the waveguide interface This term is given by equation five 2m 192w a all ZS wh n 1 3 5 In equation five u represents the dynamic viscosity of the fluid and the w and h terms have been derived to replace a radius term because the channels are rectangular ducts instead of circular pipes The setup used for LEAC chip testing had a K value approximated at 4 56x10 kg m s For active flow with a syringe pump the volumetric flow could be altered by adjusting the speed of the
85. l modifications and revisions throughout the year to troubleshoot problems with newly integrated elements and to adapt to changing goals 27 Page Microscope Sample Holder Fiber Launch Probers Base Figure V 1 Original probe station where each probe has independent cantilevered XYZ motion With magnets holding the probes in place on the large steel base one probe was used to apply a potential to a common pad on the LEAC sample as the second was brought A mechanical probe station is a device that physically acquires signals from a semiconductor A probe devices with precise positioning of probe needles onto the surface of the device station can sit on a laboratory table or can be a table in of itself In general some part of the probe station is fixed in place and another plane of the station will move very precisely with respect to the stationary portion A microscope is usually mounted on the station or the table and allows the user to view the probes as they make contact with the sample This precise motion capability can be achieved with springs and micrometers similar to those used in calipers or through the piezoelectric effect if greater precision is required Prior work on the LEAC project had been performed on a table mounted probe station that implemented two independently manipulated manual probes and a manual or piezoelectrically controlled fiber launch as shown in Figure V 1 above One probe was bro
86. l and rotational stages that provided the necessary movement of the sample for appropriate alignment 2 Planarity and Contact Allow precise vertical motion to bring probes and sample into contact while ensuring acceptable planarity tolerances The required vertical motion was provided through the use of the metal platens of the original probe station which could move vertically with respect to the table surface Once the sample is properly aligned the probe card is brought down into contact with it A benefit to this design is that it allows the microscope to stay in focus on the sample during the entire setup procedure something that becomes critical in subsequent steps The use of the platens provided a unique balance between travel distance and precision they can travel at least 4 with a precision under 10 microns while other similar translation stages could only provide about 1 5 of travel at the same level of precision Figure V 3 shows the probe card and metal slat that provides it with vertical motion for pad contact 30 Page M Metal slat of probe station E N Figure V 3 Photograph of probe station configuration at this phase in its evolution The piece labeled metal slat has vertical motion allowance Planarity was a difficult issue to solve as the weight of the extended portion of the probe card tended to pull down on the more distant probes causing them to come into contact first Despite this feature it was f
87. ld only be driven to 3 5 mA beyond the typical threshold current as given by its data sheet The laser diode used by this team had a typical threshold of 35 mA and was routinely driven at 37 5 84 Page 10 mA and 39 6 mA at an absolute maximum The gain and the exposure time of the LUCAM capture program should be set to their highest setting to allow viewing of the laser coupling into the waveguide Use the precise motion controls on the fiber launch to precisely adjust the fiber position until the laser is coupled into the waveguide Explore the coupling space to find the position that gives the best coupling Take a snapshot using the LUCAM to compare to throughout the experiment It was observed that the piezo stage as well as other non idealities caused the fiber launch to drift and thus the coupling to degrade over time Tubing placement if necessary If the desired experiment requires the use of micro fluidics they will be added at this point Make sure the microscope is in position just above the entrance to the waveguide and in focus to observe coupling The microscope must be left perfectly still now because the tubing will be connected to it Thread the tubes through the tube guides connected to the objective lens of the microscope and guide them into the punched holes in the PDMS Recheck coupling The connection of the tubes will usually move the sample enough to at least decrease coupling Finely readjust the fiber using the methods
88. lic fingers on one edge of the card If making a custom probe card the PCB can be layed out using PCB layout software such that discrete components or entire circuits can be soldered directly to the probe card The metallic fingers which are electrically connected to the probes of the probe card can be accessed through a card edge connector Figure A 9 shows the probe card that was received to be used in conjunction with the LEAC chip with a card edge connector Figure 6 4 in section VI shows a more zoomed in view of this probe card Figure A 9 Probe card as received from Alpha Probe There are two main types of probe cards The traditional blade probe card has superior mechanical stability and a higher integrity signal path than the more recently developed epoxy ring probe card 85 Page To make a blade probe card ceramic blades are inserted into lands on the probe card and soldered down These blades have probes sticking off them which are bent into the required shape for the component to be probed Because of their superior mechanical stability blade probe cards are less likely to need re planarization adjustments Blade probe cards are inferior to epoxy ring probe cards when it comes to probe densities and probe counts blade probe cards generally have a probe count less than 88 whereas epoxy ring probe cards can have probe counts greater than 2000 The epoxy ring probe card technology used to be more expensive but with r
89. lly overcome by purchasing a commercial etchant called SILOX VAPOX III from Transene This proved to selectively etch the SiO over the aluminum IV Microfluidics Fabrication One of the primary goals for this lab on a chip device was to bring the unit into real time detection In previous experiments the LEAC chip s technology was confirmed by laying a solution of antibody down over the entire chip and after appropriate incubation printing a corresponding antigen to the surface with a microarray printer This technique though useful as a proof of concept test that had a detection limit of 120 pm is not feasible for actual implementation of the device in the medical field By designing and integrating a microfluidics element to the device a blood or serum sample could potentially flow through channels of micrometer width for delivery to the waveguide wherein any present antibodies in solution corresponding to the antigens printed to the waveguide would bind and be detected by a change in photocurrent A Microfluidics Mask Design The first step towards achieving real time detection on the LEAC chip was to design and order a mask for microfluidic features The original mask design had to be small enough to fit within approximately 9 5 om so as to not cover the metal pads which would prevent probing and thus measurement Because the features used up all available space a cutting guide was also added to improve the precision of th
90. long as it is true the program will continue taking data runs until halted or switched off again Part End Code Proprietary Hytek code that is used for shutting down the Hytek This keeps the device from having errors for future runs 97 Page 98 Page Part A initialize Variables Part B Start Up Cade Pan D Changing to Current Pad Part C Once Per Run Code Part E Data Recording Part F Changing to Reference Pad Part G Reference Pad Data Recording Figure A 17 Flow Chart of LabVIEW code Part l End Code Appendix H EDS SMCC Surface Treatment Protocol This protocol was provided by Dr Lynn This protocol allows for a cheap convenient method for the covalent attachment of proteins via external thiol groups to hydroxyl terminated surfaces The bulk of this protocol was adapted from the work of Wu et al Wu et al 2006 Biosensors and Bioelectronics 21 7 1252 1263 This protocol has been successful with both glass and silicon nitride surfaces and can theoretically be possible with metal oxide surfaces or any hydroxyl terminated surface Optimally the starting material should be free of defects and macroscopic particles Required reagents EDS N 2 aminoethyl 3 aminopropyl trimethoxysilane Obtained from Gelest Morrisville PA sulfo SMCC Sulfosuccinimidyl 4 N maleimidomethyl cyclohexane 1 carboxylate Obtained from Pierce Biotechnology Pr
91. lorado State University Chemical and Biological Engineering Department Staff Faculty Postdoctoral Fellow P Wu Hogrebe Grainger DNA and protein microarray printing on silicon nitride waveguide surfaces 2005 Department of Chemistry Colorado State University 72 Page XII Acknowledgements This project would not have been a success without the support and advice or Dr Kevin Lear and Dr Dave Kisker They provided a unique senior design opportunity that was certainly a learning experience for all involved The team would also like to thank the graduate students of the optoelectronics lab Rongjin Yan Weina Wang Iris Yi Joel Kindt and Tim Erickson for their assistance in fabricating working LEAC chips training and guidance There were other contributors outside of the group that put a lot of time and effort into this project for no personal gain A special thanks is in order for Andy Nott who was invaluable during the difficulties with the probe card specifications and purchase and last but certainly not least Dr Scott Lynn for the countless hours he spent training explaining and encouraging the chemical engineers in the microfluidics and biological aspects of this project 73 Page Appendix A Project Management Below is a list of the tools and methods that we used to increase our efficiency as a team and properly manage out time Electrical and Computer Engineering Chemical and Biological Engin
92. menisci in the two reservoirs Inlet Capillary Driving Force Outlet Capillary Driving Force Flow Direction oe Figure IV 3 Capillary flow through microfluidic channels Although the evaporation can serve as an important driving mechanism for fluid flow there can be consequences of evaporation on protein or analyte concentration Evaporation on the inlet side can be limited by adding a small film of mineral oil over the sample fluid Passive flow was used for all immunoassays which are detailed in section VII and Appendix I It was found through fluorescent imaging that fluid would not flow passively through the PDMS channels when either reversibly or irreversibly bonded to a LEAC chip It was suspected that either a strongly hydrophobic chip surface or chip features significant enough to hinder capillary forces were responsible for preventing flow Further investigation of this phenomenon should be investigated in the future Because of the restraints on passive flow and the need to continuously switch between sample concentrations for a specific test active flow was implemented for this device by employing a syringe pump The available in house pump was reconfigured to pull fluid through the channels rather than to push it through to ensure a better seal between the PDMS and chip Figure IV 4shows a schematic of the pumping system where PEAK tubing extends from one reservoir to the sample and from the other reservoir to a syr
93. mporally to each other should be implemented A device like the lock in amplifier could be considered For taking data during which a change was supposed to occur sometime during the test it is useful to filter the data in order to more easily see when such changes occur To help with this a digital filter with modifiable cutoff frequency was implemented in Matlab This filter was a symmetrically applied finite impulse response raised cosine filter A plot of the amplitude spectrum of filtered and unfiltered data can be seen in Figure IX 9 below 62 Page 60 Hz Interference e f Harmonic of 60 Hz dm ap D L T BS P kd D ei a E E Se 20 30 20 20 D ww 20 oO Au 500 D a Magnitude Squared of Frequency Spectrum of Low Pass Filtered Data Ss a k i euch KS E D a Al i 50 lt 0 20 0 100 0 100 200 wg d s Frequency Figure IX 9 Magnitude squared of filtered and unfiltered data X Conclusions and Recommendations for Future Work Given that no previous senior design work had been done on this project previously and considering the feats accomplished this design project was a success In terms of LEAC chip fabrication an improved procedure that has produced a few working chips has been implemented However there remains many steps to troubleshoot and optimize before a fully successful and repeatable fabrication flow is attained in all of the disciplines of the project
94. n length run run_count dcl length run run_count dc2 for count l length run run_count dcl dark currents std count std run run_count dcl count 0 2 end for count l length run run_count dc2 dark currents std countt length run run_count dcl std run run_count dc2 count 0 2 end 126 Page dark currents std mean dark currents std 2 scalculate light currents erros excluding drift light currents std zeros nch_scan length run run_count 1c for count 1l length run run_count 1c light currents std count std run run_count lc count 0 2 end light currents std mean light_ currents _std 2 calculate photocurrents errors excluding drift photocurrents errors no drift sqrt light currents std 2 dark currents std 2 sqrt 500 sls at 500 hz photocurrents errors no drift std run run_count pc 1 0 2 std run run_count pc 2 0 2 5 sqrt 1000 sls at 1000 hz scalculate photocurrents average photocurrents 1 mean run run_count pc 1 2 photocurrents 2 mean run run_count pc 2 2 photocurrents average mean photocurrents 1 photocurrents 2 2 ScConseruct plot figure fcount fcount fcount 1 errorbar ch_scan photocurrents average photocurrents errors hold on errorbar ch_scan photocurrents average photocurrents errors no drift r errorbar ch_scan photocurrents average light currents errors c errorbar ch_scan photocurrents average dark
95. n be found in Appendix J The sulfo SMCC cross linker is then added to the surface in a sodium borate buffer This final step adds a thiol group to the surface of the chip which will immobilize any sort of biomolecule that contacts the treated surface Figure VIII 2 below shows the final chemical structure present on the surface of the chip after this treatment R SC Thiol group captures and on immobilizes antigen Ge ar O H E _ 0 Si O Cross links PDMS o Glass Figure VII 2 Molecular structure of chip surface treatment C Fluorescent Assay Imaging Before running any tests on LEAC chips it was critical as a proof of concept to verify binding with fluorescent microscopy on a cheaper surface such as a glass slide or piece of silicon nitride Understanding the binding affinity of the antibodies and how they act together is important to understanding what to expect once the reactions are tested on the waveguide of the LEAC chip To do this an assay was performed using microfluidic channels in a grid pattern Figure 8 3 below shows the results of this assay 51 Page AG85 ESAT6 0 2 mg mL BSA 0 2 mg mL OCC d a gf a AG8S 0 2 mg mL BSA a ESAT6 0 2 mg mL Figure VIII 3 ESAT 6 and AG 85 antigen antibody assay with rat a mouse control line full assay fluorescent image right cropped and labeled fluorescent image of TB antigen antibody assay Eleven channels were used to print the antigens in
96. n run_count dcl end 2 ch _scan mean run run_count pc 2 2 mean run run_count dc2 1 2 legend Test 1 Test 2 title strcat Normalized Photocurrents from run run_count run_ title xlabel Detector Number ylabel Currents uA saveas gcf strcat run run_count folder photo normalized fig saveas gcf strcat run run_count folder photo normalized jpg splot photocurrents with error bar calculate dark currents errors including drift figure fcount fcount fcount 1 dark currents zeros nch_scan length run run_count dcl length run run_count dc2 Sinitialize for count l length runf run_count dcl dark currents count mean run run_count dcl count 2 end for count l length run run_count dc2 dark currents counttlength run run_count dcl mean run run_count dc2 count 2 end dark currents errors std dark _currents 0 2 scalculate light currents errors including drift light currents zeros nch_ scan length run run_count 1lc Sinitialize for count l length run run_count 1c light _currents count mean run run_count lc count 2 end light currents errors std light_currents 0 2 calculate photocurrents errors including drift photocurrents errors sqrt light currents errors 2 dark currents errors 2 photocurrents errors photocurrents errors sqgrt 2 scalculate dark currents errors excluding drift dark currents std zeros nch_sca
97. nction leac_data_analysis new options gain voltages gain currents Fs dark_ voltages 1 ight voltages Sfunction leac data analysis new options gain _voltages gain_currents Fs dark voltages 1 ight_voltages Daniel Higley 03 2011 o oP ole 118 Page sTakes data output from LabVIEW and performs some analysis on it sSpecifically plots time series of the currents for all channels a gain splot and dark light and photo time averaged currents vs detector number Also plots the noise on each channel for the dark light and photocurrent o measurements Input arguments format dp d WP dp o soptions string which determines wether to plot time series data for each schannel see first code segment below gain_voltages data vector of voltages taken from LabVIEW gain_currents data vector of currents taken from Keithley power supply Each entry corresponds to the same indexed entry in gain voltages j a frequency of sampling for dark voltages and light voltages AP dp d d d dark voltages Data matrix taken with frequency of sampling Fs This is o Sdesioned to read voltages in in the same way that LabVIEW outputs them o Sin other words if LabVIEW output out csv for a given dark voltages sobtaining run then setting dark voltages load out csv will work fine sThe exact format of this is VM1 C1 VM2 C1 VM3 C1 VM1 C2 VM2 C2 VM2 C2 VM1 C3 VM2 C3 VM3 C3 W
98. nd likely inaccurate PECVD machine in the cleanroom This large open space allowed for a region large enough to use the Filmetrics device to measure layer thicknesses on the chip This measurement not only confirmed that enough material had been deposited onto the chip but also allowed for observation of the etching progress after small steps This advantage will become obsolete once consistent fabrication flows are completely developed and characterized but is invaluable for the current stage of fabrication The bottom right quadrant of the mask design pictured in Figure Ill 1 creates the waveguide All nine of these masks are very similar to each other differing only in the width of the waveguides Different widths were used in fabrication throughout the year but once experimentation was 16 Page feasible the 5um waveguide width was needed to increase the chances of successful laser coupling after polishing the chip as best was possible The masks can create waveguides that range from 2 5 to 5 5 um wide Finally the last quadrant of the mask design the upper right quadrant of Figure Ill 1 represents the via layer This layer is the very last step in chip fabrication as it is the mask used to etch material down to the metal pads to expose them for probing The differences between these masks are the height of the vias so that one matches to each corresponding metal pad height whose differences were discussed previously Another differ
99. nect to When a significant amount of oxides or other debris are collected in the probes it can destroy functionality and consistency of testing To fix this the probes should be cleaned when contact issues are a noticeable problem or noisy data on channels that used to not be so noisy becomes apparent There are a few easy ways to clean these probes One is by rubbing 3 microns grit silicon type lapping paper like sand paper across the probes to dislodge oxide This was the only method used for cleaning the probes previously and seemed to work fairly well If a different method is desired or this one proves to be ineffective sometime one should contact the vendor to determine what to do If a new probe card is ever ordered there are a few things which could be done differently and would possibly be beneficial The first is changing the type of probes from tungsten to tungsten rhenium Tungsten rhenium probes are less fibrous and not as prone to collecting oxides as tungsten probes This means they require less time between cleaning and possibly last longer Another thing which should be investigated is the exact location at which the edge sensors are placed on the probe card Currently they are sometimes close to coming in contact with a waveguide on the LEAC chip when they are set down It is very undesirable to come down with a probe on one of the fragile waveguides Avoiding this is relatively simple since the waveguides are so small but t
100. ng dark currents Probe Card There are a few things which could be investigated in the future related to the probe card but these are unlikely critical to the project One of the things which could be first investigated if it is shown to be necessary is looking into how to fix the currently bent probes on the probe card These seven probes were inadvertently bent during testing and lie in the middle of the set of probes They correspond to the middle of a set of photodetectors for any waveguide on the LEAC chip Thanks to the redundancy of the LEAC chip it is not imperative that these probes get fixed in order to collect useful data However if data on more channels or data on the currently inaccessible channels is desired fixing these probes should be investigated The vendor through 68 Page which this probe card was ordered says that if the probes are not too bent they can unbend them and if not they can simply replace the broken ones They also said that the cost of this is determined largely on a case by case basis If more knowledge of how much this would cost is desired the vendor should be contacted again and probably sent pictures of the broken probes An important thing to keep in mind with the probe card is that its probes require periodic cleaning in order to maintain function Currently the probes on the probe card are tungsten probes which are relatively fibrous and prone to collecting oxides from the aluminum pads they con
101. ng from above Place sample on sample holder Place the sample on the glass slide on the silver rotation stage of the sample holder It is recommended to secure the sample with double sided sticky tape A small piece of tape should be used and it is advised to touch the surface of the tape with your finger to decrease adherence To combat planarity issues it is also recommended that the glass slide and the bottom of the sample be inspected for old tape or adhesive residue Pre align sample card and fiber This is less of a concern with the additional freedoms recently added to the probe station It is still a good idea to move the sample holder roughly in place below the probe array and ensure that the fiber can reach its necessary position One can adjust the overall system at this point by moving the large aluminum plate holding the sample and fiber launch Zero the contact pads in the X dimension If the microscope camera is well aligned to the XY plane of the probe station the edge of its viewing area on the computer monitor is a good reference line While viewing the desired contact pad array through the microscope camera read out on the computer monitor align a pad array with the edge of the viewing screen using the rotation stage of the sample mount This can be done with probe card and fiber launch at a safe distance and usually will not need to be adjusted again Align the probe card to the pad array Coarsely adjust the probe card in the X
102. ng the information placed much like a book Starting in the upper left hand corner is the box labeled Initial Setup Variables which can be seen in Figure A 16 Analog Read Channel is determined by which Analog Input is hooked up to the sensor probe and can be read directly off the Hytek Save to File indicates which filepath the LabVEW Measurement File Ivm will save to This file will contain a summary of which channels were read at what time each set of data began as well as the data s mean standard deviation and variance Below Save to File is a switch with the off position deleting any file found at the save to file location before starting a new file The on position will instead append all data to the end of any file currently found there and if no file is found will start a new one The last variable the Comma Separated Values csv filepath sets where the csv will be saved This file will contain all accumulated data points for every run In the middle left box are the Run Variables shown in Figure A 11 containing parameters that can be adjusted between runs Start Channel and End Channel refer to which pads over which the data will be recorded Data will be taken sequentially from the Start Channel and end with the end Channel This will constitute one run of data Data Stream Time dictates for how long any one pad will have data recorded The minimum time is 0 1 seconds and infinite time as dictat
103. nge by tens of nanometers from one detector to the other Fortunately this issue should theoretically not be a major limitation as the device simply senses changes in the signal on each detector as opposed to absolute currents Another problem encountered during fabrication was the lack of etch control available when performing the final critical via etching step This step calls for etching through 200 to 1500 nm of SiO without etching into the metal pads used for probing that are connected to the detectors beneath the waveguide Figure IlI 5 below displays the etching process described AZ2070 Prevents Etching of rest of chip N SiO2 Insulator SiO2 Insulator SiO2 Insulator Figure III 5 Via etching procedure for exposing pads for probing Hydrofluoric acid HF was initially employed in an attempt to etch through the SiO because it is strong enough to etch through glass quickly However the HF could also etch through the relatively thin layer of aluminum contacts which rendered the chips useless The first attempt to solve this problem was by using different metals such as platinum as stop layers for the HF This solution may have been feasible had more advanced metal deposition equipment been available The 18 Page electron beam evaporation system in the CSU cleanroom was not working and many of the neighboring university labs only had aluminum available for deposition This problem was fina
104. ns in the appropriate waste container This container can be found in cabinet and is labeled appropriately a Rinse with acetone using the squirt bottle b Rinse with methanol using the squirt bottle c Rinse very thoroughly with DI water using the squirt bottle 3 Holding in the thin tweezers dry the surface off using the nitrogen tank Note Be sure to have a tight grip on the chip or slide Do not drop it so it gets dirty again or scratched gt Note if you are concerned about the surface still being dirty sonication is also a possible precaution to take First Dimension of Patterning Applying Antigens 1 Place the PDMS channels needed for this patterning step on the edge of the glass slide marked chamber slide a Make sure the PDMS is place channel side up and is very clean If it is not clean use tape to pick up any dust off of the surface 2 Place the chamber slide with the PDMS on it into the plasma chamber places between the metal grates shown below and close the plasma chamber door 104 Page wy Figure A 19 Inside Plasma Chamber 3 Run the plasma chamber at 750 mTorr and 20 Watts a Turn on the plasma chamber by first closing the green handle same as above in the tuning section plugging it in pressing the AC On button and the Start button b With the dial set to PRES wait for the pressure to pump down to 750 mTorr then switch the upper right hand knob labeled MAN
105. ns serve two purposes First they allow for a wider detection region on the waveguide by overlying a greater number of detectors than if the 100 um width had been maintained Second the expansion causes the fluid flow to slow significantly and this slowing of the sample provides for a longer residence time This residence time is needed to allow any present antibody to diffuse to the surface and react with the antigens attached to the waveguide The reaction of an antibody and its corresponding antigen is governed by kinetics Affinity coefficients k of antigen antibody reactions are typically on the order of 10 10 Msec Once the mask design was completed it was ordered as a transparency from Fineline Imaging An SU 8 mold was made by spin coating SU 8 onto a silicon wafer Because SU 8 is an epoxy based negative photoresist the portions in Figure IV 1 in white which were exposed to UV became 20 Page insoluble in the developer leaving a mold of raised channel features approximately 35 um tall The full procedure for making an SU 8 master mold can be found in Appendix C along with more detailed dimensions of the channel mask design An SU 8 mold was originally made in the CSU cleanroom However due to limitations on the maximum wafer size that could be spun the mold used for all future work was made using soft lithography in another lab on campus because using full circular wafers provided a more even coating than silicon waf
106. ntered with this setup was that the initial alignment procedure assumed it would be necessary to have the chip at the very center of the sample holder rotational stage To achieve this centered positioning several layers of glass were stacked and taped together This led to some observed instability in the sample when trying to make firm contact with the probes The second problem was the proximity of the bottom of the platen to the other pieces of hardware As can be seen in Figure V 3 above the probe card was set so far to the right on the platen that the left side of the platen was almost touching the sample holder when the probe card was in position D Second Probe Station Design with Microfluidics Integration To obtain real time data from a LEAC sample microfluidic channels as detailed in section four needed to be integrated into the system described above without disturbing or conflicting with any of the existing components The marriage of semiconductor and circuit technologies with 33 Page microfluidics presented some unique challenges that dictated the design of both the probe station and the details of the channel design It was found through preliminary calculations that active flow would be desirable as opposed to capillary driven flow Active flow put greater strain on the bond between the PDMS and the surface of the LEAC chip and to complicate matters further the interfacing of tubing to the PDMS channels tended to mov
107. o make sure this does not happen if a new probe card is ordered other positions should be looked into Team Management One hindrance to the project s progress this past year was the learning curve for the new team members and the time it took each to get up to speed on the theory technology and laboratory 69 Page techniques required by the device It would be useful to spend the first month or two reading papers presenting materials and learning basic science techniques in the laboratory as opposed to learning them haphazardly and individually along the way In addition while experiments are critical to progress it would have been useful to the validation of conclusions if the experimental work were supported by analytical solutions and simulations In terms of weekly meeting times revisions should be made to the team meeting structure to yield more efficient communication and progress While it is still relevant to alternate team leadership responsibilities it would be useful for all team members if more formal presentations of results were required at weekly meetings and all deliverables included with weekly progress reports It might also be useful to move these meetings from the middle of the working week to the very beginning in order to more clearly define goals and strategies for each week The Gantt chart was useful and should remain an integral component of the project management Finally if the team is to remain interdisciplin
108. obe configuration to avoid interfering with the device under test in case it contacts a conductive part of the device under test B Commonly used probe card parameters terminology and abbreviations Balanced contact force BCF Allowable deviation from average contact force across all probes Beam length Beryllium copper probes Blade probe card Contact resistance Depth DUT Edge sensor Edge Setting Epoxy ring probe card 88 Page The distance from a probe tip to the epoxy of an epoxy ring probe card Probes that are softer than standard tungsten probes but can be used for lower contact resistance or higher current applications than tungsten probes A specific type of probe card topology which has been overtaken by epoxy ring probe cards in popularity Attaches probes to be used to blades which are mounted into a PCB The resistance of the junction from a probe to the pad it is contacting at zero applied voltage The vertical distance from a probe tip to the bottom of the PCB Device under test In the case of probe cards this generally references the chip under test interface probe card with external circuitry A configuration of probes used to detect when probes are making contact with a device under test Specification of type of edge sensor A specific type of probe card topology which is currently very popular Has a metallic ring to which probes are precisely epoxied into Mil Planer
109. ound that all probes could generally be brought into contact effectively and within the safe flex distance of the probes so no changes were made to account for this 3 Fiber Coupling Once contact is made between the probe card and LEAC sample allow the fiber to be moved into position to couple the laser into the waveguide A fiber launch with precise XYZ motion capability either manual or piezoelectric was used to move the fiber into place once the probe card was effectively aligned The fiber launch was adapted such that its tallest structure was the optical fiber as shown in the diagram of Figure V 4 31 Page Figure V 4 Block diagram of probe station configuration and portions with movement The purpose of this design was to ensure that the fiber launch would not interfere with the probe card alignment and that they would not contact or damage each other A microscopic image of the pads in contact with probes with successful alignment of the fiber with the waveguide is seen in Figure V 5 below 32 Page Figure V 5 Probes are aligned and in contact and the fiber is aligned to the waveguide prepared for proper coupling of the laser C Problems Encountered With the First Probe Station Configuration The first configuration of the probe station met the initial design specifications and was used from late November 2010 until mid March 2011 when integration of microfluidics began to be realized One of the problems encou
110. ow a person can only sweep through channels making it cumbersome to debug individual pads or probes The reference channel logging is also currently inoperable and needs to be fixed Currently when the program logs the reference channel measurement it only saves it to the nearest integer removing any usefulness Finally a feature should be added that displays a graph with the history of the channel that can be used to compare with the current channel so that drift can be observed in real time and adjusted for accordingly Saving the data in a useful format is another consideration It is important that all of the raw data is saved for the entire run not just the mean and standard deviation Any future revisions of the code 66 Page will need to ensure that all data is saved in a coherent manner for both the signal channels and the reference channels if used The last and possibly most major improvement would be the incorporation of MATLAB m files as blocks within the LabVIEW interface It could then handle the data directly and process it in real time enabling the user to know the results of the experiment as it is happening This would allow for better response to experimental setup errors as well as save a lot of time that is currently needed for data manipulation and analysis after the experiment has already been completely performed Multiplexing Amplification Circuit The current multiplexer and amplification circuit appears to
111. part however the translation stages under the fiber launch are rather slow moving and difficult to reach In addition the sample holder is still not fixed to the base plate of the probe station Using rail type control stages may be a good way to provide the desired freedoms and make the station more usable The advantage of rail type control stages is that they are quicker and easier to adjust and generally have an increased range of travel compared to micrometer controlled translation stages One possible configuration would be to add one rail type control stage under the sample holder and another under the fiber launch The one under the sample holder would be used for coarse adjustment in the Y direction of the sample to the fiber The one under the fiber launch would allow coarse X direction adjustment Fine adjustments would then be performed using the micrometer controlled XYZ capability of the two pieces The other probe station problem that should be addressed by future work is accessing waveguides 2 and 6 As detailed in the probe station user s manual the current probe station hardware cannot access these two waveguides on a PDMS bonded LEAC sample This is due to the configuration of the LEAC sample contact areas waveguide entrances and probe card probes To allow the various pieces of hardware to fit in Configuration 3 as defined in the probe station user s 65 Page manual the fiber launch should be adjusted in such a way that
112. r ian aes 14 Ay LEAC Chip LEE EE EE EE aad 14 B Problems Encountered During Fabrication 17 IVs Microfluidics Fabrications 00heesaindnnwAns SEENEN ed 19 A Microfluidics Mask D sigtt ciincdtindndadedidadadtinindiciatidadadndanadtiadadad 19 B Fluid FlOW ss sce ekadansandadadadwtadadadwdadwdad edn data dasadetadatedasadannn ce 21 C Fluid Flow and Leaking os 02 4 Sada e de tad tach dads dere deh denen dean 24 V Brebe Statlett eg ee akan hens wean AWA aH ONAN oa SHAS Anan ana ad ae AN an ahaa an oad 27 A Introduction to Mechanical Probe Statlons 27 B First Probe Station Configuration Without Microfluidics c cccccssscceeessseceeesseecesssnseeessssseeees 29 C Problems Encountered With the First Probe Station Configuration cssccccsssseceesssseeeessees 33 D Second Probe Station Design with Microfluidics Integration 33 E Problems Encountered With the Second Probe Station Configuration cccsccccsssseceeesseeees 35 Fe Final Probe Station Configuration cec cccdeicecieicicesieacieeceeisiievesciiecisadiiiies ENNEN 35 ME Probe Card Reguitetttebtg geesde egenen 37 A Benefits of a Probe Card uge 37 B Probe Card Design Specifications and Selecton 40 C Problems Encountered With Probe Card Specification and Implementation ssosnnnseseeeseeeeee 43 VII Signal Amplification and Switching Circuit 45 A Signal Amplification and Channel Selection ccccccccsssscccssssseeecsssseceesssseeesssssee
113. r immersion times have no apparent effects on the eventual surface activity 3 Rinse each substrate with methanol followed by a thorough rinse with DI H20 Dry the substrate with N2 gas 4 Place all substrates in a 115 C oven for 20 min 5 Remove substrates from oven and allow for relaxation to room temperature If needed the samples can be stored for extended periods in a dark desiccated environment Activation of surface with sulfo SMCC This should also be carried out in a glass container that has not been silanized from the above step Again a glass petrie dish works well 1 Prepare 50 mM sodium borate solution a Add 1 55 g of boric acid FW 61 83 g mol to 500 mL DI H2O b Adjust pH to 7 5 with appx 1 2 mL of 1 M NaOH 2 Prepare sulfo SMCC solution from no weigh sulfo SMCC vials Dissolve 1 vial 2mg of sulfo SMCC into 20 mL sodium borate buffer Stir until no solids remain 3 Fully immerse all substrates in the sulfo SMCC solution for a minimum time of 3 hours 100 Page Keep the solution isolated from all light sources Longer immersion times can be used with no detrimental effects 4 Rinse all substrates sequentially with sodium borate and DI H20 Dry substrates with N2 gas 5 Store samples in a dark desiccated environment at room temperature Samples will remain active for up to two months 101 Page Appendix E Biomolecular Assay Procedure Background Biomolecule printing is an important a
114. rcos filter ccd Fs r plot_title strcat data_type Currents for Channel num2str count title plot title legend Unfiltered Data Filtered Data xlabel Time s ylabel Current uA end end sdetermine noise on each channel and plot figure plot 1l size currents 1 std currents 0 2 b l size currents filtered 1 st d currents filtered 0 2 r piot title strcat data_type Currents Noise On Each Channel Vs Channel Number title plot title legend Unfiltered Data Filtered Data xlabel Channel Number ylabel Current uA end 121 Page Drift Plot function run tests_april data_to_plot tests april data to plot Daniel Higley 04 2011 ae Ae ole o sUsed to convert run data from some of the tests conducted in April on LEAC chips sto a more usable format SInput data from folder directed to in switch case block is Sdark1CSV CSV containing data from the first series of dark current runs sDifferent rows correspond to different channels Different tests are stacked on top of eachother Slight lCSV CSV containing data from the series of light current runs sDifferent rows correspond to different channels Different tests are stacked on top of eachother dark2CSv CSV containing data from the second series of dark current sruns Different rows correspond to different channels Different tests sare stacked on top of eachother data ouput format is C D
115. rd epoxy ring probe card topology instead of a blade probe card topology With this choice huge differences for our application between these two technologies was not observed and in the future either one could be considered depending on the evolving need of the device 41 Page Another specification consideration for the probe card was the number of probes There are a over 300 independent signal pads on a LEAC chip However because the price of a probe card scales with the number of probes ordering a probe card which could connect to all of these pads simultaneously would have been extremely cost prohibitive not to mention challenging in implementation Fortunately the current LEAC chip has an easily exploitable symmetry which allows all of the probes on a LEAC chip to be probed by translating and rotating a probe card with a much smaller probe count and a much simpler probe configuration Specifically the LEAC chip has eight different waveguides which have pad locations with the exact same orientation relative to each other Each set of pads for an individual waveguide has two pads which are connected to ground flanking the sets of signal pads In this way all of the probes for a single waveguide can be probed with the same probe card on which both of the side probes have been configured to be switchable grounds By translating the probe card across this set of pads all of the pads can be measured as long as the probe card has signal probes
116. re Figure A 16 Due to the nature of the program these should not be changed For future reference they are DeviceType iUSBDAQ Device Index 0 and Input Range 0 4 096 V If a computer is currently using more than one Hytek and more than one copy of this program at once then the second copy will need to change the Device Index to 1 In the right most box is the Loop Control in Figure A 16 While this is turned on Flipped to the up position the program will continue to take runs of data as long as it can This mode is useful when taking repeat runs of data or for overnight work It has been tested at full capacity over a period of 18 hours without error When the user wants to end this mode return the switch to the off position Flipped to the down position and wait for the program to finish taking the last run of data The switch needs to be switched off while the program is still running to bring the program to a halt The Current DIO Channel and Reference DIO Channel are a bank of green LEDs that correspond to the binary signal sent from the DIOs 1 6 on the Hytek These are sent to the multiplexer to control which pad is currently read on the probe station These were added for testing purposes of the multiplexer to correspond directly to which wires should be active They are no longer needed except for manual confirmation of a working multiplexer The graph mean standard deviation variance and channel work togeth
117. rence Channel On Amgenu y Jor Start new file Core mm CSV Filepath 4 _Spring_11 Testing csv amp Current DIO Channels Reference DIO Chan 1999999 699990 Bright green is high state dark green is low Read in binary left most button is DIO 1 Channel Std Deviation p as Mean Variance fo 0 10000 15000 20000 25000 30000 35000 Data Point H 8 Figure VII 4 Version 1 0 of LEAC Data Recording and Control VI easily understandable buttons for the job Figure VII 4 shows version 1 0 the final project version and how it can be read from left to right to define integral variables before the program is run D Capabilities of LabVIEW Program The long term stability feature was proven by the eighteen hour tests that were performed on the system which was running at 8000 data points per second a hardware limit imposed by the Hytek Automation iUSBDAQ At the conclusion of the test the computer was working as fast as when the program started even after 1 4 gigabytes of data had been taken Although this case seems extreme it shows that there should be no anticipation of a software crash when left alone The other limit that was imposed was a minimum of 128 data points per second sampling rate In keeping with a user friendly design the code prevents user inputs of less than one second recording times and less than 128 points of data per second After each pad is measured all of the raw data is saved to a separ
118. rk filel strcat folder dark1CSV dark file2 strcat folder dark2CSV light file strceat folder light1csv DH DH E Sanonymous function to convert voltages to currents vtoc voltages cal voltages cal offset cal gain Sorgainize data dark voltages 1 load dark filel dark currents 1 vtoc dark voltages 1 cal dark voltages 1 break into cycles dark voltages _1 nch_scan dark currents 1 break into cycles dark currents _1 nch_scan dark voltages 2 load dark file2 dark currents 2 vtoc dark voltages 2 cal dark voltages 2 break into cycles dark voltages 2 nch_scan dark currents 2 break _into_cycles dark_ currents 2 nch_scan light voltages load light file light currents vtoc light_voltages cal light voltages break into cycles light voltages nch_ scan break into cycles light_currents nch_ scan light currents photo voltages light _voltages 1 1 dark voltages 1 1 end light voltages 1l end dark voltages 2 1 1 photocurrents light _currents 1 1 dark currents 1 1 end light_currents 1l end dark currents 2 1 1 store data in output structure run_data struct dvl dark voltages 1 dv2 dark voltages 2 lv light voltages pv photo_voltages dcl dark currents 1 dc2 dark_currents 2 lc l ight_currents pc photocurrents run title run title folder folder end 130 Page L 21lbage
119. rom test to test a more robust system of collecting and analyzing data was implemented This system made multiple measurements of the dark voltages followed by multiple measurements of the light voltages followed by multiple measurements of the dark voltages again Drifting of both the dark and light voltages could then be assessed from the data A program which takes a folder containing these three sets of data and performs some analysis on it is shown in section Drift Plot Gain Plot Analysis Code sgain data plot 3 17 m Daniel Higley 03 2011 Code sequence used to generate the gain data plot on 3 17 2011 smanual gain data voltage average voltages 0 700816 0 870122 1 04364 1 21325 1 38725 1 55743 1 73047 1 90337 2 07231 7 Smanual current vectors taken uA currents 0 99297 1 49 1 9873 2 4844 2 9817 3 4789 3 9762 4 4733 4 9707 116 Page sgenerate linear least squares fit linear fit polyfit currents voltages 1 Sdetermine correlation coefficient of data R corrcoef currents voltages splot results plot currents voltages r currents polyval linear fit currents k least_squares title strcat Linear Least Squares Fit num2str linear fit 1 x num2str linear fit 2 legend Data least_ squares title title Keithley Current Vs LabVIEW Output Voltage xlabel Curent uA ylabel Voltage V FFT Code function v U transformer t u v U transformer
120. s to accomplish all three of these goals Based on information from J Hanson 1 it was known that sucrose has a non linear relationship with RI except when dealing in a very small percentage of sucrose in water when it is nearly linear and easier to model For a baseline pure DI water was chosen because of the relative ease of acquiring and known values in literature The RI of Sucrose Diluted in Water 1 3360 f x 0 00144x 1 33298 1 3355 1 3350 1 3345 1 3340 1 3335 Refractive Index 1 3330 1 3325 1 3320 1 3315 0 0 2 0 4 0 6 0 8 1 1 2 1 4 1 6 1 8 2 Percent Sucrose in DI Water w w Figure A 24 RI of Sucrose Diluted in Water 110 Page Presented in Figure A 25 literature values are used to show that from zero to two percent sucrose in one tenth increments RI obtains a nearly linear correlation 2 To prepare for the experiment a connection was made between the reservoir of solution and the LEAC chip then from the LEAC chip to the outlet Because of the nature of the setup the amount of vertical room above the chip was limited A tube was needed that could fit into a 1 16 OD opening and be able to bend away in less than two inches without breaking and attach to a Luer Lok syringe adapter Searching around through various companies it was decided that Teflon tubing from McMaster Carr Part 5239K23 would fit in with this experimental setup PEEK tubing was initially ruled out because of the warnings by the
121. sets of three columns with a BSA control in the middle and a set of rat a mouse columns as a control line The control line as depicted by Figure VIll 3 fluoresces brightly with the detection solution making the small assay easier to find under the fluorescent microscope Shown above in Figure VIII 3 the antibody a AG 85 cross linked and bound with antigen ESAT 6 while a ESAT 6 only bound as expected to its corresponding antigen ESAT 6 This shows that a AG 85 had less specific binding D Printing Methods Various methods were tested in order to print biomolecules onto the LEAC chip Previous printing methods using an electronic printer and commercially ordered chips were difficult to adapt to this project because LEAC chip fabrication methods cannot produce a consistent 90 degree corner on the chips as a reference for the printer input This is a problem using the electronic printer because it cannot line up the correct coordinates to be printed without a perfect 90 degree corner To solve this problem hand pin printing and PDMS printing strategies were tested 52 Page Hand printing was done using the actual needle from the electronic printer The needle was dipped into the antigen solution and pressed down onto a glass slide that had been previously marked with points that would correspond with the widening in the channel design when the PDMS is aligned with a designated corner These points were marked on the glass slide by placing the PDMS
122. spect of many research projects In the context of this project tuberculosis fragments are the biomolecules used An antigen is printed and then a solution of antibodies is printed over that If the procedure is done correctly antibodies will only bind to the regions where the solution crossed the already printed antigens that correspond to the biomolecule For example the antibody anti ESAT 6 a ESAT 6 will bind to the antigen ESAT 6 Some cross linking will also likely occur A third dimension of printing is also necessary if fluorescence is to be used for verification A third antibody labeled the Detection Solution in this manual is fluorescently labeled and will bind to the tuberculosis antibodies to make them fluoresce under the microscope For an example in this manual two different tuberculosis fragments will be used One set will be the antigen ESAT 6 and the corresponding antibody a ESAT 6 The second set will be Antigen 85 and the corresponding antibody a Antigen 85 The detection solution used is a combination of goat derived and mouse derived antibodies rabbit a goat and hamster a mouse These are used because the a Antigen 85 is mouse derived thus something else derived from mouse will bind to it The a ESAT 6 has the same relationship with the goat derived antibodies Process and Procedure Selecting a Surface This procedure is applicable to printing biomolecules on glass slides silicon nitride or a microchip
123. ss 24 Figure IV 6 Fluorescing capillary fluid flow through permanently bound PDMS on glass slide left and fluorescing active fluid flow through permanently bound PDMS on LEAC chip right 25 Figure IV A PDMS chifp allge Ee eege 26 5 Page Figure IV 8 Chip mask design left PDMS mask design with cutting guide center and actual PDMS aligned to chip and permanently bound from plasma oxidation right 27 Figure V 1 Original probe station where each probe has independent cantilevered XYZ motion With magnets holding the probes in place on the large steel base one probe was used to apply a potential to a common pad on the LEAC sample as the second was brought 28 Figure V 2 Photograph of the sample holder and fiber launch system with translational and rotational stages that provided the necessary movement of the sample for appropriate alignment30 Figure V 3 Photograph of probe station configuration at this phase in its evolution The piece labeled metal slat has vertical motion allowance cccesccesteceeneceesneeeeeeceeaeecseneeeeueceeaeecetaeessnees 31 Figure V 4 Block diagram of probe station configuration and portions with movement 000 32 Figure V 5 Probes are aligned and in contact and the fiber is aligned to the waveguide prepared for proper cOUpling OF Theslasen E 33 Figure V 6 Pump positioned above probe card with tubes running through the epoxy ring and into the sample The piece of t
124. t time into the lvm file to reference when the specific run was started Part D Changing to Current Pad This is a fixed sequence of events that is run every time the channel is changed It takes the number from the for loop and converts that into a binary number before outputting it to the Hytek s DIOs After it has done this it waits the Stabilize time as defined in the Front Panel before it continues on Part E Data Recording The most important piece of code in the software this section takes all the data points amount defined by the Front Panel before saving them to disk as the csv interpreting them with math models Mean Std Deviation Variance and exporting them to both the Front Panel and the lvm file Part F Switching to Reference Pad This is entirely optional code that is enabled by a true value on the variable Reference Channel After Every Measurement When true this acts much like Part D Changing to Current Pad and instead switches to the user selected Reference Pad When false this and Part G Reference Pad Data Recording are removed from code and ignored Part G Reference Pad Data Recording This is optional code that acts much like Part E Data Recording but doing it over the reference channel Does not do advanced math or output to Front Panel Only stores the data to the lvm file Part H Loop Control This is connected to the front panel a while loop that checks if the Loop Control switch is flipped up As
125. t u Daniel Higley 2 2011 AP dp dp d Performs a dft using matlab s fft function of the data vector u with time Sindicies t Ts t 2 t 1 STs time of sampling Fs 1 Ts SFs frequency of sampling U fftshift fft fftshift u length u sperform fft v linspace 0 5 Fs 0 5 Fs length U sdetermine frequency index for DFT splot results splot v U stitle transformed data sxlabel frequency sylabel value FIR Filter Code function y leac_firrcos filter x Fs stunction y leac_firreos filter x Fs g 6 117 Page Daniel Higley 02 2011 Symmetrically applies a finite impulse response raised cosine filter to x AJP oP dp ole and returns the result as y sfilter parameters Fo 10 scutoff frequency df 5 Stransition bandwidth n 1000 default filter order changes if x is not at least 3x this length Sif data vector is not long enough to use this high of a filter order decrease the filter order if length x lt 3 n n 0 1 length x end b firrcos n Fc df Fs Sdetermine filter coefficients y filtfilt b 1 x sapply filter stest plot of frequency respons figure 12 h w freqz b 1 8000 plot w Fs 2 pi abs h stest plot of original data figure 13 plot 1 length x Fs x stest plot of filtered data dp AP d AP d oP dp AP cP CP figure 14 plot 1l length y Fs y end Currents Versus Photodetectors fu
126. tep to cross for every detector This vertical step has been shown to increase scattering from the waveguide which results in less power reaching the farther regions This scattering and reduction of power is undesirable as the largest amount of laser power possible needs to be coupled along the waveguide to ensure a quantitatively measureable evanescent field The top left quadrant of Figure IIl 1 contains the mask designs for the metal layer The one used primarily for this project was the mask nearest the middle of which a magnified view can be seen in Figure IIl 3 below 15 Page DU EE fi Ie E 8 S OO Dm mmm mm E Figure HI 37 Zoomed in view of the most used metal mask design This particular mask was also used several reasons First the design s location near the middle of the mask makes it much easier to use with the small range mask aligner machines that were available in the CSU cleanroom Second the pad sizes of this mask were optimal for the current LEAC chip testing equipment The pads of this mask though still only 100um wide were a much heftier 290um tall when compared with the 100um of the other designs This allowed for a much higher degree of rotational alignment error when bringing the probes into contact with the fabricated chip s metal pads Finally this version of chip pads was chosen for its lack of structures in the middle of the chip This feature was very useful when dealing with the old a
127. the outside of the waveguide which satisfies Maxwell s boundary condition that the tangential component of an electric field must be continuous across the materials interface The component outside of the waveguide is called the evanescent field and this is what is used to sense the presence of something on the top surface of the waveguide When a biomolecule 12 Page is attached to the surface of the waveguide its index of refraction increases to a value greater than airs RI of 1 This causes the critical angle to change in the corresponding location along the waveguide Since physics requires that power be conserved the evanescent field beneath the waveguide must decrease This decrease is detected by small silicon photodetectors placed beneath the waveguide as depicted in Figure Il 1 The silicon s resistance will increase because there is less light from the evanescent field coupled into the device This will in turn result in a lower current detected for a specific detector which is manually biased at some optimal voltage Using a probe card the user can individually switch between detectors and locally measure the corresponding photocurrent changes B Modulation of the Laser An improved method of measuring the photocurrent from the LEAC chip involves modulating the laser diode coupled into the waveguide with a function generator This modulating frequency is then input into a lock in amplifier along with the total current obt
128. the sample did not need to be in the very center of the sample holder This would only be necessary if drastic rotations were necessary or translation was limited which were not impedances in this apparatus It was found that the sample could be roughly and reasonably adjusted with the naked eye and would only need a few degrees of rotation to be adequately aligned to the probes This made it possible to put the sample on more stable footing near the edge of the rotation stage The more stable footing allowed the probe card to apply more force to the sample which produced improved contact and essentially eliminated any planarity concerns Additional Functionality There was one problem common to all the early iterations of the design not discussed nor addressed until late in the design of the probe station The limited range of the various translation stages in the system specifically the fiber launch made actual alignment fairly difficult Since the fiber motion was so limited and the probe card was fixed in place the fiber had to be moved into position under the probe card by sliding a large metal plate across the table top of the probe station This non ideal plate and the hardware attached to it weighed more than 30 Ibs The sample holder would then be slid into position with respect to the other two This process requires significant familiarity with the system and is time consuming Future versions of the probe station should work to resolve
129. tical properties of materials in contact with it This means that the target substance need not be labeled in order to detect the presence of it This feature has the advantage of not only preserving the biomolecule for any further testing but it also reduces the cost as labeling technologies can be quite costly 11 Page Figure II 1 3 D Structure of the LEAC chip As labeled in Figure Il 1 a waveguide is added to the top layer of the chip and a laser can be coupled into one end of it The index of refraction of the waveguide which is made of silicon nitride SiN is 1 8 a value higher than its surrounding materials of air and silicon oxide SiO2 which have refractive indices of 1 and 1 48 respectively Thus the incident light that hits at the proper angle remains inside the structure and propagates down the length of the waveguide without exiting This angle is known as the critical angle and can be described by Snell s Law which is given in equation one below 6 sin 1 nj In the above equation n is the index of refraction of the material adjacent to the waveguide and n is the index of refraction of the waveguide itself Any light incident on the boundary that is equal to or greater than the critical angle will be totally internally reflected and will remain in the waveguide Although the light is totally internally reflected in the structure there is a component of the electric field that is present on
130. tically switching between the individual channels on the probe card and LEAC chip 10 Page 4 Improvements to Original Circuitry To layout and make a printed circuit board amplify the signal to a readable level and design and order a probe card to interface with the circuit to improve the overall quality and efficiency of data collection 5 Biomolecular Patterning Strategies To develop successful methods for patterning onto the waveguide s surface and to test for the specificity and sensitivity of two different TB antigen antibody complexes Achieving these goals required a collaborative and interdisciplinary team dynamic encompassing expertise from electrical chemical biological and mechanical engineering To establish good communication between different aspects of the project multiple weekly team meetings were set and progress reports submitted Team leadership was altered throughout the year and a detailed Gantt chart was made together to keep the project on a schedule while also recognizing critical paths that could play a role in setting back other parts of the project The realization of the above goals including the setbacks and problems encountered with each are detailed in the following sections of this report More detail on project management can be found in Appendix A Il Background A Local Evanescent Array Coupled LEAC Chip Theory The local evanescent array coupled LEAC chip operates by sensing changes in the op
131. to a combination of insufficient point of care diagnostics and TB s predominance in impoverished regions of the world where resources for testing and treatment are limited Current methods of diagnosing TB include the skin test blood testing chest x rays and sputum cultures all of which have their limitations The skin test is the most common method for initially diagnosing a TB infection but it cannot effectively distinguish between different stages of the disease including recent exposure latent versus active infection or drug resistant TB forms For this test a small amount of purified protein derivative PPD of the bacterium is injected intradermally with a positive result indicated by a swelling at the site of PPD injection after 48 72 hours Unfortunately there is a large incidence of false positive results in patients who have had the BCG vaccine or who are infected with a latent form of TB and false negative results in immunosuppressed patients or individuals so recently infected that their immune system has not yet reacted to the bacteria The other secondary forms of TB testing are limited by either expense the need for a skilled technician or doctor time needed to obtain results or a combination of those thereof Due to the pitfalls of the current testing methods available it is clear that there is a formidable need for an innovative diagnostic strategy that could provide rapid affordable and robust testing with the abili
132. to the right to turn on the plasma a Note This knob will not stay to the right it just clicks over but the pressure will visibly drop at this point b It is also a good idea to check the power Turning the PRES dial toward FWD will show the Watts If the chamber is not running at 20 Watts adjust this using the power knob c After the plasma is on run the chamber for 18 seconds then push the STOP button unplug the machine open the green valve and wait for the pressure to equalize in the chamber so it will open again 105 Page d Once the chamber opens push the AC ON button to completely turn it off and take out the PDMS on the chamber slide 4 Using tweezers remove the PDMS from the slide and press it onto the printing surface a Push down with a finger or tweezers on top of the PDMS stamp holding it firmly onto the chip or slide for at least 30 seconds b Clean the PDMS surface with tape 5 Add antigen solution a Using a micropipette put approximately 2 5 uL volume may vary depending on channel design of the desired antigen solution into the inlet reservoir the larger punched hole of the channel Keep in mind the pattern and order of each antigen solution i Note Change out pipette tips between each solution to avoid contamination Figure A 20 Use of micropipettes 6 Incubate the chip with antigen solution in channels under a foil covered dish for 10 minutes 7 After 10 minut
133. ty to differentiate between different stages of the disease The lab on a chip diagnostic biosensor described in this paper employs a novel strategy for detecting diseases using label free technology in real time and has high potential for application in the medical industry The underlying physical phenomenon for this device is the detection of a local field shift using photodetectors and a waveguide with total internal reflection of coupled light When a nanolayer of biomolecules is added to the surface of the waveguide a change in effective refractive index causes a shift in the evanescent field distribution which is then detected as a change in photocurrent at the underlying detectors 9 Page With increasing research finding differences between molecular signatures of active latent and uninfected TB samples this device has even greater potential in potentially making it feasible to not only identify TB infected individuals but to distinguish between those with the latent versus active form Previous work on the original LEAC chip design found that a change in a thickness on the surface of waveguide as small as 30 pm can be detected which corresponds to a detection limit of 120 pm In refining and optimizing the measurement methods and electronics of the LEAC chip integrating microfluidics for real time detection and developing LabVIEW for a more user friendly interface the scope of application for this technology could
134. ubing not connected to the syringe pump is held in place by a secured MAB MELIC Ell WEE 35 Figure V 7 Mechanical fixtures ON microscope objective cccceessceesssssecesessseeeessssseeeessnseeesssseeess 36 Figure VI 1 Original Chip probing system 38 Figure VI 2 Mask design of the LEAC chip with color coding where red indicates signal pads green indicates ground pads and purple indicates the approximate locations of edge sensor touchdown39 Figure VI 3 New probe card ordered to specfications 40 Figure VI 4 Top view of the probe card showing probes and epoxy ring configuration left and a microscopic view of the probes in contact with LEAC chip pads and an edge sensor shown on the right hand side Jrtebtt utegereteg tege geet 41 Figure Vl 5 Bent probe Shien EAE EAEE E EEE E EE EAEE SE E E EE S 44 Figure VILL Block diagram of the multiplexing amplification and software automated control E EE 45 Figure VII 2 Multiplexing and Amplification Circuit before left and after redesign and PCB order FIBWE EE 46 Figure VII 3 Interface cable to connect the probe card to the switching drcunt 47 Figure VII 4 Version 1 0 of LEAC Data Recording and Control VE 48 Figure VIII 1 Antigen antibody reaction with secondary antibodies cccccssscccssssceesssseeeesssseeess 50 Figure VIII 2 Molecular structure of chip surface freatment 51 Figure VIII 3 ESAT 6 and AG 85 antigen antibody assay with rat a mouse control line full
135. ught into contact at a time followed by precise movement of the fiber into position to couple the laser into the waveguide The advantages of this type of sample centered probe station configuration are it s simplicity in construction and flexibility Unfortunately the ambitious goals of this senior design team could not be achieved with such a system One of the primary goals of this project was to streamline 28 Page the probing and data collection process To achieve this goal a probe card was used to simultaneously connect to several active areas on the LEAC sample simultaneously The probe card s role in the device as well as its specific design considerations will be discussed in the following section In brief a probe card is a printed circuit board PCB that has an array of probes attached to its underside These probes are designed and configured to correspond to the connection areas on the device for testing metal pads of the LEAC chip The design of a probe station that could utilize such a probe card is more involved than the one originally used Some of the most critical probe card considerations that were accounted for by the probing station include the linearity of the connection pads with respect to the probe card and the planarity of the sample surface with respect to the plane of the probes The probe station needed to allow for precise vertical motion to bring the aligned probe card and sample into contact with each oth
136. unacceptable effects for the very precisely aligned probe card We also considered using probes with longer probe tip lengths but it was advised that this would cause the probes to be much more fragile than usual On the topic of probe card fragility it is of note to mention that some of the signal probes in the center of the probe card were inadvertently bent during testing Figure VI 5 depicts the damage done to these bent probes during testing This limits the number of testable pads for a given system configuration as well as the total number of photodetectors that can be interfaced with using the probe card Luckily these probes were not bent in such a way as to interfere with adjacent probes leaving the remaining unbent probes viable Upon contacting the vendor that supplied our probe card we were informed that the bent probes could be realigned if the bending was not overly severe otherwise bent probes could be replaced Our vendor said that the cost of this was determined on a case by case basis It may be necessary in the future to obtain signals 43 Page from the pads to which these broken probes correspond to in which case investigation into the costs or technique for fixing the bent probes will be necessary 1 Figure VI 5 Bent probes Because our application for this probe card requires very low noise it was beneficial to place the amplifier circuit for signals exiting photodetectors on the probe card itself While
137. was reached in outputting the means and standard deviations as well as a plot of all points A full user guide for LabVIEW can be found in Appendix G VII Biomolecular Testing A Antigen Antibody Reactions The reactions that occur between antigens and antibodies in immune responses are what allow LEAC chip detection to be feasible When an antigen is present in a body the immune system will respond by binding corresponding antibodies to the foreign antigen This lock and key type binding is intended to be very specific between paired antigens and antibodies The specificity between an antibody and the antigen it is intended to bind to allow for the LEAC chip s targeted application in specific disease detection assuming an antibody intended for another purpose would not bind to the printed antigen and vice versa In Figure VIII 1 below the purple and green antibodies and antigens show the lock and key relationship of two different pairs which could be correlated to two different types of diseases or two distinct tuberculosis antigens 49 Page Fluorescing 2 Antibodies Corresponding 1 Antibodies f f Different Antigens WAVEGUIDE Figure VIII 1 Antigen antibody reaction with secondary antibodies Figure 8 1 also shows secondary antibody binding The tuberculosis antibodies used were not fluorescently labeled and for the purposes of checking the reliability of the designed printing methods fluorescence microscopy was a

Download Pdf Manuals

Related Search

Related Contents

Guía del usuario de Parallels Plesk SiteBuilder 5.0  Lenovo IdeaCentre H500s  - Jonsered  Kingston Technology ValueRAM Memory 128MB 100MHz ECC CL3 DIMM  GDR 10,8-LI Professional - Axminster Power Tool Centre  User Guide  Vornado 673 Use and Care Manual  SA 2062 (61.3 cm3)    Einhell RT-SM 430 U  

Copyright © All rights reserved. Failed to retrieve file