107
REPORT Laboratoire d’électronique et de technologie de l’information Commissariat à l’énergie atomique et aux énergies alternatives Direction de la recherche technologique MINATEC Campus | 17 rue des Martyrs | 38054 Grenoble Cedex 9 T. | F. [email protected] Page : 1 / 107 Établissement public à caractère industriel et commercial RCS Paris B 775 685 019 The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to thir d without their authorization Report on functional tests of prototypes (D8.2) Final report Date : 29/09/2014 Revision : Version 0 N / Réf. : DRT/LETI/DTBS/STD/LISA 14-175 V / Réf : Participants Demonstrator 1: IPT Demonstrator 2 : see diffusion list for CEA Demonstrator 3 : GIN/GINOLIS Name Function Date Signature Authors Demo 1 : A. Sauer-Budge Demo 2: J. Hue, G. Nonglaton Project leader, WP leader 06/10/14 06/10/14 Approval Demo 2 : J.M. Dinten Laboratory supervisor 06/10/14 Diffusion list IPT : C. Baum, T. Bastuck, A. Sauer-Budge MRS/MIN : X. Li DOL : P. Steinman PLS : S. Hamm IQS : S. Borros CET : P. Lacharmoise VTT : H. Hakalathi, J.Okkonen DUR : M. Graf COA : V. Villari PAN : R. Joachimi, J. Perl, J. Loerzer GIN : M. Känsäkoski CEA : G. Nonglaton, T. Bordy, M. Darboux, M. Domenes, C. Fontelaye, D. Lauro, F. Perraut, R. Rousier For this deliverable (8.2), the timetable is the following : - 15 th May 2014 : 1 st intermediate report - 1 st July 2014 : 2 nd intermediate report dead line extended = 21 st July 2014 (delivery date : 12/08/14) - 27 th August 2014 : final report (dead line : M24) dead line extended = 7 th October 2014 [ML2 meeting : 28 th -29 th October 2014] Confidential

REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Embed Size (px)

Citation preview

Page 1: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

REPORT

Laboratoire d’électronique et de technologie de l’information

Commissariat à l’énergie atomique et aux énergies alternatives Direction de la recherche technologique

MINATEC Campus | 17 rue des Martyrs | 38054 Grenoble Cedex 9

T. | F.

[email protected] Page : 1 / 107

Établissement public à caractère industriel et commercial RCS Paris B 775 685 019

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization

Report on functional tests of prototypes (D8.2) Final report

Date : 29/09/2014 Revision : Version 0

N / Réf. : DRT/LETI/DTBS/STD/LISA 14-175

V / Réf :

Participants Demonstrator 1: IPT Demonstrator 2 : see diffusion list for CEA Demonstrator 3 : GIN/GINOLIS

Name Function Date Signature

Authors Demo 1 : A. Sauer-Budge Demo 2: J. Hue, G. Nonglaton

Project leader, WP leader

06/10/14 06/10/14

Approval Demo 2 : J.M. Dinten

Laboratory supervisor 06/10/14

Diffusion list

IPT : C. Baum, T. Bastuck, A. Sauer-Budge MRS/MIN : X. Li DOL : P. Steinman PLS : S. Hamm IQS : S. Borros CET : P. Lacharmoise VTT : H. Hakalathi, J.Okkonen DUR : M. Graf COA : V. Villari PAN : R. Joachimi, J. Perl, J. Loerzer GIN : M. Känsäkoski CEA : G. Nonglaton, T. Bordy, M. Darboux, M. Domenes, C. Fontelaye, D. Lauro, F. Perraut, R. Rousier

For this deliverable (8.2), the timetable is the following : - 15

th May 2014 : 1

st intermediate report

- 1st

July 2014 : 2nd

intermediate report dead line extended = 21st

July 2014 (delivery date : 12/08/14)

- 27th

August 2014 : final report (dead line : M24) dead line extended = 7th

October 2014 [ML2 meeting : 28

th-29

th October 2014]

Confidential

Page 2: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 2 / 107

Modifications Date Version

Demonstrator 1 : page 2 Demonstrator 2 : page 26 Demonstrator 3 : the report will be directly sent to the prime

Page 3: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 3 / 107

Demonstrator 1 (A. Sauer-Budge)

INTRODUCTION p 4

MATERIALS AND METHODS p 5 Microfluidics-based polymerase chain reaction (on-chip PCR) Testing of second generation CETEMMSA printed heaters Testing of second generation CETEMMSA printed heaters Testing of Dolomite piezoelectric pump

RESULTS p 11 Validation of microfluidics-based PCR with Zeonex 690R chip Testing of second generation CETEMMSA printed heaters Testing of CETEMMSA T sensors Testing of Dolomite piezoelectric pump

DISCUSSION p 20 Conclusions for validation of microfluidics-based PCR with Zeonex 690R chip Conclusions for testing of second generation CETEMMSA printed heaters Conclusions for testing of second generation CETEMMSA printed heaters Conclusions for testing of Dolomite piezoelectric pump

Page 4: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 4 / 107

INTRODUCTION The primary goal of the 8.1 and 8.2 deliverables of the MultiLayer-MultiLab (ML2) project is to develop a microfluidics-based pathogen-detection system that is low cost and utilizes high-throughput manufacturing processes. ML2 draws on the latest advances in polymerase chain reaction (PCR) technologies and incorporates a microfluidic reaction chamber with printed heaters in a single laminated chip. The essential design for the assembled chip is shown in Figure 1. The key feature of our design is that the microfluidics chamber is in direct contact with a pair of CETEMMSA printed heaters set at 62°C (orange) and 95°C (red). A thermal sensor embedded in the layer directly below the heaters helps regulate and maintain the precise temperature zones. An off-chip syringe pump cycles the reaction material between the two temperature zones, mimicking the temperature changes in a classic thermal cycler-based PCR. As an alternative to the syringe pump, we also tested a Dolomite piezoelectric pump which is small enough to be incorporated into future lab-on-a-chip (LOC) applications. PCR amplification products are detected off-chip either directly by an in-house optical system (Fraunhofer CMI) or by downstream gel electrophoresis. Our chips were manufactured using Zeonex 690R (Zeon Chemicals) that has a glass transition temperature (Tg = 136°C) significantly higher than those reached during the PCR reaction. Higher Tg values reduce the possibility of channel deformation and melting in the microfluidics chamber during PCR. An alternative material, poly(methyl methacrylate) (PMMA), is ideal for low-cost industrial-scale role-to-role manufacturing procedures. Unfortunately, the Tg value for PMMA is 110°C—very close to the 95°C mark that is used for activating the PCR enzyme and separating the DNA strands. We were able to successfully mold the microfluidic structures into PMMA as well as bond the sealing coverslip to the surface without significantly deforming the channels. However, maintenance of microfluidic structures during PCR where the PMMA will reach 95°C will require strict regulation of the temperature to prevent deformation. In this report for 8.2, we report our delivery of the following products:

1. Identifying pathogen samples by on-chip PCR with Zeonex 690R demonstrator chip

2. Testing of second generation CETEMMSA printed heaters and sensors

3. Testing of Dolomite piezoelectric pump

We additionally make recommendations based on our findings for incorporation into future microfluidic-based workflows. We presented the development of the prototyping processes for this demonstrator in the report for 8.1.

Figure 1: Schematic of layers in ML2 chip. The primary component is the microfluidics chamber in the center. In the design, the DNA is amplified by PCR as the reaction components are pumped back and forth between two temperature zones generated by the printed heaters. Amplification is detected off-chip by either optic or gel electrophoretic methods.

Page 5: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 5 / 107

MATERIALS AND METHODS Microfluidics-based polymerase chain reaction (on-chip PCR) Microfluidic chip fabrication and bonding Microfluidic chips were milled in-house at Fraunhofer CMI with our ultra-precision machining center using Zeonex 690R (Zeon Chemicals) according to 5th generation design specifications. Milled chips were washed by placing chips in a bag containing soapy water and were sonicated for 30 mins. After a water rinse, chips were oven dried at 80°C for 1h. Prior to bonding the coverslip, chips were soaked twice in acetone for one minute and blown dry with air. To activate the chip surface for bonding, chips were soaked in a decalin:ethanol (35:65 v/v%) mix followed by an ethanol rinse (SigmaAldrich). A Zeonex 690R coverslip cut to match the dimensions of the chip was bonded by hot pressing the coverslip to the microfluidics chip at 134°C and 7.8 psi using our custom hot press device (Fraunhofer CMI). Microfluidic channels in the bonded chips were blocked with a 1% BSA solution for 2h and washed with DNA-free water. This blocking step was done either one day before or the same day as the PCR. Microfluidic devices were then ready for PCR. PCR reaction parameters Block heaters set to 62°C and 95°C were switched on 30 mins prior to the PCR. PCR reagents were prepared according to the manufacturer’s protocols (Primerdesign) and are described briefly in Tables 1-3.

Table 1: PCR mix composition without DNA Component Amount [µl] VeriQuestTM Mastermix 240 E. coli 0157 Primer mix 24 BSA 1% 24 PEG 8000 48 H2O 24

Table 2: On-chip PCR mix composition with DNA for one control

Table 3: Off-chip PCR mix composition with DNA

Component Amount [µl] PCR mix as described in Table 3 300 E. coli 0157 DNA 10^4 copies/µl

(or water for negative control) 100

Component amount [µl] PCR mix 15 E. coli 0157 DNA 10^4 copies/µl (or water for negative control) 5

Page 6: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 6 / 107

Experiment groups were mixed with template DNA while negative controls were mixed with water. The reaction mix was degassed prior to the PCR run to avoid air bubbles in the microfluidics and were loaded into a syringe attached to the entry port. The syringe pump was set to 0.5 µL/min for a 20 min run. After the run, the reaction mix was collected in a microcentrifuge tube and stored at 4°C until analysis by gel electrophoresis. The off-chip control PCR was set-up as shown in Table 4.

Table 4: Off-chip PCR program in thermal cycler (7500 Real Time PCR System, Applied Biosystems)

Cycles Temp [°C] Time 1 50 2 min 1 95 5 min 40 95 3 s

60 30 s

Testing of second generation CETEMMSA printed heaters Chip analog The single thermocouple in this system was embedded between a 22 x 23 x 2 mm piece of Zeonex 690R (Zeon Chemicals)

and a 200 μm thick Zenoex 690R coverslip that had been thermally bonded together. To bond the Zeonex layers to one another, each layer was treated in a 35:65 % vol mixture of decahydronaphthalene (decalin) and ethanol for 1 min, followed by a wash in 100% ethanol for 2 mins, an ethanol rinse, and drying with pressurized air. After surface activation of the Zeonex, the thermocouple was placed between the two and pressurized at 0.65 N/mm2 at 290°F (143.3°C) for 5 mins. Experimental setup As depicted in Figure 2, every heater was placed within a testing platform developed specifically for these characterizations and inspired by a similar setup utilized by CETEMMSA. Delrin (DuPont) was selected as a base material, and was also incorporated as a cover for the system, respectively 54 x 37 x 13 mm and 76 x 55 x 13 mm in size. The purpose of this encapsulation was to provide a degree of insulation against heat loss and to improve the heating within the system. After ensuring that the heater was flat against the base, the copper leads were securely pressed against the ends of the heater to establish coherent electrical contacts. These leads connected to an MPJA 9616PS power supply (Marlin P. Jones & Assoc, Inc.) which controlled the voltage and current leading into the heater. The chip analog was then clamped to the middle of the heater and underneath the Delrin cover. Attention was paid to reproducibility and uniformity of testing, such that the heater was positioned identically relative to both the chip analog and the copper leads for every experiment. The thermocouple within the chip analog was also oriented within the center of the heater.

Page 7: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 7 / 107

Figure 2: Schematic of the heater testing platform from layer (A) and top-down (B) perspectives. The Delrin cover is omitted in the top-down view

Heater tests Four heaters were tested from different material categories. Their properties are shown in Table 5.

Table 5: Heaters used in characterization testing—specifications provided by CETEMMSA

Current and voltage were increased to provide incremental changes in power delivered to the heater, and temperature readings were recorded once they had been allowed to stabilize for each power setting. Particular attention was paid to discovering the power needed to heat the chip to temperatures of 62°C and 95°C. Once a temperature close to 95°C had been reached, the power input was kept constant for at least 20 mins in a hold test and observations were made about the stability of the temperature value during this time. Using the information gleaned from these experiments, a chip starting at room temperature was heated with a power setting that anticipated a final temperature of 95°C, and the time-course of temperature values was documented until they achieved equilibrium. At this high temperature state, the power input was also removed from the system and another time-course was recorded for the drop in values back to room temperature.

Sample ID Printing substrate Reported Resistance (Ω) A1 Kapton 3.1 B1 Polyethylene terephthalate (PET) 5.9

C1 Polyethylene naphthalate (PEN) 6.4

D1 Cyclo Olefin Polymer (COP) 2.6

Page 8: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 8 / 107

Error values in temperature measurements were calculated using specifications reported by Omega and represent the sum of errors from the thermometer and the thermocouple components separately.

Testing of second generation CETEMMSA printed heaters Experimental setup Each sensor in this study was placed within a testing platform inspired by the setup for the heater characterizations (see Figure 3). As was the case previously, Delrin (DuPont) was utilized as the material for both the base and cover, each 77 x 37 x 18 mm in size. It was necessary to clamp the sensor between the two pieces of Delrin in order to achieve sufficient electrical conductivity, and also to ensure a uniform application of heat. Two copper leads were firmly attached to the cover such that, when pressed against a sensor resting on the base, they lined up with the electrical contacts of the sensor with a set distance between them. These leads connected to a 34410A digital multimeter (Agilent Technologies), which was used to record resistance measurements across the sensor. Also embedded between the Delrin was a K-type thermocouple (5SRTC-TT-K-36-36, Omega) connected to an Omega HH12 thermometer. This thermometer was utilized as a reference for the local temperature applied at the sensor surface. As with the heaters, care was taken to place each sensor within the same relative location for every experiment. The metal clamp used to compress the Delrin firmly around the sensor was itself resting on a hot plate, such that adjustments to the hot plate would conduct through the clamp and the Delrin to alter the temperature of the sensor. An alternative testing platform was developed to provide a closer approximation to the finalized chip layer design and to serve as a basis for comparison. For this arrangement, 2 mm thick Zeonex 690R (Zeon Chemical) replaced the Delrin on the top and the bottom. The Zeonex material measured 76 x 40 mm.

Figure 3: Schematic of the sensor testing platform from a top-down perspective. All layers are covered by a layer of Delrin and compressed using a clamp that sits upon a hot plate. In the alternative design, Zeonex 690R is used in place of Delrin.

Page 9: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 9 / 107

Sensor testing As with the printed heaters, sensors composed of varying materials were tested in comparison with one another. Their reported properties are shown in Table 6. In a given test, heat was applied to the system by incrementally adjusting the heat output of the hotplate upon which the apparatus stood. As the temperature increased, recordings were taken pairing resistance and temperature of the sensor. After reaching a temperature around 100 °C, the heat applied to the system was gradually reduced and similar measurements were taken. In some cases, due to time constraints, only a single direction (heating or cooling) was tested during a single experiment. However, except in certain circumstances where specifically noted, sensors were not removed from the testing apparatus in between runs, nor was the experimental setup adjusted in any way other than turning devices on or off. Error values in resistance measurements were calculated using specifications reported by Agilent. Temperature errors represent the sum of errors from the thermometer and the thermocouple components separately as documented by Omega.

Table 6: Sensors utilized for experimentation—specifications provided by CETEMMSA

Sample ID Printing substrate Reported resistance (Ω) a4 Kapton 636.4 a5 Kapton 698.7 b5 Polyethylene terephthalate (PET) 520 c5 Polyethylene naphthalate (PEN) 1080 d4 Cyclic Olefin Polymer (COP) 368.6

Testing of Dolomite piezoelectric pump The experimental pump is a peristaltic pump provided by Dolomite, whose piezoelectric diaphragm responds to changes in voltage and frequency supplied to the pump by alternately projecting and retracting the diaphragm at a certain frequency, pushing fluid in the process. In general, voltage delivered to the system allows for a higher frequency of contractions, which in turn produces a higher flow rate. In contrast, the current experimental setup for the ML2 chip utilizes a syringe pump for driving fluid through the system. A syringe pump applies a linear force with a step motor to achieve a volume displacement that will produce a fixed flow rate as long as the linear force supplied by machine can overcome the resistance in the system. The peristaltic pump alternative also displaces a certain volume of fluid on every stroke of the diaphragm, but a continuous force is not supplied in the same manner.

Page 10: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 10 / 107

To determine the theoretical application of the pump to the chip, one can begin by estimating its hydrodynamic resistance, the resistance that it presents to fluid flow at a given applied pressure. According to a simple model, the resistance provided by a channel with a rectangular cross-sectional geometry may be described as

(1)

where is the resistance measured in

, is the dynamic viscosity ( ) , is the length of the channel ( ),

is the width of the channel ( ), and is the depth of the channel ( ). Equation 1 only holds for the case where

, which is valid for this design. Units of resistance may also be reported in

after applying the necessary

conversions.

Page 11: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 11 / 107

RESULTS Validation of microfluidics-based PCR with Zeonex 690R chip The 5th generation final design of the microfluidics chip is shown in Figure 4. This chip contains two entry ports for sample and PCR reaction mix followed by an in-line mixer. The channel then traverses the 95°C region to allow for DNA polymerase activation. The reaction mix travels from 95°C to 62°C and back in one PCR cycle which takes approximately 20 seconds to complete. The reaction mix is subjected to 40 cycles before collection at the exit port. While the design of the chip ultimately allows for separate port entry of DNA sample and PCR reaction mix, for the pilot experiments described here, reactions were pre-mixed and loaded together. In this experiment, 400 µL of reaction mix and DNA from enteric pathogen E. coli O157:H7 was loaded into the chip and the flow rate was set to 1.5 µL/min. The chip was attached to the syringe pump through the Dolomite connector and placed on the heater block platforms as previously described. At the time of the results described here, we had not yet received the second generation printed heaters or the piezoelectric pump.

After the PCR run, 10 µL of on-chip PCR reaction mix was collected at the exit port for visualization by gel electrophoresis as shown in Figure 5. Simultaneously, all reaction components were run off-chip in a traditional thermal cycler-based PCR to provide a positive control. Lane 1 is the DNA ladder in base pair units. The size of the E. coli O157:H7 target amplicon is 75-base pairs. In Lane 3 is the positive control of off-chip traditional PCR showing robust amplification of the target sequence. Shown in lane 4 is on-chip amplification of the 75-base pair target amplicon demonstrating detection of pathogen-specific DNA by microfluidics-based PCR. While on-chip microfluidics-based amplification does not reach the levels obtained through traditional methods, amplicon levels are detectable and above background (as determined by comparison with negative control in lane 5).

Testing of second generation CETEMMSA printed heaters A preliminary examination of the heaters upon arrival revealed that the D1 heater had begun to delaminate in transit, though testing with this heater proceeded as described. In assessing the heater design, it was discovered that the length of the heating element on each heater extends further than the length of the microfluidic chip, such that some of the heater will hang over the sides of the chip.

Figure 4: 5th generation of microfluidics chip improves upon previous designs by the addition of a separate port for DNA

entry and a zigzag mixer.

Figure 5. Microfluidics-based amplification of pathogen-

specific DNA. Lane 4 depicts amplification of the 75-base pair

E. coli target. The positive control is shown in lane 3 and

the negative control is shown in lane 5.

Page 12: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 12 / 107

Figure 6: Data recorded from the stepwise power and hold testing conducted for heater D1. Temperature holding began at minute 120 for this heater.

As seen above in Figure 6, stepwise and hold tests were recorded for D1. The voltage and current spontaneously changed at various steps throughout the testing, even though adjustments had not been made to the power supply output. This observation persisted for all other experiments to varying degrees as well, and was a particular feature of all heaters during the low to high temperature transition experiment. During the course of their hold tests, C1 and D1 heaters both continued to rise in temperature, while A1 and B1 were found to stay constant. In all cases, however, the heaters never dropped in temperature while holding at a constant power input. After testing the heaters to obtain their operating parameters, it was then possible to plot the power dependence in terms of chip temperature, as seen in Figure 7.

0

10

20

30

40

50

60

70

80

90

100

110

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

Tem

per

atu

re (

C°)

Time (min)

Page 13: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 13 / 107

Figure 7: Comparison of heaters in terms of heating efficiency. The plot depicts results for heaters A1 (green), B1 (yellow), C1 (orange), and D1 (blue) as well as linear trend lines calculated from the respective data for each. The trend lines calculated from the graphs provided an estimate for the power necessary for each heater to heat the chip to a target temperature of 95°C (Table 5). When these settings were initially applied to each heater starting from a low temperature, the power drawn from the power supply changed in the course of the experiment, generally leading to a final temperature greater than desired (Fig 8). The consequences of this behavior are most immediately apparent in the plot for D1 around minute 44, where the voltage unexpectedly increased at this time, leading to a ramp in temperature. Attempts were made to compensate for this phenomenon by setting initial voltage and current settings lower than the predicted requirements. Despite such efforts, the target temperature of 95°C was not achieved in any of the heaters during experimentation.

0

10

20

30

40

50

60

70

80

90

100

110

0,00 0,50 1,00 1,50 2,00 2,50 3,00 3,50 4,00 4,50 5,00 5,50

Tem

per

atu

re (

°C)

Power (W)

Page 14: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 14 / 107

Figure 8: Low to high temperature transition curve for each heater, measured upon application of a fixed power to the heater. A1 (green); B1 (yellow); C1 (orange); D1 (blue). In all cases, the input power setting changed during the course of the experiment.

In the final experiment, the heat dissipation with each heater starting from a high temperature was found to be almost identical in all cases regardless of heater (Figure 9).

Figure 9: High to low temperature transition curve for each heater, recorded after eliminating the power input to each heater. Color coding is identical to that displayed in Figure 7.

0

10

20

30

40

50

60

70

80

90

100

110

0 10 20 30 40 50 60 70 80

Tem

per

atu

re (

°C)

Time (min)

0

10

20

30

40

50

60

70

80

90

100

110

0 10 20 30 40 50 60 70 80 90 100 110

Tem

per

atu

re (

°C)

Time (min)

Page 15: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 15 / 107

Testing of CETEMMSA T sensors In order to control the printed heater systems during PCR, CETEMMSA developed a series of printed thermal sensors fabricated with similar materials and methods as the heaters. These sensors consist of a continuous path of silver ink printed on the plastic substrate shaped into a repeating rectangular saw tooth pattern, and rely on this series of turns to generate an electrical resistance in the 100-1000 ohm range. The established resistance of the sensor varies in response to an applied temperature, a relationship that, when properly calibrated, can be used to inform feedback control of the heating system to achieve a static temperature. Because the output of the integrated heaters appeared to vary during experimentation, likely due to unstable material properties, a similar concern existed for the temperature sensors. While the input to the heaters may be automatically adjusted if such a phenomenon occurs -- assuming that they are driven with the proper control program -- the sensors must remain completely reliable over the course of use in order to perform their respective function. Testing of the sensors for their reliability was thus a critical objective. The temperature dependence of each sensor was assessed by incrementally increasing and decreasing the temperature and recording the resultant resistance, and these experiments were conducted multiple times to determine the degree of reproducibility. An evaluation of the testing apparatus was also conducted to assess whether any variability could be attributed to changes in sensor configuration. As was the case with the COP heaters, initial observation of the sensors prior to testing indicated that material degradation had begun to occur in all of the COP sensors before or during shipment. Regardless, because of the prior use of COP for the microfluidic layer, an initial test was conducted on sensor D4 to determine the temperature dependence of the sensor design with this material. Preliminary results are displayed in Figure 10. Unfortunately, at temperature points of 66 °C and 74 °C, the multimeter began to indicate that the measurement was overloaded, without any adjustments to the setup. Attempts to reorient the sensor to re-establish valid measurements did temporarily resolve this issue, but also eventually resulted in irreversible damage to the sensor, at which point testing could not continue.

Page 16: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 16 / 107

Figure 10: Temperature dependence of resistance in sensor D4. The multimeter displayed an overloaded resistance reading at 66°C and 74°C, at which points the setup was adjusted to re-establish coherent resistance readings. Testing could not continue after second attempt due to damage to the sensor.

Initial experimentation was also conducted for sensors B5 and C5. Although a positive linear relationship between resistance and temperature was discovered for both sensors, testing demonstrated sequentially lower resistances across the same temperature range during the cooling that followed a heating stage (Figure 11). This phenomenon was compounded with additional heating and cooling cycles (Figure 11B).

Figure 11: Resistance measurements for sensors B5 (A) and C5 (B). A) An observable drop in resistance values occurred during experimentation with sensor B5, particularly during the cooling phase as compared to heating. B) Sensor C5 displayed a similar behavior and was subjected to two rounds of heating and cooling in separate experiments. The data in the orange plot represents the second experiment in the set.

380

385

390

395

400

405

410

10 30 50 70 90 110

Res

ista

nce

)

Temperature( °C)

Page 17: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 17 / 107

Due to the absence of additional specimens composed of COP, as well as a project focus on COP and Kapton materials, subsequent testing of the sensors focused on sensors A4 and A5. During the first experiment, sensor A5 continually dropped from its initial resistance value throughout both heating and cooling stages (Figure 12). Subsequently, its temperature dependence displayed a pattern similar to that observed with B5 and C5, with each experiment resulting in a lower resistance trajectory and set point after cooling. Later in testing, more overlap occurred in the data from different experiments (Figure

12B). Sensor A4 did not follow the same trend as A5 with regard to declining resistance, though also shared the phenomenon observed for other sensors of a cyclically-lowering set of resistance values (data not shown). Despite having reported

resistances that differ by approximately 60 Ω, measurements from sensors A4 and A5 typically varied on the order of 100 Ω.

Page 18: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 18 / 107

Figure 12: Data for sensor A5 depicting all (A) and highlighted (B) regions of plot. A) Resistance values declined during the first experiment (blue), after which the resistance displayed a positive linear correlation with temperature but continued to trend

downwards for all values during each cooling cycle. Error bars are omitted for other data sets for clarity. Sequential ordering of experiments was as follows: blue, orange, green, yellow, red, purple. Yellow, red, and purple data points were obtained on the

alternative apparatus wherein Delrin was replaced with Zeonex. B) Focused area of a5 data depicting associated errors.

450

500

550

600

650

700

750

10 30 50 70 90 110

Res

ista

nce

)

Temperature (°C)

A

480

490

500

510

520

530

540

10 30 50 70 90 110

Res

ista

nce

)

Temperature (°C)

B

Page 19: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 19 / 107

An evaluation of the testing apparatus involved swapping the Delrin and Zeonex covers in an otherwise unchanged testing platform. Results of this line of experimentation may be seen in Figure 13. After repeated testing of sensors A4 and A5, both eventually stopped providing coherent resistance readings.

Figure 13: Distribution of room temperature resistance values for sensor a5 (A) and a4 (B). Sensors were switched between versions of the apparatus composed of Delrin (circles) and Zeonex (squares) during two separate experiments (blue, orange). Data for a5 in orange represent a sequence of repeatedly removing the sensor and repositioning it first into the Zeonex apparatus twice, then into the Delrin apparatus twice, for a total of eight switches.

Testing of Dolomite piezoelectric pump The experimental pump is a peristaltic pump provided by Dolomite, whose piezoelectric diaphragm responds to changes in voltage and frequency supplied to the pump by alternately projecting and retracting the diaphragm at a certain frequency, pushing fluid in the process. In general, voltage delivered to the system allows for a higher frequency of contractions, which in turn produces a higher flow rate. In contrast, the current experimental setup for the ML2 chip utilizes a syringe pump for driving fluid through the system. A syringe pump applies a linear force with a step motor to achieve a volume displacement that will produce a fixed flow rate as long as the linear force supplied by machine can overcome the resistance in the system. The peristaltic pump alternative also displaces a certain volume of fluid on every stroke of the diaphragm, but a continuous force is not supplied in the same manner. To determine the theoretical application of the pump to the chip, one can begin by estimating its hydrodynamic resistance, the resistance that it presents to fluid flow at a given applied pressure. According to a simple model, the resistance provided by a channel with a rectangular cross-sectional geometry may be described as

(1)

Page 20: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 20 / 107

where is the resistance measured in

, is the dynamic viscosity ( ) , is the length of the channel ( ),

is the width of the channel ( ), and is the depth of the channel ( ). Equation 1 only holds for the case where

, which is valid for this design. Units of resistance may also be reported in

after applying the necessary

conversions. The internal architecture of the chip is far from uniform, nor is it precisely a cuboid shape, featuring bends and a series of acute angles, but a preliminary estimate indicates that the different features of the chip have the properties outlined in Table 7.

Table 7: Estimates for internal hydrodynamic resistance of the chip using feature dimensions

Chip component Length (mm)

Width (mm)

Depth (mm)

Internal volume (μL) Flow resistance (kPa*s/μL)

Entry/exit segments 50 0.25 0.05 1 19 Mixer segment 78 0.18 0.05 1 42 Activation segment 340 0.26 0.05 4 126 PCR segment 1078 0.26 0.05 14 400 TOTAL - - - 20 587

Similar calculations were made for the tubing connected to the fluidic chip, using a cylindrical model for the resistance. However, the predicted external resistance from the tubing differed from the chip resistance value by more than three orders of magnitude, which indicates that the chip represents the largest source of resistance in the system. Given this finding, the tubing external to the chip was considered a negligible source of flow resistance. The value of total chip resistance can be easily used to predict the pressure difference that must be generated by the pump to produce a desired flow rate, assuming an uninterrupted fluid phase in the absence of gas. For a desired flow rate of 0.5

μL/min -- a rate important for PCR reactions in the current chip design -- the necessary pressure input to the system must be roughly 5 kPa higher than at the end of the tubing. Given that the maximum output of the piezoelectric pump is 90 kPa, use of the pump appears to be quite feasible under these circumstances.

Page 21: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 21 / 107

DISCUSSION

Conclusions for validation of microfluidics-based PCR with Zeonex 690R chip Microfluidics-based PCR has the potential to offer significant savings in reagent costs and time over traditional thermal cycler-based PCR. Using an in-house machined Zeonex 690R microfluidics chip, we were able to detect the enteric pathogen E. coli O157:H7 by PCR. The Zeonex 690R prototype chip was developed in order to test hot embossing protocols and to demonstrate feasibility of our designs and processes in PCR applications. Moving forward, the microfluidics-based PCR will require some optimization. These final rounds of optimization require the final chip as material properties such as roughness and composition can have a significant impact on the microfluidics chip performance. Once we receive the final chips, areas for optimization will include improving the flow rate through the microfluidics chamber and detecting amplification of the

product in situ as described below. Our target for reagent use is <2L PCR reaction mix/experiment (in a total volume of 20

L) and a total experimental time of <10 minutes. As we investigated our microfluidics chips, we were able to determine that a primary reason achievable flow rates were lower than expected was significant compression and deformation of the microfluidics channels while bonding the cover slip to the

Zeonex 690R chips in the hot press. The original design calls for a channel depth of 50 m which should allow for even flow

and minimal backpressure. However with hot press bonding, some of the channels were compressed to 30 m which significantly hampered the fluidics system. We have addressed this compression by milling new chips with deeper channels

(70 m) to allow for downstream compression. To further improve the flow rate, we have developed a hot embossing system that directly molds the microfluidics channel design into the chip. Hot embossing created extremely smooth channel features that minimized backpressure when compared to our ultra-precision machine milled chips. We were also able to hot emboss PMMA chips and anticipate that the embossing procedure will greatly facilitate final chip production. Current methods of detection involved waiting for the reaction slug to traverse all 40 cycles of the chamber for collection at the exit port. In situ detection using an in-house optics system will allow us to read fluorescently-labeled amplification directly in the channels. Identifying the cycle at which amplification crosses the limit-of-detection enables us to quantify the amount of E. coli O157:H7 in the sample. This quantitative information will greatly improve the utility of the ML2 platform. We have been reporting the results of our heater and sensor tests to the group at CETEMMSA and they have developed improved heater and sensor pairs that implement a control-loop to address the challenges we observed. These controlled heater-sensor pairs will be included in the final roll-to-roll manufactured chips.

Page 22: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 22 / 107

Conclusions for testing of second generation CETEMMSA printed heaters

As described previously, a significant complication arose during testing in that some changes in input power occurred for all heaters while experimentation was underway. For this reason, data from the stepwise test proved to be somewhat imprecise in predicting the starting power necessary to achieve a desired target temperature. In particular, the COP heater displayed considerable instability throughout all experiments, and this perhaps is reflected in the visual degradation of the printed material observed upon receipt of the heater collection. The changing voltage and current may imply a thermally-dependent resistance of the heaters, which could be an inherent feature of the printed heaters regardless of substrate. No feedback control element was present in the current testing system, and one can anticipate that application of such a feature will assist in tuning the temperature to a desired setting despite the changing resistance. A set of sensors, also provided by CETEMMSA, may sufficiently serve this purpose. However, as seen in the time-course studies, a considerable amount of time, on the order of 40-50 mins, is required for the entire system to achieve an equilibrium state when making large transitions in the temperature setting, and more time will undoubtedly be required to make small adjustments towards a specific temperature. On the other hand, as seen with the hold testing, once a satisfactory setting has been found, the heaters, especially A1 and B1, are generally quite capable of operating continuously at high temperatures. Interpolating from the trend lines calculated for the temperature-power plots will provide estimates of the power required for each heater to produce internal chip temperatures of 62°C and 95°C, as seen in Table 8.

Table 8: Estimates for power requirements of heaters to yield key PCR operating temperatures

Given the present findings, it appears that using heater A1 will require the least amount of power, 6.2 W, to generate two static temperature zones, as these are the minimum power settings of all the heaters tested. That being said, due to their relative proximity on the chip, one should anticipate that the two heaters will also heat each other, reducing the power needed by each one. Because of the apparent temperature dependence of resistance for these printed heaters, this mutual heating will likely produce additional difficulty in tuning the heaters to the desired set points. Such information can be obtained by further studying heat transfer within the chip and heater control while using multiple heaters. Another important factor that impacts the absolute data recorded in this series of experiments is the experimental setup itself. One should keep in mind that the current testing platform is only an approximation for the anticipated application of the

Heater Predicted power for 62°C (W) Predicted power for 95°C (W)

A1 2.3 3.9

B1 2.3 4.1

C1 2.8 4.9

D1 2.9 5.1

Page 23: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 23 / 107

heaters. A significant distinction is that the current ML2 microfluidic chip design is larger than the chip analog used in this study, and would thus in theory require more power to maintain the same static temperature zone. Embedding the heater and fluidics between Delrin no doubt assists in thermal insulation, which will improve efficiency of the heating element, but the specific arrangement will change in the final application. Alternative layer thicknesses and even materials may be pursued with this setup for their usefulness in other functional layers of the chip that would normally enclose the fluidic and heating layers, namely the optical and base layers. Information gleaned from the high to low temperature time-course, which represents thermal dissipation in the absence of direct heating, may be used to model the current system thermodynamically and provide guidance if future changes are made. As an additional concern, the heater design may also be scaled down to ensure that the heating element does not exceed the dimensions of the chip, as the portion of the heater that is not directly in contact with the chip is producing excess heat that will inefficiently contribute to the chip’s temperature and use power unnecessarily. Regardless of these changes, however, one can anticipate that the relative results from this experimentation are valid, in that the A1 and B1 heaters, made with Kapton and PET respectively, perform more favorably for the desired application. Along with the matters discussed thus far, future testing will address the variability between heaters composed of the same material. Also subject for subsequent inquiry is a focus on spatial uniformity of heating as measured by thermocouple arrays embedded within a new chip analog that more closely approximates the ML2 chip. Non-uniform temperature zones existed with prior heater designs, and need to be assessed thoroughly in the current design due to their potential impact to the PCR application. As was the case with this study, the goal in further experimentation will be to determine how well the current heaters may be best applied to the system as a whole towards development of a cohesive, integrated PCR lab-on-a-chip device.

Conclusions for testing of CETEMMSA T sensors Ideally, the printed T sensors would provide a well-defined and reproducible linear relationship between the temperature and resistance, as this will enable application of the sensors to a feedback circuit to control temperature of the system. Because the current testing regimen featured both a thermal source (hotplate) and sink (surrounding air) for heat during the heating stage but only a sink (air) during the cooling stage, it is also possible that a hysteresis curve could describe the relationship of the resistance and the temperature. However, it was apparent that sensor behavior followed neither relationship, as resistance values did not match at the beginning and end of each experiment. The ubiquitous decrease in resistance was also additive between experiments. Future modifications to the testing protocol can attempt to equalize rates of heating and cooling to rule out hysteresis as a factor, but it should additionally be noted that the time scale of either the heating or cooling processes was significantly slower than the frequency of recordings collected by the thermometer and the multimeter. As such, one would expect that contribution from differences in the heating rate would still be minimal, and this phenomenon would still fail to explain the permanent overall reduction in resistance.

Page 24: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 24 / 107

Testing with the D4 sensor was arrested prior to obtaining a series of data to fully characterize the sensor, so a rigorous analysis of the performance of the sensor design on COP is not possible. However, the fact that this sensor repeatedly produced overloaded resistances suggests that it was unstable with respect to the other sensors tested, which did not experience this issue. The behavior may indicate changing material properties of the sensor itself in response to heat and pressure of the system, though further experimentation with a similar sample material is necessary to corroborate any findings with this particular sensor. While the D4 sensor could be considered a more extreme case, the downward trend of resistance values upon successive heating and cooling cycles suggests that all sensors possessed some inherent issues in stability, possibly due to manufacturing technique or to features of the ink. One possible hypothesis holds that the pressure supplied by the clamp in the apparatus compressed the layer of silver made more malleable by heat, establishing better electrical conductivity by reducing the porosity of the ink. This effect will need to be reduced by exploring methods of enabling good electrical contact without requiring excessive pressure application, as well as by increasing the sensor’s overall durability. Adding to the material considerations, sensors A4 and A5 consistently differed in their resistance values, although they were printed using the same design on the same material. Some other important property or set of parameters for the sensor must then contribute to its performance. For instance, the thickness, alignment, or porosity of the ink may vary significantly. Efforts should be made to optimize uniformity of such properties, since variability in sensors that should otherwise be identical may impact mass production of the component and affect reliable incorporation into an otherwise functional chip. Despite the repeated drops in resistance values during experimentation, testing with A5 suggested that a relatively stable condition may be achieved for the sensor, as results from later experiments appeared to converge (Figure 12B). If this outcome can be reproduced, it may be utilized as a standard preparation for the sensors before calibration testing, establishing a requirement that each sensor undergo thermal cycling to achieve a more reliable performance. Further testing of future sensors that assesses this process is a significant priority moving forward, especially as variability in operation may still occur even afterwards. As subsequent sensor designs become available, continued work will focus on the stability of the sensor properties as they relate to performance. Current improvements in the sensor ink material will no doubt also assist in improving their function, especially when combined with a cycling procedure to thermally stabilize the sensors prior to testing. Also, although this study featured a significant effort to standardize how the sensors were calibrated, further effort is required to ensure that the protocol and testing conditions maximize reliability to obtain useful data. A new experimental platform has been designed that will incorporate both heater and sensor testing in the same unit, and help to mitigate the issues posed by the experiments to date. Use of this platform with improved heaters and sensors will represent a suitable advancement towards the project goal of full integration of components.

Page 25: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 25 / 107

Conclusions for testing of Dolomite piezoelectric pump Because the Dolomite piezoelectric pump is an off-chip function, while the final chip is being generated by roll-to-roll manufacturing procedures elsewhere, we will continue to characterize and optimize use of the pump. Fluids entering into the

pump need to be filtered with a pore size no greater than 5 μm. Additionally, to prevent particles in the air from entering the filter, as well as possible fluid spillage onto the control circuit, the entire device will be encased in a protective barrier constructed for the pump. The tubing compatible with the pump is larger than the current ML2 fluidic setup, so adjustments must be made to ensure proper connectivity to the chip. Once the device forms a fluidic connection to the chip, initial characterizations will determine whether the pump is powerful enough to overcome the resistance provided by the channels in the chip, as has been theoretically established. In assessing the ability of the new pump to deliver a consistent flow rate, one should evaluate sensitivity to dynamic changes in resistance caused by compliance in the tubing or by variations in viscosity due to temperature -- as is the case with the PCR temperature zones -- or phase boundaries in the presence of air. Also, at maximal frequency, one would expect that the efficiency of pumping will suffer as the diaphragm fails to open completely, reducing the theoretical flow rate at higher levels of voltage input. Such a phenomenon may be evaluated as well. Specifications for the pump list pure water as a recommended chemical, however PCR requires a salt solution with various components. One may wish to determine whether corrosion or clogging of pump may become an issue with repeated use of biochemical reagents, ideally prior to using the pump to conduct a full PCR experiment.

Page 26: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 26 / 107

Demonstrator 2 (J. Hue, G. Nonglaton)

Summary p 28

1. Introduction p 29 2. Demonstrator 2 : status from a functional point of view p 29

2.1 Status summary from a functional point of view p 29 2.2 Details concerning the functional status of the platform P2 p 31

2.3 Illustration of the (functional) results obtained with P2 p 39

2.3-a Introduction: names and compositions of various liquid solutions used with P2

2.3-b Results p 39

3. Substance P detection with P2 p 42

3.1 Few data suggest the specific detection PMMA (at an interesting level, i.e. below or around 100 ng/ml) p 42 3.2 Data show specific detection on PMMA without any ambiguity on stabilized process p 43

3.3 Fluidic protocol p 44

4. Potential instrumentation improvements concerning the LOD (LOD = Limit of detection) p 45

5. Conclusions and future work concerning P2 p 47

Page 27: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 27 / 107

Annexes Annex 1 : General views of P2 p 49 Annex 2 : Environmental applications and reagents p 49 Annex 3 : Current excitation module (not R2R compatible) p 53 Annex 4 : R2R Excitation module LGP solutions proposed by Polyscale p 55 Annex 5 : Excitation module OLED solution proposed by IPT p 57 Annex 6 : Detection module (PMT) p 59 Annex 7 : Emission filters for the substance P p 61 Annex 8 : Plastic selection based on the autofluorescence criterion for the demonstrator 2 (PMMA) p 62 Annex 9 : Surface preparation of the PMMA slide p 66 Annex 10 : How the LOD is evaluated and extrapolated if necessary? p 67 Annex 11 : Suggestions for a disposable fluidic functionalized chip in R2R (to start discussion, especially with IPT) and discussion to transfer the fluidic chamber in R2R process (discussion with IPT during a meeting in June 2014)

p 68 Annex 12 : Discussion and results concerning the LGP (Light guide plate) from Polyscale p 75 Annex 13 : Pump(s) p 80 Annex 14 : Electro valve p 85 Annex 15 : Discussions and calculations in progress with Polyscale to collect more light in modifying the geometry p 86 Annex 16: Impact of the microfluidic tubes and their preparation: Saturation of the fluidic circuit with BSA p 90 Annex 17: Various hybridization results p 92 Annex 18: Output light versus PMMA thicknesses p 101 Annex 19: Discussion with CETEMMSA p 102

Page 28: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 28 / 107

Summary LOD = Limit Of Detection AB = antibodies BSA = Bovine Serum Albumine PMT or PM : Photomultiplier tube or photomultiplier The platform 2 is fully functional for various tests (to detect the substance P) We are waiting for the IPT chamber and the Polyscale excitation module for additional tests. These components might be achieved in R2R. Dolomite electrovalve are working and implemented on P2. They are not currently used on P2. Bartels pump has been successfully tested with deionized water. Now, this pump has to be tested with PBS. The geometry with the circular PMT should allow using an excitation source with a lower optical power. It should open the possibility to use an excitation source which is R2R compatible (OLED, LGP from Polyscale, etc…). The light stability is an important factor a get a low LOD. Now, the observed behavior of the antibodies is almost the expected behavior. It is possible to achieve detection with functionalized area on PMMA (plastic slide) but not at the target concentration. The best specific detection achieved on P2 with a PMMA slide, is between 1ug/ml and 3 ug/ml (i.e. typically 2 ug/ml, it means 4000 times the target concentration). We hope to get specific detection below 100 ng/ml in a short time (200 times above the target concentration). CEA focus its effort to stabilize the functionalization process and the fluidic protocol to get “reproducible” LOD for specific detection. Then, if it seems possible, CEA will improve P2 from an operational point of view to get smaller specific LOD.

Page 29: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 29 / 107

1. Introduction The report concerning the demonstrator 2 is divided in 2 main parts. The first part is dedicated to the prototype itself (i.e. more accurately to the platform P2). This part is mainly devoted to the instrumentation (components, functions,..), which are inside this prototype. This (versatile) platform has been built to achieve various tests for the instrumentation, for the functionalization and the detection and to tests various R2R components supplied by the ML2 partners. This platform has also been built to evaluate the functionalization and the detection with plastic slides. This prototype allows validating or invalidating various choices (instrumentation, geometry, functionalization,…). In this part, the exchange with ML2 partners will be mentioned. The second part is dedicated to various tests concerning the detection of the substance P on plastics substrates, These tests are achieved on plastic slides, which are functionalized with technics which are R2R compatible. To get detection (i.e. a limit of detection (LOD)), 3 conditions have to be gathered: -A prototype, which works, -A reproducible plastic slide functionalization, -A suitable (fluidic) protocol with the suitable buffers: to prepare the samples, to clean them, to enhance the reaction between the probe (the tripod with the antibody) and the target (the substance P) etc…. Unfortunately, the slide functionalization and the fluidic protocol are not similar for the glass slide and for the plastic slide. At the beginning, the P2 development and the functionalization development have been achieved simultaneously with cross tests. From September 2014, it seems establish that: -The prototype works, -A reproducible process for the functionalization concerning the fluorescence level is validated, It is also observed that the catching of the antibodies by the tripods work. Only the specific detection of the substance P, at the target level, is missing. These conditions are necessary to develop a suitable and reproducible fluidic protocol to specially detect the substance P and to improve the capture efficiency of the substance P by the antibodies. A part of the instrumental tests and functionalization on P2 have been gathered in various annexes to illustrate the conclusions inside this report. The annexes illustrate also the exchanges with the ML2 partners. 2. Demonstrator 2: status from a functional point of view 2.1 Status summary from a functional point of view The demonstrator 2 is arbitrarily divided in many parts. The status of each part is analyzed. It will be mentioned if the components are R2R or nor R2R, or seems compatible with a R2R process from our point view (see Figure 2.1,) The comments concerning the detection are linked to the detection of the substance P (Application 1). For another application (i.e. for another dye), the wavelength of the excitation module, the optical filters and the fluidic protocol have to be adapted. The fluidic protocol will be also different since the detection is not achieved with a continuous flow.

Page 30: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 30 / 107

Function/components Status inside the platform 2

(Not R2R components) Components with compatible

with a R2R process

1-Excitation module OK May be, in progress Exchange with Polyscale

The possibility to collect more

light with a circular PMT, associated with a fiber bundle,

cannot offer the possibility to use a less sensitive sensor, but a

detection might be achieved with a lower excitation power.

2-Detection module OK No

3-Optical waveguide OK-PMMA

1 mm, 500 m, 200 m

200 m with the IPT chamber (forecasted)

May be Exchange with IPT

4-Fluidic chamber OK In progress Exchange with IPT

Optical waveguide and fluidic chamber : same component

5-Tube OK No comment

6-Pump OK Pick and place : Bartels pump Tested outside P2

(test remaining : with saline solution)

7-Electro valves OK Pick and place : Dolomite valves

Not currently used on P2 but already implemented on P2

8-Plastic surface preparation to immobilize the probes and

the probes immobilization

Selected plastic : PMMA Status: OK from a functional point of view. The result concerning the LOD on PMMA is not satisfying. (2 ug/ml reached , targeted : 0.5 ng/ml a factor 4000)

Same comment than the status inside the platform 2

9-Electronic printed elements Status : none/NA Status : none/NA

Options have been discussed due to the possibility by CETEMMSA to coat plastic surfaces with metal (see Annex 19)

10-Fluidic protocol The components of P2 should be sufficient for the fluidic protocol

Same comment than the left column

Future tests: CEA mainly waitsfor excitation Polyscale modules and IPT chamber

For more details concerning this table see the next pages

Figure 2.1: Status summary concerning the demonstrator 2 (Prototype Platform P2 and R2R components)

Page 31: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 31 / 107

2.2 Details concerning the functional status of the Platform P2 This part details the results, which are reported in the previous part.

Page 32: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 32 / 107

Function/components Status inside the platform 2 Components compatible with a R2R process

1-Excitation module -LED (centered at 515 nm) -Interference excitation filter -Ball lenses The ball lenses, and the distance between the excitation module and the functionalized spots can be adapted. In using ball lenses of 5mm or 8 mm in diameter, the light spot diameter can be adapted (typically between 2 mm or 5 mm in diameter). The optical power can be modified by software in choosing the current inside the LED.

Status : OK (For the substance P), the excitation wavelength has been shifted from 530 nm (maximum absorption for the Rhodamine 6G) at 515 nm to test if it is possible get a detection without an excitation filter. It has been experimentally established that with an excitation wavelength at 532 nm, without an excitation filter, the sensor is saturated. At the 515 nm, without an excitation filter, the sensor is not saturated. Detection measurements should be possible. A slight offset has been detected. It will have a negative impact on the LOD (Limit of detection). For a spot size of 2 mm, the maximum total power, on a functionalized spot, has been measured above 6 mW. A decrease of the optical power can be compensated by the PMT gain. This experiment has not been achieved. Be careful, when the PMT gain increases, the standard deviation of the signal increases and it has a negative impact on the LOD. The stability of the excitation module has impact on the LOD.

Solution 1 : OLED from IPT Tested outside the platform Spot : square 5 mm x 4.5 mm Centered at 513 nm Maximum power without excitation filter : 2.0 mW “Regular power without the excitation filter” : 1.4 mW Estimation of maximum OLED power without excitation filter for 2 mm x 2 mm square spot: 0.35 mW (= 20* 4/22.5, typically 20 or 40 times less than the excitation power with our current excitation module without or with an excitation filter). The power estimation for a 2 mm x 2 mm OLED does not take into account the optical divergence will have a more negative impact than the 5 mm x 5 mm OLED (see Annex 5 for the test results) With a circular PMT configuration, the excitation power of the OLED might be sufficient. (see 2-dection module of this table) Solution 2 : LGP (Light Guide Plate) from Polyscale This solution has been tested at CEA (see Annex 12). This component can be tested on P2 (mechanically and electronically compatible). Unfortunately, to extract a maximum of light, the (excitation) LEDs of this component are plugged in serial. Therefore, the P2 driver cannot deliver a sufficient current for the 2 LEDs. During the test, the LGP has been slightly damaged on the edge due to a LED crash ! For additional tests, LETI waits for new LEDs on PCB. Polyscale will probably send to CEA 2 sets of wavelengths. 1-The LEDs will be tested alone, 2-The LED will be tested with the LGP, 3-The LGP could be tested with a Rhodamine 6G solution in the chamber (to overcome the difficulty to put in coincidence the functionalized spots and the illumination areas). This solution, which seems R2R compatible, could have a drawback: the cross talking between the illumination areas and the light between the illumination areas. The stability of the incident light is also an important factor which has to be evaluated (see Annex 12) Solution 3 : LGP + BEF (Bright enhanced film) Solution 3 = Solution 2 + an adapted film This solution has not been tested. CEA did not receive any BEF.

Page 33: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 33 / 107

Function/components Status inside the platform 2 Components with compatible with a R2R

process

2-Detection module

2 geometries are possible on P2, both with a PMT. With the solution 1, a rectangular PMT is used. With the solution 2 a circular PMT associated with a fiber bundle, is used. The fiber bundle has a rectangular section (from the slide side) and a circular section (from the PMT side). In both cases, the same emission filters are used : -a long-pass colored filter, -an interference filter. With the application 1, it is not possible to replace the interference filter by a long pass color filter since the antibodies are tagged with a red dye (at least at that stage), see Annex 7. Both solutions have a similar impact on the LOD with an advantage with the circular PMT configuration (typically a factor 2 on the LOD), when they are both located orthogonally to the edge of the slide output. Nevertheless, it has been estimated that the solution with the circular PMT collects typically 150 more light than the configuration with the rectangular PMT. For a similar gain, the sensitivity of the circular and rectangular PMT are similar (slightly better for the circular PMT around 600 nm). The maximum gain, which can be used with a rectangular PMT, is higher (~ x 10). The dynamic range is divided by almost 2.5 with the circular PMT (4V versus 10V). The light collection is better with a circular PMT but the gain on the LOD is small mainly due to the standard deviation of the data.

Status : OK Remark: In collecting more light, it is possible to use a less sensitive sensor behind the circular fiber bundle. Unfortunately, to use a photodiode, a light level of a micro watt is necessary. With a circular PMT, an optimistic estimation indicates a nW level of light behind the emission filters. The other solution is the possibility to use excitation source with a smaller power (like the OLED).

No ML2 partners are involved. The detection limit of organic photodiode is not compatible with the application of the demo 2 (See Annex 6)

Organic photodiode

Page 34: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 34 / 107

Function/components

Status inside the platform 2 Components with compatible with a R2R process

3- Optical guide (one side of the optical chamber)

Plastic choice : PMMA (Based on autofluorescence level) The autofluorescence level of PMMA is compatible with the detection (similar to a microscope glass slide in BK7). The fluorescence level varies from supplier to supplier (and no thickness law have been found) (see Annex 8) To develop the functionalization process on PMMA, 1 mm PMMA thickness is used. It is easier for the functionalization development and for the tests. Nevertheless, the test of 500 microns and 200 microns PMMA slide are possible on P2. The fluorescence light level has been evaluated in evanescent wave for 3 different PMMA thicknesses (1 mm, 500 microns and 200 microns) with fluorescent tripods give similar level in epifluorescence. There is almost no difference between the light output for thickness of 1 mm and 500 microns. For a 200 microns slide, the signal is typically ~ 60% of the signal got with a 1 mm substrate. It is not always easy to keep the 200 microns PMMA slide flat with P2.

Status OK

For the current tests, the PMMA substrates are supplied to CEA by Polyscale at the wished shape (laser cutting, 25 mmx 75 mm x 1 mm, rectangular shape)

The thicknesses for a PMMA chamber in R2R are : 200 microns (or 500 microns) Above 500 microns, it is too thick for a R2R process (IPT input).

The tests on P2 indicate that more light go out from 500 microns than from 200 microns PMMA substrate.

Status : OK (In progress) (see Below : fluidic chamber)

(the optical damage with a 200 microns slide has to be studied, see figure 4.1)

4-Fluidic chamber Current tests are achieved, with a channel height of 300 microns. Many clues indicate that is not reasonable to achieve hybridization with channel height of 50 microns or 100 microns. (Difficult to add the antibody on the tripod). A 300 microns chamber is a good compromise. Indeed for a higher channel (380 microns for instance), the generation of (large) air bubbles are easier. Development of new gaskets in PDMS has been developed for an easier handle. These gaskets are larger than the previous gasket (i.e. for current tests wo R2R technics) First chamber height with this technic : 380 microns or 150 microns

Status OK For the current tests, one side of the

Towards a R2R chamber in PMMA Meeting at IPT (25

th/26

th June 2014)

See Annex 11 IPT will achieve a fluidic chamber in R2R by the “method cut and assembly” - With 2 PMMA sides (thickness = 200 microns) - With a gasket achieved with a PMMA of 200 microns + 2 optical clear adhesives layer (2 x 50 microns) it should lead to a thickness compatible with the current hybridizations (i.e. roughly 300 microns) (channel width : typically 6 mm to have the possibility to use functionalized spot up to 5 mm in diameter) After this meeting, CEA sent various specifications to IPT to insure the compatibility of the future R2R chamber with P2. The quantities of tripods (i.e. the probes),

Page 35: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 35 / 107

chamber is in PMMA and the other is in aluminum with an isoflon coating. 2 connectors in plastic are screwed on the aluminum side. These connectors are compatible with ID =1.6 mm and OD =3.2 mm (part of the fluidic circuit might be in ID = 0.8 mm and OD =2.4 mm to integrate the Dolomite valve and ID = 2.6 mm ID = 1.3 mm to integrate the Bartels pump) In June 2014, the fluidic connection between the R2R chamber and the test platform P2 has been discussed with IPT. IPT suggests using the “Dolomite connectors” or “glue connectors”. The chamber length can be adapted to be compatible with these fluidic connectors and P2.

which have to be dispensed on the PMMA, have been discussed (see item “PMMA Surface preparation to immobilize the probes”). The IPT dispensing unit from Häcker Automation GmbH might be a suitable solution in R2R process. The DURST technic cannot be tested since a too large volume is needed (incompatible with the cost).

In progress Remark : For future tests test on P2, the accurate location of the tripods are essential. (coincidence between the functionalized area and the light spot)

5-Pump Currently, Instech pumps are used on the platform. Flow rate : Between 50 ul/min and 2300 ul/min

Status : OK

Remark: An Instech pump has replaced by a same Instech pump due to a failure during the tests.

Pick and place solution: piezoelectric pump from Bartels. This pump arrived at LETI in September 2014. The Bartels pump has been continuously tested during 45 hours outside the platform in controlling its flow rate with no major trouble with desionized water (see Annex 13) The next step is to test it with phosphate buffer saline (PBS).

Status OK

Dolomite sends to CEA a piezoelectric pump for a pick and place solution. CEA evaluated this pump through various tests. This solution was not acceptable: the flow rate drastically decreases versus time (see Annex 13)

6-Tube The first implementation on the platform was silicon platinium tube (ID =1.6 mm and OD =3.2 mm) with pinch electrovalves from SIRAI. The pinch electrovalves have been chosen to avoid the cross contamination. This solution was not acceptable. It is more suitable to use Tygon R-3603 tube (or now E-3603) to avoid the tube contamination (see Annex 16 for more details). With Tygon, no “unusual” signal due to cross contamination has been observed. It was not the case with a silicon platinium tube

Status : OK Remark : Each night, the fluidic circuit is prepared for the future hybridization in introducing a given “buffer” with BSA inside during 8 hours (at 100 ul/min).

No comment

Page 36: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 36 / 107

7-Electrovalve For the hybridizations, no electrovalves are currently used. (We used the pump like an electrovalve even if the liquids can mixed together (i.e. due to the liquid diffusion)) Nevertheless, it is possible to drive 5 Dolomite electrovalves with our Labview software (electronic and software are working). All the hardware is ready, tubes with suitable diameters, connectors, etc…

Status : OK

These electrovalves use smaller tubes than the tubes currently used : ID = 0.8 mm, OD = 2.4 mm instead of ID = 1.6 mm, OD = 3.2 mm We are waiting for hybridization stabilization to directly install them on P2. No show stoppers have been identified from an instrumental point of view (ageing effect has been studied during one night) 3 tube sizes will be simultaneously used on P2 with the suitable connectors.

Pick and place solution : Electrovalve from Dolomite (see the photo in the adjacent column)

Status : OK

Page 37: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 37 / 107

8-Plastic surface preparation to immobilize the probes and the

probes immobilization

Various technics of surface preparation have been investigated by IQS and CEA only R2R compatible technics have been selected 2 R2R compatible possibilities have been investigated to prepare the PMMA surface: -One coming from IQS, -One coming from CEA. For both technics, on PMMA substrates, tripods are located on the functionalized area and AB are recognized by the tripods. At that stage, to detect the substance P, CEA chooses to focus its effort on the stabilization of the functionalization (CEA) process. Status: OK from a functional point

of view. The result concerning the LOD is not satisfying yet. Compatibility of the lab technic with R2R : no discussion in this report. The CEA work on this subject is gathered in a memo entitled: “: Report on roll-to-roll compatible chemical functionalization”

Same comments than the right column

Status: OK from a functional point of view.

The result concerning the LOD is not satisfying yet.

Remark : The cost of the probes, to test the inkjet printing at DURST, due to the required volume prohibits this technic. The IPT dispensing unit from Häcker Automation GmbH might be a suitable solution, at least for the tests. (This problem does not exist with the target product since the target product is only used during the tests. Therefore, the quantity will remain similar than the quantity currently used at the lab level) IPT have a dispensing unit from “Häcker Automation GmbH” in clean room which could be used to spot the tripods (typical total volume = 3 mL versus 100 ml or 1000 mL for the inject printing) Minimum volume has to be studied Discussion with DURST (and COATEMA) for additional tests Tripod volume (i.e. the probe volume) necessary for an inkjet printing test at DURST : 100 ml and then 1000 ml (1000 mL of tripods 0.5 MEuros ????) Therefore, to evaluate (and to debug) the inkjet printing deposition, CEA tries to develop a test based on oligonucleotide recognition on PMMA substrate (less costly). Unfortunately, a large protein has been necessary. It was not commercially available This test has been discarded.

Page 38: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 38 / 107

9-Electronic printed elements

No comment-See deliverable D7.2

The main electronic components of P2 are described inside D7.2 (M18). Most of them are coming from industrial suppliers. A laptop handles the experimental process.

No electrical components are forecasted in printed electronic. In September 2014, CETEMMSA and CEA discussed to examine what might be useful and possible to achieve at this stage. We conclude that it is useless to propose an electronic architecture based on electronic printed elements for the demonstrator 2 (see Annex 19) Nevertheless, due to the capacity of CETEMMSA to coat metal on PMMA, few suggestions were made : 1- CEA proposes that CETEMMSA examines with IPT, if it is possible to add a heater on the fluidic chamber. Even if is not compulsory for the application 1, it could help to have more reproducible results during hybridizations. Heating could be testing outside P2. Temperature Range : between 30° and 40° at a given temperature 2- CEA proposes that CETEMMSA examines with Polyscale if it makes sense to add to an absorbing or a reflecting coating on the future excitation module between the illumination areas to avoid the cross talking (i.e. see excitation module, solution 2) 3- In September 2014, CEA proposes that CETEMMSA to coat lateral edges of PMMA slides to enhance the light which will be direct towards the sensor due to additional tests. Due to additional tests during the last days, CEA is not convinced that a coating will enhance the autofluorescence light output. Moreover, it could be difficult on PMMA of 200 microns

10-Fluidic protocol The components of P2 should be sufficient for the fluidic protocol

Same comment

Page 39: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 39 / 107

2.3 Illustration of the (functional) results obtained with the prototype P2 2.3-a Introduction: names and compositions of various liquid solutions used with P2 PBS buffer : Phosphate buffer saline PVP buffer : Phosphate 0,1 M pH 7,4 with 0,1 % of PVP[poly(vinyl-pyrrolidone)], 0,15 M NaCl and 0,01 % of NaN3. EIA buffer: Phosphate 0,1 M pH 7,4 with 0,1 % BSA (Bovine serum albumin), 0,15 M NaCl and 0,01% of NaN3 Washing buffer : Phosphate 10 mM pH 7,4 with 0,05 % of Tween 20 2.3-b Results The figure 2.3a and 2.3b illustrate that P2 works well from a fluidic, optical and electronic point of view. The figure 2.3a shows a typical fluorescence decrease due to the antibodies, which grip the tripods on the PMMA. This preparation is essential to detect the substance P. Most of the time, from a functional and operational point of view, this step works well. From an operational point of view, this step is not a show stopper, even if the fluidic protocol can be improved. The figure 2.3b shows “a detection of the substance P”. It is the result that we are waited for. Unfortunately, on this graph, the reference line, before the substance P injection, is got with a PVP buffer. And the substance P is diluted in a PBS buffer (and not in PVP) buffer. Therefore, from an operational point of view, the specificity of the detection is not insured. But from a functional point of view, the prototype is working. Unfortunately, when a PBS line is used as reference (or if the substance P is diluted in PVP), the contribution of the substance P is not detected (see Figure 2.3c)

Most of the time, there is no difficulty to “quench” the tripods on PMMA with the Antibodies (AB).

It is more difficult to guarantee that the binding, with tripods and the AB, is specific Figure 2.3a: Typical AB decrease (on PMMA)

1

1,2

1,4

1,6

1,8

2

2,2

0 2000 4000 6000

Vo

ltag

e (

V)

Time (s)

Page 40: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 40 / 107

It cannot be guarantee that the increase is due to the substance P since the substance P was not diluted in the PVP buffer.

The LOD extrapolation gives an exact result if : -The signal is only due the substance P (and not to the PBS) probably a part is due to the substance P since the slope increases with the substance P concentration (not reported in this graph) - The whole signal is only due to the specific adsorption - (the linear extrapolation remains true)

Figure 2.3b: detection of the substance P from a functional point of view (Unfortunately, the specificity is not guarantee)

0,6

0,7

0,8

0,9

1

1,1

1,2

1,3

1,4

0 200 400 600 800 1000 1200 1400

Flu

ore

sce

nce

si

gnal

(a.

u.)

Time (seconds)

Substance P detection of 20 ng/ml in PBS in less than 200 seconds on PMMA (?)

LED 1

LED 2

LED 4

Linear extrapolation : LOD ~ 0,8 ng/ml (?)

PVP

PVP 20 ng/ml of SP diluted in PBS

Page 41: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 41 / 107

It clearly appears that the signal is not sensitive to the substance P (No slope change with the substance P concentration)

Unspecific detection (CEA surface preparation) - PMMA substrate

Figure 2.3c

2 reports respectively entitled “Demonstrator 2 setup. Part A : Software for the setup (version 0)” and “Demonstrator 2 setup. Part B : Tests (optical, fluidic) on platform 2” respectively from 15/11/2013 and from 22/01/2014 are available on the ML2 web site. These reports give more details concerning the software, which drives the platform P2.

0

0,5

1

1,5

2

2,5

3

0 1000 2000 3000 4000 5000 6000

Flu

ore

sce

nce

leve

l (a.

u.)

Time (seconds)

plot 1 plot 2

plot 3 plot 4

PBS only

PBS + 200 ng/ml SP PBS +1000 ng/ml SP

H83 (PMMA)

Page 42: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 42 / 107

3. Substance P detection with P2 3.1 Few data suggest the specific detection on PMMA (at an interesting level, i.e. below 100 ng/ml) Data, with a specific detection of the substance P, have been obtained on functionalization processes, which are not reproducible and stabilized (figure 3.1a, figure 3.1b). For instance, it was difficult to reproduce the initial autofluorescence of the tripod for these given processes. The figures 3.1a and 3.1b show strong clues that the specificity has been reached with PMMA substrate.

In dividing the substance P concentration by 5 (200 ng/ml instead of 1000 ng/ml), the signal is divided by 5. It is a strong clue that these 2 signals are specific to the substance P. No washing buffer (WB) has been used to get this result. The WB might be an option to avoid getting unspecific signal contribution. But the WB is also a LOD killer, if it is not well used (if the time is too long).

(The reference lines are in PVP and the substance is diluted in PBS)

Strong suspicion of a specific detection of the substance P with PMMA slides (CEA surface preparation – 2 mm spot size)

Figure 3.1a

0,00E+00

1,00E-03

2,00E-03

3,00E-03

4,00E-03

0 200 400 600 800 1 000 1 200

Slo

pe

sub

stan

ce P

on

ly (

V/s

)

Substance P concentration (ng/ml)

PMMA_wo washing buffer_2 concentrations of substance P inside PBS Spot size = 2mm x 2 mm (ML2-42-02) - g= 0,58 (Round PMT)

LOD = 35 ng/ml

LOD = 16 ng/ml

Page 43: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 43 / 107

No washing buffer has been used to get this result (Back to 2 mm spot size the specificity is back)

Specific detection of the substance P on PMMA (CEA surface preparation – 2 mm spot size) Figure 3.1b

To determine, why it is difficult to get the specificity of the substance P, it has been decided to stabilize the functionalized area process (at least from a the initial fluorescence point of view, i.e. the initial fluorescence of tripods).

1,35

1,4

1,45

1,5

1,55

1,6

1,65

1,7

0 200 400 600 800 1000 1200 1400

Plot 2

C(SP = 1000 ng/ml (inside PBS)

H85A

PBS Only

0,95

1

1,05

1,1

1,15

1,2

0 200 400 600 800 1000

H85B

plot2_H85B

C(SP = 200 ng/ml) inside PBS

PBS only

Page 44: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 44 / 107

3.2 Data show specific detection on PMMA without any ambiguity on stabilized process Recently, the specific detection of the substance P has been seen on a stabilized process. Unfortunately, the detection level is high (i.e. between 1 ug/ml and 3 ug/ml in 5 minutes). It typically means a concentration 4000 times higher than the targeted detection limit (i.e. 0.5 ng/ml in 5 minutes), based on a detection of 2 ug/ml. For the concentrations of 3 ug/ml and 10 ug/ml, specific detections have been measured. For 1 ug/ml, no specific detection has been seen. During these experiments, regeneration with the antibodies is necessary after 30 or 40 minutes to reuse the same slide for the detection (relatively short life time of the slide). Variations around this stabilized process are under progress to keep the specificity with a lower LOD. 3.3 Fluidic protocol Many problems have been solved, especially the preparation of the fluidic circuit (like its saturation with BSA= Bovine Serum albumin). Thanks to the BSA saturation, the signals are not erratic (see the annex 16). The current difficulty is now to get specific signal linked to the substance P with a functionalization on PMMA substrate and to guarantee that this signal is specific to the substance P. The fluidic protocol, which is described below, is studied in details to determine the relevance of each step. Preliminary and compulsory step: Fluidic circuit saturation with “EIA” which contains BSA (Bovine serum albumin) Each night, about 48 ml of buffer solution with BSA inside circulates during 8 hours (flow rate = 0.1 ml/minute). The BSA “sticks” on the tube and on the fluidic chamber to prevent nonspecific adsorption on the tube. This point work well and that stage it is considered as stabilized (see annex 16) Before each hybridization

a) PVP buffer, 1 ml/min, 300 seconds

b) AB (diluted in PBS) in static mode (typical time : ~ 50 minutes inside the fluidic chamber, slightly more for PMMA) The typical volume of our fluidic chamber is about 150 microliters. 300 microliters (> 150 microliters, to be sure that the AB are inside the fluidic chamber) of AB (concentration

= 50 g/ml) are sent inside the chamber. The AB concentration has to be adapted to get the best LOD (i.e. 50 ug/ml is better than 25 ug/ml for instance) The fluorescence (due to the Rhodamine 6G) of each spot, where the fluorescent tripods are immobilized, is quenched by the “red dye”, which is “gripped on the AB”. This quenching occurs when the AB is caught by the tripods. This step is finished when the fluorescence decrease becomes insignificant (typically after 50 minutes/3000 seconds) (see figure 2.3a and 2.3c).

c) PVP buffer, 1ml/min, 300 seconds

It is important to frame, inside the tubes, the (small) volume of AB by PVP (and not by PBS to minimize the diffusion, i.e. the mixing of the antibodies inside the liquid in front and behind the antibodies). Since the antibodies are diluted in PBS, this liquid should not to frame the antibodies. [in the future, electrovalves will be used]

Page 45: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 45 / 107

d) Washing buffer, 1 ml/min, 45 seconds

We work on this step to find the suitable time (or buffer), which contribute to remove the unspecific adsorption and the antibodies which are located between the functionalized areas. But, in the meantime the specific absorption to get the best LOD has to be preserved (i.e. the washing buffer also removes the specific antibodies which are on the functionalized areas) If a WB step is necessary, the time should be between 0 and 120 seconds.

e) Reference line in PBS (≈ 300 seconds)

f) Substance P diluted in PBS at given concentration according to the tests (Most of the time during 300 seconds) It is the signal line Etc… Next To regenerate the tripod, go to b) To detect again the substance P without regeneration go to e)

Before each new slide, a washing buffer is used during 20 minutes with a 1ml/min flow rate. The buffers should be modified to improve the LOD on PMMA ?

wxwxwxwxw A data processing is proposed to compare the experiments together, especially when they are achieved with the same substance P concentration. This processing leads to a LOD (see annex 10). This methodology is probably too severe. The “real LOD” will be experimentally obtained in discriminating a solution without substance P (i.e. the reference line) and with the smallest substance P concentration, which 2 different significant signals (i.e. 2 different significant slopes). 4. Potential instrumentation improvements concerning the LOD (Limit of Detection) It is difficult to quantify the potential improvements on the LOD, while the functionnlization process is not stabilized. The potential improvements could be around: -The flow rate, which contributes to the lowest LOD. During the tests, the flow rate of 1 ml/min is used. We do not anticipate a large improvement with another flow rate (but it has to be evaluated). -The spot size of the functionalized area (and a similar light size) From an instrumental optical point of view, “larger is better” for the LOD. Unfortunately, it could be different from a hybridization point of view. From an optical point view, P2 can generate various circular spot diameters. -Slight adaptation concerning the optical filters -LOD versus optical power More incident power is always better (if the functionalized areas are not damaged as it could be, see figure 4.1) But, if we want to use OLED or another excitation module with less power, it should be possible with the configuration with the circular PMT. -Suitable lightning (time for instance), suitable detection conditions (time, averaging, sample frequency, gain…).

Page 46: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 46 / 107

The damage result from a combination between optical power, spot size, irradiation time, flow rate, fluid temperature,

thermal resistivity, thermal PMMA conductivity and substrate thickness and initial fluorescence level. This point has to be experimentally verified.

Figure 4.1: Functionalized areas are damaged due to an optical incident power which is too high

Option: Discussion to collect more fluorescent light -Discussions and calculations are in progress with Polyscale to find a geometry to collect more fluorescent light

Detection geometry improvement

Discussions for simulation are in progress with Polyscale to improve the collection efficiency: geometry, shape of PMMA fluorescent guide, various reflecting technics, choice of PMT sensor (size, efficiency, etc…) See annex 6

-Similar objective with CETEMMSA in coating the slide edge to collect more fluorescent light

wxwxwxwxw Probably, the best and simple solution to collect more light is: -To have large functionalized areas (if it is compatible with the hybridization and the fluidic protocol). An excitation module similar to the Polyscale module allows using “a rectangular long shape” for instance. - To use a thick PMMA substrate (500 microns is better than 200 microns) see Annex 18 - To locate the functionalized area as close as possible from the output side (an increase of a factor 2 or 3 has been measured in using a solution of Rhodamine 6G)

xwxwxwxw

The « dark blue » circular zones are damaged areas due

to a too high optical incident power

Page 47: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 47 / 107

5. Conclusions concerning P2 and future work Conclusions Platform P2 to achieve functional tests and to receive various R2R modules It works Platform P2 ready to receive a R2R fluidic chamber from IPT in PMMA ready Platform P2 to test LGP, pump, valve, etc… ready Protocol to detect the substance P on PMMA OK, it works but… it might be better Difficulties concerning -the specificity, the targeted LOD, -the stabilization of the functionalization process and the fluidic protocol (close link) Important progresses: -Tube preparation to detect the substance P -Reduction of the antibodies volume to prepare the functionalized typically from factor 20 (from 6 ml to 0.3 ml) cost reduction. It becomes crucial, with a low antibody volume, to maintain the antibodies between a liquid, which minimizes the antibody diffusion inside this liquid, to prepare the detection of the substance P. Possibility to adapt the protocol with Dolomite valves -Possibility to use the antibodies without any electrovalve -Optical excitation setup has been adapted to generate various excitation spot sizes -Optical detection has been adapted for many configurations. The configuration with the circular PMT associated with a fiber bundle is interesting to use lower excitation power. It is good news for a potential R2R excitation module (LGP, OLED, etc…) -Smaller electrovalve and smaller pump have been (successfully) tested and evaluated they might suitable for pick and place It has to be confirmed that Bartels pump work with saline buffer -Height of a suitable fluidic channel : ≈ 300 microns (Higher height increase the probability to generate large air bubbles) Exchange with the ML2 partners During the last months, it has been possible to give more accurate specifications to the ML2 partners and to exchange with them: IPT is in on the way to develop a fluidic chamber in R2R compatible with the test platform P2 (see Annex 11). IPT OLED : the maximum accessible optical power on 5 mm x 5 mm is typically between 2.0 mW (or 0.35 mW for 2 mm x 2 mm spot size) In using the circular PMT configuration, it could make sense The long term stability of the OLED concerning the optical power has to be studied. Various exchanges between Polyscale and CEA : -concerning an excitation module with a LGP (This module should be R2R compatible and compatible with P2 for the tests), See Annex 12 - concerning the geometry to collect more fluorescent light, see Annex 15 Various Discussions between IPT/PLS and CEA Meeting at Aachen between these 3 partners (25

th/26

th June 2014)

Page 48: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 48 / 107

-Discussions and samples provided by IQS with a dedicated preparation for the functionalization -Discussions with CETEMMSA (see annex 19) - PMMA and various plastic substrates supplied by MRS, Polyscale, VTT and IPT

Future work R2R component tests We are waiting for : -The IPT chamber in PMMA, -The Polyscale excitation module. Pick and place (Dolomite valve and Bartels pum) If the PBS tests with the Bartels pump are satisfying, this pump should be installed on P2. Smaller LOD It is possible to specifically detect 3 ug/ml of substance P with P2 with a 1 mm PMMA slide. Is it possible to get a better result ? The fluidic protocol and the experimental test bench have been adapted to get a specific detection for the substance P with functionalized areas of 2 mm in diameter. The functionalization process and the fluidic protocol have to be stabilized. It is essential to reach this stabilization before analyzing without ambiguity the role of various instrumental parameters (optical filters, flow rate, LOD versus optical power,..). For the spot size of 5 mm in diameter, at this stage, any specific detection cannot guarantee. Our main objective is to improve the specific adsorption of the antibodies (AB) towards the substance P and/or to remove the nonspecific adsorption (Firstly, for the 2 mm x 2 mm spot size). The optimum fluidic protocol for the detection of substance P on PMMA might require other buffers or/and other sequences. Then, it will be nice to replace the PBS by various kinds of water. In spite of these difficulties the demonstrator 2 is operational for functional tests. To get specific substance P detection, the following parameters are necessary: -The antibodies have to be on the functionalized area (and if possible in higher concentration than everywhere and if possible only on the functionalized areas), -The substance P concentration has to enter in the chamber, -The capture coefficient of the substance P by the antibodies has to be as high as possible, -The slope of the reference line (i.e. the PBS line) has to be as low as possible. During the last 2 years, about 100 slides have been tested on the 2 platforms available at CEA (for the substance P, application 1) and about 500 slides have been functionalized and characterized for this application outside the platform.

Page 49: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 49 / 107

Annex 1 : General views of P2

This platform has been built to be versatile: to test various configurations and protocols, to test various (R2R) components. This platform can be adapted: - to answer to various questions, particularly the experiments described in the “preliminary experiments” sections of previous reports.

Figure A-1: Platform 2 (P2)

Annex 2 : Environmental applications and reagents

For the demonstrator 2, 2 environmental applications have been selected (It could be something else). -the first one is the continuous detection of microcystin inside “clear, filtered” water (Application 1), [The substance P mimics this application at lower cost, see next page] -the second one is the dosage of EE2 inside sample water (i.e. static mode, Application 2). CEA has absolutely no background concerning EE2. Therefore, it has been decided to begin with the application 1. The first tests concerning the EE2 began in April 2014. They are currently achieved outside the platforms to find a convenient protocol, which will be adapted on P2.

The reagents for the application 1 are bought from Academic French partners:

- For the antibodies, the partner is a CEA lab at Saclay (near Paris), - For the tripods (i.e. the probes), the partner is CNRS lab at Rouen.

CEA decided to search a “commercial solution “for the reagents to dose EE2. Nevertheless, one product has been specifically developed by an industrial company for this application. At that stage, LETI did not detect the EE2 outside the platform with a glass slide or a plastic slide.

Page 50: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 50 / 107

Reagent cost The antibodies are not used during the R2R process. At the beginning of ML2, about 6 mL of antibodies (concentration 25 ng/ml-glass slide process) were necessary for a continuous detection of substance P. In adapting, the fluidic protocol, this volume has been divided by 20 per detection, with a fluidic height of 300 microns (namely 300 microliters for a hybridization chamber of ≈ 150 microliters). At that stage, for fluidic height of 90 microns, 6 mL are absolutely necessary for the detection. And with such fluidic height, any detection has been achieved yet. Therefore, it seems reasonable to build R2R fluidic chamber with a height of 300 microns.

Product to be

detected Antibodies (AB) +

immobilized probes (P) Volume/concent.

Price per tests Dye

Application 1 Continuous mode

Substance P

Not commercial AB = 10 kE (10 mg) Delay : 2 or 3 months Probes = P = 6 KE or 14 kE (1mg = 500 uL à 1 mM) Delay : 3 months

Forecasted delivery at each order : 10 ml at 1mg/ml = 10 mg 1 Experiment : AB** = 0.3 ml at 50 ug/ml (plastic substrate) = 15 ug ≈ 600 experiments (glass slide 0.3 ml at 25 ug/ml, h ~ 260 um/300 um) P*** = 50 uL at 0.1 mM 100 different spotting (at the lab, 10 slides are possible per spotting) At the lab level on PMMA substrate AB : about 20 Euros per experiment (instead of 120 Euros previously) P : about 15 Euros per spotting (for 10 slides simultaneously)

Rhodamine 6G (524 nm/546 nm)

Microcystin Not commercial AB = 12 kE P = 45 kE Delay : 6 months

Forecasted delivery * : The delivery volume might smaller since it is difficult to predict the AB production and the AB purification yield (animal production) ** : the AB concentration is at least multiplied by 2 (PMMA versus glass) *** : the volume (50 uL) is mainly due to the dead volume (of spotter) Current functionalization process : 20 drops x 10 drops x 10 drops x 3 spots x 330 nL per drop = 2 uL per slide (minimum volume without taking into account the dead volume)

Figure A2-1 (Application 1 : Microcystin and substance P)

Page 51: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 51 / 107

Product to be detected

Products Delay

Volume concentration

Dye

Application 2 Dosage

EE2

Noncommercial product (custom product developed by a private company) EE2 + protein (BSA) : 9 kE Delay : 4 months (for the 1

st time ?)

Commercial products Ab59670 sheep polyclonal to EE2 50 uL at 12 mg/ml = 370 Euros ≈ 200 mL at 3 ug/ml Delay : 1 month Donkey polyclonal secondary AB to sheep IgG-H&L (Dylight ® 488) 500 ug = 180 Euros ≈ 200 mL at 2.5 ug/ml Delay : 1 month

Unknown at this stage

≈ 100 tests ?

≈ 100 tests ?

IPT asks for a lower excitation wavelength

Dylight 488 nm is

proposed and will be tested

Figure A2-2 : Application 2 (EE2)

Figure A2-3: EE2 : Detection structure

ab59670 sheep (AB which detects the EE2)

Secondary Antibody to Sheep IgG - H&L (DyLight® 488) (AB which detects the sheep AB)

EE2 + protein : 9 kE Combination specially developed by a

company for LETI

EE2

Page 52: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 52 / 107

Remark Concerning the volume of substance P probes, the current dead volume with ink jet printing at the Durst “lab level” is about:

- 100 mL for the preliminary tests It means 25 times more than the material, currently used at the LETI for 100 different spottings i.e. between 25 x 6 = 150 kE and 25 x 14 = 350 kE for only one R2R test (The price is not linear since the price is mainly due to the manpower) It is inconceivable to try such test(s)

- 1000 mL for industrial tests No additional comment.

It is the reason why CEA tries to develop a test based on oligonucleotide recognition to immobilize probes on PMMA with an inkjet printing at Durst lab. This test is not operational since CEA was not able to find a large commercial protein to enhance the coupling with the PMMA.

Probe : CY3 oligonucleotide (Fluorescent at 530 nm like the substance P tripods) Target : CY5 oligonucleotide they are caught by the CY3 oligonucleotide. They quench the CY3 fluorescence signal. A fluorescent decrease is observed. The CY5 oligonucleotide plays the AB role.

Figure A2-4: Oligonucleotide tests

Page 53: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 53 / 107

Annex 3 : Current excitation module (not R2R compatible)

The excitation module inside P1 typically emits ≈ 1.4 mW/mm

2 at ≈ 530 nm on a 2 mm diameter circular spot size

(i.e. measured after the excitation filter directly on the functionalized area). Various adaptations shave been tested. With a 5 mm spot size and a similar optical power density, a theoretical ratio on the LOD is expected, namely : [(5/2)

2 ]1/2

= 5/2 = 2.5. it is the reason why we would like to use 5 mm spot size. Unfortunately, we did not get yet specificity with 5 mm functionalized areas. Work is under progress to improve the specificity for the substance P detection with the 2 mm spot size before using the 5 mm spot size. Whatever the spot size, 4 solutions are possible for the excitation module and 3 seems R2R compatible.

1-Current LED solution (CEA) This solution works but this solution is not R2R compatible. This solution includes 4 LEDs (one for each spot). For each spot, 2 ball lenses are located in front of the 4 LEDs (i.e. for a total of 8 ball lenses). An excitation filter (only one excitation for the 4 LEDs) is located between the ball lenses. The 4 LEDs lie on an aluminum heat dissipator to stabilize the LED temperature to “always” have similar output characteristics. This point is crucial to get the best LOD. It has been improved on P2.

Figure A3-1: Current LED solution on P2

On P2, the excitation wavelength peak has been slightly shifted to shorter wavelength (i.e. from 530 nm to 515 nm) to have a better chance to remove the excitation filter (a stabilized protocol is needed to validate or invalidate this point). The size and the material of the ball lenses have been chosen to roughly get 40 mW after the excitation filter with a 5 mm spot size (i.e. slightly more than 1.5 mW/mm

2) on the functionalized areas.

(on P1, spot size diameter = 2 mm, incident power density ≈ 1.4 mW/mm2).

Why the excitation wavelength peak has been shifted to shorter wavelength? To minimize the impact of the stray light if the excitation filter is removed (to simplify the device, if possible in keeping the same LOD). For the Rhodamine 6G, a typical ratio between the absorption at 530 nm and 515 nm is: A(530 nm)/ A(515 nm) ≈ 1.00/ 0.63 ≈ 1.58 (diluted in methanol) 1.00/ 0.58 ≈ 1.72 (diluted in ethanol) (Order of magnitude since various parameters have to be taken into account like pH effect, temperature effect, ..)

Page 54: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 54 / 107

Components References Price (Euros)

4 LEDs APG2C3 530 nm

Roithner

6.75 X4 = 27.00

8 ball lenses 8.0 mm Dia S-LAH79 Ball lens

Edmund scientific (Item 47131)

69.82 x 8 = 446.88

1 Excitation filter XF1074

525AF45/25R Omega

Laser components

170 X 1 = 170.00

1 heat dissipator 29 Euros

3 plastic pieces in 3D printing Achieve by a local company according to CEA drawings

~ 150 Euros

Total for one excitation module (prototype)

~ 850 Euros

Figure A3-2: Price of the excitation module LED on P2 (Number =1)

As expected, the LOD is higher with an excitation around 515 nm instead of 530 nm (about a factor between 1.5 and 4). On P2, the LEDs are driven individually by a current driver to improve the excitation light stability. This point has an impact on the LOD. Key points: - Power density: at least > 1.5 mW/mm

2 (the highest is the best if the probes, the AB, the dyes… are not damaged

by the light) - Spot size : at least 5 mm in diameter size (the largest is the best) need to be validated during the hybridization. On the current setup, light spot sizes are circular but the functionalized areas are square (5 mm x 5 mm).

A square light should slightly improve the LOD by a theoretical factor of: 1.12 ≈ (4/ -The power density and of the excitation spectrum stabilities, -The coincidence between the excitation light and the functionalized area. Concerning, the excitation filter : Until June 2014, LETI did not find a pass-band or a pass-low colored filter in this wavelength range. From June 2014, various colored filters, have to be tested been ordered at Thorlabs Company FGB 39 (Ø25 mm BG39 Colored Glass Bandpass Filter, 360 - 580 nm.) FGB7 (Ø25 mm BG7 Colored Glass Bandpass Filter, 435 - 500 nm) Unfortunately, these filters did not match with our excitation window (ie around 515 nm)

Page 55: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 55 / 107

Annex 4 : R2R Excitation module LGP solutions proposed by Polyscale

Solution 1: Light Guide Plate = LGP (S. Hamm) Polyscale proposes to replace the current LED solution by the solution described A4-1. IPT and CEA discussed to define an excitation module compatible with the functional spot size currently used on P2. Tests are in progress (see Annex 12)

Figure A4-1: Drawing of an excitation module with a light guide plate

Polyscale has already achieved the first version of this excitation module (see figure A4-2 and annex 12).

Figure A4-2: Polyscale achievement of the excitation module with a Light Guide Plate (LGP)

CEA has in mind: - to optically characterize this module (see Annex 12), - to simulate the optical power of this source with the current LED solution (if it is possible) and to measure the associated LOD (effect of the incident power on the LOD), - to achieve hybridization with the “R2R excitation module” (if the mechanics and electronics are compatible with P2). This excitation module can be tested with or without excitation filter (at least an excitation filter in front of one spot). Polyscale and CEA have designed a mechanical piece to withstand the Polyscale excitation module (see Figure A4-3).

First test results are reported in the annex 12. For new tests, Polyscale has ordered a new PCB for the LEDs to compatible with the electronic of P2.

Page 56: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 56 / 107

Figure A4-3: Mechanical interface for the Polyscale excitation module (with P2)

Solution 2: without light guide plate (S. Hamm) Polyscale proposes another R2R compatible solution.

Figure A4-4: Drawing of an excitation module with a BEF and without a light guide plate (drawing)

LETI is waiting for the BEF.

Page 57: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 57 / 107

Annex 5 : Excitation module OLED solution proposed by IPT

In February 2014, IPT send to LETI an OLED block with 4 OLED linked by a common cathode. Each OLED needs to have its own cathode to individually be driven by the suitable current (i.e. to deliver the same optical power on each functionalized area).

Figure A5-1: OLED block supplied by IPT

After gluing electrical contacts on the OLED track, CEA achieves: - the measurement of the peak wavelength, which is close to 515 nm (similar result at IPT and CEA). - the total light power density measurement versus the current through the OLED (see figure A5-2). The power light density linearly increases with the current. The tests have been achieved until 18.5 mA and extrapolated at 40 mA : 0.031 mW/mm

2

IPT told us that, this device can be used until 80 mA, it means 0.062 mW/mm2 (0.032 mW/mm

2 x 22 ~ 1.3 mW)

Therefore, a factor 25 is missing, if an excitation filter is not necessary (or factor 50 if an optical filter is necessary) to reach the current power on P2. According to IPT, the OLED lifetime decreases with a current increase.

Additional tests at CEA : Performance damage tests at CEA : maximum current around 120 mA It leads to a maximum optical power at the damage current: ≈ 2.0 mW (5 mm x 4 mm) unfiltered Compare to ≈ 40 mW with the current filtered CEA LED on P2 (1.5 mW/mm

2 x 25 ≈ 40 mW)

CEA will try to study the impact of the optical power level on the LOD, when the functionalization process will be stabilized, but with the circular PMT which collect 100 times more light, it should possible to decrease the optical power of the excitation.

Page 58: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 58 / 107

During the next months, IPT will gather the data to determine what could be the maximum power compatible with OLED achieved with a R2R technics.

Figure A5-2: OLED characterization at LETI

dP = 7,69E-04 x I R² = 9,97E-01

0

0,002

0,004

0,006

0,008

0,01

0,012

0,014

0,016

0 5 10 15 20

Tota

l po

we

r d

esn

ity

(wit

ho

ut

filt

er)

Current (mA)

Measured data at LETI : I = 18,5 mA ==> 0,015 mW/mm2

Data got with a voltage command A current driver will be better to avoid the current drift due to the heat

Page 59: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 59 / 107

Annex 6 : Detection module (PMT)

1-Sensor/sensor configuration Detections are achieved with photomultiplier tubes (PMT).

The LOD with a PMT is few fW. For usual silicon photodiodes, LOD is few W (see for instance the presentation of Mario Kasahara from Hamamatsu entitled “detector selection guide (http://www.avsusergroups.org/pag_pdfs/pag2009_8kasahara.pdf). The LOD for the organic photodiodes, which are R2R compatible, is similar or higher than the usual silicon photodiode. Our optimistic calculations lead to few pW which reach the PMT? Therefore, to replace the PMT by an organic R2R photodiode, the target concentration has to be multiplied by 1E6. To get a lower LOD, CEA replace the rectangular PMT used by a circular PMT + a fiber bundle (i.e. a taper which transforms a rectangle array in circular array). The rectangular array is located in front of the rectangular side of the slide and the circular array is located in front of the circular PMT (see Figure A6-1).

One side is circular in order to fit with a circular PMT. The other side is rectangular to fit with the shape of the slide output

Figure A6-1: 2 Tapers (fiber bundle) achieved according to the LETI specifications

Rectangular configuration Price (Euros)

Circular configuration

Price (Euros)

Size and weight Anode sensitivity

H 11462/8289 (rectangular PMT)

~ 1300 E 95 x 50 x 38 mm3

225 g Typ. 740 V/nW

H 10722-20 (circular PMT)

1300 E 60 x 22 x 22 mm3

100 g

Typ. 150 V/nW (about 5 times lower)

Taper Between 1000 and 1800 E according the taper NA

Diameter : 33 mm Length : 100 mm

Total 1300 Euros Between 2300 and 3100 E

Figure A6-2: Prices of the circular and rectangular PMT configurations

In contact with a circular PMT: Ø = 8 mm, 50 mm

2

In contact with the rectangular side: 28 mm x 1,8 mm = 50 mm

2

(It might be adapted to the final size of the slide)

Page 60: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 60 / 107

The LOD is slightly better with a circular PMT setup (about a factor 2). As expected, the taper with the higher NA leads to a smaller LOD (better result with NA = 0.87 versus NA = 0.64). The collected light with the circular configuration (i.e. thanks to the taper) is estimated to be multiplied by about 150. Therefore, with this configuration, it is not possible to replace a PMT by a photodiode but it should be possible to use lower optical power for the excitation module (OLED, LGP,…) The setup has been modified to use the same electrical power supply for the circular PMT (5V) and the rectangular PMT (15 V): NV175 (option = NV1-4F5FY5H-F) from TDK.

Hamamatsu sells MicroPMT module with 3 x 1 mm

2 effective area (see figure A6-5). A PMMA slide

(25 mm x 0.2 mm) leads to an output section of 5 mm2

, which is not far from this area. Unfortunately, the sensitivity of this PMT is typically 10 times lower than the sensitivity of a circular PMT. A rectangular PMT geometry is not the best way to collect more light.

Figure A6-5: New micro PMT module (PMT + high voltage source)

With the help of Polyscale, calculations are in progress to see if a suitable geometry can be found to get a smaller LOD (see Annex 15).

Figure A6-4: Rectangular PMT H8289-01

Figure A6-3: Circular PMT [H10722-20] + taper (Pieces to link the PMT and the taper are

achieved in 3D printing)

8 mm x 38 mm x 15 mm

Page 61: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 61 / 107

Annex 7 : Emission filters for the substance P The nominal combination of the emission filters is: -XF3083 + OG570 -XF 3083 = interference filter -OG570 = Colored filter At that stage, it is difficult to see if an OG570 or an OG590 leads to a better LOD. But it seems that an interferential filter or at least a cut-off for the long wavelength is necessary due to the “red” fluorescence of the antibodies (see the Impact of the emission interference filter removal on figure A7)

With the 2 OG 590, there is no cut-off for the longer wavelength (red florescence of the AB dye can detected) Parasite fluorescent signals have been detected. They are probably due to the “red fluorescence” of the dye, which is linked to the AB OG570 and OG 590 are long pass filters. XF3083 is an interference pass band filter With 5 mm spot size, “stray” signal have been clearly linked to a filter change (probably due to the “red” dye, which is on AB)

2x OG 590 instead of 1 x OG570 + XF3083 (Interferential filter)

Figure A7

0

0,5

1

1,5

2

2,5

3

3,5

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000

Flu

ore

scn

ce le

vel (

a.u

in V

olt

s)

Times (seconds)

H78A

Page 62: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 62 / 107

Annex 8 : Plastic selection based on the autofluorescence criterion for the demonstrator 2 (PMMA) It has been decided to choose the material with the lowest autofluorescence around 530 nm in testing the material with the current configuration of the demonstrator 2. The ML2 consortium has pre-selected 5 plastic materials: -PMMA -PET -COC/COP -PC -TPU According the measurements, the TPU and the PET were discarded for the demonstrator 2. PC and COC/COP might be selected but PMMA is the best substrate from an “autofluorescence demonstrator 2 point of view” (see the measurements below).

wxwxwxwxxw

Excitation wavelength ≈ 530 nm Detection: experimental conditions of the demonstrator 2 The figure A8-2 shows, that there is no correlation between the autofluorescence level and the PMMA substrate thickness. The roots of autofluorescence might due to the additive or/and the impurities inside the PMMA. There is probably an impact coming from the fabrication technologies and from the suppliers. Therefore, for the demonstrator 2, CEA decided to start the development of probe immobilization on PMMA and more accurately on PMMA with one millimeter thickness (supplied by IPT coming from Mitsubishi). This PMMA substrate is too thick for the R2R technology. The 1 mm PMMA substrates have the lowest autofluorescence signal and they are convenient to handle for the technological development step and for the first tests on P2. In April 2014, IPT told us that the PMMA thicknesses for the R2R technology will be 200 microns and 500 microns. In July 2014, IPT gave us 200 microns PMMA substrates. IPT did not receive yet the 500 microns PMMA.

Page 63: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 63 / 107

Without thickness normalization With thickness normalization

Autofluorescence (AF) of a microscope glass slide = 1

PET (AF~ 10) and TPU (AF~ 100) No

PC, COC Possible (AF between 2 and 8) The bests seems PMMA (AF ≈ 1+)

Trends of the autofluorescence tests on P1

Figure A8-1

1 10 100

5013 (1100 um)

5013 (1100 um)

8007 (1200 um)

6015 (110 um)

API (1000 um)

API (500 um)

MRS (500 um)

MRS (500 um)

PC (475 um)

PC (475 um)

PMMA (250 um)

PMMA (250 um)

PMMA (250 um)

PMMA (250 um)

PET (175 um)

PET (175 um)

CO

CTP

UP

CP

MM

AP

ET

1 10 100

5013 (1100 um)

5013 (1100 um)

8007 (1200 um)

6015 (110 um)

API (1000 um)

API (500 um)

MRS (500 um)

MRS (500 um)

PC (475 um)

PC (475 um)

PMMA (250 um)

PMMA (250 um)

PMMA (250 um)

PMMA (250 um)

PET (175 um)

PET (175 um)

CO

CTP

UP

CP

MM

AP

ET

Page 64: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 64 / 107

Among the tested substrates, the PMMA substrates have a lower autofluorescence Tests of PC and PMMA from various suppliers and thicknesses

Figure A8-2

Evaluation of 18 PMMA substrates from various suppliers and thicknesses (125um-1040um) Figure A8-3

1

10

100

0 500 1000 1500 2000 2500

Thickness in microns

Autofluorescence ==> PC (square) and PMMA (triangle)

PMMA raw data

PMMA with thicknessnormalisation

PC raw data

PC with thicknessnormalisation

scope glass slide Autofluo =1

Autofluorescence (a.u.)

PMMA

0,1

1

10

100

Au

tofl

uo

resc

en

ce

( =

1 f

or

a m

icro

scip

e gl

ass

slid

e)

Autofluorescence signals (excitation light around 532 nm, P1) of various PMMA films (from 125 um up to 1040 um)

min

max

average

Sample number

Page 65: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 65 / 107

Selection of PMMA from various suppliers and thicknesses (125 um-1040 um)

Figure A8-4

Concerning the autofluorescence of the plastic substrates, the same trends are observed in the literature: (The autofluorescence of plastic materials and chips measured under laser irradiation, Aigars Piriska, lab on Chip, 2005, 5, 1348-1354 [1]) In general,

-the autofluorescence increases with adecrease, -Borofloat has the lower autofluorescence, -the ratio with PC/PMMA/COC is typically between 2 and 5 (Figure A8-5)

Figure A8-5: Data extracted from [1]

0,5

5

Au

tofl

uo

resc

ence

sig

nal

(=

1 f

or

a m

icro

scio

pe

glas

s sl

ide)

Autofluorescence signals (excitation light around 532 nm, P1) of selected PMMA films (from 125 um up to 1040 um)

min

max

average

~ 3 to 5 times AutofluoG

Autofluorescence microscope glass slide = AutofluoG

~ 2 times AutofluoG

~ 0,8 AutofluoG

~ 1,1 AutofluoG

Page 66: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 66 / 107

Annex 9 : Surface preparation of the PMMA slide

2 possibilities are studied to prepare the PMMA slide: one method from IQS and one from CEA The IQS have one drawback: the tripods have to be deposited “as soon as possible” after the surface preparation. The “as soon as possible” have not been quantified. We have not this difficulty with CEA process since all the steps are achieved at CEA. So far, it seems easier to use the CEA preparation. The CEA work concerning the PMMA functionalization is gathered in a memo entitled: “: Report on roll-to-roll compatible chemical functionalization” A stabilized process (functionalization process and fluidic protocol) is compulsory to analyze without ambiguity the effect of various instrumentation parameters (i.e. before achieving specific experiments concerning the instrumentation). Therefore, at this stage, it is impossible to efficiently work on the optical filters, to adapt the flow rate, etc… since a stabilization process has not been reached yet. Nevertheless, the current configuration with “Round PMT” leads to better result than the configuration with “Rectangular PMT”, optically, the 5 mm spot size is better than the 2 mm spot size. This last point has to be validated during the hybridizations. A fluidic protocol, which uses a “suitable” volume (a volume which is slightly higher than the volume chamber) of antibodies has been validated (i.e. typically 300 microliters instead of 6 ml).

Page 67: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 67 / 107

Annex 10 : How the LOD is evaluated and extrapolated if necessary?

The methodology, which is presented, is questionable. But this methodology has been applied to compare the experiments together. The goal is to have an objective and quantitative criterion to determine, what are the best conditions (fluidic protocol, optical configuration, functionalization method, optical parameters…), which lead towards the smallest quantification limit (QL) but above all towards the smallest detection limit (DL). DL is smaller than QL (i.e. DL < QL). The quantification limits are evaluated during 300 seconds (i.e. for a window of 300 seconds) and the tests are typically achieved during 300 seconds. 1-First of all, the slope of the substance P (SP) is evaluated in V/s in fitting the experimental data a by linear line.

The fitted data are subtracted to the experimental data. Then, the standard deviation of this result is evaluated (P). 2-If it is possible, the slope of the buffer (SB) is evaluated in V/s in fitting the experimental data by a linear line. If this

slope is negative, SB is equal to 0 (SB =0) and B = 0.

If SB > 0, the standard deviation of SB is evaluated (B). These data are obtained on on a 300 seconds window: - For the substance P - For the buffer before the substance P (i.e the reference line) (The best case is to have the substance P diluted in the same buffer than the buffer used for the reference line)

How to evaluate the standard deviation of the difference between the 2 slopes (substance P and buffer) ?

If the 2 noises are uncorrelated: uncorrelated t = (B2 + P

2)1/2

If the 2 noises are correlated: correlated t = (B + P) > (B2 + P

2)1/2

The worst case will be assumed: the noises are correlated. The symbol for the total standard deviation is t Therefore, the quantification limit (QL) is given by: The time tQ , which is necessary to quantify the test concentration C test (i.e. the concentration used during the tests, which usually last 300 seconds) is given by:

tQ = 3 t /(Sp - SB) This time will be converted in a concentration, which represents the minimum concentration which is able to quantify in 300 seconds, according this (questionable) formula:

CQL = tQ x Ctest / ttest = 3 t /(Sp - SB) x Ctest / ttest (1) It is correct to compare various CQL when they are obtained with the same ttest and the Ctest. We have many tests where ttest = 300 seconds and Ctest = 1000 ng/ml. For instance for ttest = 300 seconds and Ctest = 1000 ng/ml, if tQ = 30 seconds, it leads to CQL =100 ng/ml When CQL is ≈ 0.1 Ctests, Ctests have to be decreased. From the formula (1), 3 CQL can be evaluated:

-CQL (SP) = 3 P/(Sp) x Ctest / ttest

only the substance P parameters are taken into account. The “buffer line” is “perfect” (SB = 0, B = 0)

-CQL (SP, B) = 3 P/(Sp -SB) x Ctest / ttest

The buffer line is almost perfect (SB , B = 0)

-CQL (real) = 3 t/(Sp -SB) x Ctest / ttest

In taking into account the buffer parameters (SB, B) The calculated concentrations are independent of the PMT gain (ratio between the standard deviation and the slope, which are obtained from data store with the same PMT gain). (In this document LOD = CQL (real))

Page 68: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

REPORT

Laboratoire d’électronique et de technologie de l’information

Commissariat à l’énergie atomique et aux énergies alternatives Direction de la recherche technologique

MINATEC Campus | 17 rue des Martyrs | 38054 Grenoble Cedex 9

T. | F.

[email protected] Page : 68 / 107

Établissement public à caractère industriel et commercial RCS Paris B 775 685 019

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization

Annex 11: Suggestions for a disposable fluidic functionalized chip in R2R (to start discussion, especially with IPT) and discussion to transfer the fluidic chamber in R2R process (discussion with IPT during a meeting in June 2014)

The probes are sensitive to the light (photobleaching).

Page 69: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 69 / 107

Oxygen plasma (CEA) or corona ( IPT)

Probe ink jet

printing

Functionnalized

PMMA 1

Drilled PMMA 2

Double side adhesive tape, glue, insert ?

75 mm

25 mm

Channel height : at least > ≈ 300 microns Channel width : at least 6/7 mm (spot size up to 5 mm in diameter)

Thickness ?

Thickness (200/500 microns)?

PMMA 1

PMMA 2

Washing

Glue ?

Insert ?

Probes (in dark) Cutting

Page 70: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

REPORT

Laboratoire d’électronique et de technologie de l’information

Commissariat à l’énergie atomique et aux énergies alternatives Direction de la recherche technologique

MINATEC Campus | 17 rue des Martyrs | 38054 Grenoble Cedex 9

T. | F.

[email protected] Page : 70 / 107

Établissement public à caractère industriel et commercial RCS Paris B 775 685 019

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization

The approach A «cut and assembly» will be used for the R2R process to have a channel with a height of ~ 300 microns. The geometry G1 is selected.

From the meeting IPT-CEA (June 2014), CEA send to IPT the data necessary to insure the compatibility to test the future R2R chamber on P2 (fluidic and optical compatibility).

G1

G2

Page 71: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 71 / 107

CEA proposal (23/07/2014) CEA proposes the following design to maintain the PMMA fluidic chamber (see IGS files for the sizes).

PMMA optical guide (with the functionalized area) PMMA support : close to the holder

Figure A11-1: Picture of the system composed of PMMA fluidic chamber and its holder The holder is fixed on the translation stage. The translation stage is already on the current CEA test platform.

PMMA fluidic chamber (PMMA optical guide,

gasket, PMMA support)

Holder

Translation stage

Page 72: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 72 / 107

Figure A11-2 : Pictures of the final system (i.e. with R2R IPT chamber)

Final position the PMMA fluidic chamber (The edge of the PMMA optical guide is in contact with the PM, the functionalized areas on the PMMA optical guide are in front of the light sources delivered by the excitation module).

PM

Excitation module

Fluidic connector in this area

Mounting of the PMMA fluidic chamber

Page 73: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 73 / 107

PMMA fluidic chamber PMMA fluidic chamber elements: -PMMA optical guide: This element is in contact with the PM. This element is functionalized -PMMA support, -Gasket.

The right of the PMMA fluidic chamber is free to set up the fluidic connector.

Figure A11-3 (see the Addendum for the compatibility with the Polyscale excitation source) Important : The PMMA support is larger than the PMMA optical guide in order to maintain the PMMA fluidic chamber without deforming the fluidic channel (see the drawing for more details) To be discussed: According to the fluidic connector, it might be necessary to modify the holder to support a heavy connector compare to the rest of the PMMA chamber.

Important considerations to take into account : -The weight of the fluidic connector -The alignment of the PMMA optical guide with its PMMA support during the sticking -Dolomite connectors minimum thickness = 4 mm ? (total chamber thickness ≈ 700 microns) Addendum : Compatibility of the future IPT R2R chamber with the Polyscale excitation module It could have an impact concerning the fluidic connectors Suggestions : -Longer fluidic channel -Fluidic connectors at the opposite side of the excitation module -Something else,…

PMMA optical guide in contact with PM

Location of the fluidic connector (If necessary, the fluidic channel lengths can be extended)

PMMA support

Area used to fix the PMMA fluidic chamber

Page 74: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 74 / 107

Figure A11-4 Figure A11-5

Figure A11-6

Mid-August 2014, IPT send a mechanical design, which takes into account the remarks inside this annex and seems compatible with P2. The mechanical design and the R2R chamber are in achievement at IPT. They will be sent to CEA for when they will be ready.

Page 75: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 75 / 107

Annex 12 : Discussion and results concerning the LGP (Light guide plate) from Polyscale

Excitation module

Result of the 1st characterization of the Polyscale excitation module with CEA platform 2 (P2)

Conclusion: To go further and to be fully compatible with the power supply of P2 for a given spot, the LED of the LGP have to be electrically plugged in parallel. This task is in progress at Polyscale. Nevertheless, few characterizations and observations have been done. They are reported below.

Various observations 1-At that stage, the following point is not a big deal there is some loose Height of the PLS light source (i.e. grating height) [to be in coincidence with the functionalized areas] Smaller altitude of the grating ≈ 44 mm Wished altitude: 42.5 mm (or 45 mm in the center) 2 options to correct this point : -With a new LGP modified the altitude of the grating -With this current grating modified “the metallic flat angle bracket” Drilling at another location Another possibility: to do another shelf bracket piece with a hole located 1.5 mm lower (i.e. hole at lower height) 2- Use this excitation module to detect the Rhodamine 6G or/and to follow an antibody decrease

Use the future excitation module to detect the Rhodamine 6G diluted in water or/and to follow an antibody decrease (CEA has already bought Rhodamine 6G, which has to be diluted in water)

Page 76: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 76 / 107

With our software, the LED current is driven with a current percentage -Green visible light appears for 17% -From 20%, the emitted light does not increase It correspond to a current between 30 mA and 35 mA Explanation: the driver, which is used on the test platform P2, can at least deliver 700 mA when the LED threshold voltage is 2.8 V (case of the usual LED module used by CEA). The maximum voltage is 5 V. Therefore, the remaining (5-2.8)V =2.2 V is able to deliver up to 700 mA. For the excitation Polyscale module: Each grating is irradiated by 2 LEDs’, which are plugged in series. Therefore, the LEDs in serial lead to a threshold voltage of 2 x 2.3 = 4.6 V The remaining 0.4 V only leads to a maximum current of 35 mA.

LED driven by the software

LED switch on

1 2 2 4 3 3 4 1

At this stage, this permutation is not a real problem.

Figure A12- 1 : LED control command

λPeak ≈ 498 nm (i.e. at I ≈ 35 mA)

Figure A12-2 : LED spectrum

0

10000

20000

30000

40000

50000

60000

70000

400 420 440 460 480 500 520 540 560 580 600

(u.a

)

Wavelengh (nm)

Spectrum Of LED's polyscale

Plot1

Plot2

Front view

Front view

LED number = LED which is switched on

Page 77: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 77 / 107

LED switch on

LED 1 (mW) LED 2 (mW) LED3 (mW) LED 4 (mW)

1 1.5 (1.1) 0.71 1.5 1.3 2 1.35 3 (2.5) 1.3 1 3 1 0.9 1.3 (1.3) 1 4 0.4 0.26 0.7 1.6 (1.6)

(In bracket, optical power with the excitation time inside the software. Other measurements: Continuous irradiation)

Figure A4-3 : Optical power (with the light sensor calibration at 515 nm) and cross talk evaluation

0

0,2

0,4

0,6

0,8

1

1,2

0 20 40 60 80 100 120 140 160

Vo

ltag

e (V

)

Time (s)

Fluorescence ( PMMA slide) G = 0,53 V

LED 1; Po = 20 %

LED 2; Po = 20 %

LED 3; Po = 20 %

LED 4; Po = 20 %

a) Warmup time : typically 15 seconds

b) The stability of the output signal is not so good with LGP LEDs compare to the LED which currently

equip P2 It could be due to the current value, which is low during these tests.

LED 1 LED 2 LED 3 LED 4

A= Average (V)_ G=0.53 V With the LGP

0.98 0.89 0,89 0,89

SD= Standard deviation (V) With LGP

0.1729 0.0484 0.238 0.0467

SD/A 17,7% 5,5% 2,7% 5,4%

SD/A with current CEA LED (spot size = 5 mm)

≈ 0.3%

Figure A12-3: Fluorescence data versus time with the Round PMT

(PMMA slide of 1 mm with a set point at 20% for each group of 2 LEDs)

Page 78: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 78 / 107

Jean (CEA) Stephan (Polyscale) [22/07/2014] 1) We have mechanically open the excitation module. We should be able to remove and change PCB, where the LEDs are located (cf. Figure A12-4) ==> Therefore, we are waiting for the 4 new PCB (2 in cyan, 2 in green) that you prepare.

Figure A12-4 : Mechanical support of the Polyscale excitation module (PCB for the LED : on the side)

2) One LED has been damaged (strange: only one LED although the same current should pass through the 2

LEDs) ==> collateral damage on one side of LGP and on the small wedge (see photo A12-5).(*)

One lateral side of the LGP + the small wedge is damaged resulting from a LED failure Figure A12-5 : Mechanical support of the Polyscale excitation module + Lateral view of the LGP

3) Next steps 3a) we will replace the PCB with new one (LED connected in parallel, 2 LED per gratings) 3b) we will test the PCB electronically and optically without the LGP (to avoid the LGP damage) 3c) if everything is OK, we will add the LGP for the optical characterization 3d) and then, if possible to see what is the behavior during an antibody decrease when the PLS excitation module is used (other possibility, to detect a liquid solution with Rhodamine 6G inside)

(*)-May be you can achieve another similar LGP since one side is slightly damaged due a LED failure -It appears

that the grating zone is 1.5 mm higher ==> it is not a big deal But if you do another grating you can translate them down to 1.5 mm (all of them).

Damage on one side of the LGP

Page 79: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 79 / 107

P.S. : For discussion: It seems that the cross talk between the functionalized areas might disturb the measurements (cf. photo A12-6).

Figure A12-6: More light in the grating areas but light everywhere an additional opaque mask with

suitable openings

The light insulation might be easier for you if the distance between the 4 functionalized areas is increased (2 gratings on the right are the same location and the 2 others close to the other side ==> symmetry center: center of the optical LGP). [If the gratings are well separated, we will have a drawback : 2 functionalized areas will far from the PMT sensor ==> unfortunately, they will lead to a higher LOD.] Therefore, with a larger distance between the functionalized areas, it will be easier for you to get insulation between the various functionalized areas, For instance in sharing the current piece with 4 gratings in 4 similar smaller pieces with one grating on each. You can have "black paint" around each piece (see discussion with CETEMMSA, Annex 19) At that stage, it appears that if the light is mainly directed towards the functionalized areas, the excitation light can be find everywhere (cf. Figure A12-6). May be smaller LGP for each guide might be the solution to irradiate the suitable zone. Other solution: A dark mask with openings only in front of the gratings (see Annex 19).

Page 80: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 80 / 107

Annex 13 : Pump(s)

P2 is currently equipped with “large” peristaltic pumps (see A13-1). CEA tries to find smaller pumps since ML2 partners do not propose R2R pumps. These small pumps might be pick and place.

Current peristaltic pump on P2 Figure A13-1

A13-1 Dolomite piezoelectric pump The Dolomite proposes piezoelectric pumps (Figure A13-2), which are smaller than the current CEA peristaltic pumps. They have been tested and discarded for the demonstrator 2 due few drawbacks.

Figure A13-2: Dolomite piezoelectric pump

ADVANTAGES VERY SMALL (ALSO for the power supply plug and play) R2R : possible in pick and place The flow rate is compatible with the demonstrator 2 (see Figure A13-3) DRAWBACKS - FRAGILE no particle with a size above 10 um (Drawback to test water, to use our current saline buffer tests) - Not self-priming - The flow rate is not reversible DRAWBACK MAINLY DURING THE ADJUSTEMENT - The flow rate varies with the time (ageing effect) problem for continuous application (Figure A13-4) - reliability problem.

Page 81: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 81 / 107

Figure A13-3: Flow rate versus the frequency

Figure A13-4: The flow rate drastically decreases with the time

0

200

400

600

800

1000

1200

1400

0 50 100 150 200 250 300 350

Flo

w r

ate

L/m

in)

Frequency (Hz)

Vpp=120V; Measurement time = 8min, only the pump

0

500

1000

1500

2000

2500

0 2 4 6 8 10

Flo

w r

ate

(M

icro

lite

r/m

in)

Hours

≈ 1170 ul/min

Vpp = 120 V

F = 160 Hz

Page 82: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 82 / 107

A13-2 Piezolectric Bartels pump CEA wanted to test Bartels pump. They have been ordered in April 2014. They only arrive at CEA at the beginning of September 2014.

Figure A13-5: Bartels pump : smaller than the current CEA peristaltic pumps on P2

ADVANTAGES -VERY SMALL (ALSO for the power supply plug and play) : similar to the Dolomite pumps -R2R possible in pick and place -The flow rate is compatible with the demonstrator 2 -Self-priming -The flow rate has been tested during 45 hours and the ageing effect is almost insignificant compare to the ageing observed with the Dolomite pumps (see Figure A13-4 and A13-6) -Typical maximum flow rate ≈ 7 ml/min DRAWBACKS - The flow rate is not reversible DRAWBACK MAINLY DURING THE ADJUSTEMENT Additional test with PBS has to be done before to be implemented on P2, to determine if the salt inside the solution is a show stopper or not.

Page 83: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 83 / 107

Figure A13-6: Bartels pump: Study concerning the flow rate stability with deionized (and filtered) water

Figure A13-7: Flow rate versus the voltage amplitude at constant frequency (20 Hz)

0,920

0,940

0,960

0,980

1,000

1,020

1,040

1,060

0 10 20 30 40 50

Flo

w r

ate

(m

l/m

inu

te)

Hours

Flow rate variation during almost 48 hours with the same condition (20Hz-170 V)

y = 0,0071x - 0,1823 R² = 0,9944

0,0

0,2

0,4

0,6

0,8

1,0

1,2

1,4

1,6

1,8

0 50 100 150 200 250 300

Flo

w r

ate

(m

icro

lier/

min

)

Amplitude SRS (V)

Série1

Linéaire (Série1)

Page 84: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 84 / 107

Figure A13-8 : Power supply and control command for Bartels pumps

Page 85: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 85 / 107

Annex 14 : Electro valve

SMALLER electrovalves, than the former electrovalves on P2 (see Figure A14-1), have been tested and implemented on P2 (power supply and control command). These electrovalves have been supplied by Dolomite. They are not yet currently used on P2. An antibody decrease has been successfully achieved with these electrovalves. These electrovalves are not R2R, but they are probably pick and place.

Figure A14-1: Size of various electrovalves -The Dolomite electrovalve will be used when the fluidic protocol will be stabilized -A manifold can be implemented to decrease the volume of the experiment and the tube length.

Former EV on P2

Dolomite EV on P2

Page 86: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 86 / 107

Annex 15:

Discussions and calculations in progress with Polyscale to collect more light in modifying the geometry

Objectives :

1- Geometry to collect more light

2- Geometry to collect light on a smaller area to increase the detection level (in adapting the detection)

The current geometry is the following:

you can choose a circular functionalized area if it is easier for the simulation (even a point source) (if you take the whole geometry from the side of the functionalized area, you have water (n =1.33) and for the other sides you have air (n=1)) n PMMA (600 nm)≈ 1.49 Probably the best way for the first simulation is to assume that the functionalized area is a light source which irradiates at : - a single wavelength (600 nm for instance) - with the same amount in all directions (isotropic source) (The simulation might be done only in 2 directions in ignoring the PMMA thickness, at least to begin)

23 mm

Page 87: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 87 / 107

Part A

Q1-In assuming that 100 photons are emitted by the functionalized area, how many photons go out by the detection side ? (1-bis: The size of the slide and the location and the functionalized area size can be modified) Q2-Same question than Q1 but in adding high a reflective coating with R= 90%, whatever the incidence angle and the polarization Q3-Same question than Q1, but the detection area is located at d= 23 mm of the detection side (i.e. detection side corresponds to d= 0 mm). The detection area has the following sizes: 4 mm x 20 mm (Rectangular PMT) Q3a- When the detection area is centered Q3b- What is the vertical position for the detection area at d= 23 mm, which corresponds to highest amount of light ? What is the light level ? (We have already an idea concerning this location) Q3c- Is it possible to increase the light inside the centered detection area in adding 2 mirrors (for instance at the detection side output). It is possible to increase d (i.e. d > 23 mm is possible) Q3d- Is it possible to increase the light inside the centered detection area in adding a lens (for instance at the detection side output). It is possible to increase d (i.e. d > 23 mm is possible) Experimentally, if you increase the distance between the functionalized area and the detection side, the collected light decreases. We did not check what will be the result if the functionalized area is moved parallel to the detection side

Detection area

20 mm

d= 23 mm

Detection area

20 mm

mirror

mirror

Page 88: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 88 / 107

Q4 -Same question than Q3, but the detection area is located at d= 7 mm of the detection side (i.e. detection side corresponds to d= 0 mm). The detection area has the following sizes: 1 mm x 3 mm (Micro Rectangular PMT)

Part B

Part B In the part B, the rectangular shape of the PMMA (fluorescent light guide) and the location of the functionalized area can be adapted 1-Are you able to find a shape, which maximizes the number photon on the detection size on a smaller area ? 1-a Rectangular size compatible with the hybridization with other sizes The “U shape” of the current chamber can be modified to become a “I” chamber with a 10 mm x 50 mm for the optical guide instead of 25 mm x 75 mm 1-b Other shapes few suggestions The following shape is probably not efficient without a mirror on the edges since a lot of photons are not reflected inside the guide by internal reflection. Number photons on this side with or without mirror on the edges

Detection area

3 mm

d= 7 mm

Page 89: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 89 / 107

May it is possible to slightly bend the rays towards the sensor (for half of them), but probably it is not enough. This shape might help to have a better efficiency to decouple the fluorescent light from the guide.

Page 90: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 90 / 107

Annex 16: Impact of the microfluidic tubes and their preparation :

Saturation of the fluidic circuit with BSA

1-Microfluidic circuit (i.e. tube materials) The effect of the material tube and their preparation during the hybridizations are illustrated with the data gathered in this annex. The following graphs illustrate the progresses achieved in March and April 2014 concerning this point. Conclusion: Tygon R3603 + tube preparation with BSA saturation during each night it works and leads to low LOD (≈ 20 ng/ml on P1 with washing buffer step of 120 seconds which might be a LOD “killer” and ≈ 7 ng/ml without washing buffer step on P2, spot size = 2 mm has to be confirmed that is a only specific signal )

The fluorescence behavior is unexpected and unpredictable impossible to work Figure A16-1: Used silicon platinium tubes without BSA preparation

The fluorescence behavior allows the detection of substance P and to calculate a LOD LOD (300s, glass slide) ≈ 185 ng/ml

Same experiment between A10-1 and A10-2b, except the tube in front of the chamber: “old” tube versus “new tube”

Figure A16-2: New silicon platinium tubes without preparation (H48)

3

3,4

3,8

4,2

4,6

5

0 400 800 1200 1600 2000 2400

Vo

ltag

e (

V)

= Fl

uo

resc

en

ce s

ign

al

(a.u

.)

Time (s)

LED1

LED2

LED4

Substance P = 1000 ng/ml

PBS buffer PBS buffer

PVP buffer

4

4,5

5

5,5

6

6,5

7

7,5

0 400 800 1200 1600 2000

Vo

ltag

e (

V)

= Fl

uo

resc

en

ec

sign

al

(a.u

.)

Time (s)

LED1

LED2

LED4

PVP buffer

PBS buffer Substan

ce P =

1000

ng/ml

PBS buffer

Page 91: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 91 / 107

“Stabilization” improvement

LOD (300 s) ≈ 220 ng/ml, spot diameter = 2 mm, P2 P1 exc≈ 515 nm

Figure A16-3: New Tygon R3603 tubes without BSA preparation (H49A)

The washing step has to be adapted to increase the LOD

our first idea was to remove it but the unspecific signal has to be removed

LOD (300 s) = 19,6 ng/ml, spot diameter = 2 mm, P1 exc≈ 530 nm, V(AB) = 300 L (Washing buffer during 2 minutes)

This LOD is exact if this LOD does not include the unspecific signal

Next steps : 1- to add PBS buffer between the PVP buffer and the substance P (like A10-2) 2- to replace PBS buffer by “controlled water”

Figure A16-4 (New) Tygon R3603 tubes with BSA preparation during the previous night (H65A)

3

3,5

4

4,5

5

5,5

0 5 10 15 20

Vo

ltag

e (

V)

- fl

uo

resc

ne

c si

gnal

(a.

u.)

Time (minutes)

PVP buffer Substance P = 1000 ng/ml

PVP buffer

Washing buffer

1,9

2,1

2,3

2,5

2,7

2,9

3700 4200 4700 5200

Vo

ltag

e (

V)

-Flu

ore

scn

ece

so

gnal

(a

.u.)

Times (s)

LED 1

LED 2

LED 4

Substance P = 1000 ng/ml

PBS buffer

PBS buffer

PVP buffer

Page 92: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 92 / 107

Annex 17: Various hybridization results This annex gathers various hybridization results (see the following table).Then, few graphs extracted from these date are shown.

Hybridization Platform/ substrate

Spot size

Comments WB = washing buffer

1 H48A P2/BK7 2 mm Signal Behavior when the tubes are : -used and not saturated with BSA (no WB)

H48 B P2/BK7 2 mm Same slide : improvements with new tubes (no WB) Detection of substance P become possible (Fluidic channel ≈ 260 microns) LOD (Ref=0) =226 ng/ml, 188 ng/ml, 142 ng/ml

H48 C P2/BK7 2 mm Fluidic channel ≈ 90 microns (no WB) LOD (Ref=0) = 525 ng/ml, 360 ng/ml, 391 ng/ml LOD : lower with a 300 um fluidic channel (typically divided by 2)

2 H49 A P2/BK7 2 mm Typical filter configuration (no WB)

LOD (PBS) = 497 ng/ml, 524 ng/ml, 685 ng/ml LOD (PVP) = 205 ng/ml, 202 ng/ml, 191 ng/ml

H49 B P2/BK7 2 mm With 1 x OG 570 in front of the PMT (no WB) LOD (PBS)= 985 ng/ml, 2735 ng/ml, 3161 ng/ml LOD (PVP)= 208 ng/ml, 356 ng/ml, 318 ng/ml

H49 C P2/BK7 2 mm With 2 x OG570 in front of the PMT (no WB)

LOD (PBS) = 485 ng/ml, 315 ng/ml, 686 ng/ml LOD (PVP)= 196 ng/ml, 136 ng/ml, 265 ng/ml

H49 D P2/BK7 2 mm With an OG 590 in front of the PMT (no WB)

LOD (PBS)= 616 ng/ml, 570 ng/ml, 760 ng/ml LOD (PBS)= 249 ng/ml, 209 ng/ml, 132 ng/ml

3 H53 A P2/BK7 2 mm PBS flat signal (no WB), 300 microns LOD (PBS, sigma PBS null)=246 ng/ml, 206 ng/ml, 236 ng/ml LOD (PBS) = 483 ng/ml, 439 ng/ml, 490 ng/ml

H53 B P2/BK7 2 mm LOD (PBS, sigma PBS null) =4220 ng/ml, 3085 ng/ml, 2532 ng/ml No WB, 90 microns

4 H58 A P1/BK7 2 mm WB = 5 minutes (before PVP) LOD (PVP <0) = 40 ng/ml, 46 ng/ml, 33 ng/ml

H58 B P1/BK7 2 mm WB = 5 minutes Detection is difficult when SP is inside PVP (worse than SP in PBS)

5 H59 A P1 BK7 2 mm WB = 5 minutes (before PVP) LOD (PVP<0) = 38 ng/ml, 46 ng/ml, 52 ng/ml (similar to H58A)

H59 B P2 BK7 2 mm WB = 5 minutes (BSA Saturation) LOD (wo PVP ) = 159 ng/ml, 88 ng/ml, 92 ng/ml

A part due to the excitation wavelength shift The other part might be due to flow rate (30%), no WB

H59 C P2/BK7 2mm A flow rate increase of 10% LOD (wo PVP)= 182 ng/ml, 84 ng/ml, 127 ng/ml

6 H65 A P1/BK7 2 mm hybridization with only 300 microliters of AB WB = 2 minutes (before PVP) LOD (PVP <0 )= 61 ng/ml, 51 ng/ml, 20 ng/ml

H65 B (New LED support)

P2/BK7 2 mm WB = 2 minutes (before PVP) LOD (PVP < 0)= 95 ng/ml, 95 ng/ml, 71 ng/ml Round PMT

7 H66A and H66B

P2/PMMA (IQS)

2 mm Detection limit is improved when the AB concentration is increased from 25 ug/ml up to 50 ug/ml : 1600 ng/ml 480 ng/ml WB = 2 minutes A first substance P signal on PMMA substrate specific?

8 H68 A P2/BK7 2 mm Washing buffer (WB) effect on LOD (see graph) Is the WB is necessary or not ?

The WB removes specific signal or not ?

Page 93: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 93 / 107

9 H69 P2/PMMA

(IQS) 3 drops x 100

uM

2 mm (Higher initial fluorescence) Washing buffer = 120 seconds, Reference line = PVP LOD (without PVP ref) = 180 ng/ml (only one spot)

10 H72 A and H72 B

P2/PMMA (LETI)

New pump on P2

3 drops x 100 uM

ML2-42-02

2 mm 1000 ng/ml and 200 ng/ml

Slope proportional to the substance P concentration It is a good point (The dispersion of LOD on the 3 spots is “weak” : good point) As to be confirmed for 5 mm spot size At lower concentration With a PBS reference line of PBS (Without or with WB The suitable time)

With PVP reference line, without WB LOD (1000 ng/ml, Round, 50 ug/ml) = 37 ng/ml LOD (200 ng/ml, Round, 50 ug/ml) = 16 ng/ml LOD (200 ng/ml, Round, 25 ug/ml) = 35 ng/ml LOD (200 ng/ml, Rect., 25 ug/ml) = 57 ng/ml Substance P slope is higher with AB = 50 ug/ml than 25 ug/ml Round PMT leads to better LOD than Rectangular PMT

11 H73 P2/PMMA 5 mm Experiments at various substance P concentration (no WB) (the buffer line is : PVP not PBS in which the substance P is diluted) See the graph H73

12 H75 P2/PMMA 5 mm It cannot be guarantee that the signal is due to the substance P (the lack of specificity is possible) This lack of specificity is confirmed by H83

13 H 83 (AB = 50 ug/ml)

P2/PMMA Various tripods

conc.

5 mm There is no specificity No difference between : - PBS/200 ng/ml - 200 ng/ml SP diluted inside PBS - 1000 ng/ml SP diluted inside PBS

H84 PMMA/IQS 3 x100 uM

2mm No detection at all

14 H85 A AB = 25 ug/ml

PMMA/2mm 3 x 100 uM ?

2 mm Specificity is detected on the spot 2 (1000 ng/ml-G= 0.6V) Slope (1000 ng/ml-250-400 s) = 3.90 E-4 V/s Sigma = 3.73 E-3 V LOD (wo PBS, reference PBS) = 96 ng/ml (see graph) Real LOD = 687 ng/ml (PBS reference after the SP signal)

15 H85 B AB = 25 ug/ml

PMMA/2mm 3 x 100 uM ?

2 mm Specificity is detected on the spot 2 (200 ng/ml-G=0.6 V) Slope (200 ng/ml-250-400s) = 3.65 E-4 V/s (similar signal than with 1000 ng/ml but the reference line is higher) Sigma = 1.66 E-3V LOD (wo PBs reference PBS) = 47 ng/ml Real LOD = 99 ng/ml (PBS reference after the SP signal)

Platform 2 -From H1 to H69 pump 1 At the end of its life, the flow rate typically decreases by 30% (700 ul/min instead of 1000 ul/min) From H71, this pump has been replaced by a new one (Pump 2) -The substance P is always diluted in PBS (except one experiment to show that it worse if it is diluted in PVP) -For most of the graph: X-Axis : Time (seconds) Y-Axis : Fluorescence level (a.u. in V)

Page 94: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 94 / 107

Stray signals with “old tubes“ without any preparation Figure A17-1

Improvements in using new tubes it allows the substance P detection (compare to Figure A17-1) Figure A17-2

3

3,2

3,4

3,6

3,8

4

4,2

4,4

4,6

4,8

5

0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400

H48A (ML2-37-08)

LED1

LED2

LED4

4

4,5

5

5,5

6

6,5

7

7,5

0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200

H48B : ML2-37-08 with new tube

LED1

LED2

LED4

PVP PBS SubP = 1000ng/ml

PBS

Page 95: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 95 / 107

Same comment than Figure A17-2 Figure A17-3

The washing buffer time : has to be “calibrated” to only remove the unspecific signal The reference line has to be in PBS

Figure A17-4

1,9

2,1

2,3

2,5

2,7

2,9

3700 3900 4100 4300 4500 4700 4900 5100 5300 5500

Vo

ltag

e (

V)

-Flu

ore

scn

ece

so

gnal

(a.

u.)

Times (s)

LED 1

LED 2

LED 4

Substance P = 1000 ng/ml

PBS buffer PBS buffer PVP buffer H49A

3

3,5

4

4,5

5

5,5

0 2 4 6 8 10 12 14 16 18Vo

ltag

e (V

) -

flu

ore

scn

ec s

ign

al (

a.u

.)

Time (minutes)

H65A

Washing

buffer

PVP buffer Substance P = 1000 ng/ml

PVP buffer

Page 96: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 96 / 107

First detection of substance P with PMMA substrate:

The AB concentration is x by 2 (50 ug/ml instead of 25 ug/ml)

Figure A17-5

Washing buffer (WB ) or not ? The WB is essential if it contributes to remove the unspecific signal (otherwise, it has to be removed).

Figure A17-6

1,2

1,3

1,4

1,5

1,6

1,7

1,8

0 200 400 600 800 1000 1200 1400 1600 1800

Flu

ore

scn

ece

sig

nal

(V

)

Time (s)

PMMA, AB Concentration= 50 µM (H66B)

Slot 1

Slot 2

Slot 4

Substance P = 1000 ng/ml

0

20

40

60

80

100

120

140

0 50 100 150

LOD (ng/ml)

Time of the washing buffer (seconds)

LOD Spot1 (ng/ml)

LOD Spot 2 (ng/ml)

LOD Spot4 (ng/ml)

500 ng/ml

500 ng/ml

200 ng/ml

200 ng/ml

H68 Best LOD = 6,9 ng/ml

(P2 ==> 515 nm/2mm/Round PMT) Substrate = glass slide

(Reference line = PVP, Assumption : similar to PBS line)

Page 97: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 97 / 107

WB = 2 minutes, IQS treatment LOD (without PVP signal) = 180 ng/ml IQS surface tretament AB = 50 ug/ml and the tripods concentration has been multiplied by 2 (3 x 100 uM)

Figure A17-7

The signal is linear with the substance P concentration (Without washing buffer) LETI treatment, spot size = 2 mm

Figure A17-8

1

1,2

1,4

1,6

1,8

2

2,2

2,4

0 1000 2000 3000 4000 5000 6000 7000

Vo

ltag

e (

V)

Time (s)

Hybridation on PMMA: H69

Gain = 0.63V

y = 3E-06x + 2E-05 R² = 0,9997

0,00E+00

5,00E-04

1,00E-03

1,50E-03

2,00E-03

2,50E-03

3,00E-03

3,50E-03

0 200 400 600 800 1 000 1 200

Slo

pe

sub

stan

ce P

on

ly (

V/s

)

Substance P concentration (ng/ml)

PMMA_wo washing buffer==>2 different concentrations of substance P inside PBS Spot size = 2mm x 2 mm (ML2-42-02)

Slope SP (V/s)

DV (V/s) [-PVP]

Linéaire (Slope SP (V/s))

LOD(PVP) = 35 ng/ml

LOD(PVP) = 16 ng/ml

Page 98: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 98 / 107

Effect on the substance P concentration on the slope variation it is the signature of the specificity (Spot diameter = 5 mm)

Figure A17-9A

X-Axis : Logarithm scale H73: 5 mm spot size on PMMA, LETI surface preparation

Figure A17-9B

0,8

0,9

1

1,1

1,2

1,3

1,4

1,5

1,6

1,7

4200 4400 4600 4800 5000

Flu

ore

sce

nce

sig

nal

(a.

u.)

Time (seconds)

H73A_plot2_200ng_ml_G=0,46V

H73B_plot2_20ng_ml_G=0,46V

H73C_plot2_5ng_ml_G=0,46V

PVP

SP

0,0E+00

4,0E-04

8,0E-04

1,2E-03

1,6E-03

2,0E-03

2,4E-03

2,8E-03

0,1 1 10 100

Slo

pe

of

the

sub

stan

ce P

on

ly (

V/s

)

Substance P concentration (ng/ml)

Dilution of the substance P inside the same buffer than the reference line become essential

==> it is not the case for these experiments

H73 Experiments : detection of various substance P concentrations (on PMMA, 5mm spot size)

Page 99: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 99 / 107

It cannot be guarantee that the signal is only due to the substance P (A lack of specificity is possible) This possible lack of specificity is confirmed by H83 (spot size = 2 mm), A17-11

Figure A17-10 (H75 = 5 mm spot size)

Spot size = 5 mm Figure A17-11

1,00

1,50

2,00

2,50

0 1000 2000 3000 4000 5000 6000 7000 8000

H75A (NA=0,87)

H75A (NA=0,87)

PV

SP = 20 ng/ml

PVP

0

0,5

1

1,5

2

2,5

3

0 1000 2000 3000 4000 5000 6000

Flu

ore

sce

nce

leve

l (a.

u.)

Time (seconds)

LED 1 LED 2

LED 3 LED 4

PBS

PBS + 200 ng/ml SP PBS +1000 ng/ml SP

H83

Page 100: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 100 / 107

No washing buffer at all Back to 2 mm spot size the specificity is back

LOD estimation without taking into account the PBS line : 47 ng/ml (when tested with 200 ng/ml) and 96/ml (when tested with 1000 ng/ml)

(strange : the substance P slope seems similar for the 200 ng/ml and 1000 ng/ml)

Figure A17-12

1,35

1,4

1,45

1,5

1,55

1,6

1,65

1,7

0 200 400 600 800 1000 1200 1400

Plot 2

C(SP = 1000 ng/ml (inside PBS)

H85A

PBS Only

0,95

1

1,05

1,1

1,15

1,2

0 200 400 600 800 1000

H85B

plot2_H85B

C(SP = 200 ng/ml) PBS only

Page 101: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 101 / 107

Annex 18: Output light fluorescence versus the PMMA thickness

The output fluorescent light from 3 PMMA slides with 3 different thicknesses, functionalized with a similar a quantity of fluorescent tripods, have been tested. The initial fluorescence, obtained with the same process, has been measured with a scanner at 532 nm. The offset (signal without the tripod fluorescence) is insignificant to compare the results except when an additional slide is used with the 500 microns thickness. The conclusion is: There is almost no difference for a PMMA thickness of 1 mm and 500 microns on P2 (96 %) On P2, the fluorescence output signal is typically of 60% with 200 microns compare to a 1 mm thickness. (Experimental data obtained with the same optical power and the same PMT and the same geometrical configuration. It is more difficult to keep the substrate flat for a 200 micron thickness)

Plot 1 Plot 2 Plot 4

Slide Thickness Voltage

(V) Fluorescence

level (a.u.) Corrected

voltage (V) Voltage (V)

Fluorescence level (a.u.)

Corrected voltage (V)

Voltage (V) Fluorescence

level (a.u.) Corrected

voltage (V)

ML2-62-01

1 mm 1,35 32929 1,32 1,51 35025 1,14 1,88 44714 1,81

ML2-62-08

500 µm (+ 500 µm)

1,29 33653 1,29 1,49 39833 1,27 1,52 46554 1,52

500 µm 1,33 33653 1,33 1,52 39833 1,30 1,61 46554 1,61

ML2-62-07

200 µm (+1 mm)

0,76 33664 0,76 1,03 34634 0,77 1,18 39832 1,01

200 µm 0,74 33664 0,74 1,05 34634 0,78 1,16 39832 0,99

Final result

Ratio [500 µm/1mm] 0,96

Ratio [200 µm/1mm] 0,59

Figure A18 : Output light fluorescence versus the PMMA thickness (Functionalized area with tripod for the substance P)

Page 102: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 102 / 107

Annex 19 : Discussion with CETEMMSA Introduction: quick status of the demonstrator 2

R2R Current Study/tests in progress

Pump No Instech peristaltic pump (1 or 2 pumps)

Piezoelectric Bartels pump (Piezo Dolomite pump has been discarded : Major drawback : flow rate decreases versus time)

Electrovalve No Small dolomite electrovalve None

Optical sensor No Hamamatsu PMT Various models and configurations are investigated. PMT location tests are in progress.

Optical filters No 2 Interferential filters and 1 colored filter (glass)

Possibility to use cheaper glass filter Dye are ordered to study this point separately of the hybridization technology Other possibility : a large shift between the excitation and the emission (paper study, discussion with biologists)

Hybridization fluidic chamber (and may be the fluidic circuit distribution in a second time)

Yes Hybridization fluidic chamber PMMA substrate have been chosen since its autofluorescence is compatible with the demonstrator 2. Width : ~ 6 mm Shape : U form Thickness : ~ 300 microns Typical volume : ~150/200 ul Spot shape : circular or square Diameter size : 5mm (backup : 2 mm)

The specificity of the substance P (on PMMA) remains under study.

Excitation module 3 solutions -The current one is based on 4 LEDs with 8 ball lenses and an optical filters (Not R2R) (Typically 40 mW with excitation filter - The OLED solution (IPT). The optical power which is delivered could be a drawback (Typically 2 mW for 5 mm x 5 mm wo excitation filter) - LGP (light guide plate) which is developed by Polyscale with 2 options (with or without the BEF) Comments : the electrical plugging and control command remains the same for these 3 solutions on the platform 2

CEA would like to study the impact of the incident optical power in the detection limit. Without a stabilized protocol, it is not possible to do we will try to simulate that with a liquid dye The first tests with the Polyscale solution show stray light outside the gratings (see photos below).

Page 103: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 103 / 107

CEA proposals 1-Metallisation 1-a Chamber metallization on the lateral edge (PMMA thickness ) : 200 microns Target goal : increase the specific signal in adding a reflecting lacquer on the PMMA edge (red line below) the to collect more light that the light reflected by total reflection. CETEMMSA answer As we discussed in our last conference call, this approach could be implemented firstly by thermal evaporation as a rapid proof of concept. Then, if we see clear evidence of signal increase, metallization could be performed by means of dip coating. In our opinion, inkjet printing is not the best solution herein,

considering that printing on top of the 200 m edge is not easy. The thermal evaporation test could be done at Cetemmsa. 1-b Metallization of the “emission face“ of the LGP (between the grating zones) Target goal: to remove the excitation light between the spot area Discussion with Polyscale to determine if the printed electronics can bring interesting solution to replace the PCB where the LED are implemented (Be careful to the thermal management). Other thing… CETEMMSA comment An alternative could be to cover the areas between spots with an opaque layer capable of absorbing the “extra” excitation light. Patterned layers (either reflective or opaque) can be inkjet printed, for instance.

To add a reflecting layer to avoid the light, which is not “emitted” from the grating zones ? There are LEDs on the side of the LGP (i.e. a light guide plate). The grating on the LGP redirects the light towards the hybridization areas.

Page 104: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 104 / 107

2- Printed electronics 2-a Heaters and temperature sensor to hybridize on one side of the hybridization chamber at constant temperature The goal is get a constant temperature to hybridize always at the same temperature. According the applications, it could decrease the standard deviation between the experiments, for other applications, it could improve the limit of detection. CETEMMSA comment This would be a nice approach. Our heaters and thermal sensors developed for Demonstrator 1 would serve as starting point to adapt them to PMMA. For the current reaction, this option is not compulsory Temperature target range; between 30° and 40° with the best regulation possible CETEMMSA comment This perfectly fits within the range of work of our heaters. It could be challenging since the PMMA has probably not a good thermal conductivity. The design has to be compatible with the hybridization process and the optical detection. May be the metallization on the edges of the PMMA light guide proposed on 1-a) have to be extended on the “main faces” to uniformly heat the zone, where are located the 4 hybridization areas. Today, we have our own chamber design compatible with our platform (one side in PMMA, one side in alumina with isoflon, watertightness insured by a PDMS gasket). We have discussed in June and July with IPT to have the possibility to test a future IPT R2R chamber on our Platform 2 (optic and fluidic). Therefore, if you need the design of the future R2R fluidic chamber, you have to discuss with IPT also. CETEMMSA comment Yes, this sounds challenging, indeed. If IPT sends us the design of the R2R fluidic chamber, we could study the possibility to integrate the heater & sensor within the chamber, and eventually the approaches discussed in 1a and 1b as well. We are opened to try this at Cetemmsa. Roadmap on an IPT chamber : Step 1 Thermal tests to characterize the temperature of the future hybridization zone (can be done at CETEMMSA) Step 2 This chamber will be tested outside the Platform 2 at CEA (i.e. geneframe hybridization with static fluidic, manual filling). Step 3 If the results of the following steps are successful, this option will be implemented to be tested on the platform 2

Page 105: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 105 / 107

Additional figures for a better understanding/discussion

PMT

Valves

Excitation module

Pump

Valve Driver

Excitation module : current Driver

Power supply 1 Power supply 2

NI Cards/modules (control command) (linked by USB to the Laptop)

Page 106: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 106 / 107

Cables inside

Valve Driver

Excitation module: current Driver

Power supply 1

Power supply 2

NI Cards/modules (control command) (linked by USB to the Laptop)

Page 107: REPORT - cordis.europa.eu · REPORT Laboratoire d ... Demonstrator 2 : see diffusion list for CEA ... coverslip, chips were soaked twice in acetone for one minute and blown dry with

Report

The information enclosed in this document are the contracting parties property. It can’t be reproduced or transmitted to third without their authorization Page : 107 / 107

Demonstrator 3 (M. Känsäkoski)

The contribution for the demonstrator 3 will be send to the prime as soon as possible.