Final Report: Ultrasensitive Pathogen Quantification in Drinking Water Using Highly Piezoelectric PMN-PT Microcantilevers

EPA Grant Number: R829604
Title: Ultrasensitive Pathogen Quantification in Drinking Water Using Highly Piezoelectric PMN-PT Microcantilevers
Investigators: Shih, Wan Y. , Mutharasan, R. , Shih, W.-H.
Institution: Drexel University
EPA Project Officer: Lasat, Mitch
Project Period: January 1, 2002 through December 31, 2004 (Extended to December 31, 2005)
Project Amount: $449,713
RFA: Exploratory Research: Nanotechnology (2001) RFA Text |  Recipients Lists
Research Category: Safer Chemicals , Nanotechnology

Objective:

The objective of this research project was to develop highly piezoelectric microcantilever arrays for in situ rapid, simultaneousmultiple pathogen quantification in source water with the ability to detect pathogens using electrical means with unprecedented femtogram sensitivity.

Summary/Accomplishments (Outputs/Outcomes):

[Pb(Mg1/3Nb2/3)O3]0.63-[PbTiO3]0.37 FILM Development for Microcantilever Fabrication

[Pb(Mg1/3Nb2/3)O3]0.63-[PbTiO3]0.37 (PMN-PT) is a solid solution of magnesium niobate and lead titanate that offers one of the best piezoelectric properties. The challenge in using PMN-PT is in processing. The difficulties in making useful PMN-PT include the presence of an unwanted pyrochlore phase and lead loss at the typical sintering temperature (1,200ºC).

Novel Nano-Coating Approach for PMN-PT Synthesis

To prevent the pyrochlore phase and to lower the sintering temperature, we have developed a novel approach that requires only one calcination step to produce single-phase perovskite PMN-PT and PMN powders. First, the Nb2O5 powder was coated with Mg(OH)2 as shown in Figure 1. The Mg(OH)2-coated Nb2O5 powder then was mixed with lead acetate in ethylene glycol. (See Figure 2(a) for calcinations at various temperatures.) X-ray diffraction showed that the powder calcined at 800°C for 2 hours resulted in complete perovskite conversion, 200°C lower than a typical calcination temperature. The reduced perovskite formed by mixing Mg(OH)2-coated-Nb2O5 particles with PbO precursor resulted from the coating of Mg(OH)2 on Nb2O5, which not only prevented the direct contact between PbO and Nb2O5 but also enhanced the mixing between Mg(OH)2 and Nb2O5.

Figure 1. Optical Micrograph of Mg(OH)2-Coated Nb2O5 Particles Where the Mg(OH)2 Appears as the Dark Layer Surrounding the White Core

Figure 2. A Schematic of (a) Mg(OH)2-Coated Particles Suspended in a Pb Precursor Solution and (b) PMN Suspended in a PT Precursor Solution

Low Temperature-Reactive Sintering of PMN-PT

The powders prepared from the coating method were more reactive and easier to sinter than typical powders obtained from solid state reaction. We further suspended the PMN-PT powder in a PT precursor for sintering, allowing intimate mixing of the PMN particles with the PT precursor. The result was a reaction-enhanced sintering that lowered the sintering temperature more than 300°C. Figure 2(b) shows a schematic of PMN particles suspended in a PT precursor solution. Figure 3 shows the X-ray of PMN-PT sintered at 850, 900, and 1,000ºC. Clearly, the reaction between the PMN and PT occurred between 850 and 900ºC. Figure 4 shows the bulk density as a function of sintering temperature for PMN-PT ceramics using PMN-PT powders made from the PMN in PT precursor solution. The green bodies began to sinter rapidly above 800ºC and became fully sintered around 850ºC, which is 300ºC lower than that of the traditional Columbite method. The lower sintering temperature was attributed to a reaction sintering mechanism as evidenced in Figure 3, which showed the reaction between the PT and PMN occurred around 850ºC, which helped the sintering of the PMN-PT compact. The PMN-PT powder obtained by this novel suspension method can be sintered to 95 percent theoretical density at 850-1000°C, which was more than 200-350°C lower than the typical PMN-PT sintering temperature.

X-Ray Diffraction of PMN-PT Sintered at Various Temperatures, Which Clearly Showed That the Reaction Between the PMN and the PT Occurred at Around 850ºC to Transform Into PMN-PT

Figure 3. X-Ray Diffraction of PMN-PT Sintered at Various Temperatures, Which Clearly Showed That the Reaction Between the PMN and the PT Occurred at Around 850ºC to Transform Into PMN-PT

Figure 4. Sintering Behavior of PMN-PT Powders Prepared by the Precursor-Coated Powder Suspension Method

PMN-PT Freestanding Films

For freestanding film formation, dispersion of the PMN-PT powders was accomplished by a blend of low- and mid-molecular weight dispersing resins for a higher powder loading to result in a denser final part with less shrinkage. The freestanding films were obtained by tape casting the obtained dispersion. Figures 5 and 6, respectively, show the photograph and a scanning electron microscopy (SEM) cross-section micrograph of a 22 μm thick freestanding PMN-PT film sintered at 1,000ºC for 2 hours. The photograph shows that the film was translucent, indicative that the film was dense, as confirmed by the SEM micrograph shown in Figure 6. The freestanding film approach eliminated many of the problems associated with substrate-based thin films, such as interfacial reactions, thermal expansion coefficient mismatch, and substrate pinning/clamping effects; this allowed the films to exhibit their outstanding piezoelectric properties.

Figure 5. Photograph of a Translucent PMN0.63-PT0.37 Freestanding Film 22 mm Thick and 1 cm in Diameter

Figure 6. Cross-Section SEM Micrograph of a 22 mm Thick PMN0.63-PT0.63-PT Freestanding Firm

Giant Electric-Field-Enhanced Piezoelectric Coefficient of Freestanding PMN-PT Films

Figure 7 shows the piezoelectric coefficient d31 of a 22 μm film versus electric field where the piezoelectric coefficient, d31, was deduced from the displacement of a 2.5 mm long cantilever consisting of a PMN-PT layer electroplated with a 5 μm thick copper layer. The d31 coefficient increased to 2,000 pm/V at about 9 kV/cm. The present d31 value of 2,000 pm/V was much higher than that of commercial specially cut PMN-PT single crystals. Furthermore, to confirm the enhanced d31 coefficient, we performed direct measurement of the lateral elongation of a freestanding film when an electric field is applied perpendicular to the film. The results of the direct measurement are shown in Figure 8. The direct lateral elongation measurements also showed an enhanced d31 of 2,000 pm/V at around 10 kV/cm, consistent with that shown in Figure 7. The enhanced d31 coefficient is an intrinsic effect of the freestanding film but not an effect of a bilayer structure used in the cantilever displacement measurement as described above.

Field Enhancement of Piezoelectric Coefficient d31 in Freestanding PMN-PT Tape (Filled Triangles). For comparison, the data for commercial PZT sheet (open circles) and PMN-PT bulk ceramic (open triangles) are also shown. Up (down) triangles indicate the behavior of increasing (decreasing) field showing the hysteresis behavior.

Figure 7. Field Enhancement of Piezoelectric Coefficient d31 in Freestanding PMN-PT Tape (Filled Triangles). For comparison, the data for commercial PZT sheet (open circles) and PMN-PT bulk ceramic (open triangles) are also shown. Up (down) triangles indicate the behavior of increasing (decreasing) field showing the hysteresis behavior.

Piezoelectric Coefficient d31 in Freestanding PMN-PT Tape Measured by Lateral Displacement of a PMN-PT Rectangular Strip Under DC Fields. The large field enhancement effect is consistent with the results in Figure 5. Up (down) symbols indicate the behavior of increasing (decreasing) field showing the hysteresis behavior.

Figure 8. Piezoelectric Coefficient d31 in Freestanding PMN-PT Tape Measured by Lateral Displacement of a PMN-PT Rectangular Strip Under DC Fields. The large field enhancement effect is consistent with the results in Figure 5. Up (down) symbols indicate the behavior of increasing (decreasing) field showing the hysteresis behavior.

PMN-PT Array Microcantilevers

The freestanding PMN-PT thick filmswere electroded on one side with sputtered Ti/Pt. A metal (Cu or Sn) layer of 5 μm in thickness was then electroplated on the Pt surface as the nonpiezoelectric layer, followed by deposition of the Ti/Pt electrode on the other face of the film. Figure 9 shows the SEM micrograph of a PMN-PT/Cu bilayer. The PMN-Pt/Cu bilayer was then embedded in wax and cut to the cantilever shape with a wire saw (Princeton Scientific Precision, Princeton, NJ). Optical micrograph of resulted array PEMS 500 μm long and 800 μm wide is shown in Figure 10(a) together with the resonance spectra of the PEMS shown in Figure 10(b). Note that because the cantilevers thus fabricated had very clean interfaces, they exhibited very high Q values (above 300°C), which was much higher than we typically get with commercial lead zirconate titanate (PZT).

Figure 9. SEM Micrography of 20 mm Thick PMN-PT Plated With Copper

(a) Array PECS Fabricated by Machining Freestanding and Electroded PMN-PT Layer. The picture shows the copper-electroded side, and (b) their resonance spectra. Note the sharp resonance peaks.

Figure 10. (a) Array PECS Fabricated by Machining Freestanding and Electroded PMN-PT Layer. The picture shows the copper-electroded side, and (b) their resonance spectra. Note the sharp resonance peaks.

Detection Using a PZT/Au-Coated Glass Cantilever

The cantilever used for Salmonella t. (ST) detection had a PZT layer 0.5 mm long, 2 mm wide, and 0.125 mm thick bonded to a 0.15 mm thick glass of the same width. The cantilever had a gold-coated glass tip 2 mm in length. The gold-coated tip was cleaned with H2O2:H2SO4 (1:2 v/v) for 5 minutes and dipped in a 2 mM mercaptoprionic acid (MPA) solution for 24 hours followed by treatment with 5mg/mL EDC and 5mg/mL NHS for 30 minutes to activate the carboxyl group. The gold-coated tip then was rinsed with a phosphate buffer solution (PBS) and dipped in a 400 nM antibody (anti-CSA) solution for 30 minutes to immobilize the antibodies. The antibody-immobilized cantilever detected ST in various concentrations more than 20 times higher than that of a 5M Hz QCM (see Figure 12 (a)). The detection area of the present cantilever was 1 mm2, its mass detection sensitivity was, therefore, 8 x 10-12 g/Hz.

Figure 12. (a) Δf Versus Time in EDC and NHS Modification on MPA, Antibody Immobilization, and ST Detection, (b) Δf Versus Time for ST Detection in Different Concentrations, ( c) Δf of 30 Minute ST Detection Versus Concentration. Note that the current detection limit, 5 x 103 cells/mL, is two orders of magnitude smaller than the infection dosage and more sensitive than the commercial ELISA, where Δf is the resonance frequency shift.

Figure 12(b) shows frequency shift versus time for ST detection at different concentrations. Figure 12(c) shows frequency shift of ST detection at 30 minutes versus concentration. Note the present cantilever could detect ST at a concentration as low as 5 x 103 cells/mL, two orders of magnitude smaller than the infection dosage and also two orders of magnitude more sensitive than the commercial enzyme-linked immunosorbent assay (ELISA) with a limit of 105 cells/mL.1,2. It is expected that when cantilevers as small as 100 mm in length are made, the field-enhanced d31, together with sensor optimization, will help maintain the sharp resonance peaks in water to realize 10-14 g/Hz for in-water, in situ biosensing applications with subcellular sensitivity.

Direct Detection of Bacillus anthracis at 36 Total Spores Using PMN-PT/Sn Microcantilever in a Flow Cell With 100 Femtogram/Hz Sensitivity

A flow cell was designed, fabricated and used for Bacillus anthracis (B.a.) detection (Figure 13). In these experiments, highly piezoelectric lead manganese niobate—lead titanate (PMN-PT) freestanding films of 22 μm and 8 μm in thickness—was developed in our laboratory and fabricated into cantilevers 800 μm wide and 200-500 μm long with a 3-4 μm thick tin layer and subsequently was used for detection of B.a. spores. Prior to detection, the cantilevers were insulated to enable immersion in phosphate-buffered saline solution while still allowing for electrical actuation and detection. Following insulation, the cantilevers were functionalized with anti-B.a. immunoglobin G (IgG), using thiol binding of MPA to the platinum electrode on the PMN-PT surface and subsequent covalent linkage of antibody amine groups to carbonyls of the bound MPA by means of EDC/NHS activation. The cantilevers were mounted in a specifically designed holder; this cantilever-holder assembly then was inserted into a custom fabricated flow cell system that provided for continuous closed circuit recirculation and delivery of the spore suspension to the cantilever. Using a Cole-Parmer Masterflex C/L pump, operating at 1mL/minute to circulate the fluid, a flow speed of 1.5 cm/second was developed over the sensor surface. Furthermore, the closed-circuit nature of the flow system and total volume of only 0.8 mL asserts the fact that the number of spores actually present in the system was equal to only 80 percent of the initial spore/mL concentration. Thus, when a concentration of 20,000 spores/mL was injected into the system, only 16,000 total spores were actually available for detection. At such a concentration, the 22μm cantilever yielded a total frequency shift of 1,500 Hz, whereas the 8μm cantilever yielded a total frequency shift of 3,000 Hz. When 1,600 spores were injected into the flow cell (2,000 spores/mL), the 22μm cantilever shifted 900 Hz, whereas the 8μm cantilever shifted 1,500 Hz. The lowest amount of spores as yet presented to the 22μm cantilever was 160, which produced a shift of 500 Hz, whereas, as stated above, just 45 spores presented to an 8μm cantilever still produced a shift of 800 Hz. When SEM was used to obtain a spore coverage factor on the cantilever surface, the sensitivity of the 8μm cantilever is found to be 3 x 10-13 g/Hz. A summary of the detection sensitivity of PMN-PT/Sn microcantilevers and PZT/glass cantilevers of different mass sensitivity is shown in Figure 14. The detection of B.a. in a concentration of 45 spores/mL with working volume of 0.8 mL, just 36 total spores in suspension yielding a frequency shift of 800 Hz, is the best in situ detection sensitivity to date.

Figure 13. SEM Micrograph of B.a. Spores

Figure 14. f Versus B.a. Concentration Where Δf Denotes the Resonance Frequency Shift. Open triangles denote Δf of 8-mm thick PMN-PT/Sn microcantilevers with 3 x 10-13 g/Hz sensitivity, open triangles 22-μm thick PMN-PT/Sn microcantilevers with 1 x 10-12 g/Hz sensitivity, and crosses for PZT/glass cantilevers with 3 x 10-11 g/Hz sensitivity. The 8 mm thick PMN-PT/Sn microcantilevers achieved 36 total spores sensitivity n less than 1 mL of liquid.


Journal Articles on this Report : 7 Displayed | Download in RIS Format

Other project views: All 33 publications 12 publications in selected types All 7 journal articles
Type Citation Project Document Sources
Journal Article Campbell GA, Mutharasan R. Escherichia coli O157: H7 detection limit of millimeter-sized PZT cantilever sensors is 700 Cells/mL. Analytical Sciences 2005;21(4):355-357. R829604 (Final)
  • Abstract from PubMed
  • Other:
    Exit
  • Journal Article Campbell GA, Mutharasan R. Detection of pathogen Escherichia coli O157:H7 using self-excited PZT-glass microcantilevers. Biosensors and Bioelectronics 2005;21(3):462-473. R829604 (Final)
  • Abstract from PubMed
  • Full-text: Science Direct
    Exit
  • Other: Science Direct PDF
    Exit
  • Journal Article Detzel AJ, Campbell GA, Mutharasan R. Rapid assessment of Escherichia coli by growth rate on piezoelectric-excited millimeter-sized cantilever (PEMC) sensors. Sensors and Actuators B: Chemical 2006;117(1):58-64. R829604 (Final)
  • Abstract: Science Direct Abstract
    Exit
  • Journal Article Luo H, Shih WY, Shih WH. Comparison in the coating of Mg(OH)2 on micron-sized and nanometer-sized Nb2O5 particles. International Journal of Applied Ceramic Technology 2004;1(2):146-154. R829604 (2004)
    R829604 (Final)
  • Abstract: Wiley
    Exit
  • Journal Article Luo H, Shih WY, Shih WH. Synthesis of PMN and 65PMN-35PT ceramics and films by a new suspension method. Ceramic Transactions 2003;136():251-260 R829604 (Final)
    not available
    Journal Article Yi JW, Shih WY, Mutharasan R, Shih WH. In situ cell detection using piezoelectric lead zirconate titanate-stainless steel cantilevers. Journal of Applied Physics 2003;93(1):619-625. R829604 (2003)
    R829604 (Final)
  • Abstract: Journal of Applied Physics
    Exit
  • Journal Article Zhu Q, Shih WY, Shih W-H. In situ, in-liquid, all-electrical detection of Salmonella typhimurium using lead titanate zirconate/gold-coated glass cantilevers at any dipping depth. Biosensors and Bioelectronics 2007;22(12):3132-3138. R829604 (Final)
  • Abstract from PubMed
  • Supplemental Keywords:

    Bacteria, chemical composition, chemical sensors, drinking water contaminants, emerging pathogens, environmental monitoring, environmental sustainability, environmentally applicable nanoparticles, innovative technologies, microbial risk management, nanoengineering, nanosensors, nanotechnology, risk assessment, pathogen qualification, pathogen quantification, pathogenic quantification, pathogens, piezoelectric microcantilevers,, RFA, Scientific Discipline, Water, Ecosystem Protection/Environmental Exposure & Risk, Sustainable Industry/Business, Environmental Chemistry, Sustainable Environment, Technology for Sustainable Environment, Monitoring/Modeling, Biochemistry, New/Innovative technologies, Drinking Water, Engineering, Chemistry, & Physics, Environmental Engineering, environmental monitoring, pathogens, aquatic ecosystem, nanosensors, bacteria, aquatic organisms, chemical sensors, pathogen quantification, piezoelectric microcantilevers, nanotechnology, environmental sustainability, other - risk assessment, chemical composition, analytical chemistry, environmentally applicable nanoparticles, pathogen qualtification, pathogenic quantification, microbial risk management, emerging pathogens, sustainability, nano engineering, drinking water contaminants, innovative technologies

    Progress and Final Reports:

    Original Abstract
  • 2002
  • 2003 Progress Report
  • 2004 Progress Report