1

BioMEMS and Bionanotechnology: Integrated systems for Biology and Medicine

Rashid Bashir

Birck Nanotechnology Center, Discovery Park, School of Electrical and Computer Engineering, Weldon School of Biomedical Engineering,Purdue University, West Lafayette, Indianahttp://engineering.purdue.edu/LIBNA

July 17th – 18th, 20061st Annual Conference Methods in Bioengineering, Cambridge, MA.

2

AcknowledgementsResearch Scientists/Post-docs:• Dr. Demir Akin• Dr. Amit Gupta• Dr. Jaesung Jang• Dr. Kwan Soup Lim

Graduate Students:• Hung Chang• Rameez Chatni• Angelica Davila• Oguz Elibol• Samir Iqbal• Yi-Shao Liu• Kidong Park• Bobby Reddy• Murali Venkatesan

BioVitesse, Inc. Co-Founder

Faculty Collaborators– Prof. A. Alam (ECE)– Prof. D. Bergstrom (Med Chem)– Prof. A. Bhunia (Food Science)– Prof. D. Peroulis (ECE)– Prof. M. Ladisch (Ag& Bio Engr) – Prof. G. Vasmatzis (Mayo Clinic)

Funding Agencies• NASA Institute on Nano-electronics and

Computing (INAC)• US Department of Agriculture (Center for

Food Safety Engineering at Purdue)• NSF, NSF NSEC at OSU• National Institute of Health, NIBIB• Discovery Park at Purdue University

BBBB

3

Novel Solutions for Frontiers in Biology and Medicine

Novel Solutions for Frontiers in Materials

and InformationProcessing

Biology & medicine

Micro/Nanotechnology and Systems

Biology & medicine

Micro/Nanotechnology and Systems

Biology & medicine

Micro/Nanotechnology and Systems

Apply micro/nano-technology to develop novel devices and systems that have impact in Biology and Medicine or are bioinspired

Group Mission

4

Our Research Group (not all present)

5

Feat

ure

Size

10nm

100nm

1µm

10µm

1nm

0.1nm

100µm

On Size and Scale !

Plant and Animal Cells

Most Bacteria

Min Feature of MOS-T

(in 2005)Virus

ProteinsOne Helical Turn of DNA

Gate Insulator

Top-down

Bottoms-Up

MEMS

Nan

oele

ctro

nics

And

Nan

osca

leSe

nsor

s

Mic

roEl

ectr

onic

s&

MEM

S

Nanopores

C-C Bond

6

BioMEMS and Bionanotechnology: Tools for Biology and Medicine

DNA

mRNA

Protei

ns

Genomics

Transc

riptomics

Proteo

mics

Metabolomics

Protei

n

Interac

tions,

Metabolite

s

Pathway

s

Tissue

Organ

s

Cell-C

ell

Interac

tions

Cytomics

Nano-pores,Cantilevers,NWs, NTs, Quantum Dots,Nano-particles,

Polymers and hydrogels,Ink-jet Printing,3-D Stereo-lithography,

Micro-fluidics,Soft-Lithography,

7

BioMEMS/Biochip Sensors

• Detect cells (mammalian, plant, etc.), microorganisms (bacteria, etc.), viruses, proteins, DNA, small molecules

• Use electrical & mechanical (optical) approaches at the micro and nanoscale in biochip sensors

• Label free, small sample size, array capability, cheap

DNACells ProteinsViruses

Molecules

8

Active Research Projects

http://engineering.purdue.edu/LIBNA

Glass cover

In/Out ports Cavities/ W ells

Epoxy adhesive

Pin 700µm

Lab-on-a-chip for Detection of Live

BacteriaLiu, Li, Gomez, Huang, Geng,

Bhunia, Ladisch, Bashir

Nano-Mechanical Cantilever Sensors for Detection of Viruses

Gupta, Akin, Broyles, Ladisch, Bashir

Nanopore Sensors for DNA DetectionChang, Andreadakis,

Kosari, Vasmatzis, Bashir

Dielectrophoresis Filters an Traps for Biological Entities

Li, Akin, Bhunia, Bashir

Electrical Characterization of DNA hybridization

Iqbal, Balasubramanian, Bergstrom, Bashir

Micro-Mechanical Cantilevers for

Detection of SporesDavila, Walter, Aronson,

Bashir

1um

Silicon Resistor

Silicon dioxide

Metal Contact Area

Au/Cr Electrode

Angle View with 85 degree (SEM, 10000x)

BASIC - Bio-inspired Assembly of

Semiconductor ICsLee, Bashir

Bacterial Mediated Delivery of Particles

in CellsAkin, Bashir

Trapping/Lysing of Bacteria/Viruses In

Microfluidic DevicesPark, Akin, Bashir

9

• Microfluidic devices as petri dish on a chip- “Impedance Microbiology” on a Chip- Hybrid Devices

• Electrical detection of spore germination• Nano-Cantilevers for virus and bacterial detection• Lysing of viruses in microfluidic devices• Nanopores for single DNA molecule characterization• Field effect device array for detection of proteins and

DNA

Outline

10

Bacterial Agent Cases Hospitalizations DeathsCampylobacter spp. 1,963,141 10,539 99Salmonella spp. 1,342,532 16,102 556Clostridium perfringens 248,520 41 7Staphylococcus spp. 185,060 1,753 2Shigella spp. 89,648 1,246 14Yersinia enterocolitica 86,731 1,105 2E. coli O157:H7 62,458 1,843 52Streptococcus spp. 50,920 358 0Bacillus cereus 27,360 8 0Vibrio spp. 5,218 125 31Listeria monocytogenes 2,493 2,298 499Brucella spp. 777 61 6Clostridium botulinum 58 46 4

Pathogenic bacteria in foods is a serious concern to public healPathogenic bacteria in foods is a serious concern to public health: th: Approx. 76 million people get sick; 300,000 hospitalization, 5,0Approx. 76 million people get sick; 300,000 hospitalization, 5,000 death00 death

Needs for Rapid, Sensitive and Specific Needs for Rapid, Sensitive and Specific Detection Methods for BacteriaDetection Methods for Bacteria

(CDC, 2002(CDC, 2002

11

Technology Platforms for Rapid Detection of Live Bacteria

R. Bashir, ECE, BME A. Bhunia, Food Science M. Ladisch, BME, ABEJ. P. Robinson, BMS, BME

L. MonocytogenesOn a chip

• Rapid detection of live bacteria has wide applications– Industrial Microbiology (Pharmaceutical facility monitoring,

Food Safety)– Blood transfusion– Human Diagnostics– Global Health

• Rapid time to result limited by: – Amplification, enrichment, culture– 48-72 hours for viability determination and identification

• Center for Food Safety Engineering (www.cfse.purdue.edu) through cooperative agreement with USDA-ARS

12

Overview of Our Detection Process

Total Time less than < 4 hours~ 25 min ~ 1- 3 hr

Electrical Detection of cell Growth

On-ChipConcentration

1000X

~ 15 min

Automated Off-ChipCell Concentration And

Recovery - 1000X

Sample Off-chipConcentration

Detection/ID

Data Analysis/Results• Water

• Food• Air• Body Fluids

Sample100 – 250 ml

100-250ml fluid 100ul fluid (w. cells) Biochip sensor Readout

13

Cell Concentration and Recovery

Chen, et al. Biotech Bioengr, 2005Huang, et al. Submitted, 2006

• Proprietary technology for cell concentration and recovery in small volumes – CCRTM kit

• One-time use filter cartridge• Recovery rates ~ 50%

Membrane surface with E. coli

0.4µm pore filter – 25mm diacapture and conc. bacteria

Top downSide view

1µm

14

Massaged Hotdog Experiment

Ladsich, et al. USDA Review, 2004

15

CCR Method

Ladisch, et al. USDA Review, 2004

Automated CCRSystem

16

Recovery Results100 ml sample in PBST

67.0%0.0±0.0%0.2±0.1%4.0±1.1%3267±153162.8±14.7%3852±123

65.5%0.0±0.0%0.5±0.3%2.9±1.6%3233±80862.2±15.5%4052±122

77.7%0.0±0.0%0.4±0.1%4.9±0.9%3767±60372.4±11.6%4652±121

Total recovery (3 collections)

Leakage through

permeate side

Recovery collection 3

(viable cells)

Recovery collection 2

(viable cells)

Total cells in collection 1

Recovery collection 1

(viable cells)

Final pressure drop (psi)

Initial cell conc.

(cells/ml)

Run#

E. coli

68.2%0.0±0.0%6.0±0.6%25.1±2.6%2303±24337.2±3.9%2866±223

75.9%0.0±0.0%2.3±0.7%13.7±2.4%3717±24460.0±3.9%3066±222

70.2%0.0±0.0%0.9±0.4%12.7±2.4%3510±23656.6±3.8%3166±221

Total recovery (3 collections)

Leakage through

permeate side

Recovery collection 3

(viable cells)

Recovery collection 2

(viable cells)

Total cells in collection 1

Recovery collection 1

(viable cells)

Final pressure drop (psi)

Initial cell conc.

(cells/ml)

Run#

Listeria innocua

65.5%0.0±0.0%2.9±0.5%6.4±0.3%6133±181556.2±16.6%40109±243

56.9%0.0±0.0%2.4±0.3%17.3±0.9%4067±90737.3±8.3%40109±242

61.0%0.0±0.0%1.7±0.3%5.6±2.4%5867±20853.7±1.9%40109±241

Total recovery (3 collections)

Leakage through

permeate side

Recovery collection 3

(viable cells)

Recovery collection 2

(viable cells)

Total cells in collection 1

Recovery collection 1

(viable cells)

Final pressure

drop (psi)

Initial cell conc.

(cells/ml)

Run#

Streptococcus faecalis

Huang, Chen, Ladisch, et al.

17

67..9%0.0±0.0%1.9±0.3%11.3±0.9%30000±45854.8±0.8%28548±583

78..9%0.0±0.0%1.9±1.0%13.1±0.6%35000±415863.9±7.6%30548±582

50.1%0.0±0.0%0.0±0.0%0.0±0.0%27433±285750.1±5.2%35548±581

Total recovery (3 collections)

Leakage through

permeate side

Recovery collection 3

(viable cells)

Recovery collection 2

(viable cells)

Total cells in collection 1

Recovery collection 1

(viable cells)

Final pressure

drop (psi)

Initial cell conc. (cells/ml)

Run#

Pseudomonas fluorescens

26.0%0.0±0.0%9.6±1.4%4.0±1.3%473±9312.3±2.4%2038±183

62.5%0.0±0.0%15.0±1.6%22.3±1.1%967±6825.2±1.8%2538±182

27.3%0.0±0.0%4.3±0.1%10.8±1.7%467±6712.2±1.7%2038±181

Total recovery (3 collections)

Leakage through

permeate side

Recovery collection 3

(viable cells)

Recovery collection 2

(viable cells)

Total cells in collection 1

Recovery collection 1

(viable cells)

Final pressure

drop (psi)

Initial cell conc. (cells/ml)

Run#

Bacillus thuringiensis

18

Integrated Chips for Study ofMicroorganisms and Cells

“Lab on a Chip” with microfluidics and micro/nanosensors

Glass cover

In/Out ports Cavities/ Wells

Epoxy adhesive

Pin 700µm

Lab-on-a-chip for Detection of Live BacteriaLiu, Park, Li, Huang, Geng,

Bhunia, Ladisch, Bashir

Nanopore Sensors for DNA DetectionChang, Andreadakis, Kosari, Vasmatzis,

Bashir

Dielectrophoresis Filters an Traps for Biological Entities

Li, Akin, Bhunia, Bashir

Micro-Mechanical Cantilevers for Detection

of SporesDavila, Walter, Aronson,

Bashir

Trapping/Lysing of Bacteria/Viruses In

Microfluidic DevicesPark, Akin, Bashir

Nano-Mechanical Cantilever Sensors for Detection of VirusesGupta, Akin, Broyles,

Ladisch, Bashir

Silicon Nanowires and Nanoplates for DNA and

Protein DetectionElibol, Reddy, Nair,

Bergstrom, Alam, Bsahir

Mech/Elect.Detection

DNA, protein

Cantilevers,NanoFETs, Nano-pores

Cell Lysing

Nano-probeArray

Micro-scaleImpedance

Spectroscopy

ViabilityDetection

Conc.Sorting

On-chipDielectro-phoresis

Fluidic Ports

SelectiveCapture

Ab-based

Capture

MEMSFilters

Filters

19

Bacterial Growth !

Invented in late 1800s (Petri and Koch)

Still the most widespread means to grow and detect the

presence of bacteria !

# of

bac

teria

Time

Lag phase

Stationary phase

Gro

wth

pha

se

20

Impedance Microbiology

Energy Metabolism

SugarsOxygen

Na+ATP

OtherProcesses

CO 2

Carbonic Acid

H2O

IonChannels

Lactic AcidAcetic Acid

Cell

K+

Na+Z

Owicki et al., Biosens. Bioelectron. (1992)

1.1

1.2

1.3

1.4

1.5

1.6

1.7

0 2 4 6 8 10 12 14 16 18 20 22 24

Time (h)

Cond

uctiv

ity (m

S/c

m)

a

b

c

f

g

e

d

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

0 2 4 6 8 10 12 14 16 18 20 22 24

Time (h)pH

g f e d c b

a

(A)

(a) no cell, (b) 2.5 x102, (c) 2.39x103, (d) 2.64x104, (e) 2.66x105, (f) 2.65x106, (g) 2.7x107 CFU/ml.

pH

Yang, L, Banada, P., Bhunia, A. Bashir, R., Biotech Bioengr, 2005

Conductivity

21• Gomez, et al., Sensors and Actuators B, 2002• Gomez, et al., IEEE/ASME JMEMS, 2005

Impedance Microbiology on a Chip

• A large cell concentration can be achieved by confining a few bacteria in a small volume 107 cfu/ml = 10 cfu/nl

• On-chip miniaturization Short detection time• Electrical detection (Impedance Microbiology) Automation

1.E+00

1.E+01

1.E+02

1.E+03

1.E+04

1.E+05

1.E+06

1.E+07

0 1 2 3 4 5 6 7 8 9

Detection time in hours

Initi

al c

ell p

opul

atio

n (C

0) [C

FU/m

l]

Low number of bacteria in large volume takes long time to detect

Low number of bacteria in very small volume (in biochip!) can be detected much faster

22

Δ

[ ] 230 Re2 RMSCMm EfrF ∇= επε

Bacterial Cell Concentration on a Chip: Dielectrophoresis

( )mp

mpmpCM ε2ε

εεε,εf

+

−= )(ωε=εp

*

εp2 < εm

εmεp1 > εm

Electrode

εp2 < εm

εmεp1 > εm

Electrode

εmεp1 > εm

Electrode

DetectionC hamber

Electrodes

FlowFlow

Beadsand

bacteria

DetectionC hamber

Electrodes

FlowFlow

Beadsand

bacteria

H. Li and R. Bashir, Sensors and Actuators, 2002H. Li, D. Akin, and R. Bashir, IEEE/ASME JMEMS, 2005

Polystyrene beads : εp < εm negative DEP

Cells : εp < εm Negative DEPCells : εp > εm Positive DEP

23

A Dielectrophoretic Filter

Electrodes 20um line, 20um space

0 0.02 0.04 0.06 0.08 0.1 0.120

50

100

150

200

250

300

350

2.4μm bead

5.4μm bead

Vacciniavirus

0 0.1 0.2 0.3 0.4 0.5 0.6 0.70

50

100

150

200

250

300

Flow rate (μl/min)

yeast cells

Volta

ge (V

2 p-p

)

B. Cereus spores

Listeria innocuabacteria

Calculated

Experimental

Volta

ge (V

2 p-p

)

Flow rate (μl/min)

H. Li, et al., IEEE/ASME Journal of MicroelectromechanicalSystems. Vol. 14, No. 1, February 2005, pp. 105 - 111

Q = 0.1ul/min <v> =400um/sec

12um

350um

24

Petri Dish-on-a-Chip

PC boardw. heater

Wire-bond(with epoxy)

Micro-fluidicTubes

Micro-fluidicTubes

Edge Connector

MicroscopeObjective

BioChip

Measurementelectrodes

DEP captureelectrodes

Outlet Inlet

Measurementelectrodes

DEP captureelectrodes

Outlet Inlet

R. Gomez, D. Morisette, R. Bashir, IEEE/ASME JMEMS, 2005

25

Cell Concentration

• DEP-based concentration system collects particles from a large flowstream and diverts them to a smaller stream

Inlet

MainOutletDeviation

Outlet

DEP deviation electrodes

DEP electrodes(w. Abs and

growth detection)Fluid and Particles

(low flow)

Fluid only(large flow, no particles)

Fluid and particles

(large flow)

Fluid only(low flow, no

particles)

R. Gomez, D. Morisette, R. Bashir, IEEE/ASME JMEMS, 2005

26

102 103 104 105 106103

104

105

106

Mag

nitu

de [ Ω

]

Fits to incubation of L. innocua (~3x107 ml-1) in TA4-B6

Ch. 2. Well WB2. 39C.

102 103 104 105 106-60

-50

-40

-30

-20

-10

Frequency [Hz]

Ang

le [D

egre

es]

Meas. (2) 10/16/02. Fit (8) 10/24/02

Time

On-Chip Incubation of L. innocua in LB BrothInitial concentration: ~3x107 cfu/ml

Measured Impedance

Fitted circuit model

Time

On-Chip Incubation of L. Monocytogenes

ZwZw ZwZw

Cdi

Rs

Dielectric capacitance

Electrolyteresistance

Electrode-electrolyte interfaces

Electrode-Electrolyte

Interface Model:

Constant-angleimpedance

BjZ nw )(

1ω

=

27

90%

110%

130%

150%

170%

190%

210%

230%

250%

270%

290%

310%

330%

350%

0 2 4 6 8 10 12 14 16 18 20Time [hours]

Rel

ativ

e Ad

mitt

ance

at 1

00H

z

96%

96%

97%

97%

98%

98%

99%

99%

100%

100%

101%

1 cfu (1) 7/20/0234 cfu (1) 6/25/029 cfu (2) 3/21/02

Approximate number of colony-forming-units (cfu) in a 5.27nl volume:

95%

100%

105%

110%

115%

120%

125%

130%

0 2 4 6 8 10 12 14 16 18 20Time [hours]

Rel

ativ

e C

ondu

ctan

ce

1 cfu9 cfu34 cfu

Approximate number ofcolony-forming-units (cfu)in a 5.27nl volume: Growth

GrowthGrowth

Electrical Growth Curves

8.7x104 cfu/ml + DEP2.3x105 cfu/ml + DEP2.3x105 cfu/ml + no DEP Sterile HLB

30%40%50%60%70%80%90%

100%110%120%130%140%150%160%170%

0 2 4 6 8 10 12Time [hours]

Rel

ativ

e A

dmitt

ance

at 1

00H

z

96%97%98%99%100%101%102%103%104%105%106%107%108%109%110%111%

Rel

ativ

e Ad

mitt

ance

at 9

35.7

6Hz

~1 hr

~7.5 hrs

28

Expected Variation in Time to Detect vs. Concentration

0

1

2

3

4

5

6

7

8

9

10

1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 1E+12

On-chip Cell Concentration at Start of Incubation (cfu/ml)

Tim

e to

Det

ect (

hour

s)

Detection Time Limit

Lower Detection

Limit

Upper Detection

Limit

Dynamic Range

29

Pseudomonos Aeruginosa

0

2

4

6

8

10

12

14

1.0E+06 1.0E+07 1.0E+08 1.0E+09 1.0E+10

Initial On-Chip Concentration (cfu/ml)

Tim

e to

Det

ect (

hrs)

Escherichia Coli k12

0

2

4

6

8

10

12

14

1.0E+06 1.0E+07 1.0E+08 1.0E+09 1.0E+10

Initial On-Chip Concentration (cfu/ml)

Tim

e to

Det

ect (

hrs)

Experimental Results

30

Mediaa Initial conductivity

Exponential Growth Rate

Generation Time [h]

Lag phase Duration [h]

Maximum Population

Density LCGM 1280 µS* 0.08±0.017 B 12.9±2.77 B 2.3±0.45 AB 0.41±0.06 C

LB 1200 µS 0.07±0.004 B 12.7±0.21 B 1.7±0.25 B 0.2±0.009 D

LCGMs 1500 µS 0.04±0.021 B 28.9±2.72 A 3.44±0.52 AB 0.18±0.02 E

LBs 1500 µS 0.04±0.007 B 26.2±2.86 A 4.05±0.57 A 0.16±0.01 E

TVC 2800 µS 0.1±0.02 AB 10.3±2.45 C 3.02±1.66 AB 0.46±0.01 C

LRB 15.08 mS† 0.15±0.01 A 6.85±0.58 C 3.44±0.98 B 1.73±0.19 A

BLEB 16.00 mS 0.16±0.05 A 9.24±0.18 C 4.29±0.06 A 0.88±0.02 B

Comparison of Listeria monocytogenes growth in various low and high conductive media

• LCGM, low conductive growth medium; LB, Luria Bertani• LCGMs, LCGM with antimicrobial selective agents (Moxalactam (0.15%) and Colistinsulphate (0.1%)• TVC, total viable count media (Sy-lab, Austria); LRB, Listeria repair broth• BLEB, buffered Listeria enrichment broth; *µS- micro siemens; †mS-milli siemens

P. P. Banada, Y. Liu, L. Yang, R. Bashir, and A. K. Bhunia, International Journal of Food Microbiology, 2006.

31

Overall Detection Scheme

Biotinylated Antibody (C11E9)

SiO2 surface

~14 nm

Streptavidin

Biotinylated BSA

~10 nm

~4 nm

Streptavidin

Biotinylated BSA

~10 nm

~4 nm

* Size information are obtained from Biochemistry 2nd

edition, R. H. Garrett and C. M. Grisham, 1995.

H. Li, et al., Transducers, 2005L. Yang, H. Li, D. Akin, T. Huang, A. Bhunia, M. Ladisch, R. Bashir, Lab Chip, 2006

• Antibody to L. Monocytogenes

– 150kD IgG– Binds to a 66kD

surface protein

32

DEP on DEP on

DEP off DEP off

104 ce

lls/5μl

103 ce

lls/5μl

104 ce

lls/5μl

103 ce

lls/5μl

DEP on

DEP off

Green - Listeria monocytogenes

0

5

10

15

20

25

30

35

Ab

capt

ure

effic

ienc

y

101 103102

27.1%

18.6% 19.1%

DEP captured cell number/5 μ l

0

5

10

15

20

25

30

35

Ab

capt

ure

effic

ienc

y

101 103102

27.1%

18.6% 19.1%

DEP captured cell number/5 μ l

Red - Enterobacter aerogenes & Enterococcus faecalis

At 103 / 2μl with a ratio of 1:1:1

C11E9 Antibody capture efficiency for

L. monocytogeneswith DEP

enhancement

33

BioVitesse, Inc.• ‘Petri dish on a chip’ to miniaturize impedance microbiology• To quickly and reliably detect and identify live bacteria in 2 to 4 hours, instead of 2 to 10 days• Provides in-process quality control monitoring systems• To the industrial microbiological market (Bio/Pharma and Food Safety)

Goal – Become the leader in rapid detection and identification of live cells

BioVitesse, Inc. Company Confidential

SiliconBiochip

Chip Cartridge

CCRTM

Cartridge

Automated System

Sample, Media,Sanitizing Fluid

34

Current Status and On Going Work

• Demonstrated automated electrical detection of 10-100cfu/ml in 100ml fluids in one time use CCR and Biochip cartridge

• Serial dilution and concentration• Integrated pH and impedance sensor• Antibiotic resistance of bacteria – drug screening• Slow growing bacteria (e.g. TB) & growth of bacteria in low O2

ambient (e.g. Obligate anaerobes)• Extraction of bacteria from blood and body fluids – medical

diagnostics• Nucleic acid based identification

35

Micro-fluidic Polymer Devices for Culture of Bacteria and Spores

Input TubeOutput Tube

PDMS Cover

SiliconOxide

MetalElectrodes

Bonding with PDMS

3-D fluidic Channel

Input TubeOutput Tube

PDMS Cover

SiliconOxide

MetalElectrodes

Bonding with PDMS

3-D fluidic Channel

Input Reservoir

in 3rd Layer of PDMS

1st Layerof PDMS

2nd Layerof PDMS

ChipUnderneath

1.0E+03

1.0E+04

1.0E+05

1.0E+06

1.0E+07

0 5 10 15 20 25

Tim e (hours)

Impedance Magnitude [W

]

10:1, saturated10:1, unsaturated

2.5:1, unsaturated

2.5:1, saturated

1.0E+03

1.0E+04

1.0E+05

1.0E+06

1.0E+07

0 5 10 15 20 25

Tim e (hours)

Impedance Magnitude [W

]

10:1, saturated10:1, unsaturated

2.5:1, unsaturated

2.5:1, saturated

Chang, et al., Biomedical Microdevices 5:4, 281-290, 2003,

Silicon Base, 3 PDMS layers, Top I/O port

36

Electrical Detection of Spore Germination

• Bacillus Anthracis spores (stern strain)• Alanine & Inosine as germinants• DPA is ~10 % of spore dry weight (~10-13 g)

-5

0

5

10

15

20

0 10 20 30 40

Elapsed Time (mins)

Con

duct

ivity

Cha

nge

(%)

Germinants Only

Spore Only 1e9 Cells/ mL

Spore with Germinants1e9 Cells/ml -

Macroscale Experiments

Spore

Bacteria

Inosine

Alanine

DPA(Dipicolinic

Acid)

Liu, et al. in preparation, 2006

37

Spore Germination in Chip

Spores labeled by OC63 and captured by DEP of electrodes

Conductivity vs. Time - 750 Hz1e10 Cells/mL

-5

15

35

55

0 20 40 60Elapsed Time ( mins)

% C

hang

e of

|Con

duct

ivity

|

Spores wgerminant

Sporesonly

Conductivity vs. Time - 750 Hz1000 spores in 0.1nl = 1e10 Cells/ml

-5

15

35

55

0 20 40 60-5

15

35

55

0 20 40 60Elapsed Time ( mins)

% C

hang

e of

|Con

duct

ivity

|

Spores wgerminant

Sporesonly

Spores wgerminant

Sporesonly

Conductivity vs. Time - 750 Hz1e10 Cells/mL

-5

15

35

55

0 20 40 60Elapsed Time ( mins)

% C

hang

e of

|Con

duct

ivity

|

Spores wgerminant

Sporesonly

Conductivity vs. Time - 750 Hz1000 spores in 0.1nl = 1e10 Cells/ml

-5

15

35

55

0 20 40 60-5

15

35

55

0 20 40 60Elapsed Time ( mins)

% C

hang

e of

|Con

duct

ivity

|

Spores wgerminant

Sporesonly

Spores wgerminant

Sporesonly

Conductivity vs. Time - 750 Hz1e9 Cells/ml

-10

0

10

20

0 20 40Elapsed Time ( mins)

% C

hang

e of

|Con

duct

ivity

|

Spores wgerminant

Sporesonly

Conductivity vs. Time - 750 Hz100 spores in 0.1nl = 1e9 Cells/ml

-10

0

10

20

0 20 40Elapsed Time ( mins)

% C

hang

e of

|Con

duct

ivity

|

Spores wgerminant

Sporesonly

Comparison between Open & Close Valve Experiments(Open Valve: 700 Spores v.s. Close Valve: 100 Spores)

Whole Experiment v.s. Time

-5.00E-09

0.00E+00

5.00E-09

1.00E-08

1.50E-08

2.00E-08

Elapsed Time (minutes)

Net

cha

nge

inA

dmitt

ance

- 75

0 H

z

Close Valve - Germinant + Spore Open Valve - Germinant + Spore

Liu, et al. in preparation

• PDMS channels and valvesGlass base with metals

38

Integration of PDMS and Silicon

Transferable Microscale Metal Patterns for PDMS Metallization

(g)

(i)

(k)

(a)

Si waferAu layer

Ti layermercaptosilane layer

(h)

(j)(b)2nd Au (50 Å)

Ti (500 Å)1st Au (1500 Å)

Process flow of metal micro-patterning on PDMS: (a) substrate cleaning, (b) serial metal deposition (1500 Å of 1st Au, 500 Å of Ti, and 50 Å of 2nd Au) on Si wafer, (c) metal patterning with conventional photolithography, metal deposition (3000Å or 3rd Au), and lift off, (d) mercaptosilane treatment, (e) pouring and curing PDMS, f) peeling off PDMS from Si wafer, and (g) removal of 1st Au, Ti, followed by a short 2nd Au etch

Metal patterns are transferred to PDMS

Lim, et al. Lab Chip, 2006

39

Microscale Metal Patterns on PDMS

Minimum size of transferable pattern

40

BASIC: BIO-INSPIRED ASSEMBLY OF SEMICONDUCTOR ICs

Overall Approach:

Use of bio molecules and electro-hydrodynamics (EHD) for directed self assembly of silicon devices in 3-Dimensions

Indiana 21st Century R&T Fund

AC Electric

Field

Directed Assembly

Completed device released

in fluid

S. Lee, et al., Proceedings of MRS, 2001, 2002S. Lee, et al., Langmuir, April 2002

Molecules

41

3-D Heterogenous Integration

Displays/Electronics

Applications

• Processing temperature limits performance

– a-Silicon < 350°C– Poly-silicon < 650 °C– Single crystal < 1100 °C

• Single crystal on plastic !!– High resolution, flexible displays– Flexible Electronics

End-of-Roadmap Assembly• Multi-level 3-D micro-systems

– Higher functionality/volume• Integrated Systems

– Electronic– Optical– MEMS devices

• Nano-scale device candidates– Carbon Nanotubes– Silicon Nanowires

Lieber, et al, Feb 2nd, Sciencehttp://www.research.ibm.com/topics/popups/serious/nano/html/nhow.html

BASICC o m p lem en ta ry

m o lecu les

S e lf-A s sem b ly

D e vic e s

C o m p lem en ta rym o lecu les

S e lf-A s sem b ly

D e vic e s

42

Start Material(SOI Wafers)

Directed Self- Assembly

In Fluid

SOI NMOSFabrication

Device ReleaseIn Fluids

1st Layer Wafer 1st Layer Metal

Interconnect

1st SOI NMOS

M1 Layer

Open Contactsand Form 2nd

Device Layer

2nd Layer MetalInterconnect

2nd SOI PMOS

2nd SOI PMOS

M2 Layer

The processcan be repeated

over again tilldesired

3-Dimensional Integration

SOI PMOSFabrication

Device ReleaseIn Fluids

1st Layer Wafer

Open Contactsand Form 3rd

Device Layer

3rd SOI NMOS M2 Layer

S. Lee and R. Bashir, Advanced Materials, 2005

43

P-type silicon layer

Grown oxidePhotoresist

(c)

Silicon (100)

3-D MOSFET Body

(b)

Silicon (100)

(a)Highly doped S/D and gate regions

Nitride Spacer

(e)

(g)

Silicon (100)

Au/Cr S/D contact

Buried Oxide

(d)

Poly oxide feature Photoresist

Silicon (100)

Poly gate feature

Silicon (100)

(f)

Silicon (100)

Poly Silicon for gate

Silicon (100)

Grown poly oxide

Oxide for gate

(h)

Source

DrainGate

Transistor Process Flow

1,9-nonanedithiol

44

Transistor Release

S. W. Lee, R. Bashir, Advanced Materials, 2005

Oxide pillar

Silicon (100) (a)

45

Transistor Assembly

Randomly positioned

MOSFET in DI water

Assembled MOSFET in DI

water

Assembled MOSFET AF DI water was dried

completely

46

Electrical Results

Before release After Assembly

Rd, Rs ~ 0.5e4 Ω/μm2

47

Next Steps

• Array formation – flow through system or denser array

• Yield of release and assembly• 2-D and 3-D circuit demonstration• Orientation independent device

fabrication• Gate all-around device fabrication

48

BioMEMS Polymer/Silicon Cantilever Sensors

• Environmentally sensitive micro-patterned polymer structures on cantilevers

50 μm

Cantileverover a well

PMMA-based HydrogelPolymer

Z. Hilt, A. Gupta, O. Elibol, R. Bashir, N. Peppes

R. Bashir, J.Z. Hilt, A. Gupta, O. Elibol, and N.A. Peppas, Applied Physics Letters, Oct 14th, 2002; J. Z. Hilt, A. K. Gupta, R. Bashir, N. A. Peppas Biomedical Microdevices, September 2003, Volume 5, Issue 3, 177-184

- ΔpH = 1-10e-5 - pH = 6.5 ~ 1.9e5 H+ in 1000μm3

- ΔpH = 5e-4 change of ~ 150 H+

0

5

10

15

20

25

2.5 3.5 4.5 5.5 6.5 7.5 8.5

pH of Fluid Around the Cantilever

Def

lect

ion

at th

e tip

(μm)

Cantilevers touched the

bottom

ExperimentModel

slope = 18.3mm/pH(=1nm/5e-5ΔpH)

-

49

• Main aims:– Single molecule biophysics– Detection of hybridization – Towards direct sequencing of

single molecules

• Miniature DNA analyzer with ultra high sensitivity – Transduction events at single

molecule, nanoscale– Interface to the macroscale

Nanopore Channel Sensor Array for Characterization of Single Molecules

50

Fabrication Process

40

50

60

70

80

90

100

22500 23000 23500 24000 24500 25000Time (msec)

Cur

rent

(pA

)70

75

80

85

90

95

23144 23146 23148 23150 23152 23154

Time (msec)

Cur

rent

(pA

)

Pulses due to passage of dsDNA

Ox 1000 ASi 1400 AOx 4000 A

Handle layer

Ox 1000 A

Open submicron etch window on front side by e-beam lithography

Handle layer

Ox 1000 ASi 1400 AOx 4000 A

Ox 1000 A

Etch through the SOI wafer from backside

Ox 1000 ASi 1400 AOx 4000 A

Handle layer

Do front side etch to make a nanopore on buried oxide layer

50 – 60 um

Cont’d

Si1400 AOx 4000 A

Handle layer

Remove buried oxide layer to open a nanopore

Si1400 A

Initial opened pore is around 50-100nm

50-100nm

230-240 nm

2-4nm

Do thermal oxidation and TEM processing toshrink pore to less than 5 nm

H. Chang, F. Kosari, G. Andreakakis, M. A. Alam, G. Vasmatzis, R. Bashir, Nano Letters, 2004.H. Chang, S. M. Iqbal., E. A. Stach, A. H. King, N. J. Zaluzec, R. Bashir, , Applied Physics Letters, 2006.

~50nm Silicon

100nm SiO2

100nm SiO2

~36 nm

Not drawn to scale

200bp DNABuffer solution : 0.1 M KCl, 2 mM Tris(pH 8.5)Ag/AgCl electrodesBias : 200 mV

51

ΔI

KClConcentration

52

Explanation of the Upwards Pulses

Time

Cur

rent

(Arb

. U

nits

)

Ibulk(K+), Ibulk(Cl-), and Iint(K+).

ΔIint (K+)

ΔIbulk(K+),ΔIbulk(Cl-)

Total net current

H. Chang, et al., Nano Letters, vol. 4, No. 8, 1551-1556, 2004.H. Chang, et al. Biomedical Microdevices, 2006

- - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - DNA movement

+ ve+ ve+ ve - ve- ve- - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - -

Ipulse = k (Vbias /Lpore) 2 q (1 − φ) μK+int/b – k (Vbias /Lpore) q M* (μK+ + μCl-) Apore,

53

Voltage Dependent Pulse Reversal

54

- - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - DNA movement

+ ve+ ve+ ve - ve- ve- - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - DNA movement+ ve+ ve - ve- ve- - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - -

DNA region depleted of counterions

Iblocking

Ienhancement

I

Vbias

Increasing KCl Molarity

Vsat VT1 VT2

I Puls

e

Vbias

Increasing KCl Molarity

VT1 VT2

Voltage Dependent Pulse Reversal

55

Work in Progress

Inlet Port

BottomElectrode

oxide

Impedanceelectrodes

Top ElectrophoresisElectrode

Membrane BottomElectrode

oxide

Electrode

Glass

SiliconSubstrate

Glass

A

Impedanceelectrode

A

A’

B’

B

Micro-scale ElectrophoresisElectrode

Time

Cur

rent

(A-A’)

A’

Silicon

~50-60nm-~10-15nm

56

Nano-Mechanical Sensors for Virus Detection

Concept Device Schematic

Objectives: To develop technology for the rapid detection of virus particles in air and fluids (VacciniaVirus, Human Corona Virus)

MechanicalDetection

Cantilevers

Conc.Sorting

On-chipDielectro-phoresis

SelectiveCapture

Ab-based

Capture

Cantilever BeamElectrodes

Biological Entities

(e.g. virus)

Cantilever BeamElectrodes

Biological Entities

(e.g. virus)

Silicon Cantilever Beam

Biotinylated Bovine Serum Albumin (BSA) and BSA

Vaccinia Virus Particles

Biotinylated Antibody

Streptavidin

HCoV Virus Particles

HCoV Virus Particles

Western Reserve Strainof Vaccinia Virus

57

Piezoelectric Driven

Thermal Noise

Spectra

Cantilever Beam Length = 10 μm

Thickness = 30 nm

f0 = 270 kHz

Mass Change DetectionMass Change Detection

Microcantilever Mass Sensors

Unloaded Resonant Frequency :

Spring constant for a rectangular

shaped cantilever beam:

∗=

mkf

π21

0

3

3

4lwEtk =

⎟⎟⎠

⎞⎜⎜⎝

⎛−=Δ 22

12

114 off

kmπ

• k = spring constant• m = mass of cantilever• f0 = unloaded resonant frequency• f1 = loaded resonant frequency

⎟⎟⎠

⎞⎜⎜⎝

⎛−=Δ 22

12

114 off

kmπ

• k = spring constant• m = mass of cantilever• f0 = unloaded resonant frequency• f1 = loaded resonant frequency

Loaded Resonant frequency :

δm is the added massmm

kfδπ +∗

=21

1

3 μm 4 μm 5 μm 6 μm 7 μm 8 μm 9μm

Length of cantilever beam ( μm)

Width of cantilever beam = 1 μm

Thickness of cantilever beam = 10 nm

10 μm3 μm 4 μm 5 μm 6 μm 7 μm 8 μm 9μm

Length of cantilever beam ( μm)

Width of cantilever beam = 1 μm

Thickness of cantilever beam = 10 nm

10 μm

The frequency measurement is limited by thermo-mechanical noise on the cantilever beam.

Minimum Detectable Frequency, Δf,min =

Minimum Detectable Mass, Δm,min =

kB = Boltzmann constantT = Temperature in KelvinB = Bandwidth measurement, (= 2πf0/2Q)

Q can increase by 100X by driving the cantilevers

Minimum Detectable Mass

kQTBkf

AB

π21 0

43

4541k

mQ

TBkA

effB

Virus Mass Range

Q=5

Q=500

Bacteria

DNA bp ~ 10-21 g100kDa Protein ~ 10-19 g

Roukes, et. al.

59

Detection of B. anthracis Spore Mass in Air

• Assume equal number of cells on top and bottom of cantilever, non-uniform sizes

• Average mass of a spore in air is 367 fgcompared to expected 300 to 500 fg

Frequency change after binding of spores

Cantilever Dimensions: Length = 50 - 20μmWidth = 9μmThickness = 200nm

Loaded resonant frequency = 658 kHz Unloaded resonant

frequency = 664kHz

In air

Sensitivity - 20um long cantilevers

y = 3.6548x - 1.6627R2 = 0.9236

0

5

10

15

20

25

30

0 2 4 6 8 10

Number of B.anthracis spores

Dow

nwar

d Fr

eque

ncy

Shift

(k

Hz)

60A. Gupta, D. Akin, R. Bashir, Applied Physics Letters, 2004.L. Johnson, A. Gupta, A. Ghafoor, D. Akin, R. Bashir, Sensors and Actuators, 2006

Mass Detection of Virus Particles

f0 = 1.27 MHzf1 = 1.21 MHz

Q ~ 5k = 0.006 N/m

Δf=60kHz

• Ultimate Sensitivity ~ 6.3 Hz/ag (in air)• Min. Detectable Mass (Q = 5, Δf = 40kHz) ~ 6.3fg• Min. Detectable Mass (Q = 500, Δf = 400Hz) ~ 50ag• Work on going to integrated concentration elements• Integrate Abs on surface – in humid air !

Work featured in Popular Science, EE Times, Work featured in Popular Science, EE Times, Nature News, and CNBC TV ShowNature News, and CNBC TV Show

61

Effects of Viscous Medium

• Resonant frequency decreases• Quality factor decreases• Sensitivity decreases

Unloaded air resonant

frequency = 548.6 kHzUnloaded fluid resonant

frequency = 121.3 kHzQ = 1.8

Q = 36.3

Q =

4μπρb2 + Γr ωR( )

Γi ωR( )

– μ = mass per unit length– ρ = density of the medium– b = width of the cantilever– Γr(ωR) and Γi(ωR) are the real and imaginary parts of the hydrodynamic function Γ(ω), with ωR as the resonant frequency

Y. Martin et al., J. Appl. Phys., 61, 1987T.Braun et al., Phys. Rev. E 72, 2005T. R. Albrecht et al., J. Appl. Phys., 69, 1991

01*

2 c l

kfm mπ

=+

– k = spring constant– mc = mass of cantilever– ml = mass of liquid

62

Detection of B. anthracis Spore Mass in Fluid

• Cantilever length = 20 μmwidth = 9 μm andthickness = 200 nm

• Average measured antibody mass was 0.1 pg

• Average mass of spore was 1.85 pg

• Expected mass between 1 to 2 pg per spore

• Thermal noise driven sensitivity ~ 50-60 spores (~100pg)

• Driving the cantilevers can reduce this down to ~ 10-20pg

Sensitivity - 20μm length

y = 0.1989x - 0.8316R2 = 0.82

0

5

10

15

20

25

30

35

40

0 20 40 60 80 100 120 140 160

No. of B.anthracis spores

Freq

uenc

y sh

ift (k

Hz)

63

Final Remarks• Integration of various materials and functions on a chip• Concentration, specific capture, growth, lysing detection• Applications & fundamental biophysics• High end label free sensors that are top down fabricated

• Integrated Biochips : Sample Rapid Results• Micro-environments for single cell, cell-cell analysis, cellular

surgery and manipulation• Implantable devices for integrated diagnostic and therapeutic

devices

64

Thank You

65

Target Concentrations and Detection Limits

• Food Safety:– FDA has zero tolerance: <1cfu/100-250ml for some pathogens (e.g.

Listeria Monocytogenes)– Various limits from 1cfu/ml to 100cfu/ml at different points in

manufacturing facilities

• Water & Pharmaceuticals (USP):– WFO (Water for Operation), Bacterial Action limit <500 cfu/ml– PW (Purified Water), Bacterial Action limit <100 cfu/ml– WFI (Water for Injection), Bacterial Action Limit <10 cfu/100ml

• Human Diagnostics

66BioVitesse, Inc. Company Confidential

Video Demo

67

Bio-Aerosol Delivery Setup

Cantilever BeamElectrodes

Biological Entities

(e.g. virus)

Cantilever BeamElectrodes

Biological Entities

(e.g. virus)

Electrospray bio-aerosol generator

Vibration isolation table

Laser Doppler vibrometer

Flow channel and output aerosol collection chamber

Cantilever chip

• The electrospray aerosol generator can produce neutralized, monodisperse, nano aerosols including viruses or proteins.

• The flow and chip channel are designed to maximize the particle transport to the chip and optimize flow velocity around the cantilevers to capture viruses with DEP forces.

Flow direction

68

25 mm filter holder25 mm filter holder

47 mm filter holder47 mm filter holder

Air SpaceAir Space

CCR KiT Assembly

Uses membranes in series to process 100 mL hot dog extract into 0.1 to 1 mL sample

Ladsich, et al. USDA Review, 2004

69

Mech/Elect.Detection

DNA, protein

Cantilevers,NanoFETs, Nano-pores

Cell Lysing

Nano-probeArray

Micro-scaleImpedance

Spectroscopy

ViabilityDetection

Conc.Sorting

On-chipDielectro-phoresis

Fluidic Ports

Integrated Systems for Study ofCells and Microorganisms

SelectiveCapture

Ab-based

Capture

“Lab on a Chip” Enabled by BioMEMS and

Bionanotechnology

MEMSFilters

Filters

70

Detection of Listeria Cells

Frequency change after binding of 180 dry bacteria cells

Unloaded Resonant Frequency (85.6 kHz)

Loaded Resonant Frequency (84.7 kHz)

6.5 7 7.5 8 8.5 9.59 10x 104Frequency (Hz)

Ampl

itude

(uni

ts)

10-11

10-10

10-9

10-8

Individual Listeria innocuabacterial cells

10 μm1 μm

Non-specific binding of Listeria innocua bacterial cells to a cantilever beam

Lxm* = (x/L)m

A. Gupta, D. Akin, R. Bashir, JVST-B, 2004.Craighead, et al. APL, 77, 3, 17th July 2001, 450-452

ωo2 = k

m

71

Use of Antibodies

• Spores drift in fluid• Antibodies used to immobilize

spores onto the surface• Bovine serum albumin (BSA) used

to prevent non-specific capture• Antibodies for specific capture

Cantilever

Antibodies

Spores

10μm200X 10μm200X

No antibody

With antibody

Before Measurements After Measurements

72

Sensing Methods in BioChips

Surface Stress Change Detection

Δz

L

t

• Δz = deflection of the free end of the cantilever• L = cantilever length• t = cantilever thickness• E = Young’s modulus• ν = poison’s ratio• Δσ1 change in surface stress on top surface• Δσ2 change in surface stress on bottom surface

( ) ( )21

2 14 σσνΔ−Δ

−⎟⎠⎞

⎜⎝⎛=Δ

Etlz

Surface Stress Change Detection

Δz

L

t

• Δz = deflection of the free end of the cantilever• L = cantilever length• t = cantilever thickness• E = Young’s modulus• ν = poison’s ratio• Δσ1 change in surface stress on top surface• Δσ2 change in surface stress on bottom surface

( ) ( )21

2 14 σσνΔ−Δ

−⎟⎠⎞

⎜⎝⎛=Δ

Etlz

⎟⎟⎠

⎞⎜⎜⎝

⎛−=Δ 22

12

114 off

kmπ

mkf

π21

=

• k = spring constant• m = mass of cantilever• f0 = unloaded resonant frequency• f1 = loaded resonant frequency

Mass Change Detection

⎟⎟⎠

⎞⎜⎜⎝

⎛−=Δ 22

12

114 off

kmπ

mkf

π21

=

• k = spring constant• m = mass of cantilever• f0 = unloaded resonant frequency• f1 = loaded resonant frequency

Mass Change Detection

Mechanical Detection Electrical Detection

Conductometric Detection

Z (interface)Z (bulk)

Measurement electrodes

Measurement Volume

Measurement electrodes

Potentiometric Detection

Source Drain (+ve voltage)

ISFET Reference electrode

Current flow

Capture/sensor layer

Reference electrode

Current flow

Ions or analytes

(b) (c)

Amperometer DetectionReference electrode

Working Electrode

Reference electrode

GlucoseGOD

+ + + +

-

e-

Optical Detection

(a)

TargetProbes

Fluorescence Detection

TargetAntigens

hν1hν2

CaptureProbes

hν1hν2

hν

Bioluminescence

hν

Chemiluminescence

+ + - -

- +

R. Bashr, Advanced Drug Delivery Review, 2004