University of CyprusBiomedical Imaging and Applied OpticsBiomedical Imaging and Applied Optics

Raman SpectroscopyRaman Spectroscopy

Introduction

• Professor Sir C.V. RamanProfessor Sir C.V. RamanA New Type of Secondary RadiationC. V. Raman and K. S. Krishnan,

The Nobel Prize in Physics 1930"for his work on the scattering of light and for the

Nature, 121(3048): 501-502, March 31, 1928

1888-1970

g gdiscovery of the effect named after him"

First photographed Raman spectra

2Bangalore, India

Introduction

• The Raman Effect Inelastic Scatteringhνhνi

h(νi-νR)3

( i R)3

2

1

hνi0

S1

Virtual LevelInelastic Scattering

3ergy

• Energy transferred from incident light to molecular vibrations

2

1

0

Ene• Emitted light has

decreased energy (λi<λR)

3

S0Rayleigh (elastic) Raman (inelastic) Scattering Scattering

( i R)

difference in energy

Introduction

• Some Vibrations in BenzeneSome Vibrations in BenzeneBreathingChubby CheckerKekule

3

4

5

D C

ount

s)

1

2

3

ensi

ty (C

CD

4

400 600 800 1000 1200 1400 1600 18000

Raman Shift (cm-1)

Inte

Introduction

• Raman Spectrum of CholesterolRaman Spectrum of Cholesterol

5Hanlon et al. “Prospects for in vivo Raman spectroscopy,” Phys Med Biol 45: R1 (2000)

Introduction

• Raman Spectra pFingerprinting a Molecule

• Raman spectra are molecule specificS t t i• Spectra contain information about vibrational modes of the molecule

• Spectra have sharp f t ll ifeatures, allowing identification of the molecule by its spectrum

6

y pExamples of analytes found in blood

which are quantifiable with Raman spectroscopy

Introduction

Evolution of Raman Spectroscopy• 1928~1960

• Minor experimental advances

1960

• Late 1980s→1990s• Biomedical investigations• Advanced dispersive• 1960

• Invention of laser expands scope experiments

Advanced dispersive spectrometers

• 2000 →p p

• 1980s: rapid technological advances

• In vivo application• Optical fiber probes

Non linear spectroscopy• Fourier Transform

spectroscopy• Charge Coupled Device

• Non-linear spectroscopy

Charge Coupled Device (CCD) detectors

• Holographic and dielectric filters

7

filters• Near-Infrared (NIR) lasers

Introduction

• Applications of Raman SpectroscopyApplications of Raman Spectroscopy• Structural chemistry

Solid state• Solid state• Analytical chemistry• Applied materials analysis• Process controlProcess control• Microspectroscopy/imaging

Environmental monitoring• Environmental monitoring• Biomedical

8

Introduction

History of Biological Raman Spectroscopy• 1970: Lord and Yu record 1st protein spectrum from

lysozyme using HeNe excitation• Evolution to NIR excitation

• Decreased fluorescence, Increased penetration (mm)• 1980s:

• FT Raman with Nd:YAG and cooled InGaAs detectors (long collection times (30 min))(long collection times (30 min))

• Clarke (1987-1988): visible excitation of arterial calcium hydroxyapatite and carotenoids

• 1990s, advances in:• Lasers, Detectors, Dispersive spectrometers, Filters

Ch i9

• Chemometrics

Introduction

• Diagnostic Advantages of Raman SpectroscopyDiagnostic Advantages of Raman Spectroscopy• Wavelength selection (from UV to IR)

No biops req ired• No biopsy required• Directly measures molecules

• Small concentrations• Chemical composition• Morphological analysis

• Quantitative analysisy• In vivo diagnosis

10

Classical Raman Physics

• Interaction between electric field of incidentInteraction between electric field of incident photon and molecule• Electric field oscillating with incident frequency f :• Electric field oscillating with incident frequency fi:

0 cos(2 )i iE E f tπ=• Induces molecular electric dipole (p):

p Eα=• Proportional to molecular polarizability, α

• ease with which the electron cloud around a molecule can be distorted

• Polarization results in nuclear displacement( )

11

( )0 cos 2 Rq q tπν=

Classical Raman Physics

• For small distortions, polarizability is linearlyFor small distortions, polarizability is linearly proportional to the displacement

α⎛ ⎞∂0 0

0

...qqαα α

⎛ ⎞∂= + +⎜ ⎟∂⎝ ⎠

• Resultant dipole:( )cos 2p E E tα α πν= = +

Rayleigh Scattering

( )0 0 cos 2 ip E E tα α πν= = +

( ) ( ){ }0 01 cos 2 cos 22 i R i RE q t t

qα π ν ν π ν ν

⎛ ⎞∂ ⎡ ⎤ ⎡ ⎤+ + −⎜ ⎟ ⎣ ⎦ ⎣ ⎦∂⎝ ⎠

k RStokes Raman

{ }02 q∂⎝ ⎠

12

Anti-Stokes Raman

Photo-Molecular Interactions

1003

2 Rayleigh

Scattering

40

60

80

nten

sity

2

1

0n1 Stokes

-2000 -1000 0 1000 20000

20

1

I1

2

1

Anti-StokesStokes

Raman Shift (cm-1)

Ene

rgy

Virtual Levels

1

0n’1

ΔE=hνR

2

1

0

13

Auto- IR Rayleigh Stokes Anti-Stokes NIRFluorescence Absorption Scattering Raman Scattering Fluorescence

n0

Classical Raman Physics

• Raman scattering occurs only when the molecule is ‘polarizable’

• Raman intensity ∝ f4

0dqα∂≠

• Raman intensity ∝ f• Classical dipole radiation• Stokes shifted Raman is more intense than anti-Stokes by Boltzmann factor:

4 hf⎛ ⎞4

Rhfi RA kT

S i R

f fI eI f f

−⎛ ⎞+= ⎜ ⎟−⎝ ⎠

• Consistent with other scattering phenomena, often reported in terms of cross-section (σ [cm2]), or probability of scattering:

I I lσρ=

• ρ: density of molecules• l: pathlength

0I I lσρ=

14

p g

Characteristics of Raman Scattering

• Very weak effect7• Only 1 in 107 photons is Raman scattered

• NIR elastic scattering in tissue:• NIR absorption in tissue:

'1/ 1s mmμ ≈

1/ 10a cmμ ≈NIR absorption in tissue:• Red absorption in tissue or water:• Raman scattering in tissue or water:

1/ 10a cmμ1/ 5a mμ ≈

1/ 3R kmμ ≈• True scattering process

• Virtual state is a short-lived distortion of the electron cloud which creates molecular vibrationswhich creates molecular vibrations

• τ < 10-14 s• Strong Raman scatterers have distributed electron clouds

• C=C• π-bonds

15

Units & Dimensional Analysis

• Spectroscopic frequencies reported inSpectroscopic frequencies reported in wavenumbers [cm-1], proportional to transition energy :energy :

1E fν = = = fλE hf=

R f i i d d t f it ti

hc c λ c fλ=

• Raman frequencies are independent of excitation wavelength and reported as shifts• Wavenumbers relative to excitation frequency:

1 1Rν = −

16

Ri R

νλ λ

Units & Dimensional Analysis

• ExampleExample• NIR excitation at 830 nm: 12,048 cm-1

T pical Raman shift 1000 cm 1Rν• Typical Raman shift: ~1000 cm-1

• λR = 905 nm

Rν

• Sharp biological Raman linewidths ~10 cm-1

FWHM• ΔλR= 0.69 nm

1717

UV, Visible, and NIR Excitation

• Wavelength Selectionf• Raman signals have a constant shift

can vary excitation wavelength and get same information

• UV• + Resonance enhanced• + λR<λF filter fluorescence• - photo damage, low penetration

• Visible• + Raman ∝λ-4 ↑I vs IR• + Raman ∝λ 4 ↑IR vs. IR• - fluorescence overlaps with

Raman signal

NIR• NIR: • + Low fluorescence • + Deep penetration

18

• - Raman ∝λ-4 ↓IR vs. Vis

UV, Visible, and NIR Excitation

Applications• UVRR

• Biological macromolecules: nucleic acids, proteins, lipids• Organelles, cells, micro-organisms, bacteria, phytoplanktonOrganelles, cells, micro organisms, bacteria, phytoplankton

neurotoxins, viruses• Clinically limited: photomutagenicity

• Visible• Visible• Cells (minimal fluorescence)• DNA in chromosomes, pigment in granulocytes and lymphocytes,

RBCs hepatocytesRBCs, hepatocytes• First artery studies: hydroxyapatite and carotenoids (Clarke 1987,

1988)NIR• NIR

• Hirschfeld & Chase, 1986: FT-Raman• Tissue: artery, cervix, skin, breast, blood, GI, esophagus, brain tumor,

Al h i ’ t t b

19

Alzheimer’s, prostate, bone

UV, Visible, and NIR Excitation

• Spectroscopic 3

2

1p pAdvantages of NIR Raman

N ib ti lVisible

1

0n1

nerg

y

Virtual Level• Narrow vibrational

bands are chemical specific and rich in

Melanin 3

2

E

information• Freedom to choose

excitation wavelength

1

0n0Fluorescence Raman (inelastic)

Scatteringexcitation wavelength• Minimize unwanted

tissue fluorescence

Scattering

104

1061 c

m-1

) H2O VisibleNIR

ExcitationMelanin• optimize sampling depth

• Utilize CCD technology

102

1ε(1

0-3 M

-

HbO

Melanin

20

Utilize CCD technology100 200 500 1000 2 00010-2

Wavelength (nm)

Current Raman Instrumentation

• Laser diodesC t St bl li NIR• Compact, Stable narrow line, NIR

• High throughput spectrographs (f/1.8)• Holographic elements• Holographic elements

• Bandpass filters (eliminates spontaneous emission of lasing medium)• Notch filters (106 rejection of Rayleigh scattered laser line)

L hi hl ffi i t t i i ti• Large area, highly efficient transmission gratings• CCD detectors

• High QE (back-thinned, deep-depletion)g Q (bac t ed, deep dep et o )• Low noise (LN2 cooled)• Multichannel detection

Hi h th h t filt d fib ti b• High throughput, filtered fiber optics probes• NIR FT and scanning PMT systems no longer useful

21

Clinical Raman Systems

830 nmbandpassfilter

shutter

Diode Laser filterDiode Laser

holographic grating

notch filter

22

CCD

Fiber Raman Probe Design

Problems Fiber background NA2Problems

• Fiber backgroundFiber background μ NA2

• Distorts signal• Adds shot-noise

• Low signal collection• Raman effect is weakRaman effect is weak• Tissue is highly

diffusivediffusive

23

Fiber Raman Probe Design• Reduce Fiber Background

• Fiber background produced equally in excitation and collection fibersg p q y• Excitation laser Raman scattered light from tissue• Fiber Raman scattering, transmitted by excitation fiber*• Fiber Raman background elastically scattered from sample (and

collected)• Excitation elastically scattered and gathered by collection fibersy g y• Fiber Raman scattering by collection fibers*

x c* ∝ NA2

Excitation laser

Tissue Ramanx c

From McCreery RL

Tissue Raman

Fiber background

24

Tissue Sample

From McCreery RL “Raman Spectroscopy for Chemical Analysis,” 2000. Tissue Sample

Fiber Raman Probe Design

• Filter Transmission100

R i f I t t80

(%)

Region of Interest

60

mis

sion

(

Collection FilterExcitation Filter

40

Tran

sm

20

25

0 500 1000 15000

Raman Shift (cm-1)

Fiber Raman Probe Design

• Problems • SolutionsProblems

• Fiber background • Micro-optical filters• Short pass excitation• Distorts signal

• Adds shot-noise

• Short-pass excitation filter

• Long-pass collection filter• Low signal collection

• Raman effect is weak

Long pass collection filter

• Optimize optical designRaman effect is weak

• Tissue is highly diffusivedesign

• Characterize distribution of Raman light in tissueof Raman light in tissue

• Define optimal geometry• Design collection optics

26

g p

Fiber Raman Probe Design

Design Goals• Restricted geometry for clinical use

• Total diameter ~2mm for access to coronary arteriesy• Flexible• Able to withstand sterilization

• Designed to work with 830 nm excitation• High throughput• High throughput

• Data accumulation in 1 or 2 seconds• Safe power levelsSafe power levels• SNR similar to open-air optics laboratory system• Accurate application of models

27

pp

Fiber Raman Probe Designcollection fibers

aluminum jacketaluminum jacket

excitation fiber

long-passfilter tube

1 m

m

metal sleevesleeve

0.55short-passfilt d

retainingsleeve 1 75 mm

0.70filter rod

282 mm

sleeve

ball lens

1.75 mmSingle Ring Probe has 15 Fibers

Motz et al. Appl Opt 43: 52 (2004)

Cardiovascular Disease

The Burden of Cardiovascular Disease†The Burden of Cardiovascular Disease†

• 71,300,000 people in United States afflicted

• 910,600 deaths per year• 1 out of every 2 7 deaths• 1 out of every 2.7 deaths

• Coronary artery disease claims 653,000 lives annually

• 1 out of every 5 deaths• Economic cost: greater than $142.5 billion

29†American Heart Association, Heart and Stroke Statistics-2006 Update

Cardiovascular Disease

• Arterial Anatomy

Media

Normal Mildly Atherosclerotic Plaque Ruptured Plaque

Media

Fibrous Cap

LumenT

Atheroma

Cap

IntimaAtheroma

NC

AdventitiaT th b

30

T: thrombusNC: necrotic core

Cardiovascular Disease

• Some Current Challenges in CardiologySome Current Challenges in Cardiology• Evaluation and development of therapeutics

Etiolog of atherosclerosis• Etiology of atherosclerosis• Mechanisms of re-stenosis

• Post-angioplasty• Transplant vasculopathy

• Detection of vulnerable atherosclerotic plaques• Prediction/prevention of cardiac eventsp

31

Cardiovascular Disease

Vulnerable Plaques• Account for majority of sudden cardiac death• Frequently occur in clinically silent vessels q y y

• <50% stenosis• Effective treatments unknown • Characterized by:

• Biochemical changesg• Foam cells• Lipid pool

I fl t ll• Inflammatory cells• Thin fibrous cap (<65 μm)

C tl d t t bl

32

• Currently undetectable

Cardiovascular Disease

Standard Diagnostic Techniquesg q• Angiography

• Severity of stenosis thrombosis dense calcifications• Severity of stenosis, thrombosis, dense calcifications• Provides no biochemical information

A i• Angioscopy• Surface features of plaque, including color

N i f i f b f f• No information of sub-surface features

• Histopathologygy• Biochemical and morphological information• Requires excision of tissue

33

q

Cardiovascular Disease

• Emerging Diagnostic g g gTechniques

• Magnetic resonance imaging

• Intravascular ultrasoundMicrostructure (100 μm)imaging

• External ultrasound• Positron emission

• Microstructure (100 μm)• Optical coherence

tomographyPositron emission tomography

• Electron beam computed tomography

• Microstructure (10 μm) • Fluorescence spectroscopy

• Limited chemicaltomography• Thermography• Elastography

• Limited chemical information

• Broad spectral featuresR S tElastography • Raman Spectroscopy

• Quantitative biochemical information

Non Invasive

34

• Morphological analysisNon-Invasive

Cardiovascular Disease

0 6

0.8

1.0• Raman Spectral Pathology of Atherosclerosis • Ca hydroxyapatite

0.0

0.2

0.4

0.6y y p

• proteins

800 1000 1200 1400 1600 1800

-0.2

0.81.0

.)

• cholesterol• β-carotene

0 20.00.20.40.6

nten

sity

(a.u • β-carotene

• proteins

0.81.0

800 1000 1200 1400 1600 1800-0.4-0.2In

• collagen

-0.20.00.20.40.6 • elastin

• actin

35

800 1000 1200 1400 1600 1800-0.6-0.4

Raman Shift (cm-1)Image from medstat.med.utah.edu/WebPath/webpath.html

Cardiovascular Disease

• Coronary Artery Disease Classification: A Prospective Study

1.0 Normal Artery Non-Calcified Plaque Calcified Plaque

1.0 Normal Artery Non-Calcified Plaque Calcified PlaqueP t t C l ifi ti

0 6

0.8

on 0 6

0.8 Punctate Calcification

on

0.4

0.6

3 σ error zonealci

ficat

io

0.4

0.6

3 σ error zonealci

ficat

io

0.2

Ca

0.2

Ca

0.0 0.1 0.2 0.3 0.4 0.5 0.60.0

0.0 0.1 0.2 0.3 0.4 0.5 0.60.0

36

CholesterolNCR + Lipid CoreNCRCholesterolNCR + Lipid CoreNCR

Buschman HPJ, Motz JT, et al. Cardiovascular Pathology 10(2), 59-68 (2001)

Cardiovascular Disease

Clinical In Vivo Data: Methods• Peripheral vascular surgery

• Femoral bypass• Carotid endarterectomy• Carotid endarterectomy

• Laser power calibration set with Teflon• ~100 mW (82-132mW)100 mW (82 132mW)

• OR room lights turned off as during angioscopy• Spectra collected for a total of 5 secondsSpectra collected for a total of 5 seconds

• 20 accumulations of 0.25s each• Probe held normal to arterial wall

• Analysis of 1s and 5s data • Additional model components: sapphire, epoxy, water, HbO2

37

Cardiovascular Disease

• Clinical In Vivo Data: Calcified Plaque 50 μmClinical In Vivo Data: Calcified Plaque1

DataFitResidual

0.5

.u.)

Residual

0

Inte

nsity

(a

-0.5

800 1000 1200 1400 1600 1800Raman Shift (cm -1 )

38Motz JT et al., J Biomed Opt 11(2): 021003

Cardiovascular Disease

• Clinical In Vivo Data: Ruptured PlaqueClinical In Vivo Data: Ruptured Plaque

0 5

1DataFitR id l

0

0.5

.)

Residual

-0.5

Inte

nsity

(a.u

.

-1.5

-1

0.4 mm

0.1 mm800 1000 1200 1400 1600 1800Raman Shift (cm -1 )

39Motz JT et al., J Biomed Opt 11(2): 021003

Cardiovascular Disease

• Clinical In Vivo Data: Thrombotic PlaqueClinical In Vivo Data: Thrombotic Plaque

0 5

1DataFitResidual

0

0.5

u.)

Residual

0.1 mm-0.5

Inte

nsity

(a.

-1.5

-1

800 1000 1200 1400 1600 1800Raman Shift (cm -1 )

40Motz JT et al., J Biomed Opt 11(2): 021003

Application To Other Diseases

Normal Breast Tissue Malignant Breast Tumor1 1

0 5

1Data Fit Residual

0.5

1Data Fit Residual

0

0.5

ntes

nity

(a.u

.)

-0.5

0

ntes

nity

(a.u

.)

0In

-1

0.5In

800 1000 1200 1400 1600 1800

-0.5

Raman Shift (cm -1)800 1000 1200 1400 1600 1800

-1.5

Raman Shift (cm -1)

100 mW excitation, 1 second collection

41

Frontier Investigations

• High-Wavenumber RamanHigh Wavenumber Raman

Advantages• Higher Raman signal• Lower fluorescence

N fib b k d

Disadvantages• Broader lineshapes• Smaller spectral region

M tl li it d t li id

42www.sigma.com

• No fiber background• Distinguishes cholesterol esters

• Mostly limited to lipids• No calcification signalc

Frontier Investigations

• High-Wavenumber RamanNormal Bladder

High Wavenumber Raman

DNA

Collagen

A ti

Glycogen

Cholesteryl palmitate

Actin High-Wavenumber H&Elp: lamina propriau: urotheliumCholesteryl palmitate

Cholesteryl linoleate

Triolein

43Koljenovic S et al., J Biomed Opt 10(3): 031116 (2005)

Triolein

Conclusions

• Raman spectroscopy ‘fingerprints’ molecules byRaman spectroscopy fingerprints molecules by characterizing interactions between photons and molecular vibrationsmolecular vibrations

• Near-infrared excitation is preferred for bi di l li tibiomedical applications

• Recent optical fiber probe developments allowRecent optical fiber probe developments allow accurate real-time analysis in vivo

N f h i i f• New areas of research are promising for widespread clinical application

4444

References

• McCreery RL. Raman Spectroscopy for Chemical A l i Wil I t i N Y k 2000Analysis. Wiley-Interscience, New York, 2000.

• Ferraro JR, Nakamoto K, and Brown CW. I t d t R S t 2 d dIntroductory Raman Spectroscopy 2nd ed. Academic Press, Boston, 2003.H l EB t l “P t f i i R• Hanlon EB, et al. “Prospects for in vivo Raman spectroscopy,” Phys Med Biol 45: R1-R59 (2000).

• Mahadevan Jensen A and Richards Kortum R• Mahadevan-Jensen A and Richards-Kortum R. “Raman spectroscopy for the detection of cancers and precancers,”J Biomed Opt 1:31-70 (1996).( )

• Utzinger U and Richards-Kortum R. “Fiber optic probes for biomedical spectroscopy,” J Biomed Opt 8 121 147 (2003)

45

8: 121-147 (2003).45