Bio-materials characterization with NIRAnna Sandak & Jakub Sandak

2

Outline

Why spectroscopy?

Tendencies in biomaterials developmnet

NIR in bio-materials characterization

Trends in NIR applications

3

Goals

• to exploit the potential of near infrared spectroscopy

• to demonstrate its capabilities for bio-based materials characterization

• to highlight the potential of the NIR as a tool capable of providing complementary information to other techniques

• to present current trends in NIR developments

4

Bio-materials in construction sector

• In Italy 1 on 12 buildings is made of wood and growing tendency is observed nowadays

• Bio-based materials are often used for retrofitting of existing structures, upward construction or vertical gardens

• Buildings that use bio-materials are not just sustainable, strong and durable; they are also beautiful

5

Development priorities• Structural components

(need for developed wood products -Engineered Wood Products, high strength wood, moisture resistant sills, light-weight beams/joists/studs of bio-composites, sandwich panels for exterior walls)

• Insulation(need for compactable bats of cellulose insulation, environmentally friendly fire impregnation, high-performance insulation that provides thinner walls, insulation, optimized for soundproofing)

• Barrier Materials(need for bio-based wind and vapor barrier for moisture-proof exterior walls, waterproofing for wet areas, façade and roofing materials with improved durability/serviceability

Source: Per-Erik Eriksson: Future sustainable biobased buildings

6

Durability and performance

7

Surface properties

roughness

color

brighntess

surface free energy

wetability

gloss

pattern

temperaturesoftiness

hardness

touch experience

facture

waviness

chemical activity

contamination

resistance to dirt

outlook

resistance to scratch

resistance to abrassion

adhesion

soft feel

aseptic

durability

leveling

water repellency

matting effect

8

How to assess properties?

9

Why NIR spectroscopy? No need special sample preparation Non-destructive testing Relatively fast measurement No residues/solvents to waste Determination of many components simultaneously High degree of precision and accuracy Direct measurement with very low cost

Not self standing technology Overlapping of spectral peaks Not straightforward interpretation Needs statistics methods for data analysis

10

chemical industry, microanalysis, pharmaceutical analysis, soil,

polymers, food and beverages, surface science, fuels, textile

industry, art conservation, forensics, PAT, remote sensing

and also wood and paper industry

NIR applications

11

…but what can we see in wood?

http://www.ifb.ethz.ch/research/sinergia

12

Cellulose (40-50%)

• linear polymer• long chain

(DP 5000-10000)• Β-D-Glucopyranose

units• 1,4-glycosidic bonds• hydroxyl groups

bonds• different kind in

wood: • - amorphous,

- semi-crystalline, - crystalline

methylene groupshydroxyl groups

13

Lignin (20-30%)

Softwood (guaiacyl units) Hardwood (guaiacyl + syringyl units)

methoxyl groups

phenolic hydroxyl groups

aldehyde groups carbonyl groups

14

Hemicellulose (25-35%)

xylan

• hetero-polysaccharides• short chain

(DP 150-200)• amorphous• high reactivity• various composition

for hardwood and softwood

acetyl groups

hydroxyl groups

15

NIR & biomaterials characterization

• Chemical compositions(cellulose, lignin, extractives)

• Physical and anatomical characterization(density, moisture, calorific value, microfibryl angle, defects detection, sapwood/heartwood detection)

• Mechanical properties(compression, tension, MOE, MOR)

• Identification and monitoring others properties of wood (decay monitoring, weathering, waterlogging, thermal treatment, chemical modifications)

• Paper industry(weight, thickness, moisture, Kappa number, mechanical resistance, type of pulp/paper)

16

Virgin wood

• Species recognition• Wood provenance• Exotic wood evaluation• Biomass characterization• Clones recognition

17

Recognition of wood species

1 2 1 3 1 94 1 151 61 7 1 81 91 10 11 121 13 1 9141 9151 161 171 181 191 20

softwood hardwood exotic wood

01- Picea abies, 02-Abies alba, 03-Pinus cembra, 04-Larix decidua, 05-Pinus sylvestris, 06-Pinus ponderosa, 07-Tsuga, 08- Thuja sp., 09-Quercus robur, 10-Robinia pseudoacacia, 11-Castanea sativa, 12-Iroko, 13-Afrormosia, 14-Okumè, 15-Sipo, 16-Meranti, 17-Wengè, 18-Azobè, 19-Ipè, 20-Itauba, 21-Pau Rosado, 22-Teck, 23-Bangkirai

18

Use of CA for species classification

Picea abies Pinus cembra Larix sp. Thuja plicata Abies alba Pinus silvestris Tsuga sp.

inve

stig

ated

sam

ple

19

Wood provenance & NIRS

2163 trees of Norway spruce from 75 location

in 14 European countries2163 samples measured

x 5 spectra/sample = 10815 spectra

20

Differentation of origin

samples form 14 countries (75 locations); totally 10815 spectra was analyzed and evaluated

Austria ▲ Bulgaria ▼ Czech Republic ■ Croatia ■ Estonia Finland France Germany Hungary Italy Norway Poland Romania Sweden

21

Italy

TRENTO

Moena

Cavalese

Garda lake

22

Principal Component Analysis

PCA of spectra of wooden samples coming from six different locations (14 sites) in Trentino region

Note: second derivative, vector normalization, method: factorization (5 factors)region: 9590-5250cm-1, 5100-4160cm-1

Strada Alta Valcigolera Strada Orti Forestali, Juribello Paneveggio Cadino 06 Cadino 43 Cadino 49 Folgaria 04 Folgaria 22 Monte Sover 7 Monte Sover 11 Monte Sover 14 Monte Sover 69 Caoria Valsorda

23

Cluster Analysis

CADINO CADINO CADINO

CA – wooden samples from 3 different districts of Cadino valley Note: second derivative, vector normalization, Ward’s algorithm, region: 7089-5072cm-1

24

coconut shell tamarind grass

Identification of agricultural residues

Particleboards characterization

25

A

B

C

Raw resources: Eastern redcedar (Juniperus virginiana L.) Single- and three-layers panelsmanufactured from raw material using 9% urea formaldehyde, combination of 15% modified corn starch and 2% urea formaldehyde adhesive

12345678910111213141516171819

-0,008

0

0,008

40004500500055006000650070007500wavenumber (cm-1)

PC

2, P

C3

PC2PC3

26

Selection of biomass for optimal bio-conversion process

PCA of willow clones. Note: pre-processing: 2nd derivative + vector normalization,

17 smoothing points, region: 6869-5847 cm-1, method: factorization, 3 factors

duotursprint

turbo

wodtur

monoturstart

oltur

tur PC3

PC4

PC2

27

Willows utilization

willows

thermo - chemicalconversion chemical

conversionmechanical-chemical

conversion

combustion pyrolysis gasification fermentationanaerobic digestion estrificationpulping

panel production

heat syngas pulp&paper fiber-panel ethanol biogas biodieselcharcoalsyngas bio-oil

high lignin content high carbohydrate contenthigh fiber content

extraction

high extractives content acumulation of heavy metalsfast resprouting

manufacturing environmental engineering

phytominigphyto- remediation biofiltrationfurniturebasketry

metalsclean soilclean waterchairscontainers

cosmetic foodpharmaceutic

cream anti- phlogistic anti-

pyretic preservativeanti-

oxidantshampoo

cultivation

ap

icu

ltu

re ho

ne

y p

olli

na

tio

n construction

gre

en

a

rch

itec

ture

fen

ce

sro

ad

ma

ts

fast growingearly flowering

28

Suggested conversion paththermo-chemical

mechano-chemical chemical pharmaceut

ical phytoremediation

#1 0.93

0.07

0.13

0.96

0.19

-0.10

0.00

#2 0.98

0.04

-0.01

0.98

0.44

-0.03

0.02

#3 0.97

0.13

0.01

1.02

0.57

0.88

0.33

#4 1.05

-0.07

1.04

0.99

0.61

0.99

0.20

#5 0.89

0.13

0.19

0.80

0.57

-0.06

0.19

#6 0.05

-0.05

-0.13

1.14

0.68

0.15

0.42

#7 -0.08

0.78

0.02

0.03

0.33

-0.07

0.50

#8 0.09

1.10

-0.07

-0.11

0.06

0.00

-0.03

#9 -0.05

0.90

-0.06

0.02

0.28

0.15

0.30

#10 -0.03

0.04

0.07

0.03

0.09

0.14

-0.11

#11 -0.02

-0.05

1.01

-0.06

1.19

0.84

0.31

#12 -0.04

-0.07

1.05

-0.04

1.32

-0.15

0.22

#13 -0.02

-0.04

0.93

0.08

1.22

0.32

0.35

#14 -0.04

-0.04

0.92

0.10

1.00

0.86

0.66

#15 1.11

0.95

0.84

0.00

1.04

0.09

0.11

#16 0.16

1.12

-0.02

0.06

1.31

1.12

0.60

#17 0.06

0.06

1.07

0.01

1.10

-0.14

-0.05

high content of: high accumulation of: Discrimination rule high HHV

cellulose hemicelluose extractives Zn Pb Cu

Thresholdrule HHVexp>20.0 Xc>38% Xhe>34% Xe>11% XZn>55ppb XPb>8ppb XCu>8ppb

Average prediction

error 0.06 0.08 0.07 0.06 0.29 0.11 0.29 http://www.seco.cpa.state.tx.us/publications/renewenergy/biom

assenergy.php

29

Chemical analysis

• cellulose (by Browning)

• holocellulose (by Browning)

• lignin (T 222 om-06 TAPPI)

• pentosans (T 223 cm-01 TAPPI)

• hot water extractives (T 207 cm-08 TAPPI)

• cold water extractives (T 207 cm-08 TAPPI)

• 1% NaOH extractives (T 212 om-07 TAPPI)

• organic solvent extractives (T 204 cm-07 TAPPI)

30

NIR PLS calibration models #1of willows chemical composition (pilot study)

22

23

24

25

26

27

28

22 23 24 25 26 27 28

lignin measured (%)

ligni

n pr

edic

ted

(%)

42

43

44

45

46

47

48

49

50

51

52

42 43 44 45 46 47 48 49 50 51 52

cellulose measured (%)

cellu

lose

pre

dict

ed (%

)

17

18

19

20

21

17 18 19 20 21

pentosans measured (%)

pent

osan

spr

edic

ted

(%)

70

71

72

73

74

75

76

77

78

79

80

70 71 72 73 74 75 76 77 78 79 80

holocellulose measured (%)

holo

cellu

lose

pred

icte

d (%

)ligninr2 = 0.969RMSECV = 0.24RPD = 5.7

celluloser2 = 0.9 82RMSECV = 0. 39RPD = 7.4

pentosansr2 = 0.96 7RMSECV = 0.11RPD = 5.5

holocelluloser2 = 0.968RMSECV = 0.33RPD = 5.6

31

NIR PLS calibration models #2of willows chemical composition (pilot study)

18

19

20

21

22

23

24

25

26

27

28

18 19 20 21 22 23 24 25 26 27 28soluble in 10% NaOH measured (%)

solu

ble

in 1

0% N

aOH

pred

icte

d (%

)

2

3

4

5

6

2 3 4 5 6soluble in organic measured (%)

solu

ble

in o

rgan

ic p

redi

cted

(%)

0

1

2

3

4

0 1 2 3 4soluble in cold water measured (%)

solu

ble

in c

old

wat

er p

redi

cted

(%)

0

1

2

3

4

5

6

7

0 1 2 3 4 5 6 7soluble in hot water measured (%)

solu

ble

in h

ot w

ater

pre

dict

ed (%

)

soluble in cold waterr2 = 0.983RMSECV = 0.20RPD = 7.7

soluble in hot waterr2 = 0.983RMSECV = 0.12RPD = 7.6

soluble in organic solventsr2 = 0.981RMSECV = 0.12RPD = 7.2

soluble in 10% NaOHr2 = 0.965RMSECV = 0.37RPD = 5.4

18

19

20

21

22

23

24

25

26

27

28

18 19 20 21 22 23 24 25 26 27 28soluble in 10% NaOH measured (%)

solu

ble

in 1

0% N

aOH

pred

icte

d (%

)

18

19

20

21

22

23

24

25

26

27

28

18 19 20 21 22 23 24 25 26 27 28soluble in 10% NaOH measured (%)

solu

ble

in 1

0% N

aOH

pred

icte

d (%

)

2

3

4

5

6

2 3 4 5 6soluble in organic measured (%)

solu

ble

in o

rgan

ic p

redi

cted

(%)

2

3

4

5

6

2 3 4 5 6soluble in organic measured (%)

solu

ble

in o

rgan

ic p

redi

cted

(%)

0

1

2

3

4

0 1 2 3 4soluble in cold water measured (%)

solu

ble

in c

old

wat

er p

redi

cted

(%)

0

1

2

3

4

5

6

7

0 1 2 3 4 5 6 7soluble in hot water measured (%)

solu

ble

in h

ot w

ater

pre

dict

ed (%

)

soluble in cold waterr2 = 0.983RMSECV = 0.20RPD = 7.7

soluble in hot waterr2 = 0.983RMSECV = 0.12RPD = 7.6

soluble in organic solventsr2 = 0.981RMSECV = 0.12RPD = 7.2

soluble in 10% NaOHr2 = 0.965RMSECV = 0.37RPD = 5.4

0

1

2

3

4

0 1 2 3 4soluble in cold water measured (%)

solu

ble

in c

old

wat

er p

redi

cted

(%)

0

1

2

3

4

0 1 2 3 4soluble in cold water measured (%)

solu

ble

in c

old

wat

er p

redi

cted

(%)

0

1

2

3

4

5

6

7

0 1 2 3 4 5 6 7soluble in hot water measured (%)

solu

ble

in h

ot w

ater

pre

dict

ed (%

)

0

1

2

3

4

5

6

7

0 1 2 3 4 5 6 7soluble in hot water measured (%)

solu

ble

in h

ot w

ater

pre

dict

ed (%

)

soluble in cold waterr2 = 0.983RMSECV = 0.20RPD = 7.7

soluble in hot waterr2 = 0.983RMSECV = 0.12RPD = 7.6

soluble in organic solventsr2 = 0.981RMSECV = 0.12RPD = 7.2

soluble in 10% NaOHr2 = 0.965RMSECV = 0.37RPD = 5.4

32

Wood modifications & NIRS

• Thermal treatment• Decay • ISPM-15 • Densification• Mechanical testing• Wood after wood machining• Coatings• Weathering• Service life prediction• …

33

Wood thermal treatment

fir spruce larchparameter ML EMC ML EMC ML EMC

range (cm-1) 7502-6098 5450-4246

6102-4246 10996-10098 9303-8501 5018-4435

8505-4435 9403-5446 9403-7498 6102-5446

pre-processing 1der +MSC 1 der + VN VN 1 der + VN 1 der + VN 1 der +MSC r2 98.67 97.58 98.48 98.83 97.88 97.69

RMSECV 0.26 0.25 0.25 0.17 0.39 0.19 RPD 8.66 6.42 8.11 9.42 6.87 6.58 rank 3 3 3 4 4 5

Termovuoto: open system: (volatile products of the process are continuously removed from the kiln by means of a vacuum pump)

The treatment time: 3 hoursPressure: 0.2 bartreatment temperatures of 160, 180, 200 and 220ºC.

+5

hard

woo

ds…

195 samples measured x 5 spectra/sample = 975 spectra

34

New spectra presentationWild cherry

(Prunus avium)

no treated 160ºC 180ºC 200ºC 220ºC

-1,25

-1

-0,75

-0,5

-0,25

0am. cell OH

cell CH CH2cell CH2

cell CO OH CH2

cell OH

cell OH CH CH2

cell OH CO

cryst. cell OH

cryst. cell CH

cryst. cell OH CH

cell hemi CH

cell hemi CH CH2 OH CO

cell hemi OH CHcell hemi lig CH

extr. CHhemi C=O

hemi CH

hemi CH

hemi OH

hemi CH

hemi lig extr. CH C=C C=O

lig CH

lig CH

lig CH

lig

lig OH

H2OH2O

xylograms

35

Xylograms; different species

-1,25

-1

-0,75

-0,5

-0,25

0am. cell OH

cell CH CH2cell CH2

cell CO OH CH2

cell OH

cell OH CH CH2

cell OH CO

cryst. cell OH

cryst. cell CH

cryst. cell OH CH

cell hemi CH

cell hemi CH CH2 OH CO

cell hemi OH CHcell hemi lig CH

extr. CHhemi C=O

hemi CH

hemi CH

hemi OH

hemi CH

hemi lig extr. CH C=C C=O

lig CH

lig CH

lig CH

lig

lig OH

H2OH2O

-1,25

-1

-0,75

-0,5

-0,25

0am. cell OH

cell CH CH2cell CH2

cell CO OH CH2

cell OH

cell OH CH CH2

cell OH CO

cryst. cell OH

cryst. cell CH

cryst. cell OH CH

cell hemi CH

cell hemi CH CH2 OH CO

cell hemi OH CHcell hemi lig CH

extr. CHhemi C=O

hemi CH

hemi CH

hemi OH

hemi CH

hemi lig extr. CH C=C C=O

lig CH

lig CH

lig CH

lig

lig OH

H2OH2O

-1,25

-1

-0,75

-0,5

-0,25

0am. cell OH

cell CH CH2cell CH2

cell CO OH CH2

cell OH

cell OH CH CH2

cell OH CO

cryst. cell OH

cryst. cell CH

cryst. cell OH CH

cell hemi CH

cell hemi CH CH2 OH CO

cell hemi OH CHcell hemi lig CH

extr. CHhemi C=O

hemi CH

hemi CH

hemi OH

hemi CH

hemi lig extr. CH C=C C=O

lig CH

lig CH

lig CH

lig

lig OH

H2OH2O

-1,25

-1

-0,75

-0,5

-0,25

0am. cell OH

cell CH CH2cell CH2

cell CO OH CH2

cell OH

cell OH CH CH2

cell OH CO

cryst. cell OH

cryst. cell CH

cryst. cell OH CH

cell hemi CH

cell hemi CH CH2 OH CO

cell hemi OH CHcell hemi lig CH

extr. CHhemi C=O

hemi CH

hemi CH

hemi OH

hemi CH

hemi lig extr. CH C=C C=O

lig CH

lig CH

lig CH

lig

lig OH

H2OH2O

Black locust (Robinia pseudoacacia)

no treated 160ºC 180ºC 200ºC 220ºC

Norway spruce (Picea abies)

no treated 160ºC 180ºC 200ºC 220ºC

Silver fir (Abies alba)

no treated 160ºC 180ºC 200ºC 220ºC

European larch (Larix decidua)

no treated 160ºC 180ºC 200ºC 220ºC

Wild cherry (Prunus avium)

no treated 160ºC 180ºC 200ºC 220ºC

European beech (Fagus silvatica)

no treated 160ºC 180ºC 200ºC 220ºC

European ash (Fraxinus excelsior)

no treated 160ºC 180ºC 200ºC 220ºC

-1,25

-1

-0,75

-0,5

-0,25

0am. cell OH

cell CH CH2cell CH2

cell CO OH CH2

cell OH

cell OH CH CH2

cell OH CO

cryst. cell OH

cryst. cell CH

cryst. cell OH CH

cell hemi CH

cell hemi CH CH2 OH CO

cell hemi OH CHcell hemi lig CH

extr. CHhemi C=O

hemi CH

hemi CH

hemi OH

hemi CH

hemi lig extr. CH C=C C=O

lig CH

lig CH

lig CH

lig

lig OH

H2OH2O

-1,25

-1

-0,75

-0,5

-0,25

0am. cell OH

cell CH CH2cell CH2

cell CO OH CH2

cell OH

cell OH CH CH2

cell OH CO

cryst. cell OH

cryst. cell CH

cryst. cell OH CH

cell hemi CH

cell hemi CH CH2 OH CO

cell hemi OH CHcell hemi lig CH

extr. CHhemi C=O

hemi CH

hemi CH

hemi OH

hemi CH

hemi lig extr. CH C=C C=O

lig CH

lig CH

lig CH

lig

lig OH

H2OH2O

-1,25

-1

-0,75

-0,5

-0,25

0am. cell OH

cell CH CH2cell CH2

cell CO OH CH2

cell OH

cell OH CH CH2

cell OH CO

cryst. cell OH

cryst. cell CH

cryst. cell OH CH

cell hemi CH

cell hemi CH CH2 OH CO

cell hemi OH CHcell hemi lig CH

extr. CHhemi C=O

hemi CH

hemi CH

hemi OH

hemi CH

hemi lig extr. CH C=C C=O

lig CH

lig CH

lig CH

lig

lig OH

H2OH2O

-1,25

-1

-0,75

-0,5

-0,25

0am. cell OH

cell CH CH2cell CH2

cell CO OH CH2

cell OH

cell OH CH CH2

cell OH CO

cryst. cell OH

cryst. cell CH

cryst. cell OH CH

cell hemi CH

cell hemi CH CH2 OH CO

cell hemi OH CHcell hemi lig CH

extr. CHhemi C=O

hemi CH

hemi CH

hemi OH

hemi CH

hemi lig extr. CH C=C C=O

lig CH

lig CH

lig CH

lig

lig OH

H2OH2O

Sessile oak (Quercus peraea)

no treated 160ºC 180ºC 200ºC 220ºC

36

PLS models for TM softwoods

 y = 0.9522x - 0.1186

R2 = 0.944

-10

-8

-6

-4

-2

0

2

-10 -8 -6 -4 -2 0 2reference

estim

ated

y = 0.9782x + 0.218R2 = 0.9662

6

7

8

9

10

11

12

13

6 7 8 9 10 11 12 13reference

estim

ated

Mass loss Equilibrium Moisture Content

37

TM of poplar veneers

78910111213141516171819

-0,00001

0,000005

54505950645069507450wavenumber (cm-1)

2nd d

er o

f NIR

abs

orba

nce

NT239ºC, 1h, 250mb239ºC, 2h, 1000mb240ºC, 6.6h, 250mb

#26

#0

#37

#20

#18

#9

#15

#0 NT #26 155ºC, 1h, 250mb #9 174ºC, 22h, 250mb #20 210ºC, 1h, 250mb #18 239ºC, 1h, 250mb #15 239ºC, 2h, 1000mb #37 240ºC, 6.6h, 250mb

38

Wood sterilization

How to detect if the wood was appropriately thermally treated?

• standard ISPM No15;• Temperature: 56°C in the core of wood• Treatment time: 30minutes

39

ISPM-15Low temperature thermal treatment of wood

35

40

45

50

55

60

65

70

75

80

85

90

95

100

105

35 40 45 50 55 60 65 70 75 80 85 90 95 100 105

treatment temperature, degC

pred

icte

d te

mpe

ratu

re, d

egC

calibrationvalidation

• to investigate an effect of the temperature on wood samples of different wood species:

• Spruce (softwood with resin)• Fir (softwood without resin)• Poplar (hardwood)

• different treatment times:• 0.5 hours (30 minutes)• 1 hour• 3 hours• 6 hours

• different treatment temperatures: • 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95,

100°C• three samples for repetition/statistics• five-times spectra measurements for analysis

of weathering effect:• Conditioned after treatment• Measured after 1month• Measured after 3 months• Measured after 6 mounts• Measured after 12 mounts

468 samples measured x 5 spectra/sample x 5 repetition/sample = 9360 spectra

40

NIR-based method for verification of the ISPM-15 treatment*

chan

ges

to th

e sp

ectra

ch

ange

s to

the

spec

tra

NO TREATED

TREATED

Spectra +

model

41

891011121314

-0,000015

-0,000005

0,000005

55005600570058005900600061006200630064006500wavenumber (cm-1)

2nd d

er o

f abs

orba

nce

spruce referenceConiophora puteanaTrametes versicolor

Wood decay a c b

Trametes versicolor Postia placenta Coniophora puteana

y = 0.95x + 1.97R2 = 0.93

-10

0

10

20

30

40

50

60

-10 0 10 20 30 40 50 60reference mass loss, %

mas

s lo

ss p

redi

cted

with

NIR

, %

42

Decay type identification

Postia placenta

reference

Trametes versicolor

Coniphora puteana

Gloeophyllum trabeum

43

Archaeological wood

Q1 Q

2Q3Q4

Q5 Q

6

700-2700 years old

44

Short term waterloggingO

AK

band assessment pine oak nr wave number (cm-1) wood component functional group peat water peat water

1 4195 lignin not assigned ○ ○ ◦ 2 4268 cellulose CH, CH2 ○ ○ ○ ○ 3 4401 cellulose, hemicelluloses CH, CH2, OH, CO ○ ○ ○ 4 4546 lignin CH, C=O ◦ 5 4608 cellulose, hemicelluloses not assigned ◦ 6 4686 hemicelluloses, lignin, extractives CH, C=C, C=O 7 4739 cellulose OH ○ ○ ○ ○ 8 4808 cellulose semi-crystalline and crystalline OH, CH ○ ○ ○ 9 5051 water OH ○ ○ 10 5198 water OH center of the range ○ ○ ○ ○ 11 5245 hemicelluloses C=O 12 5495 cellulose OH, CO ◦ ◦ 13 5593 cellulose semi-crystalline and crystalline CH ◦ ○ ◦ 14 5666 not assigned CH, CH2 ◦ ○ 15 5692 not assigned CH2 16 5800 hemicelluloses (furanose / pyranose) CH ○ ○ 17 5865 hemicelluloses CH ○ ○ ◦ ○ 18 5935 lignin CH ◦ ○ ○ ◦ 19 5980 lignin CH ○ ○ 20 6126 cellulose OH ◦ ○ 21 6286 cellulose crystalline OH ◦ ◦ ◦ ◦ 22 6334 cellulose OH 23 6472 cellulose crystalline OH ◦ ◦ 24 6715 cellulose semi-crystalline OH ◦ ◦ ◦ ◦ 25 7003 amorphous cellulose, water OH ◦ ◦ ◦ ◦ 26 7092 lignin, extractives OH ○

360 samples measured x 3 spectra/sample = 1080

spectra

peat water years of exposition

0 2 4 6 8 2 4 6 8 arch

PIN

E

45

Paper & NIR

5% wheat

2nd day

7th day

14th day

5% rye 5% wheat 5% rye

Chaetomium globosum Aspergillus niger, Penicillium funiculosum, Trichoderma viride

5% wheat

2nd day

7th day

14th day

5% rye 5% wheat 5% rye

Chaetomium globosum Aspergillus niger, Penicillium funiculosum, Trichoderma viride

3240 samples measured x 3 spectra/sample =9720 spectra

46

Effect of soil type

-0,00004

-0,00003

-0,00002

-0,00001

0

0,00001

0,00002

0,00003

4100 4200 4300 4400 4500 4600 4700 4800 4900 5000 5100 5200 5300 5400 5500

wavenumber, cm-1

2nd d

eriv

ativ

e of

abs

orba

nce

0 weeksforest soil 8 weeksagricultural soil 8 weekssand 8 weeks

biodegradation of the recycled paper with addition of 5% wheat bran

47

Effect of time

biodegradation of the recycled paper with addition of 5% wheat bran in the forest soil

0

4

8

PCA analysis of NIR spectra of paper before degradation tests and decomposed in forest soil for

4 and 8 weeks. Note: spectral range: 11000-4150cm-1,

pre-processing: 2nd derivative + VN

-0,00004

-0,00003

-0,00002

-0,00001

0

0,00001

0,00002

0,00003

4100 4200 4300 4400 4500 4600 4700 4800 4900 5000 5100 5200 5300 5400 5500

wavenumber, cm-1

2nd d

eriv

ativ

e of

abs

orba

nce

0 week1 week8 week

48

PLS models for paper samples

samozerwalosc

y = 0.76x + 0.77R2 = 0.80

-1

0

1

2

3

4

5

6

7

-1 0 1 2 3 4 5 6 7

reference value (km)

estim

ated

val

ue (k

m)

porost

y = 0.75x + 0.80R2 = 0.80

0

1

2

3

0 1 2 3

reference value (3-0 scale)

estim

ated

val

ue (3

-0 s

cale

)

breaking length growth degree

49

Estimation of mechanical stressy = 0,90x + 68,91

R2 = 0,90

0

200

400

600

800

1000

1200

1400

1600

0 200 400 600 800 1000 1200 1400 1600reference force (N)

pred

icte

d fo

rce

(N)

stresses

50

why does it work?

wavenumber, cm-1

wavenumber, cm-1 wavenumbe

r, cm-1

wavenum

ber, cm-1

wavenumber, cm-1 wavenumber, cm-1

wav

enum

ber,

cm-1

wav

enum

ber,

cm-1

a) c)

b) d) in the plastic field in the elastic field

2D synchronous spectra during wood tension

40624193420042314266428142854362440145444605463246834737477947944806504951965234524254625493552055935616566256895774579357975813586358715898593259485959599460026125618762836333644964686519661966576700671167386754676967896796687369436974700170897213729873137321735274108158824786528748no

n tre

ated

norm

aliz

efir

st d

eriva

tive_

5fir

st d

eriva

tive_

15fir

st d

eriva

tive_

25fir

st d

eriva

tive

+ no

rmal

izat

ion_

5fir

st d

eriva

tive

+ no

rmal

izat

ion_

15fir

st d

eriva

tive

+ no

rmal

izat

ion_

25se

cond

der

ivativ

e_5

seco

nd d

eriva

tive_

15se

cond

der

ivativ

e_25

seco

nd d

eriva

tive

+ no

rmal

izat

ion_

5se

cond

der

ivativ

e +

norm

aliz

atio

n_15

seco

nd d

eriva

tive

+ no

rmal

izat

ion_

25

Determination coefficients (r2)

between NIR spectra amplitude and tension stresses in a relation

to the spectra preprocessing method

51

Identity test – coating recognition

Preprocessing: 2 derivative + vector normalization, 9 smoothing points

Method: factorization

Regions (cm-1): 4135-4350, 5365-5520, 5800-6000, 6290-6480, 7000-7200

Water solvent based products: #13, #26, #41

Organic solvent based products: #11, #12, #42

52

band: 4200-10000cm-1, 2 derivative, 5 smoothing points, vector normalization, 2 factors

Identification of coatings

■ Resotec ● Nordwal ■ Solas ▼ Adler aquawood ● Adler pullex ▲ Adler pullex oil

Product #1

Product #2

Product #3

Product #4

Product #5

Product #6

53

natural oak waxed oak varnished oak

Fale

jow

ka o

ak

Antique wooden floors

54

Monitoring of weathering

8 softwood, 3 hardwood, 12 exotic, 3 thermo modified wood 21 various water/organic based products with synthetic/natural/acrylic/alkyd resins/oils

55

Monitoring wood weathering

-0.000006

0

55005600570058005900600061006200630064006500wavenumber (cm-1)

2nd d

er o

f abs

orba

nce

reference1 year2 years3 years4 years

-OH

cr

ista

lline

cel

ulos

e

-CH

se

mi-c

rista

lline

cel

ulos

e

-OH

cr

ista

lline

cel

ulos

e

-CH

lig

nin

-CH

lig

nin

-CH

he

mic

ellu

lose

-CH

al

ipha

tic c

hain

s

56

Short term weathering of wood

891011121314

-0,000002

0

0,000002

55005600570058005900600061006200630064006500wavenumber (cm-1)

2nd d

er o

f abs

orba

nce

-CH S-cr cell-CH He-CH Al ch-OH Cr cell -CH L -CH S-cr cell-CH He-CH Al ch-OH Cr cell -CH L

28 days

Wind

57

Service life prediction

PCA of NIR spectra of non coated samples exposed to South (4 years of natural weathering); Reggio Emilia Roma Loninghens Macerata Udine Trento Lecce standard

EN

927-

6

EN

927-

6

EN

927-

6

1 ye

ar

2 ye

ars

2 ye

ars

2 ye

ars

1 ye

ar

1 ye

ar

Tim

e m

achi

ne

Tim

e m

achi

ne

Tim

e m

achi

ne

0 ye

ars

0 ye

ars

0 ye

ars

4 ye

ars

4 ye

ars

4 ye

ars

EN

115

07

EN

115

07

EN

115

07

3 ye

ars

3 ye

ars

3 ye

ars

EN

927-

6

EN

927-

6

EN

927-

6

1 ye

ar

2 ye

ars

2 ye

ars

2 ye

ars

1 ye

ar

1 ye

ar

Tim

e m

achi

ne

Tim

e m

achi

ne

Tim

e m

achi

ne

0 ye

ars

0 ye

ars

0 ye

ars

4 ye

ars

4 ye

ars

4 ye

ars

EN

115

07

EN

115

07

EN

115

07

3 ye

ars

3 ye

ars

3 ye

ars

a

58

Research directions

spatial measurementHyperspectral Imaging

in field measurementportable equipment

59

Weathered wood

a) b)

PCA analysis performed on not weathered (a) and sample after 28 days of weathering (b)

0 1 2 4 7 9 11 14 17 21 24 28

60

K-means classification

K‐means clustering on the mosaic of weathered wood after pre‐selecting 3 (a), 4 (b) and 5 classes (c)

b) c) A B C D

E F G H

I J K L I

A B C D

E F G H

J K L

A B C D

E F G H

I J K L

a)

61

NIR for log/biomass quality index in mountain forest

wood measurement at various harvesting moments

62

Wood defects detection

range 12500-5900cm-1

2nd derivative + VN

63

Calibration transfery = 0.93x + 1.86

R² = 0.93

25,0

27,0

29,0

31,0

33,0

25,0 27,0 29,0 31,0 33,0

Measured total lignin content (%)

NIR‐predicted

 total lignin conten

t (%)

y = 0.93x + 1.86R² = 0.93

y = 0,99x + 0,31R2 = 0,86

25,0

27,0

29,0

31,0

33,0

25,0 27,0 29,0 31,0 33,0

Measured total lignin content (%)

NIR‐predicted

 total lignin conten

t (%)

measurement of samples with equipment #1

measurement of selected samples with equipment #2equipment #2 model 

development based on samples measured with instrument #1

model #1 development

calculation of transfer matrix

selection of calibration samples by Kennard Stone 

algorithm

64

ConclussionsNIR technique has been successfully used for:• species and provenance recognition• estimation chemical composition, physical and mechanical

properties• monitoring chemical changes in wood during thermal modification,

weathering, waterlogging and decay• recognition and classification of coatings • characterization of archeological wood and cultural heritage objects• service life prediction of wooden elements and structures• paper characterization• in-field applications

Current developments in the fields of optics and electronics open new application for NIR. Its non-destructive character and simple

measurements allows assisting experts in the estimation of material properties in a fast and repeatable way

65Thank you very much