Monitoring nanostructural transformations

during film formation in waterborne coatings

using variable angle X-ray scattering

APOSTOLOS VAGIAS, GIUSEPPE PORTALE

Zernike Institute for Advanced Materials (ZIAM)

Rijksuniversiteit Groningen (RUG)

International Micro Nano Conference, December 12th 2018

“Barrier property” of coatings

❑ Solvents can unavoidably penetrate through structural defects in

applied materials.

❑ Optimal coatings needed to prevent solvent penetration

Features outlining coating quality

https://www.shutterstock.com/de 1Keddie J. et al., ACS Appl. Mater. Interfaces, 2018, 8 (50)

Structural heterogeneities in coatings

Stratification =

“out-of-plane”

heterogeneities

2

Keddie J. et al.,

ACS Appl. Mater. Interfaces,

2016, 8 (50)

Junker P. et al.,

Nature 2011, 476 (308)

Coffee-stain effect = “in-plane” inhomogeneities

https://i.ytimg.com

❑ Latex =aqueous dispersion of polymer colloids

❑ Challenging to realize uniform films (polymer incompatibility)

❑ Defects reported down to micron-sized range (not in nanoscale)

Keddie, J. L. Film Formation of Latex. Mater. Sci. Eng. R Rep. 1997, 21 (3), 101–170.

SubstrateSubstrate Substrate Substrate

Latex suspension Particle packingPacking of

deformed particlesMechanically stable

film

Water content

Stage 1 Stage 2 Stage 3

Polymer Interdiffusion (T > Tg)

Drying time

Waterborne coatings and latex film formation

3

Possibility for defects

Can probe buried nanoscopic objects at interfaces

Grazing Incidence SAXS (GISAXS)

Surface sensitive for structures

at interfaces (minimum

penetration depth, ξp~10 nm)High scattering intensity ideal

for in-situ study

2-D detectors probe at once

lateral & normal order

http://www.gisaxs.de/theory2.html (Andreas Meyer, Uni Hamburg) 4

z

yx

X-rays

ai Substrate

T

R

af

Spreading

direction

kf (= scattered

X-ray wavevector)

0.10 0.15 0.20

10-2

10-1

100

101

p (

m)

a i (

o)

ac (critical angle

of the polymer)

Depth (ξp)-resolution of GISAXS

5

Substrate (glass)

ki (= incident

X-ray wavevector)

Calculations of ξp (αi)

vs. αi, for polyacrylates

(αc ~0.1°)

Polymer Film

kiki

kf kf

Materials (from DSM)Three different polyacrylic (PA) samples investigated

6

Type PA-662 PA-661 PA-660

Hardness hard (Tg ~80° C)

soft (Tg ~4° C)

2 phases:70 % (wt)

soft / 30 % (wt) hard

Solids (%) 39.1 39.1 39.0

pH 8.1 6.8 7.0Particle size (nm), DLS

100 ±3 107 ±1 106±3

PA-662

Non- film forming coating (TFFT < Tg)

PA-660

Film forming coating (TFFT > Tg)

Synchrotron X-rays

Dutch-Belgian Beam Line

DUBBLE-BM26B

(ESRF, Grenoble)

Coating preparation and beamlines

7

MINA: Multipurpose

X-ray Instrument for

Nanostructure Analysis

(University of Groningen)

(T): Transmission(R): Refraction(S): Reflection

-0.15 -0.10 -0.05 0.00 0.05 0.10 0.15

I(q

y)

(a.u

.)qy (nm-1)

(ai =0.1o)

GISAXS data analysis

8

qy*= 2·π/d* (=0.08 nm -1)

Intensity cuts I(qy) vs. qy at qz=constant

kf

ki αi

z

yx

d* (~80 nm) : heterogeneity spacing

Depth-resolved GISAXS (PA-660)

❑(Strong) Scattering is lost at ξp ~15 μm

❑Crossover transition at ξp ~10 μm (𝛼𝑖 = 0.15°) from

structure (heterogeneities) to structure-less region

𝛼𝑖= 0° 𝛼𝑖 = 0.1°(ξp ~0.5 μm)

𝛼𝑖 = 0.2°(ξp ~15 μm)

𝛼𝑖 = 0.35°(ξp ~ 28 μm)

𝛼𝑖 = 0.15°(ξp ~10 μm)

𝛼𝑖 = 0.5°(ξp ~ 42 μm)

9

Tqy

qz RRT T

𝛼𝑖

RT

R

T

S

SR

T

S

S

Vagias A. et al, submitted to ACS Applied Polymer Materials (under review)

Depth-resolved GISAXS patterns

10

PA-661

PA-660

❑(Strong) Scattering

lost at ξp ~10-15 μm

❑Top film layer with

heterogeneities

❑Lower film layer:

no heterogeneities

❑PA-662: Control case

with strong contrast (non

film-forming) PA-662

Vagias A. et al, submitted to ACS Applied Polymer Materials (under review)

GISAXS I(qy) vs. qy cuts: proposed nanostructures

11

Sideview

-0.30 -0.15 0.00 0.15 0.300.0

2.0x10-4

4.0x10-4

6.0x10-4

8.0x10-4

I(q

y)

(a.u

.)

qy (nm-1)

ai=0.10o

ai=0.15o

ai=0.21o

ai=0.37o

glass (ai=0.37o)

increasing ai

q*

PA-660 (multiphase) PA-661 (soft) PA-662 (hard)

-0.45 -0.30 -0.15 0.00 0.15 0.30 0.45

10-3

10-2

10-1

q*

q*

I(q

y)

(counts

/s)

qy (nm-1)

ai=0.10o

ai=0.17o

ai=0.29o

ai=0.43o

ai=0.43o (S)

increasing ai

q*

-0.30 -0.15 0.00 0.15 0.300.0

2.0x10-4

4.0x10-4

6.0x10-4

8.0x10-4

I(q

y)

(a.u

.)

qy (nm-1)

ai=0.1o

ai=0.16o

ai=0.22o

ai=0.37o

glass (ai=0.37o)

increasing ai

q*

Quantifying GISAXS analysis:

nanostructural heterogeneities

12

Sideview

Topview

d*(~80 nm)

d*(~80 nm)

ddomain(~5-10 nm)ddomain(~20 nm)

PA-660 (multiphase) PA-661 (soft) PA-662 (hard)

ddomain(~60 nm)

Vagias A. et al, submitted to ACS Applied Polymer Materials (under review)

0 5 10 15 20 25 30 35 40

0.02

0.04

0.06

Film

weig

ht

(g)

time (min)

In-situ GISAXS (𝛼𝑖 = 0.15°):PA-660

At tdrying =4’ : “colloidal suspension”-like (4’)

At tdrying > 9’: partially coalesced particles

Insets: time after slot die coating (tdrying)

13

T

qz

qy

t drying= 4’ 8’

RT

S

9’

RT

S

20’

RT

S

z

yx

Macroscopic drying kinetics

Vagias A. et al. (in preparation)

0 2 4 6 872

76

80

84

88 PA-660

PA-661

PA-662

d* (

nm

)

tdrying (min)

Nanostructural transformations during drying

❑ (d*) decreases with tdrying more strongly for PA-660

❑ Nanostructural evolution during drying can be quantified 14

0 5 10 15 2010-6

10-5

10-4

10-3 PA-660

PA-661

PA-662

<a

ve

rag

e p

ixe

l in

ten

sit

y (

a.u

.)>

tdrying (min)

Vagias A. et al. (in preparation)

Conclusions GISAXS is an optimal technique to probe:

❑Rich unexplored nanostructural transformations during film drying

❑The quality of (dry) waterborne coatings at the nanoscale

(submicrometric heterogeneities)

❑depth-dependent stratification in the films

15

Perspectives

To expand the successful ex-situ/ in-situ options:

❑ couple with laser speckle imaging (structure-dynamics) (Prof. J. Sprakel, WUR) / environmental effects (airflow, humidity, aging)

❑ different chemistries (also polyurethanes, blends) / substrates (metal, wood) successfully assessed

❑ Biomedicine: polyhydroxyurethane

(PHU-QAS) dental coatings

(UMCG, P. Sharma group) 16

Untreated Ion-exchanged

Acknowledgements

• DPI for funding INCOAT (914ft16)

• DSM Coating Resins (sample synthesis, cross-sectional AFM)

• D. Hermida-Merino (DUBBLE BM26B@ESRF)

• Gert H. ten Brink (RUG)

• ALL OF YOU for your attention! 17

Monitoring nanostructural transformations

during film formation in waterborne coatings

using variable angle X-ray scattering

APOSTOLOS VAGIAS, GIUSEPPE PORTALE

Zernike Institute for Advanced Materials (ZIAM)

Rijksuniversiteit Groningen (RUG)

18

SAXS in suspensions: NP shape/size

19

SAXS modeling

𝐼 𝑞 = Δ𝜌2𝑉න𝑅1

𝑅2

𝑆 𝑞 𝑃 𝑞, 𝑅 𝐷 𝑅 𝑑𝑅

❑ Polydisperse ensemble of spheres

❑ S(q): local monodisperse approximation

20

10-3

10-2

10-1

100

0.1 1

0

0.05

0.1

0.2

0.4

I(q

)

q (nm-1

)

q-4

~q-4

Dense colloidal suspensions

S(q) simulations for liquid-

like systems at different

volume fractions φ

The peak (q*=2·π/d*) accounts

for spatial correlation between

particles (=spacing d*)

𝐼 𝑞 =𝑁

𝑉𝑉𝑝𝑎𝑟𝑡2 𝑃 𝑞 𝑆(𝑞)

q (nm-1)

(φ)

q*

d*

21

SAXS structure factor (S(q))

0.1 1

10-5

10-4

10-3

10-2

10-1

40% wt

I(q

)

q (nm-1)

5% wt

1% wt

water

dilution

SAXS in PA suspensions

❑ PA-660: (dSAXS~96nm, dDLS~106 nm)

PA-660

qmin· RNP= 4.48

(RNP , NP radius)

q* (from S(q))

22

SAXS in aqueous PA-660 suspensions

23

Appendix-

Modeling of SAXS curves in suspension

0.1 0.2 0.3 0.4 0.5 0.60.70.810-5

10-4

10-3

10-2

10-1

100

101

I*(q

) (c

ou

nts

/s)

q (nm-1)

PA-660

PA-661

PA-662

PA-661, PA-662: Polydisperse particle model

PA-660: (Polydisperse) Core-shell particle model

Monodisperse model

❑ Particle radii: PA661 (45nm) > PA662 (41nm) > PA660 (40nm)

❑ Very different nanostructure morphology in films vs. suspensions

for PA660 and PA661. PA662 shows absence of deformability!

-0.6 -0.4 -0.2 0.0 0.2 0.4 0.610-6

10-5

10-4

10-3

10-2

10-1

I(q

y)

(a.u

.)

qy (nm-1)

PA660

PA661

PA662

10-8

10-7

10-6

10-5

10-4

10-3

10-2

10-1GISAXS

PA660

PA661

PA662

SAXS

0.1 0.2 0.3 0.410-5

10-4

10-3

10-2

10-1

SAXS in aqueous suspensions

q*

q*

PA660

PA661

PA662

I* (q)

(counts

/s)

q (nm-1)

q*

Suspensions (left) vs. coatings (right)

24

25

Appendix- Features on GISAXS patterns

Depth-resolved GISAXS (PA systems)

❑ Peak intensity: PA-662> PA-660 > PA-661

❑ Different form factor and nature of scattering in hard PA-662 coating 26

-0.30 -0.15 0.00 0.15 0.300.0

2.0x10-4

4.0x10-4

6.0x10-4

8.0x10-4

I(q

y)

(a.u

.)

qy (nm-1)

ai=0.10o

ai=0.15o

ai=0.21o

ai=0.37o

glass (ai=0.37o)

increasing ai

q*

-0.30 -0.15 0.00 0.15 0.300.0

2.0x10-4

4.0x10-4

6.0x10-4

8.0x10-4

I(q

y)

(a.u

.)qy (nm-1)

ai=0.1o

ai=0.16o

ai=0.22o

ai=0.37o

glass (ai=0.37o)

increasing ai

q*

-0.30 -0.15 0.00 0.15 0.30

10-3

10-2

10-1

q*

q*

I(q

y)

(a.u

.)

qy (nm-1)

ai=0.10o

ai=0.17o

ai=0.29o

ai=0.43o

increasing ai

q*

PA-660 PA-661 PA-662

Depth-resolved GISAXS (PA-660)

❑(Strong) Scattering is lost at ξp ~10-15 μm

❑Top layer with heterogeneities 27

0.1 0.2 0.3 0.4

75

80

85

90

95

(a)

30m, PA660

7m, PA660

30m, PA661

d*(a

i) (

nm

)

ai (o)

0.1 0.2 0.3 0.4

0.0

2.0x10-6

4.0x10-6

6.0x10-6

8.0x10-6

0.0

2.0x10-5

4.0x10-5

6.0x10-5

8.0x10-5

30um, PA660

7um, PA660 30um, PA661

A (a

i) (

a.u

.)ai (

o)

I pix

el (a

i) (

co

un

t/ [

pix

els])

(b)

Cross-sectional AFM (at DSM)

Cross-sectional AFM corroborates stratification!

Rq = 3.0 nm

1 μm

Rq = 7.5 nmRq = 3.0 nmglass air

5 μm

28

GISAXS simulations

29

Randomly packed

cylinder model

Randomly packed

hard sphere model

-0.4 -0.2 0.0 0.2 0.40.0

2.0x10-2

4.0x10-2

6.0x10-2

8.0x10-2

1.0x10-1

qy (nm-1)

PA662

fit

(a)

-0.4 -0.2 0.0 0.2 0.40.0

5.0x10-5

1.0x10-4

1.5x10-4

2.0x10-4

(b)

qy (nm-1)

PA661

fit

-0.4 -0.2 0.0 0.2 0.40.0

1.0x10-4

2.0x10-4

3.0x10-4

4.0x10-4

5.0x10-4

6.0x10-4

(c)

qy (nm-1)

PA660

fit

Randomly packed

cylinder model

Object shape I(0) Dy (nm) dy h

PA662 (hard) Sphere 1.1 x 10-4 60 70 0.45

PA661 (soft) Cylinder 3.0 x 10-5 8 72 0.40

PA660 (multiphase) Cylinder 1.0 x 10-7 16 70 0.40

The local monodisperse approximation was used to take into account for the object size polydispersity described by

a log-normal distribution function. The structure factor was in all cases a Percus-Yevick function.9 The fits were

achieved using the FitGISAXS program.10

-0.4 -0.2 0.0 0.2 0.4

1000

2000

3000

4000

5000

I(q

y)

(a.u

.)

qy (nm-1)

fresh

aged (1 year)

T-annealed

after 1year aging

❑ Tannealing (= 150 °C) > Tg,DMTA,hard (~105 °C)

❑ Annealing smears out nanostructural

heterogeneities

Aging effects (𝛼𝑖 = 0.1°)

30

Fresh, non-annealedAged, non-annealed Aged, annealed

❑ T-annealing at 150 °C> Tg,DMTA,hard (~105 °C)

❑ T-annealing smears out the nanostructure

(heterogeneities evidenced by GISAXS)

Edge effects (𝛼𝑖 = 0.2°)

qz

31

qy