21. September 2016

Mitg

lied

der H

elm

holtz

-Gem

eins

chaf

t

How do hydrogen bonds influence thermophoresis?

| Simone Wiegand,

Doreen Niether, Jan K.G. Dhont

21/09/16 Folie 2

Thermophoresis – the effect

(…, thermodiffusion, Soret effect) 

No microscopic understanding

21/09/16 Folie 3

Thermophoresis – the effect

3

D - diffusion coefficient, w - concentration,DT - thermodiffusion coeff.,

Steady state

– flux,T – temperature,ST Soret coefficient

21/09/16 Folie 4

Mass effect: animationcold molecules hot molecules

cold side hot side

“kinetic gas model”higher momentum transfer from the warm side

Enrichment of the heavy particles on the cold side

21/09/16 Folie 5

Thermophoresis: Where is it used?

Application examples:  “Characterization of Soft Matter”Thermal field flow fractionation

SW., Introduction to thermal gradient related effects, in Functional Soft Matter, J.K.G. Dhont, et al., Editors. 2015, Forschungszentrum Jülich: Jülich. p. F4.1-F4.24.

21/09/16 Folie 6

Thermophoresis: Where is it used?

Mic

rosc

ale

Ther

mop

hore

sis:

Te

chno

logy

and

App

licat

ions

//Nan

oTem

perG

MB

H

Application examples:  “Biochemical reactions”Microscale thermophoresis

21/09/16 Folie 7

Hydrogen bonds: temperature effect

At low temperatures: minimization of the free energy 

F = U – TSby forming hydrogen bonds (ΔU<0).

water goes to the cold side 

At high temperatures: minimization of the free energy 

F = U – TS by entropy production (ΔS>0).

water goes to the warm side 

[Wang, Z., H. Kriegs, and SW J. Phys. Chem. B, 116 (2012) 7463.]

Assuming local thermodynamic equilibrium

21/09/16 Folie 8

Hydrogen bonds: temperature effect

[Kishikawa, Y., SW, and R. Kita, Biomacromolecules, 11 (2010) 740]

Many, but not all aqueoussystems show a similartemperature dependence

[Iacopini et al., Eur. Phys. J. E, 19(2006) 59]

0 10 20 30 40 50 60-0.1

0.0

0.1

0.2

PEO/we

lysozyme

DNA

DNA

(a)

14

PEO/water

DNA

-lactoglobine

6

dextran

dextrandextran

pullulan

temperature / °C

ST /

K-1

ST < 0

ST > 0

21/09/16 Folie 9

Validity of the empirical formula?

T

0T 1 expS T TST

T

A. Königer, et al., Philos. Mag., 89 (2009) 907.

ethanol/water

10 20 30 40

-5

0

5

10

0.051 0.100 0.152 0.1983 0.2512 0.3016 0.3538 0.3987 0.4998 0.5921

ST /

10-3K-1

temperature /°C

w =

[O. Gereben, Journal of Molecular Liquids, 211 (2015) 812-820]

• “pure water rings are formed”• clumping of like molecules

20 mol % ethanol

Breaks down at low concentrations when the homogeneity of the mixture at the molecular level is an issue.

waterrich

ethanolrich

21/09/16 Folie 10

Systematic study of amides

Urea Formamide Acetamide N-Methyl-formamide

N,N-Dimethyl-formamide

More hydrophilic

22 2 2

Why amides? “.. serve as model of the peptide bond “ [Y. Lei et al. JPC A, 107 (2003) 1574]

21/09/16 Folie 11

Temperature dependence

• at low concentrations (w 0.3):

• more flexible fit function neededto describe T-dependence athigher concentrations:

urea in water22

T

0T 1 expS T TST

T

TT expS T a bS T

[Story and Turner, Faraday Trans., 65 (1969) 1810]

20 40 600

1

2

3

w=0.5 w=0.4 w=0.3 w=0.2 w=0.1 w=0.05 w=0.02

ST /

10-3K

-1

temperature /°C

w=0.226 w=0.154 w=0.092 w=0.025

21/09/16 Folie 12

Temperature dependence

• at low concentrations (w < 0.2):

• more flexible fit function neededto describe T-dependence athigher concentrations:

0 20 40 600

1

2

3 conc. (w.f.) 0.2

0.02 0.3 0.05 0.5 0.1 0.7

ST /

10-3K

-1

temperature /°C

formamide in water

T

0T 1 expS T TST

T

2

TT expS T a bS T

[Niether, Afanasenkau, Dhont, SW, PNAS, 113(2016) 4272]

21/09/16 Folie 13

0 20 40 600

1

2

3

conc. (w.f.) 0.2

0.02 0.3 0.05 0.5 0.1 0.7

ST /

10-3K-1

temperature /°C

Structural explanationMolecular dynamic simulations[Elola & Ladanyi, JCP 125,(2006) 184506]

suggest the following picture:

conc. = ?

slope ST > 0 slope ST < 0

low w

only FA-Whydrogen bonds

higher w

also FA-FAhydrogen bonds

21/09/16 Folie 14

POSTER – P02-057

A way to achieve sufficiently high formamideconcentrations to form prebiotic nucleobases under early earth conditions

by

Doreen Niether

21/09/16 Folie 15

„log P“ a „Scale bar“ for hydrogen bonding strength?

octanolunionizedwater

[ ]log log( )[ ]

solutePsolute

Hydrophilic compound: log P < 0octanol

water

octanolwater

Hydrophobic compound: log P > 0

Marvin 16.5.2.0, 2016, ChemAxon (http://www.chemaxon.com)G. Klopman et al. J Chem Inf Comp Sci, 34 (1994) 752-781.V. N. Viswanadhan et al. J Chem Inf Comp Sci, 29 (1989) 163-172.

urea

21/09/16 Folie 16

„log P“ a „Scale bar“ for hydrogen bonding strength?

-1.30 -1.13 -1.03 -0.89 -0.64

More hydrophilic

22 2 2

Marvin 16.5.2.0, 2016, ChemAxon (http://www.chemaxon.com)

Urea Formamide Acetamide N-Methyl-formamide

N,N-Dimethyl-formamide

Log P =

21/09/16 Folie 17

„log P“ a „Scale bar“ for polar solvents ?

-10 -5 0 5-1.5

-1.0

-0.5

0.0

log

P

slope ST(T) / 10-5K-210 20 30 40 50 60 700

2

4

6

8

Ethanol (Königer) Dimethylformamide Acetamide Methylformamide Urea Formamide

ST /

K-1

T / °C

5wt% in water

O

NH2Temperaturedependence of STis correlated withlog P

[Königer, A

., et al., Philos. M

ag., 89(2009) 907]

Low concentration

21/09/16 Folie 18

Comparison: low and high concentration

urea

formamide

acetamide

NMF

DMF

0 2 4 6ST (50wt% ) ST ( 5wt% ) / 10-3K-1

@ 10°C

hydrophilic systems:increasing concentration:solute becomes more thermophobic

hydrophobic systems:increasing concentration:solute becomes more thermophilic

21/09/16 Folie 19

„log p“ scales ST change with concentration

-1.5 -1.0 -0.5 0.0 0.5-1.5

-1.0

-0.5

0.0

log

P

ST(50wt%-5wt%) /10-2K-1

O

NH2

21/09/16 Folie 20

„log p“ scales ST inrespect to c and T

correlation between log P and the change of ST with… concentration… temperature

-10 -5 0 5-1.5

-1.0

-0.5

0.0

log

Pslope ST(T) / 10-5K-2

-1.5 -1.0 -0.5 0.0 0.5-1.5

-1.0

-0.5

0.0

log

P

ST(50wt%-5wt%) /10-2K-1

O

NH2

21/09/16 Folie 21

Take home message

T

0T 1 expS T TST

T breaks down at low wdue to inhomogeneities

Log P correlates with temperature dependence of ST

Log P correlates with concentration change of ST

breaks down at high w

Thermophoresis issensitive to changes of the hydration layer

21/09/16 Folie 22

Thanks to many people and …

… thank you for your attention

FZ Jülich Jan Dhont‘s group(ICS-3)

Rio Kita‘s labKazuya EguchiTokai University, Japan

Fernando Bresme‘s groupSilvia di LecceImperial College London, GB

21/09/16 Folie 23

21/09/16 Folie 24

How do we measure?

IR-TDFRS – InfraRed -Thermal Diffusion Forced Rayleigh Scattering

Measured quantity:Intensity of the diffracted beam

[SW

et al., J. Phys. C

hem. B

, 111(2007) 14169]

Advantages:• small T • no fluorescent labeling

required• wide molecular range Disadvantages:• buffer solutions: difficult• colloids >100 nm: difficult.

Typical gradients: 1K/m

21/09/16 Folie 25

Formamide vs. N-methylformamide

O

NH2

[A. K. H. Weiss, et al. PCCP, 13 (2011) 12173]

“Whereas formamide is almost encaged by the oxygen density, the influence of the methyl group disrupts this pattern rigorously”

21/09/16 Folie 26

Dimethylformamide/water

[Lei, Y., et al., JPC A, 107(2003) 1574Vasudevan, V. and S.H. Mushrif, J. Mol. Liq., 206(2015) 338 ]

“The increases in the peaks of RDFs between water molecules are not so much caused by an increase in the structure of water as they are by the tendency of water to remain in aggregates in the mixtures.”

doubts about the Force field ?

21/09/16 Folie 27

20 40 60-10123456 Dimethylformamide

5 wt% 50 wt%

Methylformamide 5 wt% 50 wt%

Acetamide 5 wt% 50 wt%

Formamide 5 wt% 50 wt%

Urea 5 wt% 50 wt%

ST /

10-3

K-1

T / °C

Comparison: low and high concentration

21/09/16 Folie 28

Principle Microscale thermophoresisS

W., Introduction to therm

al gradient related effects, in Functional Soft M

atter, J.K.G

. D

hont, et al., Editors. 2015, Forschungszentrum

Jülich: Jülich. p. F4.1-F4.24.