Calorimetric Glass Transition Yuanzheng Yue Wuhan University of - - PowerPoint PPT Presentation

Calorimetric Glass Transition

Yuanzheng Yue

Wuhan University of Technology, China Aalborg University, Denmark

Joint ICTP-IAEA Workshop, Trieste, Italy, Nov. 6-10, 2017

Outline

• Background and motivation
• Case 1: Borosilicate and phosphate glasses

– Dulong Petit Law – The Cp  m relation – Pressure effect on fictive temperature – Structural source of the Cp change – Prediction of Tg by topological model

• Case 2: Hyperquenched (HQ) glasses

– Relaxation in multi-component oxide systems – Relaxation in metallic glasses – Tg of SiO2 – Relaxation in HQ strong glass formers (SiO2 and GeO2)

• Case 3: Mechanically vitrified glasses
• Case 4: Metal-organic framework glasses

– Evidence for polyamorprphism in ZIF-4 – Melting and glass transition of ZIFs – Ultrahigh glass-forming ability of ZIF-62

2

Outline

• Background and motivation
• Case 1: Borosilicate and phosphate glasses

– Dulong Petit Law – The Cp  m relation – Pressure effect on fictive temperature – Structural source of the Cp change – Prediction of Tg by topological model

• Case 2: Hyperquenched (HQ) glasses

– Relaxation in multi-component oxide systems – Relaxation in metallic glasses – Tg of SiO2 – Relaxation in HQ strong glass formers (SiO2 and GeO2)

• Case 3: Mechanically vitrified glasses
• Case 4: Metal-organic framework glasses

– Evidence for polyamorprphism in ZIF-4 – Melting and glass transition of ZIFs – Ultrahigh glass-forming ability of ZIF-62

3

One of the 125 big questions in science (till 2030): What is the nature of the glass transition?

Science, 2005 Numerous models about glass transition are emerging: Macroscopic models (entropy, energy, free volume), Mode- coupling theory, Frustration-based model, Elastic model (harmonic), Local expansion model, Shoving model, Liquid fragility theory, Topological model….

Angell, Science 1995 Debenedetti & Stillinger, Nature 2001 Ediger, Harrowell, J. Chem. Phys. 2012 …..

Here I focus on the calorimetric glass transition.

Calorimetric glass transition is reflected by a sudden change in heat capacity

400 500 600 700 800 900 0.8 1.0 1.2 1.4 1.6 1.8

qh=qc=10 K/min upscan downscan

Tg

Cp (Jg-1K-1)

T (K)

Cp = (dH/dT)p Cv

• Glass transition temperature (Tg) is a dynamic temperature,

measured as the onset temperature of glass-liquid transition.

• Melting temperature (Tm) is a thermodynamic temperature.

Key values for glass transition: Heat capacity (Cpg) and its jump at Tg(Cp) for a normally cooled glass

400 500 600 700 800 900 0.8 1.0 1.2 1.4 1.6 1.8

Cpl Cpg qh=qc=10 K/min

Tg

Cp (Jg

• 1K
• 1)

T (K)

Cp Cp(PO3)2 glass

d

Heat capacity for a hyperquenched glass (rockwool glass at ~106 K/s)

400 600 800 1000 0.6 0.8 1.0 1.2 1.4 1.6

64 J/g

Tg Tc

The hatched area: energy released from 1g fiber

upscan 1 upscan 2

T (K) Cp (Jg-1K-1)

Determination of the glass transition (Tg) and the fictive temperatures (Tf)

Basic equation:

• Y. Z. Yue, et al., Chem. Phys. Lett. 2002; J. Chem. Phys. 2004

 

  

f g eq c

T T pg pl p T T p

dT C C dT C C ) ( ) (

1 2

400 600 800 1000 1200 0.8 1.0 1.2 1.4 1.6 1.8

B A B A Cpg Cpl Cp2 Cp1 = Tg=941 K Tf=1141 K

T (K) Cp (Jg-1K-1)

Cpg = a + bT + c/T2 + d/T0.5

900 950 1000 1050 1100 1150

= Tg=941 K Tf=1141 K

T (K)

Glass transition

Influenced by

• Chemical composition and liquid fragility
• Thermal and mechanical history
• Types and strength of chemical bonds
• Network connectivity
• Topological degree of atomic freedom
• Atomic packing
• Microscopic heterogeneity
• Cluster structure

Outline

• Background and motivation
• Case 1: Borosilicate and phosphate glasses

– Dulong Petit Law – The Cp  m relation – Pressure effect on fictive temperature – Structural source of the Cp change – Prediction of Tg by topological model

• Case 2: Hyperquenched (HQ) glasses

– Relaxation in multi-component oxide systems – Relaxation in metallic glasses – Tg of SiO2 – Relaxation in HQ strong glass formers (SiO2 and GeO2)

• Case 3: Mechanically vitrified glasses
• Case 4: Metal-organic framework glasses

– Evidence for polyamorprphism in ZIF-4 – Melting and glass transition of ZIFs – Ultrahigh glass-forming ability of ZIF-62

10

Dulong Petit Law applies when the unit of Cp is converted to J/mol of atoms?

400 500 600 700 800 900 15 20 25 30 35 40 45

Cp (J mol-1K-1)* T (K)

*Jouls per mole of atoms, not per mole of molecules B2O3 increases

3R

20 40 60 80 23 24 25

Cpg at Tg (Jmol-1K-1)* B2O3 (mol%)

Dulong Petit Law works at Tg

Cp≈3R law works for oxide glasses at Tg

400 600 800 1000 1200 1400 1600 15 20 25 30 35 40 45 50

Cp (J mol-1K-1)* T (K)

*Jouls per mole of atoms

3R

NaPoLi CMP borosilicate basaltic Diopsite

SiO2

35Al2O365SiO2

Cp as a function of composition

100 200 300 400 500 600 50 80 110 140 170

Cp (J mol

• 1 K
• 1)

T (

• C)

75B 63B-12Si 51B-24Si 37B-37Si 24B-51Si 12B-63Si 6B-69Si 75Si

400 450 500 550 600 75 100 125 150 175 Tg Cpg Cpl

20 40 60 80 10 20 30 40 50 60 1.2 1.3 1.4 1.5 1.6 Experiment (Cp) Model (Cp)

Cp (J mol-1 K-1)

[B2O3] (mol%)

Experiment (Cp,l/Cp,g)

Cp,l/Cp,g (-)

Smedskjaer et al. J. Phys. Chem. B. 115 (2011) 12930 Configurational heat capacity (Cp) increases with increasing the B2O3/SiO2

Relation between Cp and kinetic fragility

20 30 40 50 60 10 20 30 40 50 60

Cp (J mol-1 K-1)

m (-)

Implication: There is a link between Cp to the kinetic fragility for the same series of glasses.

) 1 (

0 

 

m m T A p

g

C

The network connectivity increases with increasing B2O3, but the fragility increases.

20 40 60 80 0.0 0.2 0.4 0.6 0.8

NBO/T B2O3 (mol%)

Implication: The network connectivity is not the main controlling factor for liquid fragility.

20 40 60 80 10 20 30 40 50 60 1.2 1.3 1.4 1.5 1.6 Experiment (Cp) Model (Cp)

Cp (J mol-1 K-1)

[B2O3] (mol%)

Experiment (Cp,l/Cp,g)

Cp,l/Cp,g (-)

B2O3 mol% increase

IRO B3 SiO4

• O-

q=1.0 q=0.08

400 600 800 1000 1200 1400 1600

Relative Intensity (A.U.) Wavenumber (cm

• 1)

The IRO band is greatly enhanced by increasing B2O3 content Raman on 75q B2O3 - 75(1-q) SiO2 - 15Na2O - 10CaO q = [B2O3]/([B2O3]+[SiO2])

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

40 44 48 52 56

fragility m Total Area of IRO bands

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 20 30 40 50

Cp,conf (J mol

• 1 K
• 1)

Total Area of IRO Bands

IRO units increase Cp,conf and m

Link between Cp,conf and IRO units

The content of IRO units has a dominant contribution to the evolution of Cp,conf with composition in borate-silicate glasses.

• H. Liu, et al., PCCP, 18 (2016) 10887

Topological model and temperature dependent constraint theory

Phillips & Thorpe:

• Atomic structure of a glass- a network of bond constraints
• Each atom has 3 degrees of freedom, but they are removed by:

Two - body Linear constraints Three – body angular constraints Gupta & Mauro:

• Temperature dependent constraint theory
• network constraints vs. composition

Predicting glass properties, e.g.,Tg , m, Hv

Phillips & Thorpe, Sol .State Commun. (1985) Gupta & Mauro, J. Chem. Phys. (2009)

Se Se Se Se (a) (b) Ge Ge Ge Ge Ge Ge Se Se Se Se Se Se (a) (b) Ge Ge Ge Ge Ge Ge

Type and counting of Constraints

•  type: B-O and MNB-O linear constraints

Two  constraints at each oxygen

• Β type: O-B-O angular constraints

− Five β constraints at each Q4 unit. − Three at each Q3 unit.

• γ type: B-O-B and B-O-M(NB) angular constraints

− One g constraint at each bridging oxygen

• μ type: modifier rigidity (due to clustering)

− Two μ constraints per NBO-forming Na atom

Ranking of Constraints

Each type of constraint has its onset temperatures, which is the temperature where constraints become rigid as temperature is lowered.

T T T T

g   

  

10 20 30 40 1 2 3

O

• Na

+ Na +

Na

+

O T > T T < T < T T < T < T Tg < T < T

Atomic degrees of freedom [Na2O] (mol%)

T < Tg

Cooling

B-O M

NB-O

O B B B O O

• O

Predicting Tg by using temperature dependent constraint theory

Good prediction, but challenge for complex systems

Smedskjaer, Mauro, Sen, Yue, Chem. Mater. 22 (2010) 5358 Smedskjaer, Mauro, Youngman, Hogue, Potuzak, Yue, J. Phys. Chem. B 115 (2011) 12930.

Pressure induced enhancement of the Cp overshoot for CaP2O6 glass

500 600 700 800 900 80 100 120 140 160 180 200

2nd DSC upscan

Cp (Jmol

• 1K
• 1)

T (K)

1st DSC upscan after 500 MPa compression

780 790 800 810 820 830 120 140 160 180 200 220 Cp (Jmol-1K-1) T (K)

P (MPa) a: 500 b: 300 c: 200 d: 100 e: 20 0.1 a b c d e

Why? Pressure drives glass deep in the potential energy

• landscape. When being heated, glass absorb energy

from the surrounding, and hence enhances Cp’.

Thermodynamic consequences of compression and relaxation for CMP

300 400 500 600 700 800 100 200 300 400 500 600 700 0.5 0.6 0.7 0.8 0.9

b) a)

780 800 820 840 100 120 140 160 180

Cpl Cp(T) T2 T1

Hover

Cp (Jmol-1K-1) T (K)

Hover (Jmol-1) Sover (Jmol-1)

P (MPa)

dT C T C H

T T pl p

• ver



  

2 1

) ) ( (

dT S

T T T C T C

• ver

pl p





 

2 1

) ) ( (

Potential energy decreases during compression. Entropy change decreases during compression.

Yue, et al. J. Chem. Phys. 2007

Fictive temperature (Tf) decreases with pressure for CMP

100 200 300 400 500 600 764 768 772 776

700 750 800 850 100 120 140 160 180 A B B A

= Tf

Cp (Jmol-1K-1) T (K)

TfA (K) P (MPa)

Moynihan, et al. J. Am. Ceram. Soc. 1976 Yue, et al. Chem. Phys. Lett. 2002; J. Chem. Phys. 2007

To determine Tf correctly we should apply the correct method: Enthalpy-matching method

Structural relaxation in compressed borate glasses

Smedskjaer et al, Sci. Rep. (2014)

Outline

• Background and motivation
• Case 1: Borosilicate and phosphate glasses

– Dulong Petit Law – The Cp  m relation – Pressure effect on fictive temperature – Structural source of the Cp change – Prediction of Tg by topological model

• Case 2: Hyperquenched (HQ) glasses

– Relaxation in multi-component oxide systems – Relaxation in metallic glasses – Tg of SiO2 – Relaxation in HQ strong glass formers (SiO2 and GeO2)

• Case 3: Mechanically vitrified glasses
• Case 4: Metal-organic framework glasses

– Evidence for polyamorprphism in ZIF-4 – Melting and glass transition of ZIFs – Ultrahigh glass-forming ability of ZIF-62

26

Approaches we used

NMR HRTEM DSC Hyperquenching Ball milling Sub-Tg annealing (at T<Tg) Charaterizations

Stone Wool Milled powder Metallic glass

DSC output reflects the change of potential energy during heating or annealing

400 600 800 1000 0.8 1.0 1.2 1.4 1.6 Stone wool

Cp1 Cp2 T (K) Cp (Jg-1K-1)

DSC is a sensitive tool for detecting the energetic and structural evolution of glass

Tm Tf Tg Supercooled liquid standard glass HQG

Enthalpy Temperature

H

annealing Tf2 high Tf glass (e.g. stone wool) Tg Tf1 Tm

Collective configuration coordinate

Crystal low Tf glass (e.g. ultrastable film)

Potential energy

Z*

Yue, et al. APL 2002 Angell, et al. JPCM 2003 Hu, et al. JPC-C 2009 Qiao, et al. JACerS 2016

Hyperquenching-Annealing-Calorimetry

Hyperquenched (HQ) basalt glass

400 600 800 1000 0.6 0.8 1.0 1.2 1.4 1.6

Cp1

Teq

Cp2

T (K) Cp (Jg-1K-1)

Tc

Cp overshoot

400 600 800 1000 0.8 1.0 1.2 1.4 1.6 1.8

ta=90 min g f Ta (K) a: 573 b: 623 c: 673 d: 723 e: 773 f: 798 g: 823 e d c b a

T (K) Cp (Jg-1K-1)

An approach – for understanding the glass transition and relaxation

Excess enthalpy of fresh HQ fibers Excess enthalpy of annealed HQ fibers



  

eq c

T T p p excess

dT C C H ) (

1 2

Yue and Angell, Nature 2004

Energy ‘bird’

Yue, et al., Appl. Phys. Lett. 2002; Yue, et al. Chem. Phys. Lett. 2002

Basalt (relatively fragile)

400 600 800 1000 0.55 0.60 0.65 0.70 0.75 0.80

T (K)

d

ta a: 0 min b: 30 min c: 2 hrs d: 11 hrs e: 19 hrs f : 27 hrs g: standard

g f e c b a

Cp (Jg-1K-1)

Ta = 565 K (0.71Tg)

400 500 600 700 800 900 1000 0.8 1.0 1.2 1.4 1.6 1.8

I G F E D C A B

Cp (Jg-1K-1) T (K)

H

ta A: non-annealed B: 1 min C: 4 min D: 15 min E: 50 min F: 3.5 h G: 12 h H: 2 days I: 8 days

Onset of pre-endotherm

Ta=723 K (0.77Tg)

GeO2 (strong)

Differences in sub-Tg relaxation between a fragile and a strong system

More heterogeneous More non-exponential Less cooperative Pre-endotherm Less heterogeneous Less non-exponential more cooperative No pre-endotherm

Hu and Yue, JPC-B (2008) Yue and Angell, Nature (2004)

Double “glass transition” upon 55 days aging

400 500 600 700 800 900 1000 1.0 1.2 1.4 1.6 1.8 upscan 1 upscan 2

cooled at 106 K/s aged at 773 K for 55 days Cp (Jg-1K-1) T (K)

400 500 600 700 800 900 1000 1100

• 0.15
• 0.10
• 0.05

0.00 0.05 0.10 0.15 0.20

Energy release exotherm pre-endotherm Cp,exc (Jg-1K-1) T (K)

It shows that the relaxation is highly exponential, and hence, highly energetically heterogeneous ‘Shadow glass’ transition Yue and Angell, Nature 2004

20 40 60 80 100 120 140 160 0.00 0.01 0.02 0.03 0.04 0.05 0.06

hyperquenched annealed super-annealed crystallized

Z()

 (cm-1)

Vibrational density of state (VDOS) of HQ glasses

• VDOS peak at ~35 cm−1 in

the HQ state

• The peak is suppressed by

annealing.

• The peak disappears in

crystallized state.

• Source: topologically

diverse defects Implication: vibrational structure changes with the state of configurational excitation of the liquid.

Angell, Yue, et al. J. Phys: Cond. Mat. (2003)

Relaxation in hyperquenched 20MgO-20CaO-60SiO2 glasses – detecting structural heterogeneity

400 600 800 1000 1200 0.9 1.2 1.5

Cp (Jg-1K-1) T (K)

Remarkable! two sub-Tg energy release peaks Tg=999 K

400 600 800 1000 0.9 1.2 1.5

2nd upscan

Cp (Jg-1K-1) T (K)

1st upscan

2 sub-Tg relaxation peaks 2 kinds of structural domains?

400 600 800 1000 0.6 0.8 1.0 1.2 1.4 1.6

Tg Tc upscan 1 upscan 2

T (K) Cp (Jg

• 1K
• 1)

Each has its own structural heterogeneity. Two structural domains in the liquid state are frozen-in at high Tf.

Yue and Angell, Nature 2004, Yue, et al., Zhang, et al, JACerS 2013, 2017

Effect of sub-Tg annealing time on the Cp pattern and hence on the energetic elvolution of the two structural domains

Peak dimishes vertically (like strong systems) Peak diminshes horizontally (like fragile systems)

Two structural domains (strong and fragile ones)?

400 600 800 1000 0.55 0.60 0.65 0.70 0.75 0.80

T (K)

d

a: 0 min b: 30 min c: 2 hrs d: 11 hrs e: 19 hrs f : 27 hrs g: standard

g f e c b a

Cp (Jg-1K-1)

400 600 800 1000 0.8 1.0 1.2 1.4 1.6

Cp (Jg-1K-1) T (K)

Ta=823 K Fresh Standard 1 h 6 h 24 h 4 d (a)

400 500 600 700 800 900 1000 0.8 1.0 1.2 1.4 1.6 1.8

Cp (Jg-1K-1) T (K)

ta A: non-annealed B: 1 min C: 4 min D: 15 min E: 50 min F: 3.5 h G: 12 h H: 2 days I: 8 days Ta=723 K

Onset of pre-endotherm

Tg of SiO2 drastically drops with increasing water content

200 400 600 800 1000 1200 1400 1600 1800 0.7 0.8 0.9 1.0 1.1 1.2 1.3

1336 K

water content ~1 ppm ~1021 ppm

1434 K

Cp (Jg

• 1K
• 1)

T (K)

• Y. Z. Yue, Front. Mater. 2 (2015) 1

Comparison between the measured Cp and the calculated Cp

400 800 1200 1600 40 50 60 70 80

data Einstein Equation

SiO2 glass (<1 ppm water) qh=10 K/min

Cp (Jmol-1K-1) T (K)

Cv = 3REi(Xi) E(x) = x2ex/(ex-1)2 x = h/kT = /T where  = Einstein temperature s = 1100 K (Si vibrations) T = 370 K (transverse oxygen vibrations) L = 1220 K (longitudinal

• xygen oscillations)

Tg of silica decreases with repeating the DSC scans

400 800 1200 1600 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 SiO2 glass (~1 ppm OH) qh=10 K/min upscan 11 upscan 1 Cp (Jg-1K-1) T (K)

2 4 6 8 10 12 1300 1350 1400 1450

Tg (K) Number of DSC scans

Cristobalite formation and weakening of bonds by repeating reheating?

Enthalpy relaxation of a hyperquenched (HQ) normal glass and a HQ Silica

400 600 800 1000 1200 1400 1600 0.7 0.8 0.9 1.0 1.1 1.2

Upscan 2 Upscan 1

Tg=1356 K (1083 ºC)

SiO2 fiber T (K) Cp (Jg-1K-1)

HQ vitreous silica HQ normal glass

300 400 500 600 700 800 900 0.8 1.0 1.2 1.4 1.6

Upscan 1

Ca(PO3)2 fibers

qh=qc=20 K/min T (K) Cp (Jg-1K-1) Upscan 2

Effect of the annealing temperature Ta on the Cp,exc

0.0 0.1 0.2 0.3

d) c) b)

ta=12 hrs

Ta (K) A not annealed B 523 C 623 D 723 E 823

G E E F D D C C C B B B A A A C E D B A

HQGeO2 HQBas HQSiO2 Cp,exc (Jg

• 1K
• 1)

a)

ta=24 hrs

Ta (K) A not annealed B 873 C 923 D 948 E 1073

0.5 0.6 0.7 0.8 0.9 1.0 1.1 0.0 0.1 0.2 0.3

T/Tg (K/K)

ta=3 hrs

Ta (K) A no-annealed B 603 C 643 D 683 E 723 F 743 G 763

HQCmP T/Tg (K/K)

0.5 0.6 0.7 0.8 0.9 1.0 1.1 ta=3 hrs

Ta (K) A not annealed B 650 C 700

Effect of annealing time (ta) on Cp,exc of HQ glasses

0.0 0.1 0.2 0.3

b)

A

ta (hrs) A 0 B 0.017 C 0.25 D 3.5 E 12 F 192

Cp,exc (Jg

• 1K
• 1)

a) d)

B C A

HQGeO2 HQSiO2 c)

0.5 0.6 0.7 0.8 0.9 1.0 1.1 0.0 0.1 0.2 0.3

T/Tg (K/K)

F E D C B B CD C A

ta (hrs) A 0 B 0.25 C 3.5 ta (hrs) A 0 B 3 C 24

HQCmP

ta (hrs) A 0 B 0.11 C 1 D 9 E 27

T/Tg (K/K)

0.5 0.6 0.7 0.8 0.9 1.0 1.1 B C A

HQBas

• Y. Z. Yue, Front. Mater. 2 (2015) 1

(49m/s) (35m/s) (25m/s) (17m/s)

Cu46Zr46Al8

monotonic

Non-montonic structural response to sub-Tg annealing measured by x-ray scattering

Annealing dependence

• f the structural unit size

Annealing dependence

• f the correlation length

Critical temperature for the dramatic decreases in Rc: Tc ~ around 1.3Tg

Total structural factors PDF

Schematic scenario of the structural evolution during fragile-to-strong transition

Zhou, et al. J. Chem. Phys. (2015)

Outline

• Background and motivation
• Case 1: Borosilicate and phosphate glasses

– Dulong Petit Law – The Cp  m relation – Pressure effect on fictive temperature – Structural source of the Cp change – Prediction of Tg by topological model

• Case 2: Hyperquenched (HQ) glasses

– Relaxation in multi-component oxide systems – Relaxation in metallic glasses – Tg of SiO2 – Relaxation in HQ strong glass formers (SiO2 and GeO2)

• Case 3: Mechanically vitrified glasses
• Case 4: Metal-organic framework glasses

– Evidence for polyamorprphism in ZIF-4 – Melting and glass transition of ZIFs – Ultrahigh glass-forming ability of ZIF-62

44

Potential energy landscape

Two paths towards the glass state far from equilibirum:

• Thermally hyperquench liquids
• Mechanically hyper-mill crystals

Fiber spinner Ball mill

hyperquenching

Tg Tf

Crystal

Ultrastable glass

Tm

Potential Energy Z collective configuration coordinate

mechanical milling

Sub-Tg relaxation

Higher Tf

Contrasr in relaxation between hyperquenched and milling-derived glasses

900 700 500 300 T (K) Cp (JK-1g-1)

0.5 0.6 0.7 0.8 0.9 1.0 0.0 0.1 0.2 0.3

Main peak S2

B

HQBas

Cp (JK

• 1g
• 1)

T/Tg (K/K)

As-milled Ag3PS4

A

S1 Shoulder

Qiao, et al, J. Am. Ceram. 100 (2017) 968

360 400 440 480 520 560

• 1.5
• 1.0
• 0.5

0.0 0.5 1.0

Tg S1 Cp2 Cp (JK

• 1g
• 1)

T (K) Cp1 S2

(a)

As-milled Ag3PS4 As-quenched basalt glass

• As-milled Ag3PS4 glass is highly structurally

heterogeneous.

• The two peaks originate from - and –

relaxations.

Cp(sub-Tg)=Cp2-Cp1

Contrast in relaxation behavior between the two glasses

400 420 440 460 480 500 0.00 0.05 0.10 0.15

G F E D C B

Cp (JK

• 1g
• 1)

T (K)

Tmax (K)

A As-milled B 400 C 413 D 423 E 440 F 459 G 471

A

Energy release of both the as-milled Ag3PS4 glass (curve A) and the dynamically heated Ag3PS4 (curves B to G)

• Y. Z. Yue, et al, Appl. Phys. Lett. 81 (2002) 2983
• A. Qiao, et al. J. Am. Ceram. Soc. 100 (2017) 968-974

The milling-derived Ag3PS4 glass has similar relaxation feature to that

• f HQ glasses. But the former is structurally more heterogeneous.

Energy release of basalt glass wool after annealing at various Ta

Outline

• Background and motivation
• Case 1: Borosilicate and phosphate glasses

– Dulong Petit Law – The Cp  m relation – Pressure effect on fictive temperature – Structural source of the Cp change – Prediction of Tg by topological model

• Case 2: Hyperquenched (HQ) glasses

– Relaxation in multi-component oxide systems – Relaxation in metallic glasses – Tg of SiO2 – Relaxation in HQ strong glass formers (SiO2 and GeO2)

• Case 3: Mechanically vitrified glasses
• Case 4: Metal-organic framework glasses

– Evidence for polyamorprphism in ZIF-4 – Melting and glass transition of ZIFs – Ultrahigh glass-forming ability of ZIF-62

48

MOF glass family has emerged…

• Concerning chemical bonds, melt-quenched glasses have

3 families： – Inorganic non-metallic glasses (e.g. oxide, chalcogenide glasses, fluride…) – Organic glasses (polymer, molecular glasses…) – Metallic glasses

• A new family: Organic-inorganic hybrid glasses:

– ZIF glass, coordination polymers

49

Bennett, Tan, Yue, et al., Nature Com. 6 (2015) 8079 Bennett, Yue, Li, et al., J. Am. Chem. Soc. 138 (2106) 3484 Tao, Bennett, Yue, et al. Adv. Mater. 29 (2017) 1601705 Umeyama, et al. J. Am. Chem. Soc. 137 (2015) 864. Zhao, et al. J. Am. Chem. Soc. 138 (2016) 10818

We have succeeded in vitrifying several Zeolitic imidazolate frameworks (ZIFs)

• ZIF is a subset of MOFs

e.g., ZIF-4, structurally analogous to SiO2 Zeolitic topology Bonding unit for SiO2 Bonding unit for ZIF-4 SiO2 network

But their properties differ significantly.

Zn(C3H3N2)2

50

Coordinating bonds! Covalent+ionic mixed bonds!

Observe fascinating transitions in ZIF-4 by DSC

400 500 600 700 800 900 1000 1100

• 4
• 2

2 4 6

Cp (Jg

• 1K
• 1)

T (K)

20 K/min upscan

crystallization melting foaming

• nset of gas release

lattice collapse amorphisation LDA to HDL solvent release

80 85 90 95 100

Mass (%)

51

350 400 450 500 550 600 650 1.0 1.2 1.4 1.6

Cp (Jg

• 1K
• 1)

T (K)

ZIF-4

2nd upscan at 20 K/min

Tg=570 K

Crystal ZIFdesolvation-LDA HDA HDL phasecrystal ZIF-zni  MeltBulk glass Foam glass!

LDA: Low density amorphous phase HAD: High density amorphous phase

Crystal ZIF-4 Amorphisation Crystal ZIF-zni a) b) c)

350 400 450 500 550 600 650 700 0.8 1.0 1.2 1.4 1.6

Cp Tg of HDA Cpl

Cp (Jg-1K-1) T (K)

Cpg Upscan 2

LDL-HDL liquid transition

Cp (Jg

• 1K
• 1)

T (K)

1 2 3

Tg of HDA=563 K Cp=0.14 Jg

• 1K
• 1

Cpl/Cpg=1.1

release of solvent collapse into LDA Glass transition of HDA

upscan rate: 10 K/min 400 500 600 700 800 900 1000 1100

• 4
• 2

2 4 6

Cp (Jg

• 1K
• 1)

T (K)

20 K/min upscan

crystallization melting foaming

• nset of gas release

lattice collapse amorphisation solvent release

80 85 90 95 100

Mass (%)

Calorimetric evidence for polyamorphic transitions and glass formation

52

Potential energy landscape of ZIF-4

ZI

ZIF-4 LDA HDA/MQ G ZIF-zni

kBTm kBTg

LDA

kBTg

HDA/MQG

Potential Energy Configurational Coordinate

exo endo Hm Amorphization Quench- vitrifying

53

Co-existence of LDA and HDA phases!

480 520 560 600 640 680

LDA

300 400 500 600 700 1.0 1.2 1.4 1.6 H G F E D C B

Cp (Jg

• 1K
• 1)

T (K)

A A: 529 K B: 563 K C: 578 K D: 588 K E: 601 K F: 608 K G: 613 K H: 673 K

G H F E D C B

Heat flow (AU) T (K)

A

Rescans of ZIF-4 after the sample was scanned to different T Scan rate: 10 K/min HDA

Polyamphic transition in ZIF-4

54

Fragilities of LDA and HDA phases

LDA phase is superstrong! It is stronger than HDA phase and melt-quenched glass!

0.90 0.92 0.94 0.96 0.98 1.00 1.02 9 10 11 12 13

HDA: m=41 LDA: m=18

log ( in Pa s) Tg/T

ZIF-4

DSC data to viscosity data log  = 11.35 – logqh (Tf)

Yue, von der Ohe, Jensen, JCP (2004)

Dashed line: MYEGA fit

Mauro, Yue, Ellison, Gupta, Allan, PNAS (2009)                            1 1 15 exp 15 3 log T T m T T

g g



Compare with other systems

0.4 0.5 0.6 0.7 0.8 0.9 1.0

• 4
• 2

2 4 6 8 10 12

log (Pa s) Tg/T (K/K)

SiO2 (m=20) Anorthite (m=53)

ZIF-4 LDA (SAXS) (m=14)

ZIF-4 LDA (DSC) (m=18) ZIF-4 HDA (DSC) (m=41) T r i p h e n y l e t h y l e n e ( m = 1 1 )

56

Summary Glass transition is an fascinating complex problem. Investigation of glass transition is continuing…..

I thank all my co-authors and collaborators.

Thank you for your attention!