PALS C t ll d P Gl PALS on Controlled Pore Glass ...

PALS C t ll d P Gl PALS on Controlled Pore Glass –Porosimetry and Phase Transition m y

of Gas in Confinement

S Thraenert1 D Enke2 S. Thraenert , D. Enke , R. Krause-Rehberg1

Martin-Luther-Universität Halle-WittenbergHalle Wittenberg

Naturwissenschaftliche Fakultät II1Institut für Physik

2Institut für Chemie

Outline

Porous glass - CPGsynthesissynthesisproperties

From τ4 to dFrom τ4 to do-Ps and pick offETE (diff. geometries)

0.014

0.016

0.018

0.020

1.8 nm

Experimental resultsrelation to ETE 100000

1000000

0.006

0.008

0.010

0.012

6 2 nm

4.5 nm

2.5 nm

pdf n

(d)

pore size distributionphase trans. of gas

S mm1000

10000

8 nm

N0 5 10 15

0.000

0.002

0.0046.2 nm

d(nm)Summary0 100 200 300 400 500 600 700 800 900

10

100

40 nm

t ( )

d(nm)

PALS on Controlled Pore Glass 2

t (ns)

Controlled pore glass - CPGVYCOR-Process

alkali borosilicate

ExtractionTl

HCl/NaOH530 – 710°Cglass

ddPP 1 to 110 nm1 to 110 nmspinodal phase separationspinodal phase separationdecomposition is initiated by heat treatmentalkali rich borate phase <-> pure silicaalkali phase soluble in acid -> silica networkalkali phase soluble in acid -> silica networkpore size depends on basic materialporosity of 50 %


F. Janowski, D. Enke in F. Schüth, K.S.W. Sing, J. Weitkamp (Eds.), Handbook of Porous Solids, WILEY-VCH, Weinheim, 2002, 1432-1542.

Controlled pore glass - CPG

different geometries possiblep

homogenous microstructure

controlled pore sizemicrostructure

uniform pore sizewe can choose d (!)


D. Enke, F. Janowski, W. Schwieger, Microporous and Mesoporous Materials 2003, 60, 19-30.

Outline



0.014

0.016

0.018

0.020

1.8 nm


1000000

0.006

0.008

0.010

0.012

6 2 nm

4.5 nm

2.5 nm

pdf n

(d)


S mm1000

10000

8 nm

N0 5 10 15

0.000

0.002

0.0046.2 nm

d(nm)Summary0 100 200 300 400 500 600 700 800 900

10

100

40 nm

t ( )

d(nm)


t (ns)

Principles of PALS: pick-off annihilationPrinciples of PALSpick-off annihilation:

o-Ps captures e- with anti-parallel spinh d i lli i t ll f

positrons from 22Na: thermalize, diffuse, being trapped and annihilate

happens during collisions at walls of porelifetime (τ) decreases rapidlyτ is function of pore size: 1.5 - 142 ns

ppOR: positrons form Ps

palso for closed pore systems25 %

75 %

positronium: p-Ps -> short self annihilation

lifetime of 0.125 nso-Ps -> long self annihilation

lifetime of 142 ns (3γ)-> pick off annihilation (2γ)


> pick off annihilation (2γ)

Principles of PALS: typical spectrumtypical lifetime spectrum for CPG (here d = 20 nm):

4 ti l d t4 exponential decay components

p-Ps -> 0.125 ns

free positrons ~ 0.5 ns

o-Ps in disordered structure ~ 1.5 ns

o-Ps in pores

analysis with LT9 (LifeTime)analysis with LT9 (LifeTime)


Extended Tao Eldrup model extended TE model (calculations by EELViS):

quantum well of infinite height, but: overlap of o-Ps wave function and wall of pore -> δBoltzmann statistics ascribes explicit temperature dependence to the lifetimep p pintegrals of spherical / cylindrical Bessel functionsδ = 0.19 nm mean free path D = 4V/S = dcyl, diameter of cylinder

f th D 4V/S 2/3 d di t f hmean free path D = 4V/S = 2/3 dsphere, diameter of sphere


R. Zaleski, Excited Energy Levels and Various Shapes (EELViS), Institute of Physics, Maria Curie-Sklodowska University, Lublin, Poland

Outline



0.014

0.016

0.018

0.020

1.8 nm


1000000

0.006

0.008

0.010

0.012

6 2 nm

4.5 nm

2.5 nm

pdf n

(d)


S mm1000

10000

8 nm

N0 5 10 15

0.000

0.002

0.0046.2 nm

d(nm)Summary0 100 200 300 400 500 600 700 800 900

10

100

40 nm

t ( )

d(nm)


t (ns)

The experiments at T = 300 K

we measured CPG in a broad pore size rangepore size range

given pore sizes obtained by N2-adsorption or Hg-intrusionN2 adsorption or Hg intrusion

cubic and spherical model not sufficient for small poresp

cylindric model with δ = 0.193 nm best fit for our CPG -> calibration curve for calculating pore size

works well for RT, other T ?


The T-dependencecalculations: cylindric model with δ = 0.193 nmlth h f d d although we found good

agreement for T > 300 K temperature behavior can not be explained very well at low p ytemperaturesfor 20 nm catching effect of o-Ps at low temperatures (v. d.

l “ ll d ”) Waals power, “capill. cond.”), o-Ps bonds at the wallfor small pores -> thermal activated surface atoms > low activated surface atoms -> low T causes larger effective pore size (Ganguly et al. PPC8)

model still too simple but works well for room temperature


Pore size distribution

D τ4 σ41.8 nm 21.1 ns 14.8 ns2.5 nm 46.9 ns 17.6 ns4 5 nm 65 9 ns 18 9 ns4.5 nm 65.9 ns 18.9 ns6.2 nm 80.0 ns 19.3 ns

and its distribution by τ4 and its distribution σ4 by analysis of truncated spectra starting from 20 ns

problem of LT: limit of 142 ns is not taken into account, for large pores unphysically large σ4

distribution for 4 smaller selected pores


Pore size distribution0 016

distribution of τ4: α4(τ) = α4(λ)λ2, α4(λ) is probability density function (pdf) of 0.010

0.0120.0140.016

2.5 nm

1.8 nm

= α

4(λ)λ

2

α4(λ) is probability density function (pdf) of o-Ps annihilation rate, assumed by LT to be a log. Gaussian

0 0020.0040.0060.008

6.2 nm

4.5 nm

pdf α

4(τ) =

from distr. α4(τ) it is possible to calc.distribution of diameters of the pore:

⎞⎛

0 20 40 60 80 100 1200.0000.002p

τ(ns)

⎟⎟⎠

⎞⎜⎜⎝

⎛=

cylcyl dd

ddn 44 )()( ττα

100120140

all we need is a differentiable analytical function τ4 = τ4(dcyl):

406080

ETEFit

τ 4 (n

s)⎟⎟⎠

⎞⎜⎜⎝

⎛

+−

+= pcylcyl dd

AAA)/(1 0

2124τ 1 10 100

020

d (nm)


d (nm)

Pore size distributiondistribution norm. to 1

arrows show d directly calculated from mean o-Ps lifetime using cylindric model (1.77 nm, 3.09 nm, 4.38 nm and 5.80 nm)

this distribution contains the true variation of pore sizes true variation of pore sizes but also the effect of irregular not linear character of poresof pores

long tail for larger pores:

overestimation of α4(τ)f 4( )

nonlinear char. τ4 vs. d


to be published

Phase transition of CO2

120130

21 )

20 nm membr. at 180 K Adsorption 120

13021

m)

10 nm membr. at 180 K Adsorption

90100110

7.5

10

13.5

ize

D (n

m)

Desorption

4 (ns)

90100110

7.5

10

13.5

size

D (n

m

Desorption

(ns)

607080

4.5

4

pore

s

6

τ 4

607080

4.5

4

pore

6τ 4 (

0 200 400 600 800 1000CO2 pressure (mbar)

0 200 400 600 800 1000CO2 pressure (mbar)

we fill / degas in steps of 100 mbar

phase transition from gas to solid, p0 ~ 300 mbar at 180 K

we observe nearly no difference between adsorption and desorption curves for p > p0, small hysteresis in the end -> desorption of CO2 very easy

no complete pore filling because of pore blocking effect


no complete pore filling because of pore blocking effect

Phase transition of N2

110120130

110120130 20 nm membranes

90100110

20 nm membr at 77 K(ns)

90100110

(ns)

607080

20 nm membr. at 77 K

AdsorptionDesorption

τ 4

607080τ 4

0 200 400 600 800 100060 p

N2 pressure (mbar)50 100 150 200 250 300

60

T (K)we fill / degas in steps of 100 mbar

phase transition from gas to a liquid, p0 ~ 1000 mbar at 77 K

we observe a huge difference between adsorption and desorption curves, at 0 mbar shortest lifetime -> desorption of N2 not possible at 77 K

T-dep desorption shows interesting behavior also for other pore sizes


T dep. desorption shows interesting behavior, also for other pore sizes

Outline



0.014

0.016

0.018

0.020

1.8 nm


1000000

0.006

0.008

0.010

0.012

6 2 nm

4.5 nm

2.5 nm

pdf n

(d)


S mm1000

10000

8 nm

N0 5 10 15

0.000

0.002

0.0046.2 nm

d(nm)Summary0 100 200 300 400 500 600 700 800 900

10

100

40 nm

t ( )

d(nm)


t (ns)

Summaryfor T = 300 K we found a calibration curve for CPG

non-destructive porosimetry tool for open and closed pore-systemsmost sensitive for d = 0.5 … 10 nm

for pores d < 10 nm we can calculate a pore size distributionfirst measurements on phase transition of gas in CPGnear future:

SBA-15 (to be presented @ COPS VIII Edinburgh / Scotland, June 2008)


Acknowledgment

谢谢special 谢谢 from our group go to:

R Z l ki d hi R. Zaleski and his group (Lublin/Poland) for EELViS

G. Dlubek (Halle/Germany) for fruitful discussions

all organizers of PPC9


谢谢 for your patience!

this talk as pdf?this talk as pdf?

http://positron physik uni halle dehttp://positron.physik.uni-halle.de


PALS C t ll d P Gl PALS on Controlled Pore Glass ...

Documents

Transcript of PALS C t ll d P Gl PALS on Controlled Pore Glass ...