0% found this document useful (0 votes)
4 views9 pages

Day1988routine Analysis of the Pore Structure of Mesoporous and Macroporous Solids By

This paper investigates the pore structure of mesoporous and macroporous solids using nitrogen adsorption/desorption, mercury porosimetry, and densitometry. It highlights the limitations of these methods, particularly in the presence of macropores, and emphasizes the need for caution in quantitative analysis. The study concludes that while nitrogen data can provide reliable pore size distributions for mesoporous materials, mercury intrusion is essential for accurately characterizing more complex pore structures.

Uploaded by

moktaria.reguig
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
4 views9 pages

Day1988routine Analysis of the Pore Structure of Mesoporous and Macroporous Solids By

This paper investigates the pore structure of mesoporous and macroporous solids using nitrogen adsorption/desorption, mercury porosimetry, and densitometry. It highlights the limitations of these methods, particularly in the presence of macropores, and emphasizes the need for caution in quantitative analysis. The study concludes that while nitrogen data can provide reliable pore size distributions for mesoporous materials, mercury intrusion is essential for accurately characterizing more complex pore structures.

Uploaded by

moktaria.reguig
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

K.K. Unger et al.

(Editors), Characterization of Porous Solids 491


1988 Elsevier Science Publishers B.\'.. Amsterdam - Printed in The Netherlands

ROUTINE ANALYSIS OF THE PORE STRUCTURE OF MESOPOROUS AND MACROPOROUS SOLIDS BY


GAS ADSORPTION, MERCURY POROSIMETRY AND DENSITOMETRY

M Day, R Fletcher

ICI Chemicals and Polymers Group


Research and Technology Department
Wilton
Cleveland UK

ABSTRACT

This paper describes a study of a range of porous solids using nitrogen


adsorption/desorption at 77°K, mercury intrusion/extrusion porosimetry and
densitometry in helium and mercury. Qualitative evidence from the techniques
used indicated that the materials were representative of classes of pore
structures of differing complexity across the spectrum of meso-macropore
structures. Since commercial automated instruments and procedures are now
widely used routinely in pore structural characterization, the materials offer a
means of evaluating the validity 6f the analytical techniques as a function of
the complexity of the pore structure. It appears that only materials consisting
exclusively of mesopores can sustain quantitative analysis by these methods.
When macropores are present the analyses are less reliable and less consistent.

INTRODUCTION

The advent of commercial automated instruments for measuring nitrogen


adsorption/desorption isotherms and mercury intrusion/extrusion, with
dedicated computer faCilities for transforming data into pore size
distribution, means that quantitative descriptions of the pore
structures of solids are easily generated, but must be used with
caution.

The follOWing recommendations are contained in the IUPAC guidelines (1)


recently issued on the assessment of pore structure from nitrogen
adsorption/desorption data.

Assess microporosity via t or o<.s methods (2, 3).

2 Assess the likely formal shape of the pores in the structure via
inspection and classification of isotherm shape and hysteresis
loop: check for loop closure at p/Po~O.4.

3 Apply pore size distribution transforms only in cases where


there are no micropores, where pores are rigid in well-defined,
mesopore r3nges.

Mercury porosimetr) is an important complementary technique, but


formally agreed procedures for its use and interpretation are less
developed.
492

This paper sets out to investigate the problems associated with the
derivation of descriptions of pore structures from nitrogen and mercury
techniques using a series of porour solids which could represent classes
in a spectrum of increasing structural complexity.

2 THEORY

The theories of nitrogen adsorption/desorption and mercury porosimetry,


and the models and algebra for deriving pore size distributions are well
known (4, 5). In the case of nitrogen data, the derivation will be
suspect if:

a) The pores are not rigid, eg located between small plates.

b) Isotherm shape is affected by the presence of micropores: closure


of hysteresis loop at p/Po-",0.4 is indicative.

c) The pore sizes cover a wide enough range to render the Kelvin
equation imprecise.

On these criteria the most reliable derivation will emerge from solids
with type IV isotherms, HI hysteresis loops, representing sharp
mono-modal mesoporous structures with no micropores.

Qualifications concerning the interpretation of mercury intrusion data


are as follows:-

a) Mercury must not react or interact strongly with any part of the
pore structure.

b) The solid must not break under the pressures employed.

c) The shape of the pores eg ink bottle shapes, can affect the
interpretation of intrusion and extrusion.

3 EXPERIMENTAL

3.1 APPARATUS

Nitrogen adsorption/desorption measurements were carried out wi~h a


Micromeritics Inc 'Digisorb 2500' uSing 5-point BET plots and 43 point
adsorption/desorption isotherms. 0~4gassing was c:rried out o;-the
instrument under moderate vacuum 10 torr, at 140 C for 4 hours. The
isotherm data were output to a plotter for inspection and
classification and were transformed into o(s plots against an alumina
standard. Pore size distributions can be derived from the software
incorporated into the instrument, uSing the method of Robert (6).
Mercury intrusion was carried out using a Micromeritics Inc 'Autopore
9200' operating from 2 to 60,000 psia and hence in theory able to
penetrate pores as small as 36A o in diameter. Outgaasing on !~e
instrument is limited to evacuation at room temperature to 10 torr.
He-Hg densitometry was used as a measure of total pore volume against
which pore volumes derived from N2 and mercury could be compared.

3.2 MATERIALS

These were stable alumina-based materials derived from our laboratory,


and represent a larger body of materials consisting of six from sets 1,
3 and three from set 2. The range of their micromeritic properties was
as follows:
493

SAMPLE SET PORE VOLUME SURFACE AREA ~GE


(He/Hg) ml/g REPRESENTED m / g

0.5 - 0.8 100 - 250

2 0.2 - 0.5 50 - 150

3 0.3 - 0.6 100 - 350

All were pellets or granules not less than lmm in size, and had been
extensively dried at 110°C in air before introduction into instrumental
outgassing facilities.

4 RESULTS AND DISCUSSION

Figs 1-3 A, B compare the nitrogen adsorption/desorption isotherms and


mercury intrusion/extrusion data for sets 1, 2, 3. The data have been
normalized because they are representative of numerous individual sets
of data. Nitrogen adsorption isotherms were transformed into s plots
against an alumina standard; no detectable microporosity was observed in
them. The software contained in each instrument can be used to derive
pore size distributions from nitrogen adsorption/desorption and mercury
intrusion. These derivations and the agreement between N2 and
mercury-derived pore sizes and volumes are reported in Table 1. In
addition,_the retention of mercury after 1st cycle intrusion and
extrusion, which may be specific to the mode of operation of our
porosimeter, was noted.

The isotherms of set 1 are assigned type IV with HI hysteresis loops


according to IUPAC recommendations. Isotherms in set 2 are more
complex, clearly bimodal and are assigned as the superposition of type
IV isotherms, but there is a slight upturn as p/Po -->1, indicating that
pore structures are more complex than indicated by the type IV
assignment.

Comparison of Figs 1-3 indicates some qualitative parallelism between


hysteresis in nitrogen adsorption/desorption and mercury intrusion/
extrusion.

Qualitatively the sets can be assigned to the following three classes:-

a) Sharp monomodal - mesoporous

b) Narrow range bimodal - meso/macroporous

c) Wide range bimodal - meso/macroporous

For set 1, nitrogen and mercury-derived pore size distributions agree


well in both pore size and pore volume (typically to within 10%). Pore
size distribution derived from nitrogen and mercury data for set 2 do
not agree well, despite their qualitative similarities. In the larger
set of pores agreement in derivation of pore size and volume is poor (no
better thari 50%), and is similarly poor in the derivation of mesopore
volume. However, agreement in derivation of mesopore size is bette, but
variable (15-25%).
494

Fig 1

N2 Adsorption - Desorption Set 1

"'C
Q)
N 0.9
(1J

E
... 0.8
0
2:
0.7
o
---
Q.
I-
0.6
CIJ
...J 0.5
2
CIJ 0.4
C
ex
CIJ 0.3
2:
0 0.2
>
0.1
0.1 0.3 0.5 0.7 0.9
P/PO

Mercury Intrusion - Extrusion: Set 1

"'C 0.9
Q)
.!!1
co
..E
0.8

0 0.7
2:
l: 0.6
0

.
til
;:,
....x
0.5

w 0.4

l:
0.3
.S!til
..
;:,
....l:
0.2

0.1

0
1.4 1.8 2.2 2.6 3 3.4 3.8
o
Log 10 (Pore Diameter) A
495

Fig 2

N2 Adsorption - Desorption Set 2

"'C
OJ
N 0.9
co
...0E 0.8
2
0.7
C1
c..
'<,

l-
0.6
(/)

s 0.5
...J

(/)
Q
0.4
c:(
(/) 0.3
2
0 0.2
>
0.1
0.1 0.3 0.5 0.7 0.9
P/PO

Mercury Intrusion - Extrusion: Set 2

"'C 0.9
OJ
.!!!
co 0.8
E
...
0 0.7
2
c 0.6
0
Vl

.......
:::l 0.5
)(
w 0.4

0.3
c:
0
Vl 0.2
.......c:
:::l

0.1

0
1.4 1.8 2.2 2.6 3 34 3.8
o
Log 10 (Pore Diameter) A
496

Fig 3

N2 Adsorption - Desorption Set 3

"tl
Ql
N 0.9
It!

...0E 0.8
Z
0.7
o
'<,

0- 0.6
I-
C/)
...J 0.5
~
C/)
0.4
0
<t
C/) 0.3
Z
0 0.2
>
0.1
0.1 0.3 0.5 0.7 0.9
P/PO

Mercury Intrusion - Extrusion: Set 3

"tl 0.9
Ql
.!!!
ell 0.8
...
E
0 0.7
Z
c 0.6
.2
II)

......><
:::l 0.5

w 0.4

r::: 0.3
.2
II) 0.2
...c...
:::l

0.1

0
1.4 1.8 2.2 2.6 3 34 3.8
o
Log 10 (Pore Diameter) A
497

TABLE 1

DERIVATION OF PORE PARAMETERS FROM N2 ADSORPTION, MERCURY INSTRUSION AND


DENSITOMETRY

SET DERIVATION OF MEAN PORE VOLUME ml/g


PORE SIZE AO

MESOPORES MACROPORES MESOPORES MACROPORES TOTAL


N2 Hg NZ Hg N2 Hg N2 Hg DENSITOMETRY

95 100 - NA - 0.65 0.65 - NA - 0.65

160 176 - NA - 0.59 0.55 - NA - 0.55

2 55 65 200 600 0.12 0.10 0.05'f 0.15 0.22

38 50 300 600 0.10 0.05 0.20 0.32 0.38

3 50 55 ND 3000 0.30 0.15 ND 0.11 0.32

320 250 ND 5000- 0.10 0.06 ND 0.04 0.10


60000

T The values of macropore volume were derived from NZ adsorption/desorption data


using the procedure (6) contained in the analytical software on the commercial
instrument (Micromeritics Digisorb 2500)

The pore size distributions of samples in set 3 derived from nitrogen


adsorption/desorption data indicate monomoda1 sets of mesopores, but
fail to detect the presence of macropores (well separated from the
mesopores) which mercury intrusion data indicates. The slight upturn in
the adsorption isotherm as p/Po -->1 is an indication that the sample
has some macroporous structure, but insufficient to be detected by the
algebraic derivation of pore size distribution. Derivation of mesopore
size from nitrogen and mercury data agree to <10% at best, more
typically 15-25%. Non-detection of macropores is probably a failure in
precision and applicability of the Kelvin equation.
In all cases, entrapment of mercury on 1st cycle extrusion is of the
order of 20-30% of total intrusion. In sample sets I, 3 there is a good
degree of parallelism between intrusion and extrusion branches of the
mercury data (width 200-300AO) in the mesopore region (log d basis) and
drainage of mercury from mesopores of set 3 is similar to that from
sample 1. It appears that in these sets (where mesopores in a bimodal
system are well separated from macropores in size) the mesopores can
behave independently with regard to nitrogen and mercury and hence
exhibit similarities to systems containing only mesopores.
498

The parallelism in intrusion/extrusion is not observed in set 2, and the


width of the enclosed loop increases as pore diameter increases in the
mesopore regIon (log d DasIs). Set 2 represents interesting bimodal
systems in which the two pore systems are close enough in size to cause
an overlap, for instance, in the extrusion of mercury, and in which the
macropores are small enough to be qualitatively detected by the
Kelvin-based procedure and yet sufficiently separate in size to be
distinguishable from the true mesopores. It is possible that mercury
drains from both mesopores and small macropores.

A number of workers have compared the derivation of surface areas or


pore size distributions from nitrogen adsorption/desorption and mercury
porosimetry data for a range of porous oxides (7), compacts (8), and
crystalline polymeric filters (9), have found varying degrees of
agreement between the two techniques, and advise that they shall be used
to provide complementary data (10).

The importance of using complementary techniques of nitrogen adsorption/


desorption and mercury intrusion is obvious in the cases of sets 2 and
3. Only in case 1 can reliance on the derivation of pore size
distribution from nitrogen data be justified, because it is well
supported by mercury data. In set 3 reliance solely on nitrogen data
would even result in incorrect qualitative description of the pore
structure. In set 2 both techniques provide broadly similar qualitative
descriptions. This could be indicative of the influence of the
closeness of the distributions, and probably of ill-defined small
mesoporosity, on the algebraic derivation of pore size distribution.

In all cases there is an interesting qualitative parallelism between


hysteresis loops for nitrogen and for mercury processes.

5 CONCLUSIONS

A wide range of non-microporous materials can be classified according to


the relative size and abundance of meso and macropores. It appears that
only class 1 materials (mesoporous only) can sustain quantitative
analysis by methods now routinely accessible on modern commercial
instruments, due to the complementarity of nitrogen adsorption and
mercury intrusion data. The importance of using complementary
techniques is emphasised by class 3, where nitrogen data tends to fail
to detect macroporosity in pores_lOOOAo diameter. In class 2 both
techniques have sufficient precision to generate similar qualitative
views of complex bimodal pore structures but substantially different
quantitative assessments. Considerable care is required in analysing
this class. In all cases, qualitative assessment can be surprisingly
informative.

There is some evidence from Table 1 that the derivation of meso pore size
may be reliable in all three classes.
499

REFERENCES

IUPAC, Pure and Appl. Chern. 1985, 57, 603

2 de Boer J.H., J. Catal 1964, 3, 32

3 Sing K.S.W., Payne D.A. Chern. and Ind. 1969, 918

4 Barrett E.P., Joyner L.G., Halenda P.H. J. Amer. Chern. Soc. 1951, 73,
373

5 Drake L.C., Ritter H.C. Ind. Eng. Chern. 1945, 17, 787

6 Robert B.F., J. Colloid Interface Sci. 196/, 23, 266

7 de Wit L.A., Scholten J.J.F., J. Catal 1975, 36, 36

8 Astier M.P., Sing K.S.W. "Studies in Surface Science and Catalysis"


1982, 10, 269, Publ. Elsevier

9 Ternan M., Fuller O.M. Can. J. Chern. Eng. 1979, 57, 750

10 Lecloux A.J., 'Catalysis Science and Technology' 1981, 2, 171,


Publ. Springer - Verlag

You might also like