THE PORE SIZE DISTRIBUTION OF PORTLAND CEMENT PASTE

D N

.

WINSLOW

Final Report

THE PORE SIZE DISTRIBUTION OF PORTLAND CEMENT PASTE

To?

J. F. McLaughlin, Director Joint Highway Research Project

February 14, 1969 Projects

Fromg

C-36-61E

Ho L. Michael, Associate Director

Joint Highway Research Project

5-14-5

Files

The attached report "The Pore Size Distribution of Portland Cement Paste" is the Final Report on Phase I of the HPR Part II research project "Fundamental Studies in Portland Cement Concrete, Phase I". The report has been authored by Mr D. N. Winslow, Graduate Assistant in Research on our staff, under the direction of Professor Sidney Diamond, Mr. Winslow also used the report for his MSCE thesis. The report reviews theories of pore size distribution measurement and both theoretical and practical size limits are discussed. In this study mercury intrusion was used to measure pores of one size and capillary condensation was used for another size. Details of each technique and equipment together with results are given for cement paste samples.

The results of this Phase I also resulted in a proposal for Phasell of this research. This proposed Phase II has already been submitted for review and action. This report is submitted for the record and acceptance. It will also be forwarded to the BPR and the ISRC for their review, comment and acceptance.

Respectfully submitted,

Harold L. Michael Associate Director HLMsrg CC;

F. W. W. W. G. M.

L. L. Ho L. K, Eo

Ashbaucher Dolch Goetz Greceo Hallock Harr

R. J. V. G. F. R.

H. Harrell A. Havers E. Harvey

A. Leonards B. Mendenhall Do Miles

C. M. W. H. K. E.

F. B. T. R. B. J.

Scholer Scott Spencer J. Walsh Woods Yoder

3 JCMfC 1

n

mo MbiTuc j*raia

sua uof

Digitized by the Internet Archive in

2011 with funding from

LYRASIS members and Sloan Foundation; Indiana Department

of Transportation

http://www.archive.org/details/poresizedistribuOOwind

Find Report- Phase

I

THE PORE SIZE DISTRIBUTION OF

PORTLAND CEMENT PASTE

by Do No Winslow Graduate Assistant in Research

Prepared as Part of an Investigation Conducted by Joint Highway Research Project Engineering Experiment Station Purdue University

in cooperation with the

Indiana State Highway Commission

and the

Department of Transportation Federal Highway Administration Bureau of Public Roads

J. S*

The opinions, findings and cone lus ions sexpressed in this publication are those of the authors and not necessarily those of the Bureau of Public Roads

Not Released for Publication

Subject to Change

Not Reviewed By Indiana State Highway Commission or the Bureau of Public Roads

Purdue University Lafayette, Indiana December 18, 1968

ii

ACKNOWLEDGMENTS

This writer wishes to express his sincere appreciation to his

major professor, Dr. Sidney Diamond, for his guidance, help and understanding during the long course of this research and the writing of this report.

Also, the advice of Dr. William L. Dolch was very help-

ful and greatly appreciated.

This writer must also express his appreciation to his father, Dr.

Nathaniel M. Winslow, for his counsel and for the development and design of the major portion of the equipment used in this work. Mr. Jerry Isenburg, Mr. Everett Sutton and the Dow Chemical

Company should be thanked as they made available the use of, and

pictures from, a scanning electron microscope.

Many other people

were of great assistance in the technical phases of this work. This research was sponsored by the Indiana State Highway Depart-

ment and the Bureau of Public Roads, U.

S.

Department of Transportation

and was administered through the Joint Highway Research Project. Finally, this author wants to thank his wife, Linda, for a great

deal of patient understanding and self-sacrifice while this research was in progress.

iii

TABLE OF CONTENTS Page

LIST OF TABLES v

LIST OF FIGURES vi

ABSTRACT

V"INTRODUCTION

viii

THEORIES OF PORE SIZE MEASUREMENT 6

General Considerations . Capillary Condensation . Mercury Intrusion ....'.".'.'

6 8

APPAPATUS, EXPERPMENTAL TECHNIQUES AND REDUCTION OF MTA [ . Cement Paste Mixing and Curing Cement Paste Mixer ... Mixing Cement Paste' Curing Cement Paste Cement Paste Sampling and Drying sampling Cement Paste Techniques of Drying Cement Mercury Intrusion Porosimetry 'paste Mercury Intrusion Apparatus Technique for Measurement of MercurC TntV,*-,' D Reduction of Data to Pore Size Distributions Capillary Condensation Apparatus for CaoiUary Condensation Capillary Condensation Measurement Technique Lc cnmque Calculation of Pore Si^p w

>

a]

s

days

days

OO

ITN

t~

days

-d

co CM

days

i

cr CO IP

H -'

O

H -

^

>_'

•_'

vc

c\ J

cJ

(S/tui)

1

aumxoA

aJO,~[

*i

aATq.eTranno

Figure 11

Mercury Intrusion Pore Size Distributions for Cement Pastes at Eight Ages (water: cement ratio 0.6)

H

hi

sumioA

3.10(1

T^Q-Oi jo q.uso jaj

Figure 12

Mercury Intrusion Pore Size Distributions for Cement Pastes at Eight Ages (water :cement ratio 0.6)

kh

c

O

^>

10

3

5"

(a^s-ed ^.uauiao 3/oTJ 3) uot^djos

Figure 13

Time Rate of Sorption of Water Vapor on Cement Paste at Four Partial Pressures (water rcement ratio O.U)

-S

h5

o

o 0* ll*

o

H*

o

o

CM

CM

PL.

o

P-4

CO

o

o

C)

>

o

1

1

00

|

CXI

1 1

\

1

o

\ *D

CO




O

t- C0\O

.-.c\_-vo\Ococmo cocorococococo

Hh

O

o c-OM^co^r co_a-

IT\

t~-

t-CO

t-CO

OO

H

lf\M3

H0J roiA

s

CMtfNO

CM \0 CM

CO

o O

•P -P

CM CM ITS J- tr\ t—

O VO O CO H 1/AVO niAlAOVD t— CO CO

CO l^ CM C\CM

(h

t-j-

CO COCO C— VO^OVQ^O

ooooooSo

CJ\

CM CO

LT\

t-CO 0\co CM IT\ H CO

VsOvOvOvOvOvOvOvO

odddoooo

62

size measurement by capillary condensation is controlled by the diameter of the condensed molecules and that a pore diameter equal to about ten

molecular diameters is perhaps the lower limit of the method.

This may

be so and the data point at O.OOlUn may be purely artificial in that the

underlying assumptions of capillary condensation which were used to calculate the point are not valid.

However, both distributions had ceased

to register appreciable pore volumes in the fine range and it was desired to see if pushing the method to a smaller pore would indicate the pre-

sence of appreciable pore volumes In extremely fine pores

.

If the method

is applicable at O.OOlUu then the presence of such pores is clearly in-

dicated.

In any event the portions of the distributions for larger dia-

meters are still valid.

The arithmetic summation of pore volumes by the

Roberts method begins with the largest pores and works down.

Thus, add-

ing or not adding a final step or two at the small pore end in no way

affects the pore volumes at larger diameters.

If the distributions were

terminated at a diameter of say 0.0030|j they would be identical to those presented, up to that point. Previously, it was pointed out that the method of water vapor sorp-

tion used in this work probably allowed the water vapor to register pore volumes at correct diameters.

Further, mercury intrusion was seen to be

susceptible to improperly registering pore volumes due to the effect of

narrow entryways.

The deviations of the water vapor pore size distribu-

tions from the mercury pore size distributions are believed to arise from the improper tallying of pore volumes by mercury intrusion.

The distributions of both water rcement ratio pastes exhibit this sort of deviation in the range of diameters from O.lOOOu to 0.(AOO|i.

It

63

would seem that there is a small volume of pores within this range inside the paste which are accessible only through somewhat smaller entryways. The existence of these pores as indicated by the water vapor curves does not appear to invalidate the basic idea of a threshold diameter above

which few pores exist and immediately below which there exists a considerable volume of pores.

After this "large diameter" deviation of the pore distributions, the water vapor and mercury curves rejoin (at a diameter of about 0.j3

fc«-

VO CO

t-

In

bO

o3

s

o

f-.

O eu r-l

(d

p

Ho

ed

FH

K

•^

+•>

C C)

i

OV

*-^T3 to

>j-P

C

ed

M

as

^

T*

o