Full download ebook at ebookname.

com

Heat Air Moisture Transport Measurements on

Building Materials ASTM special technical
publication 1495 Phalguni Mukhopadhyaya

https://ebookname.com/product/heat-air-moisture-transport-
measurements-on-building-materials-astm-special-technical-
publication-1495-phalguni-mukhopadhyaya/

OR CLICK BUTTON

DOWLOAD NOW

Download more ebook from https://ebookname.com

More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Metrology of Pedestrian Locomotion and Slip Resistance

ASTM Special Technical Publication 1424 Mark I. Marpet

https://ebookname.com/product/metrology-of-pedestrian-locomotion-
and-slip-resistance-astm-special-technical-publication-1424-mark-
i-marpet/

Masonry Opportunities for the 21st Century ASTM Special

Technical Publication 1432 Diane Throop And Richard E.
Klingner

https://ebookname.com/product/masonry-opportunities-for-the-21st-
century-astm-special-technical-publication-1432-diane-throop-and-
richard-e-klingner/

Multiple Stressor Effects in Relation to Declining

Amphibian Populations ASTM special technical
publication 1443 Gregory L. Linder

https://ebookname.com/product/multiple-stressor-effects-in-
relation-to-declining-amphibian-populations-astm-special-
technical-publication-1443-gregory-l-linder/

Heat and Moisture Transfer between Human Body and

Environment 1st Edition Fohr

https://ebookname.com/product/heat-and-moisture-transfer-between-
human-body-and-environment-1st-edition-fohr/
Hazardous Air Pollutant Handbook Measurements
Properties and Fate in Ambient Air 1st Edition Chester
W. Spicer

https://ebookname.com/product/hazardous-air-pollutant-handbook-
measurements-properties-and-fate-in-ambient-air-1st-edition-
chester-w-spicer/

The Gluten Proteins Special Publication 1st Edition D.

Lafiandra

https://ebookname.com/product/the-gluten-proteins-special-
publication-1st-edition-d-lafiandra/

Building Heat Transfer 1st Edition Morris Grenfell

Davies

https://ebookname.com/product/building-heat-transfer-1st-edition-
morris-grenfell-davies/

Mechanical Thermal and Environmental Testing and

Performance of Ceramic Composites and Components A S T
M Special Technical Publication Stp 1392 Editors:
Michael G. Jenkins
https://ebookname.com/product/mechanical-thermal-and-
environmental-testing-and-performance-of-ceramic-composites-and-
components-a-s-t-m-special-technical-publication-
stp-1392-editors-michael-g-jenkins/

Neonatal Tetanus Elimination Field Guide Scientific and

Technical Publication 2nd Edition Paho

https://ebookname.com/product/neonatal-tetanus-elimination-field-
guide-scientific-and-technical-publication-2nd-edition-paho/
STP 1495

Heat-Air-Moisture Transport:
Measurements on Building Materials

Dr. P. Mukhopadhyaya and Dr. M. K. Kumaran,

editors

ASTM Stock Number: STP1495

ASTM
100 Barr Harbor Drive
PO Box C700
West Conshohocken, PA 19428-2959
Printed in the U.S.A.
Library of Congress Cataloging-in-Publication Data

Heat-air-moisture transport : measurements on building materials /

Phalguni Mukhopadhyaya and Mavinkal Kumaran, editors.
p. cm.
ISBN 978-0-8031-3422-5
1. Dampness in buildings. 2. Heat--Transmission. I. Mukhopadhyaya,
Phalguni. II. Kumaran, Mavinkal.

TH9031.H43 2007
693.8’ 93--dc22
2007040740

Copyright © 2007 AMERICAN SOCIETY FOR TESTING AND MATERIALS

INTERNATIONAL, West Conshohocken, PA. All rights reserved. This material may not
be reproduced or copied in whole or in part, in any printed, mechanical, electronic,
film, or other distribution and storage media, without the written consent of the publisher.

Photocopy Rights
Authorization to photocopy items for internal, personal, or educational
classroom use, or the internal, personal, or educational classroom use of
specific clients, is granted by the American Society for Testing and Materials
International „ASTM… provided that the appropriate fee is paid to the Copyright
Clearance Center, 222 Rosewood Drive, Danvers, MA 01923; Tel: 978-750-8400;
online: http://www.copyright.com/.

Peer Review Policy

Each paper published in this volume was evaluated by two peer reviewers and at
least one editor. The authors addressed all of the reviewers’ comments to the satisfaction
of both the technical editor共s兲 and the ASTM International Committee on Publications.
The quality of the papers in this publication reflects not only the obvious efforts of
the authors and the technical editor共s兲, but also the work of the peer reviewers. In
keeping with long-standing publication practices, ASTM International maintains
the anonymity of the peer reviewers. The ASTM International Committee on Publications
acknowledges with appreciation their dedication and contribution of time and effort
on behalf of ASTM International.

Printed in the USA

November 2007
Foreword

The First Symposium on Heat-Air-Moisture Transport: Measurements on Building

Materials was held in Toronto, Ontario, Canada on 23 April 2006. ASTM Committee C-16
on Thermal Insulation served as its sponsor. The symposium chairs and co-editors of this
publication were Phalguni Mukhopadhyaya and Mavinkal K. Kumaran of the National
Research Council, Ottawa, ON, Canada.
Contents

Overview vii

Modeling the Heat, Air and Moisture Response of Building Envelopes: What Material
Properties are Needed, How Trustful are the Predictions?—H. S. L. C. HENS 1

Improved Suction Technique for the Characterization of Construction Materials

—L. G. THYGESEN AND K. K. HANSEN 12

Water Vapor Transmission Measurement and Significance of Corrections

—P. MUKHOPADHYAYA, K. KUMARAN, J. LACKEY, AND D. VAN REENEN 21

Moisure Buffer Value of Building Materials—C. RODE, R. PEUHKURI, B. TIME,

K. SVENNBERG, AND T. OJANEN 33

Effects of Drying Conditions, Phase Transformations, and Carbonation Reactions on

Measurements of Sorption Isotherms of Building Materials—K. E. WILKES,
J. A. ATCHLEY, P. W. CHILDS, AND A. DESJARLAIS 45

Evaluation of Functional Approaches to Describe the Moisture Diffusivity of Building

Materials—G. SCHEFFLER, J. GRUNEWALD, AND R. PLAGGE 55

Determination of Equilibrium Moisture Cement of Building Materials: Some Practical

Difficulties—M. K. KUMARAN, P. MUKHOPADHYAYA, AND N. NORMANDIN 71

Inputs and Analyses: An End User’s Perspective of Heat-Air-Moisture Data

—C. P. DECAREAU, L. KAN, AND JOSEPH P. PINON 80

Interlaboratory Comparison of the Thickness of the Destroyed Surface Layer

of Closed-Cell Foam Insulation Specimens—T. K. STOVALL 88

The Effect of Air Flow on Measured Heat Transport Through Wall Cavity Insulation
—D. W. YARBROUGH AND R. S. GRAVES 94

The Effect of Vertical Distribution of Water Permeability on the Modeled Neutralization

Process in Concrete Walls—Y. KISHIMOTO, S. HOKOI, K. HARADA, AND S. TAKADA 101

v
Overview
Since the inception in 1938, ASTM Committee C16 has been working on various aspects con-
cerning the development of standards, promotion of knowledge, and stimulation of research
pertaining to the heat-air-moisture transport through building materials 共e.g. thermal insulation
materials, products, systems, and associated coatings and coverings兲. During this time the com-
mittee has seen many changes and challenges in the building construction industry and invari-
ably responded effectively to address the pressing concerns of the time.

In recent years, the building construction industry is making increased use of sophisticated
computer based design tools for moisture design of building envelopes. These design tools
invariably require well-defined heat-air-moisture transport properties of component building ma-
terials. The basics for heat-air-moisture transport through building materials had been researched
upon for a long-time all over the world. However, there is a glaring lack of unifomity in the ways
these transport processes are measured in various laboratories all over the world. The results
coming out from various test methods are also presented in many different ways. This leads to a
very confusing scenario for the end users of these material properties. Globally there is a great
need to resolve this issue urgently so that the measured material properties are reliable, consistent
and meaningful irrespective of the laboratory and personnel involved in the process.

The primary goal of the 1st symposium of “Heat-Air-Moisture Transport: Measurements on

Building Materials,” held in Toronto, Canada on April 23, 2006 was to provide a forum to
discuss the state-of-the-art research and development activities on the measurement of heat-air-
moisture transport through building materials, related to international standards. This STP pre-
sents selected peer reviewed papers from the symposium authored by renowned international
experts. The STP starts with a paper that outlines the importance and necessity of reliably
measured hygrothermal material properties, and the papers that follow deal with the individual
material properties. Some of them identify the need for improvement in existing standards and
others point towards new measurement techniques and corresponding standards.

Finally, the editors would like to acknowledge that this STP is a product of tremendous diligent
efforts of many people. In particular, the editors would like to thank ASTM symposium orga-
nizing committee, all of the authors and paper reviewers who devoted their time for this en-
deavor. Special thanks are due to Dorothy Fitzpatrick and Timothy Brooke at ASTM for their
support, timely assistance, and efficient handling of all minute details.

Dr. P. Mukhopadhyaya
National Research Council
Ottawa, ON, Canada

Dr. M. K. Kumaran
National Research Council
Ottawa, ON, Canada

vii
 Journal of ASTM International, Vol. 4, No. 2
Paper ID JAI100460
Available online at www.astm.org

Hugo S. L. C. Hens1

Modeling the Heat, Air, and Moisture Response of Building

Envelopes: What Material Properties are Needed, How
Trustful Are the Predictions?

ABSTRACT: The physical models which form the basis for the different software tools used to predict heat,
air, and moisture response of building envelopes seem quite diverse. For example, various “potentials” are
used, and each individual potential demands apparently different properties. Anyhow, when digging into the
physics and confronting the theory with what is measurable, then an array of eight basic material charac-
teristics appear, all backed by an experimental rationale. Adapting that array to the potentials in use is not
a question of creating new properties but of implementing the basic ones into potential-adapted expres-
sions. A different problem, however, is that these basic properties are macroscopic in nature, i.e., represent
the complexity of a material at the micro-scale by one single “average” number. That introduces restrictions
as to the use of the property values measured experimentally. Hence, those restrictions are typically mixed
up with the inability of the actual software packages to represent reality in full detail. Too many times,
assumed incorrectness of the property values are blamed for causing the differences found between the
predicted and real heat, air, moisture response of envelope parts. That inability, instead, should convince
researchers and building engineers that the way to gain a well balanced understanding of the heat, air, and
moisture response of envelopes is not by modeling only but by combining modeling with testing and field
experience.
KEYWORDS: heat, air and moisture transfer, models, material properties, application

Introduction
In Europe, the history of full heat, air, and moisture modeling goes back to 1932 when Hirsch published
his book on drying technology 关1兴. That book was reworked in 1956 by Krischer and Kröll 关2兴. In his
book, Krischer treats the basics of combined heat and mass transfer and applies it to industrial drying. The
chapter on moisture transport sees the water vapor resistance factor and the moisture diffusivity introduced
as main material properties that govern drying by vapor diffusion and capillary suction. Krischer anyhow
was not the only one digging in the field of knowledge. In 1955, a German translation of a book written
by a Russian scientist, Lykow, was published by VEB Verlag Technik in the DDR 关3兴. Like Krischer,
Lykow also analyzed drying and came up with analogous properties. De Vries was the last of the founding
fathers, with his theory of combined heat and moisture transfer in porous materials, published in 1958 关4兴.
In 1958–1959, Glaser advanced a simple method to control interstitial condensation in cold store walls
关5–7兴. As most materials used in cold stores are noncapillary, as the walls are airtight and as the tempera-
ture and vapor pressure difference are close to constant, steady state vapor diffusion remained as the only
moisture transport mechanism involved. Thus, the method gave quite good results for those types of
envelopes. That motivated other authors to extend the Glaser rationale to building envelopes, which was a
step too far 关8兴. None of the conditions fulfilled by cold store walls in fact are typical for building
envelopes. Transiently loaded, air-tightness not guaranteed, many materials capillary active, gravity not
negligible, etc., though the Glaser approach was charming by its simplicity. In North America, a method
analogous to the Glaser rationale was adapted, based on research done during the 1930s 关9兴.
In the years that followed the 1950s, some authors upgraded the Glaser method 关10–12兴 while others
returned to the work by de Vries, trying to measure some of the material properties he introduced, and
translating the PDEs into numerical models and computer software 关13–24兴.

Manuscript received February 13, 2006; accepted for publication January 19, 2007; published online February 2007. Presented at
ASTM Symposium on Heat-Air-Moisture Transport: Measurements on Building Materials on 23 April 2006 in Toronto, Canada;
P. Mukhopadhyaya and M. Kumaran, Guest Editors.
1
Professor at the K. U. Leuven, Department of Civil Engineering, Laboratory of Building Physics–Kasteelpark Arenberg 40,
B-3001 Leuven, Belgium.

Copyright © 2007 by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959.

1
2 MOISTURE MEASUREMENT ON BUILDING MATERIALS

Combined Heat, Air, and Moisture Models

Combined heat, air, and moisture modeling is based on the paradigms of energy and mass conservation, in
combination with diffusive and convective flow equations and a few equations of state. One should thus
expect that all models look the same and use identical material properties. On first sight, however, in the
models actually commercialized, the heat and moisture balances look quite different. The Match® software
for example is based on the following PDEs 关25兴
Air balance not considered

⳵␪
Heat balance ␳c p = ⵜ共␭ ⵜ ␪兲 + hv ⵜ 共␦v ⵜ p兲 共1兲
⳵t

⳵X
Moisture balance ␳ = ⵜ共␦v ⵜ p + kw ⵜ s兲 共2兲
⳵t
with ␳ density of the moist material 共kg/ m3兲, ␪ temperature 共°C兲, ⵜ the nabla operator reflecting a
divergence when applied on a vector and a gradient when applied on a scalar, ␭ thermal conductivity
共W/共m.K兲兲, X moisture ratio 共kg/kg兲, hv heat of evaporation 共J/kg兲, ␦v vapor permeability 共s兲, p partial
water vapor pressure 共Pa兲, kw water permeability 共s兲 and s capillary suction 共Pa兲. Potentials in the moisture
balance are vapor pressure and capillary suction.
The Wufi® software in turn uses as PDEs 关26兴:
Air balance not considered

⳵H ⳵␪
Heat balance = ⵜ共␭ ⵜ ␪兲 + hv ⵜ 共␦v ⵜ 共␾psat兲兲 共3兲
⳵␪ ⳵t

dw ⳵ ␾
Moisture balance = ⵜ共␦v ⵜ 共␾psat兲 + D␾ ⵜ ␾兲 共4兲
d␾ ⳵ t
with H enthalpy in J/K, psat partial water vapor saturation pressure at a temperature ␪ 共Pa兲, ␾ relative
humidity 共-兲 and w moisture content 共kg/ m3兲. Potentials in the moisture balance are the product of relative
humidity and partial water vapor saturation pressure, which equals partial water vapor pressure and
relative humidity.
To compare with, in his article on simultaneous transfer of heat and moisture in porous media, De
Vries 关4兴 proposed as equations for combined heat and moisture transport:
⳵H
Heat balance = ␳w ⵜ 共DT,T ⵜ ␪ + DT,⌿ ⵜ ⌿兲 共5兲
⳵t

⳵w
Moisture balance = ␳w ⵜ 共D⌿,T ⵜ ␪ + D⌿,⌿ ⵜ ⌿兲 共6兲
⳵t
the potentials being temperature and volumetric moisture ratio 共⌿兲.

Properties Involved

Heat and Moisture

The right-hand side of all equations represents transfer of heat or moisture. Seven different material
properties and five different 共pseudo?兲 potentials are used to quantify the flows. While the potentials
temperature, partial water vapor pressure, relative humidity, capillary suction and moisture ratio are well
known, only two of the seven properties are evenly well known: thermal conductivity ␭ and water vapor
permeability ␦v. The left-hand side of the equations models storage. Normally that side should contain as
many storage defining characteristics as the number of potentials intervening at the transfer side. All
except one in both the heat and the moisture balance are so tiny that they may be overlooked. Match®
 HENS ON MODELING HEAT, AIR AND MOISTURE RESPONSE 3

specifies the characteristic that governs heat storage: volumetric specific heat capacity, the product of
specific heat capacity at constant pressure 共c p兲 and density 共␳兲. Hence, sensible enthalpy used in Wufi® and
by De Vries may be written as H = ␳c p␪, meaning that the thermal balances in the De Vries equation, Wufi®
and Match® are identical. Also the moisture storage in the two models does not differ. In fact, ␳ ⳵ X / ⳵t can
be written as ⳵w / ⳵t, which in turn may be transformed into 共dw / d␾兲共⳵␾ / ⳵t兲 with the derivative dw / d␾
equal to the slope of the sorption/desorption curve of the capillary porous material. That slope is called the
specific moisture content ␰␾w, a property which mirrors the specific heat capacity. Instead, a derivative for
suction can also be used 共共dw / ds兲共⳵s / ⳵t兲兲, giving as specific moisture content: ␰sw.
In both Wufi® and Match® partial water vapor pressure activates vapor transfer. Liquid transfer,
however, is mobilized by capillary suction in Match® and relative humidity in Wufi®. That choice in
Wufi® is based on Künzel’s simplification that Fickian-like diffusion is governed by the geometry of the
porous system only, tuning vapor permeability as a constant to be measured in a dry cup test, while the
increase in apparent vapor permeability at high relative humidity is caused by surface flow in the adsorbed
water layers covering the pore walls. That flow goes in a direction opposite to the gradient in relative
humidity.
Can both liquid parts be unified? The answer is yes. A gradient in relative humidity is easily rewritten
as ⵜ␾ = 共d␾ / ds兲 ⵜ s, which means that the transport properties km and D␾ are linked by:

ds
D␾ = kw 共7兲
d␾
In moist capillary porous materials, capillary suction close to equilibrium now obeys Thompson’s law:

s = ␳wRT ln共␾兲 共8兲

meaning that the derivative ds / d␾ is changeable into a function of relative humidity, which makes the
moisture transfer coefficient D␾ transposable into a moisture permeability kw. Thus, Wufi® and Match® use
essentially the same equations and demand the same material properties. Besides, in noncapillary porous
materials, suction does not exist. Anyhow, once interstitial condensation has brought the moisture ratio
close to saturation, external pressures and gravity become the movers.
Is a same unification doable with the De Vries equations? Let us return to the basics. Moisture
transport above freezing combines two flows: vapor and liquid. The vapor flow is typically described by a
Fickian-like diffusion equation:

gv = − ␦v ⵜ p 共9兲
That equation is easily transformed into a two potentials expression:

冉
gv = − ␦v ⵜ 共psat␾兲 = − ␦v ␾
dpsat
d␪
冊 冉
ⵜ ␪ + psat ⵜ ␾ = − ␳w ␦v
␾ dpsat
␳w d␪
ⵜ ␪ + ␦v
psat d␾
␳w d⌿
ⵜ⌿ 冊
= − ␳w共D⌿,␪
v
ⵜ ␪ + D⌿,⌿
v
ⵜ ⌿兲 共10兲
with:

␾ dpsat
v
D⌿,␪ = ␦v
␳w d␪

psat d␾ psat d␾ psat

v
D⌿,⌿ = ␦v = ␦v w = ␦v 共11兲
␳w d⌿ ␰␾ ds ␰sw
or, the vapor part of the strange moisture properties De Vries uses, combines quantities of state 共partial
water vapor saturation pressure as a function of temperature兲 with two typical vapor-related material
properties: vapor permeability and specific moisture content.
The liquid part stands for capillary flow:

gl = − km ⵜ s 共12兲
As for vapor, that equation could be rewritten as:
4 MOISTURE MEASUREMENT ON BUILDING MATERIALS

gl = − ␳w 冉 km ds
␳w d␪
ⵜ␪+
km ds
␳w d⌿
冊 冉 km
ⵜ ⌿ = − ␳w 0.17 10−3 ⵜ ␪ +
␳w
km ds
␳w d⌿
ⵜ ⌿ = − ␳w共D⌿,␪
l
冊
ⵜ ␪ + D⌿,⌿
l
ⵜ ⌿兲

共13兲
with

l km
D⌿,␪ = 0.17 10−3
␳w

km ds km
l
D⌿,⌿ = = 共14兲
␳w d⌿ ␰sw
Again, the two typical material properties appear: moisture permeability and specific moisture content.
When combining liquid and vapor, parallel flow is assumed, meaning that each cut through a material
crosses pores where vapor mitigates and pores where liquid moves. Summing up both flow equations
gives:

g = − ␳w关共D⌿,T
v l
+ D⌿,T 兲 ⵜ ␪ + 共D⌿,⌿
v l
+ D⌿,⌿ 兲 ⵜ ⌿兴 = − ␳w共D⌿,T ⵜ ␪ + D⌿,⌿ ⵜ ⌿兲 共15兲
with

D⌿,␪ =
1
冉␦␾
dpsat
␳w v d␪
+ 0.17 10−3km 冊
D⌿,⌿ =
1
␳w
冉
␳ w␦ v
d␾ psat km
+
ds ␰sw ␰sw
冊 共16兲

Match® and Wufi® do not consider the temperature driven liquid flow, arguing that it is minimal
compared to the vapor part. With that simplification in mind, the De Vries moisture balance transforms into
the Match® / Wufi® moisture equation.
In the thermal balance, three flows intermix: heat conduction, vapor related sensible enthalpy mitiga-
tion, and liquid related sensible enthalpy mitigation. The last two are typically set zero, which is acceptable
as long as both mass flows are very small. As heat source or sink we have a change of state from vapor to
liquid and vice versa. That change is governed by the divergent of the vapor flow, or:

␳c
⳵␪
⳵t
冋 冉 冊 冉 冊册
= ␳w ⵜ
␭
␳w
ⵜ␪ +ⵜ
␦v
␳w
ⵜp

冋冉 冊 冉
= ␳w ⵜ
␭
␳w
ⵜ ␪ + lb ⵜ ␦v
␾ dpsat
␳w d␪
psat
ⵜ ␪ + ␦v w ⵜ ⌿
␰␾
冊册
冋 冋冉
= ␳w ⵜ
␭
冊 册 冉
␳w
+ l b␦ v
␾ dpsat
␳w d␪
psat
ⵜ ␪ + ⵜ l b␦ v w ⵜ ⌿
␰␾
冊册 = ␳w ⵜ 共D␪,␪ ⵜ ␪ + D␪,␪ ⵜ ⌿兲 共17兲

with

D␪,␪ =
1
␳w
冉␭ + l b␦ v␾
dpsat
d␪
冊
psat
D⌿,⌿ = lb␦v 共18兲
␰␾w
Again the De Vries equation is nothing else but a different way of writing the same balance as Match® and
Wufi® do. Anyhow, the potential “volumetric moisture ratio” is not a real one as its value changes at each
interface between materials. Real potentials, instead, should be continuous at interfaces, except if that
interface itself forms a resistance.
Clearly, the De Vries model demands the same six basic material properties as Match® and Wufi®.
 HENS ON MODELING HEAT, AIR AND MOISTURE RESPONSE 5

TABLE 1—Array of basic material properties.

Storage Transport
General Density 共␳兲
Heat Specific heat capacity 共c p兲 Thermal conductivity 共␭兲
Air Air permeability 共km兲 or
Air permeance coefficient 共a兲
and flow exponent 共n兲
Moisture Specific moisture content 共␰兲 Vapor permeability 共␦v兲 or
Vapor vapor resistance factor 共␮兲
Liquid Moisture permeability 共km兲

Air
None of the three models scrutinized considers air flow. Some models already do. We refer to Janssens
关27兴 and the recent versions of hygIRC 关24兴, Latenite 关19兴, and Delphin 关22兴. In general, air mitigation per
unit of surface and time through an open-porous material is governed by:
ga = − ka ⵜ 共Pa − ␳agzūz兲 共19兲
with ka air permeability, g acceleration by gravity, z height above a zero level, and ūz the unit vector in
vertical direction. As air is hardly compressible under small pressure differences and storage barely plays,
except when very fast air pressure changes are imposed which is the case when sound waves pass, the
divergent of the flow may be set zero, giving as a balance:
ⵜ关ka ⵜ 共Pa − ␳agzūz兲兴 = 0 共20兲
To describe air flow in cavities, air layers, cracks, joints, holes, voids, and fractionated layers, a flow
equation of the form ga = −a⌬共Pa − ␳agzūz兲n may be used with a the air permeance coefficient
共kg/ 共m2 . s . Pan兲 and n the flow exponent. With those two flow equations, solving a problem of air miti-
gation means transforming the building detail under scrutiny into a hydraulic network and applying
continuity in each of the nodes in the network. Hence, air flow adds a seventh 共and eighth兲 material
property to the list: air permeability or the air permeance coefficient and the flow exponent.
That hydraulic network methodology works fine as long as the air spaces are not too wide. If so, CFD
should be used to describe the complex air flows which may develop in wide air spaces.

Property Array
Table 1 summarizes the material properties needed to describe combined heat, air, and moisture transfer in
building parts. For most of them, appropriate measuring methods exist 关28–30兴.
Anyhow, before a full hygrothermal evaluation can be done, also all equations of state, the exact
geometry, the composition of the part or detail and some specific characteristics such as contact resistances
between layers, the network of cracks, air layers, voids a.o. in the part when present, etc., have to be
known.

What Do These Properties Reflect?

In General
Most models transform materials into a continuum that mixes up matrix and pores. Real materials, how-
ever, may consist of granules, bound by an intermediary paste. They may form a continuous, perforated
mass, have a bubble-like structure or consist of fibers glued together in the contact points. Heat is
conducted by the matrix and the air in the pores; the last in combination with radiation and sometimes
convection in medium and larger voids. The fact that matrix and pores contribute together, explains why
the thermal properties typify a material quite nicely. Round robin tests fit well and, if deviations are
detected, they practically always are a consequence of inaccuracies or errors in measurement.
Mass flows instead develop in the pores. Their magnitude is shaped by the geometry of the porous
system. As that geometry may differ between specimens of a same material, the air and moisture properties
show large standard deviations and do not univocally typify a material. Results of round robin tests are
6 MOISTURE MEASUREMENT ON BUILDING MATERIALS

FIG. 1—Path length.

quite disappointing, even when the samples distributed come from the same batch. The spread encountered
cannot be explained by measuring inaccuracies only, also differences in pore structure are responsible.

Water Vapor Permeability and Water Vapor Resistance Factor

Further discussion is restricted to the water vapor permeability and water vapor resistance factor. For a dry
porous material, the water vapor permeability depends on two parameters: water vapor permeability of
stagnant air and geometry of the porous system. Krischer and Kröll 关2兴 differentiated between both by
introducing the water vapor resistance factor, symbol ␮. Scientifically, the water vapor resistance factor
represents the ratio between the water vapor permeability of a layer of stagnant air and the water vapor
permeability of the material:
␦v,a
␮= 共21兲
␦v
As such, its value is shaped by the pore geometry. In fact, the only thing that counts in the case of
Fickian-like diffusion is the open surface available for the water molecules to pass through. For straight
pores with constant section, orthogonal to the surface and parallel to the gradient in partial water vapor
pressure, the water vapor resistance factor therefore equals the inverse of open porosity 共⌿o兲:
1
␮= 共22兲
⌿
If the straight pores make an angle ␣ with the normal on the surface, then the path length increases and the
water vapor resistance factor augments to:

␮=
1
冉 冊
1
⌿ cos ␣
共23兲

with ␣ the angle between the orthogonal and the inclined path 共=do / lo, see Fig. 1兲. When the section of the

再 冋 兺 冉 冊册冎
straight pores changes along the path of diffusion, the water vapor resistance factor becomes:
n m
1 ⌿ dj
␮= 兺
⌿ cos ␣ i=1 Aj
共24兲
j=1

The part between the brackets 兵 其 in the formulas is called the tortuosity of the porous system with a
symbol ⌿T. For pores crossing each other perpendicularly, transverse diffusion develops as soon as the
pores in the main direction change in shape along the path length. If shape distribution is stochastic and
pores cross the material in all directions, isotropic behavior could be expected if the representative volume
of the material under scrutiny is infinitesimally small and equal in pore system. Such material, however,
does not exist. Typically, equal representative volumes will not be found and isotropic, homogeneous
behavior will figure as an assumption, not as reality. Brick, timber, concrete, and aerated concrete are
examples of that. They all show a large spread in values for different samples of a same batch of material,
as is shown in Fig. 2. That figure gives the results of a dry cup measurement on 30 brick samples of a same
production batch.
For calculations, typically the average value of all water vapor resistance factors, measured at a same
relative humidity on several samples of a material, is used. If the sample thickness equals the layer
thickness, then the number of samples, divided by the harmonic average of all data is a better choice. One
 HENS ON MODELING HEAT, AIR AND MOISTURE RESPONSE 7

FIG. 2—Dry cup water vapor resistance factor for handmade bricks. Results for 30 samples from a same
batch (from Roels et al. [30]).

in fact may expect the same diversity as found between the samples along the surface of a layer of the
tested material, which means that the vapor will traverse a mosaic of sample large surfaces in parallel. For
the brick of Fig. 1, the arithmetic mean totals 22.1 while the harmonic mean equals 21, a difference of
5.3 %. If the sample thickness differs from the layer thickness then calculating a correct average becomes
a random exercise. The main consequence of that diversity in water vapor resistance factors for a same
material is that one-dimensional vapor flow through a flat layer is nonexisting. One will always have spots
where more and less water vapor diffuses through than the average.
If the water vapor diffusion resistance factor is a geometric characteristic of the pore system, one
should expect it to be constant, independent of relative humidity. Remarkably, this is not true. Even at low
relative humidity, the molecule thick layers, adsorbed against the pore walls of a sorption-active material,
already reduce open porosity. That reduction anyhow is so small that hardly any increase in water vapor
diffusion resistance factor will be seen. But, once capillary condensation starts, diffusion in the pores shifts
from an unhampered process to a jump from water island to water island. As a result, the path length
shortens, which is translated in a decrease of the water vapor diffusion resistance factor with increasing
relative humidity, as measured data on sand-lime stone clearly show, see Fig. 3.
But, jumping from water island to water island only does not explain the sharpness of the decrease at
very high relative humidity. Also a relative humidity driven parallel flow of absorbed water intervenes,
driving moisture back to the low relative humidity locations. In an isothermal diffusion test, relative
humidity and vapor pressure both point in the same direction, resulting in that sharp decrease of the water
vapor resistance factor. That pointing in the same direction is no longer true in a nonisothermal situation.
There, they may oppose. That fact was to some extend accounted for by the upgraded Glaser tool of the
1970s. That introduced the concept of critical moisture content. Below, only vapor moved in capillary
porous materials. Above, vapor and liquid moved, with the liquid taking the lead and displacing from
higher to lower relative humidity. Anyhow, if the above theory holds, noncapillary, hardly hygroscopic
materials should be characterized by a water vapor resistance factor independent of relative humidity.
Measurements on cork and polystyrene are proof of that as shown in Figs. 4 and 5.

FIG. 3—Water vapor resistance of sand-lime brick as a function of relative humidity (from Hens [12]).
8 MOISTURE MEASUREMENT ON BUILDING MATERIALS

FIG. 4—Water vapor resistance of cork as a function of relative humidity (from Hens [12]).

But, the water vapor resistance factor not only depends on relative humidity. Also temperature inter-
venes. In pores, which are thinner than the free path length of the water molecule, Fickian diffusion
changes into friction diffusion, also called Knudsen diffusion. While the Fickian vapor permeability of
stagnant air increases with temperature to an exponent 0.81, the Knudsen vapor permeability in a fine pore
decreases with temperature to an exponent 0.5:

␦v,K ÷ 冑 1
T
versus ␦v,F ÷ 冉 T
273.15
冊 0.81
共25兲

Traditionally, fine-porous materials are evaluated as if Fickian diffusion does the job. In reality,
Knudsen diffusion may have quite some impact, turning the water vapor resistance factor into an upward
function of temperature. Also temperature related changes in porous structure may have an impact on the
temperature dependency of the water vapor resistance factor.

Usage?
In the hygroscopic region, mass flows are assumed to engage water vapor only, be it by diffusion or by
advection, the combination of diffusion and bulk flow of vapor within moving air. Advection is mostly
neglected in massive walls composed of fine-porous materials, which leaves diffusion as the only mover.
So, mass flow calculations are performed, using measured water vapor resistance factors and measured
specific moisture ratios for the materials involved, which are all considered as perfectly homogeneous and
isotropic, while ideal diffusion contacts between layers is taken for granted.
In the capillary region, where liquid flow as a consequence of suction overwhelms diffusion in cap-
illary active porous materials, the same hypotheses are advanced: all materials homogeneous and isotropic,
ideal suction contact between layers.
Does all this fit with reality? Of course not. To start with, massive layers such as masonry, roof covers,
and lathed finishes are not homogeneous in nature. They consist of blocks and mortar joints, tiles and
overlaps, timber laths with groove and tongue joints. Mortar joints for example include voids at random

FIG. 5—Water vapor resistance of extruded polystyrene as a function of relative humidity (from Hens
[12]).
 HENS ON MODELING HEAT, AIR AND MOISTURE RESPONSE 9

TABLE 2—Water vapor diffusion resistance factor of massive concrete blocks and veneer walls, brick-laid with
the blocks (l⫻b⫻h⫽0.29⫻0.09⫻0.19 m3, density 2153 kg/m3, mortar joints 10-mm wide, from Hens [31]).

Veneer Wall, Brick-laid with the Blocks

Concrete Blocks RH Eq, Water Vapor Resistance Factor
␮50a Wall % ␮eq
50 1 59 3.0
50 2 57 3.2
a
Water vapor diffusion resistance for a relative humidity of 58 %.

and the head joints show microscopic cracks at the interfaces with the blocks. These voids and cracks
lower the equivalent water vapor diffusion resistance of veneer walls, inside leafs and massive walls
considerably compared to the water vapor diffusion resistance of the blocks used, as is shown in Table 2
关31兴.
At the same time, the veneer’s air permeability cannot be neglected, while capillary suction will be
dominated by the cracks. Cracks and voids may also activate gravity flow at the cavity side when rain runs
off. The wall should only have a cavity between the fill and the veneer but may in reality contain two air
layers, one where it should be and one between the fill and the inside leaf, activating air looping around the
insulation if containing leaks up and down. All block are perforated, which creates as many air flow paths
as there are perforations when the horizontal joints are badly filled 共Fig. 6兲. All that should withhold
modelers from tackling that type of wall with today’s advanced tools, which all assume homogeneity and
isotropy. As a consequence, measured data may deviate substantially from simulated results, a reality
which is overlooked by many authors, who blame the material property data to be the reason for the
discrepancies. They call them unreliable. The same holds for cathedralized roofs with tiled or slated decks
and a groove and tongue timber lathed inside lining 关27兴.
Things are even more complicated. Future weather is unknown. So, one has to use weather files from
the past to predict things that are coming. The inside climate is a random variable, especially in relation to
indoor humidity. Its ever moving value will depend on future outside climate, building usage, and venti-
lation habits, on the vapor sources present indoors, on interior buffering, etc. Contacts between materials
in a part are never ideal. As long as two layers only contact one another, continuity of diffusion is granted
but not continuity of suction. In case layers are poured together or in case a layer is pargetted against
another, a contact may be formed with sometimes quite different properties than the two layers in separate.
Real contacts may mix up both cases. Some work has been done on that 关32兴.
Actual models also have problems with wind-driven rain. How rain touches a surface is quite well
predictable, at least under steady state wind pressure differences 关33兴. As long as the impinging droplets
are absorbed, models master things quite well 关34兴. However, when run-off starts, preferential paths are
formed and the water moving down may get concentrated or diluted and stick on horizontal surfaces,
where it may move inward through random leaks and cracks by gravity and air pressure differences. This

FIG. 6—Difference between a virtual cavity wall, as calculated by a 1D heat, air, and moisture model, and
a real cavity wall!
10 MOISTURE MEASUREMENT ON BUILDING MATERIALS

is not tractable with the actual 1D or 2D heat, air, moisture models, as one never knows in advance where
the cracks, open joints, voids, unexpected air layers, etc. will be, and as no model counts for these random
gravity and pressure effects.
A weak point is also that most models are deterministic in nature. They calculate an assumed situation,
with geometry, material properties, in and outside climate, contact conditions, initial conditions all well
known. As said, this is fiction. Models should evaluate probability and risk, rather than a fixed scenario.
In other words, the material properties as defined and measured, do not allow simulation of how real
constructions will behave under combined heat, air, and moisture loading, except in some very specific
cases of extremely simple parts.

Conclusions
Today, many so-called full models are available to simulate heat, air, and moisture response of envelope
parts. On first sight, the balance equations used and the materials properties needed seem sometimes quite
different. When digging in the basics behind, however, it becomes clear that all use the same basic material
properties. While the thermal properties are quite representative for a given material, the mass transfer
properties may show large standard deviations for a same material. This is due to the fact they do not
characterize the material but the porous system, which may be quite variable between samples of a same
batch of material samples and interacts intensively with the relative humidity in the surrounding air, giving
rise to adsorption and capillary condensation.
Things even become more complicated when looking to real envelope parts. They do not behave as the
idealized constructions, models simulate, composed of homogeneous materials, with ideal diffusion and
suction contacts in between. Simulating their heat, air, and moisture response with the models available
today may show results which deviate largely from the actual response. Blaming the quality of the material
data for these discrepancies is not correct. One should instead critically scrutinize the models and work on
upgrades which include material heterogeneity, account for air transport, consider gravity and external
pressure induced flow through cracks, voids and air layers, take real initial conditions into account and
evaluate probability and risk.
All this proves we need a healthy mix between modeling, experiment, and field experience in com-
bined heat, air, and moisture work. Modeling helps in gaining insight, experiments allow validating some
of the steps in modeling and understanding how complex even a simplified test reality is, while field
experience is needed to gain a reliable, well balanced knowledge of hygrothermal facts and figures.

References

关1兴 Hirsch, M. , 1932, Trocknungstechnik, Springer Verlag, Berlin/Gýttingen/Heidelberg

关2兴 Krischer, O., and Kröll, K., Die wissenschaftlichen Grundlagen der Trocknungstechnik, Springer
Verlag, Berlin/Göttingen/Heidelberg, 1957.
关3兴 Lykow, A. W., Transporterscheinungen in kapillarporösen Körpern, Akademie Verlag, Berlin, 1955.
关4兴 De Vries, D. A., “Simultaneous Transfer of Heat and Moisture in Porous Media,” EOS Trans. Am.
Geophys. Union, Vol. 39, No. 5, October 1958.
关5兴 Glaser, H., Wärmeleitung und Feuchtigkeitsdurchgang durch Kaltraumisolierungen, Kältetechnik,
Vol. 3, 1958, pp. 86–91.
关6兴 Glaser, H., Vereinfachte Berechnung des Dampfdiffusion durch geschichtete Wände bei Ausscheiden
von Wasser und Eis, Kältetechnik, Vol. 11, 1959, pp. 358–364, Kältetechnik, Vol. 12, 1959,
pp. 386–390.
关7兴 Glaser, H., Grafisches Verfahren zur Untersuchung von Diffusionsvorgénge, Kältetechnik, Vol. 10,
1959, pp. 345–349.
关8兴 Seiffert, K., Wasserdampfdiffusion im Bauwesen, Bauverlag, Wiesbaden/Berlin, 1967.
关9兴 Rowley, F. B., “A Theory Covering the Transfer of Vapor Through Materials,” ASHVE Transactions,
No. 1134, July 1938.
关10兴 Vos, B. H., “Internal Condensation in Structures,” Build. Sci., Vol. 3, 1969, pp. 191–206.
关11兴 Vos, B. H., “Condensation in Flat Roofs Under Nonsteady State Conditions,” Build. Sci., Vol. 7,
1971, pp. 7–15.
 HENS ON MODELING HEAT, AIR AND MOISTURE RESPONSE 11

关12兴 Hens, H., “Theoretical and Experimental Study of the Hygrothermal Behavior of Building and
Insulating Materials During Interstitial Condensation and Drying, with Application on Flat Roofs,”
Ph.D. Thesis, K. U. Leuven 共in Dutch兲, 1975.
关13兴 Van der Kooi, J., “Moisture Transport in Cellular Concrete Roofs,” Ph.D. Thesis, Uitgeverij Walt-
man, Delft, 1971.
关14兴 Nielsen, A. F., “Moisture Distribution in Cellular Concrete During Heat and Moisture Transfer,”
Ph.D. Thesis, Thermal Insulation Laboratory, Technical University of Denmark, 1974.
关15兴 Kießl, K., “Kapillarer und dampformiger Feuchtetransport in mehrschichtigen Bauteilen,” Ph.D.
Thesis, Universität und Gesamthochschüle Essen, 1983.
关16兴 Crausse, P., “Etude fondamentale des transfers couples de chaleur et d’humidité milieu poreux
non-saturé,” Thèse de doctorat, ENSEEIHT, Toulouse, 1983.
关17兴 Kohonen, R., “A Method to Analyze the Transient Hygrothermal Behavior of Building Materials and
Components,” Technical Research Center of Finland 共VTT兲, Publication 21, 1984.
关18兴 Duforestel, T., “Bases métrologiques et modèles pour la simulation du comportement hygrother-
mique des composants du bâtiment,” Thése de doctorat, Ecole Nationale des Ponts et des Chaussées,
Paris, 1992.
关19兴 Karagiozis, A., “Overview of the 2-D Hygrothermal Heat-moisture Transport Model Latenite,”
Internal IRC/BPL report, IRC/NRC, 1993.
关20兴 Hens, H., “Heat, Air and Moisture Transfer in Insulated Envelope Parts, Modeling,” Vol. 1 of the
final report of the IEA-ECBCS Annex 24. ACCO, Leuven, 1996.
关21兴 Matsumoto, M. and Nagai, H., “An Analysis of Moisture Variations in Building Walls by Quasi
Linearized Equations for Nonlinear Heat and Moisture Transport,” Proceedings of the CIB-W40
Meeting, Borås, Sweden, 1985.
关22兴 Grünewald, J., “Diffusiver und konvektiver Stoff- und energietransport in Kapillar-porösen Baustof-
fen,” Ph.D. Thesis, Technische Universität Dresden, 1997.
关23兴 Hagentoft, C. E., “Hamstad-Modelling,” final report EU Hamstad Project, 2002.
关24兴 Maref, W., Kumaran, M. K., Lacasse, M. K., Swinton, M. C., and Van Reenen, D., “Advanced
Hygrothermal Model-HygIRC, LMBF,” Proceedings of the 12th International Heat Transfer Con-
ference,” , Grenoble, 2002, pp. 1–6.
关25兴 Rode, C., “Combined Heat and Moisture Transfer in Building Constructions,” Thermal Insulation
Laboratory, Technical University of Denmark, 1990.
关26兴 Künzel, H. M., “Verfahren zure in- und zweidimensionalen Berechnung des gekoppelten Wärme-
Und Feuchtetransports in Bauteilen mit einfachen Kennwerte,” Ph.D. Thesis, Universität Stuttgart,
1994.
关27兴 Janssens, A., “Reliable Control of Interstitial Condensation in Lightweight Roof Systems,” Ph.D.
Thesis, K. U. Leuven, 1998.
关28兴 Krus, M., “Feuchtetransport- Und Speicherkoeffizienten poröser mineralischer Baustoffe. Theore-
tische Grundlagen und neue Meßtechniken,” Ph.D. Thesis, Universität Stuttgart, 1995.
关29兴 Descamps, F., “Continuum and Discrete Modeling of Isothermal Water and Air Transport in Porous
Media,” Ph.D. Thesis, K. U. Leuven, 1997.
关30兴 Roels, S., Carmeliet, J., and Hens, H., “Moisture Transfer Properties and Material Characterization,”
EU Hamstad Project, Final Report WP1, 2002.
关31兴 Hens, H., “The Vapor Diffusion Resistance and Air Permeance of Masonry and Roofing Systems,”
Buildings and Environment, Vol. 41, June 2006, pp. 745–755.
关32兴 Qiu, X., “Moisture Transport Across Interfaces Between Building Materials,” Ph.D. Thesis, Concor-
dia University, 2003.
关33兴 Blocken, B., “Wind-Driven Rain on Buildings: Measurements, Numerical Modeling and Applica-
tions,” Ph.D. Thesis, K. U. Leuven, 2004.
关34兴 Janssen, H. and Carmeliet, J., “Hygrothermal Simulation of Masonry Under Atmospheric Excita-
tion,” Research in Building Physics and Building Engineering, P. Fazio, H. Ge, J. Rao, and G.
Desmarais, Eds., Taylor and Francis, London/Leiden/New York/Philadelphia/Singapore, 2006,
pp. 77–83.
 Journal of ASTM International, Vol. 4, No. 1
Paper ID JAI100335
Available online at www.astm.org

Lisbeth G. Thygesen1 and Kurt K. Hansen1

Improved Suction Technique for the Characterization of

Construction Materials

ABSTRACT: The suction technique is a method from soil science that is used for the study of moisture
storage capacity in porous construction materials at high relative humidity levels 共above approximately
93 %兲. The samples to be studied are placed in a pressurized container 共an extractor兲 on a water saturated
ceramic disk or membrane. The pressure corresponding to a certain relative humidity level is established
and moisture is expelled from the samples until moisture equilibrium is reached. This paper presents two
adjustments to this method and their applicability. The first adjustment is a new procedure for determining
the equilibrium moisture content of samples during the experiment. The method 共referred to as the balance
method兲 is to simply keep track of all water entering and leaving the system and use these data to estimate
the moisture content of the samples. The advantage of this approach is that the experiment does not need
to be stopped and restarted for each relative humidity level tested. The main disadvantage of the method
is that the estimation of the moisture content in the samples depends on a correct estimation of the
moisture content of the ceramic disk. The second adjustment is the development of a pressure regulation
system that keeps the pressure constant 共⫾0.015 bar兲 within an extractor for weeks.
KEYWORDS: suction technique, porous construction materials, moisture storage capacity, moisture
content

Nomenclature

␸ ⫽ Relative humidity 关⫺兴

M ⫽ Molar mass of water 关kg mole−1兴
␴ ⫽ Surface tension of water 关N m−1兴
␪ ⫽ Contact angle 关rad兴
r ⫽ Radius 关m兴
␳ ⫽ Density of water 关kg m−3兴
R ⫽ Universal gas constant 关J共K · mol兲−1兴
T ⫽ Absolute temperature 关K兴
s ⫽ Suction, i.e., the pressure difference between the pore water pressure and the ambient
total pressure 关Pa兴

Introduction
The suction technique is an experimental method for studying the relationship between the relative hu-
midity and the equilibrium moisture content of porous materials at very high relative humidity levels.
Typically, the purpose is to establish the upper part of the sorption isotherm 共i.e., above approximately 93
% relative humidity兲, but it could also be to produce samples with a high moisture content to be studied
using other techniques. At humidity levels above approximately 96–97 %, climate chambers are normally
not an option. When exposed to rain, construction materials may, however, experience such conditions

Manuscript received January 30, 2006; accepted for publication November 8, 2006; published online December 2006. Presented
at ASTM Symposium on Heat-Air-Moisture Transport: Measurements on Building Materials on 23 April 2006 in Toronto, Canada;
P. Mukhopadhyaya and M. Kurmaran, Guest Editors.
1
Associate Professors, Technical University of Denmark, Department of Civil Engineering, Building 118, DK-2800 Kgs. Lyngby,
Denmark, e-mail: [email protected], [email protected]

Copyright © 2007 by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959.

12
 THYGESEN AND HANSEN ON SUCTION TECHNIQUE FOR CONSTRUCTION MATERIALS 13

during their service life, and it is consequently of interest to study materials at these high humidity levels.
The equilibrium moisture contents of porous materials, such as brick or wood, at high moisture contents
may also be of importance during drying. Within basic research, the structure or properties of porous
materials at high moisture contents are interesting per se.
The suction technique was originally developed to study soil samples 关1,2兴, but the method has been
adapted to both inorganic and organic construction materials . The technique implies that one or more
water saturated samples are placed in a container 共an extractor兲. Pressure is applied, and moisture is forced
out of the samples through a water conductive medium and collected outside the extractor. Equilibrium has
been reached when no more water leaves the samples. The equilibrium relative humidity is calculated from
the applied pressure 共the higher the pressure, the lower the corresponding relative humidity兲 using the
Kelvin and la Place equations 共Eqs 1 and 2, respectively兲.
2 · ␴ · M · cos ␪
ln ␸ = − 共1兲
r·␳·R·T

2 · ␴ · cos ␪
s=− 共2兲
r
The Kelvin equation gives the relationship between the relative humidity and the meniscus radius
共identical to the pore radius兲, while the la Place equation gives the relationship between the applied
pressure and the pore radius. Thus, by combining the two equations the relationship between the applied
pressure and the relative humidity may be established:
ln ␸ · ␳ · R · T
s= 共3兲
M
Since the suction technique implies that moisture is removed from initially water saturated samples,
only the desorption isotherm may be studied.
Equipment for suction studies is intended for certain pressure ranges, for example, three different types
of extractors relevant for construction materials are available from SOILMOISTURE®2: Up to 5 bar
共above 99.7 % RH兲, up to 15 bar 共above approximately 99 % RH兲, and 100 bar 共above approximately 93
% RH兲. The water conductive medium, which connects the sample pore system to the tube for expelling
water out of the pressurized container, depends on the pressure range of the system. For the 100 bar system
the samples are placed on a water saturated cellulose membrane, while the other two systems use a water
saturated ceramic disk covered by a layer of wet kaolin and a wet textile cloth. The cloth is meant to
prevent kaolin from entering the samples.
This paper describes two improvements to the way suction experiments are normally performed. The
first improvement is an alternative way of registering whether the equilibrium moisture content has been
reached. The new procedure was only tested for 5 and 15 bar systems. The second improvement is a new
pressure regulation system that keeps the pressure constant in a 100 bar extractor.

The Balance Method

The traditional method for determining when equilibrium has been reached is described in a Nordtest
method for suction analysis 关7兴. This method measures only the change in the volume of water leaving the
sample. The expelled water is led into a burette, and when there has been no change for at least 48 h the
pressure is released, and samples are removed and weighed. To obtain the moisture content at a higher
pressure level the samples are put back into the extractor. Studies have demonstrated that it is necessary to
use a fresh layer of kaolin and a new water saturated ceramic disk to ensure the hydraulic contact between
the porous system of the sample and the kaolin layer 关8,9兴. This method is referred to here as the volume
method.
The newly developed method instead uses the weight of the water that has left the system to monitor
whether equilibrium has been reached. By applying strict control of the amount of water in the system it
is possible to calculate the moisture content of the sample without stopping the experiment and removing
2
Soilmoisture, Santa Barbara, CA.
14 MOISTURE MEASUREMENT ON BUILDING MATERIALS

the samples. To switch to a higher pressure 共i.e., a lower RH level兲 the pressure is simply raised without
having to stop and restart the experiment. This method is refered to as the balance method.
Characteristics of the volume method:
• The method is an established test method.
• Values for individual samples may be obtained also when they are placed on the same disk/
membrane. Samples of different materials may thus be placed on the same disk, if needed.
• Control of the amount of water entering the system is not needed.
• No estimation of the water content in the disk/membrane is needed.
Characteristics of the balance method:
• The experiment does not need to be stopped and restarted to measure the moisture content of the
samples corresponding to a certain pressure level, eliminating possible hysteresis effects from
cycles of increasing and releasing pressure. Additionally, the samples remain in the extractor and
therefore are not exposed to laboratory conditions which cause the samples to lose moisture as soon
as the pressure is released. This exposure makes it difficult to obtain correct weight measurements,
and implies that the samples, in a worst case scenario, risk losing so much moisture that their
moisture content falls below the next equilibrium moisture content level.
• The weight of the water that has left the system may only be recorded for a whole disk/membrane.
Therefore, if data are needed for individual samples, only one sample may be placed on each
disk/membrane. However, if the mean characteristics of a given material is sought, a number of
similar samples may be placed on the same disk/membrane and their mean moisture content may
then be found 共but not the standard deviation between the individual samples兲.
• Complete knowledge of all water entering and leaving the system is a requirement. This means that
during setup of the system, the dry and wet weights of all components need to be measured 共disk,
kaolin, cloth, etc.兲.
• The balance method requires a “blank” disk/membrane to be tested in order to estimate the amount
of water present within the system but outside the sample共s兲 at different pressure levels. Provided
that the same kaolin layer thickness is used for all runs these estimates need only be established
once, and may then be used henceforth.
Once established in the laboratory the balance method is less time consuming, as switching from one
relative humidity level to the next is simply a matter of raising the pressure. However, initially the method
requires collection of some background data. The most important background data are an estimate of the
equilibrium moisture content in the water conductive system inside the extractor, i.e., in the membrane or
in the disk, kaolin layer, and the cloth. Without the estimation, it is not possible to determine how much of
the water in the extractor is contained by the sample共s兲. Another estimate needed is the rate of evaporation
of water from the container that collects the water outside the extractor. The rate of evaporation is needed
to estimate the amount of water leaving the system through evaporation and to ensure the experiment is
stopped at the right time. If evaporation is ignored there is a risk of stopping the experiment prematurely.
The following two subsections describe these two estimations.

Estimation of the Moisture Content Within the Ceramic Disk, Kaolin Layer, and Cloth
The balance method calculates the amount of water within the samples by monitoring the amount of water
entering and exiting the system. Consequently, the amount of water inside the extractor that is not con-
tained by the samples 共i.e., within the disk, kaolin, or cloth兲 must be known. This can be estimated by
performing experiments without samples. In order to verify estimates based on the balance method the
suction experiments were stopped at each equilibrium moisture content level and the ceramic disks 共in-
cluding kaolin layer and cloth兲 were removed from the extractor, wrapped in plastic to prevent evapora-
tion, and weighed. The results are given in Tables 1 and 2.
Tables 1 and 2 show that individual ceramic disks have different equilibrium moisture contents. Even
those intended for the same pressure range vary. This means that the equilibrium moisture content of a
particular disk must be estimated in order to use the balance method. The tables also show that the
estimates from the balance method are a little higher than corresponding measurements from the volume
method. The differences are up to 3.4 %. The cause of the systematic error on the estimates is unknown,
but on several occasions drops of condensed water were observed on the inside of the extractor lid. Also
the disks will undoubtedly start losing moisture as soon as the pressure is released, and will continue doing
 THYGESEN AND HANSEN ON SUCTION TECHNIQUE FOR CONSTRUCTION MATERIALS 15

TABLE 1—Five bar extractor: moisture contents (MC) of ceramic disks (including kaolin layer and cloth) according to the volume method (V)
and the balance method (B) at five different relative humidity levels.

Disk I Disk II Disk III Disk IV

s ␸ MC 关%兴 MC 关%兴 MC 关%兴 MC 关%兴
关bar兴 关%兴 V B V B V B V B
1.0 99.93 19.3 19.4 18.4 18.4 21.4 21.6 18.6 18.8
1.4 99.90 19.2 19.4 18.4 18.4 21.1 21.4 18.5 18.8
2.0 99.85 18.9 19.1 18.3 18.4 20.9 21.3 18.3 18.7
2.7 99.80 18.6 18.9 18.2 18.3 20.7 21.3 18.1 18.6
3.4 99.75 18.6 18.9 18.1 18.3 20.6 21.3 18.0 18.6

so while the extractor lid is unscrewed. This implies that the balance method most likely gives better
estimates of the total amount of water inside the extractor but outside the samples. However, when
compared to the moisture content of the disk, kaolin, and cloth determined directly from weighing, the
estimates from the balance method appear to be too high.
The moisture present in the water conductive system inside the extractor is distributed between the
ceramic disk, the kaolin layer, and the cloth that make up the sandwich on which the sample is placed. This
implies that unless the exact same amount of kaolin and the same size and type of cloth is used for every
experiment, the estimates from Tables 1 and 2 are useless. In order to minimize the differences between
runs, the setup procedure needs to be standardized. For example, a special spatula with a 2 mm gap is used
to apply the kaolin layer to ensure that the layer thickness is the same every time a disk is prepared.

Evaporation of Water through a Floating Lid of Hexadecane

The criterion for equilibrium is the same for the two methods, i.e., no change in the amount of expelled
water for at least 48 h. However, initial tests showed that one cannot ignore the amount of water that
evaporates from the container in which it is collected. A floating lid of n-hexadecane was used to reduce
evaporation; however, evaporation was not completely eliminated. In order to be able to correct the weight
of the expelled water for the evaporation it was thus necessary to quantify the evaporation through the
hexadecane lid. Four different layer thicknesses were tested 共approximately 5, 10, 12, and 20 mm兲 in a
room with a temperature of 20 ° C and a relative humidity of 65 % from cylindrical containers with a
surface area of 100 cm2. Figure 1 shows that a few days after the water has penetrated the lid, the
evaporation rate is surprisingly independent of the layer thickness. It appears that only the interface
between the water and the hexadecane layer reduces the evaporation from the water. Hexadecane does not
evaporate under the conditions used here 共no weight loss for two months兲. From Fig. 1 the evaporation rate
was estimated to be approximately 0.045 g per 100 cm2 per day regardless of the lid thickness, which was
utilized in the experiments using the balance method.

Results from Sandstone Samples

The balance method was tested for two different types of sandstones 共white and green兲. Prior to the
experiment the ceramic disks were wetted using capillary saturation, while the samples were water satu-
rated using vacuum saturation. The results are shown in Figs. 2 and 3. The figures show that the balance

TABLE 2—Fifteen bar extractor: moisture contents (MC) of ceramic disks (including kaolin layer and cloth) according to the volume method (V)
and the water balance method (B) at five different relative humidity levels.

Disk I Disk II Disk III Disk IV

s ␸ MC 关%兴 MC 关%兴 MC 关%兴 MC 关%兴
关bar兴 关%兴 V B V B V B V B
3.3 99.76 18.9 19.2 18.8 18.9 19.6 19.7 18.9 19.0
5.3 99.61 18.7 19.2 18.5 18.6 19.2 19.5 18.5 18.6
9.5 99.30 18.1 18.6 18.0 18.2 18.8 19.1 17.9 18.0
10.0 99.26 18.0 18.6 17.8 18.0 18.6 19.1 17.7 17.9
13.4 99.01 17.8 18.6 17.0 17.3 18.5 19.1 17.0 17.3
16 MOISTURE MEASUREMENT ON BUILDING MATERIALS

FIG. 1—Evaporation of water through a floating lid of hexadecane of four different thicknesses. The
experiment with a 20 mm layer thickness was started on Day 13, the others on Day 1. A few drops of
hexadecane were accidentally spilled from the 20 mm container on Day 49. The temperature in the room
was not entirely stable, which explains the fluctuations around Days 22–29 and 55–69.

method tends to give slightly higher moisture contents than the volume method, about 1–2 %. The figures
also indicate that the standard deviation between runs for the balance method is about twice that for the
volume method.
The systematic difference between the two methods likely arises because the volume method tends to
underestimate and the balance method tends to overestimate the moisture content. For the volume method,
the measured equilibrium weight of the samples at these high relative humidity levels can only be too low,
never too high, as the samples will start losing moisture as soon as the pressure is released. A test was
carried out using a green sandstone sample where the sample was removed from the extractor, weighed,
exposed to the laboratory climate for 10 min and weighed again. The evaporation from the sample resulted

FIG. 2—Moisture content for pressure levels below 12 bar for green sandstone samples. Results are shown
both for the volume method (V) and the balance method (B).
 THYGESEN AND HANSEN ON SUCTION TECHNIQUE FOR CONSTRUCTION MATERIALS 17

FIG. 3—Moisture content for pressure levels below 12 bar for white sandstone samples. Results are shown
both for the volume method (V) and the balance method (B).

in a 0.2 % decrease in the calculated moisture content. Obviously, this loss may be reduced by wrapping
the sample in plastic. The loss of moisture from the samples during the release of pressure is, however,
unknown and may not be prevented. For the balance method, the measured weight of the expelled water
can only be too low, never too high, leading to overestimation of the moisture content. The water, not
registered as expelled, is assumed to be inside the extractor with a certain part of that water expected to be
contained within the samples.
That the reproducibility from run to run is poorer for the balance method than for the volume method
is most likely a result of the uncertainty of the estimate of the equilibrium moisture content of the disks.
It is possible that even more rigorous standardization of the procedure could reduce this uncertainty
somewhat.
Results obtained indicate that for sandstone the balance method appears to be about as accurate as the
volume method, but unfortunately less precise.

Remaining Problems
The suction technique may sometimes give problems that are not mentioned in the Nordtest method 关7兴.
For example occasionally the 48 h criterion is not enough time to ensure equilibrium 关10兴. Samples, both
wood and sandstone, may appear to be in equilibrium, and may then start to lose moisture again. In one
case it was calculated that the error resulting from stopping the experiment too early corresponded to a
difference in the equilibrium moisture content of up to 0.3 % 关10兴.
A problem which limits the applicability of the water balance method is the uncertainty in estimating
the moisture content of the disk/membrane. This uncertainty implies that only samples above approxi-
mately a few hundred grams may be studied using this method. The reason is that the moisture content in
the sample共s兲 is determined from the difference between the total amount of water inside the extractor and
the amount present in the disk/membrane. For example, the dry weight of a ceramic disk including kaolin
and cloth is typically around 1000 g during a run. Therefore, an uncertainty of 0.5 % moisture corresponds
to 5 g of water, which only corresponds to 1 % for a sample with a dry weight of 500 g, but 10 % to a
sample with a dry weight of 50 g. This uncertainty is the main drawback of the balance method.

Pressure Regulation of a 100 Bar Extractor

It is important that suction measurements are carried out under well-defined and stable temperature con-
ditions. This is achieved by placing the extractor in a room with a constant temperature. It is also important
18 MOISTURE MEASUREMENT ON BUILDING MATERIALS

FIG. 4—Schematic drawing of the pressure regulation system for the 100 bar extractor.

that the pressure within the extractor does not fluctuate. In the earlier version of the 100 bar system in
question, the pressure decreased by approximately one bar per day caused when the regulator did not
respond to nitrogen escaping through fittings and diffusing through the cellulose membrane.
The new system described here keeps the pressure within a narrow range for several weeks. The
system consists of a pressure transmitter to measure the pressure and send the value to a PC running an
in-house Labview program. If the pressure is below the set point, the regulation software sends a signal to
a solenoid valve, which opens for a short while and lets in a small amount of nitrogen gas in order to
re-establish the pressure level.
The system is shown in Fig. 4. The procedure for establishing a pressure of, for example, 20 bar is as
follows: Valve V2 on the extractor is closed and V1 is opened. By opening the regulator R the pressure
throughout the system is manually adjusted to approximately 19.8 bar. Valve V1 is then closed. The
default position of the solenoid valve M is closed so when the pressure is now increased to 25 bar by
adjusting R, the regions between R and V1 and R and M are affected. The control valve C is slightly open.
The pressure transmitter T signals the PC that the pressure is too low. The response of the Labview
program is to signal the solid state relay S so that the solenoid valve may open for a short, predefined time
共0.5 s兲. Henceforth, the pressure is measured once a minute and regulated as needed. The precision test
gage G was used for calibrating the pressure transmitter, but is now only used for manual inspection.
Figure 5 shows that the pressure regulation keeps the pressure constant for more than a month. The
normally used cellulose membrane was not used during this test, a 0.15 mm polyethylene foil was used
instead. Figure 5 shows that even with a tight polyethylene foil in the extractor the pressure decreases due
to leakage in the system. Without regulation the pressure would decrease by approximately 0.15 bar per
day. The figure shows that when the decreasing pressure has reached the set point, which in this case is
20.00 bar, the pressure regulation system opens the solenoid valve for a short time increasing the pressure
by 0.025 bar. From this level the pressure decreases again. In this way the pressure oscillates ±0.015 bar
around a pressure of 20.01 bar with a cycle time of about two hours. By exchanging the polyethylene foil
with the normally used cellulose membrane, the cycle time is expected to be less than two hours due to
diffusion of nitrogen gas through the cellulose membrane.

Conclusions
Two different improvements to the suction method were developed. A procedure was tested that avoids
possible hysteresis effects on the sample and is less time consuming than the standard method. The method
was found to be about as accurate as the standard method, but less precise. It was also found that the new
method gives unreliable results for samples with a dry weight less than a few hundred grams. The other
improvement developed was a pressure regulation system. The new system ensures a constant pressure
level during suction experiments.
 THYGESEN AND HANSEN ON SUCTION TECHNIQUE FOR CONSTRUCTION MATERIALS 19

FIG. 5—The pressure regulation system keeps the pressure constant 共⫾0.015 bar兲 within the 100 bar
extractor for more than a month.

Acknowledgments
A grant from the Danish Research Council for Technology and Production Sciences is gratefully acknowl-
edged 共project no. 26-02-0100兲. The authors thank Bjarke Dahl Madsen and Ulla Gjøl Jakobsen for
technical assistance.

References

关1兴 ASTM, Standard D 2325, “Standard Test Method for Capillary-Moisture Relationships for Coarse-
and Medium-Textured Soils by Porous-Plate Apparatus” Annual Book of ASTM Standards, Vol. 4.08,
ASTM International, West Conshohocken, PA, 1968, 184–190.
关2兴 ASTM, Standard D 3152, “Standard Test Method for Capillary-Moisture Relationships for Fine-
Textured Soils by Pressure-Membrane Apparatus” Annual Book of ASTM Standards, Vol. 4.08,
ASTM International, West Conshohocken, PA, 1972, pp. 301–306.
关3兴 Krus, M., “Moisture Transport and Storage Coefficients of Porous Mineral Building Materials: The-
oretical Principles and New Test Methods,” 共in German兲, Der Fakultät Bauingenieur- und Vermes-
sungswesen der Universität Stuttgart, Stuttgart, Germany, 1995.
关4兴 Janz, M., “Moisture Transport and Fixation in Porous Materials at High Moisture Levels,” Doctoral
thesis, Report TVBM 1018, Division of Building Materials, Lund Institute of Technology, Lund,
Sweden, 2000.
关5兴 Johansson, P., “Water Absorption in Two-layer Masonry Systems. Properties, Profiles and Predic-
tions,” Doctoral Thesis, Report TVBM 1024, Division of Building Materials, Lund Institute of
Technology, Lund, Sweden, 2005.
关6兴 Cloutier, A., and Fortin, Y., “Moisture Content—Water Potential Relationship of Wood from Satu-
rated to Dry Conditions,” Wood Sci. Technol., Vol. 25, 1991, pp. 263–280.
关7兴 Nordtest Method, “Building Materials: Retention Curve and Pore Size Distribution,” NT BUILD
481, 1997.
关8兴 Strømdahl, K., “Moisture Content in Building Materials—With Focus on Moisture Fixation in the
Superhygroscopic Range,” 共in Danish兲, M.Sc. Thesis, Department of Structural Engineering and
Materials, Technical University of Denmark, Lyngby, Denmark, 1997.
20 MOISTURE MEASUREMENT ON BUILDING MATERIALS

关9兴 Hansen, M. H., Houvenagehl, G., Janz, M., Krus, M., and Strømdahl, K., “Interlaboratory Compari-
son of the Measurement of Retention Curves,” Journal of Thermal Envelope and Building Science,
Vol. 22, 1999, pp. 334–348.
关10兴 Christensen, K. and Strømdahl, K., “Experimental Determination of Suction Curves for Lime Silica
Brick by Use of Pressure Plate Extractor,” 共in Danish兲, Report, Department of Structural Engineering
and Materials, Technical University of Denmark, Lyngby, Denmark, 1996.
 Journal of ASTM International, Vol. 4, No. 8
Paper ID JAI100621
Available online at www.astm.org

Phalguni Mukhopadhyaya,1 Kumar Kumaran,1 John Lackey,1 and David van Reenen1

Water Vapor Transmission Measurement and Significance of

Corrections

ABSTRACT: Water vapor transmission properties of building materials play an important role in the overall
moisture management and durability of the exterior building envelopes. The cup method, as described in
the ASTM Standard Test Methods for Water Vapor Transmission of Materials 共E 96兲, is widely used in North
America and other parts of the world for this purpose. Recently the latest ASTM standard 共E 96/E 96M–05兲
has started taking into account various corrections 共e.g., buoyancy correction, correction for resistance due
to still air and specimen surface, edge mask correction, etc.兲 while analyzing the results obtained from the
cup methods. This paper presents the results obtained from the laboratory tests carried out on more than
50 building materials. These results have been used to demonstrate the significance of various corrections
on the measured water vapor permeability or permeance of various commonly used building materials or
components. The results presented in this paper were discussed in the ASTM technical task group to
underline the importance of various corrections for the calculation of water vapor transmission properties of
various building materials.
KEYWORDS: water vapor transmission, permeability, permeance, building materials, corrections

Introduction
Water vapor transmission characteristic of a building material is the most significant moisture transport
property that is looked upon for the assessment of moisture management capability of the building enve-
lope systems. Simply put, a building envelope designer or engineer would not recommend or use a
construction material without knowing the implications of its water vapor transmission characteristics on
the overall moisture response of the building envelope. On many instances, one material is preferred over
another based on its water vapor transmission properties 关1兴. In particular, for membrane and coating
materials 共e.g., vapor barrier, sheathing paper, latex coating, etc.兲 the water vapor transmission character-
istic is the single most important property that determines its utility and effectiveness from the moisture
management point of view 关2兴. Hence, one cannot overemphasize the need to evaluate the water vapor
transmission characteristics of building materials accurately by a standard test procedure.

Research Background
There are three terminologies that are commonly used to describe the water vapor transmission properties
of building materials. These terminologies are: 共1兲 water vapor transmission rate, 共2兲 water vapor perme-
ability, and 共3兲 water vapor permeance 共defined as the reciprocal of the water vapor resistance兲. As per the
standard definitions described in ASTM C 168, Standard Terminology Relating to Thermal Insulation,
these terminologies are described in the following paragraphs:

Water Vapor Transmission Rate

The water vapor transmission rate is the steady water vapor flow in unit time through unit area of a body,
normal to specific parallel surfaces, under specific conditions of temperature and humidity at each surface.

Manuscript received May 9, 2006; accepted for publication July 24, 2007; published online August 2007. Presented at ASTM
Symposium on Heat-Air-Moisture Transport: Measurements on Building Materials on 23 April 2006 in Toronto, Canada;
P. Mukhopadhyaya and M. Kumaran, Guest Editors.
1
Research Officer, Principal Research Officer, and Technical Officers, respectively, Institute for Research in Construction, NRC
Canada, Ottawa, Ontario, Canada

Copyright © 2007 by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959.

21
22 MOISTURE MEASUREMENT ON BUILDING MATERIALS

TABLE 1—Metric units and conversion factors.a,b

To Obtain 共for the

Multiply By same test condition兲
WVT
g / 共h · m2兲 1.43 grains/共h · ft2兲
grains/共h · ft2兲 0.697 g / 共h · m2兲
Permeance
g/共Pa· s · m2兲 1.75⫻ 107 1 Perm 共in.-lb兲
1 Perm 共in.-lb兲 5.72⫻ 10−8 g/共Pa· s · m2兲
Permeability
g / 共Pa· s · m兲 6.88⫻ 108 1 Perm in.
1 Perm in. 1.45⫻ 10−9 g / 共Pa· s · m兲
a
These units are used in the construction trade. Other units may be used in other standards.
b
For all calculations 1 mm of Hg= 133.3 Pa.

G 共G/t兲
Rate of Water Vapor Transmission 共WVT兲 = = 共1兲
tA A
where G=amount of water vapor flow, kg, t=time, s, G / t=slope of the straight line, g/s, and A=test area
共cup mouth area兲, m2.

Water Vapor Permeance

The water vapor permeance is the time rate of water vapor transmission through unit area of flat material
or construction induced by unit vapor pressure difference between two specific surfaces, under specified
temperature and humidity conditions.

WVT WVT
Water Vapor Permeance 共WVP兲 = = 共2兲
⌬p S共R1 − R2兲
where ⌬p=vapor pressure difference, mm Hg 共1.333⫻ 102 Pa兲, S=saturation vapor pressure at test tem-
perature, mm Hg 共1.333⫻ 102 Pa兲, R1=relative humidity at the source expressed as a fraction 共the test
chamber for desiccant method; in the dish for water method兲, and R2=relative humidity at the vapor sink
expressed as a fraction.
Water vapor resistance 共Z兲 is the reciprocal of the water vapor permeance, 共i.e., Z = 1 / WVP兲.

Water Vapor Permeability

The water vapor permeability is the time rate of water vapor transmission through unit area of flat material
of unit thickness induced by unit vapor pressure difference between two specific surfaces, under specified
temperature and humidity conditions.
It is to be mentioned here that water vapor permeability is a property of a material and water vapor
permeance is a performance evaluation indicator/property of a component and not a property of a material.
Mathematically, permeability is the arithmetic product of permeance and thickness.

Water Vapor Permeability = Water Vapor Permeance ⫻ Thickness 共3兲

The units used to express the water vapor transmission properties of building materials depend on the trade
and location. The conversion factors for commonly used units are given in Table 1.
The water vapor permeability is a function of relative humidity 共RH兲 and temperature 共T兲. The
relationship between water vapor permeability and relative humidity is very well established 关3–5兴. For
nonhygroscopic materials, water vapor permeability values show little or no change with the variation of
relative humidity. However, for hygroscopic materials, the water vapor permeability of materials increases
with the increase of relative humidity 关6,7兴.
On the other hand, temperature dependency of water vapor permeability 共WVP兲 has been of interest to
researchers for quite some time 关3,4,8–10兴. However, the extent of this influence of temperature on the
water vapor permeability is not well known for most building materials. Though in general it is found that
 MUKHOPADHYAYA ET AL. ON WATER VAPOR TRANSMISSION MEASUREMENT 23

FIG. 1—Test chamber (controlled atmosphere) and test assembly.

the temperature effect on water vapor permeability is negligible 关11兴 but there are research findings 关12兴
that indicate an increase of water vapor permeability by 3 % per °C change 共+ve兲 in temperature. Further
investigation would be required to confirm these findings.

Test Procedure
The water vapor transmission measurements are usually done under isothermal conditions. A test specimen
of known area and thickness separates two environments that differ in relative humidity 共RH兲. Then the
rate of water vapor flow across the specimen, under steady-state conditions 共with known RHs as constant
boundary conditions兲, is gravimetrically determined.
In the Desiccant or Dry Cup Method the test specimen is sealed to the open mouth of a test dish
containing a desiccant, and the assembly placed in a controlled atmosphere 共Fig. 1兲. Periodic weighings
determine the rate of water vapor movement through the specimen into the desiccant.
In the Water or Wet Cup Method, the dish contains distilled water, and the weighings determine the
rate of water vapor movement through the specimen from the water to the controlled atmosphere. The
water vapor pressure difference is nominally the same in both methods except in the variation with
extremes of humidity on opposite sides.
ASTM Standard E 96, Test Methods for Water Vapor Transmission of Materials, prescribes two
specific cases of this procedure—a dry cup 共desiccant兲 method that gives the permeance or permeability at
a mean RH of 25 % and a wet cup 共water兲 method that gives the permeance or permeability at a mean RH
of 75 %. Various technical aspects, limitations of the test method, and procedures for analyses of the test
data are available in published literatures 关6,13–15兴.

Corrections and Why?

Corrections are important because they reduce the uncertainty of the test results and generate more
realistic/accurate water vapor transmission properties that would positively influence the design process
for moisture management. It is important that all applicable corrections are made appropriately. The
procedures for making various corrections, as summarized below, are found in the literature 关14,16–18兴.

Buoyancy Correction
The duration for one set of measurements can be many days or weeks. The atmospheric pressure may
significantly change during such periods. If the test specimen is highly water vapor resistant, the changes
in mass due to water vapor transport may be overshadowed by the apparent gravimetric changes observed.
In such cases, all gravimetric data should be corrected to vacuum or any base line pressure 共e.g., pressure
at first weighing兲. The following equation can be used for buoyancy correction 关16兴
24 MOISTURE MEASUREMENT ON BUILDING MATERIALS

TABLE 2—Materials in the database.

1. Wood 共pine, cedar, 2. Stone 共lime, sand, 3. Fiber cement board 4. Wall paper
spruce, etc.兲 granite, etc.兲
5. Gypsum board 共interior 6. Polyurethane foam 7. Oriented strand board 8. Plywood
and exterior兲
9. Brick 共cementicious, 10. Cellulose fiber 11. Mortar 12.Polyisocyanurate
clay, etc.兲 insulation foam insulation
13. Glass fiber insulation 14. Vapour barrier 15. Portland cement stucco 16. Acrylic stucco
17. Wood siding 18. Vinyl siding 19. Sheathing membrane or 20. Calcium silicate
building paper insulation

m2 ␳a共␳1 − ␳2兲
=1+ 共4兲
m1 ␳1共␳2 − ␳a兲
where m1= mass recorded by balance, kg, m2= mass after buoyancy correction, kg, ␳a= density of air, kg
m−3, ␳1= density of material of balance weights, kg m−3, and ␳2= bulk density of test assembly, kg m−3.
The density of air can be calculated using the ideal gas law for the measured atmospheric pressure and
ambient temperature. The buoyancy correction is important when measured mass changes are in the range
of 0 to 100 mg 关15兴.

Corrections for Resistance Due to Still Air and Specimen Surface

In general, if the material is highly permeable, these corrections are more significant. With known thick-
ness of the still air layer in the cup, the corresponding water vapor resistance can be calculated using the
following equation for permeability 关17兴

␦a =
2.306 ⫻ 10−5 Po
RvTP
冉
T
273.15
冊 1.81
共5兲

where ␦a=permeability of still air, kg· m−1 · s−1 · Pa−1, T=temperature, K, P=ambient pressure, Pa,
Po=standard atmospheric pressure, i.e., 101325 Pa, and, Rv=ideal gas constant for water, i.e.,
461.5 J · K−1 · kg−1.
In the absence of any measured data, the surface resistances 共i.e., inside and outside surfaces of the
specimen兲 may be approximated using Lewis’ relation 关18兴. For cup methods that follow this standard, the
total surface resistance 关14兴 should be ⬇4 ⫻ 107 Pa· s · m2 · kg−1.

Edge Mask Correction

The following equation is to be used to correct the excess water vapor transmission 共WVT兲 effect due to
edge masking 关9兴:

Percent excess WVT =

400t
␲S1
loge 冉 2
1 + e−共2␲b/h兲
冊 共6兲

where h=specimen thickness, m, b=width of masked edge, m, and S1=four times the test area divided by
the perimeter, m. If the cup assembly includes any edge masking this correction shall be made.

Effects of Various Corrections

Over a period of the past ten years or more, authors have measured water vapor transmission properties of
numerous building materials commonly used in North America 关19–21兴. These materials included almost
all the contemporary building materials used for the construction of exterior building envelopes 共Table 2兲.
The minimum thickness of the material specimen was 0.141 mm and the maximum thickness was not
over 32 mm, as required by the ASTM E 96 standard, except highly porous glass fiber 共88 mm兲 and
cellulose fiber insulation 共64.5 mm兲 materials. The water vapor transmission properties were determined
according to the wet and dry cup methods at 23± 1 ° C temperature. The dry cup tests with desiccant
method test setup but with variable chamber RH levels were carried out at three nominal chamber RH
levels: 50 ±1 %, 70± 1 % and 90± 1 %, and wet cup measurements with water method test setup but with
 MUKHOPADHYAYA ET AL. ON WATER VAPOR TRANSMISSION MEASUREMENT 25

FIG. 2—Water vapor permeability: before and after corrections.

variable chamber RH were carried out at two nominal chamber RH levels: 70± 1 % and 90± 1 %. These
tests were necessary to derive the functional relationship between RH and water vapor permeability. A
detailed discussion on the principle and test data analysis techniques can be found in the relevant publi-
cation authored by Kumaran 关6兴. Results from these tests were analyzed at first with the conventional
method using Eqs. 1 to 3, and then they were corrected appropriately using the steps outlined in Eqs. 4 to
6.A sample calculation on the test data is shown in the Appendix to this paper.
It should be mentioned here that the water vapor transmission properties presented in this paper were
measured, using high precision equipments, by the trained technical staffs. But this does not mean that the
physical quantities measured are known within a few fractions of 1 %. In addition, the basic inhomoge-
neity of all building products introduces uncertainties in the derived water vapor transmission properties
that are far greater than the uncertainties in the measurements of the basic physical quantities. The
magnitude of these uncertainties depends on the building products under investigation. A rigorous labo-
ratory test data analysis indicates that water vapor transmission property for one test specimen can be
determined well within a percent 关22兴. But when all measurements on all test specimens used are com-
bined to designate the water vapor permeability or permeance of the product, the uncertainty may be as
large as 30 % 关22兴.
The general effects of corrections on all water vapor permeability and permeance values are shown in
Figs. 2 and 3. These figures clearly quantify the differences between the permeance and permeability
values after and before corrections. These corrections are always positive 共i.e., higher values after correc-
tions are applied兲 and the differences varied between 0.003 and 84 %. The upper limit of these variations
is certainly a significantly large number to adequately justify the importance of various corrections. In
order to look further into the effects of various corrections, the following paragraphs present the results
obtained for two of the most common types of building materials: wood or wood-based materials 共i.e.,
wood, plywood, oriented strand board, etc.兲, and membranes 共i.e., sheathing membranes, building papers,
vapor barrier, wall papers, etc.兲. The water vapor permeance of the wood or wood-based components
normally stays within the relative range of high to moderate and the same for the membrane is considered
to be in the lower range.

FIG. 3—Water vapor permeance: before and after corrections.

26 MOISTURE MEASUREMENT ON BUILDING MATERIALS

FIG. 4—Water vapor permeability (wood and wood-based materials): before and after corrections.

Correction Effects on Wood or Wood-based Materials

The water vapor permeability or permeance characteristic of wood or wood-based materials is a very
important parameter for the determination of its moisture management capability in a particular applica-
tion. The water vapor permeance and permeability values of various woods 共pine, cedar, spruce, etc.兲 and
wood-based building materials 共OSB, plywood, etc.兲 are shown in Figs. 4 and 5. The thickness of the
materials under consideration varies between 9 and 20 mm. These plots clearly indicate that water vapor
permeability and permeance values of the wood or wood-based materials and components vary over a
wide range. The permeance values varied between 0.08 and 38 perm 共in.-lb兲 before correction and 0.08
and 51 perm 共in.-lb兲 after correction. The maximum overall percentage change of the water vapor per-
meance values due to applied corrections is 34 %. It also appears that the effect of corrections is larger, in
terms of percentage difference, for higher water vapor permeability or permeance values.

Correction Effects on Membranes

As far as the moisture management is concerned, the water vapor permeability or permeance characteristic
of the membranes is the most important criterion that determines its suitability for a particular application.
The membranes under consideration here include paper-based building papers, polymeric sheets, vinyl
wall paper, felt paper, self adhering and torch applied membranes, etc. The thickness of the membranes
varies within the range 0.14 and 2.44 mm. The water vapor permeability values of these membranes vary
considerably 共Fig. 6兲 and it can be clearly seen in Figs. 6 and 7 that many of these membranes are highly
impermeable 共water vapor permeance less than 0.1 perm 共in.-lb兲兲. The overall water vapor permeance
values vary between 0.004 and 55 perm 共in.-lb兲 before correction, and 0.004 and 86 perm 共in.-lb兲 after
correction. The maximum effect of correction on the water vapor permeance value is found to be about
57 % and the minimum is 0.003 %. Very much like wood or wood-based materials or components, in this
case also the percentage correction values are higher for the membranes with higher water vapor perme-
ability or permeance values.

FIG. 5—Water vapor permeance (wood and wood-based materials): before and after corrections.
 MUKHOPADHYAYA ET AL. ON WATER VAPOR TRANSMISSION MEASUREMENT 27

FIG. 6—Water vapor permeability (membranes): before and after corrections.

Water Vapor Permeance and Corrections

It has been shown so far that corrections due to buoyancy, still air resistance, specimen surface, and edge
mask can significantly change the results of the water vapor transmission tests done according to the test
method prescribed in the ASTM Standard Test Methods for Water Vapor Transmission of Materials 共E
96-00e1兲. In this paper, close to 900 test points/results have been analyzed to establish this phenomenon.
In general, it has also been observed that the effect of corrections is higher for relatively more permeable
materials. Figures 8 and 9 show the relationship between the water vapor permeance and correction effect
共% difference兲 for all the 900 test points obtained from 58 building materials. These plots make it very
clear that the effects of corrections are more significant for material components that have higher water
vapor permeance values. As mentioned earlier and quite naturally the effect of corrections is always
positive 共i.e., corrected water vapor permeance values are higher than the values before correction兲. It is
also interesting to note that the effect of correction 共% difference兲 is functionally related with the measured
permeance value before correction 共Fig. 9兲. In fact, the measured data indicate that it is possible to predict
the corrected permeance values using a simple polynomial function 共Fig. 9兲 without even doing the
detailed calculation on the correction factors. However, it should be mentioned here that this is a simplistic

FIG. 7—Water vapor permeance (membranes): before and after corrections.

FIG. 8—Water vapor permeance before and after corrections, and percentage differences.
28 MOISTURE MEASUREMENT ON BUILDING MATERIALS

FIG. 9—Relationship between water vapor permeance and corrections.

attempt to estimate the correction effects and this polynomial function is valid only for the values gener-
ated at the IRC using the existing test setup. It is very likely that the polynomial shown in Fig. 9 will
depend on the laboratory and the test setup being used.
Based on the observations presented in the above paragraphs, the experts and the members of the
ASTM C16.33 E 96 task group have decided recently that it is important that all applicable corrections be
made to all measurements that result in water vapor permeance value more than 2-perm 共in.-lb兲. For
permeance values of 2-perm 共in.-lb兲 or less, the effect correction is a negligible 2 percent or lower. These
observations or decisions have been implemented in the latest revision of the ASTM Standard Test Meth-
ods for Water Vapor Transmission of Materials 共E 96/E 96M - 05兲.

Conclusions
Close to 900 water vapor transmission properties measurement test data obtained from 58 building mate-
rials are presented in this paper with the objective to demonstrate the effect of various corrections on the
measured water vapor transmission properties. The critical observations and discussion on these results
have been used as the basis for changes in the latest revision of the ASTM Standard Test Methods for
Water Vapor Transmission of Materials 共E 96/E 96M - 05兲. The observations and discussion presented in
this paper clearly demonstrate:
1. Corrections applied to the water vapor transmission properties measurement can be significant for
many building materials.
2. Resulting corrected water vapor transmission properties are higher than the same before correc-
tion.
3. The effect of corrections is relatively higher for material components that have lower resistance to
water vapor transmission.
4. Based on these observations ASTM C16.33 - E 96 task group has recommended that all applicable
corrections be made to all measurements that result in water vapor permeance value more than
2-perm 共in.-lb兲.

Appendix
In a desiccant test on a sample of medium density glass fiber insulation the following results were recorded
共see Table A1兲.

Thickness of the specimen =25.81 mm

Test area =0.01642 m2
Mass of the test specimen =20.44 g
Mass of the desiccant =554.8 g
Initial mass of the test assembly =1.257810 kg
Thickness of air layer in the cup =15 mm
 MUKHOPADHYAYA ET AL. ON WATER VAPOR TRANSMISSION MEASUREMENT 29

TABLE A1—Recorded test data.

Barometric
Mass of the Test Chamber Chamber Pressure
Elapsed Time Assembly Change in Mass Temperature RH mm Hg
共h兲 共g兲 共g兲 共°C兲 共%兲 共kPa兲
744.7
0.000 1257.810 0.000 22.83 52.60 共99.27兲
6.067 1259.469 1.659 22.84 52.6 741.11
共98.79兲
26.633 1264.609 6.799 22.78 52.2 744.41
共99.23兲
53.150 1271.062 13.252 22.82 52.1 743.21
共99.07兲
143.767 1290.773 32.963 22.74 52.2 757.69
共101.00兲
168.283 1296.389 38.579 22.78 52.1 749.81
共99.95兲
192.883 1301.953 44.143 22.78 52.1 758.44
共101.10兲

Buoyancy Correction
The buoyancy effect will be insignificant for this set of readings as recorded changes of mass are all above
100 mg. However, for example, the corrected mass of the test assembly weight 1257.810 g 共1st reading兲
can be calculated using Eq. 4.

m1⫽mass recorded by balance, kg= 1257.810⫻ 10−3 kg

P⫽barometric pressure, Pa= 99.27⫻ 103 Pa
R⫽gas constant for dry air =287.055 J / 共kg.K兲
T⫽chamber temperature =22.83+ 273.15= 295.98 K
␳a⫽density of air, kg m−3 = 1.1684 kg m−3 ␳1⫽density of material of balance weights, kg m−3 = 8000 kg m−3
h1⫽height of the test assembly, m = 44.7⫻ 10−3 m
d1⫽diameter of the test assembly, m = 168.0⫻ 10−3 m
␳2⫽bulk density of test assembly, kg m−3 = = 4 ⫻ m1 / ␲ ⫻ d21 ⫻ h1 = 1269.4 kg m−3
m2⫽mass after buoyancy correction =1258.78⫻ 10−3 kg.

A graphic analysis of the data gives the following 共Fig. A-1兲:

A linear least-squares analysis of the data gives the slope of the straight line as 0.225± 0.002 g · h−1,
with a linear regression coefficient ⬎0.998.

FIG. A-1—Graphic analysis of the test data.

30 MOISTURE MEASUREMENT ON BUILDING MATERIALS

WVT= 0.225 g.h−1 / 0.01642 m2

=19.595 grains.h−1 . ft−2 共⬇3.81⫻ 10−6 kg· m−2 · s−1兲
S = 2775.6 Pa
R1 = 0.523
R2 = 0
Permeance= 3.81⫻ 10−6 kg· m−2 · s−1 / 共2775.6 Pa⫻ 0.523兲
=2.63⫻ 10−9 kg· m−2 · s−1 · Pa−1

Corrections for Resistance due to Still Air and Specimen Surface

Permeability of still air layer 共equation 5兲 ⫽

Ⲑ
␦a = 2.306⫻ 10−5 ⫻ 101325 461.5⫻ 共22.79
Ⲑ
+ 273.15兲 ⫻ 99860 共 22.79+ 273.15 273.15 兲
1.81

−10 −1 −1 −1
=1.98062⫻ 10 kg· m · s · Pa
Permeance of 15-mm still air layer =共1.98062
⫻ 10−10兲 / 共0.015兲kg· m−2 · s−1 · Pa−1
=1.32041⫻ 10−08 kg· m−2 · s−1 · Pa−1
Hence, the 15-mm air layer offers a vapor
resistance =1 / 共1.32041⫻ 10−08兲m2 · s · Pa· kg−1
⬇ 7.6⫻ 107 m2 · s · Pa· kg−1
Surface resistances ⬇4.0⫻ 107 m2 · s · Pa· kg−1
Total corrections for resistance due to still air
and specimen surface
=共7.6⫻ 107 + 4.0⫻ 107兲m2 · s · Pa· kg−1

Edge Mask Correction

The test assembly used does not include any edge masking. However, for example, if it includes an edge
mask of width 5 mm then the following correction is to be made.

h⫽specimen thickness, m = 25.81⫻ 10−3 m

b⫽width of masked edge, m = 5 ⫻ 10−3 m
Test area= 0.01642 m2
Perimeter= 0.4541 m
S1 = four times the test area divided by the perimeter= 4 ⫻ 0.01642/ 0.4541= 0.1446 m
Ⲑ Ⲑ
Percent excess WVT= 400⫻ 25.81⫻ 10−3 ␲ ⫻ 0.1446 loge共 2 1 + e−共2␲⫻5⫻10 兲/共25.81⫻10 兲 兲 = 9.86 %
−3 −3

The applicable corrections required for the analysis of the test results in this case are due to resistance of still air
and specimen surface.
Water vapor resistance of the test specimen+ corrections= 1 / Permeance= 共1 / 2.63⫻ 10−9兲m2 · s · Pa· kg−1 = 3.80
⫻ 108m2 · s · Pa· kg−1
The watervapor resistance of the test specimen
=共3.80⫻ 108 − 共7.6⫻ 107 + 4.0⫻ 107兲兲m2 · s · Pa· kg−1 = 2.64⫻ 108m2 · s · Pa· kg−1
Permeance of the test specimen =1 / 共2.64⫻ 108 m2 · s · Pa· kg−1兲
=3.79⫻ 10−9kg· m−2 · s−1 · Pa−1
=3790 ng· m−2 · s−1 · Pa−1
Permeability= 3.79⫻ 10−9 kg· m−2 · s−1 · Pa−1 ⫻ 0.02581 m
=9.78⫻ 10−11 kg· m−1 · s−1 · Pa−1
=97.8 ng· m−1 · s−1 · Pa−1
% difference in water vapor permeance/permeability due to corrections= 共共2630− 3790兲 / 共2630兲兲 ⫻ 100= 44.1 %.
 MUKHOPADHYAYA ET AL. ON WATER VAPOR TRANSMISSION MEASUREMENT 31

References

关1兴 Mukhopadhyaya, P., Kumaran, M. K., Tariku, F., and van Reenen, D., “Final Report from Task 7 of
MEWS Project at the Institute for Research in Construction: Long-Term Performance: Predict the
Moisture Management Performance of Wall Systems as a Function of Climate, Material Properties,
etc., Through Mathematical Modeling,” Research Report, Institute for Research in Construction,
National Research Council Canada, Vol. 132, p. 384, February 1, 2003, 共IRC-RR-132兲, URL: http://
irc.nrc-cnrc.gc.ca/fulltext/rr132/.
关2兴 Mukhopadhyaya, P., Goudreau, P., Kumaran, M. K., and van Reenen, D., “Influence of Material
Properties on the Hygrothermal Response of an Ideal Stucco Wall—Results from Hygrothermal
Simulations,” 6th Nordic Building Physics Symposium, Trondheim, Norway, 2002, pp. 611–618.
关3兴 Tveit, A., “Measurements of Moisture Sorption and Moisture Permeability of Porous Materials,”
Norwegian Building Research Institute, Report 45, Oslo, 1966, p. 39.
关4兴 Chang, S. C. and Hutcheon, N. B., “Dependence of Water Vapour Permeability on Temperature and
Humidity,” American Society of Heating, Refrigerating, and Air-Conditioning Engineers (ASHRAE)
Transactions, Vol. 62, No. 1581, 1956, pp. 437–449.
关5兴 Burch, D. M., Thomas, W. C., and Fanney, A. H., “Water Vapour Permeability Measurements of
Common Building Materials,” American Society of Heating, Refrigerating, and Air-Conditioning
Engineers (ASHRAE) Transactions, Vol. 98, Part 2, 1992, pp. 486–494.
关6兴 Kumaran, M. K., “Alternative Procedure for the Analysis of Data from the Cup Method Measure-
ments for Determination of Water Vapor Transmission Properties,” J. Test. Eval., Vol. 26, No. 共6兲, pp.
575–581.
关7兴 Kumaran, M. K., “Hygrothermal Properties of Building Materials,” ASTM Manual on Moisture in
Buildings, September 1, 2001, pp. 29–65.
关8兴 Barrer, R. M., Diffusion In and Through Solids, Cambridge University Press, London, UK, 1951.
关9兴 Joy, F. A. and Wilson, H. G., “Standardization of the Dish Method for Measuring Water Vapor
Transmissions,” National Research Council of Canada, Research Paper 279, January 1966, p. 263.
关10兴 Galbraith, G. H., Guo, J. S., and McLean, R. C., “The Effect of Temperature on the Moisture
Permeability of Building Materials,” Build. Res. Inf., Vol. 28, No. 4, 2000, pp. 245–259.
关11兴 Mukhopadhyaya, P., Kumaran, M. K., and Lackey, J., “Use of the ‘Modified Cup Method’ to Deter-
mine Temperature Dependency of Water Vapor Transmission Properties of Building Materials,” J.
Test. Eval., September 2005.
关12兴 Hedenblad, G., “Materialdata för fukttransportberäkningar,” T19:1996. ISBN 91-540-5766-3. Bygg-
forskningsrådet, Stockholm, Sweden.
关13兴 Hedenblad, G., “Moisture Permeability of Some Porous Building Materials,” Proceedings of the 4th
Symposium, Building Physics in the Nordic Countries, Espoo, Vol. 2, 1996, pp. 747–754.
关14兴 Hansen, K. K. and Lund, H. B., “Cup Method for Determination of Water Vapor Transmission
Properties of Building Materials. Sources of Uncertainty in the Methods,” Proceedings of the 2nd
Symposium, Building Physics in the Nordic Countries, Trondheim, 1990, pp. 291–298.
关15兴 Lackey, J. C., Marchand, R. G., and Kumaran, M. K., “A Logical Extension of the ASTM Standard
E 96 to Determine the Dependence of Water Vapor Transmission on Relative Humidity,” Insulation
Materials: Testing and Applications; 3rd Volume, ASTM STP 1320, R. S. Graves and R. R. Zarr, Eds.,
ASTM International, West Conshohocken, PA, 1997, pp. 456–470.
关16兴 McGlashan, M. L., “Physico-Chemical Quantities and Units,” Royal Institute of Chemistry Mono-
graphs for Teachers, No. 15, 1971, p. 8.
关17兴 Schirmer, R., ZVDI, Beiheft Verfahrenstechnik, Nr. 6, S170, 1938.
关18兴 Pedersen, C. R., Ph.D. thesis, Thermal Insulation Laboratory, The Technical University of Denmark,
1990, p. 10.
关19兴 Kumaran, K., Lackey, J., Normandin, N., van Reenen, D., and Tariku, F., “Summary Report from
Task 3 of MEWS Project,” Institute for Research in Construction, National Research Council, Ot-
tawa, Canada, 共NRCC-45369兲, 2002, pp. 1–68.
关20兴 Kumaran, K., Lackey, J., Normandin, N., Tariku, F., and van Reenen, D., “A Thermal and Moisture
Transport Property Database for Common Building and Insulating Materials,” Final Report from
ASHRAE Research Project 1018-RP, 2004, pp. 1–229.
32 MOISTURE MEASUREMENT ON BUILDING MATERIALS

关21兴 Mukhopadhyaya, P., Lackey, J., Normandin, N., Tariku, F., and van Reenen, D., “Hygrothermal
Performance of Building Envelope Retrofit Options: Task 1—A Thermal and Moisture Transport
Property Database,” IRC/NRC, National Research Council Canada, Ottawa, Client Final Report,
2004, pp. 1–37, 共B-1137.5兲.
关22兴 Kumaran, M. K., “A Thermal and Moisture Property Database for Common Building and Insulation
Materials,” American Society of Heating, Refrigerating, and Air-Conditioning Engineers (ASHRAE)
Transactions, Vol. 112, Part 2, 2006, pp. 485–497.
 Journal of ASTM International, Vol. 4, No. 5
Paper ID JAI100369
Available online at www.astm.org

Carsten Rode,1 Ruut Peuhkuri,2 Berit Time,3 Kaisa Svennberg,4 and Tuomo Ojanen2

Moisture Buffer Value of Building Materials

ABSTRACT: When building materials are in contact with indoor air they have some capacity to moderate
the variations of indoor humidity in occupied buildings. But, so far, there has been a lack of a standardized
quantity to characterize the moisture buffering capability of materials. It has been the objective of a recent
Nordic project to define such a quantity, and to declare it in the form of a NORDTEST method. The Moisture
Buffer Value is the figure that has been developed in the project as a way to appraise the moisture buffer
effect of materials, and the value is described in the paper. Also explained is a test protocol which ex-
presses how materials should be tested for determination of their Moisture Buffer Value. Finally, the paper
presents some of the results of a round robin test on various typical building materials that has been carried
out in the project.
KEYWORDS: building materials, moisture transport, humidity, buffer effect, indoor climate, material
properties

Nomenclature
Symbol ⫽ Explanation Unit
A ⫽ Area m2
bm ⫽ Moisture effusivity kg/ 共m2 · Pa· s1/2兲
Bim ⫽ Moisture Biot number -
dp ⫽ Penetration depth m
g ⫽ Moisture flux kg/ 共m2 · s兲
G ⫽ Moisture uptake kg/ m2
Ġ ⫽ Moisture release rate g/h
m ⫽ Mass kg
MBV ⫽ Moisture Buffer Value kg/ 共m2 · % RH兲
n ⫽ Air change rate h−1
p ⫽ Water vapor pressure Pa
ps ⫽ Saturation vapor pressure Pa
RH ⫽ Relative humidity %
t ⫽ Time s
tp ⫽ Period s
u ⫽ Moisture content kg/kg
V ⫽ Volume m3
Zp ⫽ Vapor diffusion resistance Pa· m2 · s / kg
␤p ⫽ Moisture transfer coefficient kg/ 共Pa· m2 · s兲
␦p ⫽ Water vapor permeability kg/ 共m · s · Pa兲
␸ ⫽ Relative humidity -
␯ ⫽ Water vapor concentration kg/ m3
␳ ⫽ Density kg/ m3

Manuscript received February 5, 2006; accepted for publication May 2, 2007; published online May 2007. Presented at ASTM
Symposium on Heat-Air-Moisture Transport: Measurements on Building Materials on 23 April 2006 in Toronto, Canada;
P. Mukhopadhyaya and M. Kumaran, Guest Editors.
1
Assoc. Prof., Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark.
2
Research Scientist and Senior Research Scientist, respectively, VTT Technical Research Centre of Finland, FIN-02044VTT,
Finland.
3
Research Director, SINTEF Building and Infrastructure, N-7491Trondheim, Norway.
4
Researcher, Lund Institute of Technology, SE-221 00Lund, Sweden.

Copyright © 2007 by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959.

33
34 MOISTURE MEASUREMENT ON BUILDING MATERIALS

Introduction
Indoor humidity is an important parameter to determine the occupants’ perception of indoor air quality 关1兴,
and is also an important parameter as a cause of processes which are harmful to the health of occupants
关2兴. Thus, it is known that humidity has an impact on both the working efficiency and health of occupants.
But due to the varying loads, the indoor humidity exhibits significant daily or seasonal variation. Materials
that absorb and release moisture can be used positively to reduce the extreme values of humidity levels in
indoor climates. Consequently, there is an interest when making performance analysis of buildings to
consider the moisture buffering properties of absorbent, porous building materials.
The interest in moisture buffering is not new. Experimental results from climatic chambers were
published along with some analytical considerations by Künzel in 1960 关3兴. Other climatic chamber tests
and field studies were carried out in Sweden in the beginning of the 1980s 关4兴. These and other early
investigations are presented in a recent overview paper 关5兴.
A method to characterize the moisture buffer effect has been proposed already in Ref 关6兴. An example
is given in Ref 关7兴 of the way in which moisture transfer between wood-based structures and indoor air
significantly reduces the peak indoor humidity 共by as much as 35 % RH兲 and increases the minimum
indoor humidity 共up to 15 % RH兲. A method to calculate the moisture penetration depth has been described
in Ref 关8兴.
In Ref 关9兴 a range of different porous building materials was investigated to find their moisture buffer
capacity on the indoor air humidity. A specially constructed climate chamber 共volume about 0.5 m3兲 was
used to measure the moisture buffering when there was a periodically varying vapor flux. End grain wood
panels showed the best buffering capacity due to the rapid diffusion and the great moisture capacity of
wood. On the other hand, cellular concrete covered by a thin gypsum plaster turned out to be the best
buffering commercial construction.
In Ref 关10兴 a room size test cell was used to investigate the moisture buffer capacity of plasterboard,
chipboard, cellular concrete, plywood, wood panels, and painted plasterboard. The relative humidity of the
test cell was a result of given rates of moisture release and removal to/from the cell air and the moisture
exchanged with the materials. The highest buffer capacity was measured for wood panels and cellular
concrete. The same setup was used in investigations where the layers of a lightweight interior wall were
added successively and walls of aerated cellular concrete were tested with and without surface treatment
关11兴. The results indicated a significant reduction of the moisture buffer effect even when rather permeable
surface treatments and layers were applied.
An approach to develop a test method and device to test the effective moisture capacity of structures
or material layers is described in Ref 关12兴. The results showed that pine with moisture transport along the
grain has the highest moisture buffer capacity while painted gypsum board has the lowest. These small-
scale tests were supplemented by a full-scale experiment together with numerical investigations. The inner
surfaces of two identical test rooms were covered by either unpainted or painted gypsum boards, unpainted
wood, or aluminum foil. Also, these results pointed out the significance of surface treatments on the
moisture buffering.
In Ref 关13兴 the hygroscopic inertia of some covering materials was studied numerically and experi-
mentally which proposed the use of inertia classes for characterization of materials. The tests were per-
formed with 12/ 12 h cyclic steps between 65 and 85 % RH.
The role of interior furnishings on the total moisture buffer capacity of a room was studied experi-
mentally and numerically in Ref 关14兴. Pieces of furniture were added step-by-step in the test room with a
given moisture production/removal rate. The results showed that it is necessary to include the moisture
buffering effect of furniture when studying whole buildings.
An attempt to derive the moisture buffer capacity of various insulation materials by using different
dynamic experiments was done in Ref 关15兴. Both isothermal and nonisothermal tests were used. The
results indicated a high sensibility of the resulting moisture buffer capacity on whether one is using
parameters from steady-state measurements or one of several types of dynamic measurements.
More results were presented and discussed during a NORDTEST Seminar on Moisture Buffer Capac-
ity in August 2003 关16兴. The workshop concluded that there was a need for a definition of the term, which
should be technically appropriate, yet comprehensible and indisputable for the industry and users that will
apply it. The workshop and the information about recent or ongoing activities on the subject was the
 RODE ET AL. ON MOISTURE BUFFER VALUE 35

background for starting this NORDTEST project, which is comprised of the following activities:
• To establish a robust definition of the moisture buffer ability of materials and material systems used
in the indoor environment.
• To define and present a test method to measure the moisture buffer ability according to the defini-
tion.
• To carry out a round robin test in order to ensure that testing laboratories are able to handle the test
methods, and to establish the first reference measurements on a limited number of representative
materials and material systems and to obtain an indication of the expected accuracy level in such
measurements.
The primary objective of the project has been the development of a NORDTEST method with a test
protocol on how materials and systems should be tested for their moisture buffer ability. It is proposed to
call this desired property the Moisture Buffer Value.
The project has now been completed and reported in Ref 关17兴. This paper will report on some of the
main results from the project.
Partners in the NORDTEST project have been the Technical University of Denmark 共DTU兲 共as project
leader兲, VTT, Finland, Byggforsk, Norway 共NBI兲, and Lund University, Sweden 共LTH兲. In addition, the
project was followed by an international reference group with participants from six other research insti-
tutions.

Definition of Moisture Buffer Value

Different Levels of Moisture Buffering

The moisture buffer performance of a room is the ability of the materials within the room to moderate
variations in the relative humidity. These variations can be seasonal or diurnal. Most attention is paid to the
moisture buffering of diurnal variations. The moisture buffer phenomena of a room can be described with
the scheme of Fig. 1.
The moisture buffer performance depends on the moisture buffer capacities of each material combi-
nation and furniture in the room together with the moisture production and air change rate and the ratio
between the material surface area and the air volume. The simplest material combination is a homogenous
material with a thin layer of stagnant air. The thickness of the boundary air layer will be determined by the
air velocity in the room.
On the next level, the system level, the moisture buffer value, which is the subject of this test proposal,
depends on the air velocity, area, and thickness of the sample.
In order to have both a theoretical, physics-based definition, and a simple and easily understandable
method for categorization, a two-fold description of the moisture buffer capacity has been chosen: A
theoretical and a practical one.

Moisture Effusivity: bm
The theoretical description of moisture buffer capacity on the material level is based on the heat-mass
transfer analogy. Well known from heat transport theory is the thermal effusivity which expresses the rate
of heat transfer over the surface of a material when the surface temperature changes. The thermal effusivity
is defined as the square root of the product of material density, specific heat capacity, and thermal con-
ductivity. By introducing the moisture effusivity, bm 关kg/ 共m2 Pa· s1/2兲兴, see Eq 1, in a similar way to the
definition of thermal effusivity, we can describe the ability of a material to absorb or release moisture.

bm =
冑 ␦ p · ␳0 ·

ps
⳵u
⳵␸
共1兲

where ␦ p 关kg/ 共m · s · Pa兲兴 is the water vapor permeability, ␳0 关kg/ m3兴 dry density of the material, u 关kg/kg兴
moisture content, ␸ 关−兴 relative humidity, and ps 关Pa兴 saturation vapor pressure. Apart from ps, which is
given by the test conditions, the other parameters in the definition of bm are all standard material proper-
ties.
36 MOISTURE MEASUREMENT ON BUILDING MATERIALS

FIG. 1—Definition scheme for the moisture buffer phenomena in the indoor environment divided into three
descriptive levels—material level, system level, and room level. On the material level the quantities are
determined with negligible influence from the surrounding climate, e.g., boundary air layers. The system
level includes material combinations where the simplest form of a material combination is a homogeneous
material with the convective boundary air layer normally present in indoor environments. Systems may
also comprise composite products. The room level includes the building and furnishing materials exposed
to the indoor air as well as moisture loads, ventilation rate, indoor climate, and other factors influencing
the moisture buffering in the room.

The moisture effusivity constitutes one theoretical possibility to express the rate of moisture absorbed
by a particular material 共i.e., as a material property兲 when it is subjected to a sudden increase in surface
humidity. The use of this quantity could be extended to cases where the humidity increase takes place in
the surrounding air, if the convective mass transfer coefficient tends to infinity.

Ideal Moisture Buffer Value

From the heat transport theory it is known how to handle surface transport phenomena in cases not only
when the surface temperature suddenly jumps from one level to another, but also when it varies according
to a sine function 共harmonic variations兲. Furthermore, by using Fourier analysis, these results can be
transformed into cases with other functional forms of the time variation of the surface conditions.
One such functional form is the signal function, which is a steadily repeating function with a high level
of the surface condition for one duration of time, and a low surface condition for another period of time.
Of interest for this project is a signal function of humidity where the high humidity is maintained for eight
hours, and the low humidity lasts for sixteen hours. Fourier analysis makes it possible to predict the
surface moisture flux versus time, g共t兲, for such an exposure. The accumulated moisture uptake G共t兲
关kg/ m2兴, respectively, moisture release, that both happen within the time period t p are found by integrating
the moisture flux over the surface g共t兲 as in Eq 2:

G共t兲 = 冕 g共t兲dt = b · ⌬p · h共␣兲冑 ␲t

t

0
m
p
共2兲

where

⬁
2 sin2共n␲␣兲
h共␣兲 = 兺
␲ n=1 n3/2
⬇ 2.252关␣共1 − ␣兲兴0.535 共3兲
 RODE ET AL. ON MOISTURE BUFFER VALUE 37

␣ 关−兴 is the fraction of the time period where the humidity level is high. For the 8 / 16 h scheme, ␣
= 1 / 3, which makes h共␣兲 = 1.007 and the accumulated moisture uptake can be expressed in a simpler form:

G共t兲 ⬇ 0.568 · bm · ⌬p冑t p 共4兲

The moisture uptake within eight hours corresponds to the moisture release during sixteen hours.
We define a Moisture Buffer Value as a characteristic of the material based on this moisture uptake/
release. The Moisture Buffer Value is expressed based on the moisture exchange from Eq 4 normalized
with the change in surface relative humidity, ⌬RH, and we abbreviate it as MBV. The MBV is proportional
p 关s 兴. The thus defined theoret-
to the moisture effusivity bm times the square root of the time period, t1/2 1/2

ical, or ideal, value of MBV is given by Eq 5, which is derived from Eq 4.

= 0.00568 · ps · bm · 冑t p
G共t兲
MBVideal ⬇ 共5兲
⌬RH
The moisture effusivity is theoretically based on material properties which are usually determined
under steady-state and equilibrium conditions. However, the buffer property represents a dynamic charac-
teristic. There may be some discrepancy between the basic material properties depending on whether they
have been determined under steady-state or dynamic conditions, and this may have some influence on the
determination of the moisture buffer value.
Since the ideal experimental conditions rarely exist, and therefore Eq 5 is only an approximation, a
calculation tool has been developed that simulates numerically the same type of experiment as used for
determination of Moisture Buffer Value using the same stationary material properties as for the calculation
of moisture effusivity. Further information on the background and theoretical approach to the calculation
tool is given in Ref 关17兴.
The ideal moisture buffer capacity is based on the assumption that the materials studied have a
thickness that exceeds the penetration depth of that material or a material combination.

Practical Moisture Buffer Value

For practical categorization of materials the Moisture Buffer Value based on an experimental method can
be used. The determination of a practical Moisture Buffer Value should involve a run time which corre-
sponds to that of typical exposure in practice—typically a daily variation. Likewise, it should be useful for
the material thickness, surface mass transfer conditions, and surface coatings normally used in practice.
The practical Moisture Buffer Value 共MBVpractical兲 indicates the amount of water that is transported in
or out of a material per open surface area, during a certain period of time, when it is subjected to variations
in relative humidity of the surrounding air. When the moisture exchange during the period is reported per
open surface area and per % RH variation, the result is the MBVpractical. The unit for MBVpractical is
kg/ 共m2 · % RH兲.
The concept of the practical Moisture Buffer Value can easily be appreciated and understood from an
experimental point of view, and likewise, it is relatively straightforward to measure.
The practical Moisture Buffer Value is determined in an experimental setup where the sample is
exposed to cyclic step changes in RH between high and low values for eight and sixteen hours, respec-
tively. For example the normal case will be a periodic exposure of 8 h at 75 % RH and 16 h at 33 %, the
size of the RH interval will then be 42 % RH. This kind of measurement sequence is shown later in Fig.3.
The value is a direct measure of the amount of moisture transported to and from a material when the
exposure is given. The value should ideally be a property of the material, but it may turn out that it is
somewhat influenced by the experimental and geometrical conditions. For instance, the practical and the
ideal moisture buffer value will only be similar if the material is homogenous and its thickness exceeds the
depth to which the periodic signal from the surface will penetrate.

Sample Thickness
The penetration depth, d p, at which the excitation of the moisture variation is only 1 % of the amplitude
at the surface, is:
38 MOISTURE MEASUREMENT ON BUILDING MATERIALS

d p,1 % = 4.61
冑 ␦ p p st p

␳0 ␲
⳵u
⳵␸
共6兲

Samples should be thicker than d p,1 %.

Surface Coefficient
In the determination of penetration depth, the periodic changes are supposed to take place on the material
surface. But in reality, the variations occur in the ambient air, and a surface resistance to moisture transfer
exists which slows down the moisture exchange. The correlation between the theoretical moisture effu-
sivity and MBVpractical can be found only in the limit when the convective mass transfer coefficient tends
to infinity—the ideal situation.
When water vapor is transmitted from within a material to the ambient air, the primary resistance to
this flow is quite often within the material itself, Z p,m. However, there is also a relatively small resistance,
Z p,s, related to the convective moisture flow over the surfaces of the material. The convective surface
coefficient for moisture transfer can be expected to be around ␤ p = 2.0· 10−8 kg· m2 · s / Pa. This corresponds
to a convective surface resistance for moisture transfer of Zs,p = 5.0· 107 Pa/ 共kg· m2 · s兲. These are normal
values for environments with an ambient air velocity around 0.1 m / s.
A Biot-number could be defined to express the ratio between the resistance to moisture flow internally
in the material and the resistance to surface moisture transfer. If the moisture penetration depth from Eq 6
is used to give a measure of the internal surface resistance, the moisture Biot-number, Bim, could be
written as:

d p,1 %
Bim = 共7兲
␦ pZ p,s
It may be suggested that Bim should have a value of at least 50 to ensure an acceptably low influence
of the surface moisture resistance.

Duration of Cycle
The MBV has its validity only for a certain declared period length, which for the NORDTEST method is
chosen to be 24 h 共8 + 16 h兲.
In Ref 关18兴 some further analysis of the Moisture Buffer Value’s sensitivity to material thickness is
given, but also to the influence of the surface mass transfer coefficient, and duration of the periodic
exposure.
Finally, it is worth noting that determination of the Moisture Buffer Value involves processes which
are nonlinear, since both vapor permeability and sorption depend on the moisture content; it involves
sorption processes which exhibit some hysteretic effects; and the processes are also somewhat temperature
dependent.

Test Protocol
The NORDTEST project defines a test protocol for experimental determination of the moisture buffer
value. The principle is based on climatic chamber tests, where a specimen is subjected to environmental
changes that come as a square wave in diurnal cycles.
The test protocol proposes to use climatic exposures which vary in 8 h + 16 h cycles: 8 h of high
humidity followed intermittently by 16 h of low humidity. The reason for the asymmetry in this time
scheme is two-fold: 共1兲 it replicates the daily cycle seen in many rooms, e.g., offices or bedrooms, where
the load comes in approximately 8 h, and 共2兲 for practical reasons during testing if the climatic chamber
conditions are changed manually, it is a scheme which is easier to keep than a 12 h + 12 h shift.
The low humidity is proposed to be 33 % RH, while the high should be 75 % RH. During the tests, it
is important that the equipment is able to make rapid RH changes. Testing should always be carried out at
23°C as a reference.
 RODE ET AL. ON MOISTURE BUFFER VALUE 39

TABLE 1—Materials tested in the round robin test, and institutions performing the tests.

Material/Product DTU VTT LTH NBI

Spruce Plywood 共pretest兲 x XX x x
Spruce boards x XX x
Precast concrete XX x x
Drywall 共paper-faced gypsum兲 x XX x
Laminated wood XX x x
Lightweight aggregate concrete XX x x
Cellular concrete XX x x
Brick XX x x
Birch wood panels x x XX
XX⫽Country responsible for supplying the material.

Specimens will normally be sealed on all but one or two surfaces so the minimum exposed surface
area should be 0.01 m2. The thickness of the specimen should be at least the moisture penetration depth for
daily humidity variations. At least three specimens should be used.
Using an accurate scale, the specimens should be weighed continuously or intermittently during the
test. At least five weight measurements should be carried out during the 8 h high humidity part of the last
cycle. A minimum of three cycles have to be carried out, and the weight amplitude must not vary by more
than 5 % from day to day. This is defined as the three stable cycles. The stable cycles are also character-
ized by the fact that the daily amounts of moisture uptake and release are approaching each other as shown
later in Fig. 4. The mass change should be plotted and normalized as mass change 共m8 hours − m0兲 per m2
and per ⌬RH to give the MBV.

Round Robin Test and Results

A round robin test has been carried out within the NORDTEST project to try the testing paradigm and to
obtain some initial results for typical building materials. In addition, and to guide the formulation of the
test protocol, a preliminary test was carried out on some spruce plywood boards that were distributed to all
project partners. The materials tested and the institutions doing the tests are listed in Table 1. Each material
is tested by three partners.
The institutions did not have quite the same experimental equipment available, and some of the
operational routines were also dissimilar, although in accordance with the common test protocol. For
example, some institutes did manual weighing of the specimen, while for others it took place by automated
logging of a scale. Thus, it has been part of the round robin test to see if it was possible to obtain similar
and agreeable results by all institutions.
Figure 2 shows a drawing and photograph of one of the climatic chambers used at DTU. The humidity
control of the chamber works by supplying it with either humid or dry air in an intermittent mode, such
that the desired humidity in the chamber is achieved.
Figure 3 shows the measured weight change response of one of the specimens when it was subjected
to cycles that varied the ambient humidity between 33 and 75 % RH for 16 and 8 h, respectively. The
choice of stable cycles and the moisture uptake is marked in the figure.
Figure 4 shows for three different specimens of the same type of material how the moisture uptake and
moisture release varied from cycle to cycle until the three stable cycles were attained. Thus the results of
measurements of one type of material give the following background for statistical analysis: three speci-
mens with results from three cycles of both moisture uptake and release—altogether 3 ⫻ 3 ⫻ 2 = 18 bids for
the MBV result. The results should be represented at least with their mean value and standard deviation
together with information about the number of tested specimens and stable cycles 共if different from 3
⫻ 3兲.
Figure 5 shows the MBV for the materials tested by the participating institutions. The bar diagrams
indicate for each material and testing laboratory the average of the MBV value and its standard deviation.
However, it seems that the order of magnitude for the tested materials is around 1 g / 共m2 · % RH兲, and
there may be approximately a factor three of a difference between the materials with the highest and
lowest MBV among those tested.
The overall conclusion on the results of the round robin was encouraging: the determination of the
40 MOISTURE MEASUREMENT ON BUILDING MATERIALS

FIG. 2—Drawing and picture of one of the climatic chambers used at DTU.

practical Moisture Buffer Value with four very different laboratory equipments gave results that are
comparable with each other. Also, consistent differences were found between the ability of different
materials to buffer variations in RH of the ambient air: materials like untreated spruce and birch boards
and cellular concrete performed as best buffers while materials like brick and concrete were able to buffer
under the half of the best buffers 共Fig. 5兲. Further discussion of the results can be found in Ref 关17兴.

FIG. 3—Determination of MBVpractical from dynamic measurements. Definition of quasisteady-state (the

three cycles inside the ellipse) and the moisture uptake and release. Also RH for the cyclic steps is given.
The rising tendency of the mass is due to the initial conditions that have been lower than the average of
the quasisteady-state conditions that will arise after some cycles.
 RODE ET AL. ON MOISTURE BUFFER VALUE 41

FIG. 4—MBV of spruce boards as a function of cycle number and uptake versus release. The last three
cycles are the stable cycles (⫽quasisteady).

Practical Application of the Moisture Buffer Value

The Moisture Buffer Value is primarily meant as a number that can be used to appraise a material’s ability
to absorb and release moisture from an adjacent space. For practical application it can also be useful as a
number for estimation of the moisture balance of rooms, as indicated by the following example.

Example
A room has dimensions 4 by 5 by 2.5 m, and thus a volume of V = 50 m3. The occupancy and activity in
the room releases Ġ = 100 g of moisture per hour. The room is clad with A = 45 m2 wall panels of spruce
board with MBV= 1.2 g / 共m2 · ⌬RH兲. Initially the room is assumed unventilated, and the storage capacity
of the room air is neglected. By how much will the indoor humidity increase during a working day 共eight
hours兲?
All the released moisture is absorbed by the spruce board, and thus, the increase in indoor relative
humidity can be calculated from the amount of absorbed moisture, and the moisture buffer value of the
wood:

FIG. 5—Moisture Buffer Values found by different institutions for the different materials. Each bar indi-
cates the average of three specimens over three stable cycles. The thin vertical line bars indicate standard
deviations.
Another random document with
no related content on Scribd:
The Project Gutenberg eBook of Astronomy in
a nutshell
 This ebook is for the use of anyone anywhere in the United
States and most other parts of the world at no cost and with
almost no restrictions whatsoever. You may copy it, give it away
or re-use it under the terms of the Project Gutenberg License
included with this ebook or online at www.gutenberg.org. If you
are not located in the United States, you will have to check the
laws of the country where you are located before using this
eBook.

Title: Astronomy in a nutshell

The chief facts and principles explained in popular
language for the general reader and for schools

Author: Garrett Putman Serviss

Release date: December 27, 2023 [eBook #72521]

Language: English

Original publication: New York: G. P. Putnam's Sons, 1912

Credits: Aaron Adrignola, Tim Lindell and the Online Distributed

Proofreading Team at https://www.pgdp.net (This file
was produced from images generously made available
by The Internet Archive/American Libraries.)

*** START OF THE PROJECT GUTENBERG EBOOK ASTRONOMY IN

A NUTSHELL ***
The Lunar “Crater” Copernicus
Photographed from Nasmyth and Carpenter's plaster-of-paris model of the moon.
In this model the topography of the moon is faithfully represented as seen with
powerful telescopes.
 Astronomy in a
Nutshell

The Chief Facts and Principles

Explained in Popular Language
for the General Reader
and for Schools

By

Garrett P. Serviss

With 47 Illustrations
G. P. Putnam's Sons
New York and London
The Knickerbocker Press
1912
 Copyright, 1912

BY

GARRETT P. SERVISS

The Knickerbocker Press, New York

 PREFACE

How many thousands of educated people, trained in the best

schools, or even graduates of the great universities, have made the
confession: “I never got a grip on astronomy in my student days.
They didn't make it either plain or interesting to me; and now I am
sorry for it.”
The purpose of the writer of this book is to supply the need of
such persons, either in school, or at home, after school-days are
ended. He does not address himself to special students of the
subject—although they, too, may find the book useful at the
beginning—but to that vast, intelligent public for whom astronomy
is, more or less, a “mystical midland,” from which, occasionally,
fascinating news comes to their ears. The ordinary text-book is too
overladen with technical details, and too summary in its treatment of
the general subject, to catch and hold the attention of those who
have no special preliminary interest in astronomy. The aim here is to
tell all that really needs to be told, and no more, and to put it as
perspicuously, compactly, and interestingly, as possible. For that
reason the book is called a “nutshell.”

The author has been sparing in the use of diagrams, because he

believes that, in many cases, they have been over-pressed. There is
a tendency to try to represent everything to the eye. This is well to a
certain extent, but there is danger that by pursuing this method too
far the power of mental comprehension will be weakened. After all,
it is only by an intelligent use of the imagination that progress can
be made in such a science as astronomy. The reader is urged to
make a serious effort to understand what is said in the text, and to
picture it in his mind's eye, before referring to the diagrams. After he
has thus presented the subject to his imagination, he may refer to
the illustrations, and correct with their aid any misapprehension. For
this reason the cuts, with their descriptions, have been made
independent of the regular text, although they are placed in their
proper connections throughout the book.

G. P. S.
April, 1912.
CONTENTS
 Part I—The Celestial Sphere.

PAGES

Definition of Astronomy—Fundamental Law of the 3-64

Stars—Relations of the Earth to the Universe—
Ordinary Appearance of the Sky—The Horizon, the
Zenith, and the Meridian—Locating the Stars—
Altitude and Azimuth—Circular or Angular Measure
—Altitude Circles and Vertical Circles—The Apparent
Motion of the Heavens—The North Star and
Phenomena Connected with it—Revolution of the
Stars round the Pole—Locating the Stars on the
Celestial Sphere—Astronomical Equivalents of
Latitude and Longitude—Parallels of Declination and
Hour Circles—“The Greenwich of the Sky”—Effects
Produced by Changing the Observer's Place on the
Earth—The Parallel Sphere, the Right Sphere, and
the Oblique Sphere—The Astronomical Clock—The
Ecliptic—Apparent Annual Revolution of the Sun
round the Earth—Inclination of the Ecliptic to the
Equator: its Cause and its Effects—The Equinoxes—
Importance of the Vernal Equinox—The Equinoctial
Colure—The Solstices—Poles of the Ecliptic—
Celestial Latitude and Longitude—The Zodiac—The
Precession of the Equinoxes: its Cause and Effects—
Revolution of the Celestial Poles—Past and Future
Pole Stars
 Part II—The Earth.
PAGES

Nature, Shape, and Size of the Earth—The Polar 67-123

Compression and Equatorial Protuberance, and their
Cause—The Attraction of Gravitation—The Mass of
the Earth: how Found—How the Earth Holds the
Moon, and the Sun the Planets—The Tides—How
the Moon and Sun Produce Tides—Spring Tides and
Neap Tides—The Atmosphere—The Law and Effects
of Refraction—Dip of the Horizon—The Aberration of
Light—Time: how Measured—Sidereal, Apparent
Solar, and Mean Solar Time—The Clock and the Sun
—Day and Night—Where the Days Begin—The
Seasons—Effects of the Varying Declination of the
Sun—Polar and Equatorial Day and Night—The
Tropics and the Polar Circles—Inequality of Length
of the Seasons—When the Seasons in the Two
Hemispheres will be Reversed—The Calendar, the
Year, and the Month—Reformations of the Calendar
—Different Measures of the Month
 Part III—The Solar System.

PAGES
The Sun—Distance, Size, and Condition of the Sun— 127-215
Temperature of the Sun—Solar Heat on the Earth,
and its Mechanical Equivalent—Peculiar Rotation of
the Sun—Sun-spots, their Appearance and Probable
Cause—Faculæ—The Photosphere—Solar
Prominences—Explosive Prominences—The Solar
Corona—Parallax, and the Measure of Distances—
Spectroscopic Analysis—How the Elements in the
Sun Reveal their Presence—List of the Principal
Solar Elements—The Moon—Origin of the Moon—
Appearance of its Surface—Gravity on the Moon—
The Phases of the Moon—Causes of the Absence of
a Lunar Atmosphere—Eclipses—How the Moon
Causes Eclipses of the Sun—The Laws Governing
Eclipses—The Shadow during a Solar Eclipse—
Eclipses of the Moon—Number of Eclipses in a Year
—The Saros—The Planets—Kepler's Laws of
Planetary Motion—Mercury—Venus—Mars, and its
So-called Canals—Theories about Mars—Jupiter, its
Belts and its Satellites—The "Great Red Spot”—
Saturn, its Rings and its Satellites—Composition of
the Rings—Uranus and Neptune—Comets, and the
Laws of their Motion—Composition of Comets—The
Pressure of Light and its Connection with Comets'
Tails—Breaking up of Comets—Meteors and their
Relations to Comets—The November Meteors and
Other Celebrated Showers—Meteorites or Bolides
which Fall upon the Earth—The Question of their
Origin
 Part IV—The Fixed Stars.

PAGES

Division of the Stars into Magnitudes—Division of the 219-257

Stars according to their Spectra—Stars Larger and
Smaller than the Sun—The Distances of the Stars—
Variable Stars—Double and Binary Stars—
Spectroscopic Binaries and how they are Discovered
—Proper Motions of the Stars—Number of the Stars
—New, or Temporary Stars—The Milky Way—The
Nebulæ—The Two Kinds of Nebulæ—Spiral Nebulæ
—The Nebular Hypothesis—Applications of
Photography to Stars and Nebulæ—The
Constellations—How to Learn the Constellations—
Their Antiquity—Description of the Principal
Constellations Visible from the Northern Hemisphere
at Various Times of the Year

Index 259
 ILLUSTRATIONS

PAGE
The Lunar “Crater” Copernicus Frontispiece

Photograph of South Polar 8

Region of the Moon

The Moon near the “Crater” 20

Tycho

Drawing of Jupiter 28

Drawing of Jupiter 28

Jupiter 38

Saturn 46

Saturn 46

The Milky Way about Chi 58

Cygni
The Great Southern Star- 64
Cluster in Centauri

Photograph of a Group of 76
Sun-Spots

Polar Streamers of the Sun, 89

Eclipse of 1889

Solar Corona at the Eclipse of 89

1871

Morehouse's Comet, October 96

15, 1908

Morehouse's Comet, November 96

15, 1908

Head of the Great Comet of 105

1861

Halley's Comet, May 5, 1910 105

The Six-Tailed Comet of 1744 112

Spiral Nebula in Ursa Major 124

(M 101)

The Whirlpool Nebula in Canes 124

Venatici
Tress Nebula (N. G. C. 6992) 132
in Cygnus

The Great Andromeda Nebula 140

Spiral Nebula in Cepheus (H. 154

IV. 76)

Nebulous Groundwork in 154

Taurus

Nebula in Sagittarius (M. 8) 162

The Great Nebula in Orion 180

Photographs of Mars 200

Schiaparrelli’s Chart of 220

Martian “Canals”

ILLUSTRATIONS IN THE TEXT

The Rational and the Sensible 12

Horizon

Altitude and Azimuth 14

Right Ascension and Declination 35

The Ecliptic and Celestial 51
Latitude and Longitude

How the Earth Controls the 75

Moon

The Tidal Force of the Moon 79

Refraction 85

Dip of the Horizon 87

Sidereal and Solar Time 93

The Change of Day 101

The Seasons 107

Parallax of the Moon 139

Parallax of the Sun from Transit 141

of Venus

Spectrum Analysis 147

The Phases of the Moon 160

Orbits of Mars and the Earth 183

Ellipse, Parabola, and Hyperbola 203

The North Circumpolar Stars 244

Key to North Circumpolar Stars 245

 PART I.
THE CELESTIAL SPHERE.
 PART I.

THE CELESTIAL SPHERE.

1. Definition of Astronomy. Astronomy has to do with the

earth, sun, moon, planets, comets, meteors, stars, and nebulæ; in
other words, with the universe, or “the aggregate of existing things.”
It is the most ancient of all sciences. The derivation of the name
from two Greek words, aster, “star,” and nomos, “law,” indicates its
nature. It deals with the law of the stars—the word “star” being
understood, in its widest signification, as including every heavenly
body of whatever kind. The earth itself is such a body. Since we
happen to live on the earth, it becomes our standpoint in space,
from which we look out at the others. But, if we lived on some other
planet, we would see the earth as a distant body in the sky, just as
we now see Jupiter or Mars.
Astronomy teaches us that everything in the universe, from the
sun and the moon to the most remote star or the most extraordinary
nebula, is related to the earth. All are made of similar elementary
substances and all obey similar physical laws. The same substance
which is a solid upon the earth may be a gas or a vapour in the sun,
but that does not alter its essential nature. Iron appears in the sun
in the form of a hot vapour, but fundamentally it is the same
substance which exists on the earth as a hard, tough, and heavy
metal. Its different states depend upon the temperature to which it
is subjected. The earth is a cool body, while the sun is an intensely
hot one; consequently iron is solid on the earth and vaporous in the
sun, just as in winter water is solid ice on the surface of a pond and
steamy vapour over the boiler in the kitchen. Even on the earth we
can make iron liquid in a blast furnace, and with the still greater
temperatures obtainable in a laboratory we can turn it into vapour,
thus reducing it to something like the state in which it regularly
exists in the sun.
This fact, that the entire universe is made up of similar
substances, differing only in state according to the local
circumstances affecting them, is the greatest thing that astronomy
has to tell us. It may be regarded as the fundamental law of the
stars.
2. The Situation of the Earth in the Heavens. One of the
greatest triumphs of human intelligence is the discovery of the real
place which the earth occupies in the universe. This discovery has
been made in spite of the most deceptive appearances. If we
accepted the sole evidence of our eyes, as men once did, we could
only conclude that the earth was the centre of the universe. In the
daytime we see the sun apparently moving through the sky from
east to west, as if it were travelling in a circle round the earth,
overhead by day and underfoot at night. In the night-time, we see
the stars apparently travelling round the earth in the same way as
the sun. The fact is, that all of them are virtually motionless with
regard to the earth, and their apparent movements through the sky
are produced by the earth's rotation on its axis. The earth turns
round on itself once every twenty-four hours, like a spinning ball.
Imagine a fly on a rotating school globe; the whole room would
appear to the fly to be revolving round it as the heavens appear to
revolve round the earth. It would have to be a very intelligent insect
to correct the deceptive evidence of its eyes.
The actual facts, revealed by many centuries of observation and
reasoning, are that the earth is a rotating globe, turning once on its
axis every twenty-four hours and revolving once round the sun every