7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties http://slidepdf.com/reader/full/forced-drying-of-hemp-lime-and-the-effects-on-its-hygrothermal-properties 1/34 Forced Drying of Hemp Lime and the Effects on its Hygrothermal PropertiesStudent: Victor Delegrego Supervisor: Mike Lawrence Department of Architecture and Civil Engineering The University of Bath 2016 AR30315 – BEng dissertation
34
Embed
Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
Hemp lime is a biocomposite made from lime and hemp (Cannabis sativa) shives. The material is used
as a non-structural shuttering for its interesting thermal and acoustic insulating properties, its air tightness
and its ‘breathability’. A common problem experienced is that the hemp shives, that is the woody core of thehemp plant, absorbs very large amounts of water during the mixing process. The release of this water is going
not be immediate, such that its thermal conductivity will rise and its breathability properties will be impaired,
while already installed in the building. The drying process can take months or even years to cease.
This research tested a new fabrication processes for hemp lime and the effects it had on the
hygrothermal properties of the material. The method consists of forced drying hemp lime as soon as it is cast
into the molds. After this process, the moisture content inside the material should be low enough to allow it
to work under project specification from day one.
In order to test these hygrothermal characteristics two different hemp lime formulations were
developed. One using high surface area lime (HSA lime) and calcium sulfoaluminates (CSA), as a setting
additive. The other using the commercial Tradical® PF70 binder. The two formulations went through bothforced (rapid) and normal (slow) drying methods.
The hygrothermal parameters tested were thermal conductivity, moisture buffer value and vapour
permeability. These results showed some sensible differences between the two drying methods. However,
it was clear that those differences were not enough to hinder the use of the forced drying method and that
this method can be considered a promissory solution for the drying problem. From both formulations, the
HAS lime with CSA was the one which seemed to perform better.
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
Abstract ................................................................................................................................................. 1 Table of contents ................................................................................................................................... 2
Table of figures ...................................................................................................................................... 3
List of tables .......................................................................................................................................... 4
Figure 1 – ‘Definition of quasi-steady state (the 3 cycles inside the ellipse) and the moisture uptake and
release’ (Rode et al. 2006) ............................................................................................................................... 12
Figure 2 – Mold with a perforated timber bottom.......................................................................................... 14Figure 3 – Shives and binders inside pan mixer .............................................................................................. 15
Figure 4 – Demolding of forced dried samples ............................................................................................... 16
Figure 5 – Very fragile sample with detached layer on top ............................................................................ 16
Figure 6 – Samples being tested with the hot wire machine .......................................................................... 17
Figure 7 – Crumbled PA sample....................................................................................................................... 20
Table I – Hemp lime formulations ................................................................................................................... 15
Table II – Results from the firs testing day, with TC in W/(m*K) .................................................................... 18
Table III - results from the second testing day, with TC in W/(m*K)............................................................... 18Table IV – results from the third testing day, with TC in W/(m*K) ................................................................. 19
Table V – MBV of samples and the moisture uptake variation of the three last cycles.................................. 21
Table VI – MBV of each type of hemp lime and statistical dispersion of samples’ MBVs ............................... 22
Table VII – Moisture buffer value (MBV) class ranges..................................................................................... 22
Table VIII – Variation from G of the last three moisture uptake measurements ............................................ 25
Table IX - Vapour permeability (δ) of each sample ......................................................................................... 26
Table X – Vapour permeability (δ), vapour diffusion resistance factor (μ) and the differences between
This dissertation aimed to investigate a forced drying process for manufacturing hemp lime and the
effect it has on the material’s hygrothermal properties. The prospects of using this method as a solution for
hemp lime’s drying problem were also discussed.
To investigate these effects, two different hemp lime formulations were prepared and subjected to
both forced and normal drying processes. Their hygrothermal characteristics were then assessed by a
number of tests and the results were compared between themselves and the ones found in the literature.
1.1 - Hemp lime
The growing concern with the depletion of natural resources and release of CO2 by the modern
construction industry, responsible for one third of the world’s energy consumption (Rahim 2015), is one of
the main motivations behind a return to more sustainable and natural building practises. Walking together
with this growing interest, research and development on natural building materials and components is
growing, with the intention of finding new, environmental friendly solutions for engineering (Benfratello et
al. 2013). It is in within this context that hemp lime, a relatively new natural building material with interesting
properties, is taking its share in the market (Shea et al. 2012) (Rahim et al. 2015).
Hemp lime, also known as hempcrete, lime hemp, hemp concrete and green concrete (Bruijn,
Johansson 2014), is a form of biocomposite concrete. Its basic formulation consists of a mixture of lime and
hemp shiv, the first being responsible for providing the mineral matrix that binds the aggregates, while the
second for providing natural fibre aggregates to the mixture (Benfratello et al. 2013). Shiv is the name given
to the woody core of the hemp plant. More information about hemp shives is described later in the text.
Lime is the most basic and standard choice of binder for hemp lime, but it can also be combined with
other materials. Because of that, some commercial brands offer binders designed specifically for hemp lime.
Those are usually a combination of lime with other components such as cement and pozzolans, which serve
to increase the hydraulic characteristics of the material. This hydraulicity is responsible for providing some
initial setting strength to the mixture, due to the quick hydration forming components, as opposed to the
following slow strength increase associated to carbonation (Sutton et al. 2011), (Hirst et al. 2010), (Shea et
al. 2012).
The use of hemp lime developed in France by the end of the 1980’s and beginning of the 1990’s for
conservation purposes. In the French region of Champagne there was a large number of historical timber
framed buildings infilled originally with a wattle and daub system, being at that time in need of repair. The
poor choice of materials in earlier restorations had only further damaged the buildings. Hemp lime was thenbrought as a possible solution and it was tested there for its first time (Hirst et al. 2010).
Hemp lime buildings are nowadays present in large numbers in France and the UK, the former having
more than 200 building by 2007 (Hirst et al. 2010). Research on the material is also carried out by other
countries, located both in Europe and North America (Colinart et al. 2012).
1.1.1 - Modern uses
Hemp lime is used as a non-structural insulation material for roofs, wall and floors (Latif et al. 2015).It possesses excellent hygric capacities, good thermal properties, good airtightness and causes a low impact
on the environment (Bruijn, Johansson 2013). It is advertised as having the ability to regulate internal air
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
humidity, or as it is technically named, for having an excellent moisture buffer capacity. This property can
even be seen as the main advantage of hemp lime over other insulation materials (Bart et al. 2014).
There are currently three major techniques for building with hemp lime:
Casting: hemp lime is cast between a temporary shuttering, which can be removed from 12 to 24
hours later. The wall is usually 300 mm thick and can be plastered after the shuttering is removed.This procedure is usually done for smaller projects, where a less expensive labour force can be used
( Ahlberg et al. 2014) (Hemcrete nd).
Spraying: hemp lime is sprayed over a temporary or permanent single sided shuttering. After
spraying, the surface can be flattened and the shuttering, if temporary, can be removed 24 hours
later. The spraying option is more suited for larger projects (Hemcrete nd).
Pre-fabrication: the production of blocks or panels containing hemp lime in an industrial context
(Shea et al. 2012). An important problem in the logistics of pre-fabricating hemp lime is the drying
process, discussed later on this text.
1.1.2 – Hemp plant
The hemp plant is an industrial variety of Cannabis sativa. It belongs to the same species of
marijuana, with both plants being visually very similar. The main difference between those plants is the
content of tetrahydrocannabinol (THC). This component is responsible for marijuana’s psychoactive effects
and it is found in high concentrations inside that plant. Hemp on the other hand, is a crop designed for
commercial purposes and contains by law a very low THC concentration, bellow 0.2%. Under this limit, it is
impossible for the plant to give the ‘high’ effect, associated with marijuana. The cultivation of hemp is
therefore approved by the EU ( Ahlberg et al. 2014).
Hemp crops are suited to various climates, with the potential of being cultivated both in Southern
and Northern Europe. It reaches 3 to 4 metres height within 4 months, about the time at which it is harvested(Shea et al. 2012). It does not require pesticides (Sutton et al. 2011) and due to its rapid growth, its leaves
block the sun light from reaching the ground, acting as a natural weed suppressor ( Ahlberg et al. 2014).
To be commercially utilized, hemp goes under a mechanical separation process. The ‘plants are cut
and dried in the sun for two weeks; then they are swingled for separating the bast (that is fibres located in
the outer stalk) from shiv, that is the wooden inner part’ (Benfratello et al. 2013). The fibres are the most
expensive part of the plant and have multiple uses, such as paper manufacture and biocomposites production
for the car industry. (Bruijn, Johansson 2014). For hemp lime, only the shiv is interesting. Shiv composes 40
to 60% of the hemp stalk and is usually marketed as animal bedding (Bruijn, Johansson 2014).
The high porosity of the shiv is of major importance to the hygrothermal behaviour of hemp lime. Some
of the shiv’s important characteristics are:
Very high water absorption, up to 406% its own weigh after 48 hours under submersion (Nguyen et
al. 2010);
A low bulk density of about 103 kg/m3 (Nguyen et al. 2010);
Low thermal conductivity (Benfratello et al. 2013);
Good acoustic insulation properties (Benfratello et al. 2013);
1.1.3 - The drying problem
As explained before, the hemp shives absorb multiple times its own weight in water while
submerged. Research conducted by Nozahic found a weight gain of around 300% in only 5 minutes. Walker
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
obtained a lower but also very high value of 225% (Walker, Pavía 2012). A significant problem is the
occurrence of this process during the hemp lime mixing. As the shives fill themselves with the mixing water,
the binder requires more to be added in order to have enough water to react. After the end of this mixing
process, hemp lime carries a considerable amount of unreacted water inside the shives, which will be slowly
released to the external environment during the following months or years (Colinart et al. 2012).
This water absorption is the root cause of a problem experienced in hemp lime construction: the
newly constructed buildings do not achieve project specifications in terms of moisture buffering and thermal
insulation. As research has shown, a rise in hemp lime’s moisture content results in a significant,
approximately linear increase in thermal conductivity of this material (Bruijn, Johansson 2013) (Korjenic et
al. 2011). At the same time, the humidity regulation properties attributed to hemp lime will not exist for
many months after construction, because the saturated shives are only able to release moisture to the
environment. Although this is a significant problem for the hemp lime industry, the current building practice
is to just let the walls dry naturally.
1.1.4 - Proposed solution
In a context of production of insulating panels containing hemp lime, a possible method was
proposed in order to deal with the drying problem. This method consists of submitting hemp lime to a forced
drying process after its casting. The idea is that after being forced dried, the hemp shives will not hold excess
water anymore and the material would be ready for use, fulfilling design specifications from day one.
Although simple, this method cannot be applied without paying attention to the materials used. That
is because the strength gain of pure hydraulic lime comes from the product of the carbonation reaction. In
this reaction the Ca(OH)2 molecules of lime react with the CO2 found in air, producing CaCO3. The problem is
that this reaction is slow. With the few hours it takes to force dry hemp lime, the reaction does not have
sufficient time to produce any sensible increase in strength, thus making hemp lime unable to support itselfstructurally. To counteract this problem and provide the initial mechanical strength needed, a proposed
solution was the addition of small amounts of calcium sulfoaluminates (CSA) to the mixture.
CSA is a low cost and eco-friendly binder with an early mechanical strength onset. Its properties come
mainly from ettringite (C6AS3H32), produced from the hydration of C4A3S. It has been used for decades as a
setting accelerator additive for lime and other binders (Telesca et al. 2014).
1.2 - Aims and objectives
Research conducted in the University of Bath showed that by adding CSA to the mixture, the resulting
hemp lime product coming from the forced dried process was acceptable from a self-supporting structural
perspective. However, as hemp lime is an insulation material, it is still necessary to know if this process has
any influences on the hygrothermal behaviour of the material and if yes, what those are. It was with the
intention of discovering these influences that the present research was conducted.
To cover the hygrothermal properties of the material, the parameters investigated and the related
methods used were respectively: thermal conductivity with the hot wire method; moisture buffer value with
the Nordtest protocol; vapour permeability with the dry cup method.
It was decided to use two binder formulations to have a better understanding on the effects of forced
drying. The first formulation used a high surface area (HSA) lime with CSA addition as binders. The secondused formulation used the commercial PF70 Tradical® binder. The hemp shives used were the same for all
cases. Both forced and normal dried samples of these compositions were tested.
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
This section describes concepts and study findings that are important to the understanding of the
experiments and results contained in this text.
2.1 - Forced drying of hemp lime
There is very little research on forced dried hemp lime and none could be found that investigated its
hygrothermal properties. Thus, the papers used as basis for this research dealt only with normally dried hemp
lime.
2.2 - Hygrothermal properties
‘Insulation is a seemingly simple product but its actual performance within a building and within the
environment is far from straightforward’ (Tucker et al. 2014). To model the performance of a building it is
then necessary to have some specific experimental parameters, relating to the hygrothermal characteristic
of the material. Some of those parameters were investigated and used as means of comparison between the
different drying processes and formulations examined. The explanation and the meaning of each of those
experiments along with their importance to the research is explained in this section.
2.2.1 - Thermal conductivity
Thermal conductivity, represented by the letters k or λ, is a number that indicates the effectiveness
of a material to conduct heat ( Ahlberg et al. 2014). In technical terms, it is the ‘rate of steady-state heat flow
(W) through a unit area of 1 m thick homogeneous material in a direction perpendicular to isothermal planes,
driven by unit (1 K) temperature difference across the material sample ’ (Shea et al 2013). This heat flow
always occurs in the direction of higher to lower temperatures, because ‘energy is transferred when
neighbouring molecules collide and higher temperature equates to a higher molecular energy, or more
molecular movement’ ( Ahlberg et al. 2014). The lower the thermal conductivity of the material, the better
are its insulation properties and thus, more desirable it is ( Ahlberg et al. 2014).
From thermal conductivity it is possible to obtain the U-value, another important parameter for thebuilding sector. The U-value can be defined as the ‘reciprocal of the sum of all thermal resistances of the
layers of the building element including resistance due to a film of air at the inner and outer surfaces’ (Shea
et al 2013), with thermal resistance being the thickness of the material divided by its thermal conductivity.
The U-value serves as a comparison parameter for building elements and takes away the need to consider
the individual properties of each material ( Ahlberg et al. 2014). It is also used as the ‘basis of many regulatory
frameworks aimed at conserving the use of fuel and power in buildings and are a fundamental part of EU
members’ methodologies for demonstrating compliance with the Energy Performance of Buildings Directive’
(Shea et al 2013).
The standard values of thermal conductivity of hemp lime are observed over a wide range. Literature
shows that these values can stay within 0.04 and 0.19, but are usually found between 0.07 and 0.09 W/mK
(Benfratello et al. 2013) (Sutton et al. 2011). The main reason for this variation is the density of the material.
By increasing the hemp to lime ratio there is a weight reduction (due to the lower density of hemp) and a
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
following decrease of the material’s thermal conductivity, due to the large superiority of hemp as an
insulating material, in an almost linear tendency (Walker, Pavía 2014) (Benfratello et al. 2013). Conversely,
when the binder to hemp ratio is increased there is a rise in thermal conductivity (Walker, Pavía 2014). In the
UK hemp lime compositions with a density lower than 300 kg/m3 have been favoured as a way of reducing
wall thickness while complying with the country’s building regulations (Barclay et al. 2014). Other parameter
which research shows to influence thermal conductivity to a lesser extent is the type of binder, with thosehaving higher hydraulicity tending to present lower thermal conductivities (Walker, Pavía 2014).
By its thermal conductivity alone, hemp lime can be considered a good material for insulation
purposes. However, recent research has shown that this parameter does not tell everything about hemp
lime’s insulation properties. This over simplification is unrepresentative of hemp lime and underestimates
the advantages of its use (Kinnane et al. 2015) (Shea et al. 2012).
Hemp lime is a material with high thermal inertia, a characteristic connected to both thermal
conductivity and volumetric heat capacity (the capacity of a material’s given volume to store heat under a
defined temperature variation). A high thermal inertia means that the modelling of a material’s thermal
properties in a steady state condition is going to give inaccurate results and that the dynamic effects are most
important to be analysed (Kinnane et al. 2015). It is also advantageous to asses a material’s insulatingproperties in a more similar situation to the external environment of a building, with its constantly varying
conditions, rarely achieving a prolonged steady state behaviour (Kinnane et al. 2015).
Research on the dynamic properties of hemp lime are being carried out by different researchers and
different parameters such as the Q24h and ts-s were found to better describe the behaviour of the material
(Shea et al. 2012).
With the aforementioned advantages of testing for the dynamic thermal properties, it may seem
reasonable to adopt them while researching hemp lime. However, most available researches on this topic
are quite recent and more would be necessary in order to stablish reference values. Thus, most studies,
including the present one, still test for the thermal conductivity value (or U-value), as those parameters are
very good for comparison purposes and are widely used by the construction industry (Walker, Pavía 2014).
Thermal conductivity can be measured by different methods and a variety of those were used in
hemp lime research. It is important to take it in consideration, as it was shown that the method employed
may have a sensible effect on the results, especially when dealing with an anisotropic, non-homogenous
material such as hemp lime (Latif et al. 2011).
For this research it was decided to use the transient hot-wire method for determining thermal
conductivity. This method consists of embedding a linear heat source, the hot wire, inside the material to be
analysed. By knowing the heat transmitted by the wire and the temperature change in a time interval, it is
possible to obtain the thermal conductivity value. ‘The mathematical model of hot wire method is based on
the assumption that hot wire is a continuous line source and by providing constant heating power through
thermal impulses it generates cylindrical coaxial isotherms in an infinite homogenous medium with initialequilibrium condition’ (Latif et al. 2011).
The hot wire method works well for materials with low thermal conductivity and is able to provide
quick and accurate results for small samples (when compared to other methods). However, a unidirectional
analysis may not be satisfactory for hemp lime (Latif et al. 2011). When hemp lime is being cast and
compacted into its shuttering or molds, it is possible for the shives to get organized in a preferential way,
perpendicular to the direction of compaction. This in turn, generates a tendency for a stratification of the
material, with intercalated layers of hemp shives and binder. As the binder matrix is the most conductive
component of the material, the measured value of thermal conductivity will tend to be lower along the
compaction direction when compared to the one perpendicular to it (Nguyen et al. 2010). It is thus necessary
to measure thermal conductivity along both directions and consider them separately.
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
As discussed before, the thermal insulating properties of hemp lime, along with the acoustic
insulation and air tightness it provides are considered good in comparison to other, more conventional
insulating materials. However, what makes it stand out is its interaction with indoor air humidity (Bart et al.
2014).
Due to hemp’s pore structure, hemp lime walls have the ability to regulate humidity in the air. When
indoor air has a decrease in its relative humidity, a vapour pressure difference between the material and the
air is generated. Then, the moisture trapped inside the hemp shives moves through hemp lime and is released
to the air. Conversely, when humidity increases in the air, the vapour pressure difference makes the hemp
fibres absorb and retain this extra moisture. Both desorption and adsorption phenomena have the effect of
damping the air’s relative humidity variation. This humidity regulation is called ‘moisture buffer effect’ and
is seen in all vapour permeable materials containing natural fibres, but it is significantly accentuated in hemp
lime (May 2005).
Due to the aforementioned effect, hemp lime is frequently categorised as a ‘breathable’ material.Behind this term are two different properties. The first one is named vapour permeability and refers to the
ability of the material to let moisture in form of water vapour pass through. The second one, called
hygroscopicity, is the total amount of this transported water that the material’s pore structure is able to store
(May 2005).
The moisture buffer effect of hemp lime is a non-negligible parameter in the hygrothermal modelling
of a building. Consequently, ignoring this hygric behaviour might result in ‘an incorrect prediction of direct
and indirect energy demands for heating/cooling because of latent heat effects, comfort condition
modifications, and heat transport parameter dependence on moisture content’ (Dubois et al. 2014).
2.2.2.1 - Moisture buffer capacity
In order to compare the effectiveness of materials to buffer variations in air humidity, the parameter
used is the Moisture Buffer Value (MBV), with higher values representing better buffering capacity and being
thus more desirable (Dubois et al. 2014).
Many methods for obtaining the MBV are used. ‘Among those, Nordtest method is the pioneering
method and is mostly used in the European context’ (Latif et al. 2015). It was therefore the one chosen for
this research. The Nordtest protocol defines three values related to moisture buffer capacity. Those are
described below, as seen in (Latif et al. 2015):
- Moisture effusivity (bm):
It is the measurement of the ability of the material to exchange moisture with its surroundings when
the surface of the material is exposed to sudden change in humidity. The equation for moisture effusivity is:
= ∗∗ (1)
Where:
- bm = moisture effusivity [kg/(m2.Pa.s1/2)];
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
Ideal Moisture Buffer Value is the theoretical determination of moisture buffer value (MBV) based
on its moisture effusivity, time period of moisture uptake and saturation vapour pressure. Ideal Moisture
Buffer Value expresses the upper limit of the moisture buffer capacity. The equation for Ideal Moisture Buffer
Value is:
≈
∆ = 0.00568 ∗ ∗ ∗ √ (2)
Where: G(t) = accumulated moisture uptake (kg/m2) and the corresponding moisture release during a time
period tp (s). The ideal moisture buffer value is measured in [g/(m2.% RH)].
- Practical moisture buffer value (MBV practical):
Practical moisture buffer value, MBV practical, is defined as the amount of moisture content that
passes through the unit open surface of the material when the material is exposed to variation in relative
humidity of the surrounding air. MBV can be expressed as:
Equation (3):
= ∆∗(ℎℎ−) (3)
The practical moisture buffer value is, as suggested by its name, obtained by practical experiments
in a laboratory. The ideal and the practical values are only comparable when the material tested is
homogeneous (Rode et al. 2006). Because of that, only the practical value is of use for hemp lime.
To test the samples according to Nordtest specifications, the samples are placed in climate chamberwith only one vapour open face. The samples are then subjected to continuous high and low humidity cycles
until the moisture uptake after each moisture absorption cycle stabilises in a quasi-steady state. Another
characteristic of the stabilization process is that the moisture uptake and the moisture release of each cycle
have similar values (Rode et al. 2006). Figure 1, obtained from Rode et al. (2006) shows a moisture
uptake/release graph depending on the RH tested, which reached the 3 stable cycles, necessary to
characterise a quasi-steady state according to the Nordtest protocol.
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
the material, as opposed to water vapour permeance, which us a property of the specimen. It is the quantity
of water vapour transmitted per unit of time through a unit area of the product per unit of vapour pressure
difference between its faces for a unit thickness.
- Water vapour diffusion resistance factor (μ):
Quotient of the water vapour permeability of air and the water vapour permeability of the material
or the homogeneous product concerned; it indicates the relative magnitude of the water vapour diffusion
resistance of the product and that of an equally thick layer of stationary air at the same temperature. In
calculating it both the vapour permeability of the test specimens and the air temperature in which those
were tested are considered. It is then possible to obtain a value that is comparable to others obtained in
different testing conditions.
- Water vapour diffusion equivalent air layer thickness (sd):
Thickness of a motionless air layer which has the same water vapour resistance as the test specimen
with the thickness d .
In this research the ‘dry cup’ method was adopted. It was used by various other authors researching
hemp lime (Latif et al. 2015)(Collet et al. 2013)(Mazhoud et al. 2016) and provides thus a good way of
comparing results. The method consists of creating a vapour pressure difference between two faces of the
samples while isolating the others. One face is maintained with 50% relative humidity while the other with
zero relative humidity, by means of a desiccant material.
Hemp lime is usually considered a material with very good vapour permeability properties when
compared to other more traditional building materials (Walker, Pavía 2014) (Bruijn, Johansson 2014). ‘Thecommon industry figure of water vapour diffusion resistance factor (μ) of lime –hemp concrete is 4.85 ± 0.24’
(Walker, Pavía 2014). Nonetheless, this value is significantly influenced by the binder aggregate ratio and the
hydraulicity of the binder, both parameters raise the resistance factor when increased (Walker, Pavía 2014).
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
The samples were fabricated by using rectangular impermeable wooden molds, being divided in two
categories according to its sizes. The first category corresponded to twelve samples measuring 10x10x10 cm,
used for the moisture buffer and vapour permeability tests. The second category was that of the samples
which measured 15x15x15 cm, being sixteen of these produced and used to assess thermal conductivity. For
the samples that passed through the forced drying process, the only difference in the molds was at their
base, made with perforated timber as shows Figure 2.
Figure 2 – Mold with a perforated timber bottom
The preparing of samples happened in two phases. The first phase dealt only with the ones subjected
to the normal (slow) drying process, while the second one only with the forced (rapid) dried samples. The
reason for this division was a logistic one: the time and equipment available demanded all the
experimentation to happen simultaneously. As it was also necessary to give time for the normal dried
samples to set for an adequate time, it was decided to offset the making of the forced dried ones such that
all samples could be tested together.
The mixing process followed was the same for both phases. The hemp shives, binders and water hadtheir weight individually measured with a scale. The shives and binders were put into a large electrical pan
mixer as shown in Figure 3. This machine was then turned on and the water was gradually added. The process
continued until the distribution was homogeneous. Right after the mixing was over, the material was filled
into molds. The molding was done by gently dispersing layers of hemp lime into the molds, taking care to
leave no voids or over compress the material.
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
It is noticeable that the TC values consistently dropped within the elapsed time from test day 1 to 3,in figures from 13 to 21%. Also, there is a remarkable proximity between these presented values and the
ones obtained a week earlier in the second testing day. It then becomes clear by the information in Table IV
that, when left in the same controlled environment for a sufficient period of time, the differences on TC
between the rapid and forced dried samples greatly diminish. However, it is not possible to assure from these
tests that this conditions would be maintained, as it would be necessary to test the samples for longer periods
of time.
3.3 - Moisture buffer testing
3.3.1 – Test procedure
For the moisture buffer experiments the smaller samples, measuring 10x10x10 cm, were used. First
of all, the samples were prepared for the test. The top part of the samples (according to their position in the
molds) were left uncovered, while all the other faces were tightly sealed with metal tape. No part of the top
face was covered with tape, therefore its dimensions were still the same as the sample’s. The intention of
this procedure was to restrict the moisture exchange between sample and exterior environment to the top
face, with its known defined area.
Figure 7 shows a sample that was lost for crumbling during the sealing process. This was due to thevery low mechanical strength of the forced dried P formulation. In this way, only 11 samples could be tested,
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
being those three HN, three HA, three PN and two PA. After the sealing was done, the samples were labelled
and put into a climate chamber as shown in Figure 8.
Figure 7 – Crumbled PA sample
The Nordtest procedure was followed as described by Rode (Rode, Grau 2008). The climate chamber
machine has the purpose of creating a controlled environment where the air temperature and relative
humidity can be manipulated. As described in the ‘2.2.2.1 - Moisture buffer capacity’ section, the samples
were submitted to periodic moisture sorption and desorption cycles. The sorption cycles lasted for 8 hours,going from 9:00 AM to 5:00 PM, and had the climate chamber set at 23 °C and 33% relative humidity (RH).
The desorption cycles went on for 16 hours and had the climate chamber working at 23 °C and 75% RH. ‘The
reason for the asymmetry in this time scheme is twofold: (1) It replicates the daily cycle seen in many rooms,
e.g. offices or bedrooms, where the load comes in approximately 8 hours, and (2) 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’ (Rode et al. 2006). After the end of each cycle the samples were briefly moved from the
chamber to have their weights measured by an accurate scale, with precision to 1/100 of a gram.
Figure 8 – Samples placed inside the climate chamber
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
The experiment can end when the weight measurements indicate that the difference of weight
amplitude between the 3 last 24 hour cycles is less than 5%. When this happens, it is said that the samples
achieved a quasi-steady state of moisture exchange. Another result of this state is that the moisture uptake
and moisture release weight have their values closer than in previous cycles. The experiment went on for 15
days, but none of the samples was able to achieve a steady state condition and due to time constraints the
experiment had to be stopped. The moisture uptake variation between the last three cycles is shown for
each of the samples in Table V, with the number after the adopted nomenclature indicating the individual
sample to which it refers.
An explanation for this instability cannot be given with certainty, however it may be related to the
fact that the samples were not always placed in the same position, after they were taken away from the
chamber to have their weights measured. It is possible that there were temperature and humidity variations
inside the chamber, such that an inconsistent displacement of samples would cause them to not reach quasi-
steady state conditions. Only the last two 24 hour cycles were the exception to this procedure, as care was
taken to place the samples in their previous positions.
Table V – MBV of samples and the moisture uptake variation of the three last cycles
Samples Cycle 1-2 Cycle 2-3 Cycle 1-3
MBV
[g/(m2.%RH)]
HA1 13% 5% 8% 1.81
HA2 10% 3% 8% 1.81
HA3 17% 23% 36% 5.06
HN1 6% 1% 7% 2.32
HN2 13% 8% 6% 2.15
HN3 8% 7% 2% 2.44
PA1 18% 6% 25% 1.27
PA2 43% 9% 49% 1.95
PN1 4% 13% 10% 2.64
PN2 13% 5% 19% 1.83
PN3 8% 1% 9% 2.13
Even though the test results did not comply with the Nordtest requirements of a quasi-steady state,
a tendency for stabilization can be seen in Table V, slightly more accentuated in the ‘Cycle 2-3’ column of the
table. According to the protocol, the moisture buffer value of the material should be calculated using theaverage of moisture uptake from the last 3 cycles. However, due to the greater tendency of stability seen in
the last two cycles, it was decided to use only those to compose the moisture buffer values.
3.3.2 - Moisture buffer value
In hands of the average moisture uptake of the last two cycles, it was possible to obtain the moisture
buffer values for each of the samples, shown in the last column of Table V. Those values were obtained by
dividing the average moisture uptake by the area of the samples (0.01 m 2) and by the difference between
the RH of the absorption and desorption cycles (42) (Rode et al. 2006). The final MBV is shown below in TableVI.
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
Table VI – MBV of each type of hemp lime and statistical dispersion of samples’ MBVs
MBV [g/(m2.%RH)] SD CV
HA 2.89 1.88 0.70
HN 2.31 0.14 0.06
PA 1.61 0.48 0.26
PN 2.20 0.41 0.18
Before analysing the moisture buffer value results, it is necessary to check the dispersion of the
experimental data. The HA samples were the ones which presented the highest dispersion, as seen by their
coefficient of variation (the ratio between the average of the samples and their SD) presented in Table VI,
which is close to one and can thus be considered a high value. The PA and PN samples presented intermediate
variations, while the HN had a very low dispersion. Clearly, it is necessary to consider the dispersion seenwithin a set of individual samples’ values, before analysing the moisture buffer value obtained from them at
face value.
On the MBV itself, the results present values that are normal but slightly inferior to what is expected
from hemp lime (Latif et al. 2015) (Rahim 2016). By using the characterization method outlined by Rode
(2006) and reproduced in Table VII, it was possible to define the PA type as good, while the HA, HN and PN
types as excellent. It is important to mention that the air velocity has an influence on the moisture buffer
value of any material, with greater velocities leading to a better MBV (May 2005). As the speed inside the
chamber was measured with an anemometer and the result turned out as zero, it is possible to assume that,
if the anemometer’s result was correct, the moisture buffer values would all be higher if tested in different
conditions. Thus, the values obtained can be compared within themselves, but they may possibly not be
representative in order to be comparable with values from the literature.
Table VII – Moisture buffer value (MBV) class ranges
MBV
class
lower
value
upper
value
Negligible 0.0 0.2
Limited 0.2 0.5
Moderate 0.5 1.0
Good 1.0 2.0Excellent 2.0 -
The HA moisture buffer value was greater than the HN one, while the PA value was inferior to the PN
one. This observation, along with the great dispersion of experimental values shown in Table VI, evidence
that the influence of the drying method on the moisture buffer capacity of hemp lime is not strong enough
for a trend to be stablished. However, the results are consistent in showing that the H formulation presented
better results than the P formulation, independent of the drying method it went through.
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
After these values were obtained, they were averaged to give the final δ value for each hemp lime
type. The vapour diffusion resistance factor (μ), which is the ratio between the air and the material’s vapour
permeability, was also calculated. The value for air’s vapour permeability was taken from BS EN 12086-2013.
It is an unitless factor, with lower values representing greater permeability, which is suitable to be compared
to test results obtained with different air conditions. The δ, the μ and the percentage difference between
same formulations under different drying methods are described in Table X.
Table X – Vapour permeability ( δ ), vapour diffusion resistance factor ( μ ) and the differences between drying
methods
δ [kg/(m.s.Pa)] μ Δ drying
HA 1.20E-10 1.74 3%
HN 1.25E-10 1.65
PA 6.99E-11 2.80 22%
PN 8.93E-11 2.38
The vapour permeability values in Table X show consistently that the H formulation has a significantlyhigher δ than the P formulation, independent of the drying method employed.
Analysing the drying methods alone within each hemp lime formulation, both the HA and PA
averages of δ are lower than their normal dried counterparts. However, the range of values for both the
forced and the normal dried samples greatly overlap with each other. As shown in Table X, the difference
between the HA and HN samples, of 3%, was much smaller than that observed between the PA and PN
samples, of 22%. However, this greater difference of the latter was heavily influenced by the HN3 sample as
can be seen in Table IX, which presents a significantly higher value than the other PA and PN samples, being
possibly just an outlier.
According to Walker (Walker, Pavía 2014) ‘the common industry figure of water vapour diffusion
resistance factor (μ) of lime –hemp concrete is 4.85 ± 0.24 measured in accordance with EN12572 for samples
with a binder:hemp:water ratio of 2:1:3 and a density of c. 400 kg/m 3’. As the samples presented a lower μ
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
The thermal conductivity results showed that the rapid dried samples had a lower thermal
conductivity in the first tests when compared to the slow dried ones. After all samples being placed in the
same controlled environment (20°C, 60%RH) for more than one and a half months, the influence of the dryingmethod greatly diminished, suggesting a tendency for stabilization at a common value for both drying
methods. However, the failure in correctly measuring the densities and moisture contents in different
occasions led to inconclusive results. What seems clear is that both formulations performed well and similarly
in the tests. Also, with the values obtained it is clear that the forced drying method does not have a sensible
negative effect on thermal conductivity of hemp lime and thus would not hinder this technique from being
used.
The moisture buffer value test did not achieve a quasi-steady state as prescribed by the Nordtest
method. However, the results obtained were still valid and useful for the analysis. From them it was seen
that a hemp lime made from a high surface area lime with a CSA additive had a consistently better moisture
buffer capacity than the hemp lime fabricated with the commercial PF70 binder.
When analysing the drying methods alone, the MBV results are not so conclusive. The samples made
with PF70 binder suggest that the forced drying process reduces the moisture buffer capacity of the material.
However, the HSA lime samples have a greater mean MBV when forced dried. This apparent advantage is
strongly influenced by one dissonant value of a forced dried sample, while all other values obtained suggest
the same behaviour pattern seen with the P formulation. Thus, it is still not possible to state conclusively if
and in what cases the forced drying process may affect either positively or negatively the material. What can
be said is that the forced drying of hemp lime does not have any hindering, incapacitating effect on its
moisture buffer properties.
The dry cup method, employed to assess the vapour permeability of hemp lime, also did not comply
with the requirements of quasi-steady state from the standard. However, the results were more precise than
those obtained for MBV and were also valid for analysis. The P formulation performed worse than the H
formulation, independent of how the material was dried.
When assessing the δ value, both hemp lime formulations had a better performance when slow
dried. The individual results of each sample also follow this tendency. Thus it can be asserted that the forced
drying of hemp lime has a negative effect on the vapour permeability value. However, the results suggest
that this influence is notably more expressive with the P formulation than with the H one.
The fact that the forced drying process lowered hemp lime’s δ would also support the hypothesis
that this drying process has a negative effect on the moisture buffer capacity of hemp lime. Nonetheless, the
lowering of δ in the tests although sensible, was nowhere near to discourage the use of rapid dried hemp
lime. The forced dried material was still very vapour permeable, even more than the standard hemp lime in
the market, showing that factors such as formulation and density may easily compensate for any negative
effects of the drying procedure.
With the results obtained, it can be safely said that there are no impediments for using the rapid
drying method tested in this research in the production of hemp lime, other than guaranteeing the minimum
mechanical strength required. The experiments have shown that hygrothermal parameters of both rapid and
slow dried hemp lime are significantly close to each other and that the differences which may arise from
them can be corrected in the formulation of the material. The forced drying method thus represents a
promising method for improving the use of hemp lime as a building material.
Comparing the two formulations tested, it was seen that the HSA lime and CSA composition provided
better overall characteristics, having either similar or better values of hygrothermal parameters when
compared to the PF70 formulation, while maintaining a much better mechanical integrity than the latter.
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
This dissertation intended to provide an understanding of the effects that forced drying hemp lime
had on its hygrothermal properties. Two hemp lime formulations were tested, differing from themselves in
the binder used. One used a high surface area lime (HSA lime) together with a small addition of calciumsulfoaluminates (CSA), while the other used the commercial PF70 binder from Tradical®.
Although mechanical strength was not researched in this paper, it was possible to notice from the
forced dried samples that those made with HSA lime had much better structural integrity when compared to
those made with PF70 binder.
Thermal conductivity was assessed with the hot-wire method and each set of samples was analysed
on two different dates. The results were initially better for the forced dried hemp lime, but in the second
testing the results of both drying methods were very close. A problem in the research was that the moisture
content and dry densities were not obtained correctly and thus the thermal conductivity values obtained
offer only limited conclusions.
To obtain the moisture buffer valued the Nordest protocol was followed. The samples were not able
to achieve the quasi-steady state described by the method, however the results were still analysed. They did
not show a consistent tendency for the behaviour of forced dried hemp lime. However, a likely hypothesis is
that this process reduces the moisture buffer value of hemp lime.
The vapour permeability results were obtained by the dry cup method following BS EN 12086-2013
and were also unable to achieve stability. Nonetheless, the results were still analysed and there was a
concrete tendency for a reduction of hemp lime’s water vapour permeability when the samples were forced
dried.
It was concluded that, although some observed differences were sensible, they did not present any
hindrance for the use of forced drying in hemp lime production, which showed to be an interesting method
to be applied in practice. From the formulations tested, the HSA lime and CSA mixture was the one judged
to perform the best overall.
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
[1] Ahlberg, J., Georges, E., Norlén, M., 2014. The potential of hemp buildings in different climates: A
comparison between a common passive house and the hempcrete building system.
[2] M. Barclay, N. Holcroft ,A.D. Shea, 2014. Methods to determine whole building hygrothermal
performance of hemp –lime buildings. Building and Environment , Volume 80, Pages 204 –212.
[3] Bart, M., Moissette, S., Oumeziane, Y. A., Lanos, C., 2014. Transient hygrothermal modelling of
coated hemp-concrete walls, European Journal of Environmental and Civil Engineering, 18:8, 927-
944.
[4] S. Benfratello,C. Capitano,G. Peri,G. Rizzo,G. Scaccianoce,G. Sorrentino, 2013. Thermal andstructural properties of a hemp –lime biocomposite. Construction and Building Materials, Volume
48, Pages 745 –754.
[5] Bruijn, P., Johansson P., 2013. Moisture fixation and thermal properties of lime –hemp concrete.
Construction and Building Materials, Volume 47, Pages 1235 –1242.
[6] Bruijn, P., Johansson P., 2014. Moisture transport properties of lime –hemp concrete determined
over the complete moisture range. Biosystems Engineering, Volume 122, Pages 31 –41.
[7] T. Colinart,P. Glouannec,P. Chauvelon, 2012. Influence of the setting process and the formulation
on the drying of hemp concrete. Construction and Building Materials, Volume 30, Pages 372 –380.
[8] F. Collet,J. Chamoin,S. Pretot,C. Lanos, 2013. Comparison of the hygric behaviour of three hemp
concretes. Energy and Buildings, Volume 62, Pages 294 –303.
[9] Dubois, S., Evrard, A., Lebeau, F., 2014. Modeling the hygrothermal behavior of biobasedconstruction materials. Journal of Building Physics 2014, Vol. 38(3), pp. 191 –213.
[10] Hemcrete, n.d. Building lime innovation.
[11] Hirst, E., Walker, P., Paine, K. and Yates, T., 2010. Characterisation of low density hemp-lime
composite building materials under compression loading. In: Second International Conference on
Sustainable Construction Materials and Technologies, 2010-06-28 - 2010-06-30, Acona.
7/25/2019 Forced Drying of Hemp Lime and the Effects on its Hygrothermal Properties
[23] Shea A, Lawrence M., Walker P., 2012. Hygrothermal performance of an experimental hemp –lime
building. Construction and Building Materials, Volume 36, Pages 270 –275.
[24] Shea, A. D., Wall, K. and Walker, P., 2013. Evaluation of the thermal performance of an innovative
prefabricated natural plant fibre building system. Building Services Engineering Research andTechnology , 34 (4), pp. 369-380.
[25] Sutton, A., Black, D., Walker, P., 2011. An introduction to low-impact building materials.
[26] A. Telesca,M. Marroccoli,M.L. Pace,M. Tomasulo,G.L. Valenti,P.J.M. Monteiro, 2014. A hydration
study of various calcium sulfoaluminate cements. Cement and Concrete Composites, Volume 53,
Pages 224 –232.
[27] Tucker, S., Latif, E. and Wijeyesekera, D. C., 2014. An experimental study of moisture buffering of
bio-insulations in lofts. Structural Survey , 32 (5).
[28] Walker, R., Pavía, S., 2012. Impact of water retainers in the strength, drying and setting of lime
hemp concrete. Bridge and Concrete Research in Ireland (BCRI), Dublin, September, Caprani &
O'Connor, 2012, 355 – 360.
[29] Walker, R., Pavía, S., 2014. Moisture transfer and thermal properties of hemp –lime concretes.Construction and Building Materials. Volume 64, Pages 270 –276.
[30] EN 12086. Thermal insulating products for building applications — Determination of water vapour