In all cases the anisotropic internal structure of the material accounts for the clear variation in both mechanical and thermal properties of the material with testing direction. This is found to be most consistent in the thermal conductivity with a significantly lower thermal conductivity for all variations of hemp-lime in the parallel loading direction. As the thermal conductivity is arguably the most important property out of those studied, it can clearly be seen from these results that it is advantageous to have compaction in the same direction as thermal loading. Traditional vertical casting processes are therefore not advantageous in this respect while pre-casting in a perpendicular direction offers a likely benefit. In the parallel direction, the impact of binder ratio on the thermal conductivity and both the compressive rupture stress and flexural strength can be considered in-line with previous studies: an increased proportion of binder enhancing the structural skeleton of the material but providing a global densification and increase in thermal conductivity. In the perpendicular direction it is logical to expect a similar trend between increasing binder content and an enhancement of physical properties, and indeed this is observed for the thermal conductivity. For the flexural strength and compressive strength a similar pattern is observed in increasing the binder content from 1:1.8 to 1:2.2 however it is not possible to extrapolate these results. An additional increase to 1:2.6 is found to have a lower than expected improvement in flexural strength, compared to the parallel direction, and a negligible impact on compressive rupture strength. It is likely that at very high binder contents the behaviour will tend to reflect that of the binder in all directions indicating that sensitivity to orientation plateaus. As particles of hemp are generally elongated in line with the stem of the plant,vertical grow rack their porous structure and mechanical properties will also be aligned. Hemp particles and indeed most bio-aggregates could then be considered isotropic themselves with a greater stiffness along the main axis.
Parallel to compaction compressive loading hemp-lime can thus be considered as transfer through the binder skeleton, the hemp particles offering limited contribution along their highly compressible secondary axis. This rupture limit would be dependent on the quantity of binder alone. In contrast, in the perpendicular direction, load is likely to be transferred in a more composite action, utilising the stiffer axis of the particles. Rupture in this case can be attributed to a failure of bond between the particles and the binder allowing localised rotational or sheering failure. As the available surface area of the hemp particles is limited, an increase to the binder content once the surface is fully utilised would have a negligible impact on the compressive strength. Such a differing behavioural model may then explain the apparently differing impact of binder content in the different loading directions. The impact of particle size distribution on the thermal conductivity can be seen from Fig. 6 to be negligible in both directions. The density of the specimens produced with differing particle size distributions were mostly consistent, , and so the total porosity may be assumed to be similar and account for this. The mechanical tests conducted on material of differing particle size distributions show a general correlation between flexural rupture stress and compressive rupture stress although no clear trend between median particle size and the mechanical properties. This is consistent with the mixed results found within the literature that report in separate studies increasing coarseness having a positive and negative impact on mechanical properties. The properties of hemp-lime and other bio-aggregate composites are often presented with respect to weight of material as it is a general trend observed widely in the literature that mechanical resistance as well as thermal conductivity increase with density. Fig. 8 plots the results from both mechanical tests and the thermal conductivity test against material average dry density values. Lines of best fit added to the data confirm the positive correlation between the properties mentioned and density that is in line with results elsewhere in the literature. In general the binder ratio is found to fit closely to the line of best fit in all cases, Fig. 8, in both parallel and perpendicular measurements with the exception of perpendicular loading. This is indicative of the observed increases in these properties being broadly associated to the increase in density of the material and thus effective structure as well as the previously discussed consideration that the perpendicular compressive strength exhibits a plateau as a certain binder content due to the mechanism of load transfer in the material.
The grading of the particle size is seen from Fig. 8 to have a more complex relation to density and to other physical properties. It is noticeable that the coarse aggregate is consistently producing stronger and more thermally conductive material than would be obtained at a similar density produced using the medium aggregate and a lower ratio of binder. This may be attributed to the larger aggregates providing a more direct thermal path and continuous loading path.This might be accounted for by the improved natural packing of particles with a lower average aspect ratio. When considered against the line of best fit, the fine grade of shiv gives no obvious benefit in terms of mechanical behaviour and so the observed benefits found in Fig. 5 can be attributed to the increase in density. Conversely it may also be noted that finer particles might give a lower thermal conductivity than may be expected for equivalent density material with a medium grade of shiv and higher binder content. This is likely attributed to the inverse of the coarser particles providing higher thermal conductivity suggesting a link between thermal conductivity and average particle size. From Table 2 and Fig. 1 it is not obviously apparent what aspect of the coarse grade of shiv might account for the seeming mechanical over performance. What is considered likely in light of these results is that a combination of factors inherent to the particle size distribution combine to determine the impact on mechanical performance. It is considered likely that this will include the mean length of particles, mean aspect ratio of particles and spread of distribution. Further study where such variables are isolated and assessed is required to establish this and could lead to easily obtainable performance increases.
A possible limitation of the presented results is the omission of testing at other material ages beyond 28 days. It is known that hemp-lime can continue to develop strength past 28 days due to the continued carbonation of the lime binder although this has been shown to vary in magnitude according to the binder and conditions of material storage. It has been previously suggested in other studies that the particle size distribution may alter the permeation of carbon dioxide into the materials and thus alter the rate of carbonation. In this particular case, in light of the binder being known to have a significant hydraulic set and being comparable to some other previously studied binders, the potential strength gains through ongoing carbonation are considered likely to be negligible; the results are therefore considered likely to be applicable to materials tested at greater ages. It is not clear and indeed unlikely that this would be replicated for pure lime binders and further work on the combined impact of aging and particle size distribution would be a useful topic for future study.The building construction industry accounted for 37 % of global energy-related greenhouse gas emissions in 2020, 10 % of which resulted from the manufacturing of building construction materials. Along with reducing energy demand and decarbonizing power supply, it is essential to address the embodied energy within the building materials and their manufacturing process for decarbonizing the global buildings and construction sector. Policies on incorporating whole-life carbon have already gained traction in countries such as the Netherlands, Denmark and France where CO2 limits are imposed on new buildings, and more countries are expected to have similar policies in place to achieve carbon neutrality as a whole. Bio-based building materials, either wood-based or containing other natural fibres are one of the solutions for producing low carbon materials. They generally have a lower embodied energy than synthetic materials, can be sourced locally, and have diverse building applications to achieve the desired performance characteristics. Plant-based materials such as cannabis grow racks, expanded cork, straw, grass, etc. are particularly well suited for providing a satisfactory thermal insulation performance owing to their porous structure and consequently low thermal conductivity in the range of 0.037 and 0.080 W⋅m− 1⋅K− 1.
However, the most frequently used thermal insulation materials in Europe are inorganic mineral fibres, e.g. glass wool and stone wool, follow by organic fossil fuel derived foams, e.g. expanded and extruded polystyrene and polyurethane, whilst all other materials only cover the remaining 1 % of the market, including plant-derived materials. A large potential saving on GHG emissions can be realized by popularising the usage of bio-based insulation materials. The lack of widespread usage of bio-based insulation materials can be attributed to their intrinsic hygroscopic nature and tendency to absorb moisture from their surroundings, and the associated durability concerns such as mould development under humid environments. Building component with the presence of organic matter is more susceptible to mould infestation than inorganic materials. And with the growth of energy-efficient buildings which are relying on air tightness and highly insulate envelope design, these buildings are found to have higher indoor humidity, which consequently supports mould germination and growth. Alarmingly, adverse health symptoms associated with exposure to indoor moulds, such as asthma, allergies and infections have been studied and established. And the use of bio-based materials has provided the optimal medium for fungal proliferation in the built environment. In view of that, it is essential to study the hygroscopic properties and mould growth potential of bio-based insulation materials, coupled with their hygrothermal performance under different environmental conditions, to overcome these concerns and to provide a factual guideline for engineers and architects. In this research, four bio-based insulation composites are selected to examine their hygroscopic properties and hygrothermal performances under predefined built environments. These composites are selected based on their low thermal conductivity, low embodied energy and preferrable composed of recycled or waste material, and commercially available.
The declared thermal conductivity of these composites is around 0.040 W⋅m− 1⋅K− 1, which is in the same range as other conventional building insulation materials. Besides, their embodied energy is considered low, since they are made of either agriculture residues or recycled materials, and do not require an energy-intensive production process. An exception is made for mycelium composite whose production methods e.g. sterilization and inoculation suggest a higher embodied energy than other bio-based materials, however, is included for its novelty as sustainable bio-based insulation material. Several authors have investigated comparable insulation materials. A comprehensive hygrothermal characterization of expanded cork for building facades is provided by Sim˜ oes et al. , where the studied cork boards are observed to have low thermal conductivity ranging from 0.037 to 0.041 W⋅m− 1⋅K− 1 with good resistance during long term durability testing. Relevant heat transfer modelling and hygrothermal simulations on the cork boards have also been carried out by the same research group and provide valuable insights into heat and moisture transport phenomena under the simulated built environment. An overview of grass-based composites is presented in, showing average thermal conductivities between 0.034 and 0.09 W⋅m− 1⋅K− 1 , and also highlighting their good sorption desorption capability. For hemp based composite, Latif et al. and Collet et al. have reported that hemp wools show higher sorption and similar vapour resistance factor as mineral wools. The thermal conductivities of hemp wool are reported between 0.038 and 0.06 W⋅m− 1⋅K− 1. Different mycelium based composites have also been developed by various research groups, e.g. mycelium-miscanthus composites by Dias et al.with reported thermal conductivities between 0.0882 and 0.104 W⋅m− 1⋅K− 1 ; mycelium-flax, mycelium-hemp and mycelium-straw by Elsacker et al.at 0.0578, 0.0404, 0.0419 W⋅m− 1⋅K− 1 respectively; and mycelium bio-foam by Yang et al. from 0.05 to 0.07 W⋅m− 1⋅K− 1. The hygrothermal models are widely used to simulate the coupled heat and moisture transport process for one or multidimensional cases, by either taking into account a single building component or a complete building envelope.