Construction and As-Built Performance of a Miscanthus Straw Bale House

Abstract

Houses constructed using straw bales have typically been built from wheat, rice, or barley straw, depending on local availability. Miscanthus is a perennial biomass crop with a high lignocellulose content that is grown on agriculturally marginal land. We describe the construction and as-built performance of what we believe to be the world’s first Miscanthus straw bale building. We describe the practical differences in working with the material that arise due to the slightly different physical properties of the baled material. The moisture content of the walls 17 months after construction was 11.3 ± 0.5% (pre-construction 10.72 ± 0.4% n.s.d). The in situ U value of the wall was 0.162 W/m²K, which compares to a reported U value of 0.189 W/m²K in wheat straw bale buildings of comparable wall thickness. Given the greater resistance of Miscanthus to biodegradation than wheat straw, its wider use as a construction material should be considered.

1. Introduction

The 17 Sustainable Development Goals (SDGs) provide a global framework for addressing environmental, social, and economic challenges. Architects and urban planners can contribute particularly to SDG 11 (Sustainable Cities and Communities) and SDG 12 (Responsible Consumption and Production) through the use of locally sourced, renewable materials that reduce the carbon footprint of construction. The opportunities and barriers to the uptake of local or sustainable materials and the extent to which this approach challenges mainstream construction practices has been studied in a variety of countries and contexts [1,2,3]. Low-energy buildings also align with SDG 7 (Affordable and Clean Energy) and SDG 13 (Climate Action) [4]. The construction industry is the largest contributor to climate change worldwide (34% of total emissions, [5]), and, consequently, interest in low-impact building materials is increasing. Furthermore, the potential to sequester carbon in building materials is seen as a potential route for greenhouse gas removal from the atmosphere in the EU [6]. Plant-based building materials are therefore increasingly regarded as a potential means for making the construction industry more sustainable, with products such as hempcrete [7], bamboo [8], cork [9], and straw [10] being increasingly studied. Since plant straw can be produced locally, is renewable, and can sequester carbon, it is regarded as a highly sustainable raw material, as evidenced by its Environmental Product Declaration [11].

Walls of buildings have been constructed from straw bales since the 19th century [10], although they remain relatively uncommon (e.g., several hundred in the UK, as estimated by Edmunds, personal communication). Today, buildings using straw as the main wall component typically use one of three approaches. The first is load-bearing construction, where straw bales are used as the structural element of the walls and a structural roof plate is used to carry the roof, as described by [12]. Concerns regarding structural stability and the ability to secure insurance and finance mean that a more common approach is to construct a structural timber frame with the bales then used as infill [10]. Prefabrication systems are also available, including one using bales in timber frame panels [13] and, more recently, compressed into open panels with twin stud frames [14].

Commonly regarded risks associated with straw as a building material include fire resistance, resistance to vermin, and resistance to decay, as discussed by [10,15,16]. The material properties of straw bale buildings are reviewed by [17]. Design details to combat the risk of decay of the building material are primarily aimed at preventing water ingress, with vapour-permeable construction aimed at preventing moisture accumulation in the walls also being common [10]. However, relatively little work has considered the nature of the straw per se, and the extent to which the type of plant straw used would impact on decay risks.

The majority of straw bale buildings built in the UK and Europe have used wheat straw as the construction material, with other plant materials having been used in other countries, e.g., rice straw in China and Japan [18]. The relative suitability of different plant straws is under-explored, with choice usually being governed by local availability [19].

Miscanthus is a member of the grass family that is widely grown ornamentally in the UK and Europe, and is regarded as a promising biomass crop for combustion and green manufacturing, with 8,760 hectares grown in England in 2023 according to [20]. In the UK, the Climate Change Committee recommend that planted areas of perennial biomass crops should increase to 700,000 hectares by 2050 [21] in order to provide the biomass required for negative carbon emissions via biomass energy with carbon capture and storage (BECCS). It is a perennial crop typically grown on lower-grade agricultural land, and has low or no inputs of fertiliser, herbicide, or pesticides after its second year, according to a review by [22]. In order to understand its suitability for combustion and biorefining, its physical and chemical properties have been extensively studied, e.g., in [23,24]. Miscanthus has a high lignocellulose content compared to wheat, rice, and barley straw. The polymer structure of lignin varies, with high S/G lignin ratios conferring recalcitrance to cell walls both by acting as a physical barrier to hydrolytic enzymes, and to irreversibly adsorb such enzymes, as discussed by [24]. As such, it may offer some protection against fungal decomposition and is also reported to hinder mould growth [25]. The silica and wax content of straw also vary significantly, and whilst both influence resistance to biological attack, their variation is highly dependent on the environment in which the plant is grown. The anatomical features of plant straws also have a significant bearing on straw degradability. In an 8-week study using litter bags in compost, Miscanthus was significantly more resistant to breakdown than wheat [26]. Given these properties, Miscanthus bales would seem to be well suited for use in construction. Whilst it has been used as roofing thatch in Japan and other countries for over 10,000 years [27], to the best of our knowledge, it has never been used as a walling material. The novelty and significance of the current study is that we (a) build what we believe to be the world’s first example of a house constructed using Miscanthus straw bale walls; (b) describe the modifications in techniques necessary to accommodate the physical characteristics of bales; (c) use a novel but simple approach to monitoring moisture content in bale walls; (d) report the as-built performance of the building; and (e) provide a thermal conductivity value for Miscanthus bales for the first time. If Miscanthus bales can be demonstrated as suitable for use in the construction industry given its composition and anatomical properties, its suitability for growing on lower-grade land, and the projected increases in planted areas in the UK and Europe [22], it could provide additional end-markets for growers and increase opportunities for carbon sequestration in building materials [28].

2. Materials and Methods

2.1. Building Description

The building is single-storey with an internal floor area of 43 m² and a mezzanine floor extending over 26 m² (Figure 1a–c). It is located in Wales, UK (53° N, −4° W). It was constructed in 2017–2018 as a self-build project by a private client, based on a design by a local architect. The client project-managed the build process, with input at key stages from a specialist straw bale building company. The building has an insulated concrete floor slab, with perimeter foundations of concrete topped with a slate plinth. The construction is a structural timber frame of Douglas fir, with seven courses of straw bale to the height of the eaves. The walls are clay, plastered internally. The roof is pitched (birch ply, blown cellulose fibre insulation, sarking board, membrane) with a reclaimed slate exterior on battens. The additional wall height at the gable ends is blown cellulose insulation in a timber stud frame. The building is connected to mains water and electricity, has 4 kW solar PVs, and has a private sewage treatment system. There is a wet underfloor heating system in the concrete slab powered from an electrically heated thermal store. The as-designed SAP rating was 82 B.

2.2. Miscanthus Bale Characteristics

Miscanthus bales were sourced from Terravesta Ltd. (Lincolnshire, UK) and were a mixture of the commercially available Miscanthus × giganteus clone and three experimental varieties with similar characteristics. The bales were produced with a Massey Ferguson 1840 square baler (Kenilworth, UK), and had 2 polypropylene strings. A subset of 20 bales was selected for basic measurements prior to construction. The bale weight was 18.3 ± 0.3 kg, with dimensions of 1060 ± 9, 450 ± 4, and 350 ± 2 mm in length, width, and height, respectively. The bale density was 108 ± 2 kg/m³ (all values mean ± standard error). The moisture content (Protimeter balemaster calibrated for straw) was 10.72 ± 0.4% (ambient temperature 16 °C, relative humidity 32%). Any bales that were irregular or damaged or had become damp whilst on site were discarded.

2.3. Air Permeability Testing

Air leakage was tested according to BS EN 13829:2001. The main door to the house was filled with an adjustable door frame, flexible canvas panel, variable speed fan (Retrotec 200-series DucTester), and a pressure and flow gauge (Retrotec DM-2 digital gauge). All external doors and windows, other than where the test equipment was mounted, were shut for the duration of the testing, whilst internal doors were kept open to ensure free movement of air within the building. Extract fans were sealed. A broken bedroom window was temporarily sealed using airtightness tape. A depressurisation test was then undertaken. Thermal imaging (Flir B335, 320 × 240 pixels, Flir Systems Inc., Wilsonville, OR, USA) was undertaken outside the building during the depressurisation test to identify points of air leakage. The building was then depressurised again whilst thermal imaging was undertaken inside the building in order to better understand pathways of air leakage. Thermal imaging was also repeated on a subsequent occasion when the difference between indoor and outdoor temperatures was greater.

2.4. In Situ Moisture Monitoring

As discussed above, moisture ingress and its potential to cause decay are a common concern with straw bale buildings [10]. The approach of permanently embedding relative humidity sensors into walls has previously been described [29]. This allows for long-term monitoring, but its utility is limited by the initial choice of sensor location. The current study therefore made use of a straw bale moisture probe in order to determine whether any areas of the walls had high moisture contents that might merit the use of long-term sensors. The survey was prompted partly by a detectable staleness being smelt during the air permeability testing and was carried out in December 2018. Locations for moisture probing were determined in consultation with the house occupier and a member of the construction team, and were focussed on the wall areas they regarded as higher risk. These were (a) the top course of bales (owing to delays to roof construction and concerns about rain ingress); (b) the wall in the south west corner (most exposed to prevailing weather during the build process); and (c) areas in proximity to door and window reveals. A ‘control’ location in a wall directly behind a woodburning stove was also probed. The locations of probing are given in Figure 2. Holes of 10 mm were drilled in the clay plaster wall internally, and a 600 mm bale probe (Protimeter Balemaster) was pushed into the bale and moisture measurements taken. The probe was marked to allow measurements to be taken at approximate depths in the wall, which were 160 mm (inner portion of wall), 280 mm (centre of wall), and 380 mm (outer portion of wall).

2.5. In Situ U Value Measurements

In situ U value measurements were carried out in accordance with ISO 9869-1 (2014) [30]. An area of wall on the east elevation was selected for taking the measurements. It was not subject to direct sunlight, and was midway up the wall, with no windows or other building elements in proximity (>700 mm to any other building element or junction). This elevation had clay plaster on the inside of the wall, and housewrap, battens, and timber cladding on the outside. A thermal imaging camera (Flir B335, 320 × 240 pixels, Flir Systems Inc., Wilsonville, OR, USA) was used to verify the uniformity of the wall area prior to testing. Heat flux plates (Hukseflux HFP01, Delft, The Netherlands) had a thin layer of Vaseline applied, followed by clingfilm, and were then pressure-fixed against the wall using vertical tele-prop poles and gutter clamps. The ambient air temperature was measured indoors and outdoors using thermistors and data was recorded on an Eltek Squirrel 850 datalogger. Additional room heating was used to increase the temperature difference between indoors and outdoors. The equipment was left in place for 27 days, and the data was later examined to determine a time period in which the conditions for the average method under ISO 9869-1 were met. Assumed values of thermal conductivities for timber cladding and clay render and internal and external surface resistance were taken from CIBSE Guide A [31] to allow a λ value to be calculated for Miscanthus straw bales.

3. Results

3.1. Construction of the Building

Construction proceeded in the usual way for a structural timber frame with bale walls, as described in detail by [19]. Following pouring of the floor slab and the preparation of strip foundations, the timber frame was erected, and the initial stage of the roof of the building was constructed (fibreboard on roof joists, later followed by roof battens and slate). The first course of Miscanthus bales was secured to a timber base plate using timber stud pins. Subsequent courses of bales were staggered by half a bale length. Where bales were to be placed where they met timbers, such as for window boxes and door frames, notches were cut in the bale ends using either a reciprocating saw or a chainsaw (Figure 3a,b). Temporary studs were placed in building corners to maintain good bale alignment and to prevent walls bowing out between courses (Figure 3c). As is the case when building with wheat straw bales, some bales required modification before use, including straightening, string realignment, and string tightening. Bales were reduced in length as required using the usual method, where a sash clamp is used to temporarily hold the straw compressed, whilst strings are cut and reknotted after the required number of straw flakes have been removed.

Any gaps around the bale ends caused by irregularities in shape were stuffed with barley straw. Following the final course of the bales being laid, a 22 mm plywood compression plate was applied and the walls were compressed using the usual approach of heavy-duty ratchet straps to apply the compressive force prior to fixing permanent polyester straps in place (Figure 4).

Prior to clay plastering of the internal walls, the wall surfaces and bale ends were smoothed by shaping with a chainsaw. In order to be in keeping with the local building aesthetic, and to provide protection against driving rain, the outer surface of the walls were finished according to their orientation (Tyvek housewrap, and then slate cladding on west elevation wrapped around to north and south, timber cladding on north and east elevations, and lime render on south elevation).

3.2. Comparisons with Using Wheat Straw Bales

Qualitative remarks were gathered from four experienced builders who had previously built with wheat straw and are summarised in

Table 1. Advantages and disadvantages of Miscanthus straw bales compared to wheat straw bales as perceived by experienced builders.

Advantages	Disadvantages
Stems seemed stronger.	Risk of skin abrasions, work gloves required.
Bale shape easier to manipulate.	Fragility of bale structure resulting from smoothness of stems and high degree of strand alignment.
Easier to cut accurate notches for window frames.	Straw too rigid to use for gap stuffing at bale ends.
	Chainsaw required for shaping of wall surfaces and bale ends.
	Uneven bale surface, bale shaping to reduce quantity of render required therefore particularly important.
	Difficulty in applying initial key coat of render by hand.

.

3.3. Air Permeability Testing

During the air permeability test, the initial inside temperature was 18.4 °C and the final inside temperature was 18.1 °C. The average outside temperature was 6.1 °C and the barometric pressure was 102 kPa. The air permeability @ 50 Pa was 8.4 m³/ (h·m²). Key areas of air leakage identified on the outside of the building using thermal imaging were multiple sections at the top of the wall and along the junction between the wall and pitched roof, and additionally the wall–roof junction at the western gable end (

Table 2. Spot temperatures from thermal imaging. Locations as indicated in Figure 5 and Figure 6.

Elevation, Location	Location	Spot Temperature (°C)
North. Junction between timber clad wall and pitched roof (Figure 5a)	A	7.9
As above	B	5.2
As above	C	4.5
South. Junction between lime rendered wall and pitched roof (Figure 5b)	A	3.5
As above	B	3.5
As above	C	2.0
As above	D	1.9
West. Junction between slate clad wall and pitched roof (Figure 5c)	A	−1.0
As above	B	−3.0
South. Junction between wall and timber frame (indoors, Figure 6b)	A	13.5
As above	B	14.7
As above	C	10.7
As above	D	12.1
As above	E	14.4

, Figure 5a–c). Thermal imaging on the inside of the building identified multiple areas where air infiltration was occurring along the junction between the clay plaster and timber frame (Figure 6a,b). There was also significant air movement through several electrical sockets, particularly in the north kitchen wall.

3.4. In Situ Moisture Monitoring

The temperature and relative humidity on the day of the survey (12 December 2018) were 18 °C and 47% (indoors) and 9 °C and 80% (outdoors). The minimum moisture content detectable by the probe is 8%, and the moisture content of the control area of the wall (point C in Figure 4) was below this. The average moisture content across all probe depths and locations (three depths at nine locations,

Table 3. Moisture contents for the sampling locations shown in Figure 2 during in situ moisture testing. All values stated in % moisture content.

Location	Inner Wall	Mid Wall	Outer Wall
A	11.2	13.3	19.2
B	8.6	8.2	8.6
D	9.1	9.8	8.0
E	10.4	10.0	11.6
F	9.3	10.4	11.8
G	11.7	10.5	11.0
H	10.3	9.7	10.2
I	13.4	13.8	17.1
J	11.9	12.8	14.1
Mean	10.7 ± 0.5	10.9 ± 0.6	12.4 ± 1.3
C (control location)	<8.0	<8.0	8.6

) was 11.3 ± 0.5%, ranging from 8% to 19.2%. The moisture contents taken from mid-bale areas (n = 9 locations) were not significantly different to the pre-construction moisture readings (n = 20 bales) taken in July 2017 (p = 0.7525, permutation test, Figure 7a). The mean moisture contents for the inner, mid, and outer bale areas were 10.7 ± 0.5%, 10.9 ± 0.6%, and 12.4 ± 1.3%, respectively, and are illustrated in Figure 7b. There was no significant difference in moisture contents between the inner, mid, and outer bale areas (p = 0.3095, permutation test), or between the different sampling locations (p = 0.1657, permutation test).

3.5. In Situ U Values

The monitoring equipment was left in place for 27 days. A time period was identified that satisfied conditions for the average method to be used to calculate heat flux. During this period, the indoor temperature averaged 22.6 °C, and the outdoor temperature averaged 9.4 °C. The U value of the wall calculated during this period was 0.162 W/m²K. Using example values from CIBSE Guide A (2006) [31], this translates to a λ for Miscanthus straw of 0.097 W/mK (

Table 4. Wall buildup on east elevation showing R values used to allow λ for Miscanthus to be calculated. Literature values given in italics, measured values in bold.

Material	Thickness (m)	λ (W/mK)	R (m2K/W)
External RSE			0.040
Timber cladding	0.018	0.015	1.385
Straw bale	0.450	0.097	4.592
Clay render	0.015	0.500	0.030
Internal RSI			0.130
Total R		=	6.176 m2K/W
U value		=	0.162 W/m2K

).

4. Discussion

Plant-based building materials are of interest to industry professionals due to their low carbon emissions during manufacture [32], the potential for local production [33], and more recently for their scope to provide the long-term storage of carbon sequestered from the atmosphere, as evidenced in EU Building Energy Performance Certificates and the EU Carbon Removals and Carbon Farming Certification (CRCF) Regulation [6]. Whilst timber is already widely used in the construction industry, the availability of sustainably managed forests may put an upper limit on the carbon sequestration potential of timber in buildings [28]. Furthermore, plants with rapid growth rates that are harvested annually could theoretically achieve sequestration much faster than timber, as discussed by [28]. The commercial Miscanthus × giganteus variety is more resistant to biological breakdown than wheat straw, although anatomical and chemical properties are highly diverse across Miscanthus genotypes. In a study of 244 genotypes, stem diameter ranged from 1.5 to 10.5 mm [34] and lignin, cellulose, and hemicellulose concentrations were also highly variable [35]. Breeding programmes are typically focussed on yield and suitability for combustion or biorefining, but given the genetic variation in Miscanthus, it would be possible to develop a variety that improved the characteristics of Miscanthus straw for construction even further. In contrast, both wheat and rice straw are co-products of grain production, and so beyond lodging resistance, stem properties are of limited interest to breeders as a trait.

The building process employed was, in most regards, very similar to constructing the equivalent timber frame building using wheat straw bales. The different physical properties of Miscanthus straw conferred both advantages and disadvantages and required a few adaptations. Most notable was the requirement for a chainsaw (as opposed to a reciprocating saw) to smooth walls prior to plastering, and the difficulties applying the initial key coat of render. The qualitative observation of high Miscanthus stem rigidity compared to wheat straw, as reported by the straw bale practitioners in the current study, reflects literature reports of strength parameters [36,37]. Reducing the straw chop length used during the baling process could be a route to achieving bales more physically similar to wheat straw (preliminary test by the authors, unpublished). The abrasiveness of Miscanthus stems would not be a barrier to its use in prefabricated panels, which are typically factory-manufactured by inserting straw into wall cassettes, followed by compression and surface strimming in a robotised process.

Air passing through the building fabric is likely to change temperature and therefore could result in the condensation of water vapour, and so airtight building details are important to reduce the risk of degradation of biobased building materials, in addition to the role that air leakage plays in heat loss from buildings. The air permeability of the building was below that required by Building Regulations in Wales at the time of construction (10 m³/(h·m²) @ 50 Pa), although limits have now been tightened to 8 m³/(h·m²) [38]. Much lower levels of air permeability have been achieved in straw bale buildings; several have achieved the Passivhaus standard of <0.6 m³/(h·m²), for example [39,40], in timber frame and load-bearing straw bale buildings, respectively. The relatively high air permeability in the current building could be a consequence of the coarse and stiff stem structure of Miscanthus (e.g., by making it more difficult to stuff gaps between bales with straw, Table 1), or because the building was not designed with high standards of air tightness in mind. It did not have a continuous airtight layer on design drawings and airtightness tapes were not used at junctions between materials. Several elements of the construction had junction details that were difficult to construct, particularly the wall–eaves junction above the top course of the straw bales, which was difficult to access owing to horizontal beams in the structural timber frame.

Air leakage pathways through building fabric are often relatively convoluted [41], as was the case in this building (i.e., the points of air movement on the inside of the building differed from those identified on the exterior). The shrinkage of clay plaster is common (it is not unique to Miscanthus bale walls), and airtightness at timber-frame-to-plaster junctions could be improved using airtightness tapes designed for junctions between plastered surfaces (e.g., Siga Fentrim, Contega PV). Air leakage through electrical sockets can be addressed using airtight back boxes and cable grommets for electrical sockets [42].

A moisture content below 15% is regarded as rendering a wheat straw bale wall at low risk of degradation [29], with others reporting transient increases to 25% as being acceptable [43]. Robustness to much more extreme conditions was reported in a climate chamber study designed to mimic local summer conditions; rice and wheat straw samples in an environment of 95% relative humidity and 35 °C did not noticeably degrade in a 12-week period. This was attributed to the protective effect of the render [18]. The same authors removed areas of render from the walls of an experimental building in a similar climate and found some degradation at the interface of the bale and render, but stated that the degradation was very localised. Moisture contents that result in the degradation of Miscanthus straw have not been reported. In the current study, there were two locations in the outer wall area where moisture contents were above 15% (as shown in Figure 7b), but this would appear to be highly localised; the mid and inner bale moisture contents were all below 15%, and there was no significant difference overall between the inner, mid and outer wall moisture contents, or between different locations in the building.

Whilst academic studies on the in situ moisture contents of bale walls have typically been in permanently instrumented buildings, e.g., in [29], bale probes as deployed in the current study provide a flexible and instantaneous method for checking moisture content in straw bale walls in a relatively non-invasive way. The method could usefully form part of the decision making process following accidental moisture ingress in order to determine whether installing monitoring equipment is warranted, to check that walls have dried out following remedial work, or as part of a condition survey. For example, in the current study, the moisture probing was undertaken because a musty/stale odour was detected during the air permeability test, but the bale moisture content was not significantly different to measurements taken on the bales 17 months earlier prior to building construction (Figure 7a), which provided considerable reassurance to the occupant that the walls were still in good condition. The demonstration of the bale probe method to monitor wall moisture content opens up opportunities for further research into the condition of the thousands of existing straw bale buildings, the majority of which have never been studied.

Strand orientation and bale density are both hypothesised to impact on the insulation value of straw bale walls, but the impact of straw type has not been systematically studied. Given the difficulties of measuring U values using heat flow meters, as discussed by [44] and reviewed by [45], differences between straw types would likely be difficult to ascertain. The U value of the 483 mm thick wall monitored in the current study was 0.162 W/m²K, compared to that indicated from studies of wheat straw bale buildings (e.g., 0.25 W/m²K for a 360 mm thick wall, 0.200 W/m²K (467 mm thick wall), 0.206 W/m²K (483 mm thick wall) and 0.189 W/m²K (483 mm thick wall)), as discussed by [43].

The λ values for a variety of straw types (wheat, rice, barley, and a range of unstated types), fibre orientations, and bale densities ranged between 0.033 and 0.19 W /mK in a review by [10]. Based on assumed R values for other wall components (Table 4), the λ for Miscanthus straw bales was calculated to be 0.097 W/mK.

The suitability of a perennial crop such as Miscanthus for construction significantly expands the potential for locally produced bio-based building materials to be used in areas where land is unsuited to annual cereal crops, provides a novel approach to decarbonising the construction industry, and increases the opportunities for Miscanthus growers. The low moisture content of the walls combined with the high resistance to bio-degradation compared to wheat straw suggests that Miscanthus bale buildings could play an important role in decarbonising the construction industry.

5. Conclusions

We have demonstrated for the first time that Miscanthus × giganteus can be used to construct straw bale houses. We demonstrate a new method for monitoring moisture content in straw bale walls, and provide the first report of a thermal conductivity value for Miscanthus bale walls. Further work to improve the physical characteristics of baled material, to monitor the long-term performance of Miscanthus bale buildings, and to investigate its use in prefabricated panels would increase the potential usefulness of the material as a carbon-sequestering building material. As discussed above, given the variation in stem structure and composition in Miscanthus varieties, plant breeders could potentially increase the plant’s suitability for the construction industry further.