Insulating solid wall houses using plant based materials
Low Carbon Manager, Judith Thornton, has been busy testing plant-based insulation materials at IBERS, Aberystwyth University. Here she reflects on her findings to date.
Last month the BEACON facility in Aberystwyth saw a flurry of activity in what used to look rather like a shed full of junk, but is becoming our testing facility for internal wall insulation systems. It is essentially a room adjacent to a barn that we use for storing biomass. The wall between the two is a double skin brick wall typical of many older brick houses and the plan is to heat up the room to ‘indoor’ temperatures, whilst allowing the adjacent barn to remain at outdoor temperatures. We can then apply a variety of insulation materials to the wall and measure heat movement across the wall by attaching a heat flux meter and associated temperature sensors.
There are a number of ways of measuring the insulation performance of materials, ranging from using probes that apply heat to a block of material on a lab bench and measuring how long it takes to dissipate, all the way through to measuring the heat flow across walls in people’s houses. In general, it is easier to control variables in the lab, but the latter systems are more similar to real life installations. Our aim in the current studies is to use equipment that would be used to measure walls in people’s houses, but under the relatively controlled conditions of a known wall and stable temperatures. We are working with Hawkland Ecological Construction, who have used hemp-lime as a cast internal wall insulation, in conjunction with Neighbourhood Construction who have been installing these systems for 10 years. The aim is to both measure the performance of the system they generally apply, and to investigate modifications to it that might improve performance. Steve Cole, an MSc student from the Centre for Alternative Technology is undertaking some of this work for his final research dissertation, alongside other measurements that will enable us to understand more about the underlying mechanisms of performance of the material. The work is being partially funded by a ‘business innovation voucher’ from the BBSRC Plants 2 Products network.
Having visited a number of installations of this type, my initial impressions are that the system improves thermal comfort for occupants in three main ways.
Mechanism 1: It improves the U value of the wall. The U value is a measure of the amount of heat that is lost through the wall in steady state condition. Stationary air doesn’t transmit heat very well, and most insulation materials are designed to trap air to take advantage of this insulating property.
Mechanism 2: Airtightness is addressed as part of the installation process. The importance of airtightness in domestic retrofit is often overlooked in favour of insulation, but having a highly insulated building, whilst neglecting to fill all the tiny gaps in building envelopes that allow air movement, makes about as much sense as leaving a window wide open all winter. Typical installations therefore include removal of timber features (e.g. architrave, timber linings, window boards), judicious use of expanding foam, and then reinstallation of the features. The potential for air movement through electrical and plumbing fittings is also investigated (draughts through electrical sockets are extremely common).
Mechanism 3: An increase in wall surface temperature. We are probably all familiar with the feeling of warmth we get when sitting under a patio heater in a pub garden, and being comfortable even though the air temperature is extremely low. Being next to a cold surface gives us the opposite sensation; we feel cold even when the air temperature is warm enough. Rooms that have been insulated with hemp-lime feel noticeably warmer, because the walls are warm rather than cold.
Whilst these are the most obvious ways in which these insulation systems work, there are likely to be other mechanisms at work. One that we are keen to investigate in the future is the extent to which these systems can be used to buffer changes in internal humidity and how this relates to thermal performance. More of this in a future blog post.
Meanwhile, back to the wall. The first job was cleaning the paint off the wall and installing the formwork that the shuttering would be attached to. This involved fixing wooden plugs into the wall, and then attaching rails to the plugs. This was more involved than if the system was being installed in a house (when only the wooden plugs would be needed), but was installed in order to allow the test samples to be kept separate and to be installed at a uniform size and consistency.
The formwork used to provide initial support to the test samples. The wall is also moistened prior to installing the samples.
Next up was sieving the material (one of the modifications we are interested in is changing the particle size) and preparing the mixes. We were then ready to start installing. The material is cast behind formwork, which is immediately removed and then moved up the wall (the material is sufficiently strong to support its own mass, and then cures over the course of a month to form a relatively robust surface which is then plastered over). We created enough formwork to do 12 panels which included the Hawkland/Neighbourhood Construction ‘standard’ installation mix, alongside variations in density, plant to binder ratio, and type of plant used. Several of these were mixes that we had previously created test cubes of (read more on that here).
Whilst it isn’t part of the usual installation protocol, we sieved the plant material to minimise the number of variables between the samples.
Mixing of bioaggregate samples. Binder and plant material are mixed initially, with water added afterwards.
The mixture is installed behind temporary plywood sheets in raises of around 100mm, which are then immediately removed and moved up the wall.
The finished set of samples, which will be left to cure.
Over the next month or so the panels will cure and by the time the weather gets cold, will be ready for testing. Watch this space.