A detailed parametric study of hygrothermal behaviour of a wall made of hemp concrete submitted to hygrothermal shock has been carried out and showed that temperature and relative humidity variations in a wall are very sensitive to thermal properties,moisture transport coefficient and sorption isotherm.Up to now,most hygrothermal tools have used the isothermal sorption curves that express the equilibrium between the moisture content and relative humidity in the representative elementary volume at a constant temperature.However,few works studied the effect of temperature on hygrothermal behaviour of building envelope.This article aims to study the effect of the temperature-dependent sorption on the prediction of hygrothermal behavior of a hemp building envelope submitted to a variation of temperature and relative humidity.First,the details for the mathematical model are shown.The models were elaborated and implemented in the Simulation Problem Analysis and Research Kernel,which is adapted to the complex problems.Then,the simulation tools are validated with experimental results obtained from the test wall realised in our laboratory.After being validated,the effect of non-isothermal conditions on the temperature and relative humidity profiles will be discussed.In the next part,the mathematical model for the coupled heat and moisture transfer in building materials will be presented.Most of the models have nearly the same origin; the main difference among them is related to particular assumptions used.In this article,the model that takes into account liquid and vapor moisture transport is used.Forms of moisture transport depend on the pore structure as well as on the environmental conditions.The liquid phase is transported by capillarity whereas the vapor phase is due to the gradients of partial vapour pressure.The comparison between the variation of temperature and relative humidity at point C obtained from the simulation using the Isoth model and the one from the experimental measurement is presented in Fig.2 and Fig.3.One can observe that the variation of temperature is very similar for the three studied cases.The model gives a quite satisfactory prediction of temperature within the wall,despite the underestimation of the maximum temperature and the overestimation of the minimum temperature.
Concerning the variation of relative humidity,cannabis grow racks the computed results did not fit to the experimental ones.This should be explained by the fact that the studied model neglected the effect of temperature on the moisture sorption capacity of the material,in which increasing temperature results in increase of the relative humidity at given water content.Therefore,the dependency of sorption characteristic on temperature has been taken into account in the physical model and the result will be presented in the following subsection.As mentioned in the subsection 2.2,both Milly’s model and Poyet’s model have been used to study the impact of non-isothermal conditions on the hygrothermal behavior of hemp concrete.Because the results obtained from the Milly’s model is very close to those from Isoth model,they are not depicted here.This subsection focuses only on the results obtained by using Poyet’s model and the comparison between its results and experimental data are shown in Fig.4 and Fig.5.As can be seen from Fig.4,the numerical results are in accordance to the experimental results.In addition,they are very close to those obtained by using the Isoth model.Fig.5 showed that compared to the final one,the Non-Isoth model results in significantly better prediction of the relative humidity variation in the tested wall.Concerning three cases studied,the results are dependent on which the sorption curve is used for the simulations.The results calculated with the adsorption allow a better prediction of the relative humidity variation than those for the model that uses an average sorption curve between adsorption and desorption or desorption curve.The maximum difference between the computed results for the model that uses adsorption curve and the experimental ones is 3% RH.This value is small compared to the accuracy of sensor inserted in the tested wall,which is ±1.5% of RH.The environmental impact of construction sector is increasing thus to satisfy the sustainable development criteria,new constructions need to recycling existing building construction materials or to use bio based materials which have low environmental impact.In this context the hygrothermal behavior of a real sized hemp concrete wall is studied in this paper.
Hemp based structural materials are more and more often used all over the world,especially in France,Great Britain or China,countries which have a long tradition of hemp cultivation.Since the eighties different research have been done on the hemp concrete.Now experimental and numerical studies are available from material scale studies to building scale analysis.Usually wall sized structure use a lime-hemp mixture.Here a blending of Prompt Natural Cement and the non-fibrous part of hemp,called shiv,was used and studied walls were exposed to weather conditions on the outside and controlled conditions on the inside.The PASSYS test cells were presented previously in.They were originally designed to test passive solar components under real weather conditions .Two of them,both located at CEA INES experiment platform in Le Bourget du Lac,south-eastern France,were used in the present experiment.They are 8.44m long,3.61m wide and 3.8m high.The cells are made of a metallic frame insulated by 48 cm of polystyrene and mineral wool to have 5 highly insulated,water and vapor proof walls.The 6th face is reserved for 3.6*3.3 m² sized testing façade here facing the South.Each cell is placed on a raised support which allows free air circulation around all sides and has a system to control indoor temperature and an independent system to control relative humidity.Indoor moisture generation is not limited in mass,generated quantities are recorded by a data logger connected to a balance on which an ultra-son moisture generator is placed.The studied wall was built in precast hemp concrete blocks system,made with Prompt Natural Cement,see Fig.1 with a concrete column.Heat bridge breaking insulation was applied around the whole structure while the outside coating covers all elements of the structure.The sensors position in the wall structure is presented in Fig.2.A vertical and two horizontal gradients,in depth and in width of the wall,are measured.Two kind of recorded data are reported here,temperature and relative humidity,heat flux measurements are not involved in this paper.
To measure RH,a capacitive sensor was used,which was protected by a manually added device in order to avoid the chemical attack of concrete on sensible element of the sensor head and the direct contact with liquid water.This method implies that we adopted the hypothesis that the sensor and its environment are in moisture equilibrium,thus recorded data is measured in air but considered in material.No weight measurement of moisture content is involved in this paper.3D behavior of the studied wall was analyzed.Measured data is presented in two section of the façade: concrete post zone and hempcrete blocks zone for a one year long period.In Fig.3.temperature profiles in both vertical sections are presented in both test cells on the outside and P4= PASSYS 4 covered with traditional lime-based coating on the outside.The data recording time step was one minute.On each part of the figure three heights are represented,bottom,middle and top of the wall as it was seen in Fig.2.Upper part shows the evolution of the temperature in the room air and under interior coating.These graphics puts in evidence that between indoor air and under inside coating only a small difference is perceived.Concerning vertical gradient,the discrepancies between the 3 heights are not significant,gap does not exceed 1°C.And it is valid for both test cells.Recorded data in the concrete post has,as expected,less variation.The curve is less perturbed by the rapid fluctuation of outdoor conditions.Nevertheless a slow seasonal variation is observable,as during winter period the concrete column temperature is lower than during the summer period,independently from indoor conditions.Outdoor temperature variation affects the measured data under outside coating independently from the presence or not of the concrete post behind as recorded data has the same variation in both presented sections.Initial conditions are the same at the three different heights.This similarity was kept during the one year measurement period.It suggests that the homogeneity boundary condition,that we supposed to be valid in the whole wall,is satisfied for temperature considerations.The same kind of presentation is following for relative humidity measurements.For a better understanding,one average data per day is printed in the figure.Fig.4.presents the two above mentioned vertical sections with always the three different horizontal cuts.The impact of initial boundary conditions is clearly seen in this figure.The intrinsic moisture content of hemp concrete influences the measured data during a period which lasts several months.The highest values were measured at the bottom of the wall while the lowest values were measured at the middle.The top section values are between the bottom and middle values.
One probable reason is the involuntary rain water leakage of the wall from outside during rainy periods.It could come from the experimental installation as the lower part of the wall was not protected against this kind of water infiltration.This effect was not measured separately and remains a suggestion.Another reason could be due to different initial moisture content,but after several month this effect was faded,cannabis grow system and moisture content varied in function of boundary conditions rather than in function of initial conditions.The different sensors converge to the same value in time which confirm the homogenous wall hypothesis which may be applied after an initial period only.This question was also studied numerically in the PASSYS 3 test cell and presented later on.A huge difference,up to 30% of RH,between the two test cells was observed.It is partly due to different initial moisture content and partly due to different outside coating even if both walls were oriented to South and they had received the same weather conditions.A parallel analysis showed that the traditional lime-based coating was more water permeable than the industrial coating,thus outside moisture is more absorbed by the coating of P4 than the coating of P3.The detailed analysis of the coatings is not the object of this paper.The 3D analysis shows that after the initializing period the differences between the 3 studied heights stay under 10%.This is corroborated by the center graphs of Fig.4.Once the initial conditions are no more important,the spatial moisture distribution in each studied wall can be considered as homogenous,except around the concrete post which has a slower drying kinetic as it is shown on the right side center graph of Fig.4.In general the presence of a slow drying process in the core of both walls is observed without reference to concrete block or column.In Fig.4.during the Decemberand January period,under the outside coating,RH values stayed around 100% in P4 while in P3 RH values progressed more in correlation to the rest of the wall than to the outside conditions.This important difference is induced by the outside coatings as it was mentioned above.The research team recently developed an innovative system with low environmental impact for the production of semi-rigid panels for thermal and acoustic insulation,obtained from recycled sheep’s wool,from Piemonte region.Starting from the previous work,a new semi-rigid panel was produced,where recycled sheep’s wool is combined with hemp technical fiber.Hemp cultivation presents huge benefits for soils in crops rotation environmental field: this culture easily adapts to different types of climate and high yields can be produced with relatively low resources; moreover it exerts a restorative action on soils,leaving a considerable amount of organic waste,which benefits the cultures that follows in the rotation crops,as stated by Assocanapa.Moreover the high annual biomass production,and its strong ability to absorb CO2 from atmosphere,hemp can be considered an interesting alternative source of energy valorization and biomass.Interest in building for hemp materials especially relies in its recyclability,hygroscopic,water vapor permeability and durability properties,its resistance to mold and fungus attacks and the porous characteristics of fibers and shives; as well as its low environmental impact and low cost.Hemp cultivation has deep roots in Italy,since medieval times,during Maritime Republics age when even in Piemonte region,particularly Carmagnola gradually became a renowned center in the production and diffusion of hemp in the region and abroad.Up to 1960 Italy was among the first countries in the world for planted area and quality of products.After a period following World War II in which hemp cultivation on our territory began gradually to disappear,it was re-introduced at the end of 1990s on a 350 ha area approximately; nowadays it is recovering importance thanks to the efforts of producers as Assocanapa and the recovered cooperation with industrial sectors on local short chain mechanism.Products obtained from hemp stalks processing are shives,fibers and dust.