Development of Highly Repellent Silica Particles for Protection of Hemp Shiv Used As Insulation Materials

materials

Article Development of Highly Repellent Silica Particles for Protection of Hemp Shiv Used as Insulation Materials

Marion A. Bourebrab 1,2,* ID ,Géraldine G. Durand 3,4 and Alan Taylor 3 ID

1 National Structure Integrity Research Centre, Granta Park, Great Abington, Cambridge CB21 6AL, UK 2 School of Engineering, The University of Edinburgh, Edinburgh EH9 3FB, UK 3 TWI Ltd., Granta Park, Great Abington, Cambridge CB21 6AL, UK; [email protected] (G.G.D.); [email protected] (A.T.) 4 Advanced Resins and Coatings Technologies Innovation Centre, School of Engineering, London South Bank University, 103 Borough Road, London SE1 0AA, UK * Correspondence: [email protected]; Tel.: +44-(0)1223-899000

Received: 1 December 2017; Accepted: 19 December 2017; Published: 21 December 2017

Abstract: New bio-materials have recently gained interest for use in insulation panels in walls, but wider adoption by the building industry is hindered by their intrinsic properties. The fact that such materials are mainly composed of cellulose makes them combustible, and their hydrophilic surface presents a high water uptake, which would lead to faster biodegradation. A hydrophobic treatment with silica particles was successfully synthesised via Stöber process, characterised, and deposited on hemp shiv. The surface of hemp shiv coated several times with 45 and 120 nm particles were uniformly covered, as well as extensively water repellent. Those samples could withstand in humidity chamber without loss of their hydrophobic property and no sign of mould growth after 72 h of exposure.

Keywords: hemp shiv; insulation; hydrophobicity; silica particles; stöber process

1. Introduction Hemp plant (Cannabis sativa L.) has recently drawn much interest in the construction sector for the manufacture of bio-based composite structures [1–4]. The plant does not require a lot of water to grow, is easy to harvest in moderate climates [5], and exhibits a range of interesting and useful properties. The woody core of the plant, usually named hemp shiv, is very porous with a low bulk density [4], which makes it a valuable candidate for use as an insulation material for both acoustic [6,7] and thermal [8–11] applications. The porosity of hemp provides the ability to create a breathable structure, where moisture can be absorbed and desorbed from its surrounding environment [3]. This characteristic enables such insulation to act as a buffer that regulates against sudden changes in temperature and humidity. One recent study demonstrated this effect by comparing the energy transfer efﬁciency of a hemp-lime wall assembly and a conventional mineral wool one when subjected to a sudden temperature drop. The hemp wall transferred 17% of energy compared to 75% for the mineral wool one [12]. The hemp-based insulating material thus better regulated the indoor air environment by moderating changes, which is likely to translate to reduced need for air conditioning, and ultimately energy savings. Extensive research has been carried out with similar hemp-lime wall assemblies for internal partitions [3,7,12–18]. In the example of a hemp-lime assembly, protection against biodegradation is offered via the highly alkaline lime present [18]. However, the protection of the hemp shiv only is a novel ﬁeld of study, with industrial impacts.

Materials 2018, 11, 4; doi:10.3390/ma11010004 www.mdpi.com/journal/materials Materials 2018, 11, 4 2 of 14

The objective of the ISOBIO project, under which the work reported here was carried out, is to develop low embodied energy bio-based insulation materials, for use in building envelopes either for new build or for retrofitting to existing buildings. Even though there are drivers to use more ecologically responsible materials in the construction sector, their composition still hinders a widespread usage. For example, hemp shiv are made from 40–52% cellulose, 22–30% lignin, and 18–28% hemicellulose, as well as pectins, ashes, waxes, and fat in smaller quantities [8,19]. The relative quantities vary considerably depending on how the hemp was harvested, the soil, the climatic conditions, etc. Nevertheless, the primary constituents are hydrocarbons and so are combustible, whilst the hydroxyl groups of the cellulose parts are hydrophilic. It becomes crucial to address these two issues if bio-materials are the main constituent of a building block. Another group of researchers under the same project have successfully demonstrated that hemp shiv could be chemically modified to become hydrophobic, by the means of a sol-gel derived, silica based thin mesoporous coating [20]. In common with many alkoxide sol-gel systems, this is an evolving dynamic system with a certain shelf-life and can be prone to gelling. Our approach differs from theirs in that hydrophobicity is introduced by depositing hydrophobic silica particles onto the hemp shiv. These particles are already fully reacted by the time of deposition. There are two reasons behind using inorganic material on the bio-aggregates: different sizes of silica particles can be deposited giving coatings of different thicknesses; and having inorganic silica creates a thermal shield that would then help dissipate heat when the material would be exposed to fire. The authors’ future work is focussed on this dual functionality of the silica particles. This study thus focusses on applying functionalised silica particles onto hemp shiv to provide hydrophobic properties and prevent degradation in humid environment, so that those bio-materials could be effectively used as insulation materials in building envelopes.

2. Materials and Methods The approach followed in this study was to develop a hydrophobic treatment to be applied on bio-based insulation materials, in particular hemp shiv. Silica particles were synthesised and functionalised to repel water, then applied directly on the shiv, according to the processes described below.

2.1. Materials Silica particles were synthesised using sol-gel methods with tetraethoxysilane (TEOS) as the silica precursor (Silanes & Silicones, Stockport, UK). Ammonium hydroxide (Sigma-Aldrich, Gillingham, UK) was diluted in-house from 28–30 wt % to 25 wt %. Industrial methylated spirit (IMS) (99% ethanol, 1% methanol), supplied by VWR International (Lutterworth, UK), was used as the co-solvent. Finally, water was deionised in-house. N-propyl trimethoxysilane (nPTMS) was used as the functionalising agent (Silanes & Silicones, Stockport, UK) with dibutyltin dilaurate (Sigma-Aldrich, Gillingham, UK) as the condensation catalyst [21]. Different size distributions of hemp shiv were kindly provided by Cavac Biomatériaux (Sainte-Gemme-La-Plaine, France): the smallest being Isoﬁn (around 2 mm in length), then G7 (7 mm), G8 (8 mm), Bioﬁbat (10 mm), and G14 (14 mm).

2.2. Silica Particles Synthesis In this study, all the silica particles were synthesised via the Stöber process [22], at room temperature, according to the authors’ previous work [23]. Three formulations of silica particles were synthesised: one that led to a particle diameter size of approximately 30 nm, and 45 nm when functionalised; one with a functionalised particle size of approximately 120 nm; and one with a functionalised particle size of 380 nm. All distributions were uniform and mono-modal (with a polydispersity index below 0.2), and formulated to reach 4.3 wt % silica at the end of the reaction. Materials 2018, 11, 4 3 of 14

The ammonia was then removed from the systems by evaporation using a Rotovap and IMS was added so that the ﬁnal content remained at 4.3 wt % silica.

2.3. Functionalisation Silica surfaces are typically hydroxyl rich, silica nanoparticles made via precipitation or pyrogenic methods have silanol densities between 4 and 6 OH/nm2 [24]. In the presence of a catalyst, alkoxysilanes are grafted onto the particle surfaces by reaction with the silanol groups (Si-OH) [25–27]. Having alkyl groups grafted on the particles gives two advantages: the silica particles are transformed from hydrophilic to hydrophobic, and the particles are prevented from aggregating due to sterical stabilisation. In this study, n-propyl trimethoxysilane (nPTMS) was reacted with the silica particles dispersed in alcohol via the use of the tin catalyst by heating at 65 ◦C for 18 h to allow hydrolysis-free condensation reaction between the alkoxy silane and silanol rich surfaces of the silica particles [21]. The relative quantities of the silane and the silica were varied to assess the impact of the silane. A simple gravimetric functionalisation ratio was deﬁned as the quantity of silane as a function of the quantity of silica in the suspension, following the example of Posthumus et al. [28].

2.4. Deposition Silica particles were deposited onto glass slides to allow accurate sessile drop assessment without the inﬂuence of the roughness of the hemp shiv. Coatings were prepared by immersing soda-lime glass microscope slides into the silica suspension and withdrawing at 100 mm/min. The glass slides were dried at 150 ◦C for one hour to remove all volatiles including excess unreacted silane. Prior to coating, the hemp shiv samples were dried in the oven at 90 ◦C for one hour. They were then immersed in a dispersion of silica particles for two minutes with constant mixing, to ensure all aggregates were coated by the silica particles. The bio-materials were then strained to remove all the liquid, and dried in the oven at 90 ◦C for one hour. This process was repeated to build up multiple layers of silica particles. The hemp shiv samples prepared are speciﬁed in Table1. The formulations used for the series of samples H4-x correspond to dispersions of functionalised 45 nm silica particles; samples H12-x with particles of 120 nm diameter; and H38-x to those treated with 380 nm diameter silica particles.

Table 1. Description of the samples with the characteristics of each treatment, where sample nomenclature refers to ﬁrstly the substrate, then silica particle size, then functionalisation level, and ﬁnally the number of coating layers.

Treatment Sample Substrate Functionalisation Level Coated × Times H0-0-0 hemp shiv G7 none H4-1-1 hemp shiv G7 1.00 1 H4-1-2 hemp shiv G7 1.00 2 H4-1-3 hemp shiv G7 1.00 3 H4-1-5 hemp shiv G7 1.00 5 H12-1-1 hemp shiv G7 1.00 1 H12-1-2 hemp shiv G7 1.00 2 H12-1-3 hemp shiv G7 1.00 3 H12-1-5 hemp shiv G7 1.00 5 H38-1-1 hemp shiv G7 1.00 1 H38-1-2 hemp shiv G7 1.00 2 H38-1-3 hemp shiv G7 1.00 3 H38-1-5 hemp shiv G7 1.00 5 H4-1.5-1 hemp shiv G7 1.50 1 H4-2-1 hemp shiv G7 2.00 1 HB12-1-2 hemp shiv G8 1.00 2 HC12-1-2 hemp shiv Bioﬁbat 1.00 2 HD12-1-2 hemp shiv Isoﬁn 1.00 2 Materials 2018, 11, 4 4 of 14

2.5. Characterisation The protocol used for characterising the different samples was as follows: Materials 2018, 11, 4 4 of 14 - Analysis of the particle size distributions (mean diameter (Z average) and polydispersity index 2.5.values Characterisation measured) of the silica suspensions; ◦ - Dip-coatingThe protocol of three used glassfor characterising slides for each the different formulation samples followed was as by follows: drying at 150 C for one hour; - Water contact angle (WCA) measurements on glass slides, to verify the hydrophobic character of - Analysis of the particle size distributions (mean diameter (Z average) and polydispersity index the dispersionvalues measured) (WCA of averaged the silica suspensions; from measurements of three to ten droplets of water on each of -the threeDip-coating glass slides);of three glass slides for each formulation followed by drying at 150 °C for one hour; - -Coating Water via contact immersion angle (WCA) of hemp measurements shiv and drying on glass at 90slides,◦C forto verify one hour; the hydrophobic character - Waterof the contact dispersion angle (WCA measurements averaged from on hemp measuremen shiv, andts of time three for towhich ten droplets the WCA of water remains on each greater thanof 90 the◦; three glass slides); - Coating via immersion of hemp shiv and drying at 90 °C for one hour; - Imaging of samples via SEM and EDX characterisation; and - Water contact angle measurements on hemp shiv, and time for which the WCA remains greater - Analysisthan 90°; ofthe environmental conditions on the treated hemp shiv: immersion in a water bath ◦ -for 24Imaging h/test of in samples humidity via chamberSEM and EDX at 35 characterisation;C/90% relative and humidity for 72 h, followed by drying -and Analysis WCA measurements. of the environmental conditions on the treated hemp shiv: immersion in a water bath Imagingfor 24 via h/test Scanning in humidity Electron chamber Microscopy at 35 °C/90% (SEM): relativeExamination humidity for of 72 the h, surfacefollowed of by coated drying hemp and WCA measurements. shiv was undertaken using SEM techniques. The samples were gold coated using sputtering methods prior to analysis.ImagingThe via Scanning images of Electron coated hempMicroscopy shiv were (SEM): produced Examination on a of Zeiss the surfaceΣigma of FEGSEM coated hemp (focussed electronshiv gunwas undertaken SEM) (Zeiss, using Germany) SEM techniques. operating The samples at 5 kV, were using gold acoated secondary using sputtering electron methods detector for prior to analysis. The images of coated hemp shiv were produced on a Zeiss Σigma FEGSEM topography analysis. (focussed electron gun SEM) (Zeiss, Germany) operating at 5 kV, using a secondary electron detector Energy-dispersive X-ray (EDX) spectroscopy: Elemental analysis of the coated surfaces was for topography analysis. undertakenEnergy-dispersive by obtaining EDX X-ray spectra (EDX) of spectroscopy: the selected Elemental areas of the analysis samples of th ande coated analysed surfaces via the was AZtec characterisationundertaken by system obtaining (Oxford EDX spectra Instruments, of the selected UK). areas of the samples and analysed via the AZtec Watercharacterisation contact system angle (WCA)(Oxford Instruments, measurement: UK). The hydrophobic character of the samples was assessed byWater measuring contact angle static (WCA) sessile dropmeasurement: characteristics. The hydrophobic In this study, character all static of the water samples contact was angle measurementsassessed by weremeasuring performed static sessile using drop a Drop characteristics. Shape Analyser In this study, DSA-100 all static (Krüss water GmbH, contact Hamburg,angle Germany).measurements For all materials,were performed between using three a Drop and tenShape 2 µ LAnalyser droplets DSA-100 of deionised (Krüss water GmbH, were Hamburg, placed onto the samples,Germany) images. For all of materials, the droplets between were three captured and ten allowing 2 μL droplets key of dimensions deionised water to be were measured. placed onto This data the samples, images of the droplets were captured allowing key dimensions to be measured. This was analysed using Krüss Advance software (Figure1). The average WCA was then reported, as well as data was analysed using Krüss Advance software (Figure 1). The average WCA was then reported, the standard deviation for each formulation of treatment applied on the hemp shiv. The measurements as well as the standard deviation for each formulation of treatment applied on the hemp shiv. ± ◦ wereThe performed measurements at room were temperature performed at (24 room1 temperatureC). (24 ± 1 °C).

Figure 1. DSA image captured via the Krüss Advance software, with the example of sample H4-1-5. FigureCourtesy 1. DSA of image TWI Ltd. captured via the Krüss Advance software, with the example of sample H4-1-5. Courtesy of TWI Ltd.

Materials 2018, 11, 4 5 of 14 Materials 2018, 11, 4 5 of 14

HumidityHumidity exposure: exposure: TheThe durability durability of of the the treatment treatment in in a a humid humid environment environment was was assessed assessed by by placingplacing loose loose coated coated hemp hemp shiv shiv in in a a humidity humidity ch chamberamber (Vötsch (Vötsch VC VC 0020, 0020, We Weississ Technik, Technik, Germany). Germany). ThereThere is is currently currently no no benchmark benchmark test test to to assess assess th thee durability durability of of a a hydrophobic hydrophobic coating coating deposited deposited on on bio-materialsbio-materials in in humid humid environment. environment. It It was was decided decided to to set set the the parameters parameters of of the the humidity humidity chamber chamber atat 35 35 °C◦C and and 90% 90% relative relative humidity humidity (RH) (RH) accord accordinging to to recommendations recommendations from from the the ISOBIO ISOBIO project. project. TheThe materials materials were were taken taken out out every every 24 24 h h to to measur measuree the the WCA WCA to to monitor monitor any any loss loss of of water water repellence repellence duedue to to the environment. TheThe testtest was was terminated terminated when when traces traces of of mould mould growth growth were were visible visible on at on least at leastone sample.one sample. The undamagedThe undamaged samples samples were were dried dried at room at room temperature temperature for 5 for days. 5 days.

3.3. Results Results

3.1.3.1. Assessment Assessment of of Functionalisation Functionalisation TheThe dispersion dispersion of of unfunctionalised unfunctionalised particles particles deposited deposited as as a asingle single layer layer on on glass glass slides slides gave gave a ◦ watera water contact contact angle angle of 21°, of demonstrating 21 , demonstrating a hydrophilic a hydrophilic character. character. All other All formulations other formulations of nPTMS- of ◦ functionalisednPTMS-functionalised silica had silica water had contact water angles contact of angles118.5 ± of 0.8°. 118.5 This± 0.8is presented. This is presentedgraphically graphically in Figure 2. in TheFigure quantity2. The of quantity silane ofused silane did usednot appear did not to appear have a to significant have a signiﬁcant influence inﬂuence on the repellence on the repellence to water to whenwater measured when measured on glass on slides. glass slides.

FigureFigure 2. 2. WaterWater contact contact angle angle measurements measurements (the (the standard standard deviation deviation is shown is shown as error as errorbars) on bars) glass on slides.glass slides.

3.2.3.2. Uniformity Uniformity of of Deposition Deposition NineNine samples samples of of hemp hemp shiv, shiv, without without and and with with su surfacerface treatment treatment with with silica silica particles particles (H0-0-0, (H0-0-0, H4-1-x,H4-1-x, and and H12-1-x) H12-1-x) were were examined examined using using SEM SEM imaging. imaging. Several Several samples samples of of hemp hemp shiv shiv coated coated with with thethe same same formulation formulation were were imaged, imaged, at at four four magnifications: magnifications: ×1000,×1000, ×5000,×5000, ×10,000,×10,000, and ×25,000. and ×25,000. This allowedThis allowed a representative a representative overview overview of how of un howiform uniform the deposition the deposition of each of coating each coating was. was. Only a partial coverage of the surface of the hemp shiv was achieved by single or two layers of Only a partial coverage of the surface of the hemp shiv was achieved by single or two layers of silica particles, which corresponds to samples H12-1-1 and H12-1-2 (Figure 3b,c). As further layers silica particles, which corresponds to samples H12-1-1 and H12-1-2 (Figure3b,c). As further layers were deposited, more complete coverage of the surface of the shiv was achieved, as shown in Figure 3d were deposited, more complete coverage of the surface of the shiv was achieved, as shown in Figure3d for the sample H12-1-3. At greater magnification for sample H12-1-2 (Figure 3c), it was possible to for the sample H12-1-3. At greater magnification for sample H12-1-2 (Figure3c), it was possible verify the size of the particles deposited. Particles with a maximum diameter of 152 nm and a to verify the size of the particles deposited. Particles with a maximum diameter of 152 nm and a minimum of 90 nm were measured from the image. This fits well with expectations and the assessment minimum of 90 nm were measured from the image. This fits well with expectations and the assessment of the particles size when in suspension, 120 nm [23]. of the particles size when in suspension, 120 nm [23]. The EDX spectra of two coated samples and of the untreated hemp shiv are represented in Figure 4. Even though the EDX characterisation method is not quantitative, it is still possible to differentiate

Materials 2018, 11, 4 6 of 14

The EDX spectra of two coated samples and of the untreated hemp shiv are represented in Figure4. Even though the EDX characterisation method is not quantitative, it is still possible to differentiate whichMaterials chemical 2018, 11 elements, 4 are in abundance on the surface of the hemp shiv. The ﬁrst spectrum6 of 14 shown in Figure4a belongs to untreated hemp shiv (sample H0-0-0–Figure3a) that presented a lot of carbon which chemical elements are in abundance on the surface of the hemp shiv. The first spectrum shown and oxygen, originating from the cellulose of its surface. The same two elements were still present in in Figure 4a belongs to untreated hemp shiv (sample H0-0-0–Figure 3a) that presented a lot of carbon the secondand oxygen, spectrum originating (Figure from4b), the with cellulose the addition of its surface. of silicon The same from tw theo elements silica particles. were still present As seen in from the SEMthe second image spectrum in Figure (Figure3b, the 4b), corresponding with the addition sample of silicon was from only the coated silica onceparticles. with As silica seen from particles, and thosethe SEM only image partially in Figure covered 3b, the the corresponding surface, which sample is why was elements only coated from once cellulose with silica could particles, still be seen. Expectedly,and those the only highest partially peak covered of the the third surface, spectrum which is (Figure why elements4c) of samplefrom cellulose H12-1-5 could shows still be far seen. greater levelsExpectedly, of silicon. the In thishighest sample, peak theof the signal third relatingspectrum to (Figure silicawas 4c) of the sample most H12-1-5 intense shows compared far greater to those of otherlevels elements. of silicon. This In sample this sample, wascoated the signal ﬁve relating times, to and silica SEM was image the most shows intense that compared the coverage to those by silica particlesof other of the elements. hemp shivThis surfacesample waswas almostcoated five complete times, and uniformSEM image for shows three (H12-1-3–Figurethat the coverage 3byd) and silica particles of the hemp shiv surface was almost complete and uniform for three (H12-1-3–Figure 3d) more layers of silica. and more layers of silica.

Figure 3. Cont.

Materials 2018, 11, 4 7 of 14

Figure 3. SEM (Scanning Electron Microscopy) images of samples: (a) H0-0-0 at ×10,000 magnification; Figure(b) 3. H12-1-1SEM (Scanning showing a Electronpartial coverage Microscopy) at ×10,000 images magnification; of samples: (c) (H12-1-2a) H0-0-0 at ×25,000 at ×10,000 magnification; magnification; (b) H12-1-1and (d) showing H12-1-3 ashowing partial coveragea complete at and×10,000 uniform magnification; coverage of (thec) H12-1-2hemp shiv at × surface25,000 at magnification; ×10,000 andmagnification. (d) H12-1-3 Courtesy showing of aTWI complete Ltd. and uniform coverage of the hemp shiv surface at ×10,000 magnification. Courtesy of TWI Ltd.

Materials 2018, 11, 4 8 of 14

Figure 4. EDX spectra of H0-0-0, H12-1-1, and H12-1-5: (a) sample H0-0-0; (b) H12-1-1, coated once Figurewith 4. EDX 120 nm spectra silica ofparticles; H0-0-0, and H12-1-1, (c) H12-1-5, and coated H12-1-5: five ( atimes.) sample The peaks H0-0-0; for (Aub) H12-1-1,are due to coated the gold once with 120coating nm silicaapplied particles; for SEM andimaging. (c) H12-1-5, Courtesy coated of TWI ﬁveLtd. times. The peaks for Au are due to the gold coating applied for SEM imaging. Courtesy of TWI Ltd.

Materials 2018, 11, 4 9 of 14

3.3. Water Repellence of Loose Coated Hemp Shiv Materials 2018, 11, 4 9 of 14 The static WCA was measured for samples H4-1-1, H4-1.5-1, H4-2-1, H12-1-1, H38-1-1, and H0-0-0. The3.3. influence Water Repellence of the hydrophobic of Loose Coated coatingHemp Shiv was successfully demonstrated by comparing untreated pieces ofThe hemp static shiv WCA (H0-0-0) was measured with different for samples formulations H4-1-1, H4-1.5-1, of hydrophobic H4-2-1, H12-1-1, silica particles H38-1-1, applied and on hempH0-0-0. shiv. The Particles influence of 45 of nm the diameterhydrophobic were coatin usedg inwas samples successfully H4-x-1, demonstrated with an increasing by comparing amount of alkoxysilane.untreated pieces Doubling of hemp the quantityshiv (H0-0-0) of nPTMS with different did not formulations have a significant of hydrophobic effect on thesilica WCA, particles as can be seenapplied in Figure on 5hemp, which shiv. supports Particles theof 45 earlier nm diameter findings were on glassused in slides. samples Sample H4-x-1, H12-1-1 with an with increasing particles of 120amount nm diameter of alkoxysilane. also showed Doubling a WCA the similarquantity to of the nPTMS other did samples, not have approximately a significant effect 114◦ on the average WCA, as can be seen in Figure 5, which supports the earlier findings on glass slides. Sample H12-1-1 across the four treated samples. The WCA for sample H38-1-1 was just over 90◦, indicating a less with particles of 120 nm diameter also showed a WCA similar to the other samples, approximately hydrophobic114° on average nature across than thethe H4-x-1four treated and H12-1-1samples. samples.The WCA The for particlessample H38-1-1 used for was this just formulation over 90°, had a meanindicating diameter a less of hydrophobic 380 nm, and nature were than applied the H4-x-1 as a single and H12-1-1 layer. samples. The particles used for this formulationThe six samples had a mean were diameter immersed of 380 in anm, bath and of were deionised applied water as a single for 24 layer. h to verify the durability of ◦ the coating.The six Upon samples drying were the immersed samples in at a 65bathC of for deionis 10 min,ed water the WCA for 24 was h to measuredverify the durability again (Figure of 5). As expected,the coating. the Upon untreated dryingsample the samples showed at 65 a °C significant for 10 min, reduction the WCA in was the measured measured again WCA, (Figure which 5). would suggestAs expected, that water the was untreated still trapped sample in showed the structure a sign ofificant the hempreduction shiv in even the aftermeasured drying. WCA, Sample which H38-1-1 withwould 380 nm suggest particles that water also demonstratedwas still trapped a in significant the structure reduction of the hemp in theshiv measured even after drying. WCA comparedSample to theH38-1-1 value prior with to 380 immersion. nm particles The also four demonstrated samples with a particlessignificant of reduction 45 and 120 in nmthe diametermeasured maintainedWCA theircompared hydrophobic to the value behaviour prior to after immersion. immersion The four in water. samples The with smaller particles silica of 45 particles and 120 thusnm diameter provided an maintained their hydrophobic behaviour after immersion in water. The smaller silica particles thus effective barrier to liquid water within and on the hemp shiv. provided an effective barrier to liquid water within and on the hemp shiv.

Figure 5. Water contact angle measurements on hemp shiv before (dark shade) and after immersion Figure 5. Water contact angle measurements on hemp shiv before (dark shade) and after immersion in in a water bath for 24 h (lighter shade). The error bars represent the standard deviation of measurements a water bath for 24 h (lighter shade). The error bars represent the standard deviation of measurements for each type of sample (several hemp shiv were analysed). for each type of sample (several hemp shiv were analysed). 3.4. Increasing the Silica Content and Hydrophobic Period 3.4. Increasing the Silica Content and Hydrophobic Period As was clear from the SEM images (Figure 3), each piece of hemp shiv should be coated more thanAs once was to clear ensure from more the complete SEM images coverage (Figure of the3 ),surface each by piece silica. of It hemp was also shiv proven should that be increasing coated more thanthe once quantity to ensure of silica more coated complete increased coverage the WCA of the of surface more than by silica.10° (from It was 115.5° also for proven H12-1-1, that coated increasing theonce, quantity to 125.9° of silica for H12-1-5 coated coated increased five times) the WCA (Figure of 6 moreand Table than 2). 10 Having◦ (from more 115.5 silica◦ for on H12-1-1, the surface coated once,provides to 125.9 more◦ for barrier H12-1-5 effect, coated thus ﬁve water times) is less (Figure likely 6to and be absorbed Table2). by Having the hemp more shiv. silica on the surface provides more barrier effect, thus water is less likely to be absorbed by the hemp shiv.

Materials 2018, 11, 4 10 of 14 Materials 2018, 11, 4 10 of 14

FigureFigure 6. 6.WCA WCA and and time time measurements measurements forfor thethe series of samples samples H4-1-x, H4-1-x, H12-1-x, H12-1-x, H38-1-x, H38-1-x, and and H0-0-0, H0-0-0, showingshowing the the beneﬁcial beneficial effect effect of of the the quantity quantity of of silica silica applied applied onto onto the the hemp hemp shiv shiv on on both both the the WCA, WCA, and theand period the period of sustained of sustained water water repellence. repellence.

Table 2. Summary of results, after Figure 6. Table 2. Summary of results, after Figure6. H4-1-x H12-1-x H38-1-x H0-0-0 WCA (°) H4-1-x 114.4 H12-1-x 115.5 H38-1-x 93.1 H0-0-0 52.0 coated × 1 timeWCA to (◦ )90° (min) 114.4 13.0 115.5 9.5 93.1 7.0 52.0 0.1 coated × 1 ◦ time to 90WCA(min) (°) 13.0 113.0 9.5117.9 7.0121.6 0.1 coated × 2 timeWCA to (◦ )90° (min) 113.0 15.5 117.9 14.7 121.6 23.0 coated × 2 ◦ time to 90WCA(min) (°) 15.5 115.4 14.7116.5 23.0122.8 coated × 3 timeWCA to (◦ )90° (min) 115.4 18.5 116.5 18.5 122.8 21.5 coated × 3 ◦ time to 90WCA(min) (°) 18.5 124.2 18.5125.9 21.5128.0 coated × 5 WCA (◦) 124.2 125.9 128.0 coated × 5 time to 90° (min) 22.7 22.5 24.5 time to 90◦ (min) 22.7 22.5 24.5 Cellulose is naturally hydrophilic, which means that, if the hemp shiv are left untreated, liquid waterCellulose penetrates is naturally quickly in hydrophilic, the structure which of the means bio-aggregates. that, if the When hemp a droplet shiv are of left water untreated, is deposited liquid wateron the penetrates surface quicklyof a piece in of the hemp structure shiv, of it theis abso bio-aggregates.rbed in three When to five a dropletseconds of[29]. water However, is deposited when on theanalysing surface of coated a piece hemp of hemp shiv, shiv, hydrophobicity it is absorbed was in threeachieved to five as the seconds droplet [29 remained]. However, on whenthe surface analysing of coatedthe shiv hemp with shiv, a contact hydrophobicity angle greater was achievedthan 90° asfor the considerably droplet remained longer. onThe the time surface for which of the a shiv single with a contactdroplet angle maintained greater a than contact 90◦ anglefor considerably greater than longer. 90° with The the time substrate for which was a monitored single droplet for the maintained series of samples H4-1-x, H12-1-x, and H38-1-x. The results are shown in Figure 6. There is a clear a contact angle greater than 90◦ with the substrate was monitored for the series of samples H4-1-x, correlation between the number of deposited layers on the hemp shiv and the retention of H12-1-x, and H38-1-x. The results are shown in Figure6. There is a clear correlation between the hydrophobic behaviour, which went up to nearly 25 min for sample H38-1-5, and above 20 min for 5 number of deposited layers on the hemp shiv and the retention of hydrophobic behaviour, which went samples. This method has its flaws as some of the water can evaporate during the experiment, but up to nearly 25 min for sample H38-1-5, and above 20 min for 5 samples. This method has its flaws as overall it can be a good first assessment of prolonged hydrophobic behaviour of a surface that usually someabsorbs of the water. water This can enhanced evaporate resistance during the to experiment, water absorption but overall is due it to can the be barrier a good effect first of assessment the silica. of prolongedIndeed, the hydrophobic samples coated behaviour five time of as surface (H4-1-5, that H12-1-5, usually an absorbsd H38-1-5) water. were This produced enhanced so resistancethat more to waterinorganic absorption material is due was to deposited the barrier between effect of the the hemp silica. shiv Indeed, and thethe sampleswater, therefore coated five preventing times (H4-1-5, the H12-1-5,attraction and of H38-1-5) liquid water were by produced the hydroxyl so thatgroups more of the inorganic cellulose. material was deposited between the hemp shiv and the water, therefore preventing the attraction of liquid water by the hydroxyl groups of the cellulose.

Materials 2018, 11, 4 11 of 14

Materials 2018, 11, 4 11 of 14 3.5. Humidity Tests 3.5. Humidity Tests The bio-materials analysed in this study are aimed to be used as insulation as the core of The bio-materials analysed in this study are aimed to be used as insulation as the core of panels panelsin walls. in walls.The reason The reasonfor this foris the this ability is the of ability the hemp of the shiv hemp to shivbuffer to moisture buffer moisture thanks to thanks their to theirmicroporosity microporosity [3]. This [3]. breathable This breathable material material then allows then water allows vapour water to vapour permeate, to permeate, and, when and, coated when coatedwith silica with particles, silica particles, prevents prevents absorption absorption of liquid of water. liquid When water. vapour When con vapourdensates condensates within the within panel, the panel,it is important it is important that the that bio-materials the bio-materials continue continue to repel to liquid repel liquidwater. water. Moreover, Moreover, bio-materials bio-materials are aresubject subject to natural to natural biological biological decay, decay, which which is usua is usuallylly triggered triggered by bya humid a humid and and warm warm environment. environment. ToTo assess assess the the effecteffect ofof humidityhumidity on the treated treated bio-aggregates, bio-aggregates, samples samples H0-0-0, H0-0-0, HB12-1-2, HB12-1-2, HC12-1-2, HC12-1-2, ◦ andand HD12-1-2 HD12-1-2 werewere placedplaced in a humidity chamber chamber at at 35 35 °CC and and 90% 90% of ofrelative relative humidity. humidity. The The three three treatedtreated samplessamples werewere coatedcoated with the the same same formulation formulation of of hydrophobic hydrophobic silica silica particles, particles, but but were were depositeddeposited on on three three different different grades grades of of hemp hemp shiv. shiv. As As the the performance performance of of each each type type of of coating coating has has been assessedbeen assessed in the in previous the previous sections, sections, verifying verifying the waterthe water repellence repellence on on bigger bigger pieces pieces of of hemp hemp shiv shiv can givecan insightfulgive insightful information information on the on inﬂuencethe influence (or lack(or lack thereof) thereof) of the of the substrate. substrate. The The WCA WCA of theof the loose materialsloose materials was measured was measured at regular at intervals. regular interval The changes. The in repellencechange in ofrepellence the samples of throughoutthe samples this experimentthroughout is this shown experiment in Figure is shown7. Untreated in Figure hemp 7. Untreated shiv (sample hemp H0-0-0) shiv (sample became H0-0-0) more hydrophilicbecame more the longerhydrophilic they were the longer exposed they to were humidity, exposed with to thehumi WCAdity, decreasingwith the WCA from decreasing 68◦ to 39◦ fromin 48 68° h of to exposure. 39° in The48 h absence of exposure. of data The for absence 72 h of of exposure data for 72 is dueh of ex toposure the development is due to the of development fungal growths of fungal which growths led to the samplewhich beingled to discarded.the sample Thisbeing period discarded. of exposure This period marked of exposure the termination marked ofthe the termination test. On the of otherthe test. hand, coatedOn the hemp other shivhand, consistently coated hemp maintained shiv consistently a WCA maintained of 115◦ on a averageWCA of across115° on the average different across samples, the alldifferent throughout samples, the threeall throughout days of humidity the three exposure. days of humidity Upon termination exposure. ofUpon the experimenttermination andof the after experiment and after drying the samples, the final WCA remained consistent with the values drying the samples, the ﬁnal WCA remained consistent with the values obtained during the humidity obtained during the humidity exposure, with no sign of microbial growth. The presence of silica exposure, with no sign of microbial growth. The presence of silica particles on hemp shiv again proved particles on hemp shiv again proved to act as a barrier to liquid water, but also to prevent mould to act as a barrier to liquid water, but also to prevent mould from developing. from developing.

Figure 7. Humidity chamber test (35 °C and 90% RH). The measurements were executed every 24 h ◦ Figureuntil signs 7. Humidity of mould chambergrowth were test visible (35 C on and at 90%least RH).one sample, The measurements and after drying were at room executed temperature every 24 h untilfor five signs days. of mould Only very growth little were variation visible of on WCA at least during one sample,the experiment and after for drying the coated at room samples temperature was forshown, ﬁve days.and mould Only was very prevented little variation from developing. of WCA during the experiment for the coated samples was shown, and mould was prevented from developing.

Materials 2018, 11, 4 12 of 14

4. Discussion Materials 2018, 11, 4 12 of 14 Due to their microporosity and cellulosic content, hemp shiv are hydrophilic and have a very high potential4. Discussion of water uptake, as they can absorb up to three times their own dry weight [7]. Coating the hemp shiv with hydrophobic silica particles is potentially beneficial in two ways. The coating may act Due to their microporosity and cellulosic content, hemp shiv are hydrophilic and have a very as a barrier to liquid water, so droplets would just sit on the surface of the hemp shiv; however, if the high potential of water uptake, as they can absorb up to three times their own dry weight [7]. Coating silicathe particles hemp shiv are with small hydrophobic enough, they silica would particles not blockis potentially the primary beneficial pores in oftwo the ways. hemp The shiv coating and water vapourmay could act as can a barrier still permeate to liquid allowingwater, so thedroplets hemp would shiv tojust retain sit on its the buffer surface properties. of the hemp shiv; however,We demonstrated if the silica particles in this studyare small that enough, the more they functionalisedwould not block silicathe primary was present pores of onthe thehemp surface, the greatershiv and thewater water vapour contact could anglecan still was. permeate Imaging allowing with the SEM hemp techniques shiv to retain demonstrated its buffer properties. that several layers ofWe coating demonstrated led to a more in this complete study that and the uniform more func coveragetionalised of silica the hemp was present shiv. The on additionalthe surface, layers alsothe increased greater the water time itcontact took for angle a droplet was. Imaging of water with to beSEM completely techniques absorbed,demonstrated from that five several seconds to beyondlayers 25 of min. coating This led demonstrated to a more complete that functionalisedand uniform coverage silica of particles the hemp could shiv. beThe used additional as a hydrophobic layers also increased the time it took for a droplet of water to be completely absorbed, from five seconds to treatment for bio-materials, as they provided an effective barrier to water (as illustrated in Figure8). beyond 25 min. This demonstrated that functionalised silica particles could be used as a hydrophobic treatmentThe presence for bio-materials, of inorganic as materialthey provided between an effe thective shiv barrier and theto water environment (as illustrated protected in Figure it from8). the effects biodegradation,The presence of inorganic as well asmaterial reduced between the potential the shiv and attraction the environment of water protected by the hydroxyl it from the groups of theeffects cellulose. biodegradation, The bond as between well as reduced cellulose the andpotent silicaial attraction was strong of water enough by the to hydroxyl withstand groups exposure of to humidthe conditionscellulose. The and bond immersion between in ce water,llulose whenand silica small was particles strong (45enough or 120 to nmwithstand diameter) exposure were to applied on thehumid hemp conditions shiv. and immersion in water, when small particles (45 or 120 nm diameter) were appliedThe silica on the particles hemp shiv. also brought microroughness to the surface of the shiv. This characteristic was demonstratedThe silica particles by other also researchers brought microroughness [20] by the means to the ofsurface a 3D of optical the shiv. profilometer. This characteristic Using such characterisationwas demonstrated technique by other could researchers be valuable [20] for by quantifyingthe means of howa 3D much optical silica profilometer. was effectively Using such deposited characterisation technique could be valuable for quantifying how much silica was effectively onto the bio-aggregates after being coated multiple times. deposited onto the bio-aggregates after being coated multiple times.

Figure 8. Coloured water droplets onto hemp shiv H12-1-5, with a WCA > 125°. Courtesy of TWI Ltd. Figure 8. Coloured water droplets onto hemp shiv H12-1-5, with a WCA > 125◦. Courtesy of TWI Ltd. 5. Conclusions 5. ConclusionsIn this study, a hydrophobic treatment was applied onto hemp shiv that are starting to be used in insulation for buildings. When functionalised silica particles were coated several times onto the In this study, a hydrophobic treatment was applied onto hemp shiv that are starting to be used in hemp shiv, a uniform and complete coverage of the surface was achieved. We successfully insulationdemonstrated for buildings. that such When treatment functionalised provided silica hemp particles shiv with were extensive coated several water timesrepellence: onto thethe hemp shiv,hydrophilic a uniform character and complete of the coverageuntreated ofhemp the shiv surface was wasmodified achieved. to durably We successfully hydrophobic demonstratedonce the thatbio-materials such treatment were provided coated with hemp functionalised shiv with extensive silica particles. water repellence:Mould growth the hydrophilicwas delayed characterwhen of the untreatedexposed to hemphumidity, shiv whilst was modiﬁed the liquid to water durably repe hydrophobicllent property onceof the the treated bio-materials hemp shiv were was coated withmaintained functionalised despite silica the particles.humid conditions. Mould growthThe treatm wasent delayed developed when in exposedthis study to could humidity, be a viable whilst the liquidsolution water for repellent use on bio-materials property of thethat treatedneed to hemprepel liquid shiv waswater, maintained whilst preserving despite the the integrity humid of conditions. the Theinsulation treatment panel developed in common in this environmental study could beconditions. a viable There solution are forof course use on still bio-materials further tests that to be need to carried out before adopting this treatment in the construction sector, such as additional moisture repel liquid water, whilst preserving the integrity of the insulation panel in common environmental tests, mechanical tests, biodegradation tests, etc. However, the promising results described here give conditions. There are of course still further tests to be carried out before adopting this treatment in a good first assessment of feasibility. Before being fully adopted in the construction sector, such the construction sector, such as additional moisture tests, mechanical tests, biodegradation tests, etc.

Materials 2018, 11, 4 13 of 14

However, the promising results described here give a good ﬁrst assessment of feasibility. Before being fully adopted in the construction sector, such treatment and materials will have to be fully tested according to building codes. Considering the only recent use of bio-materials, new standards and test methods should be developed for those speciﬁc materials.

Acknowledgments: The author would like to acknowledge colleagues from NSIRC and TWI Ltd., particularly the FCR team at TWI Ltd., Nadia Sid for coordinating the ISOBIO project, and Sheila Stevens for her help with the SEM and EDX. Colleagues from the School of Engineering at the University of Edinburgh are thanked for their support and interest, especially Luke A. Bisby and Rory M. Hadden for the supervision. The authors would especially like to acknowledge the ANNIC 2017 Scientific Committee for permitting this paper to be published following the 3rd edition of the Applied Nanotechnology and Nanoscience International Conference. The work presented here was carried out under the ISOBIO project, with funding from the European Union’s H2020 research and innovation programme, under grant agreement No. 636835. Author Contributions: Marion A. Bourebrab is the main author of most of this work. Alan Taylor and Géraldine G. Durand have contributed as her advisors, as well as by proposing the subject. Marion A. Bourebrab and Alan Taylor conceived and designed the experiments. Marion A. Bourebrab, Alan Taylor and Géraldine G. Durand analysed the data. Marion A. Bourebrab and Alan Taylor wrote the article, with the contribution of Géraldine G. Durand. Conflicts of Interest: The authors declare no conflict of interest. The founding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results. The ISOBIO project aims to develop and bring new bio-based insulation panels and renders into the mainstream for the purpose of creating more energy efficient buildings. The contents of this publication are the sole responsibility of the authors and can in no way be taken to reflect the views of the European Union.

References

1. Pacheco-Torgal, F.; Jalali, S. Cementitious building materials reinforced with vegetable fibres: A review. Constr. Build. Mater. 2011, 25, 575–581. [CrossRef] 2. Khazma, M.; Goullieux, A.; Dheilly, R.M.; Queneudec, M. Coating of a lignocellulosic aggregate with pectin/polyethylenimin mixtures: Effects on flax shive and cement-shive composite properties. Cem. Concr. Compos. 2012, 34, 223–230. [CrossRef] 3. Tran Le, A.D.; Maalouf, C.; Mai, T.H.; Wurtz, E.; Collet, F. Transient hygrothermal behaviour of a hemp concrete building envelope. Energy Build. 2010, 42, 1797–1806. [CrossRef] 4. Magniont, C.; Escadeillas, G.; Coutand, M.; Oms-Multon, C. Use of plant aggregates in building ecomaterials. Eur. J. Environ. Civ. Eng. 2012, 16, s17–s33. [CrossRef] 5. Ebskamp, M.J.M. Engineering flax and hemp for an alternative to cotton. Trends Biotechnol. 2002, 20, 229–230. [CrossRef] 6. Glé, P.; Gourdon, E.; Arnaud, L. Acoustical properties of materials made of vegetable particles with several scales of porosity. Appl. Acoust. 2011, 72, 249–259. [CrossRef] 7. Arnaud, L.; Gourlay, E. Experimental study of parameters influencing mechanical properties of hemp concretes. Constr. Build. Mater. 2012, 28, 50–56. [CrossRef] 8. Kymäläinen, H.R.; Sjöberg, A.M. Flax and hemp fibres as raw materials for thermal insulations. Build. Environ. 2008, 43, 1261–1269. [CrossRef] 9. Lenormand, H.; Mahieu, A.; Leblanc, N.; Vivet, A. Nouvelles Agroressources Pour Panneaux de Particules 100% Biosourcés. Colloque 01 Ecomateriaux. 2014. Available online: https://hal.archives-ouvertes.fr/hal- 01144554/document (accessed on 20 December 2017). 10. Korjenic, A.; Petránek, V.; Zach, J.; Hroudová, J. Development and performance evaluation of natural thermal-insulation materials composed of renewable resources. Energy Build. 2011, 43, 2518–2523. [CrossRef] 11. Nykter, M. Microbial Quality of Hemp (Cannabis sativa L.) and Flax (Linum usitatissimum L.) from Plants to Thermal Insulation; University of Helsinki: Helsinki, Finland, 2006. 12. Evrard, A.; De Herde, A. Hygrothermal Performance of Lime-Hemp Wall Assemblies. J. Build. Phys. 2010, 34, 5–25. [CrossRef] 13. Piot, A.; Béjat, T.; Jay, A.; Bessette, L.; Wurtz, E.; Barnes-Davin, L. Study of a hempcrete wall exposed to outdoor climate: Effects of the coating. Constr. Build. Mater. 2017.[CrossRef] Materials 2018, 11, 4 14 of 14

14. Mostefai, N.; Hamzaoui, R.; Guessasma, S.; Aw, A.; Nouri, H. Microstructure and mechanical performance of modified hemp fibre and shiv mortars: Discovering the optimal formulation. Mater. Des. 2015, 84, 359–371. [CrossRef] 15. Shea, A.; Lawrence, M.; Walker, P. Hygrothermal performance of an experimental hemp–lime building. Constr. Build. Mater. 2012, 36, 270–275. [CrossRef] 16. Benfratello, S.; Capitano, C.; Peri, G.; Rizzo, G.; Scaccianoce, G.; Sorrentino, G. Thermal and structural properties of a hemp-lime biocomposite. Constr. Build. Mater. 2013, 48, 745–754. [CrossRef] 17. Florentin, Y.; Pearlmutter, D.; Givoni, B.; Gal, E. A life-cycle energy and carbon analysis of hemp-lime bio-composite building materials. Energy Build. 2017, 156, 293–305. [CrossRef] 18. Hirst, E.A.J. Characterisation of Hemp-Lime as a Composite Building Material. Ph.D. Thesis, University of Bath, Bath, UK, 2013. 19. Vignon, M.R.; Garcia-Jaldon, C.; Dupeyre, D. Steam explosion of woody hemp chenevotte. Int. J. Biol. Macromol. 1995, 17, 395–404. [CrossRef] 20. Hussain, A.; Calabria-Holley, J.; Schorr, D.; Jiang, Y.; Lawrence, M.; Blanchet, P. Hydrophobicity of hemp shiv treated with sol-gel coatings. Appl. Surf. Sci. 2018, 434, 850–860. [CrossRef] 21. Taylor, A.; Durand, G.G.; Alvarez Tirado, M.; Sid, N.; Mycock, S. Functionalisation Method for Metal Oxide Particles. U.S. Patent WO2017/093759A1, 8 June 2017. 22. Stöber, W.; Fink, A.; Bohn, E. Controlled growth of monodisperse silica spheres in the micron size range. J. Colloid Interface Sci. 1968, 26, 62–69. [CrossRef] 23. Bourebrab, M.A.; Sid, N.; La Rosa, A.; Durand, G.G.; Taylor, A. Identification of a threshold of water content during Stöber silica formation–influence on particle characteristics. 2017, Manuscript in preparation. 24. Zhuravlev, L.T. Surface characterization of amorphous silica—A review of work from the former USSR. Colloids Surf. A Physicochem. Eng. Asp. 1993, 74, 71–90. [CrossRef] 25. Lin, J.; Siddiqui, J.A.; Ottenbrite, R.M. Surface Modification of Inorganic Oxide Particles with Silane Coupling Agent and Organic Dyes. Polym. Adv. Technol. 2001, 292, 285–292. [CrossRef] 26. Carrot, G.; Rutot-Houzé, D.; Pottier, A.; Degée, P.; Hilborn, J.; Dubois, P. Surface-initiated ring-opening polymerization: A versatile method for nanoparticle ordering. Macromolecules 2002, 35, 8400–8404. [CrossRef] 27. Liu, Y.; Hsu, C.; Wang, M. A novel approach of chemical functionalization on nano-scaled silica. Nanotechnology 2003, 14, 813–819. [CrossRef] 28. Posthumus, W.; Magusin, P.C.M.M.; Brokken-Zijp, J.C.M.; Tinnemans, A.H.A.; Van der Linde, R. Surface modification of oxidic nanoparticles using 3-methacryloxypropyltrimethoxysilane. J. Colloid Interface Sci. 2004, 269, 109–116. [CrossRef][PubMed] 29. Tavisto, M.; Kuisma, R.; Pasila, A.; Hautala, M. Wetting and wicking of fibre plant straw fractions. Ind. Crops Prod. 2003, 18, 25–35. [CrossRef]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).