lable at ScienceDirect

Food Control 73 (2017) 91e100

Contents lists avai

Food Control

journal homepage: www.elsevier .com/locate/ foodcont

Evaluation of polyurethane foam materials as air filters against fungalcontamination

Mark Briffa a, b, c, Stephen Decelis c, Jean-Pierre Brincat d, Joseph N. Grima d, e,Ruben Gatt d, Vasilis Valdramidis a, *

a Food Studies and Environmental Health, Faculty of Health Sciences, University of Malta, Msida, Maltab Applied and Biomedical Science, Faculty of Health Sciences, University of Malta, Msida, Maltac Mycology Laboratory, Mater Dei Hospital, Msida, Maltad Metamaterials Unit, Faculty of Science, University of Malta, Msida, Maltae Department of Chemistry, Faculty of Science, University of Malta, Msida, Malta

a r t i c l e i n f o

Article history:Received 20 February 2016Received in revised form26 April 2016Accepted 19 May 2016Available online 20 May 2016

Keywords:FiltrationFoamsStorageFungiEnvironmental airSynthetic polymer

* Corresponding author.E-mail address: vasilis.valdramidis@um.edu.mt (V

http://dx.doi.org/10.1016/j.foodcont.2016.05.0280956-7135/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

Current air filters used in food processing or storage facilities are expensive and disposable. The ability touse polyurethane foam as air filters against fungal spores would be beneficial as they are both cheap andre-usable. The aim of this study was to evaluate the air filtration capabilities, in terms of fungal spores, ofa selection of polyurethane foam(s) of differing combinations of pores per inch (PPI) (50 and 90 PPI) andthickness (15 and 20 mm). Environmental air was used as a source of fungal spores and membranefiltration was used to assess the filtration capabilities of the foams. Spores capable of passing through thefoams were captured on cellulose nitrate membrane filters and quantified in CFU counts. Apart from the50 PPI foam of 15 mm thickness, all the foam samples were effective at significantly reducing the numberof spores. The PPI was found to be 2 times more influential on the efficiency of the foam material thanthe foam thickness. This may be explained through the higher number of pores present and the decreasein thickness of the ribs composing the microstructure of the foam as shown through scanning electronmicroscope (SEM) micrographs. These studies show that reticulated polyurethane foams at the selectedPPI and thickness can be used as effective air filters against fungal spores.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Airborne particles of biological origins, known as bioaerosols,may include, amongst other living material, fungi and fungal en-tities (Yoon et al., 2008). Highly-resistant fungal spores are found inall environments, including air inside buildings and facilities. Thischaracteristic of fungal spores would not be an issue were it not forthe fact that fungi are capable of causing seriously adverse effects;they can damage buildings (May, 2001; Miller, Rand, & Jarvis,2003), they can cause spoilage of certain foodstuffs in food stor-age facilities and they may cause allergic diseases of the airways inindividuals who inhale the spores (Portnoy, Barnes, & Kennedy,2004; _Zukiewicz-Sobczak et al., 2013). There are several ways bywhich bioaerosols can be controlled. These include ultravioletgermicidal irradiation (UVGI), air ionisation, dielectric barrier

. Valdramidis).

discharge and others (Griffiths, Bennett, Speight, & Parks, 2005;Jankowska, Reponen, Willeke, Grinshpun, & Choi,.2005; Ko, First,& Burge, 2002; Park, Yoon, Kim, Byeon, & Hwang, 2009; Park &Hwang, 2013; Schmid et al., 2013; Woo et al., 2012). The mostcommonly used means of eliminating bioaerosols from the internalenvironment of facilities is by air filtration (Brincat et al., 2016).

Air filtration systems are important for different types of pro-tected environments, such asmedical and food processing facilities,and less commonly in food storage environments. The most com-mon air filtration systems to reduce the abundance of fungal sporesin protected environments (especially in medical wards) are HighEfficiency Particulate Air (HEPA) filters. HEPA filters have beenshown to reduce the incidence of invasive aspergillosis (AbdulSalam, Karlin, Ling, & Yang, 2010; Alberti et al., 2001; Araujoet al., 2008; Vonberg & Gatsmeier, 2006). Studies have shownthat fungal spores are much more abundant in environments thatlack an air filtration system (50e500 CFU per m�3) than in thosewhich have functioning and effective air filtration systems(0e50 CFU per m�3) (Brenier-Pinchart et al., 2009; Dassonville

M. Briffa et al. / Food Control 73 (2017) 91e10092

et al., 2008; Sautour et al., 2009). In order for this to be true, airfiltration systems must be appropriately maintained and replacedwhen necessary so as to ensure the utmost quality of filtration. Infact, it is not uncommon for outbreaks (most commonly involvingAspergillus fumigatus) to occur due to the negligence of faults andcontaminated air filtration systems (Lutz, Jin, Rinaldi, Wickes, &Huycke, 2003; Mu~noz et al., 2004; Reboux et al., 2014).

The importance of optimally functioning air filtration systems isnot restricted to medical environments. The quality and quantity ofair in food manufacturing and storage facilities is regulated so as toprevent or significantly reduce the growth of fungal organismswhich would otherwise cause severe spoilage of food stuffs,particularly grains, flours, nuts, ripening cheeses and fruits (Isaac,1996). The primary concern of fungal contamination of food stuffsis the generation of mycotoxins, which are toxins generated bytoxigenic fungal organisms within certain temperature and relativehumidity ranges. Fungal species commonly of interest in terms offood contamination include those belonging to the Aspergillus,Alternaria, Claviceps, Fusarium, Penicillium and Stachybotys genera(Mili�cevi�c, �Skrinjar, & Balti�c, 2010). The durability and heat resis-tance of these toxins makes them particularly difficult to eliminateas they are not destroyed by common cooking procedures such asfrying and boiling (Magan & Olsen, 2004; Mili�cevi�c et al., 2010).When consumed, food stuffs contaminated with mycotoxins cancause a variety of acute and chronic health disorders in bothhumans and animals alike (Mili�cevi�c et al., 2010). As a result,considerable attempts are made to prevent the contamination offood stuffs by fungal organisms (Magan & Olsen, 2004).

The regulation of air is most commonly brought about byfiltration, which eliminates potentially harmful micro-organismsfrom the air that is being introduced into the facility by capturingand retaining them in the filter medium (EHEDG, 2006). Fungalspores may enter food production and storage facilities via drains,doorways, hatches, disinfection tunnels, compressed air supplies,raw materials, packaging, people handling and poorly designed,cleaned and maintained air filtration systems, amongst others. Thelatter source of airborne contamination is of particular interest.Functioning air filtration systems are designed to distribute fullyconditioned air as required at the desired rate. The air handlingsystems must be able to not only remove contaminating micro-organisms from the air by filtration, but also to reduce or preventgrowth of micro-organisms, prevent their ingress, curtail cross-contamination, direct bioaerosols away from the food stuffs andnot act as an additional source of contamination (EHEDG, 2006).Systemsmust be designed in such away to allowmaintenance to beperformed on the filter media. Lack of proper maintenance mightlead to the filter itself becoming a source of spores in the down-stream air, which is known to be a huge health risk in hospitals(Price, Simmons, Crow, & Ahearn, 2005).

Polyurethane is a synthetic polymer which forms from the re-action of a polyol (a long chained alcohol) with a diisocyanate.Depending on the type of polyol used, different types of poly-urethane can be formed. The reticulated type of polyurethane foamis a porous, low-density type of foam characterised by a three-dimensional skeletal structure lacking membranes between thestrands (Polyurethane Foam Association, 1994). The porosity of thefoam is typically 95% but it can be as high as 98% (Gliganic, 2008).Currently reticulated foam is used in water filters and air filtersfound in air-conditioning units. No previous studies haveadequately assessed the true potential of reticulated polyurethanefoam as an air filter against fungal spores.

In light of the above, the aim of this study was to assess the airfiltration capabilities in terms of fungal spores of a selection ofpolyurethane foams of differing combinations of pores per inch(PPI) and thickness so as to identify the foam PPI and thickness

combination that is most effective for filtering fungal spores in airenvironments of different temperature and relative humidity.

2. Materials and methods

2.1. Sample collection

Outside air at the area of Msida in Malta (site of the Faculty ofHealth Sciences, University of Malta) was selected as the source offungal spores. A filtration manifold with 3 ports was attached to anair flow meter which was in turn attached to a vacuum pump(Sartorius AG, Goettingen, Germany). The vacuum pump was sub-sequently connected to a power supply. The components werelinked together by means of rubber tubing (See Fig. 1). Sterile cel-lulose nitrate membrane filters of pore size 0.45 mmwere placed onthe base of each manifold port (Sartorius AG, Goettingen, Ger-many). The ends of the tubing and the entirety of the setup weresealed by means of multiple sheets of parafilm M sealing film(Bemis NA, Neenah, Wisconsin) so as to ensure that the setup wasthoroughly air-tight. A lack of air-leaks was confirmed by closingeach of the valves of the 3 manifold ports, switching on the vacuumpump and noting an absence of air flow in the flow meter.

Three acrylonitrile butadiene styrene (ABS) plastic adapters,generated by a 3D printer, were designed in a manner that allowedfor them to be inserted into the manifold ports while providing aplatform into which any foams under assessment could be placed.The dimensions of the platforms were 45 mm � 45 mm � 20 mm.The adapters were sealed to the 3 manifold ports by means offurther sheets of parafilm M sealing film.

Membrane filters were put into place by means of sterilisedforceps and the adapters were duly sealed to the body of themanifold using sheets of parafilm M. The sampling of air occurredas the vacuum pump was switched on, air was drawn throughfoams supported by the adapters in the manifold ports, and sporeswere collected on the membrane filters. The necessary volume ofair was sampled by running the vacuum pump for a period of timewhile the flow rate was measured by an air flow meter. Uponcompleting the sampling, the membrane filters were placed onto aseparate Dichloran Rose Bengal Agar (DRBA) (Sigma Aldrich, St.Louis, USA) with Chloramphenicol plate. The inoculated plateswere incubated in a cooling incubator (Leec P3C, Leec Limited, UK)at 25 �C for 72 h. Following the incubation period, the number ofcolony forming units present on the membrane filters was counted.

A pilot study was performed to determine the ideal volume ofair (or rather theminimumnumber of spores) to be sampled for theevaluation of the filtration aptitude of the foam materials. Thevolume of air sampled had to be large enough to allow some fungalspores to pass through the foam samples in order to reveal thedifferent filtering capabilities of the different foams. The sampledvolume of air had to also yield a countable number of colonies uponculturing. Therefore, the volumes of air sampled varied between10 L and 1000 L. All experiments were performed on 3 separatedays by running triplicates each time.

2.2. Evaluation of air filtration capabilities of polyurethane foams

The polyurethane foam samples (Articoli Resine Espanse (ARE),Italy) were cut using a Proxxon Thermocut 230/E (Proxxon, F€ohren,Germany). Sheets of foamwere cut along the direction of maximumcell elongation to thicknesses of 15 mm and 20 mm. Four poly-urethane foams were selected for evaluation of air filtration capa-bilities against fungal contamination; 50 PPI foam of 15 mmthickness, 50 PPI foam of 20 mm thickness, 90 PPI foam of 15 mmthickness and 90 PPI foam of 20 mm thickness.

The experimental runs carried out consisted of the use of all 3

Fig. 1. Components of air filtration experimental setup.

M. Briffa et al. / Food Control 73 (2017) 91e100 93

manifold ports for both the analysis of the foams’ filtering capa-bilities and controls. These runs differed in that foam analysisentailed positioning foam samples in the adapters when samplingair whereas the controls involved sampling air in the absence ofthese foam samples. For the control, the volume of air filtered wassubdivided between the 3 manifold ports such that CFU’s of thecontrol would be obtained by adding the CFU’s grown on themembrane filters of each individual manifold port.

When the foam samples were being analysed, the area aroundtheir periphery was made air-tight by the application of a generousamount of petroleum jelly to ensure that all the air being sampled

Fig. 2. Scatter Plot illustrating the variability of

was passing through the foam. Furthermore, in this case, 3 timesthe amount of air used in the control was sampled. In this way, thevolume of air used in the control passed from each manifold port(assuming that equal amounts of air pass through each port).

The assessment of each foam consisted of three tests: a controlrun, a run to sample air through 3 identical foam samples (1 oneach manifold port), and another control run, in that order. Thecaptured spores were then calculated based on the differencesbetween the control and the foam samples. This was repeated onthree separate days for each of the polyurethane foams. Fresh foamsamples were used each time (the foam samples were not re-used).

CFU counts at different sampling volumes.

Fig. 3. Assessment of the filter efficiency for all the tested ppi and thickness.

M. Briffa et al. / Food Control 73 (2017) 91e10094

Crucially, the temperature and relative humidity at the time ofsampling was monitored and noted during all the runs using aPro’sKit digital temperature and humidity meter.

2.3. Identification of fungal organisms trapped by the foam filters

Once the quantification of the colony forming units was carriedout, the CFUs present were described accordingly. Each physicallydifferent CFU was sub-cultured onto fresh Sabouraud DextroseAgar with Chloramphenicol (SDC) plates by stab inoculation toobtain pure fungal cultures. Sub-culturing was performed by col-lecting a visible amount of fungal material from the CFU on theDRBC plate with a heat-sterilised straight wire and stabbing(inoculating) an SDC plate at 4 ends. The plates were subsequentlyincubated at room temperature for 5e7 days.

The sub-cultured colonies were observed following the incu-bation period. If a plate was found to be contaminated, the desiredCFU was sub-cultured onto a new SDC plates and re-incubated asrequired. Pure fungal cultures were initially assessed macroscopi-cally by describing the colony present in terms of colonialmorphology, colonial colour and diffusible pigment.

The pure fungal colonies were identified by means of micro-scopic examination of their conidia and conidiogenesis. The col-onies were assessed microscopically by the adhesive tape mounttechnique using Lactophenol Cotton Blue (LPCB) dye as a mountingfluid. Occasionally, a lack of distinctive conidia or asexual structureswas present on the adhesive tape mount. This equated to the or-ganism remaining unidentified such that a means of promoting theproduction of conidia was required. This involved sub-culturing the

unidentified organism onto Potato Dextrose Agar (PDA) and/orCorn Meal Agar (CMA) by stab inoculation. The media were incu-bated at room temperature for 24e48 h and once sufficient growthwas present, the organisms were examined microscopically by theadhesive tape mount technique discussed above. Organisms whichwere still not producing conidia were incubated further and theadhesive tape mount technique was repeated after a few moredays. Fungi which were persistently unable to produce conidiawere naturally described as such.

2.4. Scanning electron microscope (SEM) imaging

Micrographs of the 50 PPI and 90 PPI polyurethane foams, at amagnification of 100�were captured using a Carl Zeiss Merlin FieldEmission scanning electronmicroscope (SEM)with an InLens signaldetector, a scan speed of 8 (the rate by which the image is scanned;in this case the SEM takes about 10 ms per line) and an EHT (ExtraHigh Tension) of 3.00 kV. The foam samples were cut in cubeshaving a size of approximately 5 � 5 � 5 mm, making sure that thesurface being examined was that of the foam’s rising direction. Thesides and bottom of the cubes were covered with carbon tape.

The thicknesses of 20 randomly selected ribs of both of thefoams were measured from the micrographs using an SPM datavisualization and analysis tool, Gwyddion 2.4. The measurementsobtained were converted to mm through:

mr ¼�pmsm

�sv

Fig. 4. Air filtration efficiency of the analysed foams.

M. Briffa et al. / Food Control 73 (2017) 91e100 95

where mr is the measurement in micrometers, pm is the measure-ment of the rib thickness from the micrograph, sm is measurementof the scale and sv is the scale value.

2.5. Data analysis

In order to investigate the effects of the thickness and theconventional foam PPI of the foam samples on the air filtrationefficiency, the two independent variables were scaled andweighted according to the following formula:

Scaling and Weighting of Foam PPI or Thickness Measurements

¼�Original Measurment �Minumum Value

Maximum Value�Minimum Value

�

�Weight Factor

The weight factors were selected in such a way as to obtain thehighest correlation value (r2 value) for the sum of the two modifiedindependent variables against the air filtration efficiency. The effectof the two variables on the efficiency was studied by combining theconventional foam PPI and thickness into a single variable andattempting to correlate this variable with the air filtration effi-ciency. The variation of the number of CFUs captured with theambient temperature and percentage humidity was also investi-gated in order to determinewhether these environmental variablesaffect both the number of spores in the air (which is expected as perthe pilot study), as well as their likelihood of being trapped by thepolyurethane foams.

3. Results

3.1. Selection of the volume of air to be sampled

The pilot study was performed to determine the volume of air tobe sampled when carrying out the experiments. In other words, theminimum number of spores that needed to be collected in thecontrol so as to obtain coherent results that were independent ofthe CFU count number needed to be evaluated. Initial sampling ofdifferent volumes of air on three separate days gave a clear indi-cation that variability of CFUs/L was very large when small volumesof air were sampled (or rather when small number of CFUs wascollected). This variability seemed to decrease at larger samplingvolumes (larger number of CFUs collected). Fig. 2 demonstrates thevariability of CFUs/L obtained at different volumes of sampling.Each data point represents a CFU count obtained after sampling aspecific volume of air. The three readings obtained when sampling10 L, 20 L, 40 L, 60 L or 80 L of air are considerably dispersed acrossthe range of CFU/L at any given volume. Alternatively, the CFU/Lresults generated when sampling 100 L, 200 L, 300 L, 400 L or1000 L of air are not highly variable at any given volume in terms ofCFUs/L.

As is evident in Fig. 2, it was opted to sample on the side ofcaution, i.e., minimum variability, for a volume of at least 400 L. Byusing this volume an adequate number of CFUs would be of10e60 CFUs on any given membrane filter. Note that as indicated inthe Methodology section, the control is obtained by addition ofthree separate membrane filters found in each of the three mani-folds ports. The pilot study was performed between August and

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 0.5 1 1.5 2 2.5 3 3.5

Foam

Filt

ratio

n E

ffic

ienc

y (%

)

Mean of Weighted and Scaled Foam PPI and Thickness

Fig. 5. Correlation between combined conventional foam PPI and thickness and foam filtration efficiency with a correlation coefficient r2 value of 0.8231.

M. Briffa et al. / Food Control 73 (2017) 91e10096

September in the region of Msida, Malta. Preliminary results haveshown that seasonality and location enhance the microbialvariability.

3.2. Assessing the efficacy of foam filters

Fig. 3 illustrates the amount of fungal spores captured on themembrane filters with each of the tested foams. The 50 PPI foam of15 mm thickness significantly reduced the amount of fungal sporesin the sampled air on day 2, but this was not evident on days 1 and3. As opposed to the 50 PPI foam of 15 mm thickness, the 50 PPIfoam of 20 mm thickness significantly reduced the amount offungal spores in the sampled air on days 1, 2 and 3. The 90 PPI foamof 15 mm as well as the 20 mm thickness foam significantlyreduced the amount of fungal spores in the sampled air in all days.

The percentage of fungal spores eliminated from the air by thefoam on each day was calculated using the following formula:

Percentage Efficiency ¼ 100

��Average CFU=L ofFoam Analysis

Average CFU=L of Control

�

Note by Percentage efficiency one understands the % of CFUseliminated from the air.

The results obtained are shown in Fig. 4, which illustrates thefiltration efficiency of each of the foam materials which wereanalysed.

The efficiency of the 90 PPI foam of 20 mm thickness is notsignificantly different to that of the 90 PPI foam of 15mm thickness,but it is significantly different to that of the 50 PPI foams. When theefficiency of the 90 PPI foam of 15 mm thickness was compared

with that of the 50 PPI foam of 20 mm thickness no significantdifferences were observed while it is significantly different to thatof the 50 PPI foam of 15 mm thickness. It is evident that the effi-ciency of the 50 PPI foam of 15 mm thickness is significantly lessthan that of the other foams.

3.3. Assessing the effect of foam properties and environmentalconditions on the foam efficacy

The relationship between the sum of the scaled and weightedfoam thickness and conventional PPI, and the air filtration effi-ciency of the foam is shown in Fig. 5.

In order to obtain the maximum r2 value, the weight factor ofthe conventional PPI was set to be twice as large as theweight valuefor the foam thickness. This value was obtained by varying theweight factor linearly until the highest r2 valuewas obtained. The r2

value of 0.8231 obtained indicates that both the conventional foamPPI and its thickness contribute positively and linearly (within therange tested) to the air filtration efficiency of the foammaterial, andthat these two variables may be considered to account for a veryhigh (~80%) percentage of the observed variance in CFUs, the restbeing attributable to unknown causes such as natural variation inthe number of spores in the air.

In order to check whether environmental factors affect thefiltration capabilities of the foams, the temperature and relativehumidity of the air sampled was measured during all experiments.These were then plotted against the measured filtration efficiency,in order to ascertain whether there is a trend (Fig. 6).

There seems to be next to no effect of the temperature or rela-tive humidity on the filtration efficiency (see Fig. 6). In both figuresa data point corresponding to a negative efficiency is present. These

-10

0

10

20

30

40

50

60

70

80

90

100

12 14 16 18 20 22 24

Eff

icie

ncy

(%)

Temperature (ºC)

50 PPI 15 mm 50 PPI 20 mm 90 PPI 15 mm 90 PPI 20 mm

-10

0

10

20

30

40

50

60

70

80

90

100

40 45 50 55 60 65 70 75 80

Eff

icie

ncy

(%)

Realtive Humidity (%)

50 PPI 15 mm 50 PPI 20 mm 90 PPI 15 mm 90 PPI 20 mm

(a)

(b)

Fig. 6. Relation between filter efficiency (%) versus temperature (�C) (left) and filter efficiency versus Relative Humidity (%) (right).

M. Briffa et al. / Food Control 73 (2017) 91e100 97

Fig. 7. SEM micrograph of 50 PPI conventional polyurethane foam.

Fig. 8. SEM micrograph of 90 PPI conventional polyurethane foam.

M. Briffa et al. / Food Control 73 (2017) 91e10098

points have likely occurred due to the natural variability of thequantity of fungal spores in air in addition to the fact that itoccurred with the least efficient foam.

3.4. Organism specific filtration efficiency of foam filters

Based on the microscopic analysis discussed in Section 2.3 thefungal organisms trapped by the tested reticulated polyurethanefoam filters and the corresponding levels are presented hereunder:

- Cladosporium spp. (1.26 � 10�2 to 3.82 � 10�2 CFUs/L)- Penicillium spp. (0e2.10 � 10�2 CFUs/L)- Yeasts (1.88 � 10�3e4.86 � 10�3 CFUs/L)- Geotrichum candidum (0e1.25 � 10�3 CFUs/L)- Aspergillus niger (0e2.50 � 10�3 CFU/L)- Aspergillus terreus (0e3.75 � 10�3 CFUs/L)- Aspergillus flavus (1.00 � 10�3e1.00 � 10�2 CFUs/L)- Aspergillus fumigatus (0e1.25 � 10�3 CFUs/L)- Alternaria alternata (0e1.81 � 10�3 CFUs/L)- Fonsacaea pedrosoi (1.25 � 10�3e2.50 � 10�3 CFUs/L)- Rhizopus oryzae (0e1.25 � 10�3 CFUs/L)- Zygomycete spp. (0e1.25 � 10�3 CFUs/L)

- Sepedonium spp. (0e1.25 � 10�3 CFUs/L)

In addition, a considerably large number of colonies remainedunidentified due to an absence of conidia upon microscopic ex-amination (range of 2.50 � 10�3 to 2.00 � 10�2 CFUs/L). Othercolonies which appeared as ‘glossy’ circles on the cultured mem-brane filters remained unidentified as they were unable to grow asufficient amount of fungal material in order for a macroscopicdescription to be made and for sub-culturing to be carried out(range of 2.58 � 10�2 to 6.29 � 10�2 CFUs/L).

3.5. Scanning electron microscope (SEM) micrographs

Micrographs of the 50 PPI and 90 PPI conventional polyurethanefoams, at a magnification of 100�, were captured. The SEM mi-crographs of each of the foams are displayed in Figs. 7 and 8.

The thicknesses of 20 randomly selected ribs of both of thefoams were measured. The results obtained were recorded in thetable below (See Table 1).

The average rib thickness of the 50 PPI foam was found to be53.65 mmwith a standard deviation of 6.67 whilst that of the 90 PPIfoam was established to be 37.15 mm with a standard deviation of6.75.

4. Discussion

Most of the foam materials were found to be significantlyeffective at reducing the number of fungal spores in the filtered air.A clear trend is visible in Fig. 4, whereby the filtration efficiency ofthe conventional polyurethane foams increases with increasing PPIand foam thickness. An increase in PPI results in (1) smaller poressizes, (2) a higher amount of pores and (3) thinner ribs, all of whichaid the filtration process. Thinner ribs have such an effect as theyincrease the inertia on the spores passing through them by way ofthe larger surface area to volume ratio they provide. Studies showthat pore size usually affects the initial and final filtration effi-ciencies, while the rib diameter has a significant effect on the rate ofincrease of filtration efficiency (Seok, Chun, Song, & Lee, 2015).

When correlating the conventional foam PPI and thickness withthe efficiency of the foam material, the PPI was weighted to be 2times more relevant than the thickness in determining the filtra-tion efficiency of the foam material. This is likely to be the casesince an increase in PPI results in more ribs and cells in a given area,and also thinner ribs.

Utilising a polyurethane foam filter as an air filter in food pro-cessing/storage facilities or hospitals would require that it be of arelatively high PPI. Although important, the thickness is notnecessarily the priority in the design of the filter. A well-designedconventional polyurethane foam filter would provide a usefulfiltering tool, as presently demonstrated by its use in car filters andHVAC systems (New Dimension Industries, 2015).

Studies show a major disagreement as to the nature of therelationship between the seasonal variation of fungal spores andthe environmental variables. Some studies found that the numberof airborne spores was positively correlated with relative humiditybut negatively correlated with temperature such that highernumbers were observed during autumn and winter than in springand summer (Asan, Sarica Okten, & Sen, 2010; Bartzokas, 1975).Alternatively, other studies observed that a larger quantity ofairborne fungal spores was present during the warmer and morehumid summer months than during the winter season (Ceter &Pinar, 2009; Oliveria, Ribeiro, & Abreu, 2005; Shelton, Kirkland,Flanders, & Morris, 2002).

The two variables that were measured were the temperatureand relative humidity. Neither of the two variables seemed to have

Table 1Thickness of foam ribs.

50 PPI conventional polyurethane foamRib thickness (mm)

90 PPI conventional polyurethane foamRib thickness (mm)

44 2747 3953 3543 3158 4551 3151 3553 3756 4348 3548 4661 3553 4671 3051 2763 5157 4158 3151 3756 41

M. Briffa et al. / Food Control 73 (2017) 91e100 99

an effect on the filtration efficiency of the foam materials. Unsur-prisingly, both of these variables were found to influence the CFUcount in the control, in a manner whereby the number of fungalspores in the air increased with an increase in the temperature and/or relative humidity. This partially explains why a greater diversityof organisms was observed in the summer season, than in thecolder and less humid winter months.

5. Conclusions

In this work the air filtration capabilities in terms of fungalspores of a selection of polyurethane foams of differing combina-tions of pores per inch (PPI) and thickness were assessed. It wasfound that apart from the 50 PPI foam of 15 mm thickness, all thefoam samples were effective at significantly reducing the numberof spores. The PPI was found to be 2 times more influential on theefficiency of the foam material than the foam thickness. The highernumber of pores present and the decrease in thickness of the ribscomposing the microstructure of the foam explained the afore-mentioned observations.

Future studies could involve performing an in vitro study of thefiltration capacities of the foams prior to analysing them in terms ofoutside air filtration. This would involve preparing a suspensioncontaining a known quantity of genus/species-specific fungalspores and passing them through the foam materials being ana-lysed in a closed system. This would provide a considerably moreapt ground for comparison between the polyurethane foam undertest and other established filtration materials.

Acknowledgments

This research was partially funded from the European Union’sSeventh Framework Programme, REPEAR Project, under grantagreement nº 604733, FP7-SME-2013-SME AG".

References

Abdul Salam, Z. H., Karlin, R. B., Ling, M. L., & Yang, K. S. (2010). The impact ofportable high-efficiency particulate air filters on the incidence of invasiveaspergillosis in a large acute tertiary-care hospital. American Journal of InfectionControl, 38(4), 1e7.

Alberti, C., Bouakline, A., Ribaud, P., Lacroix, C., Rousselot, P.,Leblanc, T.,…Aspergillus Study Group. (2001). Relationship between

environmental fungal contamination and the incidence of invasive aspergillosisin haematology patients. Journal of Hospital Infection, 48(3), 198e206.

Araujo, R., Carneiro, A., Costa-Oliveira, S., Pina-Vaz, C., Rodrigues, A. G., &Guimaraes, J. E. (2008). Fungal infections after haematology unit renovation:evidence of clinical, environmental and economical impact. European Journal ofHaematology, 80(5), 436e443.

Asan, A., Sarıca Okten, S., & Sen, B. (2010). Airborne and soilborne microfungi in thevicinity hamitabat thermic power plant in Kirklareli city (Turkey), their sea-sonal distributions and relations with climatological factors. EnvironmentalMonitoring and Assessment, 164(1e4), 221e231.

Bartzokas, C. (1975). Relationship between the meteorological conditions and theair-borne fungal flora of the Athens metropolitan area. Mycopathologia, 57(1),35e38.

Brenier-Pinchart, M. P., Lebeau, B., Quesada, J. L., Mallaret, M. R., Borel, J. L.,Mollard, A.,…Pelloux, H. (2009). Influence of internal and outdoor factors onfilamentous fungal flora in hematology wards. American Journal of InfectionControl, 37(8), 631e637.

Brincat, J.-P., Sardella, D., Muscat, A., Decelis, S., Grima, J. N., Valdramidis, V. P., et al.(2016). A review of the state-of-the-art in air filtration technologies as may beapplied to cold storage warehouses. Trends in Food Science & Technology, 50,175e185.

Ceter, T., & Pinar, N. (2009). Atmospheric concentration of fungus spores in Ankaraand the effect of meteorological factors in 2003 period. Bulletin of Microbiology,43(4), 627e638.

Dassonville, C., Demattei, C., Detaint, B., Barral, S., Bex-Capelle, V., & Momas, I.(2008). Assessment and predictors determination of indoor airborne fungalconcentrations in Paris newborn babies’ homes. Environmental Research, 108(1),80e85.

European Hygienic Engineering & Design Group (EHEDG). (2006). Guidelines on airhandling in the food industry. Trends in Food Science & Technology, 17, 331e336.

Gliganic, R. (2008). Where reticulated polyurethane foam’s a fit. Retrieved fromhttp://machinedesign.com/archive/where-reticulated-polyurethane-foam-s-fit.Last Accessed 19.02.16.

Griffiths, W. D., Bennett, A., Speight, S., & Parks, S. (2005). Determining the per-formance of a commercial air purification system for reducing airbornecontamination using model micro-organisms: a new test methodology. Journalof Hospital Infection, 61, 242e247.

Isaac, S. (1996). To what extent do airborne fungal spores contribute to respiratorydiseases and allergic reactions in humans? Mycology, 10, 31e32. Retrieved fromhttp://www.fungi4schools.org/Reprints/Mycologist_articles/Isaac_answers/V10pp031-032spore_allergies.pdf. Last Accessed 19.02.16.

Jankowska, E., Reponen, T., Willeke, K., Grinshpun, S. A., & Choi, K. J. (2005).Collection of fungal spores on air filters and spore reentrainment from filtersinto air. Journal of Aerosol Science, 31, 969e978.

Ko, G., First, M. W., & Burge, H. A. (2002). The characterization of upper-room ul-traviolet germicidal irradiation in inactivating airborne microorganisms. Envi-ronmental Health Perspectives, 110, 95e101.

Lutz, B. D., Jin, J., Rinaldi, M. G., Wickes, B. L., & Huycke, M. M. (2003). Outbreak ofinvasive Aspergillus infection in surgical patients, associated with a contami-nated airhandling system. Clinical Infectious Diseases, 37(6), 786e793.

Magan, N., & Olsen, M. (2004). Mycotoxins in Food: Detection and control. Wash-ington, US: Woodhead Publishing Limited.

May, J. C. (2001). My house is killing me. Baltimore: Johns Hopkins University Press.Mili�cevi�c, D. R., �Skrinjar, M., & Balti�c, T. (2010). Real and perceived risks for myco-

toxin contamination in foods and feeds: challenges for food safety control.

M. Briffa et al. / Food Control 73 (2017) 91e100100

Toxins (Basel), 2(4), 572e592.Miller, J. D., Rand, T. G., & Jarvis, B. B. (2003). Stachybotrys chartarum: cause of

human disease or media darling? Medical Mycology, 41, 271e291.Mu~noz, P., Guinea, J., Pel�aez, T., Dur�an, C., Blanco, J. L., & Bouza, E. (2004). Noso-

comial invasive aspergillosis in a heart transplant patient acquired during abreak in the HEPA air filtration system. Transplant Infectious Disease: An OfficialJournal of the Transplantation Society, 6(1), 50e54.

New Dimension Industries. (2015). Industries served. Retrieved from http://www.ndi-us.com/industries.html. Last Accessed 19.02.16.

Oliveria, M., Ribeiro, H., & Abreu, I. (2005). Annual variation of fungal spores inatmosphere of porto: 2003. Annals of Agricultural and Environmental Medicine,12(2), 309e315.

Park, C. W., & Hwang, J. (2013). Susceptibility constants of airborne bacteria todielectric barrier discharge for antibacterial performance evaluation. Journal ofHazardous Material, 244, 421e428.

Park, J. H., Yoon, K. Y., Kim, Y. S., Byeon, J. H., & Hwang, J. (2009). Removal of sub-micron aerosols and bioaerosols using carbon fiber ionizer assisted fibrousmedium filter media. Journal of Mechanical Science and Technology, 23,1846e1851.

Polyurethane Foam Association. (1994). Joint industry foam standards and guidelines.Retrieved from http://www.pfa.org/jifsg/jifsgs15.html. Last Accessed 19.02.16.

Portnoy, J. M., Barnes, C. S., & Kennedy, K. (2004). Sampling for indoor fungi. Journalof Allergy and Clinical Immunology, 113(2), 189e198.

Price, D. L., Simmons, R. B., Crow, S. A., Jr., & Ahearn, D. G. (2005). Mold colonizationduring use of preservative-treated and untreated air filters, including HEPAfilters from hospitals and commercial locations over an 8-year period(1996e2003). Journal of Industrial Microbiology and Biotechnology, 32(7),319e321.

Reboux, G., Gbaguidi-Haore, H., Bellanger, A. P., Demonmerot, F., Houdrouge, K.,

Deconinck, E.,…Millon, L. (2014). A 10-year survey of fungal aerocontaminationin hospital corridors: a reliable sentinel to predict fungal exposure risk? Journalof Hospital Infection, 87(1), 34e40.

Sautour, M., Sixt, N., Dalle, F., L’Ollivier, C., Fourquenet, V., Calinon, C., et al. (2009).Profiles and seasonal distribution of airborne fungi in indoor and outdoor en-vironments at a French hospital. Science of the Total Environment, 407(12),3766e3771.

Schmid, S., Seiler, C., Gerecke, A. C., Hachler, H., Hilbi, H., & Frey, J. (2013). Studyingthe fate of non-volatile organic compounds in a commercial plasma purifier.Journal of Hazardous Material, 256e267, 76e83.

Seok, J., Chun, K. M., Song, S., & Lee, S. (2015). Study on the filtration behavior of ametal fiber filter as a function of filter pore size and fiber diameter. Journal ofAerosol Science, 81, 47e61.

Shelton, B., Kirkland, K., Flanders, W., & Morris, G. (2002). Profiles of airborne fungiin buildings and outdoor environments in the United States. Applied and Envi-ronmental Microbiology, 68(4), 1743e1753.

Vonberg, R. P., & Gastmeier, P. (2006). Nosocomial aspergillosis in outbreak settings.Journal of Hospital Infection, 63(3), 246e254.

Woo, M., Grippin, A., Anwar, D., Smith, T., Wu, C., & Wander, J. D. (2012). Effects ofrelative humidity and spraying medium on UV decontamination of filtersloaded with viral aerosols. Applied and Environmental Microbiology, 78,5781e5787.

Yoon, K. Y., Byeon, J. H., Park, J. H., Ji, J. H., Bae, G. N., & Hwang, J. (2008). Antimi-crobial characteristics of silver aerosol nanoparticles against Bacillus subtilisbioaerosols. Environmental Engineering Science., 25, 289e293.

_Zukiewicz-Sobczak, W., Sobczak, P., Krasowska, E., Zwoli�nski, J., Chmielewska-Badora, J., & Gali�nska, E. M. (2013). Allergenic potential of moulds isolated frombuildings. Annals of Agricultural and Environmental Medicine, 20(3), 500e503.