Monitoring Indoor Air Quality in Additive Manufacturing environment

HAL Id: hal-02965017

https://hal.archives-ouvertes.fr/hal-02965017

Submitted on 12 Oct 2020

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Monitoring Indoor Air Quality in Additive Manufacturing environment

Shirin Khaki, Maud Rio, Philippe René Marin

To cite this version:

Shirin Khaki, Maud Rio, Philippe René Marin. Monitoring Indoor Air Quality in Ad- ditive Manufacturing environment. Procedia CIRP, ELSEVIER, 2020, 90, pp.455 - 460.

�10.1016/j.procir.2020.01.113�. �hal-02965017�

ContentslistsavailableatScienceDirect

Procedia CIRP

journalhomepage:www.elsevier.com/locate/procir

Monitoring Indoor Air Quality in Additive Manufacturing environment

Shirin Khaki, Maud Rio

^∗

, Philippe Marin

CNRS, Université Grenoble Alpes, Grenoble INP, G-SCOP, F-380 0 0 Grenoble, France

a rt i c l e i n f o

Keywords:

Advanced industrial sustainability Cleaner production

Additive Manufacturing

Environmental and social analysis and assessment

a b s t r a c t

AnIndoorAirQuality(IAQ)monitoringandassessmentisnecessaryinordertosafeguardthewellbeing oftheoccupants.ThisresearchaddressesthisissueinanAdditive Manufacturing(AM)platformofan educationalinstitute.Performingairmonitoringrequiresquantifyingtheplatformuserexposures.Regu- lations,localpolicies,aswellasISOstandardsarewelldevelopedforofficeenvironmentsandindustrial sites.Howevertheindicatorsandthepollutionlevelsallowedfornewtoolsandtechnologiesareyetto bedefined.Thisis specificallythecase inFrance forAMeducational purpose.ApracticalIAQassess- mentprotocolisaddressedinthisresearchquestioning:thetypeofemissionstomeasure,andtheway tomeasure them onanAM platform.The methodincludesaselection ofthe assessmentparameters forthe IAQ,the associatedIAQ indicatorsand theassociated sensorsto measureemissions rates.The assessmentprotocolhasbeenappliedtoGINOVAS.martGrenoble-AlpseducationalFabLabforstudents andresearchers.EmissionshavebeenmeasuredfromavarietyofAMtechnologies,mergingmaterialand gasinputresources.Thechoiceofsensorandmeasurementmethodisdiscussedinthispaper.Themea- surementresultscoveravarietyofparticlesandgasemissions.Thispaperconcludesonthelimitsand theopportunitiesofanIAQdynamicassessmentprotocoltomakeanemergingmanufacturingplatform eagertodevelopthenumberofmachinesusingvarious materialstypes,acleanerenvironmentforits users.TheMakerscommunitycouldbeinterestedtousesuchaprotocolinanystandardFabLab.

© 2020TheAuthor(s).PublishedbyElsevierB.V.

ThisisanopenaccessarticleundertheCCBY-NC-NDlicense.

(http://creativecommons.org/licenses/by-nc-nd/4.0/)

1. Introduction:qualitymonitoringFabLabs

Indoor Air Quality (IAQ)regulations, local policies, as well as ISO16000:2004forofficeenvironmentsandindustrialsitesareun- clearaboutdefiningtheAdditiveManufacturing(AM)technologies indicators and level of emissions safeguarding the well being of themachine users.AMindustry isarapidlygrowingmarket field (Mendes,2017)producingcomplexstructures(Graff,2016)aswell ascustomparts.3Dprintingandlasercuttingtechnologiesarein- deedwidely usedatuniversitiestosupportthe processofdesign development (McDonnell, 2016). Precise recommendations about theinfrastructurerequiredtomonitorandmaintainanacceptable levelofIAQareparticularlylackinginFranceinFabricationLabo- ratories(FabLabs)foreducationalpurposes.Thenoveltyofthisre- search is,therefore,toproposean IAQassessmentprotocolbased on: (1)severalmainrelevantindicatorsandmeasurement instru- ments, covering specific VOCs and PM emitted in AM and laser cutting technologies (2)a methodadapted to academics andFa- bLabcontextsusage,integratingtherelatedactorsintomonitoring

∗ Corresponding author.

E-mail address: maud.rio@g-scop.eu (M. Rio).

andassessingVOCsandPMforpreventinghealthdamageinaone yeartimeperiod.

Section2addressestheresearchobjectivesrefiningthetypes ofVOCsandPMemissionstobemeasuredandthewayto measurethembasedonaliteraturereviewincluding

experimentalfeedbacks.Section3proposesanintegratedprotocol forIAQmanagementinFabLabsbasedontherelevantindicators andtheusageofsuchmanufacturingplatforms.Section4studies theapplicationoftheprotocoltotheGINOVAplatformin Grenobleandconcludesontheapplicabilityoftheproposition.

RecommendationstointegratethisprotocoltoanyAMcontextare formulatedinconclusion. Nomenclature

AM AdditiveManufacturing FabLab FabricationLaboratories IAQ IndoorAirQuality

VOCs VolatileOrganicCompounds PM ParticulateMatters

UFPs Ultrafineparticles

https://doi.org/10.1016/j.procir.2020.01.113

2212-8271/© 2020 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license.

( http://creativecommons.org/licenses/by-nc-nd/4.0/ )

456 S. Khaki, M. Rio and P. Marin / Procedia CIRP 90 (2020) 455–460

2. Paperissue:detectingandpreventingexposuretonew indoorpollutantsemittedinFabLabs

2.1.FabLabsarenotasafeguardedareaforhealth

FabLabs are publicly accessible workshops providing accessto digitalmanufacturingtechnologyandelectronicstoolstoeveryone tofacilitate approaching theDoIt Yourself(DIY)concept (Troxler, 2016) – a‘make (almost)anything’ place (Gershenfeld, 2012). In- novation,invention,learningandsharingideasarethenfacilitated.

TheFabFoundation countsmorethan 1750FabLabs in109 coun- triesused at the moment.They are often ownedby a school or universityor are sponsoredby huge corporations foreducational orrecreationalpurposes.Standardmachinesandprocessesshared include3Dprinter,lasercutterfor2Dand3Dstructuresmanufac- turing,high-resolution ComputerNumericalControl(CNC)milling machine,as well asseveral electronic components andprogram- mingtools.Theutilizationrateofthosemachinesandthematerial quantitiesusedarehowevermuchvaryingineachplatform.

Despite this considerable non-industrial manufacturing activ- ity,thetypesandmagnitudeoftheemissionsoccurringthereare not oftenpublishedon-site orin specific journals.Several litera- turestudiesdemonstrate that3D printers emit ultrafineparticles (UFPs)besidespotentially dangerous Volatile OrganicCompounds (VOCs)such asStyrene, Butanol, orEthylbenzene (Wojtyła,2017; Azimi,2016; Unwin, 2013). The compliance oftheseplatforms to IAQstandardsisunclear.Analysingtheoccupant’sexposuretothe hazardouspollutantsoccurringtherethrough aprotocolis,there- fore,anecessity.

2.2.Principalindoorpollutantsandparametersforhealthdamages assessments

The indoor air presents some high concentration of out- door substances captured from outside (World Health Organiza- tion,RegionalOfficeforEurope,2010).Additional aircomponents (William, 1992) are generally originated from the indoor equip- ment, and from the occupant’s behaviour. The indoor air pollu- tioncomponents are sorted inseveral categories,based onphys- icalproperties, chemical properties,adversehealth effects,or froma pollution source.The chemical properties classification forinstance commonly distinguishes: the chemical pollutants, such as volatile organiccompounds(VOCs),nitrogenoxide(NOx),carbonmonoxide (CO), polycyclic aromatic hydrocarbons (PAHs), phthalates; from theorganic contaminants such as mold, householdallergens from dustmites,pollens;andfromthephysicalpollutantsincludingpar- ticlesandfibres(asbestos,artificialmineralfibres,etc.).Intermsof adverseeffects onhumanhealth,theVOCsandaldehydes aremost oftenthecauseofirritationoftheeyesandrespiratorytract.Some ofthem,suchasbenzeneandformaldehyde, arefurtherclassified as ‘carcinogenic to humans’ by the International Agency for Re- searchonCancer.

Assessingthehealthdamagecausedbythepotentialpollutants inFabLabsis,therefore,a factorofthepollutantstoxicityandcon- centrationsparameter:upto900organiccompounds,particles,mi- crobes,andallergens.Therelatedexposureparameter:inhalation,in- gestion,or dermal contact (Lunetto, 2018). The exposure-response relationshipparametertotherelatedbodyexposedtoasubstance.

2.3.FocusingonIndoorAirQualityinFabLabs

The composition andtheconcentration ofthegeneratedVOCs andUFPsemissionsareaffectedbyseveralparameterssuchasthe filamentmanufacturedtype,theextrusiontemperature,orthebed temperature(Kim, 2015). New andinventive filaments enter the

marketeveryyear.HoweverABS– ahighemitter(Graff,2016)– is stilloneofthemostcommonlyusedfilamentswithPLA,andHigh ImpactPolystyrene(HIPS).

TheUFPsandVOCsemissionsconcentrationincreaseswiththe number of printers working together. This concentration affects humanhealth (Peters, 1997).Quantifying pollutionscausedby an additionalmachineisstillnotclearlyestablished.Someeffectsare howeverscientificallyproven.Forinstance,anElectronBeamMa- chine (EBM) using materials powder of 10 to 100

μ

^{m size in-}

creases the risk of dust explosions (Lunetto, 2018). As another example, the highlytoxic vapours,particulates, and metal fumes fromthesubstrate(mainlyplastics,woods,andmetals) produced with laser cutter (Yun-Jung, 2016), generate specks of dust that mightincludemetalssuchaschromium,lead,nickel,orcadmium.

This dust thickness also affects the cutting efficiency by making thebeamdifficulttodiffuse.Those examplesshow thatascreen- ing processisyet missingfordetectingIAQproblemsin FabLabs:

identifying the potential emission sources, the related measure- mentdevices,aswellasthelimitsnottobecrossedtoguarantya healthysafeworkplace(Hui,2009).

Thisresearch, therefore,aims at determiningthe physicaland chemical characteristicsofthe emittedparticles andVOCspoten- tially released in a FabLab to define the proper device to detect them through a general protocol. Measuring the realistic human exposuretothesepollutantsisnecessaryforanyIAQimprovement strategiesinthegivencontext.

3. Researchproposition:aprotocolforIAQmanagementin FabLabs

3.1. TheIAQassessmentparametersandprocessforFabLabs

Any personal complaints are an early demonstration of IAQ problems,addressing a necessity forimprovementactions. A first assessmentisrequiredtodeterminethethreats,estimatingtheex- posure, analysing the impact of the sources emissions, and pro- vidingquantified data forthe appropriate measures to be taken.

A continuous measurement of the indoor pollutants would en- able identifying IAQ problems and adjust solutions. A significant amount oftime, resource andeffortare required to obtainsome accurateresultsinmediumtolongterm(Hui,2009).Themainas- sessmentparameters toconsiderare:theFabLabinternalorganiza- tion, the Makers’ practiceand behaviour,their time spent inside, in addition to the energypolicies and building systemtechnolo- giesinplace,withinthelocalrequirementandstandardsforIAQ.

Detectingthepollutionpatternsoccurringbyaquestionerorbyin- terviewing theMakersis crucial(e.g.daily, weekly, seasonally),as the pollution ratewill be influenced by: the ventilation andex- traction localisationandmode, the machines’utilisation rate, the materials and liquids used, the humanactivities andmovements in the rooms, aligned with the season and the time of the day.

Thedatameasurementperiodshould,therefore,bechosenaccord- ingly in a systematic risk assessment strategy–within a dynamic andpracticalprotocol(Hui,2009).APlan-Do-Check-Act(PDCA)asa commonapproachintheindustryischoseninthisresearchtosupport IAQmanagementprotocolin FabLabsandprevent theusers’health damage.

3.2. TargetedpollutantsinFabLabs

The frequentair pollutantscontributingto weakeningIAQ are CO₂, NO₂, formaldehyde (HCHO), CO, SO₂, total volatile organic compounds (TVOCs)andairborne particulatematters(i.e.PM2.5 andPM10).Airtemperature,velocity,andhumiditylevelsarefur- therIAQ factorsassociated withoccupants’comfortzone (Abdul-

Wahab, 2015). The oxygenis not classifiedasa factor that could affect poorIAQ, buta sufficient oxygen concentration level must be maintained. The mostcommon andhigh-end levels of indoor airpollutantsinEuropeanCountriesandtheircomparisontoWHO guidelinestandardscanbefoundinJantunen(2011).Forinstance, formaldehyde is typically found indoors, at a concentration rate varying from20to 80μg/m³.However, samplingevery toxicpol- lutantisalmostimpossible.Hence,theFabLabairsamplingwillbe based onan assessmentstrategy, targetingspecific pollutantsde- tectedduringtheinitialassessmentassignaturepollutants,moni- toringthemfor:aregulationrequirement,checkingtheimpactim- plementationofachange,ahazardevaluationincaseofanewAM machine, newmaterialsused,etc. Thesignature pollutantstotar- geti.e.generatedfromtypicalmachinesusedinAMplatformsare takenfromParticulateMatters(PM)andTVOCs.

3.2.1. Targetingparticulatematters(PM)andhealtheffects

PMs are solid orliquid particles distributed evenly inthe air.

Theyremainairborneforprolongedperiodsowingtotheirparticle size, which could range from1 to 10,000 nm (cf. WHOreports).

The smaller the particle size, the more difficult it is to be mea- suredandcontrolled. PM10includesall particleswithadiameter of fewer than 10

μ

^{m and therefore PM2.5 – ultrafine particles.}

These particles can generate environmental and health damages, aswellasclimaticimpacts (Jaffrezo,2018).The mainexposureto theseparticlesisthroughinhalation.LargeparticlessuchasPM10 can be eliminated fromthe body by sneezing, coughingorswal- lowing.Particle smallerthan10μmindiametermaygodeepinto thelungs andpotentially tothe bloodstream.Thegreatestrisk to healthiscausedbythefineparticles(diameterinferiorto2.5μm) orPM_2.5(Graff,2016).Theeffectsofinhalingthisparticulatemat- ter includeasthma,lung cancer,respiratorydiseases, cardiovascu- lardisease,prematuredelivery,birthdefects,lowbirthweight,and prematuredeath.Understandingsuchpropertiesandtheprocesses thataffectPMevolutionreferringtotherelatedresearchfieldsare relevanttocontinuouslyimprovetheFabLabIAQ(Jaffrezo,2018).

3.2.2. Targetingvolatileorganiccompounds(VOCs)

Volatile organic compounds are a set of substancesbelonging to differentchemical classes having incommon their capacityto evaporate more or less quickly at room temperature. WHO has sortedtheseorganicpollutantsaccordingtotheirboilingpointinto 3categories:VeryVolatile,VolatileorSemi-VolatileOrganicCom- pounds (mainly found in the form ofgas in the air). The short- term exposure to VOCs may cause eye irritation and respiratory tract, headaches, dizziness, aswell asvisual disorders.Long-term exposures mayresultinmoreserious symptoms likefatigue,loss ofcoordination,damage totheliver,kidneys,andcentralnervous system.VOCsincludeformaldehyde,d-Limonene,toluene,acetone, ethanol,2-propanol,andhexanalsubstances.Thesummationofall detected VOCsisnamedTotalVolatileOrganicCompound (TVOC).

Indoor hygiene and Indoor Air Quality consider TVOC as a ma- jorindicator.ScientificstudiesoftenindicatetheTVOCraterather thandetailingtheanalysedVOCs.Inthiscase,theTVOCcomposi- tionshouldbeclarified(cf.WHOGuidelines).

3.2.3. Guidelinesvalues

IAQstandardguidelineseditedbytheEuropeanCommissionare usuallyprovidingreferencevaluesincaseofdevelopinganewfa- cility. In contrastto the outdoor air, very few specific provisions existto dategoverningthe qualityofindoor airinnon-industrial premises.Differentregulationsandguidelineshavebeenhowever definedby variousinternationalagencies. Thesestandardsareei- ther based on health concerns or acceptable levels for occupant comfort.

The European Harmonised Framework on IndoorMaterial La- bellingSchemesincludescommoncoreandtransitionalcriteriaon testing and evaluation methodologies related to indoor products chemicalemissions(Kephalopoulosetal.,2012).Thisisareference pointforharmonizingdifferentlabellingsystemsandforidentifi- cationofpotential overlapping inthe existingtest methodologies andsensoryevaluation(Bravi,2019).Someguidelinesandpolicies specificallyexistforpreventingpersonal exposures.Mostofthese guidelinesaredevelopedbyinternationalscientificcentresinclud- ingWHOandtheEnvironmentalandOccupationalHealth&Safety (ANSES).Theagreementcovers thetypesofpollutantsbutdiffers invaluedefinitions,andinthelegallybindingstatusofpollutants.

NoofficialguidelinevalueshavebeenestablishedintheEU’sEuro- peanCommission(Abdul-Wahab, 2015).Some Europeancountries aswellasCanada,theUSA,andAustraliahavetheirownindividual regulationsandrecommendationsfortheassessmentofIAQ.These proposedvaluesdependonthemethods ofsamplingandanalysis developed by their national training partners (Bravi, 2019). Reg- ulations andrecommendationsalso differin their legallybinding status.

France wasone of the first countriesaddressing indoor envi- ronmental pollutions in a form of legislation. To deal with the health issue surrounding IAQ and providing the public author- ities with useful information to manage this risk, the ANSES has been conducting expert assessments for ten years on the development of IAQ Guidelines integrating the WHO and the French IAQ Observatory (OQAI) analyses. Those guidelines aim to protect the general public from any harmful effects of air- borne exposure to substances. Quantitative value limits for CO₂, NO₂, formaldehyde, CO, SO₂ and some other type of VOCs are provided. Recommended measurement methods are also sug- gested. However, there is no clear guideline defined for PM or TVOCs.

3.2.4. Anassessmentmethodologyrequired

After the preliminary analysis of the location of interest, the numberofsamplingpoints,samplingduration andfrequencyand samplingmethodsshouldbedetermined.Eachbuildingspacecov- eringanactivitycanbeidentifiedasa‘zone’.Azoneisdefinedas asetofspaceswhicharecharacterisedwiththefollowingcriteria:

(1)ventilatedwiththe sameairdiffusion strategyandbe served bythesameairhandlingunit;(2)havesimilar activities,thermal loadandpollutantemission; and(3)spaces withsimilar compli- antrecordsorhostingmorerequiringoccupants(Asadi,1013).The minimum number ofsampling points in each zone is calculated bya givenformula, combiningthenumberofsamplingpointsN, andareaofconcernzonesinsquaresmeter.AccordingtoISOstan- dard16000:2004,samplingmustalsobetakingatleast1maway fromthewalland1to1.5abovethefloorsinceapproximatelythe averagebreathingzone.The samplingtimeshouldbe perfectlyin linewith the objectives. Short term sampling– about 60 min – in caseof suspecting highconcentrations; long term sampling – fromseveralhourstoafewweeks– forassessingaverageairpol- lution.ForanyIAQguidelinereferencevaluecompliance,thesam- plingdurationmustbealignedtoreferencevalueone.Inaddition, backgroundmeasurements andpost-operation measurements are recommendedtohighlightthedecaytime.Thesamplingmethods eventuallydependonthetargetedpollutants.Forexample,aVOCs analysisisnowadaysdominatedbytheuseofpurge-and-trapfol- lowedby a gas chromatography (GC) orby gaschromatography- massspectrometry(GC/MS),orinsome casesbythermaldesorp- tion(toairpage).Thebestmetrictousedependsonthemeasure- mentspecificfocus.

Inthisresearch,thesamplingmethodsarealignedtotheANSES guidelines.

458 S. Khaki, M. Rio and P. Marin / Procedia CIRP 90 (2020) 455–460

3.3.ApracticalIAQassessmentprotocolbasedonaPDCA managementsystem

To sum-up,the IAQassessment protocoladdressed inthis re- search is questioning the type of emissions to measure, andthe waytomeasure them ina FabLab. Inpractice thisprocess starts witha walk throughinspection inorder tounderstandthe space usage,theactivitiesthataretakenplacedinside,themachinesand materialsused,aswellastheventilationsystem. Anyinformation fromthe FabLab users can point-out on a new parameter to in- cludeintheassessmentfactors.Aquestionercanbeusedtointer- viewthemonaregularbasis.Theassessmentfactorsarethenused toplanthe auditprotocolindetail: whatshouldbe measuredand howtheyshouldbe measured,aimingto findsolutiontothecom- plaintsmadeby the Makers; evaluate the effectiveness of a new mitigation process tomaintain a targetedIAQ level;be (atleast) complianttothecompetentauthoritiesguidelines, andabletore- portthemtherelatedproof.

Thentheimplementationstagebegins:theobtainedresultscan becompared withthe appropriatestandardsconsidering thatthe exposure time varies for each type of occupant (long term ex- posure for the regular staff and short term forstudents and re- searchers).If the resultsare not compliant withthe standards, a setofriskmanagementactionsshouldbetaken.

The plannedactions ateach stepofthiscycleassistinvarying andprioritising theproblems,enablingtheimplementation ofan auditprocedure,an actionplan,itscontrol,andtheglobalcontin- uousimprovementsovertime.

4. Casestudy:applyingtheprotocolforIAQmanagementof VOCsandPMsinaGrenobleeducationalFabLab

4.1.Step1:Plan

Context: GINOVA is an inter-university technological platform includinga technical platform and numerous spaces and project roomslocatedonthesiteoftheGrenobleSchoolofIndustrialEngi- neering.Itisoneofthe5platformsmanagedbyS.martGrenoble- Alps,formerAIPPrimecaDauphiné Savoie.Thisplatformworkson the model ofFacLabs, a concept similar to the FabLab one inte- grated within a university environment: a place of experimenta- tionandcollaborationbutonlyopen tostudents, researchersand schoolpartners.

Machines,filters,andmaterial:GINOVAisequippedwithmuch- advanced equipment dedicated to the simulation of mechanical systemsandworkflows,prototypingandadditivetechnologies.Be- sides3Dprinters,Lasercutterand,CNCmillingmachine,theplat- formisequippedwithEBMandZ-Ultramachines.The3Dprinters are Fused Deposition Modelling (FDM) Zortax printers, equipped withHEPAfilters.Materialsusedfor3DprintingareABS,PLA,TPE, Z-ABS,andULTRAT, respectively.Thelaser cutterisusedgenerally in4mainmodes:cuttingPMMAthickness3mm,5mm,or8mm, andengraving. OnlyPMMA5mm waschosen forcuttinganden- gravingwiththelasercutinthiscasestudy.

Concerns: the main issues reported by the staff and regular usersare relatedto the laser cutterand3D printers. 3D printers are placed in a non-ventilated area, potentially leading to fumes andparticleaccumulation.Thelasercutterisequippedwithalo- calventilationexhaustsystem.However,despiteclearinstructions, most students forget to turn the ventilation on. This generates odorousfumes.Inaddition,thefiltersaresometimesoverusedand insteadof purging the air,it discharges morepolluted air in the room.This issue hasled to headache, dizziness, andirritationto nose,throat,andeyes,thataretheshort-termeffectstoairpollu- tionexposure.

Based on such problems and additional one captured at this initialstage theIAQobjectivesandlimits,thetargetedpollutants, the type of the sensor, and the measurement procedure have been defined. The audit plan has been prepared in3 phases: 1- Choosingappropriate guidelinesfromtheANSESones.2-Targeting pollutants:PMs(UFPs)andVOCsincludingbenzene,trichloroethy- lene,tetrachloroethylene,ethylbenzene,toluene,acetaldehyde,and acrolein.3-Organisingthe measurementmethod:choosinginstru- ments,rentingtoolsandsamplinglocations.

Chosen measureinstruments: the VOCs were collectedby cus- todian SMPE syringes, withthe flow rate of 35mL/min. Each of themeasurementstookplaceoveracourseof5min,with175mL airbeingcollectedandanalysedbyaportablegaschromatograph- massspectrometer(GC/MS,TorionT-9,PerkinElmer)GC/MSanaly- siscanreadilyandaccuratelysegregatecomplexcompoundsfound intheair.Lessvolatilechemicalsmoveslowerthanmorevolatile chemicals and are therefore separated by a GC/MS. GC/MS can measure the amount of every chemical present in a given sam- plesimplyby comparingtheparticularchemicaltopre-measured standard. VOCs were identified by matching their spectral mass characteristicsandretentiontimesusingalaboratory-specificspec- tralmassdatabase.Thisdatabasecontainsapproximately700VOC andhasbeenvalidatedbythelaboratoryforanalysisusingthelab- oratoryspecificsystemsfollowingamethodcombiningEPATO-15 and ISO standard 16000-6 guideline. The PM was detected by a Naneos– PartectorinstrumentationmeasuringtheLungDeposited SurfaceArea(LDSA)ofnanoparticlesbasedonanon-contactelec- trical detectionprinciple. LDSAconcentration isa relevantmetric for evaluating the negative health effects of aerosol particles by deep penetration into the lungs. The concentration rangeof this sensorisfrom0to12,000μm²/cm³andthesizerangeofthede- tectedparticlesisbetween10nmand10μm.Thesensormeasures theparticleswithatimeresolutionof1s.The limitsofexposure tonanoparticlesmeasuredbyLDSAsensorsaredefinedasaverage between50and250μm²/cm³referringtothesensormanufacturer.

Thestandardobjectchosenfor3DprintingwasareducedsizeAM modelproposed bythe NationalInstitute of Standards andTech- nology(NIST)developedasatestparttoevaluatetheperformance ofsuch technologies(Moylan,2012).SampleZone:thearea ofthe platform has beendivided into 6 differentzones. Zones 1 and2 were ‘conventional test rooms’forassessing the concentration of thepollutions.7adjunctprinterswereplacedinZone3.TheEBM, laser-cut,andWAMwerealsoplacedinzones4,5and6,respec- tively.

4.2. Step2:do

TheVOCexperimentswereconductedoverfourdistinctopera- tionalperiods:(1)anoverallbackgroundmeasurementofthelabo- ratorybeforeanyoperationbegins,(2)abackgroundmeasurement adjacenttothetargetedmachinesandzones(i.e.closeroom1and 2),(3)duringthemachinesoperations– approximately60minfor 3D printers,6 minfor thelaser cut, (4) post-operationmeasure- ments,inordertohighlightthedecaytime,i.e.thetimenecessary forhaving theenvironment concentrations.The recorded average temperaturerangedfrom26°Cto 28°C.The VOCmeasurements were conducted in one day, with thehelp of3 agents of Perkin Elmer Company and3 researchers. The UFP measurements were conductedover a course of5 consecutivedays onlywiththe re- searchers and local staff able to use the differentmachines (e.g.

EBM).Sincethecasestudyobjectivewastoidentifythemaximum concentration value of VOCs andUFPS, short-term samplingwas carriedout duringthe60minfor3D printers,and6minforthe lasercut.

Table 1

Particulate matters concentration for laser cutting in μm ²/cm ³.

Fig. 1. PMs concentration in μm ²/cm ³ over time (abscise values in seconds) for PMMA 5 mm laser engraving and cutting using a LDSA machine.

4.3. Step3:check

Duringthemeasurements,theaveragebackgroundemissionof thelabwas25

μ

^m2/cm3,thetemperatureandtherelativehumid- itywere28°Cand45%,respectively.Allthemeasurements,except theEBM,wereconductedin3phases:pre-operation,duringoper- ation, andpost-operation.PMresultsforlaser cut,cf.Fig.1:over the engraving process the emission concentration remained in a steady-state,justasthepre-operationmeasurement.

As soon asthe cutting starts, the concentration begins to in- creasewithout muchdifference betweenwithandwithout filtra- tionmodes.A100-spauseafterthecuttingprocessisrespectedto let the remaining fumes be extracted fromthe chamber towards thefilters,lettingthemachineandthecutplatetocooldown.The window is openedat the post-operationphase: the highest con- centrationratereachesamaximumof469

μ

^{m2/cm3 fornoventi-}

lationand90

μ

^{m2/cm3 withthe}ventilation.Thesteady-statewas still not achievedattheendofthepost-operation (decayperiod) measurements.

4.4. Step2:act

Table1addressesthesensorrecommendationstothePMscon- centrationpresented inFig. 1(laser cut).As an averagewiththe ventilationPMsstayunderthelimitedrecommendation(ingreen- compliant). This limit is exceeded during the post-operation pe- riod without the ventilation(in red-danger). As an average with- out ventilation and at the maximum recorded with the ventila- tion,the ratesrecordedare dangerous forhumanexposureup to a giveninhalation timeperiod (inorange-to control).Similarly, to the lasercut measurement resultspresented inthispaperinFig.

1andTable1theoverallresultsshowedrelativelyhighPMseven ifonemachineatatimewasconsidered.Inaddition,theplatform hadthelowest yearutilisationrateinJulyandwithnoconsider- ationofaccumulationofPMs.Accordingly,thepriorityforfurther andmoreextensiveinvestigationshouldbeon3Dprintersandthe

laser cutter, due to their intensive use and relatively highemis- sionrate.Also,an appropriatepalliation methodshould beputin placestraightaway,especiallyfor3Dprinterssinceresultsshowed thatthefiltrationsystemwasnotaseffectiveasexpected.Specific mitigation:theemissionof3Dprinters,forinstance,couldbemit- igatedbyoperatingthe3Dprintersinsideasealedenclosurewith novelairfiltrationtechnologiese.g.photocatalyticfiltration,which isan activetechnologyforthedecompositionofinorganicandor- ganicpollutants(Wojtyła,2019).Anothersolutionistoimplement a2-layeredfiltration system, inordertomakesurethat eventhe smallestparticlesare trapped.Generalemissioncontrol strategies mitigatingthepollutionrateinallzoneswouldincludesome op- erationofstandaloneaircleaners,installingaspotventilationsys- temandupgradingthecentralHVACfiltration.Someofthesecon- trolstrategiesmaybe morecost-effectiveandpractical thanoth- ers.Forexample,ahighflowrateventilationsystemthatexhausts outdoors is likely cost-prohibitiveand impractical in many loca- tions.Whileusingstandaloneaircleanermaysignificantlyreduce the PM concentrations in all zones a substantial energy penalty wouldbegenerated(Parham,2017).Conclusion:incaseoffacinga highconcentration someshort termsolutionsincludinginstalling aspotventilationsystemandupgradingthecentralHVACfiltration couldbeputintheplace.Howevermoreextensivestudiesshould beconductedtoquantifythesignaturepollutants,pollutantsgen- eratedinhigherquantity,ormosthealthaffectingpollutants.

5. Conclusion

AM technology opens to endless design opportunities and is particularlyusefulforstudents.However, extracaution shouldbe takeninFabLabsusingmanymachinesconcurrently,especially in poorly ventilatedspaces orwithout the aidof the particlefiltra- tion system. This research, therefore, proposesa method for as- sessingtheIAQinAMenvironments.Thecasestudydemonstrated that the PDCA basedrisk analysisand assessmentapproach pro- posedasa generalprotocolisadapted tothe context ofFabLabs.

This methodwas implemented in GINOVA platform in Grenoble, toassesstheoccupant’sexposuretopotentialpollutantsandverify thecredibilityofthisapproach.The measurementsweredesigned followingthemostordinaryusagemode ofeach ofthemachines.

Theresultsrepresentedapotentialexposureforusersofthisplat- formtoahighconcentrationoftheparticles.Thisisespeciallythe casefor3D printersandthelaser cut,mostcommonlyused ma- chinesinFabLabsingeneral.Anextendedanalysisisnowrequired to quantify the signature pollutantcompounds for each machine in each zone and define an appropriate mitigation plan for IAQ.

Currentresearchincludestheaccumulationofcontaminantsinthe caseofseveralmachinesworkingtogetherbythesimulationsoft- wareCONTAM,whichisamulti-zoneIndoorAirQualityandven- tilationanalysiscomputer programdesignedtohelppredictionof contaminant concentrations to determine the Indoor Air Quality performance ofbuildings. Thiswill supportthe PDCA continuous improvementstagesbyinvestigatingtheimpactsofimplementing additionalairqualitycontroltechnologyinagivenFabLab,tomake anefficientdecision.

460 S. Khaki, M. Rio and P. Marin / Procedia CIRP 90 (2020) 455–460 DeclarationofConflictInterest

Theauthorsdeclarethattheyhavenoknowncompetingfinan- cialinterestsorpersonalrelationshipsthatcouldhaveappearedto influencetheworkreportedinthispaper.

Acknowledgements

AcknowledgementstotheS.martnetwork,GINOVAactors,UGA researchcolleaguesthatparticipatedinthisresearch.Thecompany PerfinElmerisgreatlythanksforthetechnicalexpertiseandhigh- qualitymeasurementdevices.

References

Abdul-Wahab, et al., 2015. A review of standards and guidelines set by international bodies for the parameters of indoor air quality. Atmos. Pollut. Res. 6 (5), 751–

767. doi: 10.5094/APR.2015.084 .

Asadi , et al. , 2013. A systematic indoor air quality audit approach for public build- ings. Environ. Monit. Assess. 185, 865–875 .

Azimi, et al., 2016. Emissions of ultrafine particles and volatile organic compounds from commercially available desktop three-dimensional printers with multiple filaments. Environ. Sci. Technol. 50 (3), 1260–1268. doi: 10.1021/acs.est.5b04983 . Bravi, L. , 2019. Additive Manufacturing: Analysis of the Economic Context and Eval- uation of the Indoor Air Quality, with a Total Quality Management Approach.

Ph.D. Dissertation. Univertita Degli Studi Di Urbino Carlo Bo, pp. 1–315 . Gershenfeld, N , 2012. How to make almost anything, the digital fabrication revolu-

tion. Foreign Affairs 91 .

Graff, et al. , 2016. Evaluating measuring techniques for occupational exposure dur- ing additive manufacturing of metals. J. Ind. Ecol. 21 (1), 120–129 .

Hui , et al. , 2009. An indoor air quality monitoring and assessment protocol for air conditioned offices: a case study of Hong Kong. E. Green Build. Energy Resour.

513–524 .

Jaffrezo J-L. Particules, polluants:de quoi parle-t-on ? [Online] 2018.

Jantunen , et al. , 2011. Promoting Actions for Healthy Indoor Air (IAIAQ). Euro- pean Commission Directorate General for Health and Consumers, Luxembourg, pp. 1–45 .

Kephalopoulos , et al. , 2012. European Commission’s effort s in harmonising the test- ing and health based evaluation of construction products emissions in relation to CE marking. In: Proceedings of the 2012 Healthy Buildings Conference, 1 . Kim, et al., 2015. Emissions of nanoparticles and gaseous material from 3D printer

operation. Environ. Sci. Technol. 49 (20), 12044–12053. doi: 10.1021/acs.est.

5b02805 .

Lunetto , et al. , 2018. Comments about the human health risks related to additive manufacturing. Sustain. Des. Manuf. 95–104 .

McDonnell , et al. , 2016. 3D printing in the wild: a preliminary investigation of air quality in college maker spaces. In: Proceedings of the 27th Annual Interna- tional Solid Freeform Fabrication Symposium, pp. 2456–2469 .

Mendes, et al., 2017. Characterization of emissions from a desktop 3D Printer. Jour- nal of Industrial Ecology 21, 94–106. doi: 10.1111/jiec.12569 .

Moylan, S.P. , et al. , 2012. Proposal for a standardized test artifact for additive man- ufacturing machines and processes. In: Proceedings of the Solid Freeform Fabri- cation Symposium, Austin, Texas, pp. 902–920 .

Parham, et al., 2017. Predicting concentrations of ultrafine particles and volatile organic compounds resulting from desktop 3D printer operation and the im- pact of potential control strategies. J. Ind. Ecol. 21 (1), 107–119. doi: 10.1111/jiec.

12578 .

Peters , et al. , 1997. Respiratory effects are associated with the number of ultrafine particles. A. J. Respir. Crit. Care Med. 155 (4), 1376–1383 .

Troxler, P , 2016. Fabrication Laboratories (Fab Labs). The Decentralized and Net- worked Future of Value Creation. Springer, pp. 109–127 .

Unwin, et al., 2013. Airborne emissions of carcinogens and respiratory sensitizers during thermal processing of plastics. Annals Occup. Hygiene 57 (3), 399–406.

doi: 10.1093/annhyg/mes078 .

William, F , 1992. Understanding Indoor Air Quality. Bradford O. Brooks .

Wojtyła, et al., 2017. Is 3D printing safe? Analysis of the thermal treatment of ther- moplastics: ABS, PLA, PET, and nylon. J. Occup. Environ. Hygiene 14 (6), 80–85.

doi: 10.1080/15459624.2017.1285489 .

Wojtyła , et al. , 2019. 3D printer as a potential source of indoor air pollution. Int. J.

Environ. Sci. Technol. 17, 207–218 .

World Health Organization, Regional Office for Europe, 2010. WHO Guidelines for Indoor Air Quality: Selected Pollutants. WHO .

Yun-Jung , et al. , 2016. Characterization and exposure assessment of odor emissions from laser cutting of plastics in the optical film industry. Aerosol. Air Qual. Res.

16 (9), 2216–2226 .