materials
Article Iron Carbide@Carbon Nanocomposites: A Tool Box of Functional Materials
Chiara Defilippi 1 , Mariam Omar Ali Mukadam 1, Sabina Alexandra Nicolae 2, Martin Richard Lees 3 and Cristina Giordano 1,*
1 School of Biological and Chemical Sciences, Chemistry Department, Queen Mary University of London, Mile End Road, London E1 4NS, UK; c.defi[email protected] (C.D.); [email protected] (M.O.A.M.) 2 School of Engineering and Materials Science, Queen Mary University of London, Mile End Road, London E1 4NS, UK; [email protected] 3 Department of Physics, University of Warwick, Coventry CV4 7AL, UK; [email protected] * Correspondence: [email protected]; Tel.: +44-(0)20-7882-6605
Received: 23 November 2018; Accepted: 8 January 2019; Published: 21 January 2019
Abstract: Iron carbide (Fe3C) is a ceramic magnetic material with high potential for applications in different fields, including catalysis, medicine imaging, coatings, and sensors. Despite its interesting properties, it is still somehow largely unexplored, probably due to challenging synthetic conditions. In this contribution, we present a sol-gel-based method that allows preparing different Fe3C@C nanocomposites with tailored properties for specific applications, in particular, we have focused on and discussed potential uses for adsorption of noxious gas and waste removal. Nanocomposites were prepared using readily available and “green” sources, such as urea, simple and complex sugars, and chitosan. The nanocomposite prepared from chitosan was found to be more efficient for CO2 uptake, while the sample synthetized from cellulose had optimal capability for dye absorption and waste oil removal from water.
Keywords: metallic ceramics; nanocomposites; urea glass route; iron carbide; chitosan
1. Introduction The powerful combination of iron and carbon to improve material performances has been known since antiquity. According to a legend, the art of the carburization of iron might have been developed by an Anatolian population called Chalybes, who could have also forged the famous sword of Julius Cesar. Mythology aside, it was discovered that metallurgy employing combinations of iron and carbon dates back even to prehistory when meteorites were used as the first sources of iron [1]. Beads found in tomb 67 at Gerzeh, which were made by using meteoritic iron as the metal source, represent one of the earliest examples of exploitation of iron in Egypt [2]. Even the superior hardness of Damascus sabre seems to arise from iron carbide nanoparticles reinforcing iron metal [3]. Nowadays, the best and most known representative of the iron/carbon winning combination is surely steel but, with the advent of nanoscience, the iron/carbon blend can be finely adjusted toward designed applications. In this way, besides a mere control of the elements ratio (as for the case of steel), we can prepare iron-based (nano)composites, as this study will also show, where carbon acts as functional matrix (contributing to a higher surface area, improved electron transport, and additional conductivity to the final system), while the iron-based phases represent the (nano)fillers, providing the final material with magnetic properties. Here, a whole tool box of possibility unfolds, where final composition, size, and morphology can be tailored for specific applications. The range of applications can be even broadened if, alongside pure iron, iron carbide is also considered. Iron carbide has high
Materials 2019, 12, 323; doi:10.3390/ma12020323 www.mdpi.com/journal/materials Materials 2019, 12, 323 2 of 15 Materials 2018, 11, x FOR PEER REVIEW 2 of 15 thermalharsh conditions and chemical are resistance,required an andd where, might alsousually, be used pure for iron applications or iron whereoxides harshdo not conditions survive. areFurthermore, required and in its where, most usually,simple cementite pure iron form, or iron it is oxides less toxic do not than survive. pure iron Furthermore, but, at the insame its mosttime, simplemore magnetic cementite than form, iron it is oxide, less toxic and than can, pure for ironinstance, but, atalso the be same considered time, more in magneticnanomedicine, than iron for oxide,example, and as can, contrast for instance, agent for also MRI be and considered drug deliver in nanomedicine,y [4]. For this for reason, example, toxicity/biocompatibility as contrast agent for MRIwas andstudied drug both delivery in vitro [4]. Forand thisin vivo reason, with toxicity/biocompatibility cytotoxicity essays, involving was studied several both cells,in vitro suchand as inmacrophage, vivo with cytotoxicity mouse fibroblast, essays, human involving breast several adenocarcinoma, cells, such as macrophage,and human ovarian mouse fibroblast,carcinoma. human These breasttests were adenocarcinoma, performed on and differ humanent iron ovarian carbides, carcinoma. such as These Fe7C tests3, Fe were5C2, and performed Fe3C, and on did different not show iron significant negative side effects [4,5]. Their magnetic properties make iron-based materials also easily carbides, such as Fe7C3, Fe5C2, and Fe3C, and did not show significant negative side effects [4,5]. Their magneticrecoverable properties after use, make simply iron-based by applying materials an extern alsoal easily magnetic recoverable field, thus after favouring use, simply their by reuse applying and anwaste external collection magnetic [6]. field, thus favouring their reuse and waste collection [6]. BesidesBesides biomedicalbiomedical applications,applications, iron iron carbide carbide can can be be used used as as sensors, sensors, coatings coatings [7 ],[7], and and catalysts catalysts in in processes such as Fischer–Tropsch, ammonia decomposition [8,9], and CO2 hydrogenation [10]. processes such as Fischer–Tropsch, ammonia decomposition [8,9], and CO2 hydrogenation [10]. ConsideringConsidering the crucial problems of environmentalenvironmental pollution from air and wastewater [[11,12],11,12], chemicalchemical sensingsensing and and pollutant pollutant absorbent absorbent are are surely surely important important for thefor detectionthe detection of noxious of noxious gases, gases, such such as CO2 [11]. Iron-based materials, such as iron oxides [12–14] are used as magnetic gas sensors as CO2 [11]. Iron-based materials, such as iron oxides [12–14] are used as magnetic gas sensors in differentin different applications applications [15 ].[15]. SuitableSuitable functionalizationfunctionalization of iron carbidecarbide nanoparticles can also be usedused forfor thethe preparationpreparation ofof novelnovel magneticmagnetic fluids.fluids. AnAn overalloverall representationrepresentation ofof ironiron carbidecarbide applicationsapplications isis reportedreportedin inScheme Scheme1 .1.
Scheme 1. Overview of the manifold application of iron carbide-based materials. Scheme 1. Overview of the manifold application of iron carbide-based materials. Despite its high potential, iron carbide-based materials are still not as broadly considered as the correspondingDespite its oxides high potential, or parental iron metal carbide-based [16]. This materials is probably are due still to not the as difficulties broadly considered related to as their the synthesis,corresponding such asoxides chemical or parental vapour metal deposition [16]. This (CVD)-assisted is probably pyrolysis due to the [7 ],difficulties and ammonolysis, related to which their requiresynthesis, high such temperatures, as chemical vapour rendering deposition iron carbide (CVD)-assisted production pyrolysis more expensive [7], and ammonolysis, but also hindering which controlrequire overhigh the temperatures, size and morphology rendering of iron the carbi finalde product production when preparedmore expensive at the nanoscale but also [hindering17,18]. controlIn thisover paper, the size we and present morphology how iron carbide@carbonof the final product nanocomposites when prepared can beat the prepared nanoscale and shaped[17,18]. to changeIn thethis final paper, material we present functionality, how iron adjusting carbide@ca its compositionrbon nanocomposites (e.g., incorporating can be prepared a useful and amount shaped of iron),to change size, andthe crystallinity,final material but functionality, also changing adjust propertiesing its such composition as magnetism, (e.g., surface incorporating area, absorption, a useful andamount more, of whichiron), size, in turn and result crystallinity, in a material but also with changing higher potential properties for such biomedical as magnetism, applications surface (e.g., area, in hyperthermiaabsorption, and treatment), more, which diagnostic in turn (e.g., result as a contrastin a material agent inwith MRI), higher absorption potential of toxicfor biomedical molecules, sensorapplications technology, (e.g., in catalytic hyperthermia efficiency, treatment), etc. Here, diagno we willstic (e.g., only as show a contrast a few agent model in examples, MRI), absorption but the applicationof toxic molecules, range can sensor be much technology, broader. catalytic In particular, efficiency, we etc. have Here, focused we will our only attention show ina few preparing model materialsexamples, with but higherthe application magnetization range and/or can be larger much surface broader. area forIn particular, future potential we have application focused in our the attention in preparing materials with higher magnetization and/or larger surface area for future detection of noxious gases, such as CO2, or absorption of pollutants such as inorganic dyes or scrap oilspotential from water,application keeping in the in minddetection that of an noxious effective gases, gas sensor/adsorbent such as CO2, or absorption should be of cost-effective pollutants such and highlyas inorganic sensitive, dyes to or detect scrap low oils concentrations. from water, keepin Here, nanocompositesg in mind that an with effective higher gas surface sensor/adsorbent area promote should be cost-effective and highly sensitive, to detect low concentrations. Here, nanocomposites with higher surface area promote contact surfaces to the analytes, and have increased sensitivity and gas adsorption compared to bulk samples [11,19].
Materials 2019, 12, 323 3 of 15 contact surfaces to the analytes, and have increased sensitivity and gas adsorption compared to bulk samples [11,19]. To overcome the synthetic difficulties, we have employed a sustainable route, based on a sol-gel process, to prepare the nanocomposite, that allows using lower reaction temperatures compared to classical synthesis, and to adjust morphology, particle dimension, and surface area of the final material simply by varying the carbon source and reaction conditions [18].
2. Materials and Methods
2.1. Iron Carbide Synthesis
Iron(III) acetylacetonate (1 g, Fe(acac)3, Sigma-Aldrich 97%, F300, Gillingham, UK) was used as iron source and mixed with approximately 2.5 mL of ethanol to obtain a clear solution. The solution was then mixed with a suitable carbon source to obtain the final iron carbide upon heat treatment. As carbon source, several compounds were explored, going from small molecules such as urea (CO(NH2)2, Sigma-Aldrich 99%, U5128), to sugars such as D-glucose (analytical reagent grade, Fisher Scientific,10391201, Loughborough, UK), D(+)-sucrose (analytical reagent grade, Fisher Scientific, 10647421), and chitosan (MW: 100,000–300,000, VWR, ACRO349050500) but also a polymer-based sugar such as cellulose (Sigmacell, Type 20, 20 µm Sigma-Aldrich, S3504). The cellulose was heat ◦ ◦ treated prior to its utilization (21 C/min ramping until 800 C under N2 flow, no dwelling) to remove possibly absorbed molecules on its surface. The carbon source strongly affects the texture of the carbon matrix, the composition of the final material but also its surface area and porosity. Specific synthetic details are reported in Table1, while Figure1 reports the chemical structure of the several carbon sources used.
Table 1. Experimental details of samples prepared using different carbon sources and heat-treated at 800 ◦C under nitrogen flow. R refers to the molar ratio between the carbon source and the metal-salt.
Average Crystallite Size (nm) Carbon Source Phases Present * Elemental Analysis (wt %) # by PXRD ** by TEM Fe C (76%) (Fe0 Urea R = 3 3 N/A 10–30 N% = 0.15, C% = 41.3 (24%)) Cellulose 1.2% Fe3C 36 30–80 N% = 0, C% = 87 Fe C (76%) (Fe0 Glucose R = 8 3 1724 (Fe0) 20–60 N% = 1, C% = 97.4 (24%)) Glucose R = 16 Fe3C 63 15–50 N% = 0, C% = 90 Fe C (78%) (Fe0 Sucrose R = 8 3 36 10–40 N/M (22%)) Fe C (72 (Fe0 Sucrose R = 10 3 81 10–45 N/M (28%)) Fe C (71%) (Fe0 Sucrose R = 12 3 8655 (Fe0) 10–50 N/M (29%)) Chitosan 0.42% Fe3C 73 10–25 N% = 2.3, C% = 82.7 Chitosan 0.21% Fe3C 60 6–30 N/M Chitosan 0.42% Fe C 46 10–40 N% = 6, C% = 94 slower heating 3 *: For the samples including Fe0, the percentages of the two phases, calculated using the software Match!®, are reported between brackets. **: Calculated on the main peak of the Fe3C PXRD pattern using the Scherrer equation. 0 For Fe3C phase, calculation refers to the peak with d value of 2.06682 Å that corresponds to plane (102). For Fe phase, calculation refers to the peak with d value of 2.02671 Å that corresponds to plane (110). #:Theoretical value C wt %: 22%. N/M not measured. Materials 2019, 12, 323 4 of 15 Materials 2018, 11, x FOR PEER REVIEW 4 of 15 Materials 2018, 11, x FOR PEER REVIEW 4 of 15
Figure 1. Molecular structures of the used carbon sources. Figure 1. Molecular structures of the used carbon sources. For chitosan and celluloseFigure 1. samples, Molecular the structures percentage of the reported used carbon refers sources. to wt % between iron precursors and carbonThe heat source. treatment (up to 800 °C for 2 h under nitrogen flow) leads to a black (matt or shiny dependingThe heat on treatment the carbon (up (up source) to 800 powder °C◦C for for made2 2 h h under of crystalline nitrogen Feflow) flow)3C and leads carbon, to to a black as ascertained (matt or shiny by X- 0 ray investigation. In some cases, the presence of Fe was also observed3 (see PXRD patterns vide infra). depending on thethe carboncarbon source)source) powder powder made made of of crystalline crystalline Fe Fe3CC and and carbon, carbon, as as ascertained ascertained by by X-ray X- rayinvestigation.A representation investigation. In some Inof somethe cases, procedure cases, the the presence is presence reported of Feof in0 Fewas Scheme0 was also also 2. observed observed (see (see PXRD PXRD patterns patterns vide vide infra). infra). A Arepresentation representation of of the the procedure procedure is reportedis reported in in Scheme Scheme2. 2.
Scheme 2. Overview of the synthesis of Fe3C samples. Scheme 2. Overview of the synthesis of Fe3C samples. Scheme 2. Overview of the synthesis of Fe3C samples. 2.2.Dyes Dyes and and Oil Oil Uptake Uptake Procedure Procedure Dyes and Oil Uptake Procedure MethylMethyl orangeorange (Sigma-Aldrich,(Sigma-Aldrich, ACS reagent,reagent, dyedye contentcontent 85%,85%, 114,510)114,510) andand methylenemethylene blueblue (Sigma-Aldrich,(Sigma-Aldrich,Methyl orange ≥≥97.0%,97.0%, (Sigma-Aldrich, 66,720) 66,720) solution solution ACS were reagent, were prepared prepared dye bycontent dissolving by dissolving85%, the114,510) powder the and powder of methylenethe dye of thein water blue dye (Sigma-Aldrich,into waterachieve to the achieve ≥desired97.0%, the 66,720)final desired concentration. solution final concentration. were preparedSolutions by Solutionswere dissolving kept were in the the kept powder da inrk the ofto the darkreduce dye to in natural reduce water tonaturalphotodegradation. achieve photodegradation. the desired final concentration. Solutions were kept in the dark to reduce natural photodegradation.AsAs preparedprepared samples samples powderspowders (around 5 mg)mg) werewere mixedmixed withwith thethe dyedye solutionssolutions (5(5 mL)mL) andand UV–VisUV–VisAs measurementspreparedmeasurements samples werewere powders performedperformed (around over over time 5time mg) to to we measure measurere mixed the the with dye dye concentration. theconcentration. dye solutions (5 mL) and UV–VisVegetable measurements oil (Tesco®, were pure performed sunflower over oil) time was to added measure to water the dye containing concentration. the Fe3 C-based sample Vegetable oil (Tesco®, pure sunflower oil) was added to water containing the Fe3C-based sample andand theVegetablethe affinityaffinity oil ofof (Tesco®, thethe compoundcompound pure sunflower waswas recordedrecorded oil) was byby visualaddedvisual inspection. inspection.to water containing the Fe3C-based sample and the affinity of the compound was recorded by visual inspection. 2.3.Techniques Techniques Techniques ToTo evaluateevaluate thethe crystallinitycrystallinity ofof thethe samplessamples andand thethe phasesphases present,present, X-rayX-ray powderpowder diffractiondiffraction (PXRD)(PXRD)To evaluate measurementsmeasurements the crystallinity werewere performedperformed of the samples onon XPERTXPER anTd PanalyticalPanalyticalthe phases EmpyreanEmpyreanpresent, X-ray DiffractometerDiffractometer powder diffraction (Cu(Cu KKαα (PXRD)radiation,radiation, measurements λλ= = 0.1540.154 nm,nm, atwereat 4545kV kVperformed and 40mA, 40 mA, on Malvern MalvernXPERT Panalytical Panalytical Ltd,Empyrean Royston, Diffractometer UK) equipped (Cu with K ananα radiation,X’CeleratorX'Celerator λ = detector.detector. 0.154 nm, at 45kV and 40mA, Malvern Panalytical Ltd, Royston, UK) equipped with an X'Celerator detector.
Materials 2019, 12, 323 5 of 15 Materials 2018, 11, x FOR PEER REVIEW 5 of 15
To obtainobtain informationinformation about the morphology and large-scale homogeneity of the samples, scanning electron microscopy (SEM, Thermo Fisher Scientific,Scientific, Hillsboro, USA) was performed on a FEI inspect F instrument. The samples were loaded on carbon-coated stubs and coated by sputtering an Au prior to imaging. To observeobserve thethe sizesize andand shapeshape ofof thethe particles,particles, transmissiontransmission electron microscopy (TEM, JEOL (U.K.) Limited, London, UK) was performed on a JEOL 2010 operatedoperated at anan accelerationacceleration voltage of 200 kV. SamplesSamples werewere ground ground and and then then suspended suspended in in ethanol. ethanol. One One drop drop of of this this suspension suspension was was put put on aon holey a holey carbon-coated carbon-coated copper copper grid grid of300 of 300 mesh mesh and and left left to dry. to dry. Elemental analysis was conducted using a Thermo ScientificScientific FLASH 2000 Elemental Analyzer (Thermo Fisher Scientific, Scientific, Hillsboro, Hillsboro, USA). USA). Information Information about about the the surface surface area/CO area/CO2 adsorption2 adsorption of the of thematerial material were were obtained obtained using using Quantachrome Quantachrome Nova Nova 4200e 4200e (Anton (Anton Paar Paar GmbH, GmbH, Graz, Austria). NN22 adsorption-desorption isot isothermsherms were measured at −−196196 °C.◦C. The The samples samples were degassed at 150 ◦°CC under vacuum overnight, overnight, prior prior to to the the adsorption. adsorption. BE BETT theory theory was was used used to calculate to calculate surface surface area. area. For ◦ Forthe theCO2 CO adsorption2 adsorption experiments, experiments, thethe samples samples were were degassed degassed at at150 150 °CC under under vacuum vacuum overnight. overnight. Measurements were performedperformed atat 11 barbar andand 00 ◦°C.C. Magnetic (M vs. H) measurements measurements were perfor performedmed using a Quantum Design Magnetic Property Measurement System (MPMS, QuantumQuantum Design, Inc., San Diego, USA) squid magnetometer. A small quantity of each sample was placed in a gel capsule and this capsule was then placed in a plastic drinking straw. straw. M M vs. vs. H data H data were were collected collected swee sweepingping the magnetic the magnetic field from field +50 from kOe +50 (5 T) kOe to − (550 T)kOe, to −and50 then kOe, back and thento +50 back kOe to at +50 T = kOe 300 atK.T Ultraviolet-visible= 300 K. Ultraviolet-visible (UV–Vis, Agilent (UV–Vis, Technologies Agilent Technologies LDA UK LDALimited, UK Limited,Stockport, Stockport, UK) spectroscopy UK) spectroscopy measurem measurementsents were were performed performed using using Cary Cary 100 100 Agilent spectrophotometer to measure thethe absorbanceabsorbance ofof thethe dyedye solutions.solutions.
3. Results and Discussion The crystal structure of the finalfinal material wa wass ascertained by PXRD study, using ICDD (PDF4+ version) database. Using urea as carbon source, crystalline iron carbide was obtained (Figure 2 2,, upperupper pattern) alongside a small amount of amorphousamorphous carbon (*(* marked peak)peak) and Fe00. On On the the other other hand, hand, when cellulose was used as C source, the peak of carbon is more defineddefined and prominent,prominent, indicating the formationformation of of a quasi-crystallinea quasi-crystalline (possibly (possibly turbostratic) turbostratic) phase, phase, alongside alongside the pattern the pattern of iron carbide.of iron Incarbide. this case, In this the case, presence the presence of metallic of ironmetallic was notiron observed. was not observed. For the cellulose For the sample,cellulose the sample, increased the backgroundincreased background at lower angles at lower could angles be due could to thebe due presence to the of presence amorphous of amorphous carbon. carbon.
Figure 2. PXRD patterns of samples prepared using urea (R = 3, upper pattern) and cellulose (1.2% 0 downer pattern). pattern). The The expected expected patte patternrn from from the thedatabase database for pure for pureFe3C and Fe3C Fe and0 are Fe reportedare reported as vertical as 0 verticallines for lines comparison for comparison (Fe3C, (Feblack3C, blacklines, lines,ICDD: ICDD: 03-065-2413 03-065-2413 and and Fe0 Fe redred lines, lines, ICDD: ICDD: 01-087-0721. 01-087-0721. Marked peak (*) is attributed to carbon (ICDD 01-077-7164).
The presence of the carbon matrix seems to affect,affect, also, the final final size of the iron carbide nanoparticles, which appeared to be bigger than in the presence of urea (as ascertained by Scherrer calculation ofof the the mean mean crystallite crystallite size size and and also observedalso observed by TEM by investigation,TEM investigation, Figure3 ).Figure The presence 3). The presence of the carbon matrix drastically increases the surface area of the sample prepared with cellulose, which was determined to be above 200 m2/g, against the lesser value of 50 m2/g for the
Materials 2019, 12, 323 6 of 15
Materials 2018, 11, x FOR PEER REVIEW 6 of 15 of the carbon matrix drastically increases the surface area of the sample prepared with cellulose, which 2 2 wassample determined prepared to using be above urea 200 (see m Table/g, against 2). On the the lesser other value hand, of the 50 mmagnet/g forization the sample of the prepared sample usingprepared urea via (see urea Table was2). much On the higher other than hand, that the obse magnetizationrved for the ofcellulose-derived the sample prepared sample via (100 urea emu/g was muchvs. <20 higher emu/g, than see that Table observed S1), which for the means cellulose-derived a better suitability sample (100of the emu/g former vs. for <20 application emu/g, see where Table S1),magnetization which means plays a better the main suitability role, while of the cellulose-de former for applicationrived samples where are more magnetization suitable for plays applications the main role,where while surface cellulose-derived area is crucial. samples are more suitable for applications where surface area is crucial.
(A) (B)
Figure 3.3. TEMTEM imagesimages ofof samplessamples preparedprepared withwith ((AA)) ureaurea RR == 33 andand ((BB)) cellulose.cellulose. ParticleParticle dimensiondimension histograms areare reportedreported inin FigureFigure S1.S1.
Table 2. Surface area and CO2 adsorption (CO2 ads) values for the samples prepared. Table 2. Surface area and CO2 adsorption (CO2 ads) values for the samples prepared. 2 Carbon Source Surface Area (m /g) Max CO2 ads (mmol/g) Carbon Source Surface Area (m2/g) Max CO2 ads (mmol/g) UreaUreaR = 3 R = 3 43 43 0.58 0.58 Cellulose 1.2% 211 0.98 Cellulose 1.2% 211 0.98 Glucose R = 8 239 1.11 ChitosanGlucose 0.42% R = 8 131 239 1.11 1.37 ChitosanChitosan 0.21% 0.42% 55 131 1.37 1.76 Chitosan 0.42%Chitosan slower heat 0.21% ramping 194 55 1.76 1.2 Chitosan 0.42% slower heat ramping 194 1.2 In the case of urea and cellulose, from the TEM images, it can be seen that the inorganic nanoparticlesIn the case (metallic of urea in and nature, cellulose, i.e., with from higher the TEM electron images, density it can and be therefore seen that showing the inorganic higher contrast)nanoparticles are distributed (metallic in on nature, a matrix, i.e., possiblywith high turbostraticer electron carbon,density inand line therefore with PXRD showing observation higher (broadcontrast) carbon are distributed peak around on 26 a◦ matrix,2θ). These possibly results turb areostratic similar carbon, to what in reported line with previously, PXRD observation where the catalytic(broad carbon graphitization peak around of carbon 26° 2θ by). These iron was results also are observed similar [ 20to]. what reported previously, where the catalyticRegarding graphitization the SEM of study, carbon samples by iron prepared was also usingobserved urea [20]. and cellulose as carbon source form a quiteRegarding compact structure, the SEM wherestudy, “wrapped”samples prepared structures using can urea be alsoand observedcellulose as (Figure carbon S2). source These form rods a couldquite compact be due to structure, thin sheets where of iron/carbon, “wrapped” which structures wrap can into be rods, also as observed previously (Figure observed S2). These [9]. rods couldIn be order due toto investigatethin sheets theof iron/carbon, role of the carbon which source wrap ininto the rods, formation as previously of a carbon observed functional [9]. matrix, celluloseIn order was replacedto investigate with simplethe role sugar of the molecules, carbon source such as in glucose the formation and sucrose. of a carbon functional matrix,A similar cellulose trend was to replaced the cellulose-derived with simple sugar samples molecules, was observed such as when glucose glucose and wassucrose. used as carbon source.A similar Also in trend this case, to the both cellulose-derived iron carbide and sample carbons was peaks observed are observed when (Figureglucose4 was). From used Figure as carbon4, it cansource. be seen Also that in this for highercase, bothR, the iron carbon carbide peak and is ca lessrbon intense peaks and are broader observed while, (Figure for lower4). From urea/metal Figure 4, ratioit can ( Rbe), seen residual that Fefor0 higheris observed. R, the carbon peak is less intense and broader while, for lower urea/metal ratio (R), residual Fe0 is observed. Although glucose is a reducing sugar, a higher amount (i.e., higher R) hinders Fe0 formation, possibly towards the formation of ‘reduced’ phases.
Materials 2019, 12, 323 7 of 15 Materials 2018, 11, x FOR PEER REVIEW 7 of 15 Materials 2018, 11, x FOR PEER REVIEW 7 of 15
Figure 4.4. PXRDPXRD pattern pattern of samples prepared using glucose as carbon source with two different 0 glucose/FeFigure 4. PXRD molar molar patternratios ratios ( (RRof).). samplesThe The expected expected prepared pattern pattern using from glucose the database as carbon for purepuresource FeFe 33withC and two Fe 0different are also 00 reportedglucose/Fe asas molar verticalvertical ratios lineslines for(R). comparison The expected (Fe(Fe 3pattern3C, black from lines, the ICDD: database 03-065-2413 for pure and Fe Fe3C andredred lines, lines,Fe0 are ICDD: ICDD: also 01-087-0721.reported as vertical The marked lines for peak comparison (*) is attributedattributed (Fe3C, blackto carbon lines, (ICDD(ICDD ICDD: 01-077-7164).01-077-7164). 03-065-2413 and Fe0 red lines, ICDD: 01-087-0721. The marked peak (*) is attributed to carbon (ICDD 01-077-7164). 0 AlthoughThe presence glucose of the is carbon a reducing matrix sugar, imparts, a higher to the amountglucose-derived (i.e., higher samples,R) hinders a higher Fe surfaceformation, area possibly(>200The m2/g, towardspresence see Table the of the formation2). carbon ofmatrix ‘reduced’ imparts, phases. to the glucose-derived samples, a higher surface area (>200TheTEM m2/g, presence investigation see Table of the2). carbonrevealed, matrix for both imparts, ratios to the(R glucose-derived= 8 and 16), the samples, formation a higher of medium-sized surface area 2 (>200nanoparticlesTEM m /g, investigation see (20–60 Table nm2). revealed,on average), for mostlyboth ratios aggl omerated(R = 8 and and 16), ill-shaped, the formation dispersed of medium-sized in the carbon matrix.nanoparticlesTEM As investigation expected (20–60 nmfrom revealed,on the average), PXRD for patternmostly both ratios aggl(seeomerated (FigureR = 8 and4), and for 16), ill-shaped, R the= 16, formation nanoparticles dispersed of medium-sized in are the slightlycarbon nanoparticlessmallermatrix. Asthan expected for (20–60 R = nm8,from but on the the average), PXRDpresence mostlypattern of bigger agglomerated(see Figureand ill- 4),shaped and for ill-shaped,R aggregates = 16, nanoparticles dispersed can be also in are the seen slightly carbon (see matrix.Figuresmaller 5). Asthan Interestingly, expected for R = from 8, butthe the thelower PXRD presence degree pattern ofof (see biggercrystallization Figure and4), ill- forshaped forR = the 16, aggregatescarbon nanoparticles phase can arein be R slightly also= 16 seenseems smaller (see to thanalsoFigure affect for 5).R Interestingly,=the 8, nanoparticles’ but the presence the lower morphology, of degree bigger andofpossib crystallization ill-shapedly due to aggregates afor lack the of carbon ‘protection’. can be phase also seenInin fact,R (see= 16for Figureseems R = 8,5 to ).a Interestingly,graphiticalso affect carbon the the nanoparticles’ shell lower around degree morphology,the of crystallizationparticles was possib observed, forly the due carbon towhich a lack phase might of in‘protection’. actR = as 16 a seemsprotective In tofact, also coatingfor affect R = (see8, the a nanoparticles’insetgraphitic Figure carbon 5A). morphology, shell around possibly the particles due to was a lack observed, of ‘protection’. which might In fact, act for asR a =protective 8, a graphitic coating carbon (see shellinset aroundFigure 5A). the particles was observed, which might act as a protective coating (see inset Figure5A).
(A) (B) (C) (A) (B) (C) Figure 5. TEM images of samples prepared using glucose as carbon source with different glucose/iron Figuremolar ratio 5. TEM (A) images and (B )of R samples = 8 and (prepared C) R = 16 using. Particle glucose dimension as carbon histograms source withwith are differentreporteddifferent glucose/iron inglucose/iron Figure S1. molar ratio (A)) andand ((B)) R = 8 and ( C) R = 16 16.. Particle dimension histograms are reported in Figure S1. SEM images of the glucose-derived samples are reported in Figure S3, where a rod-like structure can beSEM observed images (similar of the glucose-derived to those observed samples for the are case reportedrepo of rtedthe urea-derived in Figure S3, wheresample). a rod-like Those “needles” structure canare absent be observed in the (similarsample prepared to those observed with higher for the urea/metal case of the ratio urea-derived (R = 16). sample). Those Those “needles” “needles” are absentInterestingly, in the samplesample a completely preparedprepared different withwith higherhigher trendurea/metal urea/metalis observed ratio when ( R replacing = 16). 16). glucose with sucrose. The use ofInterestingly, sucrose as carbon aa completelycompletely source differentdifferentallows for trendtrend the isformationis observedobserved of whenwhen iron carbide replacingreplacing only glucoseglucose at a lower withwith sucrose.sucrose.sugar/metal The usemolar of sucroseratio, although asas carboncarbon carbon sourcesource and allowsallows a significant forfor thethe formationamountformation of ofofiron ironiron zero carbidecarbide were onlyformed,only atat aa as lowerlower ascertained sugar/metalsugar/metal in all molarcases by ratio, PXRD although although investigation carbon carbon and(see and aFigure asignificant significant 6). The amount amountresidual of of metaliron iron zero iron zero were increases were formed, formed, with as higher asascertained ascertained R (i.e., in the all in allinitialcases cases by sugar byPXRDPXRD amount) investigation investigation to the detriment (see (see Figure Figure of 6).the 6The ).carbide The residual residual phase, metal which metal iron irondisappears increases increases withcompletely with higher higher forR (i.e., RR ≥(i.e., the16. theinitialAlongside initial sugar sugarwith amount) increasing amount) to the to Rdetriment the, the detriment peak of ofthe carbon of carbide the carbidealso phase, reduces phase, which in whichdisappears intensity disappears and completely becomes completely for broader, R ≥ for16. ≥ RindicatingAlongside16. Alongside thatwith the increasing with formation increasing R ,of the theR peak, iron the peak ofphase carbon of has carbon alsonot also supportedreduces reduces in the inintensity intensitygraphitization and and becomes becomes of carbon. broader, broader, The indicatingreasonindicating behind that this thethe formationformationtrend requires of thethe a dedicated ironiron phasephase stud hashasy notnot and supportedsupported specific equipment thethe graphitizationgraphitization (e.g., EXAFS ofof carbon.carbon. for in Thesitu reasonmeasurements), behind this and trend this requiresrequiresis in progress. aa dedicateddedicated In any studystud case,y and SEM specificspecific study equipmentequipment reveals the (e.g., homogeneity EXAFS for inof situthe samplesmeasurements), (see Figure and S4). this is in progress. In any case, SEM study reveals the homogeneity of the samples (see Figure S4).
Materials 2019, 12, 323 8 of 15 measurements), and this is in progress. In any case, SEM study reveals the homogeneity of the samples Materials 2018, 11, x FOR PEER REVIEW 8 of 15 (seeMaterials Figure 2018, S4). 11, x FOR PEER REVIEW 8 of 15
Figure 6. PXRDPXRD pattern pattern of of samples samples prepared prepared with sucrose at different molar ratios ( R).). The expected Figure 6. PXRD pattern of samples prepared with0 sucrose at different molar ratios (R). The expected pattern from the database for pure Fe3CC and Fe 0 areare also also reported as vertical lines for comparison 3 0 pattern from the database for pure Fe3C and0 Fe are also reported as vertical lines for comparison (Fe(Fe33C,C, red red lines, lines, ICDD: 03-065-2413, and Fe 0 blackblack lines, lines, ICDD: ICDD: 01-087-0721. The The marked marked peak peak (*) (*) is (Fe3C, red lines, ICDD: 03-065-2413, and Fe0 black lines, ICDD: 01-087-0721. The marked peak (*) is attributedattributed to carbon (ICDD 01-077-7164). attributed to carbon (ICDD 01-077-7164). From TEM analysis (Figure 77)) thethe particlesparticles ofof samplesample RR == 8 8 (Figure (Figure 7A)7A) are mostly agglomerated, From TEM analysis (Figure 7) the particles of sample R = 8 (Figure 7A) are mostly agglomerated, while, byby increasingincreasing R, R, the the particles particles become become smaller smaller and lessand aggregated,less aggregated, for all ratios,for all andratios, the particlesand the while, by increasing R, the particles become smaller and less aggregated, for all ratios, and the particlesare surrounded are surrounded by the carbon by the matrix. carbon matrix. particles are surrounded by the carbon matrix.
(A) (B) (C) (A) (B) (C) Figure 7. TEMTEM images images of samples prepared with sucrose (A)) RR == 8, (B) R = 10, and (C) R = 12. Particle Particle Figure 7. TEM images of samples prepared with sucrose (A) R = 8, (B) R = 10, and (C) R = 12. Particle dimension histograms are reported in Figure S1. dimension histograms are reported in Figure S1. InIn order order to to understand whether whether the the initial initial complexation complexation of of the the iron iron source source may may play play a a role, role, In order to understand whether the initial complexation of the iron source may play a role, chitosan was was also explored as a carbon source. Thanks Thanks to to the the presence of NH 2 groups,groups, a a stronger chitosan was also explored as a carbon3+ source. Thanks to the presence of NH2 groups, a stronger bonding between chitosan and Fe3+ waswas expected expected [21]. [21]. Furthermore, Furthermore, chitosan chitosan decomposes decomposes at at lower lower bonding between chitosan and Fe3+ was expected [21]. Furthermore, chitosan decomposes at lower temperaturestemperatures compared to cellulose[22]. cellulose [22]. temperatures compared to cellulose[22]. 0 As expected, all samples prepared with chitosan form pure Fe 3C,C, and and no Fe 0 waswas observed, observed, As expected, all samples prepared with chitosan form pure Fe3C, and no Fe0 was observed, although the chitosan/iron molar molar ratio ratio had had no no influence influence on on the the final final composition composition of of the sample, as although the chitosan/iron molar ratio had no influence on the final composition of the sample, as well as the heating rate. PXRD patterns of the chitosan samples are reported in Figure8 8.. well as the heating rate. PXRD patterns of the chitosan samples are reported in Figure 8.
Materials 2019, 12, 323 9 of 15 Materials 2018, 11, x FOR PEER REVIEW 9 of 15
Figure 8. PXRD pattern of samples prepared with chitosan.chitosan. The expected pattern from the database
for pure Fe 3CC is is also reported as vertic verticalal lines for comparison (Fe 3C,C, black black lines, lines, ICDD: ICDD: 03-065-2413). 03-065-2413). Marked peak (*) is attributed to carbon (ICDD 01-077-7164).
On the other hand, the sample’s lack of homogeneity and carbon phase phase looks looks less less defined, defined, compared to the cases described previously. Fe3C nanoparticles (darker spot in the TEM images) are smaller and spheroidal (Figure9 9).). SEM images of chitosan samples are reported in Figure S5.
Figure 9. TEM images(a) of samples prepared with(b)different amounts of chitosan( (cA) ) 0.42%, (B) 0.21%, and (C) 0.42%, slower heat ramping. Particle dimension histograms are reported in Figure S1. Figure 9. TEM images of samples prepared with different amounts of chitosan (A) 0.42%, (B) 0.21%,
andSEM (C images) 0.42%, ofslower chitosan heat samplesramping. areParticle reported dimens inion Figure histograms S5. are reported in Figure S1.
N3.1.2 and N2 COand2 COadsorption2 Adsorption test Test
N2 adsorptionadsorption measurements measurements were were conducted conducted to to me measureasure the surface area of the samples, while CO2 adsorptionadsorption tests tests were were performed to study the adsorption potential of our materials towards sensors applications. The adsorption capacity ( Ca) was calculated using the formula reported below:
Volume of gas CO2 adsorbed cc/g Volume of gas (CO2) adsorbed (cc/g) Ca = cc (1)(1) Molar volume cc of gas CO2 Molar volume mmol of gas (CO2) where the volume of gas adsorbed corresponds to the y-axis and, for the adsorption capacity, capacity, the maximum value is considered, and molar volume is the volume occupied by one mole of a substance (chemical element or chemical compound) at a gi givenven temperature and pressure 1 atm. The obtained values are reported in millimoles of adsorbed gas per gram of the adsorbent.adsorbent. The molar volume of CO2 at standard conditions (273 K and 1 bar) is equal to 22.4 L/mol [23]. The results are reported in CO2 at standard conditions (273 K and 1 bar) is equal to 22.4 L/mol [23]. The results are reported in Table2 2 and and Figure Figure 10 10.. Figure 10 shows that chitosan samples possess higher adsorption ability for CO2. From the comparison of chitosan-derived samples with different chitosan amount, we can safely say that the better adsorbance is not just a question of surface area, with chitosan 0.42% samples having a higher surface area than chitosan 0.21%.
1
Materials 2019, 12, 323 10 of 15 Materials 2018, 11, x FOR PEER REVIEW 10 of 15
FigureFigure 10. 10.CO CO22 uptakeuptake vs.vs. pp//p00 forfor different different samples. samples.
FigureBy comparing 10 shows the that results chitosan achieved samples from possess CO2 higheradsorption adsorption with the ability results for reported CO2. From in thethe comparisonliterature (see of Table chitosan-derived 3), it can be said samples that our with mate differentrials nicely chitosan compete amount, withwe those, can with safely the say advantage that the betterof being adsorbance magnetic is and, not justtherefore, a question easily of recoverable, surface area, but with also chitosan having 0.42% potential samples in magnetic having a highersensor surfaceapplications. area than chitosan 0.21%. By comparing the results achieved from CO2 adsorption with the results reported in the literature
(see Table3), it canTable be 3. said Reported that ourvalues materials of CO2 adsorbed nicely compete (ads/mmol/g) with those,for different with thematerials. advantage of being magnetic and, therefore, easily recoverable, but also having potential in magnetic sensor applications. Max CO2 ads (mmol/g) (Temperature of Surface Area Material Reference Analysis) (m2/g) Table 3. Reported values of CO2 adsorbed (ads/mmol/g) for different materials. AMC* 4.66 (273 K) 2475 Max CO2 ads (mmol/g) 2 CeO2–AMCMaterial 3.13 (273 K) Surface Area (m /g) 1906Reference (Temperature of Analysis) [24] CuO–AMCAMC * 4.66 2.40 (273 (273 K) K) 2475 2097 CeO –AMC 3.13 (273 K) 1906 NiO–AMC 2 3.30 (273 K) 2006 [24] CuO–AMC 2.40 (273 K) 2097 Mesoporous MgO 1.82 (not reported) Not reported NiO–AMC 3.30 (273 K) 2006 [25] Non-porousMesoporous MgO MgO 1.82 0.45 (not (not reported) reported) Not reported Not reported [25] AC Non-porous** MgO 0.45 (not 2.61 reported) (298 K) Not reported Not reported [26] AC ** 2.61 (298 K) Not reported CNTs 1.08 (323 K) Not reported[26] CNTs 1.08 (323 K) Not reported MgAlO 1.13 (623 K) 161 MgAlO 1.13 (623 K) 161 [27] [27] CaAlO CaAlO 0.87 0.87 (623 (623 K) K) 26 26 N–rGO–ZnON–rGO–ZnO 3.55 3.55 (298 (298 K) K) 1122 1122 [28] [28] MOF–MgMOF–Mg *** *** 1.1 1.1 (298 (298 K) K) Not reported Not reported MOF–CuMOF–Cu 0.48 0.48 (298 (298 K) K) Not reported Not reported[29] [29] MOF–Fe 1.1 (298 K) Not reported MOF–Fe 1.1 (298 K) Not reported ZnO–nanoporous carbon 4.42 (273 K) 706 [30] ZnO–nanoporous MCM-41 0.62 (298 K) 1129 4.42 (273 K) 706 [31] [30] carbonMeso Al2O3 0.84 (298 K) 271 MCM-41*: AMC = activated mesoporous carbon. **: 0.62 AC (298 = activated K) carbon. ***: MOF = metal organic 1129 frameworks. [31] Meso Al2O3 0.84 (298 K) 271 *: ToAMC observe = activated whether mesoporous the carbon carbon. source **: hasAC an= ac influencetivated carbon. on the ***: magnetic MOF = metal behavior organic of the frameworks. as-prepared samples, magnetometry measurements were performed at 300 K, and the results reported in Figure 11 and TableTo observe S1. Different whether magnetization the carbon source behaviors has an are in observedfluence on for the samples magnetic prepared behavior with of varying the as- carbonprepared source, samples, as well magnetometry as metal/C-source measurements ratio which, were in performed turn, means at that 300 theK, and different the results materials reported have thein Figure potential 11 and to beTable used S1. in Different a range magnetization of different applications. behaviors are The observed magnetization for samples of all prepared the samples with studiedvarying wascarbon saturated source, in appliedas well fieldsas metal/C-source above ~10 kOe. ratio The which, saturation in magnetization,turn, means thatσM, the varies different from 6materials emu/g for have chitosan the potential 0.42% up to to be 100 used emu/g in a forrange urea ofR different= 3 sample. applications. All samples The exhibit magnetization some magnetic of all hysteresisthe samples (see studied inset of was Figure saturated 11 and in Table applied S1) with fields coercive above fields~10 kOe. ranging The saturation from 87 up magnetization, to 400 Oe. σM, varies from 6 emu/g for chitosan 0.42% up to 100 emu/g for urea R = 3 sample. All samples exhibit some magnetic hysteresis (see inset of Figure 11 and Table S1) with coercive fields ranging from 87 up to 400 Oe.
Materials 2018, 11, x FOR PEER REVIEW 11 of 15 Materials 2019, 12, 323 11 of 15
100 20
50 0
-20 -3 0 3 0
Chitosan 0.42% (emu/g) Chitosan 0.21% M -50 Chitosan 0.42% lower ramping Cellulose 1.2% Glucose R=8 Urea R=3 -100 T = 300 K
-60 -40 -20 0 20 40 60 H (kOe)
Figure 11. M vs. H curves collected at T = 300 K for the different samples. Figure 11. M vs. H curves collected at T = 300 K for the different samples. It must be noted that magnetization and magnetic hysteresis are dependent on the amount of carbon present in the sample, and all samples reported here have a comparable magnetic It must be noted that magnetization and magnetic hysteresis are dependent on the amount of material/carbon ratio by mass (the only exception is the urea sample, with the lowest carbon amount) carbon present in the sample, and all samples reported here have a comparable magnetic as reported in Table1. material/carbon ratio by mass (the only exception is the urea sample, with the lowest carbon amount) 3.2.as reported Dyes and in Oil Table Uptake 1.
DyesFe and3C-based oil uptake nanocomposites were tested for possible applications as sorbent for contaminants in wastewater. Samples were treated with two different dyes as model molecules, i.e., methyl orange Fe3C-based nanocomposites were tested for possible applications as sorbent for contaminants in (MO) and methylene blue (MB). MO and MB were selected as model systems since they are well-known wastewater. Samples were treated with two different dyes as model molecules, i.e., methyl orange materials and can be easily quantified via UV-Vis spectroscopy. (MO) and methylene blue (MB). MO and MB were selected as model systems since they are well- Furthermore, MB is a cationic dye while MO is an anionic dye. This ensures that the filtration is notknown solely materials based on and electrostatic can be easily interaction. quantified via UV-Vis spectroscopy. TheFurthermore, solutions initially MB is a testedcationic were dye 50 while mg/L MO both is an for anionic MO and dye. MB This and, ensures after 72 that h, a the clear filtration solution is wasnot obtainedsolely based for cellulose on electrostatic sample ininteraction. the case of MO, with a decrease in colour also for MB; such distinct differencesThe solutions were not initially observed tested for were the other 50 mg/L samples both tested.for MO Theand picturesMB and, are after reported 72 h, a clear in Figure solution 12. Fewas3C synthesizedobtained for from cellulose cellulose sample was thenin the believed case of to MO, be the with most a decrease promising in candidate colour also for for dye MB; removal such fromdistinct water. differences were not observed for the other samples tested. The pictures are reported in Figure 12. Fe3C synthesized from cellulose was then believed to be the most promising candidate for dye removal from water.
Materials 2019, 12, 323 12 of 15 Materials 2018, 11, x FOR PEER REVIEW 12 of 15
(A) (B)
(C) (D)
Figure 12. Dye sorption on FeFe33C samples over time, with a concentrationconcentration ofof 5050 mg/Lmg/L both for MO (A,,B) and MB ((C,,D).).
Clearly, alsoalso inin this case, surface area does not play the major role, and another factor should be responsible for the uptake. As an example, the concentration over time calculated from UV–Vis analysis for MO is reported in Figure S6, and a decrease in the concentration concentration of MO was observed for the first first few hours, followed by the release of small quantityquantity of dye overover time.time. To calculatecalculate thethe concentrationconcentration of MO withinwithin thethe samesame conditions,conditions, a calibration curve was constructed, and the extinction coefficientcoefficient (ε) waswas calculated.calculated. The calibration curve details can be −4 × -4 found inin FigureFigure S7.S7. TheThe initialinitial concentrationconcentration of of the the solution solution without without Fe Fe3C3C is is 7.64 7.64 ×10 10 M,M, so almost half of the MO was immediatelyimmediately absorbed in the material.material. The recyclability of the compound was tested twice, and the sample was washed several times with water and ethanol to remove all the absorbed dyedye and dried. In all cases, a large amount of dye was released. Vegetable oil was also tested as a realreal exampleexample ofof waterwater contamination since it is commonly poured intointo the the sink sink when when used used for for cooking. cooking. Firstly, Firs thetly,sample the sample was mixed was mixed with water with andwater the and powder the almostpowder immediately almost immediately deposited deposited on the bottomon the bottom of the vial,of the with vial, the with addition the addition of the of oil, the and oil, the and Fe the3C sampleFe3C sample (from (from cellulose) cellulose) moved moved to the to hydrophobic the hydrophobic phase. phase. When When a magnet a magnet was was moved moved close close to the to waterthe water phase, phase, no oil no mixed oil mixed with it.with The it. test The with test oilwith is reportedoil is reported in Figure in Figure13. This 13. material This material could then could be appliedthen be applied for contaminants for contaminants in wastewater, in wastewater, such as cookingsuch as cooking oil, since oil, it leavessince it the leaves water the ‘clean’. water ‘clean’.
Materials 2019, 12, 323 13 of 15 Materials 2018, 11, x FOR PEER REVIEW 13 of 15
(a) (b)
FigureFigure 13. 13. PicturesPictures of of the the test test of of Fe Fe33CC sample sample from from cellulose cellulose both both in in water water ( (aa)) and and in in a a mixture mixture of of water water andand vegetable vegetable oil oil ( b).). By By using using a a magnet magnet (see (see bottom bottom of of Fe 3CC in in oil-water oil-water solution, solution, picture picture on on the the far far right),right), the the contaminant contaminant is not carried into the water.
4.4. Conclusions Conclusions IronIron carbide@carbon nanocomposites nanocomposites were were prepar prepareded using using different different carbon carbon sources, sources, with with the aimaim of controlling morphology, surface area, and absorption properties of the finalfinal material. After characterization, the materials were tested for different applications, such as CO adsorption uptake, characterization, the materials were tested for different applications, such as CO22 adsorption uptake, and dye and oil removal from water. The best CO adsorption performance was found for samples and dye and oil removal from water. The best CO2 adsorption performance was found for samples prepared from complex carbon source structures, such as chitosan, with a variation in the CO uptake prepared from complex carbon source structures, such as chitosan, with a variation in the CO2 uptake byby varying varying the the quantity quantity of of chitosan chitosan used used and and the the heat heat treatment; treatment; this this finding finding is is especially especially interesting interesting because no direct relation with the surface area was observed. Further investigation over the CO because no direct relation with the surface area was observed. Further investigation over the CO2 adsorption behaviour of these samples was also studied to extend Fe C-based materials for CO adsorption behaviour of these samples was also studied to extend Fe33C-based materials for CO2 sensors, especially considering that, to the best of our knowledge, this was the first time where CO sensors, especially considering that, to the best of our knowledge, this was the first time where CO2 uptakeuptake was reported.reported. HigherHigher saturationsaturation magnetization magnetization was was observed observed for for urea ureaR =R3 = sample 3 sample but, but, for for the theother other nanocomposites, nanocomposites, the the presence presence of extra of ex carbontra carbon should should be taken be taken into into account. account. TheThe best best sample sample for for dye dye uptake uptake was was the the one one prepar prepareded from from cellulose, cellulose, with with an an almost almost immediate immediate uptakeuptake of dye dye by by the the nanocomposite nanocomposite powder. powder. This This samp samplele proved proved to be be also also applicable applicable for for oil oil removal removal fromfrom water water with no powder present in the water.
SupplementarySupplementary Materials: TheThefollowing following are are available available online online at http://www.mdpi.com/1996-1944/12/2/323/s1 at www.mdpi.com/xxx/s1, Figure S1: Particles, dimensionFigure S1: Particleshistograms dimension for the samples histograms discussed for the samplesin the manuscript. discussed in (A) the urea manuscript. R = 3, (B) (A) cellulose, urea R = (C) 3, (B)glucose cellulose, R = (C) glucose R = 8, (D) glucose R = 16, (E) sucrose R = 8, (F) sucrose R = 10, (G) sucrose R = 12, (H) chitosan 0.42%, (I) 8,chitosan (D) glucose 0.21%, R (L) = 16, 0.42% (E) chitosansucrose withR = 8, slower (F) sucrose rate, Figure R = 10, S2: (G) SEM sucrose images R of= 12, samples (H) chitosan prepared 0.42%, with (A)(I) chitosan urea and 0.21%,(B) cellulose (L) 0.42% respectively, chitosan Figurewith slower S3: SEM rate, images Figure of S2: samples SEM images prepared of usingsamples glucose prepared with differentwith (A) glucose/ironurea and (B) cellulosemolar ratio respectively, (A) R = 8 andFigure (B) RS3:= SEM 16, Figure images S4: of SEM sample imagess prepared of samples using prepared glucose usingwith sucrosedifferent with glucose/iron different sugar/iron molar ratio: (A) R = 8; (B) R = 10; (C) R = 12; (D) R = 16, Figure S5: SEM images of samples prepared molar ratio (A) R = 8 and (B) R = 16, Figure S4: SEM images of samples prepared using sucrose with different with different amount of chitosan (A) 0.42%, (B) 0.21%, (C) 0.42% slower heat ramping, Figure S6: Variation in the sugar/ironconcentration molar of ratio: MO in(A) cellulose R = 8; (B) sample R = 10; over(C) R time, = 12; Figure(D) R = S7:16, CalibrationFigure S5: SEM curve images for methyl of samples orange, prepared in theinsert with differentdetails of amount the linear of regressionchitosan (A) are 0.42%, reported, (B) 0.21%, Table S1:(C) Coercive 0.42% slower field, Hheatc, and ramping, the saturation Figure S6: magnetization, Variation in σtheM, concentrationper gram at T of= 300MO K in for cellulose the different sample samples over time, studied. Figure S7: Calibration curve for methyl orange, in the insert detailsAuthor of Contributions: the linear regressionConceptualization, are reported, Table C.D. andS1: Coercive C.G.; Methodology, field, Hc, and C.D.,the saturation M.O.A.M., magnetization, S.A.N., M.R.L. σM, andper gramC.G.; at Validation, T = 300 K for C.D. the and different C.G.; Formalsamples Analysis, studied. C.D., S.A.N. and C.G.; Investigation, C.D., M.O.A.M, S.A.N. and M.R.L.; Resources, C.G.; Data Curation, C.D., S.A.N.; Writing–Original Draft Preparation, C.D. and C.G.; AuthorWriting–Review Contributions: and Editing, Conceptualization, C.D. and C.G.; C.D. Supervision, and C.G.; C.G.; Methodology, Project Administration, C.D., M.O.A.M., C.G. S.A.N., M.R.L. and C.G.;Funding: Validation,This research C.D. and received C.G.; Formal no external Analysis, funding. C.D., S.A.N. and C.G.; Investigation, C.D., M.O.A.M, S.A.N. andAcknowledgments: M.R.L.; Resources,SEM C.G.; and Data TEM Curation, experimental C.D., S.A.N. data were; Writing–Original provided by the Draft Nanovision Preparation, Department C.D. and ofC.G.; the Writing–ReviewQMUL. XRD experimental and Editing, data C.D. were and providedC.G.; Supervision, by the Material C.G.; Project Research Administration, Institute of the C.G. QMUL. N2 and CO2 adsorption were provided by the School of Engineering and material science of QMUL. Magnetic measurements Funding:were provided This research by the Department received no of external Physics, funding. University of Warwick. C.G. and C.D. are grateful to Titirici group for the N2 and CO2 adsorption measurements. C.G. and C.D. acknowledge the QMUL for financial support. Acknowledgements: SEM and TEM experimental data were provided by the Nanovision Department of the Conflicts of Interest: The authors declare no conflict of interest. QMUL. XRD experimental data were provided by the Material Research Institute of the QMUL. N2 and CO2 adsorption were provided by the School of Engineering and material science of QMUL. Magnetic measurements
Materials 2019, 12, 323 14 of 15
References
1. Pigott, V.C. The Archaeometallurgy of the Asian old World; University of Pennsylvania Museum of Archaeology and Anthropology: Philadelphia, PA, USA, 1999; ISBN 0924171340. 2. Johnson, D.; Tyldesley, J.; Lowe, T.; Withers, P.J.; Grady, M.M. Analysis of a prehistoric Egyptian iron bead with implications for the use and perception of meteorite iron in ancient Egypt. Meteorit. Planet. Sci. 2013, 48, 997–1006. [CrossRef] 3. Levin, A.A.; Meyer, D.C.; Reibold, M.; Kochmann, W.; Pätzke, N.; Paufler, P. Microstructure of a genuine Damascus sabre. Cryst. Res. Technol. 2005, 40, 905–916. [CrossRef] 4. Yu, J.; Chen, F.; Gao, W.; Ju, Y.; Chu, X.; Che, S.; Sheng, F.; Hou, Y. Iron carbide nanoparticles: an innovative nanoplatform for biomedical applications. Nanoscale Horiz. 2017, 2, 81–88. [CrossRef] 5. Tang, W.; Zhen, Z.; Yang, C.; Wang, L.; Cowger, T.; Chen, H.; Todd, T.; Hekmatyar, K.; Zhao, Q.; Hou, Y.; et al.
Fe5 C2 Nanoparticles with high MRI contrast enhancement for tumor imaging. Small 2014, 10, 1245–1249. [CrossRef][PubMed]
6. Wang, K.; Qiu, G.; Cao, H.; Jin, R. Removal of chromium (VI) from aqueous solutions using Fe3O4 magnetic polymer microspheres functionalized with amino groups. Materials 2015, 8, 8378–8391. [CrossRef][PubMed] 7. Sajitha, E.P.; Prasad, V.; Subramanyam, S.V.; Mishra, A.K.; Sarkar, S.; Bansal, C. Size-dependent magnetic properties of iron carbide nanoparticles embedded in a carbon matrix. J. Phys. Condens. Matter 2007, 19, 46214. [CrossRef] 8. Giordano, C.; Kraupner, A.; Wimbush, S.C.; Antonietti, M. Iron carbide: An ancient advanced material. Small 2010, 6, 1859–1862. [CrossRef] 9. Giordano, C.; Kraupner, A.; Fleischer, I.; Henrich, C.; Klingelhöfer, G.; Antonietti, M. Non-conventional
Fe3C-based nanostructures. J. Mater. Chem. 2011, 21, 16963. [CrossRef] 10. Rodriguez, J.A.; Liu, P.; Stacchiola, D.J.; Senanayake, S.D.; White, M.G.; Chen, J.G. Hydrogenation of CO2 to Methanol: Importance of metal–oxide and metal–carbide interfaces in the activation of CO2. ACS Catal. 2015, 5, 6696–6706. [CrossRef] 11. Sharma, S.; Madou, M. A new approach to gas sensing with nanotechnology. Philos. Trans. A Math. Phys. Eng. Sci. 2012, 370, 2448–2473. [CrossRef]
12. Mirzaei, A.; Hashemi, B.; Janghorban, K. α-Fe2O3 based nanomaterials as gas sensors. J. Mater. Sci. Mater. Electron. 2016, 27, 3109–3144. [CrossRef] 13. Matatagui, D.; Kolokoltsev, O.V.; Qureshi, N.; Mejía-Uriarte, E.V.; Saniger, J.M. A magnonic gas sensor based on magnetic nanoparticles. Nanoscale 2015, 7, 9607–9613. [CrossRef][PubMed] 14. Su, Y.-C.; Lin, C.T.; Chung, W.Y.; Wu, W.J. A room temperature operated carbon dioxide sensor based on EB-PANI/PEDOT:PSS sensing material. Allsensors 2016, 2016, 23–26. 15. Gooneratne, C.; Li, B.; Moellendick, T. Downhole applications of magnetic sensors. Sensors 2017, 17, 2384. [CrossRef] 16. Schliehe, C.; Yuan, J.; Glatzel, S.; Siemensmeyer, K.; Kiefer, K.; Giordano, C. Iron nitride and carbide: From crystalline nanoparticles to stable aqueous dispersions. Chem. Mater. 2012, 24, 2716–2721. [CrossRef] 17. Zhu, K.; Ju, Y.; Xu, J.; Yang, Z.; Gao, S.; Hou, Y. Magnetic nanomaterials: chemical design, synthesis, and potential applications. Acc. Chem. Res. 2018, 51, 404–413. [CrossRef] 18. Giordano, C.; Antonietti, M. Synthesis of crystalline metal nitride and metal carbide nanostructures by sol-gel chemistry. Nano Today 2011, 6, 366–380. [CrossRef] 19. Dhahri, R.; Leonardi, S.G.; Hjiri, M.; Mir, L.E.; Bonavita, A.; Donato, N.; Iannazzo, D.; Neri, G. Enhanced
performance of novel calcium/aluminum co-doped zinc oxide for CO2 sensors. Sens. Actuators B Chem. 2017, 239, 36–44. [CrossRef] 20. Glatzel, S.; Schnepp, Z.; Giordano, C. From paper to structured carbon electrodes by inkjet printing. Angew. Chem. Int. Ed. 2013, 52, 2355–2358. [CrossRef] 21. Dutta, P.K.; Dutta, J.; Tripathi, V.S. Chitin and chitosan: Chemistry, properties and applications. CSIR 2004, 63, 20–31. 22. Arora, S.; Lala, S.; Kumar, S.; Kumar, M.; Kumar, M. Comparative degradation kinetic studies of three biopolymers: Chitin, chitosan and cellulose. Arch. Appl. Sci. Res. 2011, 3, 188–201. Materials 2019, 12, 323 15 of 15
23. Cohen, E.R.; Cvitas, T.; Frey, J.G.; Holström, B.; Kuchitsu, K.; Marquardt, R.; Mills, I.; Pavese, F.; Quack, M.; Stohner, J.; et al. Quantities, Units, and Symbols in Physical Chemistry, 2nd ed.; Blackwell Scientific Publications: Oxford, UK, 1993; ISBN 978-0-632-03583-0.
24. Li, M.; Huang, K.; Schott, J.A.; Wu, Z.; Dai, S. Effect of metal oxides modification on CO2 adsorption performance over mesoporous carbon. Microporous Mesoporous Mater. 2017, 249, 34–41. [CrossRef]
25. Kumar, S.; Saxena, S.K. A comparative study of CO2 sorption properties for different oxides. Mater. Renew. Sustain. Energy 2014, 3, 30. [CrossRef]
26. Younas, M.; Sohail, M.; Leong, L.K.; Bashir, M.J.; Sumathi, S. Feasibility of CO2 adsorption by solid adsorbents: a review on low-temperature systems. Int. J. Environ. Sci. Technol. 2016, 13, 1839–1860. [CrossRef] 27. Wang, X.; Yu, J.; Cheng, J.; Hao, Z.; Xu, Z. High-temperature adsorption of carbon dioxide on mixed oxides derived from hydrotalcite-like compounds. Environ. Sci. Technol. 2008, 42, 614–618. [CrossRef]
28. Li, W.; Yang, H.; Jiang, X.; Liu, Q. Highly selective CO2 adsorption of ZnO based N-doped reduced graphene oxide porous nanomaterial. Appl. Surf. Sci. 2016, 360, 143–147. [CrossRef] 29. Queen, W.L.; Hudson, M.R.; Bloch, E.D.; Mason, J.A.; Gonzalez, M.I.; Lee, J.S.; Gygi, D.; Howe, J.D.; Lee, K.; Darwish, T.A.; et al. Comprehensive study of carbon dioxide adsorption in the metal–organic frameworks M
2 (dobdc) (M = Mg, Mn, Fe, Co, Ni, Cu, Zn). Chem. Sci. 2014, 5, 4569–4581. [CrossRef] 30. Chen, B.; Ma, G.; Kong, D.; Zhu, Y.; Xia, Y. Atomically homogeneous dispersed ZnO/N-doped nanoporous
carbon composites with enhanced CO2 uptake capacities and high efficient organic pollutants removal from water. Carbon 2015, 95, 113–124. [CrossRef]
31. Yu, C.-H.; Huang, C.-H.; Tan, C.-S. A Review of CO2 capture by absorption and adsorption. Aerosol Air Qual. Res. 2012, 12, 745–769. [CrossRef]
© 2019 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/).