Optimization of the graphene oxide synthesis by experimental design

(1)

Bachelor Thesis Scheikunde

Optimization of the graphene oxide synthesis by

experimental design

door

K. C. van Rijn

15 Juli 2017

Studentnummer 10781846 Onderzoeksinstituut Van 't Hoff Institute for Molecular Sciences Onderzoeksgroep

Heterogeneous Catalysis and Sustainable Chemistry

Verantwoordelijk docent Dr. N. R. Shiju

Begeleider T. K. Slot

(2)

1

Abstract

Using a method combining experimental design and principal component analysis, the relations between experimental and product parameters in the modified

Hummers oxidation of graphite to graphene oxide by Chen et al.1_{were analysed,} and there was found to be an inverse relation between the reaction temperature and degree of oxidation, an inverse relation between the exfoliation temperature and degree of oxidation, and a relation between the polydispersity of the graphene oxide structure and the heat generated during the reaction.

The thermal decomposition of the graphene oxide product was studied using XRD, and it was found that the structure of the product degrades even at temperatures as low as 40°C. This degradation continues up to a temperature of 200°C, at which the product starts to fully decompose.

The reaction was also attempted using large graphite flakes and amorphous carbon powder, although the large graphite flakes did not react to a reasonable degree, and the carbon powder returned a black powder of equal structure.

(3)

2

Samenvatting (NL)

Grafiet is een van de vormen waarin koolstof in de natuur voor kan komen, en bestaat uit een grote hoeveelheid gestapelde laagjes. Een zo een laagje noem je grafeen, en dit materiaal heeft veel bijzondere eigenschappen. Zo geleid het elektriciteit, is het enorm sterk, dun en flexibel. Een afgeleide van grafeen,

grafeenoxide, bestaat uit dezelfde laagjes, maar deze laagjes bevatten in dit geval ook heel veel zuurstof op het oppervlak, en kan chemisch worden omgezet in grafeen. Om grafeenoxide te maken is een hele sterke oxidator nodig, die de zuurstof op de laagjes kan zetten, in dit geval gebruiken wij mangaan heptoxide, maar andere oxidatoren kunnen ook gebruikt worden. Als het grafiet en de mangaan heptoxide in geconcentreerd zwavelzuur bij elkaar worden gebracht, gaat het

zwavelzuur eerst onder invloed van het mangaan heptoxide tussen de laagjes zitten, waardoor de laagjes uit elkaar gedrukt worden, zoals in de onderstaande afbeelding te zien is. Nadat de laagjes uit elkaar gedrukt zijn kan de mangaan heptoxide er ook tussen, en plakt het overal zuurstof op de laagjes.

Het materiaal dat zo gevormd wordt bevat veel OH groepen, net als een alcohol, en kan dus goed in water oplossen en komt zo los van de rest. In dit onderzoek wordt onderzocht onder welke omstandigheden, zoals temperatuur of andere toegevoegde stoffen, dit het beste gaat. Door het op een aantal verschillende manieren te

proberen, en vervolgens een zogenaamde hoofdcomponentenanalyse uit te voeren is het mogelijk om uit te vinden wat de onderlinge relaties zijn tussen de

omstandigheden die gevarieerd zijn, en dus ook welke richting we op moeten om het het beste te doen. Een hoofdcomponentenanalyse is een wiskundige behandeling waarbij door met matrices te rekenen een aantal vectoren worden gemaakt, die aangeven in welke richting de relaties tussen de ingevoerde waarden liggen. Door dit te doen hebben wij verschillende relaties ontdekt, waaronder twee relaties tussen de hoeveelheid zuurstof in het product en de temperatuur, en een relatie tussen de warmte die bij de reactie vrijkomt en de structuur van de laagjes die geproduceerd worden.

(4)

3

Introduction

Graphene oxide(GO) has recently attracted the attention of the scientific community because of its structural2_{, electrochemical}3_{and even optical properties}4_{. Graphene} oxide is also an intermediate in the manufacture of certain types of graphene and expanded graphite, two related materials that have also become increasingly popular within recent years. Graphene oxide, also commonly referred to as graphitic oxide, or graphite oxide in bulk form, has a structure consisting of carbon plates containing high amounts of oxygen functionalities in the form of epoxides, hydroxyl, phenol, ketone and carboxylic acids.5,6_{The material also often contains sulphur based} functionalities.7_{An example of a GO structure is given in figure 1.}

Figure 1: An example of a graphene oxide structure showing epoxy, hydroxyl, phenolic, ketone and carboxylic acid functionalities.

The oxidation of graphite to graphene oxide was first performed by Benjamin C. Brodie in 1859, more than a century ago, using a fuming nitric acid and potassium chlorate mixture.8_{As this procedure requires multiple reactions and produces the} explosive and toxic chlorine dioxide gas, a safer and faster method was developed by Hummers and Offeman in 1957 using potassium permanganate and sodium nitrate in sulphuric acid.9_{More recently, in 2013, a method relying on the binary} system of potassium permanganate and sulphuric acid was used to perform the oxidation, which simplifies the system, and prevents the formation of toxic nitrogen dioxide gas has been developed by Chen et al.1

The mechanism of the reaction is thought to proceed in a two-steps.10_{The first step} is the oxidative intercalation of a weakly coordinating anion into the graphite lattice forming a graphite intercalation compound. The intercalating anion depends on the type of reaction used, and hydrogen sulphate and nitrate are most often the

intercalating species, but this depends on which acids are present.

The second step in the mechanism is effected by the main oxidizing agent, which can also be the oxidizer in the first step. After the graphite has been intercalated with the anions, the main oxidizer can move into the structure and oxidize the surface of the mixture, forming the characteristic oxygen functionalities. The main oxidizer for the production of graphene oxide is either potassium permanganate or potassium

(6)

5

chlorate, depending on the method used. The process is shown graphically in figure 2. These two processes can occur simultaneously in the same reaction mixture or separately, again, depending on the procedure.

Figure 2: The reaction of graphite to graphene oxide, firstly, hydrogen sulphate oxidatively intercalates into the graphite structure, followed by the action of the main oxidant on the surface and sides of the graphene sheets to introduce oxygen functionalities.

Exfoliation of the intermediate bulk graphitic oxide can be effected in multiple ways, including heating to induce partial decomposition,11_{dissolving in basic aqueous} solution,12_{and sonication of a graphite oxide suspension.}13

Experimental design

Experimental design is a very broad term used to describe the design of procedures that are used to explain variations in a system by varying input parameters. These parameters and their effect on the system can be studied either one at a time or using a factorial experiment, varying multiple parameters at the same time.14

Factorial experiment is often more useful, as it can uncover more complex relations between a multitude of variables.

In order to process the experimental data obtained after the experiments, a

processing technique is needed to uncover the relations between the data obtained. One such technique, principal component analysis (PCA), uses matrix calculations to find the relations between data vectors.15_{In order to perform PCA, the data is first} converted to a matrix, after which the covariance matrix of the data matrix is

calculated. The covariance matrix contains all values for the covariance of each data column with the other columns, and thus indicates their relation. Mathematically, one can determine the eigenvectors and eigenvalues of this matrix, which will indicate the relations which describe the system and the relative amount by which the system

(7)

6

is described by each vector, respectively. These eigenvectors can then also be used to define the data in terms of the principal components, which is the data translated using the eigenvectors instead of the original units. During this step, less important data relations can be omitted from the result by not including their eigenvectors during the final transformation. In our results, we will perform the PCA step by step and explain our findings.

The goal for this project is to find which reaction variables affect the reaction and product and in what way they do. In order to achieve this, a set of reactions will be performed using a strategically chosen set of variables. Because not all variables can be directly controlled due to the nature of the reaction and because of the large dimension of the data, the analysis technique of principal component analysis will be used to find the relations between the experimentally obtained variables, chosen variables and product properties.

(8)

7

Results and discussion

In order to make a principal component analysis (PCA), an array of reactions were performed using different reaction conditions, we chose to do the PCA on both reaction and exfoliation temperature, as the product is very sensitive to high

temperatures, but the reaction itself won’t proceed at ambient or lower temperature. As seen in the temperature log in figure 4, the reaction is strongly exothermic,

showing a sharp increase in temperature starting when the reaction mixture reaches a temperature of around 30°C, thus we use 30°C as the minimum temperature. Because manganese heptoxide is known to decompose explosively above 55°C16_, 50°C was used as the maximum temperature for the reaction, although only normal decomposition was observed for the sulphuric acid solution of manganese

heptoxide, even at higher temperatures. For the exfoliation temperature, both the 95°C used in the article on which this synthesis method is based1_{, and a much} milder 60°C were used.

These reactions were examined using XRD, as XRD is known to give the distance between graphene sheets in graphite and its derivatives, which increases with an increase of oxygen functionalities on the carbon surface, thus, it can be used to give an indication of the degree of oxidation of the material. Morimoto et al. have

performed a study on the relationship between these parameters.17_{All graphene} oxide products yielded XRD spectra similar to that in figure 3, containing a graphene oxide peak and a second order peak at slightly varying positions of about 8-12°.

(9)

8

In addition to the XRD, the temperature of each reaction was also monitored and logged in order to record the exothermic reaction during the first part of the synthesis and monitor any abnormalities during the rest of the reaction, as shown in figure 4.

Figure 4: Full picture of a typical temperature plot, including description of important steps in the process. The figure is plotted right to left in a millivolt scale with 1mV = 1.12°C and 1 div = 0.85min.

(10)

9

In order to quantify the exothermic reaction, a blank “reaction” was run using only water and monitoring the temperature during heating. From this blank, a heating model was made using a hill equation of order 3, as this was found to give the best model for the system. This hill equation, including offsets in time and temperature is given below. 𝑇(𝑡) =(𝑡 − 𝑡0) 3_{∙ (𝑇} 𝑠𝑒𝑡− 𝑇0) (𝑡 − 𝑡0)3+ 𝑡𝑎𝑡 𝑇1 2 3 + 𝑇0

Where T is the temperature, t is the time, t0 is the time at which heating starts, T0 is the temperature at t0, Tset is the target heating temperature, and tat T½ is the time at which the temperature is halfway between T0 and Tset. By the design of this function, all variables are known and controlled, except tat T½, which is dependent on

experimental factors such as the amount of water in the heating bath, contact with the heating plate, and heating plate used and must be fitted to the temperature plot. Digitizing the first part of the temperature plot and subtracting the fitted blank heating plot from the temperature plot, as done below in figure 5, we get a plot containing two peaks, the first one being caused by the heating after changing the water/ice bath for a heating bath with water at room temperature, and the second by the exothermic reaction. For the principal component analysis we can numerically integrate the second peak to get the TΔt value, which is the surface of the peak and gives us a relative value for the energy released during this reaction.

Figure 5: Digitized plot of the start of the reaction, showing reaction temperature (gray), fitted blank heating temperature(red) and the difference between these(blue).

(11)

10

Product and side-products

The product of the reaction is relatively difficult to isolate, as it has a tendency to form a gel when washing with water or other general solvents, but it was found that the residual salts and acid could be separated by dialysis. Before performing dialysis, it is best to let the mixture settle and decant as much of the sulphuric acid as possible and replace it with water, as the high acidity and dehydrating properties of the concentrated solution can damage the dialysis membrane. During dialysis, the graphene oxide, a yellow powder in acid solution, forms a thick dark brown gel, as shown in figure 6a. Diluting this gel creates a yellow colloid or solution, shown in figure 6b. The gel can be dried to yield a dark brown film, the thickness of which depends on the surface area of the container and the amount of gel.

Figure 6 a:Product after dialysis, forming at thick gel. b:Dilute solution of the product, showing its yellow colour.

Looking carefully at the reaction mixture, a precipitate was noticed which was

different from the graphene oxide precipitate. The precipitate was washed out of the container using acetone, and the precipitate was washed several times using

acetone, as it was found to dissolve in water. The precipitate was found to actually contain two different compounds, one which forms larger crystals, while the other is a fine microcrystalline powder. The two compounds were separated using a sieve, and were examined using XRD as seen in figure 7. When compared to a database, it was found that the larger crystalline precipitate was potassium bisulphate, while the microcrystalline powder was manganese sulphate monohydrate, the artificial form of the szmikite mineral.18

(12)

11

Figure 7: XRD spectra of the two precipitates found in the reaction mixture, potassium bisulphate and manganese sulphate monohydrate.

(13)

12

Principal Component Analysis

In order to perform Principal Component Analysis(PCA) on our data, a reduced table of the data, table 1, was converted to a matrix form. This reduced data table only contains the data values that could influence the degree of oxidation, also excluding all values that were constant throughout all experiments. These values are reaction set temperature, reaction maximum temperature, calculated heat production during active reaction as shown in the general results, water addition time, water addition maximum temperature, exfoliation temperature, XRD peak width at half height and XRD peak position, respectively.

Table 1:Reduced experiment data table

Ex. No. T reaction set (°C) T reaction max. (°C) Heat production (Tex*Δt) Quench time (min) Tmax quench (°C) T exf. (°C) Graphene oxide XRD WaHH (2θ) Graphene oxide XRD position (2θ) 1 40 53.2 181.9 8 78.4 95 1.2 9.25 2 50 58.8 122.2 8 86 95 0.7 10.7 3 30 48.2 162.9 8 87.4 95 0.8 9.9 4 40 52.1 100 4 92.2 95 0.75 9.9 5 40 46.5 64.5 58 57 95 0.75 9.2 6 40 56 133.9 51 54 60 0.65 8.2 7 50 61.3 97 50 36 60 0.75 8.7

Conversion to matrix form yielded the below displayed matrix, without transposing columns and rows.

In order to prepare our matrix for PCA, a mean was calculated from the matrix.

The mean was subtracted from the original data matrix to yield a new data matrix with a mean of zero.

(14)

13

The data was standardized to prevent bias based on the units of the original data.

The covariance matrix was calculated from this adjusted data set.

This covariance matrix shows relations between the columns of our data, and from this matrix the eigenvectors were calculated. The eigenvectors are shown in the rows of the matrix, and represent the covariance of the columns from the original dataset, shown as component vectors.

The corresponding eigenvalues were also calculated. These represent how much each of the eigenvectors defines the original matrix. Each eigenvalue corresponds to the above eigenvector on the same row.

(15)

14

These eigenvectors and eigenvalues can be used to directly derive the related values in the system or be used to calculate the principal components, the data rewritten using the eigenvectors as the new dimensions of the matrix. The full sized principal component matrix is too complex to yield any useful information, but using only part of the eigenvectors, a reduced matrix can be calculated. The full principal component matrix is shown below.

The reduced matrix calculated using only the first two eigenvectors is shown below.

Plotting these values gives the plot in figure 8below.

Figure 8: Plot of the reduced principal components.

The points in this matrix are however randomly distributed throughout the system, and no further data can be extracted.

(16)

15

From these data we can determine which variables are most dependent on each other. The eigenvalues indicate that only the first four eigenvectors hold relevant information, the rest can be ignored.

The first eigenvector shows a positive relation between the XRD peak position, water quench temperature, and exfoliation temperature, and a negative relation with water addition time, likely due to the inverse relation between water addition and increasing temperature. This indicates the primary variable for the properties of the graphene oxide product to be the temperature of the exfoliation, including the temperature during water addition and thus a negative relationship with the oxidation degree of the product.

The second eigenvector shows a positive relation between the XRD peak position, reaction temperature set, and maximum reaction temperature, showing a negative relationship between the reaction temperature and oxidation degree of the graphene oxide product.

Although the fourth eigenvector shows no special relations, the third vector reveals an unexpected, but strong relation between XRD peak width and heat generated during the reaction. We hypothesize this is due to the overoxidation of graphite causing the oxidizing agent to be consumed before the product is fully oxidized, causing a distribution between highly oxidized small graphene oxide plates and larger less oxidized graphene plates, giving a broad distribution in the graphene oxide peak on XRD and causing heating of the reaction mixture as the oxidation to carbon dioxide releases more energy.

(17)

16

Testing of additives

Although the sulphuric acid/potassium permanganate system is the simplest system for oxidizing graphite to graphene oxide, many different combinations have been used, including the addition of phosphoric acid used by Marcano et al.19_{and the} usage of sodium nitrate in the original Hummers’ synthesis9_{. Here, we explored the} usage of both phosphoric acid and perchlorate, an exceptionally non nucleophilic anion which is known to intercalate into graphite under oxidizing conditions, even better so than sulphuric acid20_{. The intercalation of perchlorate into graphite was} tested by bubbling dry air into a graphite suspension in azeotropic perchloric acid, which was found to be successful, as seen in figure 9, when compared to the literature.20

Figure 9: XRD of perchlorate intercalate in graphite using perchloric acid and potassium permanganate, including a graphite reference.

Both additives had however no effect on the structure of the product produced, as determined by XRD, as well as no significant effects on the reaction temperature. Unfortunately, we didn’t have access to equipment to determine the surface

functionalities, so it is not possible to tell if there’re any differences compared to the material produced without additives. Using a technique such as X-ray photoelectron spectroscopy would give more insight into the effect of the additives.

(18)

17

Analysis and optimization of the drying procedure

Graphene oxide being sensitive to high temperatures, the drying and heating of the wet purified material is important for the quality and properties of the final product. In order to do this, a sample of dried graphene oxide was sequentially heated at

different temperatures and examined using XRD to determine structure integrity. Weight was also measured, but the small sample size prevented accurate

measurement. The sample was initially dried at 30°C and the temperature was increased stepwise to 260°C, heating for 15 minutes at each temperature before cooling to room temperature and performing measurements, the spectra of which are shown in figure 10.

Figure 10: XRD spectra of graphene oxide heated to different temperatures, shown as waterfall chart. Peak position is indicated.

The XRD spectra recorded during the experiment clearly show the deterioration of the graphene oxide at even moderate temperatures, small changes in structure starting even at 40°C. Upon further heating, the graphene oxide starts to decompose further, and at 200°C, the structure dramatically collapses as the GO material is reduced to a disordered graphite structure. This is also accompanied with a weight reduction to roughly half the original weight, indicating near full expulsion of oxygen surface functionalities. In order to better show this change, a heatmap graph was made using the XRD data displayed in figure 10, shown below in figure 11. In this figure, the shift in peak position is shown more clearly, including the second order GO peak and graphitic peak.

(19)

18

Figure 11: Heatmap composed of combined XRD graphs of the heating of graphene oxide. The strong signal on the left is graphene oxide, the middle bottom signal is the second order signal of graphene oxide and the top right signal is graphitic.

(20)

19

Alternate carbon sources

In addition to the commercial fine graphite powder, other sources of carbon were explored. First, we tried to perform the reaction using large commercial graphite flakes (~1mm), but only traces of this material would react and mostly graphite flakes were recovered after dialysis. Grinding up this graphite material and sieving to select for particle size under 57μm produces a fine graphite powder, which was found to react the same as the commercial fine powder.

In addition to graphite, an amorphous carbon material was tested. The carbon

material was prepared from the acid catalysed dehydration of sucrose, forming a fine carbonaceous powder. This powder was then used in a reaction both before and after calcination at 600°C. When using the carbon before calcination, a vigorous reaction was observed during addition of permanganate, and a purple colour was observed, although a different shade than that of permanganate solutions. We were able to recreate this purple colour by reduction of dilute manganese heptoxide in sulphuric acid using small amounts of either ethanol or acetone, as seen in figure 12, although it remains unknown what species is responsible for this colour.

Figure 12: Colours of a: manganese heptoxide in sulphuric acid b:permanganic acid in water c: manganese heptoxide solution after reduction using ethanol.

After this violent reaction, no further reaction was observed during the rest of the process, and the purple colour slowly started fading away about halfway into the process. Using XRD, it was found that the structure of the carbon material was, however, unchanged. The reaction was also attempted using the calcined carbon, but no reaction occurred and XRD indicated no change in structure.

(21)

20

Conclusion

The oxidation of graphite was performed under an array of different conditions, and the results were analysed using principal component analysis(PCA). The PCA revealed that both the reaction temperature and the exfoliation temperature are inversely related to the degree of oxidation, although separately, and the dispersity of the degree of oxidation of the material is related to the heat generated during the reaction, although no way to control the heat generation has been found.

All reactions performed on the powdered graphite yielded a mixture of sulphuric acid, water, yellow graphene oxide powder, potassium bisulphate precipitation, and

manganese sulphate monohydrate precipitation. Performing dialysis on this product yielded an aqueous gel of dark brown graphene oxide that could be dried to yield the purified product.

The decomposition of the product during heating was researched and it was found that, while full decomposition only occurs between 200°C-250°C, the product does degrade and the structure collapses at temperatures as low as 40°C.

It was also discovered that the graphite needs to be very finely divided in order to properly react, tolerating particle sizes only slightly higher than 50μm. Performing the reaction on amorphous carbon was found to not alter its structure, although reaction can occur on the surface functional groups of the carbon. Performing the reaction in the presence of the additives phosphoric acid and sodium perchlorate didn’t

influence the structure of the final product significantly, although it could alter surface functionality. In order to find the exact effect of the reaction on the amorphous

carbon or the effect of these additives, further analysis needs to be performed in order to identify the difference in surface functionalities.

(22)

21

Experimental

Equipment

The Rigaku MiniFlex II desktop X-ray diffractometer equipped with the default copper based x-ray source was used to record all XRD spectra (Cu Kα: 0.15418nm). A Fritsch pulverisette 2 mortar grinder equipped with steel pestle and mortar was used to grind the large graphite flakes.

Chemicals

Finely powdered graphite was obtained from Merck B.V. Graphite flakes were

obtained from Sigma-Aldrich Chemie N.V. All other chemicals used were obtained in high purity from their respective commercial sources.

Procedure for general graphene oxide synthesis

The graphene oxide synthesis method used is an adaptation of the method used by Chen et al.1

Sulphuric acid(140ml) was added to a 1 litre round bottom flask and cooled on ice. Graphite powder(6.0g, 0.5mol carbon equivalent) was added to the round bottom flask and the mixture was vigorously stirred. Carefully, over the course of about 12 minutes, potassium permanganate(18.0g, 0.114mol) was added to the mixture in small portions while keeping the temperature under 15°C to yield a green

manganese heptoxide solution. The ice/water bath was replaced by a water bath, and the mixture was heated to 40°C, while monitoring the temperature of the reaction mixture to prevent and monitor a thermal runaway. After 30 minutes, the mixture was cooled on a water/ice bath.

Water(150ml) was carefully added while the mixture was still vigorously stirred and the temperature was kept below 95°C. The mixture was then heated to 95°C for 15 minutes to exfoliate the graphitic oxide. The mixture was allowed to cool to 60°C and a 10% hydrogen peroxide solution in water(10ml) was slowly added to quench the reaction and reduce all oxidized Mn species to Mn2+_{, effervescence stopped after} adding roughly 5ml. The mixture was transferred to a bottle with ventilation cap and left to stand for a few days to decompose leftover hydrogen peroxide and let the graphene oxide settle on the bottom.

Excess water/sulphuric acid was decanted and the sludge was suspended in an approximate equal volume of water. The resulting suspension was transferred into a dialysis tube and dialysis was performed in water until the pH of the water remained constant (pH≈6). The graphene oxide, which is now a thick, dark brown gel, was dried at 25°C under a flow of nitrogen until dry to yield the final graphene oxide product. XRD samples were prepared by applying the wet graphene oxide gel onto an XRD plate and drying it under a nitrogen atmosphere at no more than 25°C.

(23)

22

Procedure for reaction temperature and exfoliation temperature variation

To perform the reaction temperature variations, the water bath was heated to either 30°C or 50°C instead of 40°C. To perform the exfoliation temperature variations, the addition speed of the water was adjusted to keep the temperature under 60°C and/or the water bath was heated to only 60°C.

Procedure for graphene oxide synthesis using a phosphoric acid additive

The general graphene oxide synthesis method was used, with a water addition temperature and exfoliation temperature of 60°C and 140ml of 1:9 85% phosphoric acid:sulphuric acid instead of 140ml of sulphuric acid.

Procedure for graphene oxide synthesis using a phosphoric acid additive

The general graphene oxide synthesis method was used, with a water addition temperature and exfoliation temperature of 60°C and an addition of sodium perchlorate(1.2g, 0.01mol) to the sulphuric acid (use of sodium perchlorate over potassium perchlorate is important, as it’s solubility is over 100x greater than that of potassium perchlorate).

Procedure for the production of powdered graphite from graphite flakes

Graphite flakes were ground up in a mortar grinding machine for 1h. The clumps of graphite formed this way were broken up on a sieve and sieved through a 57μm mesh. The <57μm particles were collected and the larger particles were recycled to the first step.

Procedure for the production of amorphous carbon from sugar

Sucrose(varying weight) was added to a large beaker(1l per 60g of sucrose) and concentrated sulphuric acid(2ml/3g of sucrose) was added to the beaker. The mixture was stirred using a glass stirring rod until the mixture heated up and expanded to fill the beaker. The black cake was broken up into smaller pieces and heated to 160°C for 15 minutes to finish the reaction. A large excess of water was added and the black mass was filtered off and washed using water, ethanol and acetone. The product was dried at 160°C and crushed to yield a very fine black powder.

Procedure for the production of calcined amorphous carbon

The amorphous carbon prepared using the previous method was heated in a loosely covered calcining dish at 600°C in air for 2h to yield a very fine black product.

(24)

23

Acknowledgements

I would like to thank my main supervisor, Dr. N.R. Shiju, for generously offering me this research project. I’d also like to give my thanks to my daily supervisor, T. Slot Msc, for his help throughout the project and his help during the writing of the report. I also thank all members of the Heterogeneous Catalysis and Sustainable Chemistry research group for their helping me various times during my time there. And lastly, I’d like to thank prof. dr. J.H. van Maarseveen for being the independent reviewer for my report and presentation.

(25)

24

References

(1) Chen, J.; Yao, B.; Li, C.; Shi, G. An Improved Hummers Method for Eco-Friendly Synthesis of Graphene Oxide. Carbon 2013, 64, 225–229.

(2) Tang, L.; Li, X.; Ji, R.; Teng, K. S.; Tai, G.; Ye, J.; Wei, C.; Lau, S. P. Bottom-up Synthesis of Large-Scale Graphene Oxide Nanosheets. J. Mater. Chem. 2012, 22 (12), 5676–5683.

(3) David, L.; Singh, G. Reduced Graphene Oxide Paper Electrode: Opposing Effect of Thermal Annealing on Li and Na Cyclability. J. Phys. Chem. C 2014, 118 (49), 28401–28408.

(4) Liu, Z.; Wang, Y.; Zhang, X.; Xu, Y.; Chen, Y.; Tian, J. Nonlinear Optical

Properties of Graphene Oxide in Nanosecond and Picosecond Regimes. Appl. Phys. Lett. 2009, 94 (2), 021902.

(5) He, H.; Klinowski, J.; Forster, M.; Lerf, A. A New Structural Model for Graphite Oxide. Chem. Phys. Lett. 1998, 287 (1), 53–56.

(6) Kumar, H. V.; Woltornist, S. J.; Adamson, D. H. Fractionation and

Characterization of Graphene Oxide by Oxidation Extent through Emulsion Stabilization. Carbon 2016, 98, 491–495.

(7) Eigler, S.; Dotzer, C.; Hof, F.; Bauer, W.; Hirsch, A. Sulfur Species in Graphene Oxide. Chem. – Eur. J. 2013, 19 (29), 9490–9496.

(8) Brodie, B. C. On the Atomic Weight of Graphite. Philos. Trans. R. Soc. Lond.

1859, 149, 249–259.

(9) Hummers, W. S.; Offeman, R. E. Preparation of Graphitic Oxide. J. Am. Chem. Soc. 1958, 80 (6), 1339–1339.

(10) Hegab, H. M.; Zou, L. Graphene Oxide-Assisted Membranes: Fabrication and Potential Applications in Desalination and Water Purification. J. Membr. Sci.

2015, 484, 95–106.

(11) Talyzin, A. V.; Szabó, T.; Dékány, I.; Langenhorst, F.; Sokolov, P. S.; Solozhenko, V. L. Nanocarbons by High-Temperature Decomposition of

Graphite Oxide at Various Pressures. J. Phys. Chem. C 2009, 113 (26), 11279– 11284.

(12) Dreyer, D. R.; Park, S.; Bielawski, C. W.; Ruoff, R. S. The Chemistry of Graphene Oxide. Chem. Soc. Rev. 2009, 39 (1), 228–240.

(13) Lavin-Lopez, M. P.; Valverde, J. L.; Sanchez-Silva, L.; Romero, A. Solvent-Based Exfoliation via Sonication of Graphitic Materials for Graphene Manufacture. Ind. Eng. Chem. Res. 2016, 55 (4), 845–855.

(14) Fisher, R. A. The Arrangement of Field Experiments. In Breakthroughs in Statistics; Springer Series in Statistics; Springer, New York, NY, 1992; pp 82– 91.

(15) Pearson, K. On Lines and Planes of Closest Fit to Systems of Points in Space. Philos. Mag. 1901, 2 (11), 559–572.

(16) Simon, A.; Dronskowski, R.; Krebs, B.; Hettich, B. The Crystal Structure of Mn2O7. Angew. Chem. Int. Ed. Engl. 1987, 26 (2), 139–140.

(17) Morimoto, N.; Kubo, T.; Nishina, Y. Tailoring the Oxygen Content of Graphite and Reduced Graphene Oxide for Specific Applications. Sci. Rep. 2016, 6, srep21715.

(18) Anthony, J. W.; Bideaux, R. A.; Bladh, K. W.; Nichols, M. C. Handbook of Mineralogy; Mineralogical Society of America: Chantilly, VA 20151-1110, USA. (19) Marcano, D. C.; Kosynkin, D. V.; Berlin, J. M.; Sinitskii, A.; Sun, Z.; Slesarev, A.;

Alemany, L. B.; Lu, W.; Tour, J. M. Improved Synthesis of Graphene Oxide. ACS Nano 2010, 4 (8), 4806–4814.

(26)

25

(20) Ying, Z.; Lin, X.; Qi, Y.; Luo, J. Preparation and Characterization of

Optimization of the graphene oxide synthesis by experimental design

Bachelor Thesis Scheikunde