Characterization of Synthetic dyes by Comprehensive Two-Dimensional Liquid Chromatography

(1)

Bachelor Thesis Scheikunde

door

Studentnummer

Onderzoeksinstituut

Onderzoeksgroep

Verantwoordelijk docent

Dhr. prof. dr. ir. P.J.

(Peter) Schoenmakers

Begeleider

Dhr.

(2)

Analyse van Synthetische kleurstoffen met behulp van 2-dimensionale

vloeistofchromatografie

Ongeveer halverwege de negentiende eeuw was een wetenschapper genaamd William H. Perkin, bezig met de ontwikkeling van een medicijn tegen malaria. Terwijl hij hard aan het werk was op het lab, maakte hij per ongeluk in plaats van werkend medicijn een felle paarse poeder. Deze mooi gekleurde poeder is uiteindelijk geregistreerd als de eerste echte synthetisch gemaakte kleurstof. Perkins product was zo populair dat hij en zijn medewetenschappers zich richtten op de ontwikkeling van meer van dit soort kleurstoffen, en tegen het eind van de negentiende eeuw waren er al meer dan vierhonderd kleuren gemaakt en geregistreerd. Het succes van deze uit het laboratorium afkomstige kleurstoffen zorgde ervoor dat zij al gauw de natuurlijke kleurstoffen, die al sinds de prehistorie gemaakt en gebruikt werden, hadden vervangen. Hun succes was te danken aan het minder arbeidsintensieve en langdurige productieproces en hun desondanks fellere kleuren. Maar waarom is dit interessant voor wetenschappers? Waarom zou een scheikundige zich bezig houden met de analyse van deze stofjes. Om te beginnen is er veel interesse naar het productieproces van die tijd, de negentiende eeuw is immers al ruim honderd jaar geleden. Maar nog belangrijker is de interesse in het degradatie proces die deze kleurstofjes over de loop van tijd ondergaan. Blootstelling aan UV-licht voor misschien wel meer dan honderd jaar, zorgt ervoor dat een kleurstof gaat vervagen. Door te analyseren hoe dit gebeurd, is het misschien mogelijk om dit degradatie proces te vertragen, of om op z’n minst te kunnen achterhalen hoe het er origineel uit heeft gezien. Daarom is dit project opgezet. Het doel van dit project was om een scheidingsmethode te ontwikkelen die de verschillende soorten kleurstoffen (zuur en basisch) kan tegelijkertijd kan scheiden. Dit wordt gedaan met behulp van vloeistofchromatografie. Eerder onderzoek heeft al aangetoond dat het moeilijk is om zure en basische kleurstoffen tegelijkertijd te kunnen scheiden, hierom wordt er tijdens dit onderzoek gebruik gemaakt van 2-dimensionale vloeistofchromatografie. Dit betekent dat we de scheiding gaan laten plaatsvinden op basis van 2 verschillende scheidingsmechanismes, maar dan wel tegelijkertijd! Deze methode wordt niet alleen gebruikt omdat het moeilijk is om deze twee soorten kleurstoffen van elkaar te scheiden, maar ook omdat de hoeveelheid synthetische kleurstoffen zo groot is, dat het onmogelijk lijkt om ze allemaal van elkaar te kunnen scheiden op slecht één chromatogram. Een chromatogram bestaat namelijk uit pieken, die elk een representatief zijn voor een unieke stof. Er is echter maar gelimiteerd ruimte op zo’n chromatogram om deze pieken weer te geven en het chromatogram van een één dimensionale methode heeft gewoon niet genoeg ruimte om de hoeveelheid verschillende pieken die dit onderzoek omvat weer te kunnen geven. Een 2-dimensionaal chromatogram biedt echter wel genoeg ruimte voor een dergelijk grootschalig onderzoek. Daarom is tijdens dit project gewijd aan het

ontwikkelen van een methode waarmee een mix van welgeteld 54 verschillende synthetische kleurstoffen van elkaar gescheiden kunnen worden, wat uiteindelijk is gelukt. Zelfs scheiding van de zure en basische kleurstofjes is hiermee bereikt. Dit is bewerkstelligd door twee scheidingsmethodes (Reversed Phase en Ion

Exchange Chromatography) met elkaar te

combineren en zo een 2-dimensionaal chromatogram te creëren. Dit is jammer genoeg slechts de eerste stap geweest naar het uitvoeren van degradatie studies, welke

juist zo belangrijk zijn. _{Voorbeeld van een chromatogram van een} 2-dimensionale methode (IEC × RP)

(3)

Characterization of Synthetic Dyes by

Comprehensive

Two-Dimensional

Liquid Chromatography

Abstract:

The analysis of early synthetic dyes is interesting for several reasons. But most importantly, it can help with the identification of possible degradation products that have formed over time. The analysis of these degradation products can help with the clarification of the original appearance and/or the composition of these dyestuffs. To analyse these dyestuffs UHPLC can be used. Because of the great sample amount and the difficult separation of acidic and basic dyes simultaneously, comprehensive 2-dimensional LC was applied to increase peak capacity and induce better separation. To separate the dyestuffs based on hydrophobicity and charge, reversed phase ion pair chromatography and strong anion exchange chromatography were used. The application of these two separations methods in a comprehensive IEC×RP setup led to the separation of 54 different dyestuff samples. As well as the simultaneous separation of acidic and basic dyes.

(4)

1. Introduction

1.1 Dyestuffs

When William H. Perkin was trying to synthesize an anti-malarial drug, quinine, in 1856, he accidentally discovered a brilliant purple dye called Mauve, what is now considered the first ‘real’ synthetic dye.[1]_{The synthetic dye was such a great commercial success that within a few years,}

scientists developed even more synthetic dyes and by the end of the century, in 1897, already 404 different dyestuffs had been synthesized. Even though natural colorants had been used since the prehistoric times, the huge success of the synthetic dyes led to their disappearance, as they were quickly replaced by the synthetic ones.[2]

The analysis of these synthetic dyes can be very useful for various reasons. For example, it is helpful to understand not only the production process, but also to explain the degradation processes or even to figure out how to slow down the rate of decay of objects of art. To understand more about the original appearance and obtain even more knowledge of

the historical context of particular objects, an option is to identify the dyestuffs that were used for the dying of textile.[3]

It may also be interesting and useful to identify the degradation products that might have formed over time whilst the dyestuff decay. The identification of these degradation products can help with the clarification of the original appearance and composition of the dyestuffs back when they were first made and used, maybe several hundreds of years ago. One way to analyse these early synthetic dyestuffs and their degradation products is by using ultra high-performance liquid chromatography (UPLC), a technique that enables rapid analysis and quick identifications of the different classes of dyestuffs.[2]

In general, the early synthetic dyestuffs can be divided into different categories. There are acid, basic, mordant and direct dyes. This division is based on the binding mechanism that takes place when the dyestuff binds to textile fibers. Of these four, the most important and frequent class is the acid dye class. This class of early synthetic dyes derive their name from the fact that they are applied to textile fibers in an acid dye bath. The functional group that is responsible for the bond with the textile fiber that forms during this dying process is the acidic group(s) they possess, such as a sulfonic acid and/or a nitroso group. Thus basic dyes, as one would suspect, are applied using basic solutions. The groups responsible for these basic characteristics are one or more primary or secondary amine groups. Direct dyes are used in a neutral environment

and bind directly to the textile fiber, this bond is caused by _{Figure 1: Examples of a Direct, acid,} basic and mordant dye

(6)

hydrophobic interactions. Even though functional groups may be present in the structure of the dyes of this group, they do not take part in the binding process to the fiber. The most complex class is that of the mordant dyes, where a transition metal, like chromium, aluminum, tin or iron, are responsible for the bond between dye and fiber.[16]_{Examples of these different categories of dyes can be seen in}

Figure 1.

1.2 Liquid Chromatography

Ultra high-performance liquid chromatography is the more improved form of the standard High-performance Liquid Chromatography (HPLC). It works along the same principles, except where HPLC is limited to pressures of 400 bar, UPLC can go up to 1200 bar. This enables the use of columns packed with smaller particles and as a result shorter run times are obtained. In liquid chromatography, a mobile phase is utilized, in the form of a flowing liquid, and a stationary phase is present, which consists of sorbents packed inside a column.[4]_{The combination of these two phases}

will induce the separation of the sample components as a result of differences in partitioning between the two phases. These interactions will cause them to each have a different degree of retention and leave the column at varying times.

There are different types of HPLC techniques, that involve different stationary and mobile phases. Some of the most commonly used techniques are: Reversed Phase Chromatography (RPLC), Normal Phase Chromatography (NPLC), Ion-Exchange Chromatography (IEC), Hydrophilic Interaction Chromatography (HILIC) and Size-Exclusion Chromatography (SEC). Each of these techniques rely on a different type of interaction, e.g. with NPLC where the sample components are separated on basis of their affinity for a polar stationary phase.[5]_{But not just molecular properties like polarity can be a}

determining factor in liquid chromatography. When using the SEC mode, the size is all that matters. In size-exclusion chromatography, separation arises from the relative size or hydrodynamic volume of macromolecules.[6]_{Consequently, the sample is separated on size. Both of these techniques are very}

useful when trying to separate very polar components or macromolecules respectively, but seeing that the dyestuffs analysed during this research are neither, we neglect these two techniques and focus on the remaining ones, all of which will be explained extensively later on.

1.3 Application of Comprehensive Two-Dimensional Liquid Chromatography to Analysis as Dyestuffs 1.3.1 Why use Two-Dimensional Liquid

Chromatrography?

Even though there is a lot of knowledge and information on the early synthetic dyestuffs, analyzing them can be difficult. Firstly because most of these early dyestuffs that were created in the 1900s were not created to be perfectly pure to begin with, as long as they worked well enough, they were good enough. On top of that there is also the possibility that the dyes will have decayed

(7)

more degradation products, making these samples hard to separate and identify. Also considering the limited peak capacity of 1D chromatography, and the enormous amount of synthetic dyes that have to be analysed, it was decided that one-dimensional chromatography would not offer sufficient separation power. Previous research of one-dimensional chromatography with these type of samples furthermore showed that it was difficult to separate acidic and basic dyes simultaneously. Therefore comprehensive two-dimensional chromatography (LC×LC) was introduced, to separate these compounds and degradation products on a single chromatogram. Not only to allow for greater peak capacities, but also to be able to separate the acidic and basic dyes simultaneously2]

With the fundamentals of liquid chromatography explained, this section will address the application of the techniques for the analysis of several dyestuffs. It will explain how the combination of two different LC methods will allow for even greater separation and result in a comprehensive two-dimensional chromatogram. While two-dimensional liquid chromatography often has been applied by simply taking interesting cuts (heartcut) from a first dimension and running these through a second dimension, the entire first dimension eluent is subsequently subjected to the second separation dimension in comprehensive two-dimensional chromatography (LC×LC).

Two-Dimensional liquid chromatography is a technique that makes use of two different types of subsequent columns. The columns are placed in series and the analytes will be run through both of them subsequently, having the analytes retain using two varying factors, for example hydrophobicity and size. This will result in a chromatogram with two separate dimensions. An example of a contour plot (2D) can be seen in figure 2.[9] _{For the dyestuffs, reversed phase and anion exchange}

chromatography were chosen to use for the two dimensional analysis. These retention mechanisms were selected due to the hydrophobic backbone and the varying charges that the dyestuffs display.

1.3.2 Orthogonality and eluent compatibility

When performing two dimensional chromatography there are certain aspects that have to be considered. One of those aspects is the orthogonality of the two different separations, where the separation arising from the two separation dimensions is needs to be as statistically independent as possible from one another. Indeed, it is not useful to separate the samples twice based on the same separation principle. And on the other hand the eluent compatibility between the two dimensions is something that needs to be considered, as it can create restrictions on the possible 2-D combinations. This is because the eluent from the first dimension is principally the sample solvent in the second dimension and will not always be compatible with this second dimension, for example, when your second dimension is normal phase and the mobile phase of the first dimension consist of mainly water, the normal phase separation will not be successful, as the samples will elute with the water and have little to no retention.[11]

1.4 Retention Mechanisms

1.4.1 Anion-Exchange Chromatography

Unlike reversed-phase and normal-phase chromatography, analytes are not separated based on hydrophobicity in ion-exchange chromatography. With ion-exchange chromatography the separation is based on differences in electrostatic interactions. It is designed for the separation of differently charged or ionisable molecules. In ion-exchange chromatography, the mobile phase typically is an aqueous buffer and the stationary phase is an inert organic matrix covered with ionisable functional

(8)

groups that carry an exchangeable oppositely charged counter-ion (either cation or anion).[8]_{If the}

analyte contains a charge, opposite to that of the organic matrix, it will displace the counter-ion and adsorption of the analyte to the matrix takes place. There are two types of ion exchange chromatography: weak and strong. When executing weak ion exchange chromatography the samples are pH dependent and will elute when the pH is changed in such a way that they no longer remain charged and desorption takes place, also resulting in the elution of the analyte. Strong ion exchange, on the other hand, depends on the increase of a similarly charged species within the mobile phase. This species will then compete with and eventually displace the analyte bound to the matrix surface. By gradually increasing the salt concentration of the mobile phase, the affinity of interaction between these salt ions will exceed the interaction of the analyte charges, resulting in their displacement and thus their elution. The separation arises due to differences in the strength of the bond between the stationary phase and the varying samples. Should a compound have more than one oppositely charged group, it will bind more strongly to the matrix than when a compound is singularly charged, thus they need higher concentrations of competing salts to be displaced.

The LC technique that was selected for the first dimension separation consist of strong anion exchange. As most of the selected dyestuffs carry varying charges, some of them having negative charges ranging from 1- to 3-, therefor strong anion exchange chromatography appeared to be a good choice for the first part of the separation..

1.4.2 Reversed Phase Chromatography with Ion Pair

Reversed-Phase, Ion-exchange and Hydrophilic Interaction Chromatography are all unique techniques that separate components on different aspects. RPLC is one of the most popular techniques this is because most relevant analyte mixtures comprise of molecules with differences in hydrophobicity. RPLC consists of a relatively polar mobile phase and a hydrophobic (nonpolar) stationary phase. Modern day reverse phase stationary phases mostly consist of permanently bonded hydrophobic groups, e.g. octadecyl (C-18) bonded groups, on a silica support. The separation of this technique is attributed to the solvophobic or hydrophobic interactions that take place.[4][7]

The separation technique that was chosen for the second dimension, is Reversed Phase Chromatography (RPLC). This is because the selected dyestuffs mostly consist of a carbon backbone, build up from aromatic rings and several functional groups. Thus dyestuffs are organic hydrophobic compounds, which will have good interaction with the hydrophobic stationary phase of the Reversed Phase Column (in this case the organic C-18 chains inside the column). This interaction will cause the retention needed to have a good separation of the analytes.[7]

RPLC is a technique that is very dependent on the affinity between the analytes and the organic stationary phase and solvent, but also their affinity with the water. When studying the functional groups that the dyestuffs contain, one can see several charged group. Charges like this will cause extreme polarity, decreasing their hydrophobicity, which will result in loss of retention. It is possible to prevent this from happening by using a so called ‘ion pair’ and thus performing Ion Pair chromatography (ICP). Because when adding an oppositely charged ion to the mobile phase, it will pair up with the negatively charged functional groups of the analytes, neutralising them entirely. This will cause them to retain during the reversed phase chromatography like neutrally charged compounds would, the ion pair can even add to the retention depending on its qualities. Should it have high reversed phase properties (like long carbon tails), it will enhance the reverse phase retention and therefor retain the compounds it has paired with even more than it normally would have.[10]

(9)

But for the execution of IEC×RP, a cation was chosen with low reversed phase interaction, this was done because it is important to take orthogonality into consideration. When taking into account that each negative charge a compound contains will pair up with a cation of the ion pair, the more negative groups they have, the more they will pair up with the ion pair. This ion pairing will result in increased retention per pair, making them separate not only on the reversed phase properties they have, but also based on the amount of negative charges the analyte carries. For regular RPLC this would not be a problem, often even a desired effect, but for the orthogonality of a comprehensive 2D IEC×RP method, it’s necessary to keep the separation on basis of charge to a minimum during the second dimension, as it is the main basis of the separation of the first ion exchange dimension.

1.4.3 Hydrophilic Interaction Chromatography

Hydrophilic Interaction Chromatography is a chromatographic technique that is a variant of NP-LC, where the retention of the analytes is caused by partitioning between analyte and a hydrophilic stationary phase and as mobile phase a relatively hydrophobic eluent, usually something like 5-40% water in ACN. HILIC is used to separate small polar compounds. The retention increases as the polarity of the mobile phase decreases, this results in greater retention for the more polar compounds (which is the opposite of RP-LC). [17][12]

(10)

2. Experimental

2.1 Equipment

For the method development and analysis, an Agilent 1290 Infinity 2D-LC setup was used. The system utilized two Agilent 1290 Infinity Binary Pumps (G4220A), each equipped with a Jet Weaver Mixer 380 µL (G4220-60012), two Agilent 1290 Infinity Thermostatted Column Compartments (G1316C) for both columns, of which the compartment for the second dimension column was equipped with an Agilent 2-position 8-port valve (G4236A), a schematic representation of this valve can be seen in figure 3. The system also featured an Agilent 1290 Infinity Autosampler (G4226A) and two Agilent 1290 Infinity Diode Array Detectors (G4212A) with Agilent Max-Light Cartridge Cell 60 mm (G4212-60007) and Agilent Max-Light Cartridge Cell (10 mm, V(o) 1.0ul) (G4212-60008) flow cells for the first and second dimension respectively. An Agilent 1290 Infinity In-Line Filter (G5067-4638) was located directly preceding the first dimension column.

The system was controlled by a computer with Agilent OpenLAB CDS Chemstation Edition (Rev. C.01.04 [35]). Data was processed and analysed using MATLAB 2013a.

2.2 Ion pair chromatography

For Ion Pair Chromatography (IPC) the following method was applied. The mobile phase consists of a gradient of [A]: Water/Acetonitrile (95:5) with 10 mM tetramethyl ammonium hydroxide (TMA), brought to pH 3.0 with formic acid. And [B]: Acetonitrile/Buffer[A] (95:5). The mobile phase was delivered at a flow rate of 1.000 mL/min. The gradient profile can be seen in table x. Separation was performed on a C-18 column (Agilent ZORBAX Eclipse Plus RRHT 50x4.6mm, 1.8 µL)

The column was placed in a column oven and was kept at a constant temperature of 25⁰C. Figure 3: Visual representation of the

(11)

The method above was performed in the same manner with tetrabutyl ammonium hydroxide (TBA) instead of TMA.

Table 1: HPLC gradient for Ion Pair Chromatography, all changes are linear

Time (min) %A: Water/

Acetonitrile with 10 mM TMA (95:5, v:v) pH 3.0 %B: Acetonitrile/Buffer[A] (95:5, v:v) 0 100 0 0.50 100 0 12.50 0 100 13.00 0 100 15.00 100 0

2.3 Strong Anion Exchange Chromatography

For Strong Anion Exchange analysis the following method was applied. The mobile phase consists of a gradient of [A]: water/acetonitrile (1:1) and [B]: water/acetonitrile (1:1) with 100 mM ammonium sulphate. The mobile phase was delivered at a flow rate of 0.500 mL/min. The gradient profile can be seen in table 2. Separation was performed on a SAX (strong anion exchange) column (150 mm x 5 mm, 2.1 µm I.D.).

The column was placed in a column oven and was kept at a constant temperature of 25⁰C.

Table 2: HPLC gradient for Strong Anion Exchange, all changes are linear

Time (min) %A: water/

acetonitrile (1:1) %B: 100 mM (NH4)2SO4 in water/acetonitrile (1:1) 0 100 0 0.50 100 0 10.50 0 100 14.50 0 100 15.00 100 0 16.00 100 0

2.4 Two Dimensional method

For two dimensional HPLC two different systems were evaluated and combined. The mobile phase of the first system consists of a gradient of [A]: water/acetonitrile (1:1) and [B]: a buffer solution of water/acetonitrile (1:1) with 100 mM ammonium sulphate. The mobile phase was delivered at a flow rate of 0.010 mL/min. The gradient profile can be seen in table 3. For this system, separation was performed on a SAX (strong anion exchange) column (150 mm x 5 mm, 2.1 µm I.D.)

The second system has a mobile phase that consists of a gradient of [A] a buffer solution of 10 mM tetramethylammonium hydroxide (TMA) in water/acetonitrile (95:5) that was brought to pH 3

(12)

with formic acid, and [B] an acetonitrile/buffer (95:5). The mobile phase was delivered at a flow rate of 2.500 mL/min. The gradient profile can be seen in table 4. For this system, separation was performed on a C-18 column (Agilent ZORBAX Eclipse Plus RRHT 50x4.6mm, 1.8 µL)

For the performance of comprehensive two dimensional HPLC the systems are linked with an 8-port valve, containing two 40 µL loops.

Both columns were placed in a column oven and were kept at a constant temperature of 25 ⁰C.

Table 3: HPLC gradient for first dimension (IEC) system, all changes are linear

acetonitrile (1:1) %B: 100 mM (NH4)2SO4 in water/acetonitrile (1:1) 0 100 0 150.00 45 55 170.00 20 80 224.00 0 100 254.00 0 100 264.00 100 0 314.00 100 0

Table 4: HPLC gradient for second dimension (RP) system, all changes are linear

Time (min) %A: Water/

Acetonitrile with 10 mM TMA (95:5, v:v) pH 3.0 %B: Acetonitrile/Buffer[A] (95:5, v:v) 0 100 0 2.2 0 100 2.4 0 100 2.6 100 0 3.0 100 0 2.5 Sample Preparation

Standards of the 54 dyestuffs were analysed and are listed in Table b. In this table the colour index names and numbers are listed, along with other relevant information[1]_.

For HPLC analysis, the pure dyestuffs were dissolved in water:methanol (1:1) at a concentration of 5000 ppm. Prior to injection, they were diluted in water:methanol (1:1) to a concentration of 250 ppm. When a sample was still an unclear solution (one could not see through it), it was diluted further to a concentration of 25 ppm. Of the 54 selected dyestuffs a mixture was made according to the ratios represented in Appendix 1. Of the separate samples, a volume of 2.5 µL was injected, of the mixture a volume of 20 µL was injected.

(13)

3. Results and Discussion

3.1 Ion pair chromatography

When performing reversed phase chromatography with charged samples, using an ion pair can influence the results of the chromatography significantly. This is called Ion Pair chromatography (IPC). For the purpose of demonstrating this, a comparison of a chromatogram with and a chromatogram without the ion tetramethyl ammonium (TMA) can be seen in Figure 4. The chromatogram shows that when an ion pair is present, the separation is significantly improved and results in a higher peak capacity. As explained earlier, this is the result of the neutralising effect that the ion pair of choice (TMA) exhibits.

While the effect of the presence and absence of an ion pair such as TMA is quite evident, changing the type of ion pair is also of a great influence on the eventual retention. To register the effects that different ion pairing can

have on reversed phase chromatography, another experiment was done to demonstrate these effects. For this, the exact same measurements were performed with the ions tetramethylamine (TMA) and tetrabutylamine (TBA) in the mobile

13 Figure 4: Overlay of UV chromatograms of the separation of a mixture of 54 synthetic dye

samples by ion-pair chromatography with 5 mM TBA (orange) and TMA (blue) in the mobile phase at pH 3.0. Gradient analysis . Mobile phase A: buffer, B: acetonitrile. Chromatogram recorded at 254 nm. Flow: 1.0 mL/min, Column: Agilent ZORBAX Eclipse Plus RRHT 50x4.6mm, 1.8 µm. Injection volume: 1.0 µL.

(14)

phase, both the molecular structure of TMA and TBA can be seen in figure 5. As can be seen in this figure, TBA’s four carbon tails are much longer than those of TMA. These long butyl groups will increase the hydrophobicity and thus cause extra reversed phase retention. More than the methyl groups of TMA will. But a large molecule like TBA will also shield off the analyte is has paired with, making the eventual retention less about the compounds own reversed phase interaction and more about the ion pairing that takes place at each negatively charged site. The example of this can be seen in figure 6. When the compounds are paired with TBA, the retention for the paired compounds is significantly bigger than when TMA is used. As can be seen in this chromatogram, the greater part of the compounds elute later on in the chromatogram. The most hydrophobic compounds can be found here, but when using TBA all of the charged compound also elute around this time. While when using TMA, the compounds elute more evenly distributed, leaving us with a greater separation and peak capacity.

3.2 Strong Anion Exchange Chromatography

The first dimension of the comprehensive 2-dimensional LC×LC mechanism is strong anion exchange (SAX) chromatography, as mentioned earlier, this type of HPLC will allow for the separation of the dyestuffs that are carrying one or more negative charges. The ones that contain more negative charges will need higher concentrations of exchanging ions to elute. They stay punt until the gradient of the salt concentration has increased. Therefore they will be retained longer than those with only one or no charge (or positive charge). An example of this can be found in figure 7. The compounds that have no charge or those that are positively charged, exhibit little to no retention by the SAX column whatsoever. They will elute quickly and therefore appear at the beginning of the chromatogram. The charged analytes however, are retained by their negative charges and thus more to the middle and the back of the chromatogram.

As some of the dyestuffs have multiple negative charges (like those with a charge of 3-), a very strong buffer anion was required to compete and exchange with them. Therefore ammonium sulphate was chosen as the anion exchange ion, as it turned out that SO42- is an anion that is suitable

to compete and exchange with these highly negative dyestuffs. Unlike the anion acetate, used regularly with anion exchange chromatography, that was tried previous to sulphate. But it was not strong enough to compete with the triple charged samples.

Figure 6: Overlay of UV chromatograms of the separation of a mixture of 54 synthetic dye samples by ion-pair chromatography with 10 mM TMA (blue) and no ion pair (red) in the mobile phase at pH 3.0. Gradient analysis . Mobile phase A: buffer, B: acetonitrile. Chromatogram recorded at 500 nm. Flow: 1.0 mL/min, Column: Agilent ZORBAX Eclipse Plus RRHT 50x4.6mm, 1.8 µm. Injection volume: 1.0 µL.

(15)

Figure 7: Overlay of UV chromatograms of the separation of a mixture of 54 synthetic dye samples by strong anion-exchange chromatography with ammonium sulfate in the mobile phase. Gradient analysis. Mobile phase A: water/acetonitrile (1:1), B: 100 mM ammonium sulfate in water/acetonitrile (1:1). Chromatogram recorded at 254 nm. Flow: 0.5 mL/min, Column: Agilent PL-SAX 150x2.1mm, 8 µm. Injection volume: 1.0 µL.

(16)

3.3 Van Deemter

To limit the impact of the flow rate on the band broadening of the analyte bands, it is important to determine the optimal flow rate by using the van Deemter equation. This is an equation that predicts the optimum velocity at which the peak broadening is at a minimum. Peak broadening happens due to several properties of a separation. The van Deemter equation can be seen below.

H= A+

B

u

+C ∙u

Here, H represents the height of a theoretical plate (HETP), A the eddy diffusion (m), B the longitudinal diffusion coefficient (m2_s-1_),

_C

_{the mass transfer resistance coefficient and}

_u

the linear velocity in m s-1_.

With this equation it is possible to determine the lowest possible plate height H. The lower H, the more narrow the peaks. The A in this equation stands for the eddy-diffusion. This is the diffusion that is caused by irregularities in the packing of the column. Because of these irregularities in the packing the different pathways an analyte particle can take will have varying pathlengths, which will cause varying retention times for particles of the same analyte, which leads to peak broadening. Would the column have been ideally packed, all of the possible pathways would have been equally long and there would be no eddy-diffusion and thus no peak broadening.

B

, on the other hand, is the diffusion-coëfficient and is the parameter that includes the dispersion throughout the mobile phase. If a measurement is to take a very long time, the dispersion will increase more and more. Therefore it is important to have a velocity that allows for minimal dispersion. Finally, there is the C parameter, which represents the contribution as a result of the resistance to mass transfer. The exact description of the C parameter is rather complex and outside the scope of this study. In short, the C term accounts for the fact that the equilibrium of analytes between the stationary phase and mobile phase is constantly disturbed as the mobile phase is constantly moving.

By using this equation it was determined that for the reversed phase column (the second dimension of the 2D-LC), at all higher flow rates, the theoretical plates arrived at a minimum. Therefor the maximal flow rate that the column could take was used for the second dimension (2.500 mL/min).

(17)

3.4 Comprehensive 2D-chromatography

By applying the two mechanisms discussed above, RP-IPC and SAX chromatography, two-dimensional chromatography was executed.

The first 2D method that was applied resulted in the chromatogram shown in figure 8. This chromatogram was made with a method that contained 5 mM of TMA in the reversed phase mobile phase and 10 µL of the mixture of 54 synthetic dyes was injected. Also the gradient for the first dimension was longer than that of the eventual method. This longer gradient can be seen in Table 5. The result of this is a chromatogram with decent separation and peak capacity.

Table 5: Initial gradient for the first dimension of the 2D-LC measurements

acetonitrile (1:1) %B: 100 mM (NH4)2SO4 in water/acetonitrile (1:1) 0 100 0 270.00 0 100 300.00 0 100 310.00 100 0 360.00 100 0

Figure 9 shows the result of the increase of the injection volume from 10 uL to 20 uL.

0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 175 150 125 100 75 50 25 0

First Dimension retention (min) - Strong Anion-Exchange

S e c o n d D im e n si o n r e te n tio n ( s) I o n -P a ir R e v e rs e d -P h a se C 1 8

Figure 8: Fig. 39 - LCxLC-UV chromatogram of a mixture of 54 synthetic dyes at 254 nm. Run 0005. Injection volume: 10 µL. First dimension: [B.026] Agilent PL-SAX 150x2.1mm, 8 µm, flow: 0.01 mL/min. Second dimension: [B.014] Agilent ZORBAX RRHT Eclipse Plus 50x4.6mm, 1.8 µm, flow: 2.4 mL/min. Measured on Agilent 1290 Infinity 2D-LC. Ammonium sulfate was used as buffer displacement ions in the first dimension in a gradient from 0 to 100 mM. Tetra methyl ammonium hydroxide was used as ion-pair in the mobile phase buffer of the the second dimension at a concentration of 5 mM.

(18)

However, to establish greater visibility of the smaller peaks a greater sample volume (20 µL) was injected. This modification resulted in an increase of intensity of the peaks, as can be seen in Figure 9.

The following step of the optimisation process was to decrease the amount of unused space in the chromatogram. This was done by introducing quick gradients. Therefore a new gradient method was introduced for the first dimension, which reduced its run time by 44 minutes. The modified gradient can be seen in Table 3. By applying this gradient the following chromatogram was established (Figure 10). 0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 175 150 125 100 75 50 25 0

First Dimension retention (min) - Strong Anion-Exchange

S e c o n d D im e n si o n r e te n tio n ( s) I o n -P a ir R e v e rs e d -P h a se C 1 8

Figure 9: LC×LC-UV chromatogram of a mixture of 54 synthetic dyes at 254 nm. Injection volume: 20 µL. First dimension: [B.026] PL-SAX 150x2.1mm, 8 µm, flow: 0.01 mL/min. Second dimension: [B.014] Agilent ZORBAX RRHT Eclipse Plus 50x4.6mm, 1.8 µm, flow: 2.4 mL/min. Measured on Agilent 1290 Infinity 2D-LC. Ammonium sulfate was used as buffer displacement ions in the first dimension in a gradient from 0 to 100 mM. Tetra methyl ammonium hydroxide was used as ion-pair in the mobile phase buffer of the second dimension at a concentration of 5 mM.

(19)

When looking at the chromatogram displayed in Figure 10, one sees that in the middle region of the chromatogram the peaks are blotched. It seems like they are tailing in the second dimension, the reversed phase dimension. To decrease this tailing, that seemed to happen for just the negatively charged analytes, the Ion Pair concentration in the mobile phase was increased to ensure better ion pairing with the negatively charged compound. Thus the concentration was increased from 5 mM to 10 mM of TMA.

By using 10 mM TMA in the mobile phase of the reversed phase of the second dimension the following chromatogram was achieved (Figure 11). This chromatogram displays slightly sharper peaks in the middle regions and high peak capacity in the earlier regions of the first dimension (40 to 80 minutes) and throughout all of the second dimension. As the measurement progresses, the peaks lose their sharpness and become more stretched out, which is due to the common qualities of ion exchange column, that rarely produces sharp peaks for strongly charged components. The figure also displays the effective gradients, which is the method gradient corrected with the dwell time.

The downside of increasing the TMA concentration is the appearance of an extra system peak that is very present in the entire chromatogram.

Figure 10: LC×LC-UV chromatogram of the separation of a mixture of 54 synthetic dye with effective gradients. Dual gradient analysis. First dimension: Strong Anion-Exchange, Agilent PL-SAX 150x2.1nm, 8 µm, 0.01 mL/min. Second Dimension: Ion-Pair Chromatography, Agilent ZORBAX Eclipse Plus RRHT 50x4.6mm, 1.8 µm, 2.4 mL/min. Chromatogram recorded at 254 nm. Injection volume: 20.0 µL. Loopsize: 40 µL.

(20)

Figure 11: LC×LC-UV chromatogram of the separation of a mixture of 54 synthetic dye. Dual gradient analysis. First dimension: Strong Anion-Exchange, Agilent PL-SAX 150x2.1nm, 8 µm, 0.01 mL/min. Second Dimension: Ion-Pair Chromatography, Agilent ZORBAX Eclipse Plus RRHT 50x4.6mm, 1.8 µm, 2.4 mL/min. Chromatogram recorded at 254 nm. Injection volume: 20.0 µL. Loopsize: 40 µL.

(21)

3.5 Hydrophilic Interaction Chromatography

HILIC was performed to see whether it would be a viable option for the 2-dimensional chromatography. But the results showed that there was not enough interaction of the samples with the HILIC column to establish enough retention for good enough separation. Thus HILIC was discarded.

(22)

4. Conclusion and Perspectives

Anion-exchange, reversed-phase and HILIC columns were studied for the separation of old synthetic dyes samples. None of these columns yielded full separation of all of the relevant synthetic dyes. As a result, a study to the combination of two separation dimensions was initiated and a comprehensive two-dimensional liquid separation system was successfully developed. The system utilized an Agilent PL-SAX (150 mm x 2.1 mm, 8 µm) column for the first dimension and a Agilent ZORBAX Eclipse Plus RRHT (50 mm x 4.6 mm, 1.8 µm) column for the second dimension. The obtained peak capacity obtained by combination of these two columns was approximately 3000.

The LCxLC system was successfully used for the analysis and separation of 54 synthetic dyes samples. By using this 2D-LC method the simultaneous separation of basic and acidic dyes was established. It was furthermore concluded that using HILIC will not establish enough retention to be a viable option for analysing these dyestuffs.

Now that a separation has been established the 2-D method is open to optimization. A possible way to do this is by reducing the run-time even more, which is still over 5 hours with the current method. This can be done by speeding up the gradients even more often, as there are still ‘empty spaces’ left in the chromatogram. Other than this it can be relevant to investigate the ion-pair effects more or even investigate alternative separation mechanisms, as RP and IEC are not the only mechanisms applicable to these dyestuffs. Lastly, and most importantly, an important perspective of this research, the original perspective in fact, is to do degradation studies. By manually degrading the samples with UV-light for example, one can apply the developed 2-D method to analyse the degradation.

(23)

5. References

[1] Christie, R. (2014). Colour chemistry. Royal Society of Chemistry.

[2] Erdmann, H. (1902). G. Schultz und P. Julius. Tabellarische Übersicht der künstlichen organischen Farbstoffe. Vierte Auflage, herausgegeben von Dr. Gustav schultz. Berlin, R. Gaertners Verlag (H. Heyfelder), 1902. Angewandte Chemie, 15(30), 767-767.

[3] van Bommel, M. R., Berghe, I. V., Wallert, A. M., Boitelle, R., & Wouters, J. (2007). High-performance liquid chromatography and non-destructive three-dimensional fluorescence analysis of early synthetic dyes. Journal of Chromatography A, 1157(1), 260-272.

[4] Dong, M. W. (2006). Modern HPLC for practicing scientists. John Wiley & Sons.

[5] Snyder, L. R., Kirkland, J. J., & Dolan, J. W. (2010). Normal‐Phase Chromatography.

Introduction to Modern Liquid Chromatography, Third Edition, 361-402.

[6] Barth, H. G., Jackson, C., & Boyes, B. E. (1994). Size exclusion chromatography. Analytical

chemistry, 66(12), 595R-620R.

[7] Snyder, L. R., Dolan, J. W., & Gant, J. R. (1979). Gradient elution in high-performance liquid chromatography: I. Theoretical basis for reversed-phase systems. Journal of Chromatography

A, 165(1), 3-30.

[8] Cummins, P. M., Dowling, O., & O’Connor, B. F. (2011). Ion-exchange chromatography: basic principles and application to the partial purification of soluble mammalian prolyl oligopeptidase. In Protein Chromatography (pp. 215-228). Humana Press.

[9] Jandere, P. (2007) Column Selection for Two-Dimensional LCxLC. LCGC Europe, 20(10) (pp. 510-525).

[10] Cui, L., Wen, J., Zhou, T., Wang, S. and Fan, G. (2009), Optimization and validation of an ion-pair RP-HPLC-UV method for the determination of total free iodine in rabbit plasma: application to a pharmacokinetic study. Biomed. Chromatogr., 23: 1151–1159. doi: 10.1002/bmc.1237

[a] Dugo, P., Cacciola, F., Kumm, T., Dugo, G., & Mondello, L. (2008). Comprehensive multidimensional liquid chromatography: theory and applications. Journal of

Chromatography A, 1184(1), 353-368.

[11] Schoenmakers, P. J., Vivó-Truyols, G., & Decrop, W. M. (2006). A protocol for designing comprehensive two-dimensional liquid chromatography separation systems. Journal of

Chromatography A, 1120(1), 282-290.

[12] Stoll, D. R., Li, X., Wang, X., Carr, P. W., Porter, S. E., & Rutan, S. C. (2007). Fast, comprehensive two-dimensional liquid chromatography. Journal of Chromatography A, 1168(1), 3-43.

(24)

[13] Alpert, A. J. (1990). Hydrophilic-interaction chromatography for the separation of peptides, nucleic acids and other polar compounds. Journal of chromatography A, 499, 177-196.

[14] Society of Dyers and Colourists., & American Association of Textile Chemists and Colorists. (1971). Colour index. Bradford.

[15] Small, H. (2013). Ion chromatography. Springer Science & Business Media.

[16] Hofenk de Graaff, J. H., Roelofs, W. G., & Bommel, M. R. V. (2004). The colourful past:

origins, chemistry and identification of natural dyestuffs. Archetype publications;

Abegg-Stiftung.

[17] Hemström, P., & Irgum, K. (2006). Hydrophilic interaction chromatography. Journal of

(25)

Appendix 1: List of selection of synthetic dyes

Color Code Name C.I. Name Charge

Mix proportion

1735 Indigo Carmine Acid Blue 74 2- 4

2028 Alizarine Red S Mordant red 3 1- 1

2030 Indigotin vat blue 1 0 4

3016 Orange II Acid orange 7 1- 1

3034 Alizarine mordant red 11 0 4

3147 Alizarin yellow Mordant yellow 1 1- 1 3203 Diamond green B basic green 4 1+ 2 3231 Diamond green G basic green 1 1+ 4

3289 Chrysoidine basic orange 2 1+ 1

3290 Azo Fuchsine 6B acid violet 7 2- 4

3408 Methyleen Blue Basic Blue 9 1+ 1

3652 Auramine basic yellow 2 0 1

3712 Nigrosin acid black 2 ? 4

3742 Crystal violet basic violet 3 1+ 2

3824 Uranine A acid yellow 73 2- 1

3835 Crystal ponceau 6R acid red 44 2- 2

3855 Orange GG acid orange 10 2- 4

4104 Rhodamine B Basic Violet 10 1-/1+ 1

4328 Eosine A Acid Red 87 2- 1

4335 Fuchsin basic violet 14 1+ 4

4412 Chrysoin acid orange 6 1- 1

4434 Methyl Violet 2B Basic Violet 1 1+ 4 4518 Water Blue IN Acid Blue 93 3-/1+ 4

4711 Rhodamine 6G Basic Red 1 1+ 1

4958 Croceine Orange G Acid Orange 12 1- 1

4989 Fast Red AV Acid Red 88 1- 2

4995 Fast Red B Acid Red 17 2- 4

5000 Ponceau RR Acid Red 26 2- 4

5027 Azo Flavine 3 R acid orange 1 1- 4

5302 Patent Blue V Acid Blue 3 ? 2

5305 Erythrosine acid red 51 2- 1

5316 Tartrazine Acid Yellow 23 3- 4

5347 Metanil yellow acid yellow 36 1- 4

5349 Amaranth acid red 27 3- 1

5365 Flavazine L Acid Yellow 11 1- 4

5367 Amido Naphtol Red G Acid Red 1 2- 4

5528 Congo red direct red 28 2- 4

5641 Yellowish light green SF acid green 5 3-/1+ 4 5706 Brilliant Yellow Direct Yellow 4 2- 4

6344 Ponceau 3RO acid red 25 2- 2

6531 Quinoline Yellow acid yellow 36 0 4

6556 Cotton scarlet acid red 73 2- 4

(26)

6887 Orange IV Acid Orange 5 1- 2 6923 Naphthol yellow S acid yellow 1 2- 4

6928 Orange I Acid Orange 20 1- 1

7088 Safranine T Basic Red 2 1+ 1

7098 Victoria Blue B Basic Blue 26 1+ 2 7177 Fast acid magenta B acid red 33 ? 4

7690 Wol red B acid red 115 2- 4

7759 Martius yellow acid yellow 24 0 1

7966 Murexide Mordant Dye ? 4

8511 Vesuvine BA Basic Brown 1 ? 4