EXPERIMENT 1
HIGH PERFORMANCE LIQUID CHROMATOGRAPHY
This is a two-week experiment and you will follow the schedule outlined below. The prelab assignment for this experiment involves using a computer simulation of HPLC instrumentation. The week before you begin Experiment 1, please see the Manager of Advanced Undergraduate Laboratories and check out the simulation diskette. This diskette must be returned when you begin Week 2 of this experiment.
EXPERIMENT SCHEDULE
WEEK 1
* Discussion of the results of your pre-lab assignment. * Introduction to the HPLC instrumentation. * Preparation of solvents, analytes and instrument startup. * How to perform a chromatographic analysis. Experiment 1.1. Investigate
chromatographic response as a function of detector wavelength. * Set up autosampler to perform overnight data collection for two experiments: Experiment 1.2. Theories regarding the partitioning process in HPLC.
Experiment 1.3. Chemometric analysis of retention data.
* You MUST come to lab THE NEXT DAY and collect the data from Experiments H.2 and
H.3. BEFORE YOU COME TO LAB FOR WEEK 2 you should use this data to perform
the following calculations: I. Complete the analysis of Experiment 1.2 - partitioning experiment. II. Calculate lnk' values for each solute in Experiment 1.3. WEEK 2
* Discussion of your results from Experiment 1.2 - partitioning experiment.
* Chemometric analysis of retention data from Experiment 1.3.
* Experiment 1.4: Problem-Solving: Optimized Separation of a Mixture.
You will be given a mixture containing some of the following solutes: phenol, 4-nitrophenol, p-cresol, 2,5- dimethylphenol (xylenol), methyl benzoate, anisole, benzene, phentole, toluene, p-xylene, 4- methoxyphenol, 4-nitrotoluene, fluorobenzene. You will use the remainder of the lab period (1.5-2 hours) to optimize the chromatographic separation of this mixture. You will need to select and optimize the mobile phase composition, and to decide whether and how to use a mobile phase gradient. The goal is to achieve optimum separation with maximum efficiency of time and resources (solvents).
EXPERIMENT 1
HIGH PERFORMANCE LIQUID CHROMATOGRAPHY
PURPOSE
This experiment has two goals: (i) to introduce you to the power of high performance liquid chromatography (HPLC) as an analytical tool and (ii) to investigate the use of important computational strategies in the analysis of multivariate data. You will perform four experiments:
EXPERIMENT 1.1: Introduction to chromatographic separation by HPLC. Investigation of chromatographic response as a function of detector wavelength.
EXPERIMENT 1.2: New theories regarding the partioning process. Retention data will be compared to partition coefficients for compounds partitioned between an octanol-like phase and water. The octanol-water system is commonly used because it mimics the partitioning which occurs between cell membranes and aqueous phases in living organisms.
EXPERIMENT 1.3: A chemometric technique known as principal component factor
analysis (PCFA) will be used to search retention data for trends which may not be apparent upon simple visual inspection. These trends may be employed for subsequent prediction of retention behavior.
EXPERIMENT 1.4: Problem-solving: Optimized separation of a mixture.
INTRODUCTION AND THEORY
High performance liquid chromatography (HPLC), or liquid chromatography (LC) is an extremely important analytical technique which is used for the separation, identification and quantitation of complex mixtures. In chromatography the chemical components in a mixture are carried through a stationary phase by the flow of a gaseous (gas chromatography) or a liquid (liquid chromatography) mobile phase. Separation in liquid chromatography is achieved by means of differences in the interactions of the analytes with BOTH the mobile and stationary phases. To plan a separation by LC, the user must select both a stationary phase and a mobile phase appropriate to the analytes under investigation. In addition, the user must identify chromatographic conditions that will maintain the sharpness of analyte bands as a sample moves through the stationary phase column to the detector.
The mobile phase must be chosen to ensure solubility of the sample solutes. For the stationary phase, microparticulate silica (bare or chemically modified) is used almost universally because its high surface area accentuates the differences in solute-stationary phase interactions. The use of a stationary phase that interacts strongly with solutes relative to solute mobile-phase interactions will result in very long retention times, a situation which is not analytically useful. Hence the stationary phase must be selected so as to provide weak to moderate solute interactions relative to those in the mobile phase. As a consequence, the nature of the solute governs the type of LC selected. The stronger interactions should occur in the mobile phase to ensure sample solubility and ready elution, while the stationary phase should be responsive to more subtle differences among the solutes. For example, polar neutral compounds are usually better analyzed using a polar mobile phase together with a nonpolar stationary phase that distinguishes subtle differences in the dispersive character of the solutes. One of the powerful aspects of HPLC is that the mobile phase can be varied to alter the retention mechanism. Modifiers can be added to the mobile phase to control retention. For example, pH is an important variable in aqueous mobile phases.
The conventional classification of separation modes in LC is that of liquid-solid, liquid-liquid, ion-exchange, and size-exclusion chromatography. In reality, the precise classification of a system into one of the first two categories can be difficult because the experimental evidence often suggests a mixture of retention mechanisms. Four general classes of LC can be distinguished from this perspective.
1. Normal-phase chromatography calls for the use of a polar stationary phase in conjunction with a non-polar (dispersive) mobile phase.
2. Reverse-phase chromatography, the opposite possibility, calls for the use of a non-polar stationary phase and a polar mobile phase (composed of one or more of the solvents water, methanol, acetonitrile, and tetrahydrofuran). The majority of chromatographic analyses falls into one of these two classes.
3. Ion-exchange chromatography involves ionic interactions. In this case the mobile phase must support ionization to ensure solubility of ionic solutes. The stationary phase must also be partially ionic to promote some retention. Consequently, the interactions with the stationary phase are strong, and this is usually reflected in longer analysis times and broad peaks.
4. Size-Exclusion chromatography involves separations based on molecular size alone and ideally requires that there be no energetic interaction of the solutes with the stationary phase.
In this experiment you will use reverse-phase LC, using a C18H37 - derivatized silica stationary phase, to investigate the retention behavior of a series of benzenoid derivatives. The interactions of solutes with an alkyl- bonded phase, such as the C18 bonded-phase silica, are dispersive, and a polar mobile phase is conventionally used in conjunction with these stationary phases. The non-polar stationary phase makes these systems very useful for separating organic compounds with slight differences in the backbones or side-chains. Clearly, the ability to alter the mobile phase to influence the retention of a solute presents the analyst with a dilemma; how to select the best mobile phase from a very large number of possibilities. Two approaches are used. The first is to systematically test large numbers of mobile phases and then select the one in which the best separation is achieved. The second is to use known trends in retention properties in a few mobile phases to predict retention in many mobile phases.8 The use of computer-based experimental design and data analysis has led to significant advances in both approaches. Simplex optimization methods (similar to those discussed in Experiment F) have been used to optimize mobile phase composition in the first approach. Chemometric analysis of multivariate retention data has led to advances in the second approach. Because reliable prediction schemes have yet to be developed, the first approach is usually employed when attempting a separation by LC.
EXPERIMENT 1.2: NEW THEORIES REGARDING THE PARTITIONING
PROCESS IN HIGH PERFORMANCE LIQUID
CHROMATOGRAPHY - (HPLC)
INTRODUCTION AND THEORY
Cell membrane studies have made a contribution to our understanding of the partitioning process in chromatography. Within a living organism, the interaction of chemicals with cells may be dependent on partitioning between the aqueous biosystem and the hydrophobic cell membrane. For example, Richett1 found that toxic effects of ethers, alcohols, aldehydes, and ketones are inversely proportional to their solubility in water.
A model has been developed to explain this phenomenon. This model takes an unsubstituted compound, such as benzene, as a reference compound. Mathematically, the phenomenon has been described by Hansch et al.4,5,6 as
log {K(SX) / K(SH)} = (X). [1]
K(SH) and K(SX) are the partition coefficients of SH, an unsubstituted reference compound, and SX, a substituted reference compound, respectively; (X) is the hydrophobic substituent constant, which describes the contribution of substituent X to the lipophilicity of structure SH when X replaces a H atom in SH; and the constant reflects the characteristics of the solvent pair used in determining the partition coefficient. The common two-phase system used as a model for the cell-membrane/aqueous-phase interaction is an octanol/water system.
In this experiment you will analyze mixtures containing benzene and substituted benzene derivatives by HPLC. On the column, these compounds will be partitioned between the stationary phase, octadecyl (C 1 8) silica, and the mobile phase, methanol/water. It is theorized that the same lipophilic character which plays a role in cell- membrane interaction will play a role in the chromatographic partitioning process occurring in column chromatography. Retention data will be used to calculate partition coefficients for different components in the analyte. This will allow calculation of Hansch-model parameters for the functional groups in the substituted benzene derivatives.
Uracil, a compound which does not interact with the C18 stationary phase, is used to determine the simple retention time (to) of a non-partitioning species. For compounds which do partition, the net retention time (tn) is given by
tn = tr - to [2]
where tr is the measured retention time.
The partition coefficient K for a given component is proportional to its value of tn and therefore using equation (1) we may write:
log [3]
This allows calculation of parameters for the different aromatic substituents (X). These values may then be compared with values determined in the octanol/ water system used in cell interaction studies.
EXPERIMENT 1.3: CHEMOMETRIC ANALYSIS OF RETENTION DATA
Note: Chemometric analysis of multivariate data is also discussed in the theory section of Experiment B. The mathematical basis of factor analysis is outlined in Appendix A at the end of Experiment B.
INTRODUCTION
Using computerized instrumentation, it is now possible for the chemist to study complex multicomponent systems governed by numerous independent and/or interdependent variables. This has created a new problem: how do we extract useful information from large data sets? A sub-discipline of analytical chemistry termed CHEMOMETRICS has been developed to solve complex multivariate chemical problems.
In chemometrics, it is important to understand the concept of MULTIVARIATE data. It is probably overly simplistic in most real chemical problems to assume that a measured property is determined exclusively by a single variable. In numerous types of analysis, however, we do make this assumption or at least try to regulate the system so that only a single variable influences our measurement. For example, we assume that only a single species absorbs appreciably at a particular wavelength (as in atomic absorption spectrometry) and that only a single species reacts with a titrant in acid-base titrations. In real systems, however, we frequently find that more than one variable affects the measurements we make. An excellent example is the measurement of the activities of certain enzymes, which may depend on pH, temperature, concentration of enzyme, concentration of substrate, competing reactions, etc. When confronted with so complex a chemical system, the analyst has traditionally attempted to control all but one variable, thereby simulating what is called a univariate system. When the system is a true unknown, i.e. the analyst does not know what variables influence the system, a more sophisticated approach is required. If more than a few variables affect the system, it is appropriately called a MULTIVARIATE system. Techniques developed to deal with multivariate data include multiple regression, pattern recognition, partial least squares modelling and the technique you will use in this experiment, factor analysis.
PRINCIPAL COMPONENT FACTOR ANALYSIS
Factor analysis is a powerful mathematical technique based on eigenvalue-eigenvector analysis of a data matrix.7 Essentially factor analysis involves the transformation of the
n- orthogonal axes (representing the variables) that span the data space into n new axes (representing linear combinations of the variables), such that these new axes lie along the direction of maximum variance. This concept can be visualized with the help of the two-dimensional example in Figure 1. It is obvious from Figure 1(a) that the direction of maximum variance lies neither along the x-axis nor along the y-axis, but rather along some direction between them, i.e. along some combination of x and y. Similarly, the axis describing the direction of the second greatest amount of variation away from the principal direction of variance is coincident neither with x nor with y. Figure 1(b) depicts the identical distribution to that of Figure 1(a), but referred to a new set of axes f1 and f2, such that f1 represents the direction of greatest variance and f2 the direction of greatest variance orthogonal to f1. Now if the variation along f2 is minimal compared to that along f1, then it could justifiably be argued that the combination of x and y represented by f1
FIGURE 1: The relation between data distribution and (a) variable axes x and y and (b) factors f1 and f2. Note that the origin of the factor space actually lies at the midpoint of the data distribution; the factor axes in (b) are merely intended to emphasize the directions of variance.
is adequate in describing the distribution of data points in the two-dimensional space spanned by x and y. In other words, a reduction in the dimensionality of the data point distribution from two to one has been achieved.
In the case of an n-dimensional problem, factor analysis yields up to n orthogonal factors (linear combinations of the original variables) lying along, respectively, the axis of largest variance, the axis of second largest variance, the axis of third largest variance, and so on. Often the number of factors needed to describe, say 95% of the sample variance is less than n, so that factor analysis essentially affords a technique whereby the dimensionality of the parameter space can be reduced. The task of the chemist is then to interpret, in chemical terms, those factors extracted out of the data matrix by factor analysis.
Principal Component Factor Analysis (PCFA) extracts, from the data themselves, the axes (or eigenvectors) that best span the data matrix. The first eigenvector is computed such that the sum of the magnitudes of the projections of all points on that vector is a maximum, i.e. as much variation in the data as possible lies along the direction of the first eigenvector. The projection of each data point on the eigenvector will be the coordinate of that datum along the vector. The second eigenvector is chosen, orthogonal to the first, so that as much as possible of the remaining variation lies along this vector. Subsequent vectors and the projections of data thereon are constructed in like manner until all the variation in the data can be described in terms of the extracted eigenvectors and associated coordinates along these vectors. The data matrix is thus decomposed into two matrices, the row cofactor matrix and the column cofactor matrix, which are composed of the coordinates and eigenvectors respectively:
[D]PCFA = [R]abstract [C]abstract
data row column (4)
matrix matrix matrix
NOTE: The mathematical basis for Factor Analysis is explained in a tutorial in Appendix A following Experiment B.
Since this factor analytical solution is purely mathematical and is devoid of physical meaning, these matrices are called abstract matrices. The columns of [R]abstract are called abstract factors.
Since the abstract solution should involve a physically meaningful number of factors, determination of n, the correct factor "size", is a particularly important step. As a result of this step we obtain an estimate of the complexity of the data space, information normally lacking even for the simplest chemical problems.
Factor analysis can be successfully applied when the measured response of a system can be modeled as a linear sum of independent terms. We will take just such an approach to model the liquid chromatographic behavior of several solutes in several different solvents. If we measure capacity factors (see below) by chromatographing the compounds in several different solvents, we create a data matrix consisting of capacity factors for each compound at each solvent composition.
The capacity factor (k', called "k prime") is defined as follows:
k' = (5)
where tr is the time between the injection of the sample and the peak for the solute on the chromatogram and t0 is the time for the elution of an unretained compound (uracil in this experiment). The capacity factor is a derived unit which allows comparison of LC behavior of solutes in different chromatographic systems (in fact there is some controversy as to whether this measure is truly comparable for species with very different chemical characteristics). In the case of capacity factors, it is convenient to use the natural logarithm of k' (lnk') for the factor analyses. Thus for r compounds chromatographed in c solvents, the resulting data matrix will have the form:
Solute Solvent 1 Solvent 2 ... Solvent c
Compound 1 lnk'11 lnk'12 ... lnk'1c
Compound 2 lnk'21 lnk'22 ... lnk'2c
. . . . .
. . . . .
Compound r lnk'r1 lnk'r2 ... lnk'rc
where the subscripts define the row and column positions of the data within the matrix.
In using factor analysis, we model the data matrix as being composed of measurements where each measurement is the sum of an unknown number of independent terms. For example, we may think of lnk' as consisting of several parts, each due to a particular type of interaction. Examples of the types of interactions which may influence our system are dispersion forces, dipole-dipole, and ion-dipole interactions, solubility of the solute in the mobile phase, acidity of phenolic groups and symmetry effects.
In the current experiment we will determine the influence of mobile phase composition on retention (i.e., lnk') and will therefore consider the problem as interactions between the solute and solvent rather than between the solute and stationary phase. In a simple equation format we express the component model of lnk' as
lnk' = lnk*1 + lnk*2 + ... + lnk*n (6)
or more compactly as
lnk' = lnk*x (7)
where the subscripts refer to the xth interaction generating a contribution (lnk*) to the overall lnk' for the solute. To take our model a step further, we presume that each component contributing to the overall lnk' is the product of some property of the solute (called the row cofactor, R, since solute is the row designee) and some property of the solvent (called the column cofactor, C, by similar reasoning). Hence, equation (7) is expanded to
lnk'ij = RixCxj (8)
where lnk'ij is the datum corresponding to the ith compound (solute) in the jth solvent composition and x designates the row or column cofactor representing the xth of n interactions. In equation (8), Rix is the row cofactor specific to compound i for interaction x; this cofactor indicates the magnitude of the influence on lnk' for the compound caused by interaction x. Likewise, Cxj is the cofactor indicating the importance of interaction x in solvent j. The abstract cofactors conceptually separate the interactions into solute and solvent components.
In matrix notation we can succinctly express the decomposition of the data matrix by equation (4). The row cofactors in the abstract row cofactor matrix are, as mentioned before, indicative of properties of the solutes themselves, so we can use them to differentiate between solutes of various types. The results of our factor analysis will give use the matrix [R] in a format similar to the following table:
ABSTRACT ROW FACTOR MATRIX
SOLUTE X 1 2 3 ...
SOLUTE 1 .49654 .2608 -.51959 ...
SOLUTE 2 .30191 .2987 -.31958 ...
SOLUTE 3 .11954 .3235 -.16482 ...
SOLUTE 4 -.04242 33083 -.05541...
. . . . . . . . . . . . . . .
In practice, the factor analysis will generate as many factors (i.e., columns in the row matrix) as there are columns in the original data matrix. By examining the EIGENVALUES (produced by the factor analysis program) for the matrix, we can determine how many factors are significant to the system of interest. If the number of significant factors is three or less, the data can be graphed and visually inspected for trends.
In this experiment, in which four solvent compositions are used, each solvent could be assigned to an axis of a coordinate system and the lnk' values could be plotted. The final map of the data could be visually examined for trends. Unfortunately, the coordinate system described above has a dimensionality higher than three and is not amenable to visualization. One of the most important outcomes of performing factor analysis is that the dimensionality of the data is often reduced.
As stated previously, the eigenvectors account for the variation in the data. The first of these accounts for as much of the data variation as possible, the second accounts for as much of the remaining variation as possible, and so on. Associated with each eigenvector is an eigenvalue, which indicates the importance of the vector. When the eigenvalue for a particular vector is divided by the sum of all of the eigenvalues, the result gives the fraction of the overall variance accounted
FIGURE 2: FACTOR 3 VS. FACTOR 1 FOR SEVERAL SOLUTES
for by the eigenvector. By cumulatively adding these fractions together we can find out how many vectors are required to reproduce the data to within experimental error. In this experiment we will estimate the percent uncertainty in measured retention times. Factors which contribute to the overall variance in the data by an amount less than this experimental error, will be considered insignificant. The number of significant eigenvalues (and eigenvectors) indicates the underlying dimensionality of the data.
For our particular system, the plot will be constructed using the abstract row factor matrix. Each of the cofactors in the matrix will be treated as a coordinate in a Cartesian coordinate system (e.g., column 1 becomes X, column 2 becomes Y, etc for each solute), and we can map the data set in the "space" of the row cofactors. Our hope is that compounds with similar properties will show up in the map as groups or "clusters" in the space.
An example of a cluster plot may be found in Figure 2. Note that the three types of compounds (nitro-, carbonyl-containing and alcohols) in the map are "clustered" according to the functionality present in the molecule, as shown by the broken borders. This map also shows that the two heterofunctional compounds, -- p- nitrobenzaldehyde and p-nitroacetophenone, lie in the intersection of the nitro and carbonyl clusters. Further information about factor analysis is available in reference [7] and in Experiment B.
EXPERIMENT 1.4: OPTIMIZATION OF SEPARATION BY HPLC
We expect you to use your knowledge of HPLC and the HP 1050 instrument (gained during the first week of Experiment 1.1) to optimize the separation of an unknown mixture in the second week.
INTRODUCTION
The objective of optimization in chromatography is to find the set of conditions required to achieve the "best" separation in a given situation. Definitions of the goals of HPLC optimization might include the following:
* to solve the problem with minimum cost
* to achieve a separation goal with the minimum of time and effort
* to produce the "best" separation possible of a given sample
* to select the mobile phase and column stationary phase combination that gives baseline separation in a given time
* to achieve an optimum combination of elution speed, sample size and solute resolution
One of the first steps in the optimization process is the definition of the objectives. Instrumental aspects must also be taken into account. The separation will depend heavily on the column and detector performance and the capabilities of the pumping system. Also important will be any limitations or advantages that accrue from the integrator or data station. By and large, and certainly in this experiment, selection of the mobile phase composition is the primary objective. Your goal is choice of mobile phase composition and elution timetable to achieve adequate separation of the analytes within a reasonable analysis time. A brief introduction to the concept of gradient or programmed elution is appropriate at this stage.
GRADIENT ELUTION
Complex mixtures often include solutes with a wide range of retention times. To ensure resolution of the less retained compounds, a weak mobile phase is frequently required, with the result that strongly retained compounds elute only after a prohibitively long time. The problem can be solved by gradient development, a process in which a strong mobile phase is mixed in increasing proportions with the weak phase to produce a progressively stronger mobile phase. With gradient elution, weakly retained compounds are resolved under conditions of the weak mobile phase, and the subsequent increase in the mobile phase strength shortens the retention time of compounds that interact more strongly. The use of gradient elution in LC is analogous to temperature programming in GC separations (see Experiment F).
The mobile phase gradient can be generated discontinuously by a series of discrete additions or continuously (and even in non-linear fashion). The actual changes in phase composition are accomplished by precision metering from the different solvent resevoirs (as described in the instrumentation section).
Gradient elution often permits all components to be eluted with peaks of similar shape. This can be compared with many isocratic (constant mobile phase composition) separations where early eluting peaks are narrow and often densely packed whereas later eluting peaks are broad, often with excessively long retention times.
EXPERIMENTAL SECTION
INSTRUMENTATION
This experiment uses a Hewlett-Packard 1050 high performance liquid chromatographic system which consists of the following components (listed from top to bottom as you view the instrument):
MOBILE PHASE RESEVOIRS
The HP 1050 HPLC has four 1 liter gas-tight glass solvent resevoirs labeled A, B, C, and D. Resevoirs, A, B, and C contain water, methanol, and acetonitrile respectively, and resevoir D is not used. All solvents must be filtered through 0.45µ filters prior to introduction to the resevoirs. It is very important to remove dissolved gases and particulate matter from the solvents. The former produce bubbles in the column and thereby cause band spreading. In addition, both bubbles and dust interfere with detector performance. The solvents are degassed by vigorous sparging, in which dissolved gases are swept out of solution by fine bubbles of an inert gas (helium in this case) that is not soluble in the mobile phase. Once degassing is complete, the resevoirs remain pressurized using a slow flow of helium during experiments.
COLUMN
The 100 mm x 4.6 mm i.d. stainless steel chromatographic column is packed with a homogeneous bed of uniform spherical silica gel particles which are 5 microns in diameter with a pore size of 120 Å. Some of the silanol (SiOH) groups on the surface of each particle have been derivatized with an eighteen carbon alkane (C18H37) chain which gives the surface a certain degree of hydrophobic character. Octadecyl modified silica is one of the most common stationary phases used in reverse-phase LC. Chemical modification of the surface is preferred to coating by adsorption, because the latter surface may be removed by dissolution in the mobile phase.
Separation in LC is achieved by means of differences in the interactions of the analytes with both the mobile and stationary phases. In this experiment you will perform reverse-phase LC using a nonpolar stationary phase (C18) and a polar mobile phase.
DETECTOR
Methods of detection used for HPLC include UV absorption, IR absorption, fluorescence, refractive index, electrical conductivity, mass spectrometry, electrochemical and radiochemical. Many modern LC instruments use diode array UV detectors. As well as the traditional two-dimensional presentation of detector signal against time, it is possible to display the UV spectra of the components giving rise to different chromatographic peaks. Differences in the spectrum between the front and tailing edges of a peak can be used to detect unresolved impurities. Such spectra can be captured in 'real time' without the need to interrupt the eluent flow. It is possible to construct pseudo-three-dimensional, or isometric plots, of absorbance, wavelength and time. (Note: This type of plot can be generated using the computer simulation used in your pre-lab assignment. You are encouraged to investigate this.)
The HP 1050 instrument uses a diode array UV detector with a deuterium lamp source. All wavelengths in the range 190-600 nm are monitored simultaneously during elution. Unfortunately, the HP 1050 LC data system is not capable of analyzing more than three discrete wavelengths during the course of a chromatographic run. By careful selection of the wavelength(s) used for detection, the response of a particular analyte can be enhanced at the expense of possible interferences.
PUMP
The requirements for the pumps used in liquid chromatography are severe and include (1) the generation of pressures up to 6000 psi, (2) pulse-free output, (3) flow rates ranging from 0.1 to 10 ml/min, (4) flow reproducibilities of 0.1% or better, and (5) resistance to corrosion by a variety of solvents. This instrument uses a quaternary pump with a proportioning multi-channel gradient valve. Changes in mobile phase composition are accomplished by proportional passage of solvent from each resevoir through this valve. After passage through the valve, located between the solvent resevoirs and the pump, the solvents are fed into a low-pressure mixing chamber. The mixed solvent, or mobile phase, then enters the high-pressure pump to be pumped through the column. Mobile phase changes can be accomplished easily and quickly which is especially useful for reequilibration following a gradient run. On startup, the pump should be primed according to instructions from your TA.
SAMPLE INJECTION SYSTEM
The ideal sample injection system places the sample at the top of the column in a sharp well- defined plug of minimum thickness. This instrument uses a programmable and automated sample injection system. Samples for analysis are placed in 2 ml glass vials. A crimper tool is used to seal a septum cap on each vial. The vials are placed at fixed locations in a 21-vial carousel. A nitrogen-pressurized pneumatic syringe system is used to aspirate a controlled volume of sample from the vial chosen for analysis. This sample is transferred to the sample loop of the injection valve. Rotation of the injection valve results in the sample loop being inserted in the flow path of the mobile phase just before it enters the head of the column as shown in Figure 3. In this instrument, the sample volume is determined by the volume aspirated by the needle up to a maximum volume of 100 µl.
COMPUTER
A desktop computer is interfaced to each LC module through an IEEE-488 interface. Instrument control, data acquisition and storage, and chromatographic peak integration are all initiated from the computer.
FIGURE 3: Sample Injection valve. The seven-port valve allows three pairs of ports to be switched synchronously. In the FILL position the mobile phase flows directly to the column through one pair of ports, while the other ports allow new sample to enter the sample loop. Rotation of the valve to the INJECT position directs the mobile phase flow into the sample loop, leading to injection of sample onto the column.
EXPERIMENTAL PROCEDURE - SOLVENT PREPARATION AND INSTRUMENT STARTUP
1. You may need to refill some of the solvent resevoirs before beginning. FILTER ALL SOLVENTS (water, methanol, acetonitrile) through 0.45 micron Nylon 66 membrane filters by vacuum filtration. Use the filtration apparatus provided, under the direction of your TA.
2. Using a CLEAN funnel, transfer the filtered solvents to the appropriate resevoir:
A: Water B: Methanol C: Acetonitrile Replace the top of each resevoir and twist slightly to form a gastight seal.
3. Helium is used for sparging the solvents (see above) and nitrogen is used for the pneumatics of the autosampler. Open both the helium and the nitrogen tank valves to deliver the PRESET flow of gases (helium: 2-4 bar (30-60 psi); nitrogen: 5 bar (75 psi)). DO NOT adjust the pressure or serious damage to the system may result.
4. Depress the ON/OFF button on the left-front of the resevoir tray to ON and then ASK YOUR TA to regulate the gas flow through the solvents. Sparge for 5-10 minutes with a vigorous stream of helium and then ASK YOUR TA to reduce the flow until only a very slow constant bubbling persists. The valves should be left in this position for the duration of the experiment. DO NOT ADJUST THE VALVES YOURSELF - THEY ARE VERY EASILY DAMAGED AND EXPENSIVE TO REPLACE.
5. Samples for analysis should be dissolved in acetonitrile, or one of the other mobile phases, and introduced into the 2 ml sample vials using a syringe and SYRINGE FILTER, or an adjustable pipet and PIPET FILTER. IT IS ESSENTIAL THAT ALL SAMPLES BE FILTERED TO REMOVE PARTICULATE MATTER. Seal each vial using the crimping tool, as demonstrated by your TA.
6. Place each vial in one of the numbered slots in the 21-slot carousel as directed in Table 1.
7. Depress the ON button for each of the autosampler, pump and detector modules and then switch on the computer using the switch on the adjacent surge protector. The HPLC application software will automatically be invoked and you will see a picture of a pair of chromatographic columns!
NOTE: The detector lamp should be switched on and allowed a warm-up period of 30
minutes before beginning the first analysis.
8. The autosampler, pump and detector will be initialized and the top-level menu will shortly appear on the monitor. At the bottom-right of the screen three boxes will report the status of the autosampler, pump and detector. These should be in READY status, unless you forgot to switch on any of the modules! The box on the upper right is used to report the status of analyses in progress and the box on the upper left is used to monitor the detector response (at a selected wavelength) as a function of retention time (i.e., to display the chromatogram in progress).
9. Prime the pump following directions from your TA.
10. ou are now ready to set up an analysis. This is done following the directions in Appendix 2.
EXPERIMENT 1.1 - PROCEDURE
CHROMATOGRAPHIC ANALYSIS - Investigation of chromatographic response as a function of detector wavelength
The purpose of this study is to teach you how to use the instrument and to give you some practice in setting up an analysis. You will investigate the utility of multiwavelength detection and consider the importance of response factors in quantitation of the components of a mixture.
1. You will be given a sample containing phenol, toluene, nitrobenzene, and p-nitrotoluene, together with uracil which is used as a void volume marker. This sample should be filtered and introduced into a 2ml sample vial as described previously. Load the vial into the autosampler carousel in position 12.
2. Following the directions in Appendix 2, Part I, edit a method to analyze this sample under the following conditions:
Flow Rate: 1.00 ml/min Stop Time: 20.00 min Mobile Phase Composition: 60% methanol/40% water Injection Volume: 10 microliter Detector Wavelength: 254, 280, AND 330 nm (simultaneously) Output: Send chromatogram and area report for each
wavelength to the printer.
This method should be stored in the file EXPT1.M
3. Following the directions in Appendix 2, Part II.a., create a single vial sequence which uses method EXPT1.M to analyze the vial in location 12 of the carousel. This sequence should be saved as EXPT1.S.
4. Following the directions in Appendix 2, Part III, run the analysis and print the chromatogram and area report for each of the three detector wavelengths.
EXPERIMENTS 1.2 AND 1.3 - PROCEDURE
1. The autosampler carousel should already be loaded with vials containing the analytes required for Experiments 1.2 and 1.3 as follows:
TABLE 1: LOCATION OF SAMPLES IN AUTOSAMPLER CAROUSEL
__________________________________________________________________
Position in
Sample Sample Tray* __________________________________________________________________
Experiment 1.2
Sample 1: uracil, benzyl alcohol, benzene 1
toluene, bromobenzene, p-xylene
Sample 2: uracil, phenol, nitrobenzene, nitrotoluene 2 Experiment 1.3 benzene 3 p-xylene 4 4-methoxyphenol 5 toluene 6 phenol 7 p-cresol 8 ethylbenzene 9 4-chlorophenol 10 acetonitrile ("blank vial" - used to allow equilibration period 11 between changes in mobile phase)
Experiment 1.1 uracil, phenol, toluene, nitrobenzene, nitrotoluene 12
Experiment 1.4 unknown mixture for method optimization 13 __________________________________________________________________
*Before proceeding, you MUST check that each vial is in the correct designated location.