Introduction to Chromatography

Theory

Bioanalytical Chemistry
Spring 2004

The Theory of Chromatography

Plate theory - older; developed by Martin &
Synge
Rate theory - currently in use today

Plate Theory - Martin & Synge

1954 Nobel Laureates
View column as divided into a number (N)
of adjacent imaginary segments called
theoretical plates
within each theoretical plate complete
equilibration of analytes between stationary
and mobile phase occurs

Plate Theory - Martin & Synge

1954 Nobel Laureates
Significance?
Greater separation occurs with:
greater number of theoretical plates (N)
as plate height (H or HETP) becomes smaller

L = N H or H = L / N
where L is length of column, N is number
of plates, and H is height of plates

N can be Estimated
Experimentally from a
Chromatogram
N = 5.55 tr2 / w1/22 = 16 tr2 / w2
where:
tr is retention time;
w1/2 is full width at 1/2 maximum height
w is width measured at baseline

Choice of Column Dimensions

Nmax = 0.4 * L/dp
where:
N - maximum column efficiency
L - column length
dp - particle size
So, the smaller the particle size the higher the
efficiency!

Efficiency Relative to Analysis

Time

today
90 mm L
3 um

today
150 mm L
5 um

1970s
300 mm L
10 um

10

100

Analysis Time, min

First Important Prediction of

Plate Theory

Bandspreading - the width of bands

increases as their retention time
(volume) increases

Problem:
A band exhibiting a width of 4 mL and a
retention volume of 49 mL is eluted from a
column. What width is expected for a band
with a retention volume of 127 mL eluting
from the same analyte mixture on the same
column?
ANS: 10.4 mL

Second significant prediction of

plate theory

The smaller HETP, the narrower the

eluted peak

Plate Theory - Practical

Considerations
Not unusual for a chromatography column
to have millions of theoretical plates
Columns often behave as if they have
different numbers of plates for different
solutes present in same mixture

Rate Theory
Based on a random walk mechanism for the
migration of molecules through a column
takes into account:
band broadening
effect of rate of elution on band shape
availability of different paths for different
solute molecules to follow
diffusion of solute along length

Van Deemter Equation

H = A 1/3 + B/ + C
where:
H is HETP (remember want a minimum!)
is mobile phase velocity
A, B, and C are constants

Van Deemter Equation

H = A 1/3 + B/ + C
first term - rate of mobile phase movement
through column (often just a constant)
second term - longitudinal solute diffusion;
solute concentration always lower at edges of
column so solute diffuses longitudinally
third term - equilibration is not instantaneous

Resolution
Ideal chromatogram exhibits a distinct
separate peak for each solute
reality: chromatographic peaks often
overlap
we call the degree of separation of two
peaks:
resolution = peak separation
average peak width

Resolution
Resolution = tr / wavg
lets take a closer look at the significance of
the problem:

Resolution
So, separation of mixtures depends on:
width of solute peaks (want narrow)
efficiency
spacing between peaks (want large spacing)
selectivity

Example
What is the resolution of two Gaussian
peaks of identical width (3.27 s) and height
eluting at 67.3 s and 74.9 s, respectively?
ANS: Resolution = 2.32