Faculty of Engineering

Department of Chemical and Process Engineering

Name of Experiment: Construction and Operation of an

Extraction System
Abstract
The aim of this experiment is to firstly construct a fixed bed leaching system for the
extraction of solute from a bed of solids and then to test the system by extraction of dried sorrel
using hot water.

In achieving the aim of the experiment, pieces of pipe were measured, cut, threaded and
assembled on to an already made extraction system. Pre-heated water was then allowed to flow
through a bed of dried sorrel contained in a cylindrical column and the resulting solution was
collected into the water bath. A pump was used to continuously recirculate the solution through
the bed of dried sorrel. Samples of the solution were taken during equally spaced intervals during
the extraction process and were quantitatively analysed by UV-Vis spectroscopy.

Values of transmittance obtained were used to plot a Time-Transmittance graph and

conversion of transmittance to absorbance allowed an Absorbance-Time graph to be plotted.
From the respective graphs, it was deduced that concentration of the sorrel increased with time
and the time at which maximum extraction occurred is 20 minutes or 1200 seconds. The rate of
extraction was estimated to be 1.55 x 10-4 s-1.

The Time-Transmittance graph illustrated that the time decreased as transmittance

increased whereas the Absorbance-Time graph illustrated an increase in absorbance as time
increased. This suggests that time and transmittance are inversely related whilst absorbance and
time are directly related. This relationship between the quantities for both cases however, are not
uniform. This can be seen by the random shape of the graphs with increasing and decreasing
gradients. By Beer Lamberts law, since absorbance and concentration are directly related, a
measure of the change in absorbance of the samples is essentially a measure of the extraction
rate. Though the extraction rate varied at particular instances in the extraction process, an
estimate was still computed. The increase in concentration of sorrel in the solution can be
physically observed by the deepening of the red colour of the samples collected.

From the results obtained from the UV-Vis analysis, one can deduce that the extraction of
the sorrel that occurred was a transient process since the concentration of the sorrel changed
continuously with time. Furthermore, there were no signs of the system approaching a state of
equilibrium since even at the uttermost set of absorbance values obtained, an increase was
clearly visible.

Some errors which may have inevitably occurred during the experiment include heat
losses from the extraction system to the environment and also evaporation of the water thereby
inherently increasing the concentration of the sorrel solution. Potential errors were predicted and
specific precautions were taken to avoid it. These include thoroughly wiping the cuvette to
remove any contaminants and also ensuring that there were no air bubbles in the sample and
reference cuvettes. An important recommendation that can be made for persons performing this
experiment is to insulate the extraction system to reduce heat losses to the environment. This can
be done by simply filling the air spaces at the top of the cylinder with an insulating material such
as cotton or wool or even insulating the pipelines. More accurate results would be obtained and
evaluative statements made from the experiment would be more definitive.
Appendix
Introduction
Extraction, in the context of Chemical Engineering, refers to the removal or separation of
a component from a mixture of components by means of a solvent. Extraction is mostly practiced
in the food and pharmaceutical industries to flavour foods, produce oils from seeds and nuts
using organic solvents, purify heat sensitive materials and recover valuable substances from
reactions. There are two main types of extraction: leaching or solid extraction and liquid
extraction. The equipment used to carry out the extraction process are designed based on the
fundamental principles of the specific type of extraction.

In this experiment, extraction due to a fixed bed leaching system is examined. In this
system, a dried solid containing the material to be extracted interacts with a liquid solvent, such
that, there is dissolution of the solute into the solvent thereby being separated from the solid,
followed by diffusion of the solute in the solvent through the pores and cells of the solid. A
system in which the solvent is recirculated through the stationary solid bed, is referred to a semi-
batch system and is known to have a higher percentage recovery of solute as compared to the
corresponding batch system (Unit Operations of Chemical Engineering, Leaching and
Extraction, pg. 740). A system that is able to carry out the extraction as mentioned must consist
of a column containing the dried solid and a pump and pipelines to mobilize and the solvent in a
cyclic motion. Also included in the arrangement of the extraction system is a volumetric meter to
measure the flow of solvent through the system, ensuring a constant volumetric flow.

There are several factors that affect solid-liquid extraction. These include, particle size,
solvent, temperature and agitation. The smaller the particle size, the greater the specific area.
There would therefore be better contacting between solid-liquid phases and hence a higher
extraction rate. The solvent is selective and ideally removes only the substance to be extracted.
The viscosity of the solvent plays an important role in the extraction process since solvent is
allowed to trickle down the column and a solvent with low viscosity would be move through the
compacted solid particles with greater ease than a more viscous solvent. Temperature is directly
related to rate of diffusion of the solute into the solvent, as well as, the solubility of the solute in
the selective solvent. Additionally, there is an inverse relationship between temperature and
viscosity. A higher temperature would suggest a lower viscosity of the solvent and concentrated
product solution thereby having an enhanced fluid circulation as compared to room temperature
conditions. Leaching is a mass transfer operation and agitation of the mixture would certainly
increase the rate of mass transfer. Thus, these factors must be taken into consideration when
designing and constructing an extraction system. (Introduction to Food Process Engineering
(Second Edition), pg. 454)

The solid extraction or leaching in the manner previously described where solvent is
passed over a dried solid can be seen as an unsteady state process since the concentration of the
solution produced changes continuously with time. More specifically, the solution becomes more
concentrated with solute as time progresses. Ultraviolet-Visible (UV-Vis) Spectroscopy is a
common method used to quantitatively analyse the solute extracted. When a sample of the
solution is subjected to light of wavelength 200nm to 800nm, that is, the UV-Vis region of the
electromagnetic spectrum, provided that it contains non-bonding and / or pi electrons, light is
absorbed as electrons are promoted to the higher energy, anti-bonding orbitals. The structural
features responsible for the absorption of light in UV-Vis region are called chromophores. The
light emerging on the opposite side of the sample can be measured and the ratio of the intensities
of the incident and emergent light gives a measure of transmittance and concentration of
chromophores or extracted solute particles. Typically, a sample of the solution produced from
the extraction process that has a low transmittance suggests a high concentration of solute
extracted and therefore efficient extraction. It must be noted that the quantity of light transmitted
can indicate the quantity of light absorbed. Absorbance and transmittance are related by the

equation A=log 10 ( T1 ) and where transmittance is given as a percentage, Absorbance can be

calculated from the equation A=2−log10 (%T ).

The relationship between absorbance and concentration is summarised in Beer Lambert’s

equation, A=∈c l. This equation suggests a direct relationship between absorbance and
concentration and if the molar absorptivity and path length are kept constant, a sample with
higher concentration would absorb a greater quantity of light. Additionally, where the
concentrated product solution is coloured, that is, where it absorbs light in the UV region, a rich
or dark colour indicates a high concentration of solute.
Apparatus
 5cm glass column  Water bath
 Pump  Rotameter
 Pipework and fittings  Gate valve
 Dried Sorrel  Spectrophotometer
 Hacksaw  Tape measure
 Thread Seal  Flat file 3/8” Stock and Die
 Couplings  Pipe Wrenches
 Pipe vice  Stock and die

Rotameter
Cylinder
containing
dried sorrel

Gate valve

Pump
Water bath

Sorrel extraction apparatus used

 The Thermo Scientific Helios™ Zeta spectrometer used

Spectrometer details:
Model – Helios Zeta

Serial number – HEDN095025

CAT – 9423UVZ1002E
Procedure
1) A piece of pipe of length 28 inches was measured and cut using a measuring tape and
hacksaw respectively.
2) The edges of the pipe was filed to make even using a flat file.
3) The pipe was threaded using 3/8 inches Stock and Die.
4) Thread seal was applied to the pipe in the direction of the threads.
5) Elbows were fitted on to the ends of the pipe using a wrench and a pipe vice.
6) A pipe was connected to the top of the extraction system and was clamped and the other
was connected at the bottom. The pipe on the top connected the cylinder to the piping
above the rotameter whilst the pipe at the bottom connected the water bath to the pump.
7) The water bath was turned on.
8) The pump was turned on, the flow rate was adjusted to the midpoint of the rotameter and
water was allowed to recycle through sorrel in 2: G.V.F. column. The sorrel was then
fully immersed.
9) A sample was taken every 2 minutes for 20 minutes.
10) Transmission measurements were taken using the Spectrophotometer at 460nm.
Treatment of Results
Table 1: Percentage transmission obtained for each sample at 460nm
Sample Time (seconds) Transmittance (%)
1 120 46.13
2 240 46.24
3 360 45.38
4 480 35.93
5 600 33.63
6 720 33.90
7 840 32.18
8 960 29.32
9 1080 29.05
10 1200 27.43

Table 2: Absorbance calculated for each sample at 460nm

Sample Time (seconds) Absorbance
1 120 0.336
2 240 0.335
3 360 0.343
4 480 0.445
5 600 0.473
6 720 0.470
7 840 0.492
8 960 0.533
9 1080 0.537
10 1200 0.562
 Time-Transmittance Graph
1400

1200

1000

800
Time (s)

600

400

200

0
0.25 0.3 0.35 0.4 0.45 0.5

Transmittance

Absorbance-Time Graph
0.6

0.55

0.5
Absorbance

0.45

0.4

0.35

0.3
0 200 400 600 800 1000 1200 1400

Time (s)
2) From Beer Lambert’s law, Absorbance, A = ɛ c l

If the path length and molar absorptivity are assumed to be constant, then, A ∝ c

Absorbance can be used as a measure of concentration of sorrel present in the respective

samples. Maximum extraction is said to occur when the concentration of sorrel is the highest.
This is certainly at the end of the experiment where absorbance is the highest, and time at which
maximum extraction occurs is seen in the last plotted value, that is, at 20 minutes or 1200
seconds.

3) The rate of extraction can be measured by the rate of change of absorbance with time. This is
obtained from the gradient of the Absorbance-Time graph.

Assuming a linear response of absorbance with time, the equation of the Absorbance-Time graph
is y = mx + c (equation of a straight line)

where m is the gradient or slope of the graph and c is the y-intercept.

The gradient of the graph and hence the rate of extraction was found to be 1.55 x 10-4 s-1.
Calculation
Conversion of % transmittance to absorbance

Derivation of formula

I
%T = ∗100
I0

I %T
=
I 0 100

lg ( II )=lg ( %T
0 100 )

lg ( II )=lg ( 100
0
%T )

lg ( II )=lg 100−lg ⁡(%T )

0

lg ( II )=2−lg ⁡(%T )
0

However, Absorbance = lg ( II )
0

∴ Absorbance=2−lg ⁡(%T )

Sample calculation for Absorbance using % Transmittance obtained for sample 1

Absorbance=2−lg ( 46.13 )

Absorbance=0.336
Calculation of rate of extraction

Obtaining gradient of the Absorbance-Time graph plotted using points (1400, 0.54) and
(240, 0.36).

∆ Absorbance
Slope , m=
∆ Time

0.54−0.36
m=
(1400−240)s

m=1.55∗10− 4 s−1

¿ , m=0.0093 min−1

Hence, rate of extraction was found to be 1.55 x 10-4 s-1.

Discussion
Leaching, as performed by the constructed fixed bed leaching system, is a semi-batch
process and is inherently transient in nature. Due to the fact that the extraction of sorrel from the
dried solid state is a mass transfer operation, it occurs continuously and the concentration of
sorrel changes over a period of time. A sample of the solution collected at a particular instant in
time is therefore expected to have a lower concentration of sorrel particles as opposed to one that
is collected at a later instant in the extraction process. Since transmission is inversely
proportional to the number of chromophores present (the absorbing species) and hence the
amount of sorrel present per unit volume of solution, the more concentrated sorrel solution
collected at a later instant in time would exhibit a lower transmittance than the sample collected
at the earlier time. It is expected, therefore, that a graph of Time against Transmission is such
that time decreases with Transmittance, that is, an inverse relationship between the two
quantities and negative gradient. Conversely, a plot of Absorbance against Time would be such
that absorbance increases as time increases, that is, a direct relationship and positive gradient.

The general expectations of the Time-Transmittance and Absorbance-Time graphs were

met. There is however, an unexpected deviation in values of transmittance and hence absorption
obtained for samples 1and 2, and also, 5 and 6. It shows an increase in transmittance while time
increases. This goes against the theory presented and may have occurred due to unexpected
adsorption of solute by the solid bed during the time interval of 120s to 240s and 600s to 720s,
malfunctioning of the spectrometer and / or due to errors which may have occurred while
performing the experiment as mentioned below.

With respect to the gradient of the graphs, the Time-Transmittance graph possesses a
negative gradient whilst the Absorbance-Time graph possesses a positive gradient. However, it
must be stated that the gradients vary randomly as time increases. Since the rate of extraction is
indicated by the gradient of the graphs, this suggests that the rate of extraction of the fixed bed
leaching system used was not constant and shows no clear trend on how it changes as the
extraction progresses.

Theoretically, for a fixed bed leaching system as used in this experiment, it is expected
that the Time-Transmittance and Absorbance-Time graphs would experience a general increase
and decrease in gradients respectively and perhaps show convergence to a particular
transmittance and absorbance value as time becomes large. This was not observed from the
graphs plotted due to limitations of the experiment itself, such as, the time period over which
extraction was investigated was too small (20 minutes) to observe the general trend of extraction
rate and also, the volume of water used was probably too large to investigate a small change in
concentration of solute in the solvent. Furthermore, the fact that neither of the graphs showed
convergence to a particular value as time became large suggests that the extraction process did
not approach equilibrium and concentration of sorrel in the solution was still considerably less
that that contained in the bed of dried sorrel. Nevertheless, the rate of extraction was estimated
by finding the gradient of the best fit straight line through the plotted points. The rate of
extraction was found to be 1.55 x 10-4 s-1.

The wavelength of light used in this experiment is 460nm. Its significance lies in the fact
that 460nm exists in the range of wavelength of the UV-Vis spectrum and more specifically, it
exists in the range of wavelengths which sorrel absorbs UV-Vis light thereby allowing the
samples to be analysed by UV-Vis spectroscopy. The wavelength at which maximum absorbance
occur is preferentially used when analysing samples by UV-Vis spectroscopy.

Two observations made during the extraction process are the colour of the collected
samples eventually darkened and the temperature of the galvanised steel pipe increased as time
progressed. Since sorrel gave the samples the characteristic red colour, an increase in
concentration of sorrel caused the colour of the samples to deepen. Additionally, the fact that a
red colour was observed suggests that the wavelength of light absorbed corresponds to its
complementary colour, which is, blue-green. This is consistent with the theory since the range of
blue-green light is 440nm to 570nm (according to chemistry.msu.edu) and the wavelength of
light used in the analysis was 460nm. The increase in temperature of the pipe, on the other hand,
was simply due to heat transfer of the hot water pumped from the water bath to the walls of the
pipe. Since the pipe was initially at thermal equilibrium with the surrounding temperature when
no water flowed through it, an increase in temperature was observed as hot water began to flow
and heat was transferred from the water to the pipe by conduction. The purpose of using water at
an elevated temperature as the solvent was simply to increase the rate at which extraction of the
sorrel would have occurred thereby making the change in concentration of sorrel extract in the
water more measureable and hence facilitate a better quantitative analysis.

Sources of Error:

1) Heat losses to the surroundings could have affected the rate at which extraction occurs
thereby adversely affecting results obtained.
2) The inevitable loss of water due to evaporation could have contributed to the increase in
concentration of the collected samples.
3) Distilled water was used in the reference cuvette whereas the sorrel was dissolved in tap
water (evident from the calcium carbonate at the start-up of the extraction process). Tap
water may contain traces of minerals which might have absorbed UV-Vis light during the
analysis and a comparison of the sample to reading (I) to that obtained from the distilled
water (I0) may not be an accurate measure of concentration.

Precautions:

1) The connecting pipe at the bottom of the extraction system was elevated by a wooden
block instead of directly placing into the water bath so that the liquid in the bath would be
sucked up by the pump instead of air thereby preventing presence of air bubbles in the
system and cavitation and destruction of the pump.
2) The cuvettes were thoroughly wiped on the smooth surface through which light passes.
This is to ensure that no soils or prints made on the surface of the cuvette from handling
would not affect transmission of light through the sample.
3) The sorrel samples obtained from the column were placed in containers that were labelled
to avoid confusion and incorrect placement of desired sample into the spectrophotometer.
4) It was ensured that there were no air bubbles in the cuvette.
5) The circular shape of the galvanised pipe was preserved throughout the threading process
so that the elbows can fit snuggly at its ends and hence prevent undesired loss of extract.
Conclusion
A fixed bed leaching system was successfully constructed and tested by examining the extraction
of sorrel from a bed of dry sorrel using water as the solvent. After analysis of the samples
obtained at different instances in the extraction process, it was found that the concentration of
sorrel increased as the time increased. A graph of time against transmittance yields one that
shows a somewhat inverse relation between transmittance and time. An absorbance time graph
showed the opposite. As time increased, so too did absorbance. For both graphs however, the
change in gradient of the graph varied randomly and therefore it can be concluded that the rate of
extraction of the constructed leaching system cannot accurately be predicted. Nevertheless, by
approximation, an estimate rate was found to be 1.55 x 10-4 s-1. During the extraction process, 2
interesting observations were made. Firstly, the red colour of the solution darkened as time
progressed and secondly, the pipe carrying the hot water became noticeably warmer as compared
to its initial temperature.
References
1) MacCabe, W. L., & Harriot, P. (2001). Chapter 23 (Leaching and Extraction).Unit
operations of chemical engineering (6. ed., pp. 739-740). New York [u.a.: McGraw-Hill.

2) Smith, P. G. (2011). Introduction to food process engineering (2nd ed.). New York:

Springer.

3) Rousseau, R. W. (1987). Chapter 10 (Leaching-Organic Materials). Handbook of

separation process technology (p. 540). New York: J. Wiley.

4) Michigan State University, Department of Chemistry. (n.d.). UV-Visible Spectroscopy.

Retrieved April 14, 2014, from
http://www2.chemistry.msu.edu/faculty/reusch/virttxtjml/spectrpy/uv-vis/spectrum.htm