1

Chapter 5 Problems

1. Show that the units are correct for Eq. 5.1.15 with  in M-1 cm-1, A in m2, and V in liters in
Eq. 5.1.6.

Answer: Substituting Eq. 5.1.6 into Eq. 5.1.15 gives:

d[B] 2.303 Jo l B A 
= 2.303 JoB l [A] =    [A]
dt  NA h V
with Jo in J m s ,  in M cm , l in cm, A in m , and V in L. The concentrations are in mol L-1.
-2 -1 -1 -1 2

Substituting in just the units:

-2 -1 -1 -1 2
J m s (mol L cm )(cm)m 
mol L s = 
-1 -1
 L  [mol L-1]
 mol-1 (J s)(s-1)  
No other unit conversions are necessary.

2. Show that the photochemical rate constant for an optically thin solution is independent of path
length for a cell with a uniform cross section. For example, a cell with uniform cross section
includes cylindrical cells and rectangular cells where the volume is given by the area of the
solution exposed on the face of the cell, a, multiplied by the path length, V = al. Determine any
unit conversion factors in the final result.

Answer: The photochemical rate constant for the formation of a secondary photoproduct from
Eqs. 5.1.6 and 5.1.19 is:
2.303 Jo l B A 
jB =    with l in cm, A in m2, and V in L.
 NA h V
The total volume of the solution is given by V = al, where l is the reaction cell path length.

Lens
A l
Substituting V = al into Eqs. 5.1.6 and 5.1.19 gives:
2.303 Jo l BA 2.303 Jo  BA
jB =    =   
 NA h al  NA h  a 
which is independent of path length. In other words, as the path length increases, the volume of
the solution also increases, keeping the increase in moles per unit volume per unit time constant.
 2

However, now we need to be careful about units. With J o in J m-2 s-1,  in M-1 cm-1, and jB in s-1
the conversions needed are:
2.303 Jo  B
jB =   (100 cm/1 m)(1 m3/1000 L)
 NA h 
In other words, putting in just the units:
-2 -1 -1 -1
J m s (mol L cm )
s = 
-1
 (100 cm/1 m)(1 m3/1000 L)
 mol-1 (J s)(s-1) 

3. A chemical actinometer is a solution with known quantum yield that can be used to find the
incident intensity in photochemical experiments. The ferrioxalate actinometer uses the reaction:
h
2 Fe3+ + (C2O4)2-  2 Fe2+ + 2 CO2
A ferrioxalate concentration of 0.15 M is normally used for actinometry, which is optically thick.
A common light source for photochemical reactions is the 366 nm emission line of a mercury
lamp. The quantum yield for the ferrioxalate reaction at 366 nm is 1.18. 1 The progress of the
reaction is monitored using the visible absorption of the ortho-phenanthroline complex of Fe2+ at
522 nm. The molar absorption coefficient of the Fe 2+ complex at 522 nm is 8650. M-1 cm-1. The
ortho-phenanthroline complex for Fe3+ is very weak and transparent at 522 nm. The following
experiment was used to determine the incident intensity for a photoreactor. A solution of 0.15 M
ferrioxalate was irradiated for 10.0 min. A 1.00 mL aliquot was withdrawn and diluted with
water to a total volume of 100.0 mL in a volumetric flask. The absorbance of this solution in a
1.00 cm pathlength cuvette at 522 nm was 0.410. Calculate the incident flux in mol L -1 s-1.

Answer: The plan is to use the Beer-Lambert Law to calculate the concentration of Fe 2+ in the
aliquot. Then the dilution factor is used to calculate the Fe 2+ concentration in the actinometer
solution. Since the solution is optically thick, we assume a zeroth-order reaction and, using Eq.
5.1.14, we calculate the incident intensity.
Using the Beer-Lambert law: A = lc the concentration of the aliquot is:
c = A/l = 0.410/(8650. M-1 cm-1 1.00 cm) = 4.74x10-5M
The dilution factor for the absorbance sample is 1/100 giving the concentration of Fe 2+ in the
original actinometer as [Fe2+] = 4.74x10-3 M. Assuming zeroth-order kinetics, Eq. 5.1.14:
d[Fe2+]
= B Ja = B Jo
dt
which integrates to [Fe2+] = [Fe2+]o + B Jo t, where the initial concentration is zero, [Fe2+]o = 0.
Using the known photochemical quantum yield, B = 1.18, and the Fe2+ concentration at 10.0
min gives the incident intensity as:
[Fe2+] = B Jo t
 3

4.74x10-3 M
or Jo = [Fe2+]/(B t) = = 6.70x10-6 mol L-1 s-1
1.18 (10.0 min)(60 s/1 min)

4. A high power mercury lamp produces 219.0 W m -2 at 366 nm at the surface of a

photochemical reaction cell (see Problem 3). Assume the cross-section of the incident beam is
1.00 cm2 and the solution volume is 10.0 mL. Calculate the incident flux in mol L-1 s-1.

Answer: (a) The energy of 366 nm light in kJ mol-1 is:

E = NA h = NA hc/ = 6.022x1023 mol-1 (6.626x10-34 J s)(2.998x108 m s-1)/366x10-9m =
= 3.268 x105 J mol-1 = 326.8 kJ mol-1
Using Eq. 5.1.6 gives the flux in mol L-1 s-1:
Jo  A  219.0 J m-2 s-1 1 cm2 (1 m/100 cm)2
Jo =  =   = 6.70x10-6 mol L-1 s-1
NA hV 3.268 x105 J mol-1 0.0100 L 
(which, by the way, corresponds to the answer from Problem 3).

5. When p-nitroanisole and pyridine are photolyzed in aqueous solution the reaction is:

NO2
h +
N
+ + NO2-
H2 O
N
O
CH3
O
CH3

The quantum yield for a solution containing 1.00x10-5M p-nitroanisole and 0.0100 M pyridine
in 1% acetonitrile is 4.65x10-3 at 366 nm.2 The molar absorption coefficient at 366 nm of p-
nitroanisole is 1990 M-1 cm-1. Calculate the photochemical rate constant and write the rate law
for an optically thin solution assuming the incident flux is 6.70x10 -6 mol L-1 s-1 for a 10.00 cm
path length reaction cell.

Answer: For an optically thin solution, using Eq. 5.1.19:

jB = 2.303 JoB l
= 2.303 (6.70x10-6 mol L-1 s-1)( 4.65x10-3)( 1990 M-1 cm-1)(10.0 cm)
= 1.43x10-3 s-1
The rate constants for environmental problems is often given in days -1:
jB = 1.43x10-3 s-1 (60 s/1 min)(60 min/1 hr)(24 hr/1 day) = 123 day-1
Using Eq. 5.1.17 gives the rate law:
 4

d[B]
= jB [A] = 1.43x10-3 s-1 [PNA]
dt
where [PNA] is the concentration of p-nitroanisole. This reaction is commonly used as an
actinometer for environmental studies.

6. A solution with a known photochemical quantum yield can be used to calculate the incident
light flux during a photolytic reaction. Such solutions are called chemical actinometers (see
Problems 3 and 5). Consider the reaction of p-nitroacetophenone and pyridine:
NO2

h +
+ N NO2-
H2O
+
N
C
O CH3
C
O CH3

A chemical actinometer and a solution of p-nitroacetophenone and pyridine were simultaneously

photolyzed at 366 nm in reaction cells with identical geometry. The path length of the reaction
cell is 1.00 cm. The quantum yield for the actinometer is 4.65x10 -3 with a molar absorption
coefficient 1990 M-1 cm-1. The photochemical rate constant for the actinometer is determined to
be 1.43x10-3 s-1. The molar absorption coefficient of p-nitroacetophenone is 160. M-1 cm-1.2 The
photochemical rate constant for 1.00x10 -5 M p-nitroacetophenone and 0.100 M pyridine is
4.18x10-5 s-1. Calculate the quantum yield for the p-nitroacetophenone and pyridine reaction at
366 nm.

Answer: For an optically thin solution, using Eq. 5.1.19:

jB = 2.303 JoB l
The incident flux calculated from the actinometer is:
jact 1.43x10-3 s-1
Jo = =
2.303 act l 2.303 (4.65x10-3)(1990 M-1 cm-1 1 cm)
= 6.71x10-5 mol L-1 s-1
Then applying Eq. 5.1.19 for the p-nitroacetophenone, PNAP, reaction gives:
jPNAP 4.18x10-5 s-1
PNAP = =
2.303 Jo l 2.303 (6.71x10-5 mol L-1 s-1)(160. M-1 cm-1 1 cm)
= 1.69x10-3
This reaction is also commonly used as a chemical actinometer. This actinometer is appropriate
with slower environmental reactions than are covered by the PNA actinometer in Problem 5.
 5

7. Consider the following reversible first-step mechanism for a first-order photochemical

reaction:
jA* kR
A + h 
 A* A*  B
kD
where kD is the combined rate constant for all the non-photochemical deactivation processes,
with kD = kf + kISC + knr. Derive Eq. 5.1.24 directly from this mechanism. [Hint: Express the rate
law in the form of Eq. 5.1.17 and then use Eq. 5.1.19]

Answer: The plan is to use the steady-state approximation to find the overall rate law for the
formation of products in the form of Eq. 5.1.17 and then find the photochemical quantum yield
using Eq. 5.1.19. The reactive intermediate is the molecular excited state, A*. Step 1 is to write
the rate law for the formation of product:
d[B]
= kR [A*]
dt
Step 2 for the steady-state mechanistic scheme is to find the rate law for the formation of the
reactive intermediate:
d[A*]
= jA* [A] – kD [A*] – kR [A*] = 0
dt
Step 3: Solving for the concentration of the reactive intermediate, A*:
jA*
[A*] = [A]
kD + k R
With only one unknown we can proceed to Step 5 and find the overall rate law by substituting
the steady state A* concentration back into the rate law for the formation of product:
d[B] jA*
= kR [A]
dt kD + k R
Comparing this last equation with Eq. 5.1.17 and 5.1.19, that is j B = B jA*, gives:
jA*
jB = kR = B jA*
kD + k R
which is an important equation in its own right. This equation shows the relationship between j B,
which is the rate constant for the overall production of the product while k R is the rate constant
for the formation of product from the molecular excited state. So solving for the photochemical
quantum yield and using the definition of k D = kf + kISC + knr gives:
kR
B =
kf + kISC + knr + kR

8. Anthracene fluorescence is quenched by halogenated compounds like CCl 4. A Stern-Volmer

quenching study was completed giving the fluorescence intensities, as a function of CCl 4
 6

concentration, in the following table.3 The intensities are in arbitrary units. The fluorescence
lifetime in the absence of CCl4 is 5.03 ns. Calculate the quenching rate constant.
[CCl4] (M) 0 0.02 0.04 0.08 0.12
Intensity 2437 1860 1490 1110 893

Answer: Using Eq. 5.1.31 in the absence of quencher, kR = kq[Q] = 0 gives:

kf + kISC + knr = 1/ = 1/5.03x10-9 s = 1.99x108 s-1
Using Eq. 5.1.28, in terms of the intensity ratio:
kq[Q]
Io/I = o/f = 1 +
kf + kISC + knr
a plot of Io/I versus [CCl4] is constructed. The slope is determined by a least squares fit using the
spreadsheet and linest():

[CCl] Intensity Io/I

2.800
0 2437 1.000
y = 14.385x + 1.026
0.02 1860 1.310 2.600
0.04 1490 1.636
R2 = 0.9986
2.400
0.08 1110 2.195
0.12 893 2.729 2.200
2.000
slope 14.3850 1.0260 intercept
I o /I

1.800
± 0.3139 0.0212 ±
r2 0.9986 0.0302 s(y) 1.600
F 2099.9195 3.0000 df 1.400
ssregression 1.9203 0.0027 ssresidual
1.200
1.000
0 0.02 0.04 0.06 0.08 0.1 0.12
[CCl4] (M)

From the slope the quenching rate constant is:

kq = slope (kf + kISC + knr) = 14.38 (1.99x108 s-1) = 2.86x109  0.06x109 s-1

9. Run a numerical simulation for the Chapman mechanism for the rate constants and
concentrations appropriate at an altitude of 25 km. A table of appropriate constants is given
below.4,5S Determine the steady-state concentration of ozone using Eq. 5.2.11 and by numerical
simulation.

Altitude j1 k2 j3 k4 [M] [O2]

km s-1 cm6 molecule-2s-1 s-1 cm3 molecule-1s-1 molecule cm-3 molecule cm-3
25 3.0x10-12 1.2x10-33 5.5x10-4 6.9x10-16 9x1017 1.8x1017
40 5.7x10-10 9.1x10-34 1.9x10-3 2.2x10-15 8.1x1016 1.7x1016
 7

Answer: We proceed by editing the MatLab files listed in Addendum 5.7. A longer maximum
time is necessary to achieve a steady state. The changes necessary to the files listed in the
Addendum define the rate constants:
% Constants for 25 km:
j1=3.0e-12;
k2=1.2e-33;
j3=5.5e-4;
k4=6.9e-16;
M=9.0e17;

and the main file is changed to give:

% Set the initial values
Xo = [0 1.8e17 0];
% Set the total integration time in seconds
maxTime = 4.0e7;

The steady-state concentration of ozone that results is 1.645x10 13 molecules cm-3. A short Excel
spread sheet was written to solve the quadratic equation for the steady-state ozone concentration
from Eq. 5.2.11:
j1 3.00E-12 s-1
k2 1.20E-33 cm6 molecule-2 s-1
j3 5.50E-04 s-1
k4 6.90E-16 cm3 molecue-1 s-1
[M] 9.00E+17 molecule cm-3
[O2] 1.80E+17 molecule cm-4

a 7.59E-19
b 7.45E-10
c -2.10E+08
O3 (+root) 1.66E+13 molecule cm-6
(-root) -1.66E+13 molecule cm-7

The listed a, b, and c cells are the normal coefficients for ax 2 + bx + c = 0. The analytical value,
1.66x1013 molecules cm-3, is differs slightly from the limiting numerical value because of
numerical error. Stiff methods are usually not as accurate as the much slower 4 th order Runge-
Kutta algorithm, ode45().

10. Nitrogen oxides catalyze the destruction of ozone and must be taken into account in accurate
stratospheric models. The reactions and the rate constants appropriate for 25 km are:
k5
NO + O3  NO2 + O2 k5 = 3.4x10-15 cm3 molecules-1 s-1
k6
NO2 + O  NO + O2 k6 = 1.1x10-11 cm3 molecules-1 s-1
j7
NO2 + h  NO + O j7 = 7x10-3 s-1
The values for the rate constants j1-k4 and [M] at 25 km are given in Problem 9. Guesses for the
starting concentrations for NO and NO2 that you can use are:
 8

[NO]o = 8.0x108 molecules cm3 and [NO2] = 1.0x109 molecules cm3

Add these three reactions to the numerical simulation outlined in Addendum 5.7 to find the
change of the ozone concentration with and without catalysis. You can use MatLab or MathCad,
or any numerical routines that employ stiff methods.

Answer: The rate laws are:

d[O]
= 2 j1 [O2] – k2 [O][O2][M] + j3 [O3] – k4 [O][O3] – k6 [NO2][O] + j7 [NO2] 1
dt
d[O2]
= – j1 [O2] – k2 [O][O2][M] + j3 [O3] + 2 k4 [O][O3] + k5 [NO][O3] + k6 [NO2][O] 2
dt
d[O3]
= k2 [O][O2][M] – j3 [O3] – k4 [O][O3] – k5 [NO][O3] 3
dt
d[NO]
= – k5 [NO][O3] + k6 [NO2][O] + j7 [NO2] 4
dt
d[NO2]
= k5 [NO][O3] – k6 [NO2][O] – j7 [NO2] 5
dt
The rate of formation of odd oxygen species is obtained by adding Eqs. 1 and 3 to give:
d([O3]+[O])
= 2 j1 [O2] – 2 k4 [O][O3] – k5 [NO][O3] – k6 [NO2][O] + j7 [NO2] 6
dt
The concentrations are set to X(4) = [NO] and X(5) = [NO 2]. The revised “chapman.m” method
file for MatLab is called “chapmanNOx.m”:
function dX = chapmanNOx(t,X);
% Differintial equations for the Chapman Mechanism.
% Based on work by Farhan Akhtar, School of Earth and Atmospheric Sciences
% Georgia Institute of Technology:
% [Link]
% Constants for 25 km:
j1=3.0e-12;
k2=1.2e-33;
j3=5.5e-4;
k4=6.9e-16;
k5=3.4e-15;
k6=1.1e-11;
j7=7.0e-3;
M=9.0e17;
%Calculations
dX = zeros(5,1) ;
dX(1) = 2*j1*X(2)-k2*X(1)*X(2)*M+j3*X(3)-k4*X(1)*X(3)-k6*X(5)*X(1)+j7*X(5) ;
dX(2) = -j1*X(2)-k2*X(1)*X(2)*M+j3*X(3)+2*k4*X(1)*X(3)+k5*X(5)*X(3)+k6*X(5)*X(1) ;
dX(3) = k2*X(1)*X(2)*M-j3*X(3)-k4*X(1)*X(3)-k5*X(4)*X(3) ;
dX(4) = -k5*X(4)*X(3)+k6*X(5)*X(1)+j7*X(5) ;
dX(5) = -dX(4) ;

And the revised main program is:

% The differential equations are defined in the -m file: 'chapmanNOx.m'
%
clear
% Set the initial values
 9

Xo = [0 1.8e17 0 8e8 1e9];

% Set the total integration time in seconds
maxTime = 4.0e7;
trange = [0 maxTime];

% solve the differential equations

[T,X] = ode15s(@chapmanNOx,trange,Xo);

These calculations give a steady-state ozone concentration of 8.7 4x1012 molecules cm3, which is
a factor of 0.53 smaller than the case without NO and NO 2. However, the exact change in ozone
concentration is very dependent of the values of the rate constants and the total NO and NO 2
concentration. Hopefully, you noticed that the graph of [O2] shows a slow, constant increase for
long times. This increase is from numerical errors that are caused by the use of the less accurate
stiff differential equations solver. However, using the more accurate ode45(), a 4 th order Runge-
Kutta solver, requires many hours of computation time.

11. Problem 10 lists the three reactions that supplement the Chapman mechanism to account for
the catalytic destruction of ozone caused by NO and NO2. (a) At steady state, show that the rate
law for odd oxygen species can be expressed as:
d([O3]+[O])  k6 [NO2]
= 2 j1 [O2] – 2 k4 [O3][O]1 +
dt  k4 [O3] 
The term in parentheses is called the enhancement factor, :
 k6 [NO2]
 = 1 +
 k4 [O3] 
which determines the extent of the catalysis of the destruction of ozone by NO and NO [Link] the
rate constants given in Problems 9 and 10 along with the following rough estimates for the
steady-state concentrations to estimate the enhancement factor at 25 km. These concentrations
are from the results of Problem 10 at 25 km:
[O3] = 8.74x1012 molecules cm-3 [NO2] = 1.45x109 molecules cm-3

Answer: The rate laws are:

d[O]
= 2 j1 [O2] – k2 [O][O2][M] + j3 [O3] – k4 [O][O3] – k6 [NO2][O] + j7 [NO2] 1
dt
d[O2]
= – j1 [O2] – k2 [O][O2][M] + j3 [O3] + 2 k4 [O][O3] + k5 [NO][O3] + k6 [NO2][O] 2
dt
d[O3]
= k2 [O][O2][M] – j3 [O3] – k4 [O][O3] – k5 [NO][O3] 3
dt
d[NO]
= – k5 [NO][O3] + k6 [NO2][O] + j7 [NO2] 4
dt
d[NO2]
= k5 [NO][O3] – k6 [NO2][O] – j7 [NO2] 5
dt
The rate of formation of odd oxygen species is obtained by adding Eqs. 1 and 3 to give:
 10

d([O3]+[O])
= 2 j1 [O2] – 2 k4 [O][O3] – k5 [NO][O3] – k6 [NO2][O] + j7 [NO2] 6
dt
Eq. 6 involves both the NO and O atom concentrations. One of these intermediate concentrations
can be eliminated using the steady-state approximation. Assume that the reactions are at steady
state. The rate law for reaction 5 can then be set equal to zero. Adding Eqs. 5 and 6 eliminates
[NO] as a variable and gives an additional destruction term:
d([O3]+[O])
= 2 j1 [O2] – 2 k4 [O3] [O] – 2 k6 [NO2][O] 7
dt
production destruction destruction

which can be rearranged to give:

d([O3]+[O])  k6 [NO2]
= 2 j1 [O2] – 2 k4 [O3] [O]1 +
k4 [O3] 
8
dt 
production destruction

 k6 [NO2]
 = 1 +
k4 [O3] 
with 9

Substituting in the values of the constants from Problems 9 and 10 and the steady-state
concentrations of O3 and NO2 gives:
 k6 [NO2]  1.1x10-11 cm3 molecules-1 s-1 (1.45x109 molecules cm-3)
 = 1 +
k4 [O3]   6.9x10-16 cm3 molecule-1s-1 (8.74x1012 molecules cm-3)
= 1+

= 3.6
The ozone destruction processes are accelerated by a factor of 3.6 by NO and NO 2 catalysis. This
value is very sensitive to the values of the rate constants, especially the photolytic rate constant
j7, as shown in the next problem.

12. Nitrogen oxides catalyze the destruction of ozone and must be taken into account in accurate
stratospheric models. The reactions and the rate constants appropriate for 25 km are:
k5
NO + O3  NO2 + O2 k5 = 3.4x10-15 cm3 molecules-1 s-1
k6
NO2 + O  NO + O2 k6 = 1.1x10-11 cm3 molecules-1 s-1
j7
NO2 + h  NO + O j7 = 7x10-3 s-1
Show that the ratio of the NO2 and NO concentrations at steady-state are given by the
relationship:
[NO2] k5 [O3]
=
[NO] k6 [O] + j7
Calculate the steady-state ratio assuming that the O atom concentration is small enough that
k6[O] << j7. Assume [O3] = 8.74x1012 molecules cm-3 (as in Problem 11).
 11

Answer: The rate laws are:

d[NO]
= – k5 [NO][O3] + k6 [NO2][O] + j7 [NO2] = 0 1
dt
d[NO2]
= k5 [NO][O3] – k6 [NO2][O] – j7 [NO2] = 0 2
dt
At steady-state both are equal to zero. Either Eq. 1 or 2 can be solved for the NO 2 to NO ratio:
k5 [NO][O3] = k6 [NO2][O] + j7 [NO2] 3
[NO2] k5 [O3]
= 4
[NO] k6 [O] + j7
This relationship can also be used to find the enhancement factor in Problem 11 in terms of the
NO and O atom concentrations. Assuming k6[O] << j7 and substitution of the constants gives:
[NO2] k5 [O3] 3.4x10-15 cm3 molecules-1 s-1 (8.74x1012 molecules cm-3)
= =
[NO] j7 7x10-3 s-1
= 4.2
This relationship shows that the ratio of NO2 and NO and the O atom concentrations are sensitive
to the rate constant for the photolytic decomposition of NO2. Photolytic rate constants are
dependent on time of day, time of year, and altitude and so are highly variable. The range of
photolytically active wave lengths for the decomposition of NO2 is 300-400 nm. Direct O3
photolysis occurs in the ranges of 200-300 and 450-650 nm.
O3+ h(200-300, 450-650 nm)  O2 + O
NO2 + h(300-400 nm)  NO + O
The UV-A range is 320–400 nm and the UV-B range is 280–320 nm.

13. The combustion of carbon sources, such as coal or charcoal, in limited amounts of oxygen
produces carbon monoxide. Carbon monoxide is a commonly used reducing agent, especially in
metallurgy. The reaction of carbon with high temperature steam produces carbon monoxide and
hydrogen gas. Carbon monoxide and hydrogen are the feed stocks for industrial processes like
the Fischer-Tropsch process, which can be used to produce transportation fuels from coal or
biomass. Carbon monoxide readily adsorbs onto charcoal surfaces. The equilibrium surface
loading of CO on charcoal at 0°C is given in the following table. Determine the Langmuir
coefficient for this system.

Pco (bar) 0.0973 0.240 0.412 0.720 1.176

 (mmol g-1) 0.113 0.248 0.378 0.573 0.787

Answer: Based on Eq. 5.3.11 or 5.3.18, using the general form “abx/(1+bx)” in the “Nonlinear
Least Squares Curve Fitting” applet gives the following results and corresponding curve-fit plot.
 12

======= Results ============ 0.0009

a= 0.001823 +- 0.000061 0.0008
b= 0.643 +- 0.033 0.0007
---------------------------------------------

gamma (mol g -1 )
0.0006
sum of squared residuals= 8.473e-11
0.0005
stand. dev. y values= 0.000005315
correlation between a & b= -0.989 0.0004
0.0003
0.0002
0.0001
0
0 0.5 1 1.5
Pco (bar)

The fit gives b = 0.643± 0.033 bar-1. A double reciprocal plot also works well, but again the
nonlinear fit parameters are better because the uncertainties are handled better. Inverting Eq.
5.3.18:
b PA 1 1 1
 A = max gives = +
1 + b PA A max b PA max

Pco  1/Pco 1/ 10000

(bar) (mol g-1) (bar-1) (mol-1 g) 9000 y = 801.44x + 646.38
0.0973 0.000113 10.27671 8859.907 8000 R2 = 0.9998
0.240 0.000248 4.167778 4027.23 7000
1/gamma (mol -1 g)

0.412 0.000378 2.427832 2647.382 6000

0.720 0.000573 1.389259 1744.076 5000
1.176 0.000787 0.850567 1270.541 4000
3000
The linear fit gives b = 0.81 bar-1, which is 2000
outside the error bounds from the non-linear 1000
fit. 0
0 5 10
1/Pco (bar -1 )

However, the double reciprocal plot does help verify Langmuir behavior. Notice the very high
correlation coefficient between fit parameters in the non-linear fit, -0.989. The values of the
maximum binding capacity and the Langmuir coefficient are highly correlated. To resolve this
issue, to generate results that can be published, experimental points at higher P CO are necessary.

14. Antibody-antigen interactions are very strong and very specific. The interaction between a
protein, bovine serum albumin, and anti-BSA immunoglobulin G (IgG ) was determined using
SPR. BSA was attached to a gold surface and the IgG was flowed over the surface at constant
concentration. The results for the observed association rate constant are given in the table below.
The dissociation rate constants, from nonlinear curve fitting from the time courses, were
averaged over each run and found to be 5.94x10-5 s-1. Find ka, the association equilibrium
 13

constant, KA, and the dissociation equilibrium constant, KD. Convert KD to picomolar units, pM,
which is typical of the conventional choice of units in the medicinal chemistry literature.
[IgG] (nM) 10.0 4.00 1.60 0.640
kobs (s-1) 0.00623 0.002914 0.001578 0.00057

Answer: From Eq. 5.4.10, kobs = ka [A]o + kd, a plot of kobs versus the concentration of IgG
flowing over the surface, [A]o, should yield a straight line. The linest() fit is for k obs vs. [IgG] in
units of nM, so the final slope must be multiplied by 1x109 to convert to M-1 s-1.

[IgG] nM [IgG] (M) kobs (s-1)

0.007
10 1.00E-08 0.00623
4 4.00E-09 0.002914 0.006 y = 5.845E+05x + 4.499E-04
1.6 1.60E-09 0.001578 R 2 = 9.933E-01
0.005
0.64 6.40E-10 0.00057

kobs (s -1)
0.004
In nM concentration units: 0.003
slope 0.00058452 0.00045 intercept
± 3.3891E-05 0.000185 ± 0.002
r2 0.9933 0.000247 s(y) 0.001
F 297.467 2 df
ssregression 1.812E-05 1.22E-07 ssresidual 0
0.00E+00 5.00E-09 1.00E-08
[IgG] (M)

The resulting ka = 5.85x105  0.34x105 M-1 s-1, giving KA = 5.85x105 M-1 s-1/ 5.94x10-5 s-1 =
9.84x109 M-1 and KD = 1/KA= 1.02x10-10 M. In picomolar units:
KD = 1.02x10-10 M (1 pM/1.0x10-12M) = 102. pM = 1. x102 pM

15. SPR is a commonly used technique in immunology. The interaction between a protein,
porcine serum albumin, PSA, and anti-PSA immunoglobulin G (IgG ) was determined using
SPR. The anti-PSA IgG was attached to a gold surface and PSA was flowed over the surface at
constant concentration. The results for the observed association rate constant are given in the
table below.6 The dissociation rate constants, from nonlinear curve fitting from the time courses,
were averaged over each run and found to be 1.02x10-3 s-1. Find ka, the association equilibrium
constant, KA, and the dissociation equilibrium constant, KD. Convert KD to nanomolar units, nM,
which is typical of the conventional choice of units in the medicinal chemistry literature.
[PSA] (nM) 7.18 21.5 66.4 201. 601.
kobs (s-1) 0.0122 0.00189 0.00189 0.0297 0.0641

Answer: From Eq. 5.4.10, kobs = ka [A]o + kd, a plot of kobs versus the concentration of PSA
flowing over the surface, [A]o, should yield a straight line. The linest() fit is for k obs vs. [PSA] in
units of nM, so the final slope must be multiplied by 1x109 to convert to M-1 s-1.
 14

[PSA] nM [PSA] (M) kobs (s-1) 0.07

7.18E+00 7.18E-09 0.0122
y = 8.321E+04x + 1.383E-02
2.15E+01 2.15E-08 0.0189 0.06
R2 = 9.903E-01
6.64E+01 6.64E-08 0.0189
0.05
2.01E+02 2.01E-07 0.0297

kobs (s -1 )
0.04
In nM concentration units: 0.03
slope 8.3207E-05 0.013825 intercept
± 4.7595E-06 0.001358 ± 0.02
r2 0.99028 0.00235969 s(y) 0.01
F 305.627 3 df
ssregression 0.0017018 1.6704E-05 ssresidual 0
0.00E+00 2.00E-07 4.00E-07 6.00E-07
[PSA ] (M)

The resulting ka = 8.32x104  0.48x104 M-1 s-1, giving KA = 8.32x104 M-1 s-1/ 1.02x10-3 s-1 =
8.16x107 M-1 and KD = 1/KA= 1.23x10-8 M. In nanomolar units:
KD = 1.23x10-8 M (1 nM/1.0x10-9M) = 12.3. nM = 1. x101 nM

16. It is not necessary in dynamic SPR measurements to wait for the surface adsorption to reach
equilibrium. However, if the time course for the association does essentially reach equilibrium,
the equilibrium values can be fit to the Langmuir adsorption isotherm. Such equilibrium SPR
experiments provide an alternative method to determine the equilibrium dissociation constant
that is complementary to dynamic measurements. Comparison between equilibrium and dynamic
results helps to determine experimental uncertainties. The limiting refractive index values from
the time course measurements for the system in Problem 15 are given below. 6 By fitting the
results to a Langmuir adsorption isotherm, determine the equilibrium dissociation constant in
nanomolar units. The units typical for SPR instrument output are micro-refractive index units, or
RIU.
[PSA] (nM) 0 7.18 7.18 19.7 59.2 181.3 538.5
R (RIU) 0 23.6 26.5 38 55.6 57.6 58.8

Answer: Based on Eq. 5.3.11 or 5.3.18, using the general form “abx/(1+bx)” in the “Nonlinear
Least Squares Curve Fitting” applet gives the following results and corresponding curve-fit plot,
below. The non-linear fit value for K A = 9.52x107  1.0x107 M-1. Then KD = 1/KA = 1.05x10-8 M
= 10.5 nM. The value from Problem 16 was 12 nM, which agrees within experimental
uncertainty.
 15

======== Results =========

70
a= 61.36 +- 1.55
b= 95200000 +- 10000000 60
--------------------------------------------- 50
sum of squared residuals= 22.66
40

mRIU
stand. dev. y values= 2.129
correlation between a & b= -0.6582 30

20

10

0
0.0E+00 2.0E-07 4.0E-07 6.0E-07
[PSA ] (M)

A double reciprocal plot also works well, but again the nonlinear fit parameters are better
because the uncertainties are handled better:

[PSA] (M) R (RIU) 1/[PSA] (M-1) 1/R 4.5E-02

0 0 y = 1.710E-10x + 1.639E-02
7.18E-09 23.6 1.39E+08 4.24E-02 4.0E-02 R2 = 9.773E-01
7.18E-09 26.5 1.39E+08 3.77E-02
1.97E-08 38.0 5.06E+07 2.63E-02 3.5E-02
5.92E-08 55.6 1.69E+07 1.80E-02
1/R

1.81E-07 57.6 5.52E+06 1.74E-02 3.0E-02

5.38E-07 58.8 1.86E+06 1.70E-02
2.5E-02

2.0E-02

1.5E-02
0.0E+00 5.0E+07 1.0E+08 1.5E+08
1/[PSA ] (M-1)

17. The rate of decomposition of NH3 was determined as a function of the initial partial pressure
of H2 and is inhibited by product formation:7
NH3 (g)  ½ N2(g) + 3/2 H2 (g)
The initial pressure of NH3 in each run was 100 mm Hg, and varying amounts of H2 gas were
added to the reaction vessel at the beginning of the reaction. The catalyst was platinum and the
reaction was run at 1138C. The results are given below. Show that the dependence on the
product, H2, partial pressure is described by Eq. 5.5.17.
PNH3 in 120 s PH2 initially added
33 50
27 75
16 100
10 150
 16

Answer: First we approximate the initial rate as o = PNH3/120 s. Then a plot of the initial rate
as a function of 1/PH2 should give a straight line in accordance with Eq. 5.5.17. A spreadsheet
was set up and the plot was constructed:

PNH3  PH2,o o (mm Hg s-1) 1/PH2,o 0.025

33 50 0.275 0.020
27 75 0.225 0.013 y = 0.0634x + 0.0011
16 100 0.133 0.010 0.020 R2 = 0.9327
10 150 0.083 0.007
0.015

rate o
0.010

0.005

0.000
0.000 0.100 0.200 0.300
1/PH2,o

The inverse dependence on PH2 is verified by the linear plot, to within experimental error.

18. Hydrogen is a clean burning substance that is being suggested as a transportation fuel.
However, hydrogen is costly to produce. One proposal is to use solar thermal energy to provide
the energy necessary to convert water into hydrogen gas. The Sulfur-Iodine cycle consists of
three coupled reactions, which add to give the dissociation of water:
H2SO4 (l)  SO2 (g)+ H2O (g) + ½ O2 (g) (>850°C)
I2 + SO2 + 2 H2O  2 HI + H2SO4 (>120°C)
2 HI  I2 + H2 (>450°C)
net: H2O  H2 (g) + ½ O2 (g)
The first step is the most unfavorable thermodynamically and kinetically. The reaction occurs in
two steps:
H2SO4 (l)  SO3 (g) + H2O (g)
SO3 (g)  SO2 (g) + ½ O2 (g)
The decomposition of SO3 has a negligible rate at 800°C without a catalyst. Mixed chromium-
iron oxide catalysts have been proposed for the gas phase decomposition of SO 3.8 The
heterogeneous decomposition of SO3 on Fe1.6Cr0.4O3 in a constant flow reactor has been studied
as a function of temperature. The percent yields at several temperatures are given in the table,
below. The residence time of the reactant in a constant flow reactor is constant with temperature,
so the percent yield is directly proportional to the reaction rate. Verify Arrhenius behavior and
determine the activation energy.
T (°C) 550 600 650 750 800
SO2 yield % 1.2 5.5 10.7 52.7 79.3
 17

Answer: The ln(rate) and ln(yield) can be used interchangeably for this constant flow reactor
data. A plot of ln(yield) versus 1/T will verify Arrhenius behavior with the slope = –E a/R. A
spreadsheet was set up, the plot was constructed, and the slope determined using linest():

T (C) SO2 yield T (K) 1/T (K-1) ln(yield) 5.0

550 1.2 823.2 0.001215 0.182
4.5
600 5.5 873.2 0.001145 1.705
4.0
650 10.7 923.2 0.001083 2.370
3.5 y = -14533x + 18.076
750 62.7 1023.2 0.000977 4.138
3.0 R2 = 0.9857
800 79.3 1073.2 0.000932 4.373

ln(yield)
2.5
2.0
1.5
slope -14532.895 18.076 intercept
1.0
 1011.114 1.087 
r2 0.986 0.236 [Link]. Y 0.5
F 206.588 3.000 df 0.0
ssreg 11.506 0.167 ssresid 0.0009 0.001 0.0011 0.0012 0.0013
1/T (K-1)

Given the experimental uncertainty, Arrhenius behavior may be justified. The lowest
temperature point, which is the point with the largest absolute and relative uncertainty, deviates
significantly from the line. Several additional data points at low temperatures would be necessary
to judge if systematic curvature exists for the plot. Given Arrhenius behavior, the activation
energy is:
Ea = –slope R = 148  23 kJ mol-1
The nonlinear fit using the “Nonlinear Least Squares Curve Fitting” applet, however, provides a
very important lesson:
=========== Results ============= 90
a= 4130000 +- 4750000 80
b= -11630 +- 1200 70
----------------------------------------------------- 60
sum of squared residuals= 43.02 50
yield

stand. dev. y values= 3.787 40

correlation between a & b= -0.9994 30
20
10
0
800 900 1000 1100
T (K)

The activation energy from the non-linear fit is 97  10 kJ mol-1, which is outside the range of
the experimental uncertainties. Why the large difference? The correlation coefficient between the
fit values is very large at -0.9994, showing a very strong correlation between the fit values of A
and EA. You wouldn’t have noticed this problem without the non-linear fit and you would have
underestimated the uncertainty in the activation energy. In fact, using the linearized form of the
data in the “Nonlinear Least Squares Curve Fitting” applet using the simple line option, “ax + b”,
also gives a very large correlation coefficient between the slope and the intercept of -0.9953. The
 18

results from this experiment are best given as 1.x102 kJ mol-1 to avoid over-representing the
uncertainty.

19. Derive the rate law for a bimolecular heterogeneous reaction with stoichiometry: A + C P.
Assume that reactant C is strongly adsorbed to the catalytic surface and A is weakly adsorbed.

Answer: The reaction mechanism can be approximated by:

kAa k

A (g) + B  A + C  P+ B 1
kAd
where the surface concentration of C is determined by the rapid pre-equilibrium:
kCa

C (g) + B  C 2
kCd
The rate law for the formation of products is constructed similarly to Eq. 5.5.2:
1 d
= k [A] [C] 3
 dt
with the pre-equilibrium concentration of free binding sites dominated by the fractional coverage
for C, in analogy with Eq. 5.5.15 :
[C]  [B]o 4
[B]o
and [B]  5
bC PC
with bC = kCa/kCd which is the equilibrium constant for the surface adsorption of C, and P C the
partial pressure for C. The rate law for the formation of A bound on the surface from Eq. 1 is:
d[A]
= kAa [B] PA – kAd [A] – k[A][C] = 0 (steady-state) 6
dt
where [B] is given by Eq. 5. Solving for [A] using the steady-state approximation and
substituting Eq. 5 gives:
kAa [B]o
[A] = P 7
kAd – k[C] bC PC A
Substitution of Eq. 7 for the surface concentration for A into Eq. 3 and using Eq. 4 for the
surface concentration of C gives:
1 d kAa [B]o
= k P [B]o 8
 dt kAd – k[B]o bC PC A
For the rate law with respect to the disappearance of A, using Eq. 5.5.5:
 19

dPA kAa /V [B]o

– = k(RT) PA [B]o 9
dt kAd – k[B]o bC PC
Gathering together all of the constants defines the observed rate constant:
kAa /V [B]o
kobs = k(RT) [B]o 10
kAd – k[B]o bC
and then the final rate law is:
dPA PA
– = kobs 11
dt PC
In some cases the rate constant for the surface reaction, k, is assumed to be very small so that
the observed rate constant can be written:
bA
kobs = k(RT) /V [B]o2 12
bC
where bA = kAa/kAd, which is the equilibrium constant for the surface adsorption of A.

20. Determine the integrated rate law for surface catalysis from an adsorbed monolayer using
Eq. 5.5.7.

Answer: Starting with Eq. 5.5.7 and separating variables:

1 + b A PA  1 
dPA =  +1 dP = – k dt
bA P A  bA P A  A
The integration limits start at t= 0, PA = PAo:
 A  1 +1 dP = –  t k dt
P
  bA P A  A 0
PAo
The sum in the first integral can be split into two terms:
1 PA PA
ln(PA|PAo + PA|PAo = – kt
bA
1 PA
ln + (PA – PAo) = – kt
bA PAo

If A is weakly adsorbed, then bA is small. Notice that if bA is small, then the (PA – PAo) is
negligible compared to the logarithmic term and this last equation reduces to a simple first-order
reaction, Eq. 5.5.8:

1 PA PA
ln  – kt and ln  – bA kt
bA PAo PAo
 20

On the other hand, if bA is large, the logarithmic term is negligible and the reaction is zeroth
order, as expected from Eq. 5.5.9:

(PA – PAo)  – kt

21. We assumed a pre-equilibrium mechanism to determine the rate for a heterogeneously

catalyzed reaction according to Eq. 5.5.1:
ka k
A (g) + B 
 A  P (g) + B
kd
The rate law for the reaction, in terms of the products is then given by:
d[P]
= k(RT) (/V) [A]
dt
To give a better approximation, use the steady-state approximation to determine the rate law.
Then show that the more exact rate law reduces to Eq. 5.5.6 using a suitable approximation.

Answer: Eq. 5.5.1 is equivalent to the Michaelis-Menten mechanism, so the derivation for this
rate law should be parallel to our previous derivation. The plan is to use the steady-state
approximation on [A] and then use the mass balance for the total concentration of surface
adsorption sites, [B]o = [B] + [A], Eq. 5.3.1.
The rate law for the appearance of surface bound A for this mechanism is identical to Eq.
5.5.11:
d[A]
= kAa [B] PA – kAd [A] – k[A] 1
dt
In steady state, for [A] as the reactive intermediate:
ka [B] PA
ka [B] PA – kd [A] – k[A]  0 or [A] = 2
kd + k 
Solving the mass balance equation, Eq. 5.3.1, for the concentration of free surface sites gives:
[B] = [B]o – [A] 3

and substitution into Eq. 2 gives:

(kd+k)[A] = kaPA([B]o – [A]) or (kd+k+kaPA)[A] = kaPA[B]o 4
Solving for [A] gives:
ka PA [B]o
[A] = 5
kd + k  + k a P A
 21

Dividing both numerator and denominator by kd and using the definition of the Langmuir
coefficient, bA = ka/kd gives:
bA PA [B]o
[A] = 6
1 + k/kd + bA PA
Substitution of Eq. 6 into the rate law for appearance of product gives the final result:
d[P] bA PA [B]o
= k(RT) (/V) [A] = k(RT) (/V) 7
dt 1 + k/kd + bA PA
If we assume that kd >> k then the k/kd term is negligible and this last equation reduces to the
pre-equilibrium mechanism result, Eq. 5.5.6.

Literature Cited
1. A. Leifer, The Kinetics of Environmental Aquatic Photochemistry, American Chemical
Society, Washington, DC, 1988. pp. 149-50.
2. A. Leifer, The Kinetics of Environmental Aquatic Photochemistry, American Chemical
Society, Washington, DC, 1988. pp. 106-10.
3. W. Legenza, C. J. Morzzacco, “The Rate Constant for Fluorescence Quenching, An
undergraduate experiment using the Spectronic 20,” J. Chem. Ed., 1977, 54, 183-184.
4. F. Akhtar, “The Chapman Reactions: A model of the Production and Destruction of
Stratospheric Ozone,” School of Earth and Atmospheric Sciences, Georgia Institute of
Technology:
[Link]
5. E. Harvey, R. Sweeney, “Modeling Stratospheric Ozone Kinetics, Part I The Chapman
Cycle,” J. Chem. Ed., 1999, 76, 1309.
6. T. Ryan, Reichert Analytical Instruments, Depew, NY, private communication.
7. C. N. Hinshelwood, R. E. Burk, “,” J. Chem. Soc., Transactions, 1925, 127, 1105-17.
8. A. M. Banerjee, M. R. Pai, K. Bhattacharya, A. K. Tripathi, V. S. Kamble, S. R. Bharadwaj,
S. K. Kulshreshtha, “Catalytic decomposition of sulfuric acid on mixed Cr/Fe oxide samples
and its application in sulfur–iodine cycle for hydrogen production,” International Journal of
Hydrogen Energy, 2008, 33(1), 319-326.