Cre 2 Lab Manual

V.G.E.C Chandkeda | CRE II Lab Manual

Experiment – 1 (Annular Reactor) Date: / /20 Aim:

To determine various curves like E, F, E0, and F0 for the flow of Water through annular reactor and to predict conversion for a First order reaction of a known rate constant.

Apparatus:

Annular reactor, Tracer injection device, Stop watch, Sample Bottles, Titration set.

Chemicals:

10 ml 2N Potassium Permanganate (KMnO4) solution, 50 ml 2N Sodium Hydroxide (NAOH) solution, 1000 ml 0.1N Hydrochloride acid (HCL) solution, Phenolphthalein (indicator).

Fluid used:

Water

Theory: If we know precisely, what is happening within the vessel, i.e., if we have a complete velocity distribution map for the fluid, then we are able to predict the behavior of a vessel as a reactor. Though fine in principle, the attendant complexities make it impossible to use this approach. Residence Time Distribution studies predict extent of non-ideal flow in real reactor. Information on RTD obtained from tracer measurements can be directly used to predict the performance of the reactor in which non-ideal flow occurs. RTD is the distribution of the time taken by various elements of the taking different routes to pass through the reactor. Distribution Curves: E, the Age distribution of fluid leaving a vessel, it is the distribution needed to account for non-ideality in flow. It is represented by equation: ∞

∫ Edt 0

Also, Edt – Fraction of exit stream between t and (t+dt). C-curve: Normalized response of an impulse or a data function is called C curve. For Normalizing the measured concentration ‘C’ is divided by ‘Q’ which is the area under curve: ∞

i.e. Q=

∫ Cdt 0

And normalizing concentration,

P a g e 1 | 42


Ci Q

C=

The plot of, Ci vs. ti gives so called ‘C’ – curve. The area under this curve is: ∞ ∞ ∫ Cdt = ∫ CQi = 1, 0 0 ∞

As,

Q=

∫ Cdt 0

= 1, hence C=E

F-curve: The response curve of the tracer in exit stream from the reactor, for a step input, is called F C curve, This response is measured as C 0 . Relation between E and F is given by, ∞

F=

∫ Edt 0

Above distribution curve, based on dimensionless time units, are denoted by subscript Ɵ i.e. t E ɵ , F ɵ and C ɵ curves, where Ɵ = t . Conversion directly from tracer information: Linear process: A variety of pattern can give the same tracer output curve. For liner processes, however, these all results in the same conversion; consequently we may use any convenient flow pattern to determine conversions, as long as the pattern selected gives the same tracer response curve as the real reactor. The simplest pattern to use assumes that each element of fluid passes through the vessel with no intermixing with adjacent elements, the age distribution of material in the exit stream telling how long each of these individual elements remains within the reactor. Thus for reactant A in the exit stream ∞

CA =

∫ C Aelement Edt

i=0

These conversion equation can be solved either graphically for any pattern of the most important measure is the location of the distribution. This is called the meant value or the centroid of the distribution. This for a C versus t curve the mean is given by

P a g e 2 | 42

V.G.E.C Chandkeda | CRE II Lab Manual ∞

∫ tCdt 0 ∞

t=

∫ Cdt 0

If the distribution curve is known at a number of discrete time values ti, then

∑ tiCi ∆ ti ∑ Ci ∆ ti

t=

The next most important descriptive quantity is the spread of the distribution. This is commonly measured by the variance σ 2, defined as ∞

∞

∫ ( t−t ) 2Cdt σ

2

=

0

∞

∫ Cdt

∫ t 2Cdt =

0

– (t)2

∞

∫ Cdt

0

0

Again in discrete from σ

2

=

∑ ( ti−t ) 2 Ci ∆ ti ∑ Ci ∆ ti

=

∑ ti2 Ci ∆ ti ∑ Ci ∆ ti

- (t)2

The variance represents the square of the spread of the distribution and has units of (time) 2. It is particularly useful for matching experimental curves to one of a fairly of theoretical curves. When used with normalized distribution for closed vessel these expressions simplify somewhat. Thus for a continuous curve or for discrete measurements at equal time intervals mean becomes∞

∫ t Edt = ∑ Ei 0 ∑

t=

tiEi

=

∑ tiEi ∆ t

=

∫ t 2 Edt

And the variation becomes∞

σ

σ

2

2

=

=

∞

∫ ( t−t ) 2 Cdt 0

∑ ti2 Ei ∑ Ei

=

0

∑ ti 2 Ei ∆ t

- (t)2

- (t)2

Procedure: 1. Set up the apparatus as shown in the diagram. P a g e 3 | 42


2. Find out the Mean Residence Time for the reactor at a particular fixed flow rate, using the KMnO4 solution. Mean Residence Time is the time required for complete color removal after injection of the KMnO4 Solution. 3. Now keeping the flow rate constant inject 2ml 2N NaOH solution and take the samples of the outlet stream from the reactor at a definite interval of time. 4. Titrate the samples against 0.5N HCL Solution and examine their concentration. 5. Carry out the necessary calculations and plot E, F, E ɵ , and F ɵ curves. 6. Find out conversion for first order irreversible reaction. Results: For the first order irreversible reaction, Conversion =

%

Conclusion:

P a g e 4 | 42


P a g e 5 | 42


Experiment – 2 (Sine Tube Reactor) Aim

Apparatus: set.

Date: / /20 : To determine various curves E, F, E θ, and Fθ for the flow of water through sine tube reactor and to predict conversion for a first order reaction of a known rate constant. Sine Tube reactor, Tracer injection device, Stop watch, Sample bottles, titration

Chemical : 10ml 2N Potassium Permanganate (KMnO4) solution, 50ml 2N Sodium hydroxide (NaOH) solution, 1000ml 0.1N Hydroxide acid (HCL) solution, Phenolphthalein (indicator). Fluid used: Water Theory: If we know precisely what is happening within the vessel, i.e. if we have a complete velocity distribution map for the fluid then we are able to predict the behavior of a vessel as a reactor. Through fine in principle, the attendant complexities make it impossible to use this approach. Residence time distribution studies predict extent of non-ideal flow in real reactor information on RTD obtained from tracer measurement can be directly used to predict in which non-ideal flow occurs. RTD is the distribution of the time taken by various elements of taking different routes to pass through the reactor. Distribution Curves: E, the Age distribution of fluid leaving a vessel; It is the distribution needed to account for nonideality in flow. It is represented by equation: t

∫ Edt 0

Also, Edt-fraction of exit stream between t and (t+dt) C curve: Normalized response of an impulse or a data function is called C curve. For normalizing the measured concentration ‘C’ is divided by ‘Q’ which is the area under curve: ∞

Cdt i.e. Q=∫ 0 And normalizing concentration, C=

Ci Q

The plot of, ci vs. tt gives so called ‘C’- curve. The area under this curve is:

P a g e 6 | 42


∞

C

∫ Cdt =∫ Qi dt =1, 0

0

As, Q = Cdt =1, hence C=E F curve: The response curve of the tracer in exit stream from the reactor, for a step input, is by, t

F=∫ Edt 0

Above distribution curve, based on dimensionless time units, are denoted by subscript θ i.e. Eθ, Fθ and Cθ curves where

θ=

t ´t

.

Conversion directly from tracer information: Linear process: A variety of flow pattern can give the, same tracer output curve. For linear processes however, these all result in the same conversion; consequently, we may use any convenient flow pattern to determine conversions, as long as the pattern selected gives the same tracer response curve as the real reactor. The simplest pattern to use assumes that each element of fluid passes through the vessel with no intermixing with adjacent, the age distribution of material in the exit stream telling how long each of the individual elements remains within the reactor. Thus for reactant A in the exit stream. ∞

´ A = ∫ CA element Edt C t=0

These conversion equations can be solved either graphically or numerically for any pattern of flow, thus the performance of non-ideal flow reactor can be determined precisely given the residence time distribution and rate constant for the first order reactions. The mean and variable: It is frequently desirable characterize a distribution by a few numerical values. For this purpose, the most important measure is the location of the distribution. This is called the mean value or the centroid of the distribution, Thus for a C versus t curve the mean is given by

P a g e 7 | 42


∫ tCdt 0 ∞

´t ≡

∫ Cdt 0

If the distribution curve is known at a number of discrete time values /, than t C ∆t ´t ≅ ∑ i i i ∑ Ci ∆t i The next most important descriptive quantity is the spread of the distribution. This is 2 commonly measured by the variance σ , defined as-

∞

∞

∫ ( t- ´t ) Cdt ∫ t2 Cdt 2

σ2 =

0

=

∞

∫ Cdt

0

∞

- ( ´t )

2

∫ Cdt

0

0

Again in discrete form σ

2

2 ∑ ( t i - ´t ) Ci ∆t i ∑ t 2i Ci ∆t i ´ 2 ≅ = -( t) ∑ Ci ∆t i ∑ Ci ∆t i

The variance represents the square of the distribution and has unit of (time) 2. It is particularly useful for matching experimental curves to one of a fairly of theoretical curves. When used with normalized distribution for closed vessels these expressions simplify somewhat. Thus for a continuous curve or for discrete measurement at equal time intervals mean becomes ´t =∫ tEdt ≅ ∑ ∞

t i Ei

∑ Ei

0

=∑ t i E i ∆t

And the variance becomes ∞

∞

σ =∫ (t−´t ) Edt =∫ t 2 Edt −( ´t )2 2

2

0

0

2 t i Ei ∑ σ= − (´t )2=∑ t 2i E i ∆ t−(´t )2 ∑ Ei 2

P a g e 8 | 42


Procedure: 1. Set up apparatus as shown in diagram. 2. Find out the mean residence Time for the reactor at a particular fixed flow rate, using the KMnO4 solution. Mean residence Time required for complete color removal after injection for KMnO4 solution. 3. Now keeping the flow rate constant inject 2ml 2NaOH solution and take the sample of the outlet stream from the reactor at definite interval of time, 4. Titrate the samples against 0.5N HCL solution and examine their concentration Eθ F 5. Carry out necessary calculations and plot E, F, and θ curves. 6. Find out conversion for first order irreversible reaction.

Result: For the first order irreversible reaction, conversion = Conclusion:

P a g e 9 | 42


P a g e 10 | 42


Experiment – 3 (Spiral Coil Reactor) Aim:

Date: / /20 To determine various curves like E, F, E ɵ , and Fɵ for the flow of water through spiral coil reactor and to predict conversion for a first order reaction of a Known rate constant.

Apparatus: Spiral coil reactor, tracer injection device stop watch sample bottles titration set. Chemicals: 10 ml, 2N Potassium permanganate solution, 50 ml, 2N sodium hydroxide solution, 1000 ml, 0.1 N hydrochloric acid solution, Phenolphthalein (indicator) Fluid used: Water Theory: If we know precisely, what is happening within the vessel, i.e. if we have a complete velocity distribution map for the fluid, then we are able to predict the behaviour of a vessel as a reactor. Though fine in principle, the attendant complexities make it impossible to use this approach. Residence time distribution studies predict extent of non-ideal flow in real reactor. Information on RTD obtained from tracer measurement can be directly used to predict the performance of reactor in which ideal flow occurs. RTD is the distribution of the time taken by various element of the taking different routes to pass through the reactor. Distribution Curves: E, the age distribution of fluid leaving a vessel; It is the distribution needed to account for nonideality in flow, It is represented by equation: ∞

∫ Edt 0

Also, Edt fraction of exit streams between t and (t + dt) C curve: Normalized response of an impulse or a data function is called C curve. For normalizing the measured concentration ‘C’ is divided by ‘Q’ which is the area under the curve: ∞

i.e. Q =

∫ Cdt 0

And normalizing concentration,

P a g e 11 | 42


C=

Ci Q

The plot of ci vs. ti gives so called ‘C’ curve. The area under this curve is ∞

∞

∫ Cdt

=

0

∫ Ci Q 0

dt = 1

∞

As Q =

∫ Cdt 0

= 1, hence C = E

F curve The response curve of the tracer in exit stream from the reactor, for a step input is called F curve this response is measured as C/C0. Relation between E and F is given by, t

∫ Edt 0

Above distribution curve, based on dimensionless time units, are denoted by subscript ɵ i.e. t−¿ t Eɵ, Fɵ and Cɵ curves, where ɵ = ¿ Conversion directly from tracer information: Linear process A variety of flow pattern can give the same tracer output curve. For linear processes, however these all result in the same conversion ; consequently we may use any convenient flow pattern to determine conversion , as long as the pattern selected gives the same tracer response curve as the real reactor. The simplest pattern to use assumes that each element of fluid passes through the vessel with no intermixing with adjacent elements, the age distribution of material in exit stream telling how long each of these individual element remains within the reactor .Thus for reactant A in the exit stream ∞

CA =

∫ C A element Edt

t=0

These conversion equation can be solved either graphically or numerically for any pattern of flow. Thus the performance of non –ideal flow reactors can be determined precisely given the residence time distribution and the rate constant for the first order reactions. The Mean and Variance:

P a g e 12 | 42


It is frequently desirable to characterize a distribution by a few numerical values. For this purpose the most important measure is the location of the distribution. This is called the mean value or the centroid of the distribution. Thus for a C vs. t curve the mean is given by, ∞

∫ tCdt 0 ∞

t- =

∫ Cdt 0

If the distribution curve is known at a number of discrete time values l, then

t- =

∑ tiCi Δti ∑ Ci Δti

The next most important descriptive quantity is the spread of the distribution. This is commonly measured by the variant σ2, defined as t∞

∞

∫ (t−´t )2Cdt σ2 =

0

∞

∫ Cdt

∫ t 2Cdt =

0

0

∞

∫ Cdt

´ -( t

)2

0

Again in discrete form

∑ (t−´t ) 2 Ci ∆ t ∑ Ci ∆ ti

2

σ =

∑ ti2 Ci ∆ t ∑ Ci ∆ ti

=

´ -( t

)2

The variance represents the square of the spread of the distribution and has units of (time) 2. It is particularly useful for matching experimental curves to one of a fairly of theoretical curves. When used with normalized distribution for closed vessel these expressions simplify somewhat thus for continuous curves or for discrete measurements at equal time intervals mean becomes ∞ -

t=

∫ tEdt 0

=

∑ tiEi ∑ Ei

=

∑ tiEi ∆ t

And variance becomes

P a g e 13 | 42


∞

(t −´t ) 2 Edt σ2 = ∫ 0

2

σ =

∑ ti2 Ei ∑ Ei

=

´ -( t

∫ t 2 Edt 0

)2 =

´ -( t

∑t 2E∆t

)2

´ -( t

)2

Procedure: 1. Set up the apparatus as shown in the diagram. 2. Find out the Mean residence Time for the reactor at a particular fixed flow rate, using the KMnO4 solution. Mean Residence Time is required for complete colour removal KMnO4 solution. 3. Now keeping the flow rate constant inject 2ml 2N NaOH solution and take the samples of the outlet stream from the reactor at a definite interval of time. 4. Titrate the samples against 0.5N HCL solution and examines their concentration. 5. Carry out necessary calculations and plot E, F, and E ɵ and F ɵ curves. 6. Find out conversion for first order irreversible reaction. Results: For the first order irreversible reaction, conversion =

%

Conclusion: .

P a g e 14 | 42


Experiment – 4 (Laminar flow system) Aim:

Date: / /20 To study the Fθ curves from the E curves and compare with theoretical result for the given Laminar flow System.

Apparatus:

Tubular reactor (laminar Tube), Tracer injection device, Stop watch, Sample bottles, titration set.

Chemical:

10ml 2N Potassium Permegnet (KMno4) solution, 50ml 2N Sodium hydroxide (NaOH) solution, 1000ml 0.1N Hydroxide acid (HCL) solution, Phenolphthalein (indicator).

Fluid used: Water Theory: A tubular reactor with laminar flow can be taken has good approximation to segregated flow. If dispersion model is not used and the basic assumption of dispersion due to molecular diffusion is negligible since the flow is segregated and the velocity profile is known, RTD can be calculated theoretically. The actual experimental result can be compared with the theoretically result to get the deviation from ideal laminar flow. Fθ curve for ideal laminar flow: The velocity profile foe laminar flow of fluid through tubular reactor of radius ‘R’ is,

{ ( )}

r u=2ū 1R

2

(1)

Where, u is average velocity (cm/sec) R is radius (cm) Consider ‘r’ at which fluid spends time‘t’ inside the reactor. All the fluid, flow between reactor center (axis) and up to distance ‘r’. It will take less time to pass through the reactor than the fluid at ‘r’. This is given by r

q=∫ 2πrudr 0

(2)

Substituting ‘u’ from (1) into (2) and integrating,

P a g e 15 | 42


q=4π ū

{

2

2

r r 2 4R 2

}

(3)

Also, total flow rate 2

Q=4π R ū

Fθ =

q R

(4)

From (3) and (4), Fθ =

Equation (5) yields

Fθ

{

3

3

4 r r − 3 3 R 2 4R

}

(5)

for laminar flow system

Conversion directly from tracer information: Linear process: A variety of flow pattern can give the, same tracer output curve. For linear processes however, these all result in the same conversion; consequently we may use any convenient flow pattern to determine conversions, as long as the pattern selected gives the same tracer response curve as the real reactor. The simplest pattern to use assumes that each element of fluid passes through the vessel with no intermixing with adjacent, the age distribution of material in the exit stream telling how long each of the individual elements remains within the reactor. Thus for reactant A in the exit stream. ∞

´ A = ∫ CA element Edt C t=0

These conversion equations can be solved either graphically or numerically for any pattern of flow, thus the performance of non-ideal flow reactor can be determined precisely given the residence time distribution and rate constant for the first order reactions. The mean and variable: It is frequently desirable characterize a distribution by a few numerical values. For this purpose, the most important measure is the location of the distribution. This is called the mean value or the centroid of the distribution, Thus for a C versus t curve the mean is given by

P a g e 16 | 42


∫ tCdt ´t ≡

0 ∞

∫ Cdt 0

If the distribution curve is known at a number of discrete time values /, than ´t ≅ ∑

t i Ci ∆ t i

∑ Ci ∆ t i

The next most important descriptive quantity is the spread of the distribution. This is 2 commonly measured by the variance σ , defined as∞

∞

∫ ( t−´t ) Cdt ∫ t2 Cdt 2

σ 2=

0

=

∞

∫ Cdt

0

∞

−( ´t )

2

∫ Cdt

0

0

Again in discrete form σ

2

2 ∑ ( t i−´t ) C i ∆ t i ∑ t 2i C i ∆t i ´ 2 ≅ = −( t ) ∑ C i ∆ ti ∑ C i ∆t i

The variance represent the square of the distribution and has unit of (time) 2. It is particularly useful for matching experimental curves to one of a fairly of theoretical curves. When used with normalized distribution for closed vessels these expressions simplify somewhat. Thus for a continuous curve or for discrete measurement at equal time intervals mean becomes ∞

´t =∫ tEdt ≅ ∑

t i Ei

∑ Ei

0

=∑ t i E i ∆t

And the variance becomes ∞

∞

σ =∫ (t−´t ) Edt =∫ t 2 Edt −( ´t )2 2

2

0

0

P a g e 17 | 42


t 2i E i ∑ 2 2 2 σ= − (´t ) =∑ t i E i ∆ t−(´t ) ∑ Ei 2

Procedure: 1. Set up apparatus as shown in diagram. 2. Find out the mean residence Time for the reactor at a particular fixed flow rate, using the KMnO4 solution. Mean residence Time required for complete color removal after injection for KMnO4 solution. 3. Now keeping the flow rate constant inject 2ml 2NaOH solution and take the sample of the outlet stream from the reactor at definite interval of time, 4. Titrate the samples against 0.5N HCL solution and examine their concentration Eθ F 5. Carry out necessary calculations and plot E, F, and θ curves. 6. Find out conversion for first order irreversible reaction.

P a g e 18 | 42


Result: For the first order irreversible reaction Conversion 1. For given laminar system = ……………..% 2. For ideal PFR = ……………...% 3. For ideal CSTR

= ……………. %

Conclusion:

P a g e 19 | 42


P a g e 20 | 42


Experiment – 5 (Tanks in Series Model) Date: / /20 To study the F0 curves from the E curves and predict the conversion of given first order reaction with known rate constant, by applying tanks in series model

Aim :

Apparatus:

Tubular reactor, Tracer injection device, stop watch, Sample bottles, Titration Set.

Chemicals:

10 ml 2 N Potassium Permanganate (KMnO4) solution, 50 ml 2N Sodium Hydroxide (NaOH) solution, 1000 ml 0.1N Hydrochloric acid (HCL) solution, Phenolphthalein (indicator)

Fluid used:

Water

Theory: In this model we view the fluid to flow through a series of equal sized ideal stirred tanks, and the one parameter of this model is the no. of tanks in this chain for W tanks in series. N−1

EƟ = ( Nti )E =

N (Nθ) ( N−1)!

e−Nθ

Where, ti = mean residence time in one tank t = Nt mean residence time in the N tank system Ɵi = t/ti = Nt / tֿ Ɵ = t/tֿ= t/Ntiֿ The variance for the system of tanks in series is found to be σ

2

σ θ2

´ 2 = N (t i)

=

σ2 ´t 2

=N

=

´t N

2

( )

=

´t N

2

( )

1 N

Procedure: 1. Set up the apparatus as shown in the diagram. 2. Find out the mean residence time for the reactor at a particular fixed flow rate, using the KMnO4 solution. Mean residence time required for complete colour removal after injection of the KMnO4 solution. P a g e 21 | 42


3. Now keeping the flow rate constant inject 2ml NaOH solution and take the samples of the outlet stream from the reactor at a definite interval of time titrate the samples against 0.5N HCL solution and examine their concentration. 4.

P a g e 22 | 42


5. Carry out necessary calculations and plot E, F, EƟ and FƟ curves. Find out conversions for first order irreversible reaction. Results: For the first order irreversible reaction, conversion 1. Experimental value 2. From equation of Tanks on series model 3. For Ideal PFR 4. For Ideal CSTR

= = = =

% % % %

Conclusion:

P a g e 23 | 42


P a g e 24 | 42


Experiment – 6 (Dispersion Model) Date: / /20 Aim:

To study the Fɵ curves from the E curves and predict the conversion of gives first order reaction with known rate constant, by applying Dispersion Model

Apparatus:

Tubular Packed Bed reactor, Tracer injection device, Stop watch, sample bottle titration set

Chemical:

10 ml 2 N potassium permanganate (KMnO4) solution, 50 ml 2N sodium Hydroxide (NaoH) solution, 1000ml 0.1 N Hydrochloric acid (HCL) solution, Phenolphthalein (indicator).

Fluid used: Water Theory: Consider a plug flow of fluid, on top of which is superimposed some degree of back mixing or intermixing, the magnitude of which is independent of position within the vessel. This condition implies that there exists no stagnant packet and no gross bypassing or shortcircuiting of fluid in the vessel. This is called the dispersed plug flow model, or simply dispersion model. Note that with varying intensities of turbulence or intermixing the predictions of this model should range from plug flow at one extreme to mixed flow at other. As a result the reactor volume for this model will lie between those calculated for plug and mixed flow. Since the mixing process involves a shuffling or redistribution of material either by slippage or eddies, and since this is repeated a considerable number of times during the flow of fluid through vessel we can consider these disturbance to be statistical in nature, somewhat as in molecular diffusion. For molecular diffusion in the x direction the governing differential equation is given by Fick’s law∂C ∂t

=D

∂2 C ∂ X2

Where, D is the coefficient of the molecular diffusion, is a parameter which uniquely characterizes the process. In an analogous manner we may consider all the contributions to back mixing of fluid flowing in the x direction to be described by a similar form of expression, or

P a g e 25 | 42


∂C ∂t

∂2 C = D ∂ X2

Where, The parameter D, which we call the longitudinal or axial dispersion coefficient, uniquely characterized the degree of back mixing during flow. We use the terms “longitudinal” and “axial” because we wish to distinguish mixing in the direction of flow from mixing of in the lateral or radial direction, which is not our primary concern. These two quantities may be quite different in magnitude. For example, in stream line flow of fluids through pipes, axial mixing is mainly due to fluid velocity gradients whereas radial mixing is due to molecular diffusion alone. t tu In dimension from where z = (ul+x)/L and ɵ = t = L , the basic differential equation representing this dispersion model becomes ∂C ∂θ

Where, the dimensionless group

=

( )

( uLD )

2

D ∂ C uL ∂ z2

∂C - ∂z

called the vessel dispersion number, is the

parameter which measure the extent f axial dispersion. Thus, D → 0 Negligible dispersion, hence plug flow uL

( )

(DuL ) →∞

Large dispersion, hence mixed flow

This model usually represent quite satisfactory that deviates not too greatly from plug, thus real packed bed and tubes (long ones if flow is streamline).

P a g e 26 | 42


Procedure: 1. Set up the apparatus as shown in the diagram. 2. Find out mean residence time for the reactor at a practical fixed flow rate, using the KMnO4 solution. Mean Residence Time is the time required for complete colour removal after injection of the KMnO4 solution. 3. Now keeping the flow rate constant inject 2ml 2N NaOH solution and take the sample of the outlet stream from reactor at a definite interval of time, 4. Titrate the sample against 0.5N HCl solution and examine their concentration. 5. Carry out necessary calculation and plot E, F, E and Fɵ curve 6. Find out conversion for first order irreversible reaction. Result: For the First order irreversible reaction, Conversion 1. Experimental value 2. From equation of dispersion model 3. For Ideal PFR 4. For Ideal CSTR Conclusion:

P a g e 27 | 42


P a g e 28 | 42


Experiment – 7 (Mixed Flow Reactor) Aim: dFθ =E θ dθ

Date: / /20 To study the F0 curves from the E curves and predict the conversion of given first order reaction with known rate constant, for mixed flow reactor. To verify the relation

Apparatus: Continuous Stirred Tank Reactor, Tracer injection device, stop watch. Sample bottles, Titration set. Chemicals:

10 ml 2 N Potassium Permanganate (KMnO4) solution, 50 ml 2N sodium Hydroxide (NaOH) solution, 1000 ml 0.1 N Hydrochloric acid (HCl) solution, Phenolphthalein (Indicator)

Fluid used:

water

Theory: The real stirred tank For most applications the real stirred tank with sufficient agitation can be taken to approximate the ideal of mixed flow. There are some cases, however, where deviation from this ideal should be considered, for example in large tanks with insufficient agitation and for fast reactions where the of reaction is short compared to the time for mixing and for achieving uniformity of composition. It is here that mixing models are needed. Not only will these be useful for the real stirred tank, they will have numerous other applications. Such as Io represent the distribution of chemicals and drugs in animals and man. Distribution curves: E, the Age distribution of fluid leaving a vessel; It is the distribution needed to account for non-ideality in flow. It is represented by equation: ∞

∫ Edt 0

Also Edt-fraction of exit stream between t and (t + dt) C-curve: Normalized response of an impulse or a data function is called C curve. For normalizing the measured concentration ‘C’ is divided by ‘Q’ which is the area under curve: ∞

Q=

∫ Cdt 0

And normalizing concentration, P a g e 29 | 42


Ci Q

C=

The plot of, Ci Vs. tt , gives so called 'C’- curve. The area under this curse is: ∞ ∞ Ci Cdt ∫ ∫ = 0 Q dt = 1, 0 ∞

As, Q =

∫ Cdt 0

= 1, hence C=E

F Curve The response curve of the tracer in exit stream from the reactor, for a step input, is called F curve .This response is measured as C/C0. Relation between E and F is given by, ∞

F=

∫ Edt 0

Above distribution curve is based on dimension less time units, are noted by subscript i.e. FƟ and CƟ curves. Where

θ=

θ

t ῖ

P a g e 30 | 42


Procedure: 1. for impulse input 1. Set up the apparatus as shown in the 2. Find out the Mean Residence Time for the reactor at a particular fixed flow rate, using the KMnO4 solution. Mean Residence Time is the time required for complete color removal after injection of the KMnO4 solution. 3. Now keeping the flow rate constant inject 2ml 2N NaOH solution and take the samples of the outlet stream from the reactor at a definite interval of time, 4. Titrate the samples against 0.5N Hecla solution and examine their concentration 5. Carry out necessary calculation and plot E, F, EƟ, and FƟ curves. 6. Find out conversion for first order irreversible reaction. 2. for step There is only once difference in the procedure for step input, here at time t = 0 reactor is filled with solution containing tracer (here 0.125N NaOH solution) Result Conversion: 1. Impulse in put: a. For given reactor b. For ideal PFR c. For ideal CSTR

= = =

2. Step input a. For given reactor b. For ideal PFR c. For ideal CSTR

= = =

P a g e 31 | 42


P a g e 32 | 42


Experiment – 8 (Slurry Reaction kinetics) Date: / /20 Aim:

To study the slurry reaction kinetics

Theory: In this case which contains reactant A is bubbled through liquid B, which contains suspended solid catalyst surface, to react with B. for this system following resistance are viewed to act in series. STEP 1: Reactant A must cross the gas film to reach the gas-liquid interface. STEP 2: A must cross the liquid film and reach the main body and liquid. STEP 3: A must cross the liquid film surrounding the catalyst particles with the liquid component B. STET 4: A then reacts on the surface of the catalyst particles with liquid component B. Let us suppose that B is in large excess; for the surface reaction to be first order with respect to A with rate constant, let us also be define

Q =

interfacial surface volume of liqiud

Q1 =

Surface of suspended catalyst Volume of liquid

And

Then the rate for this individual step is given by, −1 dNA r K Ag P – P Ai – A1 = V1 = ai ( A ) dt =

K A 1 a1

=

K A1 a s ( CA1 – CAS )

=

K1 a s CAs

(

CAi – CA1 )

And on combining we find,

P a g e 33 | 42


–

1 HA 1 = + KAg a 1 K A1 a 1

r A, L

+

HA K'

A1 as

HA + K1 a s

PA

……………. (1) Gas-liquid Interface

catalyst surface

It is important to know whether the primary resistance lies at the gas- liquid interface or dt the surface of the particles. This is easily found by changing separately gas-liquid interracial areas and the amount catalyst in the liquid. This will determine how best to scale up the operation. The following rearrangement of the equation (1) is often in correlating data in kinetic studies of the system:

–

PA r A,t

1 = ( K Ag

HA + K A1

)

1 a1

HA + ( K

+

' A1

HA k1

1 ) as

=

C1

1 1 + C2 a1 as

The hydrogenation of the organic compound represent a typical application of slurry reactions. Since pure hydrogen invariably is used the gas phase resistance just doesn’t exists in these operations. Supporting Element: Slurry reactor can use very fine catalyst particle and this can lead to problem of separating catalyst from liquid trickle beds. Unfortunately there large particles in trickle bed means much lower reaction rates with regard to rate the trickle be commonly hold its down. For very slow reaction on porous solid where pure different diffusion limitations don’t appear even for large particles. For very fast reaction on non- porous catalyst coated particles. Overall, the trickle bed is simple. The slurry reactor usually has a rate and fluidized bed is somewhere in between. Application: a. The catalyst hydrogenation of petroleum fraction to remove sulphur impurities. b. Hydrogen is very soluble in the liquid, high pressure is used, while the impurity is present in the liquid in low concentration. c. The catalytic oxidation of the hydrocarbon with air or oxygen since oxygen is not very soluble in the liquid while the hydrocarbon could well be present in high concentration, we would end up in entrance.

P a g e 34 | 42


Experiment – 9 (Solid Catalyzed Reactor) Date: / /20 Aim: To study the experimental method for finding the rate kinetics for solid-catalyzed reactors. Theory: Any type reactor with known catalyzing pattern may used to explore the kinetics of catalyticreactions. Since ,only one fluid phase is present in these reaction and therefore rate can be found as with homogeneous reaction; the only special precautions to observe is to make sure that the performance equation used is dimensionally correct and that its terms are carefully and precisely defined. The experimental strategy in studying catalytical kinetics, usually involves measuring the extent of conversion of gas passing in steady flow through a batch of solid. Any flow pattern can be used as long as the pattern known. If it isn’t known then the kinetics cannot be found. A batch reactor can also be used. In turn, we discuss the following experiment devices    

Differential (flow) meter Integral plug flow reactor Mixed flow reactor Batch reactor for both solid and gas

Differential reactor: We have a differential flow reactors when choose it, consider the rate to be constant at all points within the reactor, since rate is concentration dependent. This assumption is for shallow small reactor, but this is not necessary so, take an example of a slow reaction where the reactor can be large or for zero order reaction kinetics where composition changes can be large. For each run in a differential reactor the plug flow performance equation becomes, XA ,∈¿ XA ,∈¿ dXA= XA , out −XA ,∈

¿

( −r ' A ) avg

W dXA 1 =∫ XA , out = ∫ XA , out ¿ ' FAO ¿ −r ' A (−r A ) avg ¿ Integral reactor: When the variation in the reaction rate within a reactor is so large that, we have an integral reaction, since rates are concentration dependent. Such large variation in rate may be expected to occur, when the composition of reactant fluid changes significantly in passing through the reactor. We may follow one of the procedures in searching for a rate equation. P a g e 35 | 42


Integral analysis: Here, a specific mechanism with its corresponding rate equation is put to the test by integrating the basic performance equation to give similar. XA

W dXA =∫ FAO 0 −rA Differential analysis: Integral analysis provides a straight forward rapid procedure for testing some of the simpler rate expressions, however the integrated forms of these expressions become unwisely with more complicated rate expression. In this situation the differential method of analysis becomes more convenient. −rA=

dXA dXA = dW /dFA W d( ) FA

Using this equation we can find the rate of the reaction.

P a g e 36 | 42


Experiment – 10 (Rate controlling Step) Date: / /20 Aim: Determination of rate controlling step Theory: The kinetics and rate controlling steps of fluid solid reactions are detected by noting rate. The progressive conversion of particles is influenced by Particle size and operating temperature. This information of particles can be obtained in various ways depending on the facilities available and the material at the end. The following observations are guide to experimentation and interpretation of experiment data. Temperature: The chemical steps in usually much more temperature sensitive than physical steps, hence, experiments at different temperatures should be easily distinguish between Ash or film diffusion on the other hand as the controlling step. Tune: In the figure 1 and 2 shows that the progressive conversion of spherical solids when chemical reaction film diffusion and ash diffusion in turn control result of kinetics runs compared with this predicted word should be indicated the rate controlling steps, unfortunately the difference between the ash diffusion and chemical reaction are controlling step is not great. It may be marked by the scatter in experimental data. Conversion time curve analogous to these in fig. 1 and 2 using particle size. Particle size: The time needed to achieve the same fraction for particles of different but unchanging size is given by, t<*R p or fjm diffusion controlling. (The important drop as Reynolds number rises) T cc R p or jj diffusion controlling. T ce R p or chemical reaction controlling. Kinetics runs with different sizes of particles can distinguish between reaction in which the versus Film resistance; When a solid ash form during reaction the resistance of gas phase reactant through this ash is usually much greater than the gas film surrounding the particle. Hence in presence of a nonflaking ash layers film resistance can safely be ignored in addition ash resistance is unaffected by change in gas velocity.

Predictability of film resistance; P a g e 37 | 42


The magnitude of film resistance can be estimated from dimensionless correction such as equation. K y dP y = 2 + 0.6 (Sc)1/3(Re)1/2 = 2 + 0.6 (N / ρD )1/3(dpmρ / D

μ )1/2

Thus, it can be observed rate, approximately equal to the calculated rate, suggest that film resistance controls. Overall versus Individual resistance: If a plot of individual rate co-efficient is a function of temperature. The figure shows the overall co-efficient given by equation cannot be higher than any of the individual co-efficient. With those observations; we can usually discover with a small carefully planned experimental programme which is the controlling mechanism. Let illustrate the inter play of resistance with the studied gas-soild reaction of pure carbon particles with oxygen. C + O2 CO2 B + A  Gaseous product With rate equation: −1 sec

dNB dt

=

−1 dR ρ 4 πR B dt

dR = -ρB dt

= K-11CA

Since no ash is formed at any time during reaction, we have a case of kinetics of shrinkage particles for which two resistances at most.

P a g e 38 | 42


Experiment – 11 (Diagnosing reactor ills) Date: / /20 Aim

:

To study the Diagnosing reactor ills.

Theory : Flow models can be of different levels of sophistication and the compartment models of chapter are the next stage beyond the very simplest stages that assumes the extremes of plugflow mixed flow. In the compartment model. We consider the vessel and flow through it as follows: Total volume V Vp = plug flow reactor. Vm = mixed flow region. Vd = stagnant region with in the vessel. Total through flow is Va = Active flow that through the plug and mixed flow region. Vb = By pass flow. Vr = Recycle flow. By comparing the ‘E’ curve for the real vessel with the theoretical curve for various combination of compartment and the rough flow, we can find which model best fits the real vessel of course the fit will not be perfect, however model of this kind are often a reasonable approximation to the real vessel. The ‘E’ curve looks like for various combinations of the above elements certainly not all combination. Diagnosing reactor ills: Combined models are useful for diagnosing purpose to pinpoint faulty flow “and suggest causes. For example, if you except plug flow and you know f = V/v. Stim trim curve means reasonable good flow. Late tracer is parring material balance say it cannot happen so the only expansion are,

P a g e 39 | 42


P a g e 40 | 42


Experiment – 12 (Kinetic reagent for mass transfer and reaction) Date: / /20 Aim:

To study the kinetic reagent for mass transfer and reaction.

Theory: Here, we have three factors to consider what happens in the gas film in the liquid film and in the main body of the liquid as shown in figure. All short of special forms of rate constant K, K G and KL the conversion ratio of reactant P O/PB and Henry’s law constant HA. If runs out there are eight cases to consider going from the extreme of infinitely first reaction rate (mass transfer control) to the other extreme of very slow reaction rate (no mass transfer resistance need be considered.) The eight special cases each with its particular rate equation are from infinitely first to vary slow reaction as follow. Case A: Instantaneous reaction with low Cb. Case B: Instantaneous reaction with higher Cb. Case C: Fast reaction in liquid film with low Cb, Case D; Fast reaction in liquid film with higher Cb. Interface behavior for the liquid phase reaction aA(From Gas) + bB (Liquid) = Product (Liquid) For the complete range of rate of the reaction and of the mass transfer. Case ‘E’ and ‘F’ intermediate rate with reaction in the film used in the main body of the liquid. Case G = slow reaction in main body but with film resistance. Case H – Slow reaction, no mass transfer resistance. We discuss these special cases which present their particular rate equation. Later after we present the general rate equation

-rA,L =

1 H HA 1 A + + k Ag ai k A aF k cs Fl

The absorption of A from gas is larger when reaction occurs within the liquid film than for straight mass transfer. Thus, for the same concentration as the two boundaries of the liquid film use, hence,

(

Liquid film enhancement, E factor

)

=

backup of A when reactionoccurs ( raterateofofbackup of A for straight mass transfer )

P a g e 41 | 42


Since, CAi, CA, CD, CBi in two cases The value of E is always greater, or equal to one. The only problem have is to evaluate E, the enhancement factor E is dependent on two quantities, E i= the enhancement factor for an infinitely fast reaction H= maximum possible conversion in the film composed with max. Transport the film.

P a g e 42 | 42

Cre 2 Lab Manual

Recommend Documents