Saturday, 10 March 2018

Batch Size selection for Scale-up & Feasibility Study

Hiii all, back after a long time .....!!!

Because of my stringent schedules, Can't check the comments and sorry for not replying to those, Apart from that i'm quite happy as followers were getting increased day to day, in parallel it was desperate as i'm not getting any author requests. 

Anyway, recently i've received a mail from a Process Engineer working in one of the top Pharma company, his name is Mr. Ram Prakash, and the query is, 

"How to select a batch size during scale up ? As mostly management is not interested in going for green field projects, how to select a appropriate batch size ?"





So, as per me the query is slightly a common issue, because in the early stages of process engineering during learning about scale-up techniques there will be lots and lots of things, but while coming to implementation, a process engineer should always think in company's favour and the decision that is taken should be in company's interest. As management always want high returns with low investments.

Rookie Process Engineer's may not accept this, but after gaining some experience one should accept it.
Anyway, i'll explain you clearly with basic calculations and a general case study.







Usually Scale-up shall be done based on 5 parameters, those are:

1) Equating power per unit volume,
2) Equating Reynold's number,
3) Equating Tip speed,
4) Equating Shaft speed,
5) Equating Froude's number.




For additional information based on above mentioned parameters, Please visit: 
https://pharmacalc.blogspot.in/2015/05/api-scaleup-calculations.html

Scale-up's are most common in industries with certain amount of risk. Always there will be a some risk under every task, But a process engineer is treated as a true professional when the executed scale-up get succeeded with calculated risk. For that there may be many ways, out of which Power per unit volume is a basic and traditional path. Everyone knows that, but mostly not applied practically . 





Mostly the equipment selection shall be done based on availability (no offense to anyone). But the scale-up batch size selection most probably rests with a process engineer. And the available chance need to be capitalised properly.

Lets start with a case study, Consider a reaction and we need to scale it up.

Let it be some condensation / dehydration reaction. 

and initially the reaction [35 Kg batch size] is carried out in a 500 L SSR in small scale.

2000 L SSR is available for Scale-up,  Agitator remains the same as PBT.

Below is the Calculation :

Parameters
UOM
Values
Formulae
Reactor Capacity (REV)
L
500
-
Batch Size (B)
Kg
35
-
Reaction mass volume (RV)
L
265.3
7.6 volumes
% Occupancy (O%)
-
53.06
REV/V
Cylindrical height  (H)
m
0.865
-
Reactor ID  (D)
m
0.815
-
L/D 
-
1.06134969
H/D
Agitator Dia (d)
m
0.46
-
% sweep (S%)
-
56.4417178
(d/D)*100
Torispherical Volume (Tv)
m3
0.04817956
0.089x(D^3)
Cylindrical Volume (Cv)
m3
0.45102538
0.785x(D^2)xH
Volume in Cylindrical section (Rcv)
m3
0.26525182
(Rv-Tv)/1000
Cylindrical height of Rxn. Mass (Rh)
m
0.50871378
Rcv/(0.785x(D^2))
L/D for RM
-
0.62418868
Rh/D
Agitator type
-
PBT
-
Np
-
5
-
RPM 
-
96
-
RPS (N)
rps
1.6
RPM/60
Density of the Rxn. Mass ®
Kg/m3
780
-
P
J/S
329.013496
Np x r x (N^3) x (D^5)
P
KW
0.3290135
P/1000
V
m3
0.2653
Rv/1000
P/V
KW/m3
1.24015641
P/V

So as per the batch size of 35 Kg, the power per unit volume is 1.24 KW/Cu.m.





To learn how Height and Dia of reactor is calculated, follow below link:
http://pharmacalc.blogspot.com/2016/03/how-to-find-reactor-heat-transfer-area.html
To learn how to determine the volume occupied in a torispherical dish, follow below link:http://pharmacalc.blogspot.com/2016/03/how-tocalculate-volume-occupied-by.html

Now, we need to consider some random batch sizes, [ kindly don't ask me how to consider those random too, that depends on our common sense],


and perform the same calculation like we have done for 35 Kg batch size,



Parameters
UOM
Values
Formuale
Capacity
L
2000
-
Batch Size
Kg
70
80
90
95
100
110
120
-
Reaction mass volme
L
530.6
606.4
682.2
720.1
758
833.8
909.6
7.6 volumes
% Occupancy
-
26.53
30.32
34.11
36.01
37.9
41.69
45.48
REV/V
Cylindrical height 
m
1.79
1.79
1.79
1.79
1.79
1.79
1.79
-
Reactor ID 
m
1.4
1.4
1.4
1.4
1.4
1.4
1.4
-
L/D 
-
1.28
1.28
1.28
1.28
1.28
1.28
1.28
H/D
Agitator Dia 
m
0.56
0.56
0.56
0.56
0.56
0.56
0.56
-
% sweep
-
40
40
40
40
40
40
40
(d/D)*100
Torispherical Volume
m3
0.244
0.244
0.244
0.244
0.244
0.244
0.244
0.089x(D^3)
Cylindrical Volume
m3
2.754
2.754
2.754
2.754
2.754
2.754
2.754
0.785x(D^2)xH
Volume in Cylindrical section
m3
0.53
0.606
0.682
0.72
0.758
0.834
0.909
(Rv-Tv)/1000
Cylindrical height of Rxn. Mass
m
0.345
0.394
0.443
0.468
0.492
0.542
0.591
Rcv/(0.785x(D^2))
L/D for RM
-
0.25
0.28
0.32
0.33
0.35
0.39
0.42
Rh/D
Agitator type
-
PBT
PBT
PBT
PBT
PBT
PBT
PBT
-
Np
-
5
5
5
5
5
5
5
-
RPM 
-
96
96
96
96
96
96
96
-
RPS (N)
rps
1.6
1.6
1.6
1.6
1.6
1.6
1.6
RPM/60
Density of the Rxn. Mass 
Kg/m3
780
780
780
780
780
780
780
-
P
J/S
879.76
879.76
879.76
879.76
879.76
879.76
879.76
Np x r x (N^3) x (D^5)
P
KW
0.88
0.88
0.88
0.88
0.88
0.88
0.88
P/1000
V
m3
0.53
0.61
0.68
0.72
0.76
0.83
0.91
Rv/1000
P/V
KW/m3
1.66
1.45
1.29
1.22
1.16
1.06
0.97
P/V


After performing the calculation, it is clear that power per unit volume of only three batch sizes 90, 95, 100 Kgs were comparable to that of standard, and remaining four batches were out of bounds.





So, the conclusion is that if we are proceeding with 2000 L SSR, 90 Kgs, 95 KGs, 100 Kgs batch size is preferable and can be proceeded with calculated risk.

That's it....!!!

For automated Calculator follow the below link and download the spread sheet.

Reference Screen:











If any queries pl do comment or use contact me page.

Pl feel free to comment / contact.


About The Author


Hi! I am Ajay Kumar Kalva, Currently serving as the CEO of this site, a tech geek by passion, and a chemical process engineer by profession, i'm interested in writing articles regarding technology, hacking and pharma technology.
Follow Me on Twitter AjaySpectator & Computer Innovations



30 comments:

  1. Dear ajay,
    its nice post shared by you...but by increasing the RPM we can match power per unit volume ratio.
    as you mentioned above formulae P=Np x r x (N^3) x (D^5 RPM is only parameter that can be vary & it directly proportion to power.

    ReplyDelete
    Replies
    1. Dear Mahesh,

      Great catch, but in some cases when you want to carry out reaction in Base mode then increasing RPM will have serious effect on agitator coupling, in that case if you want to enhance RPM then you need to increase torque, for that the shaft dia should be increased and the motor capacity should be more.

      So one change will have impact on many things, which need to be carefully evaluated.

      Regards,
      AJAY K

      Delete
  2. When we calculate RPM for scale with above Dimensionless number like Nre,Power No or tip but all are not giving same RPM so which one will be considered??

    ReplyDelete
  3. Dear Ajay,
    Please send the post if possible on solvent recovery & what are the parameters to be set like feed rate,Reflux rate ,Bottom temp & Top temp of column etc consider basically azeotropic Distillation.

    ReplyDelete
  4. How did you get reaction mass volume(RV) as 265.3 Liters?

    ReplyDelete
    Replies
    1. Dear Sri,

      35 x 7.58 (~7.6) = 265.3 L.

      Regards,
      AJAY K

      Delete
    2. Dear sir,

      I meant what is 7.58?

      Delete
    3. Dear Sri,

      those are the volumes of reaction mass at that particular reaction step, it may vary from product to product, anyway to this post, i'll be adding a video tutorial.

      Regards,
      AJAY K

      Delete
  5. Dear Ajay
    let me know How do you calculate the steam pressure requirement based on temperature basis 140 deg c

    ReplyDelete
    Replies
    1. Dear Sir,

      As per me there wont be any particular calculation for that, but just we need to follow steam tables,
      Pl follow below link:

      http://pharmacalc.blogspot.com/2018/04/steam-tables.html

      So, the Steam Pressure shall be 1.4 bar i.e., 1.4276 Kg/cm2.

      Still any queries pl feel free to ask.

      Regards,
      AJAY K

      Delete
  6. Dear Ajay

    Im working in process engineering department, one of the reaction i am handing reactants to products conversion time 43 hrs(Reaction time ), I have to increase the productivity with out changing the reactor for that i need to reduce the total reaction time for that

    what are the calculation are required in engg points of you to increase the productivity and reduce the time cycle
    On what parameters can i make the changes
    Can you please explain by taking the example

    ReplyDelete
    Replies
    1. Dear Sir,

      What type of reaction is that ?
      What are the type of byproducts that are formed during the reaction ? whether they are washable in later stages or not ?
      If possible share the brief process, so that i can help you.

      Regards,
      AJAY K

      Delete
  7. Please share unprotected sheet

    ReplyDelete
    Replies
    1. Dear Nitin,

      Share your mail id, so that i can share you,

      Regards,
      AJAY K

      Delete
  8. Mr Ajay Kumar, can you please provide me the notes on water ejector and steam ejector type vaccume pump

    ReplyDelete
    Replies
    1. Dear ,

      Share you mail id / contact me at pharmacalc823@gmail.com.

      Regards,
      AJAY K

      Delete
  9. Dear ajay sir,
    Please tell me the role process engineer in pharmaceutical industry as im fresher?

    ReplyDelete
    Replies
    1. Hey buddy,

      The question is clear, and i'll tell you far clear answer,
      Pharma directly implies manufacturing, in a manufacturing industry always the production department is king, and all other departments are supportive departments to production.
      And coming to our role, as a supporting team our job is to provide some momentum to production through
      1. Speed(time-cycle reduction of the manufacturing operations),
      2. Robustness(address the loopholes in the manufacturing operations),
      3. innovations(implementing new techniques which can produce higher outputs through minimum inputs),
      4. Recoveries(enhancing the reuse potential of the solvents),
      5. Cost improvements(Reducing the excess raw materials consumption through optimization),
      6. Safety(Eliminating the existing hazards in the manufacturing process which acts as bottle necks),
      7. Quality management(Enhancing the quality measures to improve/optimize the output quality).

      I think the above mentioned things are a part of our chemical engineering. So a chemical engineer, these are our responsibilities in Pharma.
      If nay queries contact me through the contact me page.

      Next time pl comment with your good name.

      Best Regards,
      AJAY K

      Delete
  10. Dear sir,
    When we need to scale up from lab scale to pilot scale we need to consider the lab scale stirrer blade power number to calculate the power/vol value. right? So can you please suggest me how much should I take for lab scale stirrer blade?

    ReplyDelete
    Replies
    1. Dear Sir,

      Usually it is very difficult to scale up from lab scale to pilot scale, because while coming to higher batch sizes in plant scale, we need to select facility based on availability. Mostly in Lab scale they will use Half moon agitator and RBF for experiments, so we can't judge / match the L/D ration of RBF and reactor.

      So never scale up from lab to pilot[that's my suggestion], based on the MOC and based on the scale of agitation select the reactor & RPM,
      If required we can match power per unit volume from pilot to plant large scale.

      If nay queries feel free to comment / contact me through the Contact me Page available at top of the page.

      Best Regards,
      AJAY K

      Delete
  11. nkhese@gmail.com
    Require unprotected sheet

    ReplyDelete
    Replies
    1. Hey Nitin,

      Surely, kindly reach out at pharmacalc823@gmail.com

      Regards,
      AJAY K

      Delete
  12. dear Sir,

    can you explain how scale up and select the filtration equipment from pilot to plant

    ReplyDelete
    Replies
    1. Dear ,

      Pl find the below links,
      https://www.pharmacalculations.com/2018/08/evaluate-filtration-feasibility-in-anfd.html
      https://www.pharmacalculations.com/2017/11/filtration-equipment-types-of-filters.html
      https://www.pharmacalculations.com/2019/02/scaleup-process-equipments-pharma.html

      Best Regards,
      Ajay K

      Delete
  13. Dear Ajay,
    krishnajilavand@gmail.com
    Kindly sent unprotected sheet
    Regards,
    Krishna

    ReplyDelete
  14. But be practical management always ask to go for at least 80% occupancy of the reactor, in that case batch size will be around 240 kg's.
    what can be the consequences, if we go for full volume available by simply matching the RPM from pilot to commercial scale.

    ReplyDelete
    Replies
    1. Dear ,

      Management will always encourage to enhance capacity with minimum possible resources but as an engineer we need to assess the possibility/feasibility, making wrong decision will impact the existing productivity too.

      Management will encourage you upto some level and if result got negative you have to face the rage of management.

      Yah, as like you mentioned we can go upto 80% occupancy and sometimes i've gone upto even 96% occupancy also based on the available nominal volume. But remember one thing, increasing the occupancy shouldn't impact the mixing characteristics, and they should remain same.

      To avoid any serious consequences which might arise due to matching RPM, we have use the dynamic similarity, kinematic similarity and matching power per unit volume.

      Any decision without backup might knock us out of the organisation and the engineering certificate wont speak that time.

      If possible take some experiments in laboratory auto reactors and try to simulate with 80% occupancy. if you cant afford auto reactors, then just assess the case of 80% occupancy with practical knowledge i.e., agitators having top side pumping will push the reaction mass on to the top which might increase the load and might effect the heat transfer area.

      And if its a RCI / turbine agitators, then we have to assess the requirement of impeller stages and the gap between the impeller stages.

      Hope you understand and if you can make me the query more detailed, i'll try to help you on this. Pl reach me at pharmacalc823@gmail.com.

      Best Regards,
      AJAY K

      Delete
  15. From above post p/v ofbatch size 90 95 100 comparible that of standard remaining batch sizes are out of bounds means what the standard?

    ReplyDelete
  16. Dear Sir,

    You have consider the same power number but as per my knowledge it varies with Reynold number and Reynold number is changing dur change in agitator diameter....Then why u consider the power number same i.e. 5

    ReplyDelete
    Replies
    1. Hii Nilesh,

      You are right, Power number varies with Reynolds number. But in the case which i've considered the variation is low.

      Delete