Ref: 110615_02To: Alex WoltornistFrom: Maxine Ankora, Fletcher Horowitz, Colton Mitchell, Lisa GouldingDate: October 9, 2015Subject: Project Design Memo II

Objective: To design a reactor plant to purify crude lactic acid to polymer grade lactic acid while considering the effect of feedstock reactor yields, and design choices on plant economics and environmental impact.

Abstract: This memo proposes a base case process flow diagram for a plant that purifies lactic acid. Initially, the design does not reach product specifications. The base case was designed solely for the production of LA, without taking into account the efficiency of the reactors, separation units, or economics. Modifications to the plant were considered to make the plant more efficient and cost effective and will continue to be iterated upon. For example, recycle streams for methanol were added to the block flow diagram and the final three distillation columns were re-tuned, allowing for the removal of one of the columns in addition to improved product flow rate. The cost of the plant was analyzed, and will almost certainly decrease as the operation design improves. The Quality Lab Control layout and labor flow were proposed. The labor flow is presented in a block flow with even work division amongst two people for eight hour shifts.

Attachments:I. Unit Operations Summaries

Base Case SimulationIn the base case simulation of the LA purification plant, the goal was to get the

simulation to converge and meet specifications. The simulation was run and can be seen in Figure 1. The stream summaries are shown in Table 1.

The crude lactic acid (LA), containing 57.1 wt% LA, 38.1 wt% water, and 4.8 wt% acetic acid (AA), and methanol solution, containing 99.6 wt% methanol and 0.4 wt% water, are entering the reactor heated by a heater to 120C. The two streams are being mixed prior to entering a PFR. A PFR was used for the first reactor because we first tried a reactive distillation column and there was not a high enough conversion. In a reactive distillation column, the mixture reacts and then separates. Perhaps the reactant components were being distilled before they had a chance to react. With a PFR there is no evaporation. It maintained contact between the reactants. The PFR is operating adiabatically. It has 23 tubes of length 50m and diameter 0.5m. The unit operation summary for the PFR can be seen in Table 2 in Appendix 1. The reaction happening in the PFR is the esterification of LA and methanol into methyl lactate and water. The kinetics for this reaction were found from Troupe’s paper, “The kinetics of methanol-lactic acid reaction”. The following equation was used to calculate the reaction constant, k, with a reference temperature of 100.

log 104 k=10.91369−3193.3/TThe activation energy was calculated using the Arrhenius equation. The pressure at the

reactor inlet was set to 5 atm. The PFR is loaded with 500000 grams of catalyst with an assumed bed voidage of 0.7. The stream tables shown in Table __, shows that the initial crude LA reacts in the PFR 83%. The stream entering the PFR, FD-MIX, contains 0.296 kmol/hr of LA, 1.185 kmol/hr methanol, 0.997 kmol/hr water, and 0.047 kmol/hr methyl lactate. The stream leaving the PFR, PFR-EFF, contains 0.05 kmol/hr of LA, 0.246 kmol/hr methyl lactate, 0.938 kmol/hr methanol, and 1.243 kmol/hr water. The lactic acid reacted with the methanol to produce methyl lactate, ML, and water. The residence time in the PFR is 62.186 days, as seen in Table__. The effluent from the PFR is sent through a RADFRAC distillation column, called METHDIS in the PFD.

The METHDIS column was used to separate out remaining methanol in the stream. The column was designed to have 50 stages, has a total condenser, a kettle reboiler, a reflux ratio of 1, and a specified distillate rate of 1 kmol/hr. the heat duty to the reboiler is 0.01633 Gcal/hr and the condenser heat duty is -0.01558 Gcal/hr. The complete summary of the unit operations for METHDIS can be seen in Table 3 in Appendix 1. The stream exiting the top of the column, METHREC, mainly contains methanol, 0.935 kmol/hr, and some water, 0.065 kmol/hr. The stream exiting the bottoms of the column, BOTTOMS, contains 1.179 kmol/hr water, 0.05 kmol/hr LA, 0.003 kmol/hr methanol, 0.037 kmol/hr AA. The BOTTOMS stream was heated through HEATER1 to 138C and then enters a Distl distillation column, DIST1.

DIST1 was designed to have 30 stages where the feed stage was above stage 10. The reflux ratio was set to 0.5 and the distillate to feed mole ratio was 0.5. The condenser pressure and reboiler pressures were both set to 2 bar. The condenser heat duty was 0.010785 Gcal/hr and

the reboiler heat duty was 0.010398 Gcal/hr. The complete summary of the unit operations for DIST1 can be seen in Table 4 in Appendix 1. The tops of this column come off as WASTE1, containing 0.432 kmol/hr water, 0.013 kmol/hr AA, and 0.009 ML. This gets rid of a lot of excess water. The goal of DIST1 and DIST2 were to separate out the water so that the AA could also be separated out. There was some issues with the separation of AA because it was so dilute with water. The bottoms stream of DIST1, called MLAC in the PFD, contains 0.432 kmol/hr water, 0.05 kmol/hr LA, 0.03 kmol/hr AA, and 0.246 kmol/hr ML. The bottoms is then fed to DIST2.

DIST2 is a Distl distillation column which, like DIST1, separates more water and AA. This column was designed with 30 stages and the feed enters above stage 15. The reflux ratio was set 10 0.6 and the distillate to feed mole ratio was set to 0.6. The condenser pressure was set to 2 bar and the reboiler pressure was set to 10 bar. The condenser heat duty was 0.006923 Gcal/hr and the heat duty for the reboiler was 0.008945 Gcal/hr. The complete summary for the unit operation of DIST2 can be seen in Table 5 in Appendix 1. The stream entering DIST2, MLAC, enters at 133.6 C. This was deemed okay because the boiling point of AA is 118.1 C, and the boiling point of water is 100C, while the boiling point of methyl lactate is 144 C. The tops of the column, WASTE2, contains 0.746 kmol/hr water, 0.003 kmol/hr methanol, 0.008 kmol/hr AA, and 0.001 kmol/hr ML. Now, all the methanol and water are out of the process stream. The stream leaving the bottom of the column, METLACP, contains only 0.05 kmol/hr LA, 0.016 kmol/hr AA, and 0.237 kmol/hr ML. This stream is then heated by HEATER2 and sent into the hydrolysis reactor.

The second reaction was found using the equilibrium constants found in the Troup paper at a reference temperature of 100. The following equation was used to calculate it. Again the Arrhenius equation was used to calculate the activation energy.

K=k1/k−1

Where K = 2.69 at 100C.The hydrolysis reactor was chosen to be an RStoic model, called HYDROLY. An RStoic

model was used for the second reactor because all that is needed is a material balance, which was completed in the first memo, and a conversion, which we could determine from the specifications for the desired product. The reactor had an operating temperature of 300 K and pressure of 1 atm. The outlet temperature of the fluid was 26.85 C. The heat duty of the reactor was -0.00626539 Gcal/hr. The complete summary for the unit operation of HYDROLY can be seen in Table 6 in Appendix 1. Water was also added to the reactor to drive the reaction forward to produce LA and methanol. The exit stream from the reactor, PRODUCT, contained 1.198 kmol/hr water, 0.239 kmol/hr LA, 0.189 kmol/hr methanol, 0.016 kmol/hr AA, and 0.047 kmol/hr ML. The PRODUCT stream then entered DIST3.

DIST3 was a RadFrac model distillation column. DIST3, 4, and 5 are all trying to further purify the final product of LA. DIST3 was designed with 50 stages, a total condenser, a kettle reboiler, a distillate to feed ratio of 0.7, and a reflux ratio of 0.5. The reboiler heat duty was 0.02079 Gcal/hr. The complete summary for the unit operation of DIST3 can be seen in Table 7

in Appendix 1. In DIST 3, the tops stream, WASTE3, takes out most of the water. The WASTE3 stream contains 0.991 kmol/hr water, 0.189 kmol/hr methanol, and 0.0033 kmol/hr AA. The bottoms stream from DIST3, LA-ML, contains 0.207 kmol/hr water, 0.239 kmol/hr LA, 0.013 kmol/hr AA, and 0.047 kmol/hr ML. LA-ML then passes through a heater to cool it from 124C to 105C so that the rest of the water would leave the next distillation column, DIST4.

DIST4 is a Distl model distillation column. It was designed with 45 stages and the feed stage coming in above stage 13. The reflux ratio was set to 1 and the distillate to feed mole ratio. the condenser pressure was set to 1 bar and the reboiler pressure was set to 6 bar. The condenser heat duty was 0.00498037 Gcal/hr and the reboiler heat duty was 0.007734 Gcal/hr. The complete summary for the unit operation of DIST4 can be seen in Table 8 in Appendix 1. The tops of DIST4, WASTE4, took out the rest of the water. WASTE4 contained 0.207 kmol/hr water, 0.13 kmol/hr AA, and 0.033 kmol/hr ML. This leaves 0.296 kmol/hr LA, and 0.037 kmol/hr ML in the bottoms stream, LA-ML2.

Now all that needs to be done is the separation of LA and ML. This is pretty easy because the boiling point of LA is 119 C and the boiling point of ML is 144 C. LA-ML2 is passed through a HEATER4 to 123C and then sent to DIST5 which is also a Distl model distillation column. DIST5 was designed with 30 stages with the feed stream coming in above the 12 stage. the reflux ratio was set to 2 and the distillate to feed mole ratio was set to 0.5. The condenser pressure was set to 2 bar and the reboiler pressure was set to 6 bar. The condenser heat duty was 0.005311 Gcal/hr and the reboiler heat duty was 0.0073995 Gcal/hr. The complete summary for the unit operation of DIST5 can be seen in Table 9 in Appendix 1. The tops, WASTE5 contained 0.112 kmol/hr LA and 0.014 kmol/hr ML. The bottoms stream, PURELA, contained only LA, 0.127 kmol/hr.

Proposal for Plant Modifications

Possible Future Events Affecting the ResultsThere is always a risk of natural disasters happening anywhere in the world that a plant is

set up. It could be an earthquake, a hurricane, tornado, etc. Any of these things could damage the plant physically. Flooding, fires, structural damage, or damage to equipment are all possibilities. Natural disasters cannot be avoided, but, based on the location the buildings could be enhanced to combat the most common natural disasters in the area.

Equipment failure can lead to decreased production of purified lactic acid. Fouling of piping or a failure of any of the unit operations will result in a shutdown of the plant for maintenance which would delay production and cause a decrease in product per year.

Labor strikes can cause major disruptions to industry, commerce and the lives of many people who aren't even connected to the strike itself. This is because if workers strike, production stops and customers of the company are no longer getting product. To prevent strikes from happening, our workers would be paid fairly and management will listen to any issues they have and try to resolve them expeditiously.

A paradigm shift that leads to a higher demand of biodegradable plastics would cause a need for increased production of lactic acid which may require our plant to become larger. A paradigm shift away from products that require lactic acid would cause our plant to go out of business if we are unable to convert it to something still useful.

If something happens to the providers of methanol or the crude lactic acid it could cause a change in the operating costs of the plant. If one of the providers runs out of business then the price of the raw material will increase because there is one less provider for the same amount of demand. On the other hand, if a new provider comes to market, it could cause the raw materials to be cheaper.

If there is a regulation change for any of the chemicals used in the process it could cause our plan to fall below environmental standards requiring the potential addition of new equipment or better equipment.

Recommendations Regarding Cost, Efficiency, Energy, and EnvironmentThe tuning of the plant and the addition of recycle streams to our process will increase

the efficiency of the plant as well as save money in resources and equipment. Currently, we have minimal pre-heating or cooling of streams, for the next block we will look to not only add recycle streams but also optimize the conditions in all reactors and separators. This will require researching the optimal conditions for each separation. If the equipment can be made to work more efficiently, one or more distillations columns can be removed as well as the length of our PFR, which is currently 55 meters, to be shortened. For example, in the base case system, nearly 47% of the product lactic acid is in a waste stream (WASTE5).

The plant must also be sized correctly as our current base case only produces 100,000 kg/year of product. However, much of this can be compensated with better-tuned reactors and separation systems. With some of the proposed modifications, such as a decrease in the feed to

distillate ratio of the final distillation column, the lactic acid product flow rate can increase from 100,000 kg/year to approximately 200,000 kg/year at high purity without changing the feed stream flow rates. These modifications also remove a superfluous distillation column, which will reduce capital and maintenance costs.

The next modifications to consider are a fix for the 62-day residence time PFR, more thorough recycle streams, and further tuning of the reactors and columns. These modifications are in that order of priority.

Proposal for Quality Control LabThe layout for the Quality Control lab is shown in Figure 2. The equipment was already

available to test the samples, since this lab already tests a similar product. The equipment just needs to be placed, turned on and staffed appropriately. The Lab Information Management System (LIMS) is placed in between the two HPLC’s because both scientists are using the LIMS in between using the HPLC’s. The sample dilution area is situated so that both staff member one and staff member two can be dilute samples at the same time if they need to.

The lab will need to be staffed by two people on each of the three eight hour shifts. The staffing was decided based on the amount of tasks needed to be completed and the amount of time in one shift of work. The lab runs 24/7. One shift is eight hours with two ten minute breaks and one 45 minute lunch break. Taking out breaks, the actual amount of time spent working in a single shift is 415 minutes. In a single day of three shifts, 1245 minutes are spent working. The calculated values for time per task and variability are shown in Table 10. The breakdown of tasks for each person is shown in the block flow diagram in Figure 3. The labor flow for each person is also shown in Figure 2.

References

Bauer, James E., Mario L. Occelli, et al. “Heterogeneous catalyst structure and function: review

and implication for the analysis of dissolved organic carbon and nitrogen in natural waters.”

Marine Chemistry, 41 (1993) 75-89. Online: Accessed October 7, 2015.

https://aquaticbiogeochem.osu.edu/sites/aquaticbiogeochem.osu.edu/files/1993%20Bauer

%20Occelli_Marine%20Chem.pdf.

Clark, James H. and Jeffery J.E. Hardy. “Towards Sustainable Chemical Manufacturing:

Polylactic Acid-A Sustainable Polymer?”. Sustainable Development in Practice: Case Studies

for Engineers and Scientists. Ed. by Adisa Azapagic, Slobodan Perdan, and Roland Clift. West

Sussex, England: John Wiley&Sons, Ltd, 2004. Online: Accessed October 7, 2015.

http://www.researchgate.net/publication/228017202_Towards_Sustainable_Chemical_Manufact

uring_Polylactic_Acid__A_Sustainable_Polymer.

FDA. "Enforcement Report." US FDA. US FDA, 5 July 2012. Web. 7 Oct. 2015.

FDA. "Questions and Answers on Current Good Manufacturing Practices (CGMP) for Drugs."

US FDA. US FDA, 5 July 2012. Web. 7 Oct. 2015.

Gaffar, Tayyba, Muhammad Irshad, et al. “Recent trends in lactic acid biotechnology: A brief

review on production to purification”. Journal of Radiation Research and Applied Sciences, 7,

p.222-229. 2014. Online: Accessed October 7, 2015.

http://ac.els-cdn.com/S1687850714000314/1-s2.0-S1687850714000314-main.pdf?

_tid=98323a32-6d1a-11e5-8dd4-

00000aab0f02&acdnat=1444239849_5b2e7042cdb1809c736de8b6fe8d1f95.

“Material Safety Data Sheet: Acetic Acid.” Science Lab. Accessed October 8, 2015.

https://www.sciencelab.com/msds.php?msdsId=9922769.

“Material Safety Data Sheet: Methyl Alcohol.” Science Lab. Accessed October 8, 2015

http://www.sciencelab.com/msds.php?msdsId=9927227

“Material Safety Data Sheet.” Sigma Aldrich. Accessed October 8, 2015.

http://www.sigmaaldrich.com/MSDS/MSDS/DisplayMSDSPage.do?

country=US&language=en&productNumber=230340&brand=ALDRICH&PageToGoToURL=h

ttp%3A%2F%2Fwww.sigmaaldrich.com%2Fcatalog%2Fproduct%2Faldrich

%2F230340%3Flang%3Den.

“Material Safety Data Sheet: Sulfuric Acid.” Science Lab. Accessed October 8, 2015.

https://www.sciencelab.com/msds.php?msdsId=9925146.

“Material Safety Data Sheet: Water.” Science Lab. Accessed October 8, 2015.

https://www.sciencelab.com/msds.php?msdsId=9927321.

“Methanol.” National Pollutant Inventory. http://www.npi.gov.au/resource/methanol

Randhawa, A., & Akram, K. (2012). Optimization Of lactic acid from a cheap raw material:

sugar cane molasses. Pakistan Journal of Botany, pp. 333-338.

Sanz, Maria Teresa, et al. "Kinetic study for the reactive system of lactic acid esterification with

methanol: Methyl lactate hydrolysis reaction." Industrial & engineering chemistry research 43.9

(2004): 2049-2053.

“Sulfuric Acid.” National Pollutant Inventory. http://www.npi.gov.au/resource/sulfuric-acid

Safety Data Sheet. Fisher Scientific. Accessed October 8, 2015.

https://www.fishersci.com/shop/msdsproxy?productName=BP26615&productDescription=L-

(%2B)-LACTIC+ACID+FREE+5G&catNo=BP2661-

5&vendorId=VN00033897&storeId=10652.

Troupe, Ralph A., and Kenneth A. Kobe. "Kinetics of Methanol–Lactic Acid Reaction-Reactions

with 85% Acid." Industrial & Engineering Chemistry 42.5 (1950): 801-810.bm

UN. (1987). “Report of the World Commission on Environment and Development: Our

Common Future.” World Commission on Environment and Development.