Poeb140 Lee
Poeb140 Lee
Poeb140 Lee
-1
For food safety reason, mandatory 2.5 mg kg (ppm) The washed oil exits the POS tank with less than 0.8%
maximum level of 3-MCPDE in refined palm oil are and 0.03%, moisture and impurities respectively. The mill
enforced as followed: vacuum drier is able to reduce the moisture down to below
0.13%. The dried washed oil moisture and impurities
i. 1 January 2021: Compliance to EU member countries’ combined complied to the commercial contractual limit of
requirement. 0.25%.
Retention
tank
Reacts
Mixer
Existing vacuum
drier: Prod. Oil
Mixing stage
In the separation process, apart from the washed oil, water separation. The system operated for six months
the spent washed water (SWW) is continuously discharged before being switched to the POS tank to replace the
by POS tank during operation. This SWW is directed to machine.
the mill’s oil reclaim tank for oil recovery as a last line of
defence in oil loss prevention prior to discharging to the The objective of the POS tank was to reduce
effluent treatment ponds. maintenance cost and power consumption associated with
the machine and to prevent future capital expenditure for
The tank is emptied by vacuum drying the remaining a stand by separator. It was proven to be a success during
oils during its periodic cleaning. The balance of the SWW is the industrial trial.
drained to the oil reclaim tank for recovery and discharge.
The emptied tank is then washed, cleaned, and readied Under the terms of the MPOB matching grant
for the next process operation. The frequency of cleaning memorandum of agreement, the POS tank was patented
depended on the mill’s management practice. In our case, by MPOB and IJMEO. The author designed and invented
the tank is emptied and cleaned every month. the POS tank. The POS tank was commissioned on 22
October 2019.
CPO washing can significantly reduce the inorganic
chloride (IOCL) content in CPO. However, due to the Milling process and sources of CPO
chemical properties of the organic chlorine (OCL), The mill’s design is conventional with 2.5 mt cage
washing is not effective for its removal. The process has horizontal sterilisers, P15 presses, three phase decanters
been demonstrated to effectively remove about 90% of and empty fruit bunch (EFB) presses.
TCL (IOCL) in the mill with the remaining are deemed to
be mostly OCL. During production trial, steriliser condensate and
EFB liquid were processed in the mill clarification
PERFORMANCE MONITORING section together with the crude oil from the presses.
Thus, the washed CPO consisted of oils from
History of the trial these streams.
The CPO washing system was installed in Desa Talisai
under the MPOB’s matching grant program in 2019. The Our own estates operations were well controlled and
CPO washing system came with a separator for CPO and indicated by low FFA of below 3%.
Sampling and Testing of Total Chlorides (TCL) setting the process conditions to deal with high content of
Each sample was a daily hourly composite sample. TCL in CPO feed. It may not be a perfect solution, but it will
In this writeup, the industrial trial period was carried out help to mitigate the challenges.
between October 22, 2019 to February 28, 2021. A total of
357 samples were collected and analysed. TCL removal and TCL in washed oil
During the trial period, the washing process was able
The industrial trial meant for the washing system to to remove on average, 89 % of the TCL. The TCL content
operate as part of the milling process. in washed CPO was below 2 ppm. The washed oil is
transparent and cloudy visually compared to unwashed
Samples were analysed by an independent accredited oil, as shown below (Figure 2).
laboratory in Sandakan. The testing procedure was that
each sample was tested in duplicate for validation. The Cloudy appearance
Cloudy appearance
of the of the CPO
CPO before
average values were reported. The testing methodology before
mixingmixing
was based on ASTM 4929 Procedure B. This said
laboratory is listed in MPOB’s web site, participating
Clear oil after
in MPOB’s inter laboratory cross checking verification washing
Clear oil after washing
program.
During this trial period, the washing process processed Dirt in the CPO
Dirt in the CPO
before mixing
46 293 mt CPO and 213 704 mt FFB. The process before mixing
–1
throughput was 12 mt CPO hr .
This is no dirt in the
This is noprod
washed dirt inoil.
theItwashed
is a
RESULTS AND DISCUSSION prod oil. It is a shadow.
shadow.
The SWW pollutant strength was measured with 5 process time must be taken into consideration in comparing
samples. Results are shown in Table 1. Both BOD and the results between before wash and production oil. Thus,
COD levels were significantly lower than the effluent averaged results were used in this
treatment ponds’ design limits. The SWW pollutant comparison.
strength and quantity (5% to 10% of CPO) did not overload
the effluent treatment ponds. The impact of washing using POS tank on both FFA
and DOBI is neglectable. The FFA content increased by
It was further observed that the oil and grease content 2% while DOBI reduced by 1%.
averaged at 318 ppm, or 0.0318%, indicated that oil loss
in SWW was low. Elimination of purifiers
Prior to CPO washing, the mill uses 2 units of purifier
Phosphorus and Iron removal
and vacuum dryer to achieve the required moisture and
In the course of our trial, apart from chlorides removal, impurity level of 0.25%, in the CPO. An additional benefit
the removal of phosphorous and iron by washing was of POS tank was the elimination of purifiers. This reduced
evaluated. A total of 38 samples were taken and the power, operation cost and future replacement capital
averaged results are presented the Table 2. expenditure.
The washing process removed phosphorous and The average oil moisture and impurity content of
iron from the oil by 33% and 32.5% (after POS tank), washed oil after POS tank, before vacuum drying, in the
respectively. The removed compounds were water soluble. trial were 0.63% and 0.02%, respectively (Table 4). The
The washed oil benefits the refinery though reduction of purifier design required a maximum 1% moisture and 0.1%
bleaching earth consumption. impurity in the CPO feed. The purifiers reduced these to
0.5% (moisture) and 0.02% (impurity) prior to feeding to
Effects of washing on FFA and DOBI the vacuum dryer. The POS tank proved to be able to
The effects of washing on the oil quality were evaluated achieve the same result with 0.02% impurities at purifier
in the trial with results tabulated in Table 3. The delay in outlet.
The vacuum dryer design required the CPO feed with Operation experience and cost
maximum 0.5% moisture. The machine was able to reduce The CPO washing system with POS tank is easy to
it to 0.08%. However, our contractual CPO quality of operate with little training. In comparison with the separator,
moisture and impurity is 0.25%. There was sufficient buffer the POS tank incurred no maintenance cost to date. It
to take higher moisture in the feed with 0.63% moisture. does not require electrical power to achieve CPO and
The drier was able to take high moisture in the washed water separation and does not need daily cleaning unlike
oil and still able to dry it to comply to the contractual limit. the machine. In our six months of separator operation,
There were times the moisture in the feed were greater wear and tear were observed with wear grooves found in
than 0.63% at about 0.8%. the bowl and its components.
The POS tank is able to make the purifier redundant The cost of tank is an advantage as it does not
due to its ability to achieve low moisture and impurity in the breakdown like a mechanical machine or subjected to
washed oil for the vacuum dryer so that it can lower the oil concentrated wear and tear. Furthermore, a standby
moisture to meet the contractual limit. tank is not required, unlike mechanical machine, thereby
minimising capital investment. The tank capital cost is
The mill has stopped using the purifier since the lower than that of the machine. Thus, the design objective
installation of washing plant with POS tank. of the POS tank has been achieved.
Impact of high FFA and low DOBI on TCL removal 1. The average TCL content in CPO feed was 20.07
It is quite unlikely that high FFA and low DOBI have ppm and TCL reduction was 95%. The TCL content in
any impact on TCL removal. Sungai Sabang palm oil mill CPO after POS tank was 0.976 ppm.
No. 1, our second washing plant, which was commissioned 2. The average FFA and DOBI were 6.25% and 2.22%
recently during the rainy and flooding season indicated respectively and TCL removal was achieved.
that TCL could be removed to below 2 ppm, despite the oil 3. The moisture and dirt content in the CPO feed were
containing more than 5% FFA, ranging between 6% to 8%, 1.04 % and 0.02% respectively. The POS tank
and DOBI below 2.3. reduced the moisture content to 0.46 %. This was
good enough to bypass the existing purifiers. After
The second washing plant incorporated with POS tank vacuum drier, the production oil moisture content
In March 2022, the second plant was successfully was 0.19%. Subsequently, the mill has stopped using
commissioned and performed as per design these machines.
specification. The average results are summarised in The phosphorous and iron content were reduced by
Table 5. 25% and 28% respectively.
The average results were meant as indication due to These results are similar to the Desa Tailsai palm oil
limited sampling size. mill’s washing plant performance.
TABLE 5. RESULTS FROM SECOND PLANT AT SUNGAI SABANG PALM OIL MILL NO. 1
10 - 26 March 2022 commissioning results SWW, Spent Wash Water
Day Sampled 14 days Chloride (ppm) 154.94
FFB processed (mt) 7171 pH 4.46
3
Total CPO washed (m ) 1369 TDS 3082.50
CPO washing plant running hours 119 Oil Losses % by solvent (wet basis) 0.28
Production oil
Average TCL (ppm) 0.9757 FFA (%) 6.48 Moisture (%) 0.19 Phosphorus (ppm) 13.88
% reduction in FFA vs % reduction P vs
% CL reduction 95.1% 4% Impurity (%) 0.02 26%
before mixing before mixing
DOBI 2.15 Iron (ppm) 1.37
% reduction in DOBI vs % reduction Fe vs
3% 38%
before mixing before mixing
Is POS tank applicable in refinery? IOCL will convert to OCL and washing cannot remove
The washing process with POS tank system with OCL. If OCL becomes difficult to manage, it is possible
extension can be applied in the palm oil refinery. that refineries may impose a limit of OCL to fortify their
quality control and assurance.
The refinery system will need additional process
systems. Dedicated unwashed oil and washed oil storage ACKNOWLEDGMENT
tanks are needed to support the system. A vacuum drying
system is needed to lower the washed oil moisture level The author wishes to express sincere gratitude to the
to 0.25%. The refinery’s waste water treatment plant following entities and people: MPOB, Director-General
needs to be reviewed for its existing treatment capability to MPOB, 3-MCPDE team: Ms. Rosidah, Dr. Azmil and, Mr.
handle the waste water generated by the washing plant. A Roddy for their ongoing support and opportunities during
washing plant recycling system is needed to return the oil the trial, and patent works, as well as their consent to
in the system back to the feed tank. This process is patent publish the results. The author also would like to thank
pending. you KLK Agri Oils Sdn. Bhd. for their consent to publish
the trial results and IJMEO Sdn Bhd especially Mr. Joseph
CONCLUSIONS AND WAY FORWARD Tek for their supports and believes in the POS tank design
and development. Last and not least, to the management
At the time of writing, the mills and refineries are finalising and staffs of Desa Talisai and Sungai Sabang palm oil
the most suited location to carry out the CPO washing. mills for their commitments and dedication in the plant’s
MPOB’s mandate is in place for the compliance of construction, commissioning, operation and carrying out
–1
3-MCPDE of 2.5 ppm (mg kg ) in refined oil products the trial.
from 1 of January 2023 for both national consumption and
export. Food safety is the paramount importance, followed REFERENCES
by upholding our Malaysian palm oil reputation.
EFSA Journal (2016). Risk for human health related to the
On the world’s stage, Malaysia’s success in achieving presence of 3- and 2-monochloro-propanediol (MCPD),
3-MCPDE quality parameter on refined oils means that and their fatty acid esters, and glycidyl fatty acid esters in
the buying countries will exert pressure on other palm food, DOI: 10.2903/j.efsa.2016.4426
oil producing countries to comply. It will give our country
an edge over others in supplying better quality products. EFSA Journal (2018) Update of the risk assessment on
Perhaps, it may even command monetary premium. 3-monochloropropane diol and its fatty acid esters, DOI:
10.2903/j.efsa.2018.5083
On the local front, the palm oil industry will settle on
the location of CPO washing eventually. Some mills may EU (2020). Official Journal of European Union L 310/2,
choose to wash and some refineries will carry out the Commission Regulation (EU) 2020/1322 of 23 September
washing. Refineries will certainly establish TCL limit in 2020 amending Regulation (EC) No 1881/2006 as regards
CPO just like FFA, M+I, and DOBI, which is likely to be maximum levels of 3-monochloropropanediol (3-MCPD),
as followed: 3-MCPD fatty acid esters and glycidyl fatty acid esters in
certain foods.
1. Washed CPO: 2 ppm of TCL or less.
2. Unwashed CPO: TCL value is yet to be determined. Hamlet, C G; Asuncion, L; Velíˇsek, J; Doleˇzal, M; Zelinkov́,
Z and Crews, C (2011). Formation and occurrence of
For mills who choose not to wash, the unwashed oil TCL esters of 3-chloropropane-1,2-diol (3-CPD) in foods: what
limit will be critical. It will be a fixed limit like 5% FFA and we know and what we assume. European Journal Lipid
below. Assuming that, these mills experience fluctuation of Science and Technology 113 (3), 279–303. https://doi.
like in the trial, penalty may be incurred for failing to meet org/10.1002/ejlt.201000480
the specification if there is no effective means of quality
control. It is advisable for the mills to verify occurrence of Norliza, S; Syed Mohd. Hadi, S H; Ahmadilfitri, M N and
the fluctuation and its magnitude. Ahmad Jaril, A (2019). 3-MCPD ester in palm oil production
chain: Sime Darby’s experience. Palm Oil Engineering Ahmadilfitri, M N and Ahmad Jaril, A (2019). Production
Bulletin No 132, p26-31 of high-quality crude palm oil and low 3-MCPD ester
RBD palm oil, Palm Oil Engineering Bulletin No 131,
Rahmat, N; Syed Mohd Hadi, S H; Norliza, S; Syahril p24-28.
Anuar, M R; Yosri, M S; Mohammed Faisal, M Y;