In this article, a refiner needed a solution to recover export-oriented naphtha contaminated by methanol (MeOH). This is an actual case in which one of three 30,000-m3 capacity tanks used to store naphtha export was found to be severely contaminated with MeOH. A water-washing solution was applied to reprocess the naphtha in-situ and to remove MeOH from the naphtha, along with returning the storage tank back to continuous operations.
Problem-solving approach. The approach to solving problems such as this type is threefold. First, understand the root cause of the incident and rectify it. Second, develop a theoretical basis to resolve the problem and validate it. Third, evaluate and implement a practical solution.
Naphtha contamination with oxygenates, such as MeOH, is a costly problem for any refinery. Reprocessing or re-blending naphtha is a risky proposition, especially when only limited storage is available. A creative and scientific water-washing solution was identified to remove oxygenates from the naphtha. Understanding the chemistry of the problem is only the first step. Substantial laboratory and engineering work was necessary to successfully identify and to validate the solution.
PROBLEM: MeOH CONTAMINATION IN NAPHTHA TANK
This Middle East refinery exports straight-run naphtha (SRN) from the crude distillation unit (CDU). There are three storage tanks available to store, blend and certify the naphtha prior to shipping. The refinery also operates a tertiary amyl methyl ether unit that uses imported MeOH as a feedstock.
Following a routine MeOH unloading at the refinery, 25,000 m3 of SRN product in an export 30,000-m3 storage tank was later found to be contaminated. This naphtha failed a product-certification test. Contamination results were further confirmed by a third-party laboratory.
Test results of the SRN showed an oxygenate content of 240 ppm against the maximum acceptable level of 50 ppm to obtain a quality certificate. Unfortunately, the SRN of this tank could not be exported. Now, there were only two naphtha tanks left in operation. The quantity of material and oxygenates prevented meaningful re-blending and reprocessing. Contamination of the other two tanks remained a real and immediate threat to plant operations. During the incident, naphtha rundown from the process plant was continuously showing acceptable oxygenate content that was less than 50 ppm.
| Fig. 1. Water is a polar molecule with |
positive charges on one side and negative
charges on the other.1
Step 1: Identify and rectify root cause.
The investigating team successfully determined the root cause for the MeOH contamination. The investigation revealed that both the meters installed on the line to the jetty for naphtha export and the meter on the MeOH import/unloading line share the same meter prover. A single cross valve on the meter prover was mistakenly left open. Pressure differential allowed imported MeOH from the vessel to flow unimpeded to the naphtha-product line from the CDU to the storage tank during unloading of the cargo vessel.
Rectification of the problem involved strict new operating procedures for the meter and meter prover. Operator training was improved, and it focused on careful handling of equipment with checklists and verification by foremen. For the long-term solution, a separate meter and meter prover would be installed.
Different ways of disposing the off-spec naphtha were studied. However, disposal could impose higher financial losses to the company due to contamination of the naphtha lines to the jetty. Moving the material to other tanks had a higher inherent risk of cross-contamination for the remaining product-storage tanks. At that time, no buyer could be found to purchase this batch of off-spec naphtha. Unfortunately, the MeOH-contaminated naphtha remained in the tank for several months as various options were considered.
Step 2: Theory and validation.
The rule for determining if a mixture becomes a solution is that polar molecules will mix to form solutions and nonpolar molecules will form solutions, but a polar and nonpolar combination will not form a solution. Both MeOH and water are polar. So extraction of MeOH in an aqueous solution is a feasible pathway. The geometry of the atoms in polar molecules is such that one end of the molecule has a positive electrical charge and the other side has a negative charge. Nonpolar molecules do not have charges at their ends. Mixing molecules of the same polarity usually results in the molecules forming a solution.
Low-molecular-weight alcohols, such as MeOH, are completely soluble in water. Because of their polar structure, the alcohol molecules actively associate with water molecules through the hydrogen bonds. The hydrogen bonds are strong enough to prevent separation of the water/alcohol mixture by distillation, as shown in Fig. 2.2
| Fig. 2. MeOH hydrogen bonds and polarity.2 |
Various molecules may mix and dissolve in each other if they have approximately the same polarity. In the case of water and MeOH, this is the situation. The hydrogen of the OH group on the alcohol is polar in the same manners as the water molecule.
| Fig. 3. Lab results of water washing |
of contaminated SRN.
Solubility of MeOH in naphtha.
In terms of polarity, MeOH is a strong polar molecule, and aromatics, such as toluene, are slightly polar. Paraffins, such as hexane, are nonpolar. Aromatics will be temporarily polarized within the vicinity of a polar molecule (MeOH), and the induced and permanent dipoles will be mutually attracted (Debye Interactions). However, MeOH is not completely soluble in streams, such as SRN that contain low levels of aromatic compounds. Paraffinic/naphthenic hydrocarbons (HCs) comprise 90 wt% of the SRN, and the remaining 10% are aromatic HCs. Therefore, the MeOH and naphtha are not soluble in any large ratios.
SRN, depending on the crude type processed, normally contains 8 wt%10 wt% of aromatics. MeOH solubility in aromatics is temperature dependent. Essentially above 0°C, for every percentage of aromatics present, 0.5% of MeOH will be soluble. Following this rule, it is expected that the SRN can dissolve up to a maximum of 4 wt%5 wt% MeOH.
Laboratory testing was proposed and arranged. Test samples with different water concentrations were added to known volumes of the off-spec naphtha0% water content in naphtha was the control sample with 1%, 5%, 10% and 20% water concentration standards tested. To investigate the effect of thorough mixing, the samples were analyzed with and without a magnetic stirrer used. Table 1 summarizes the lab results.
Another set of tests was done on the samples from the contaminated tank to measure the effect of water washing at different vol% of water to remove the various oxygenates from the contaminated naphtha. Test results show that water washing removed the majority of the MeOH content from the naphtha while other oxygenates were not affected. Table 2 lists these test results at different water-wash volumes with and without mixing. Fig. 4 shows the appearance of the SRN after water washing at different vol% of water with a one-hour settlement time and the settled water drained from the sample. These tests showed that there was not much difference in haziness of the naphtha when different volumes of wash water were used. The lab report can be summarized as:
MeOH and total oxygenate content decrease dramatically to within specs (50-wppm maximum) when the contaminated naphtha was water-washed with subsequent mixing (by a magnetic stirrer similar to actual tank mixing). The MeOH content remained high when mixing was not done.
There was no change in color and the product was not hazy.
There is only a slight increase in water content after the sample remains stagnant if water is not drained.
| Fig. 4. Haziness of treated naphtha after |
water washing with different water volumes.
The tests also confirmed the understanding that, if water washing is done together with mixing, MeOH removal would be more efficient. Based on these results and the lab report, it was also decided that the contaminated naphtha should be washed with demineralized (DM) water.
Lab results showed that 10% water addition to the naphtha with mixing would reduce the MeOH content from 190 ppm to 1.4 ppm, while 1% water can reduce it from 190 ppm to 22.7 ppm. Subsequently, it was decided to inject only 1% of water wash and to drain after mixing, and repeat several times, until the total oxygenate concentration dropped to less than 50 ppm. Using this method, less DM water would be used, thus, limiting cleanup costs and time to recover the product naphtha.
| Fig. 5. Process scheme for Option 1: |
Direct water injection to tank.
Step 3: Effective implementation.
Now that a lab-scale solution was available, the emphasis shifted to execution. Several ideas were considered, with three of the most viable choices listed here:
Option 1. Direct water injection to tank. Water can be pumped directly into the tank T6217C. After injection, the SRN could be mixed with the aid of an available tank mixer. Advantages of this process were:
Water can be introduced through a larger nozzle (4-in. size).
The associated lines would not be contaminated.
The procedure can be done several times. In case of failure, the other two naphtha tanks would remain available for rundown and dispatch.
The disadvantages included:
Mixing will require a longer time.
Mixing may not be as effective as circulating the SRN to and from the tank.
| Fig. 6. Tank mixing patterns for Option 1: |
A and B.
The tentative time required for each cycle of water, assuming 1% DW will be mixed to the naphtha tank and drained after mixing and settlement, are summarized Tables 3 and 4.
Option 2. Water injection via export piping. Naphtha inventory of the tank can be circulated by way of marine-loading pump. Water can be put into the suction line of the pump0.75-in. nozzle with two nozzles with a capacity of 3 m3/hr per nozzle. The resulting mixing could be done by the pump itself and would not rely on the effectiveness of the tank mixer. The advantages of this option include:
There is thorough mixing of SRN and water
The mixing time will be shorter
The experiment can be carried out several times. In case of failure, the other two tanks will be available for rundown and dispatch.
However, the disadvantages are:
Associated pipelines will have to be flushed thoroughly with on-spec naphtha
Limitations would have to be imposed on the scheduling of naphtha shipments.
| Fig. 7. Processing scheme for |
Option 2: Water injection via export piping.
Option 3. Water injection and mixing using remaining tanks. Water can be sent to one of the other tanks (T6217 A/B. The T6217C can be transferred to it. The advantages from this option include:
There is thorough mixing of SRN
The mixing time will be less, as the SRN can mix while it is filling the tank.
Conversely, the disadvantages are:
Only one tank will be available for operation.
If the procedure fails for any reason, then the additional tank also contains contaminated naphtha.
Associated pipelines will have to be flushed thoroughly with on-spec naphtha
Naphtha shipment schedules would be affected.
Option 1 was selected as the preferred method. As per the plan, 1% DW or 250 m3 of DW would be injected directly to the tank. The tank mixer would be used to mix the SRN and DW, followed by tank settling and draining of settled water. This procedure would be repeated as required until the naphtha is completely washed and meets all oxygenate specifications.
Successful water-washing plan.
Water injection to the tank started on Feb. 26 during the day shift. Table 5 shows the result of oxygenates, MeOH and chlorine content of the naphtha before and after the water washing. Water draining started right after nine hours of settling. After the first water wash, the oxygenate level dropped to 60 ppm, close to spec, from the average result of 240 ppm. Therefore, after the water was drained, a second water-wash operation started on March 4, after which the total oxygenates dropped to 40 ppm; both were acceptable and on-spec. Fig. 8 shows how the oxygenate level changed with water washing. Table 6 summarizes the moisture content of the SRN before and after the water-washing operations. The SRN then received a quality certificate and it was successfully exported. The refinery continues to successfully operate with all three naphtha tanks in service, and with no further incidents of MeOH contamination. HP
| Fig. 8. Oxygenate content changes over |
time with water washing.
|The author |
||Farzad Ovaici received his MSc degree in chemical engineering from Shiraz University in 1978. In 1979, he began his career with Bandar Imam Petrochemical Co. in Iran. In 1980, he moved to the Isfahan refinery. Mr. Ovaici was later responsible for reconstruction and rehabilitation of Abadan refinery, a 630,000-bpd refinery. This refinery severely damaged due to the Iran-Iraq War. In 1992, he joined Tabriz Petrochemical Co., and was assigned as project director for EB/SM, and different polystyrene plants. Later, he was assigned as chairman and managing director of Tabriz Petrochemical Co. In 2000, he became the managing director of Kala Naft Canada Ltd. Mr. Ovaici received an M. Sc. degree in engineering from the Chemical and Petroleum Engineering School of University of Calgary. He is a member of the Association of Professional Engineers Geologists and Geophysicist of Alberta Canada. In 2005, he moved from Canada to Oman and joined Oman Refinery Co. as the general manager, of the Mina Al-Fahal Refinery. Later, he was promoted to general manager of the two refineries in Oman Refineries and Petrochemical Co. Mr. Ovaici joined Al-Ghurair Energy as the managing director, of refining and petrochemicals and is based in Dubai, UAE. In addition to his position in Al-Ghurair Energy, Mr. Ovaici is currently chief executive officer of Libyan Emirates Oil Refining Co. |