With increasing supplies of gas-based petrochemical feedstocks coming on stream in the Middle East, now is a good time to consider reconfiguring refineries to boost production of chemicals for export to Asia
In the Middle East gas field development is continuing at a rapid pace, recording an average annual growth rate of about 9% between 2000 and 2011, according to the International Energy Agency. Qatar and Iran represent almost 60% of the total regional supply. As ethane and liquefied petroleum gas (LPG) are good feedstocks for petrochemical plants as well as condensates for refineries, the recovery of liquids from natural gas is an added value.
It is worth noting that a typical regional condensate such as North Field Condensate (NFC) – being light (°API 57.95), sweet (0.23% sulfur content), and low in metals content – is an excellent raw material for a hydroskimming refinery. In addition, growing availability of feedstocks and low-priced natural gas provide competitive advantages for additional capacities in the Middle East.
Strong growth of petrochemical and oil demand in Asia, driven mainly by increasing population and high economic performance, is making this area an attractive market for major players. According to the latest release of the ICIS database (2013), oil consumption is expected to increase at a higher rate than supply, resulting in a growing deficit over the next few years.
In the South Asia and Pacific (SAP) region a shortage of road diesel is expected to reach over 50m tonnes by 2025 while gasoline should achieve almost 20m tonnes as well as LPG. Finally, kerosene demand should have a minor impact on the regional deficit if it is compared with the other oil products (see SAP balance chart on page 29).
Based on 2012 statistical trade data, northeast Asia shows a deficit of about 950,000 tonnes of mixed xylenes while maintaining a surplus of benzene (577,000 tonnes) and toluene (almost 260,000 tonnes). The SAP region records a shortage of 165,000 tonnes of toluene, a surplus of benzene (753,000 tonnes) and mixed xylenes (almost 70,000 tonnes). In addition, the spread between benzene, toluene, xylene (BTX) demand and supply should grow over the next few years unless additional investments are made in petrochemical capacity.
Therefore, in this context, it is interesting to analyse a hypothetical hydroskimming refinery located in the Middle East and export-oriented towards the Asian market. For this type of plant, reformate represents the key interface between the refinery and petrochemical plant, as it is the most important component for gasoline blending as well as a precious feedstock for BTX production.
Therefore, two cases have been taken into consideration. The first (a non-integrated plant) is a simple refinery equipped with a condensate splitter (110,000 bbl/day of capacity), a naphtha splitter (66,000 bbl/day), an isomerisation unit (26,000 bbl/day), a continuous catalytic reforming unit (40,000 bbl/day) plus the necessary hydro-desulfurisation processes.
The second (an integrated plant) includes a simple petrochemical complex (32,000 bbl/day) that produces benzene, toluene and mixed xylenes (see chart on page 27).
Process integration, in terms of main streams, is based on recovery of aromatics from reformate through the BTX complex and backflows (raffinate and C9+) blending, after extraction, into the gasoline pool.
The main drivers are, on one side, feedstocks costs and, on the other, product prices as well as the quality of the products required from the reference market.
Condensate average prices have been indexed to Dubai crude oil while oil products are referenced mainly on the Singapore market. Price quotations for gas oil (0.05% sulfur content) are not available therefore it has been evaluated starting from gas oil (0.5%) FOB Singapore adding the differential price taking into account the sulphur content difference into gas oil products available on the international market. What are the fundamental differences between an integrated and non-integrated site? What is the gross refining margin for each plant? What are the benefits and disadvantages of integration?
Typically, a linear programming (LP) model is used for solving the issues above in order to maximise refinery profit. Through this optimiser tool, it is possible to select the optimal refinery operations in terms of output and, consequently, revenues.
Running an LP model, the answer is that the refinery handles about 11,980 tonnes/day of condensate providing globally about 11,900 tonnes/day of products with a recovery of 99.4%. The properties of feedstock and products are summarised in the table on page 26.
As percentage yields to weight of condensate are about 58% of the total for light distillates and 36% for middle distillates, NFC results in a suitable feedstock in transportation fuel products.
In the table on page 29 we have reported daily feedstock processing and its cost, plus the product slate that optimises the refinery operation and the value of each product. In terms of the full range of naphtha processing, about 36 wt% of light naphtha goes to isomerisation or to market while 64 wt% of heavy naphtha is for reforming capacity.
According to the simulation, the isomerisation process required to increase low octane of C5/C6 normal paraffins, produces almost 2,400 tonnes/day of isomerate for the non-integrated configuration and less than 250 tonnes/day for the integrated model. A 3,100 tonne/day surplus of naphtha is available.
Total available volumes of distillates remain unchanged for both schemes while the amount of gasoline shows a decrease from about 6,200 tonnes/day to 700 tonnes/day between non-integrated and integrated plants because 94% of total reformate availability is addressed to BTX manufacturing.
In addition, analysing NFC assay, N+2A of heavy naphtha (called feed index) results are high (60.7), making it a good feedstock for reformer unit.
Reformate production in the two cases is another interesting issue. The LP model suggests two different types of reformate for the non-integrated plant: RON 102 and RON 90. There are two opposite drivers. In the non-integrated site the reformer needs to operate more intensively for octane boosting, therefore losing some volumes of reformate (total production: 3,720 tonnes/day). By contrast, in an integrated plant it is necessary to decrease the operations of reformer severity by maximizing the availability of reformate for the BTX plant. This operating condition causes a slight increase of total reformate volumes (3,894 tonnes/day) compared with the first case.
Again LPG (30% propane and 70% butanes) production is, by and large, lower for a plant that incorporates a BTX complex because the lower reformer intensity causes a minor production of gaseous streams.
The share of road diesel corresponds to about 23% by weight of the total product supply and is constituted mainly by hydro-treated gas oil and hydro-treated heavy gas oil streams that account for almost 83% and 10% of total blending volume, respectively.
To meet the required specification of sulfur content (0.05% maximum), the desulfurisation of heavy gas oil (0.413% sulfur) is necessary before the blending process.
Therefore, the refinery sees zero fuel oil production. The increasing demand for high -value cleaner transportation fuels and the decline of low-value heavy residue market represent a competitive advantage for the product slate released by this type of refinery.
Standard sulfur specification for jet kerosene requires a maximum content of 0.3 wt%. As the kerosene cut has 0.23 wt% of sulfur, the refinery configuration is not equipped with a kerosene hydro treating unit.
Hydrogen sulfide coming from different refinery gas streams, after an amine treatment unit, is converted in elemental sulfur by a conventional Claus process. The small quantity of sulfur recovered, about 20 tonnes/day, can be shipped to Asia as feedstock for fertilizer manufacturing.
The petrochemical complex produces about 2,600 tonnes/day of BTX, 822 tonnes/day of raffinate and 229 tonnes/day of C9+, and accounts for about 22% by weight of total product supply. The LP model evalutates a higher gross refining margin for the integrated site compared with the non-integrated model.
Indeed, according to price scenarios for 2012, as the weighted average of BTX’s premium to gasoline is slightly over $160/tonne, the model calculates, in the first case, a gross margin of $1.214m/day while in the second case $1.023m/day, providing almost 19% of incremental profitability. Finally, gross refining margin, in terms of US$/bbl, is $15.8/bbl for the favourable case while it is $13.3/bbl for the unfavourable. Integration could be deeper if the plant produces only benzene and paraxylenes.
Synergies are evident not only in terms of incremental profitability but also as security of feedstock supply to the petrochemical industry, logistics optimisation, saving on transport costs, reduction in utilities system costs (heat, steam, water, hydrogen and power) and re-processing of by-products (raffinate) to add more valuable products. Again, refining and petrochemical integration increases the level of flexibility and mitigates risks of market volatility.
Indeed, with increasing gasoline demand, the refiner can optimise operations to maximise production of fuels (toluene and mixed xylenes can be also blended into gasoline pool). Conversely, if fuel requirement is decreasing, the value of specific refinery streams can be higher in chemicals production.
Among the main disadvantages of integration, it is interesting to highlight the higher initial investment, increased complexity of the plant, more technical problems, and more complex planning.
However, environmental challenges based on more and more strict constraints in the transportation fuels, low refining margins and growing competitiveness provide excellent reasons for assigning to refinery streams valuable assets for manufacturing products other than conventional transportation fuels. It is apparent that the integration process adds mutual value to operations of refining and petrochemical industry.
Italo Righi contributed to this article
Paolo Scafetta is a chemical engineer and joined Parpinelli TECNON (now ICIS Consulting) in 2001. He is engaged in annual multi-client reports, as well as single-client studies. Contact: email@example.com