INDUSTRIAL CHEMISTRY AND ENGINEERING

The Propylene Gap: How Can It Be Filled?

by Jeffrey S. Plotkin

September 14, 2015

Propylene is the second most-produced building block (after ethylene) in the petrochemical industry; its 2014 global demand is estimated at 89 million tonnes. Propylene may play second fiddle to ethylene in market demand, but it leads the orchestra in diversity of derivatives and supply sources. It is the starting point for numerous materials.

Figure 1
Courtesy of IHS

Polypropylene is by far the major propylene derivative (Figure 1). But propylene is also the raw material for producing an array of more specialized products, including the extensive cumene value chain (phenol/acetone/bisphenol A/polycarbonate), epoxies, phenol–formaldehyde resins, propylene oxide, acrylic acid, acrylonitrile, n-butyl alcohol, 2-ethylhexanol, and isopropyl alcohol.

And on the supply side, the variety of propylene production technologies is almost a match for propylene’s derivative list. Historically, the main source of propylene for use in the petrochemical industry was as a coproduct from ethylene-driven steam crackers (see box). The amount of propylene coproduced in a steam cracker, however, depends mostly on the nature of the feedstock.

For example, in a pure-ethane cracker, very little propylene is coproduced. As feedstock trend heavier, (i.e., changing from ethane to propane to butane to naphtha), propylene coproduct yields increase. Since 2008 or so, the availability of low-cost ethane in the United States has brought about two supply challenges for propylene with regard to ethylene crackers:

  • Existing crackers with the flexibility to crack very light feeds are being used to do so and therefore produce less and less propylene.
  • New ethylene steam crackers (some under construction and others recently announced) are all designed to crack only ethane—with little or no capability to coproduce propylene.

Projection: Ethylene up, propylene down

Petroleum Cracking

“Cracking” is one of the most commonly used processes in petroleum refineries. Larger hydrocarbon molecules are broken down, or “cracked”, into smaller ones.

Figure 2
Courtesy of IHS Chemical

An interesting view of this scenario is shown in Figure 2. In this chart, the ratio of ethylene to propylene (E/P) production is indexed from 2000 to 2020 for North American (NAM) steam crackers. From 2000 to 2007, the E/P production ratio was more or less constant. But as inexpensive shale gas–based ethane became available starting in 2008–2009, propylene coproduction relative to ethylene production declined. As new pure-ethane crackers come on stream in 2015 and beyond, it is projected that ethylene production will increase, but propylene production will stay low.

Note, however, that with the recent sharp decline in oil prices, the cost advantage of ethane cracking has diminished;  and cracker operators are likely to opt to crack heavier feeds and produce more propylene than is projected in the figure.

In any case, the quantity of propylene that comes from North American ethylene crackers is significantly reduced, and the industry must turn to other sources to keep up with propylene demand. The US petroleum industry is fortunate to have a large gasoline production capability; with this capacity come large amounts of refinery-grade propylene.

Current options

Refineries use a process called fluidized catalytic cracking (FCC, see box) to crack heavy molecules with long carbon chains that are found in petroleum fractions from vacuum gas oil down to the lighter naphtha that is needed for gasoline production. In the course of this cracking process, some of the molecules are overcracked; and very short-chain molecules, including ≈5% propylene, are produced.

Propylene yields also can be increased by adding specially formulated zeolites to FCC catalysts to make so-called enhanced FCC catalysts. But before refinery-grade propylene makes its way to the petrochemical industry, the refiner must make a decision about its disposition.

In gasoline production, propylene has value in a process known as alkylation. In alkylation, propylene reacts with isobutane to give a highly branched seven–carbon atom paraffinic mixture called alkylate. Because it is highly branched, alkylate has a high octane number and is often used as a blendstock in gasoline formulation. Thus, the refiner must constantly keep an eye on the value of octane compared with the price of chemical- or polymer-grade propylene to determine where to place this refinery-grade propylene.

Making propylene on purpose

Internationally, even if ethylene-cracker coproduct propylene and FCC-derived propylene are combined, propylene supply is still ≈12–14% short of the amount needed to satisfy demand. To fill this supply gap, the industry has turned to several “on-purpose” propylene (OPP) technologies. The on-purpose process that is slated to grow the most during the next several years is selective propane dehydrogenation (PDH) to propylene. Currently, the only PDH plant in the United States is operated by Flint Hills Resources (formerly Koch Petroleum Group), but several more are under construction or consideration.

Olefin metathesis

One OPP technology is olefin metathesis, in which ethylene and normal butenes are combined and then disproportionated to give propylene. There are two such units in the United States: one owned by LyondellBasell and the other jointly owned by BASF and Total. The economics of olefin metathesis as a way to make propylene depend on the relative prices of ethylene and propylene. As propylene values rise more than ethylene, this technique becomes increasingly attractive.

Methanol-to-propylene

The newest on-purpose way to produce propylene is methanol-to-propylene (MTP). In it, methanol (typically made from natural gas or coal) is passed over a zeolite catalyst. The methanol reacts with itself to give substantial yields of propylene, along with heavier liquid hydrocarbon byproducts. This innovative way to make propylene overcomes the historical barrier against using natural gas or coal to make olefins. BASF is planning a gas-based MTP plant in the United States; if it goes ahead, it will be the first outside of China.

Types of Cracking

Thermal cracking
In this early “brute force” method, heavy (high-boiling) petroleum fractions are heated to 700–900 ºC at high pressure (≈7000 kPa) to produce smaller hydrocarbons and even heavier, tarlike materials or petroleum coke.

Steam cracking
In this variant of thermal cracking, heavier, often saturated, hydrocarbons are diluted with steam and heated to similar temperatures, but for much shorter residence times. The products are smaller, unsaturated molecules and some hydrogen.

Fluid catalytic cracking
FCC is widely used to make gasoline, diesel fuel, and liquefied petroleum gas (LPG). In it, solid acidic catalysts such as zeolites are powdered and suspended (fluidized) in a flow of hydrocarbon feedstock at 750–850 ºC for short contact times.

Hydrocracking
In this version of FCC, heavier unsaturated hydrocarbons are converted to smaller, saturated ones for producing kerosene and diesel fuel. The temperature and pressure vary widely, depending on the feedstock and target product.

Isobutyl alcohol-to-propylene

New OPP technologies are being developed all of the time. For instance, in a recent patent application (US Pat. Appl. 20150239801, Aug. 27, 2015), Total Research and Technology discloses catalysts and operating conditions that allow the conversion of isobutyl alcohol to propylene. The Total invention is based on the initial dehydration of isobutyl alcohol, which can be made via fermentation, to isobutene. The isobutene is then catalytically cracked to propylene.

Hydrogenation of acetic acid

In another process for making propylene, workers at Archer Daniels Midland and Washington State University (Pullman) disclose catalysts that promote the conversion of acetic acid and hydrogen to good yields of propylene (US Pat. Appl. 20150239800, Aug. 27, 2015). Acetic acid can be made economically from inexpensive natural gas and potentially represents a low-cost route to propylene.

Summary

Predicting the eventual commercialization of any technology is tenuous, especially for on-purpose routes to propylene. These new approaches must withstand the rigors of scale-up and long-term catalyst life tests; and the economic viability of these new processes must be sustainable. The current volatile oil-price scenario makes it difficult to forecast economic performance. The availability of propylene from ethylene steam crackers is directly related to the price spread between oil-based naphtha and the natural gas liquids ethane and propane.