So in the good old days of Penn crude-- this is a simple now I'm holding in my hand of Pennsylvania crude. You can how fluid that is, or low viscosity. You can really see that it's almost drinkable. I'm tempted to actually take a sip from this now, but that's probably against regulations here in the museum.
Now for this you would not need any hydrotreatment because this has virtually no sulfur in it. But the current crude oil, which would be much more viscous than this, much more aromatic-- this essentially paraffinic crude-- then you would really need hydrotreatment to remove sulfur, nitrogen, or metals associated with this. This is an extinct crude oil. This is the early Pennsylvania crude.
Here we see some samples of Pennsylvania crude oil, the first group that was drilled and produced in the United States. And Penn crude, or Pennsylvania crude, is a very special crude oil. It is the sweetest. The sweet means, in this case, not really with a lot of sugar, but very low sulfur.
So you can see the light color of the crude oil. This is actually as it came from underground without any refining. This is very rare now, and Penn crude is pretty much extinct these days. This was so sweet that some entrepreneurs could actually sell this as a remedy.
There was one man who was Samuel Kier in Pittsburgh, I think he was in Canal Street. He put the Penn crude in pint bottles and sold them as remedies for different ailments, such as stomach aches, headaches, or growing mustaches on young boys who wanted to have a mustache to show off.
So sour crude oils, obviously, are what we have today. That means high sulfur crude oil, which would require quite a bit of hydrotreatment to remove sulfur, unlike the sweet Penn crude grade oil.
Having talked about the separation processes and the conversion units, we are now ready to talk about the finishing processes-- that's the third kind of processes using petroleum refining. Now finishing is done essentially to make sure that the product that is leaving the refinery is compliant with the required performance specifications, such as octane number for gasoline, or cetane number for diesel fuel. And also the environmental regulations, like sulfur, nitrogen, or metal contents of these fuels that are leaving the refinery to be sold in the marketplace.
So the finishing processes are hydrotreatment and blending. We do categorized them into these two main categories. In hydrotreatment the point is to remove the heteroatom, whatever that is, sulfur, nitrogen, or metal, with the help of a catalyst and hydrogen. So the objective is to use the minimum amount of hydrogen and make the minimum amount of change in the hydrocarbon structure of your feed materials to remove the sulfur, nitrogen, or metal out.
Minimizing hydrogen is important because hydrogen is a very expensive chemical material. And, of course, hydrotreatment or finishing is not a place to make the chemical changes desired in the hydrocarbon skeleton, or hydrocarbon structure. Conversion processes do that.
So in hydrotreatment then we would need catalysts. These are typically supported catalysts. The support is alumina, silica, in some cases, mixed oxides. And the metals typically molybdenum, cobalt or nickel that are put on these supports.
We need these metals to dissociate molecular hydrogen, so that it can actually react with these heteroatom species. In hydrodesulfurization, we remove sulfur as H 2 S, which is an acidic gas, nitrogen as ammonia, which is a base. So the purpose of hydrogen is to really seek out and find that heteroatom, and pull that out of the hydrocarbon structure as H 2 S, for sulfur, and ammonia from the nitrogen-containing species.
The metals typically of vanadium and nickel are separated as sulfides on catalyst surfaces-- a pretty interesting chemistry, as we will discuss this in lesson. So hydrotreatment would give us desirable heteroatom content, or regulated heteroatom content in the products.
With blending we need to look into all the specifications that are needed for a given product. For example, for diesel fuel the viscosity, or pour point, could be important. And typically, in a refinery to make a product like gasoline or diesel, you will be blending a large number of streams. Remember, there are quite a few different streams coming from different conversion processes, or separation processes, to be blended to make these final products.
For gasoline, it's the octane number. So we will go through some of the procedures we can use to calculate the physical properties like pour point or viscosity of these blends, to make sure that they actually follow the specifications needed for these products.
You would see that many of these calculations are nonlinear. If you take, say, sample A and sample B, blend them together, the viscosity of that blend would not be the average of two, using essentially a linear mixing formula. So there are correlations that were developed to incorporate these non-linearities into calculating, determining, the final properties of the blends from multiple streams to make the final product from the refinery.
Following our discussion on separation and conversion processes, this lesson will cover the third type of refining processes – finishing processes. Finishing processes include hydrogenation for stabilization of petroleum products, hydrotreating to remove heteroatoms (S, N, and metals) and product blending to attain the product specifications and assure compliance with environmental and government regulations on petroleum fuels and materials. The finishing step is the last stage before the hydrocarbon streams from different units leave the refinery as commercial fuels and materials. Therefore, the challenges involved both hydrotreatment and blending operations are diverse and complex. The constraints on the commercial fuels that need to be satisfied simultaneously range from composition and performance specifications to seasonal fluctuations in demand for different fuels and materials. A brief overview of only the basic concepts in finishing operations is presented in this lesson.
By the end of this lesson, you should be able to:
Please refer to the Course Syllabus for specific time frames and due dates. Specific directions for the assignments below can be found in this lesson.
Readings: | Petroleum Refining, by J. H. Gary and G. E. Handwerk, Chapter 9 (Hydrotreating) and Chapter 12 (Product Blending) |
---|---|
Assignments: | Exercise 8 Exam 2. Will cover material in Lessons 6-9. Exam 2 is found in the Exam 2 Module. |
If you have any questions, please post them to our Help Discussion Forum (not email), located in Canvas. I will check that discussion forum daily to respond. While you are there, feel free to post your own responses if you, too, are able to help out a classmate.
Hydrogenation, or adding hydrogen to unsaturated hydrocarbons, is used for stabilization of petroleum products and aromatic reduction [1]. One particular application of hydrogenation is to saturate unstable olefins and di-olefins that are implicated in producing gums (high-molecular-weight sticky semi-solid material) during the storage of fuels, such as gasoline and jet fuel. Gum formation is detrimental, particularly, to the operation of fuel injectors in combustion engines. Narrow passages found in fuel injectors can be partially or completely plugged with deposition/accumulation of gums on flow surfaces, causing engine failures. Figure 9.1 on the next page shows examples of hydrogenation of an aromatic compound (alkylated naphthalene) and an olefin. The objective of hydrogenation is just adding hydrogen to unsaturated hydrocarbons using precious metal (Pt, Pd) or Ni catalysts at low temperatures to avoid cracking or other chemical changes.
[1] Petroleum Refining, by J. H. Gary, G. E. Handwerk, M. J. Kaiser, 5th Edition, CRC Press NY, 2007, Chapter 9, Hydrotreatment, pp. 195-203.
As a natural substance, crude oil contains heteroatom (S, N, and O) compounds as well as metals (mainly Ni and V) in addition to hydrocarbons. The sulfur content of crude oils normally ranges from 0.05 to 6 wt %, nitrogen content varies from 0.1 to 1 wt %, oxygen from 0.1 to 2.0 wt %, and metals from 10 to 1000 ppm. Once crude is fractionated by distillation, the heteroatoms (particularly S, N, and metal compounds) are distributed in the products. Heteroatom compounds are mainly associated with the higher boiling fractions and the residues that are rich in aromatic compounds. The presence of heteroatom compounds in petroleum products is undesirable because they reduce fuel stability, contribute to the emission of pollutants, and damage engines. Heteroatom compounds in refinery streams (e.g., basic nitrogen compounds, sulfur compounds, and metals) can deactivate catalysts and promote coke formation [2].
The light products such as LPG and naphthas have low concentrations of sulfur and require minimal treatment, such as absorption in alkaline solvents (e.g., H2S or mercaptan sulfur) or conversion of mercaptans to sulfides to eliminate odor. For example, the UOP MeroxTM process is widely used to remove H2S and mercaptan sulfur (Merox extraction) or to convert mercaptan sulfur to less-objectionable disulfides (Merox sweetening) [2]. The Merox process can be used to treat liquids such as LPG, naphtha, and kerosene. Removing sulfur from higher molecular weight and more complex molecules requires hydrotreatment processes that use hydrogen.
[2] S. Eser and M. R. Riazi, “Crude Oil Refining Processes” In Petroleum Refining and Natural Gas Processing, Editors: M. R. Riazi, S. Eser, J. L. Peña, ASTM International, West Conshohocken, PA, 2013, p. 119.
Desulfurization of fuels is commonly achieved by catalytic hydrodesulfurization (HDS), in which the organic sulfur species are converted to H2S and the corresponding hydrocarbon, as in the following reaction: R-SH + H2 → R-H + H2S. Here R represents an alkyl group, such as methyl (CH3–), or ethyl (C2H5–). Figure 9.2 shows the type and relative reactivity of different sulfur species in HDS reactions. The reactivity of R-SH (mercaptan, or thiol) compounds is higher than that of disulfides (R-S-S-R). The H2S is easily removed from the desulfurized oil by absorption in a gas treatment unit and subsequently converted to elemental sulfur by the Claus process, as will be discussed in Lesson 10.
Some HDS reactions are shown in Figure 9.3. Note that the objective of hydrotreatment reactions is to take the heteroatom out with minimum hydrogen consumption. The ideal scenario is to get the hydrogen atoms to break the carbon-sulfur bonds and to remove sulfur as H2S. This may not always be possible because of problems with the accessibility of sulfur atoms to hydrogen atoms. Sulfur ring compounds such as methylated dibenzothiophenes have the lowest reactivity in HDS reactions because they are planar compounds, and in some methylated DBTs the S atom is shielded by the methyl groups (See Figure 9.4). Removing sulfur from these compounds requires more hydrogen consumption in order to saturate the aromatic rings in dibenzothiophene (DBT) to form non-planar compounds. Remember that, as opposed to aromatic compounds, cycloalkanes are not planar compounds. Therefore, saturation of the aromatic rings in methylated DBTs eliminates the steric hindrance (shielding) of methyl groups to make the S atom accessible to active sites on the catalyst surface for removal as H2S. The geometry of this proposition is better understood when one looks at the configuration of active sites on the HDS catalysts shown in Figure 9.5.
The HDS catalysts consist of sulfided molybdenum supported on Al2O3. Sulfided molybdenum surfaces have vacancies (with missing S atom in the sequence) that act as active sites on the catalyst surface (Figure 9.5). In addition to Mo, HDS catalysts also contain other metals, such as cobalt (Co) as promoters, the main function of which is believed to dissociate H2 to atomic species that readily react with S. The HDS mechanism involves inviting the sulfur atom at the HDS target to sit in the vacancy and once the S is in the vacancy, dispatching H atoms to clip C-S atoms to make H2S that will leave the vacancy and make it available for the next action. You can see that this scenario would get into trouble if S cannot sit in the vacancy, as would happen with the compound 4,6 dimethyldibenzothiophene (Figure 9.4 on the previous page and Figure 9.5 below). This calls for saturating the aromatic rings to twist the methyl groups aside to make the S atom fit in the vacancy to be taken away as H2S. The list below lists the relative HDS reactivities of methylated DBT compounds relative to HDS of unsubstituted DBT.
One can see above that the HDS reactivity of 4,6 dimethyldibenzothiophene (4,6 dimethylDBT) is one-tenth of unsubstituted dibenzothiophene (DBT), because of the shielding of the S atom by methyl groups as discussed above. Moving the methyl groups away from the sulfur atom to 3,7 positions (See Figure 9.4) increases the HDS reactivity 15- fold to 1.5 times of DBT. Interestingly, having methyl groups away from the S atom (as in 3,7 and 2,8 positions) increases the HDS reactivity relative to that of DBT. Methyl substitution on DBT away from the S atom increases the HDS reactivity by promoting adsorption of the compounds on the catalyst surface and may weaken the C-S bonds on DMDBTs. In contrast, even one methyl group shielding the S atom (4-methyldibenzothiphene in list above) reduces the reactivity of the methylatedDBT to one-fifth of the reactivity of DBT.
Hydrodenitrogenation (HDN) is a similar process to HDS in which hydrogen is used to remove nitrogen, and for this reason during HDS the nitrogen content of fuels is also reduced. Figure 9.6 shows different types of nitrogen-containing compounds in crude oil and refinery unit products. In addition to reducing the N content in fuels as a finishing process, basic nitrogen compounds such as pyridine and quinoline shown in Figure 9.6 should also be removed as a pretreatment step to protect the acidic catalysts used in processes such as FCC. This is similar to HDS being an important pretreatment process for feedstocks that would be exposed to noble metal catalysts (such as Pt in a catalytic reforming process) that are poisoned by sulfur compounds. Figure 9.7 shows examples of HDN reactions and catalysts used for HDN. Similar to HDN reactions for quinoline, shown in Figure 9.7, pyridine (C5H5N) can be reduced to pentane (C5H12) and ammonia (NH3) by adding 5 molesH2 in three steps. The overall HDN reaction for pyridine is C5H5N + 5H2 → C5H12 + NH3.
Similar to HDS and HDN, hydrodemetallation (HDM) is carried out: 1) as a pretreatment operation to remove metals (mostly nickel and vanadium) to protect the catalysts in subsequent conversion reactions, and 2) for removing metals in petroleum fuels to prevent corrosion in furnaces and toxic emissions from combustion engines. As mentioned in Lesson 1, metals (Ni, V) are mostly incorporated in the cage-like structures of highly stable organometallic compounds called porphyrins. As seen in Figure 9.8, to remove the metals, the cage structure of porphryins needs to be broken up. Similar to HDS and HDN, hydrotreatment on a Mo/Al2O3 catalyst first breaks up the chemical bonds in porphyrins to free the metals. The metals are then reacted with H2S and precipitated as metal sulfides, e.g., NixSy, on catalyst surfaces. Note that this is different from HDS and HDN where the heteroatoms are removed as gas (H2S and NH3, respectively). [Think: What would be the significance of removing metals as metal sulfides as far as the morphology of the HDM catalysts is concerned? Save your answer for the Self-Check question later in this lesson. ] Metals Ni and V deposited on the catalysts can be recovered by post treatment of the used catalysts.
In addition to HDS, HDM, and HDN, there may be a need for hydrodeoxygenation and hydrodehalogenation processes to remove O, or Cl in petroleum products [1].
[1] Petroleum Refining, by J. H. Gary, G. E. Handwerk, M. J. Kaiser, 5th Edition, CRC Press NY, 2007, Chapter 9, Hydrotreatment, pp. 195-203.
Process objectives, conditions, and configurations are similar for all hydrotreatment processes [1]. As an example, HDS list, below, lists the project objectives and selected conditions for HDS processes. Minimization of cracking (or any other chemical change that is not needed for removing heteroatoms) and minimization of hydrogen consumption are the two principal objectives of hydrotreatment processes for reducing the heteroatom concentrations to the desired levels in the reaction products. These objectives are achieved by careful selection of process conditions that need to be adjusted based on the chemical composition of the feedstock. You can see in HDS list, that more severe reaction conditions (e.g., higher reaction temperatures, higher hydrogen/feed ratios, and lower space velocities) are necessary for treating heavy residue compared to treating light distillates. Catalysts have shorter lives in residue treatment processes, as well, because of harsh reaction conditions and more excessive deposition on catalysts during the treatment of heavy residue.
Objectives
Process Conditions |
Light Distillate |
Heavy Residue |
Temperature,&;ºC |
300-400 |
340-425 |
LHSV, h-1 | 2-10 | 0.2-1 |
H2/Oil, scf/bbl | 300-2000 | 2000-10000 |
Catalyst life, years | 10 | ~1 |
Note: LHSV (Liquid Hourly Space Velocity) – the volumetric flow rate of liquid reactant passing through a reactor divided by the volume of the catalyst in the reactor, measured in units of h-1. Lower LHSV (i.e., longer exposure to catalyst in the reactor) is needed for more challenging hydrotreatment tasks.
Figure 9.9 shows a process flow diagram for HDS. The feed is introduced into a fixed-bed catalytic reactor along with hydrogen, and the products are sent to a high-pressure separator to separate H2 and H2S gases from the hydrocarbons. The gases are sent to a scrubber with a basic solution (containing ethanol amine or diethanol amine) to remove H2S that is sent to the sulfur recovery unit and H2 is recycled to the treatment reactor. Desulfurized products are sent to a fractionator to separate C3 and C4 alkanes (LPG) and the liquid products.
For HDS, HDN and HDM sequential reactors can be configured with different catalysts optimized for the targeted heteroatom removal. Figure 9.10 shows staging of the HDM and HDS operations. The first step in a series of reactors is always HDM because of the unique removal mechanism of the metals (Ni and V), as solid materials on catalyst surfaces, catalysts with large pores in the support materials are needed so that the problems with reactor plugging can be avoided. Following the removal of metals, HDS and HDN can be conducted on medium-pore and small-pore catalysts, in some cases simultaneously in the same reactor or the same catalyst bed.
Staging of Hydrotreatment Processes
-Feed w/high metals + high S content
-First HDM
-H2S, NixSy, VxSy
-Mo/Alumina, large pores
-Finish HDM then HDS
-CoMo/Alumina, medium pores
-Finish HDS
-CoMo/Alumina, small pores
-Product
Increasing sulfur contents of the crude oils available for the U.S. refineries on one hand, and decreasing sulfur contents of regulated fuels on the other, have increased the sulfur production in the U.S. refineries as a by-product to 40,000 tons/day [4]. This is sufficient capacity to meet all sulfur demand for the chemical industry, particularly for H2SO4 production used in many chemical industries. One should note that this quantity of sulfur comes only from desulfurization of the fuels that are subject to sulfur regulation. A significant fraction of sulfur in crude oil (>60%) is concentrated in the heavy ends (e.g., residual fuel oil, asphalt) that are not regulated for sulfur content.
[1] Petroleum Refining, by J. H. Gary, G. E. Handwerk, M. J. Kaiser, 5th Edition, CRC Press NY, 2007, Chapter 9, Hydrotreatment, pp. 195-203.
[4] Petroleum Refining, by J. H. Gary, G. E. Handwerk, M. J. Kaiser, 5th Edition, CRC Press NY, 2007, Chapter 12, Product Blending, pp.257-270.
Product blending plays a key role in preparing the refinery products for the market to satisfy the product specifications and environmental regulations. The objective of product blending is to assign all available blend components to satisfy the product demand and specifications to minimize cost and maximize overall profit [5]. Almost all refinery products are blended for the optimal use of all of the intermediate product streams for the most efficient and profitable conversion of petroleum to marketable products. For example, typical motor gasolines may consist of straight-run naphtha from distillation, crackate (from FCC), reformate, alkylate, isomerate, and polymerate, in proportions to make the desired grades of gasoline and the specifications.
Basic intermediate streams can be blended into different finished products. For example, naphthas can be blended into gasoline, or jet fuel streams, depending on the demand. Until the 1960s, the blending was performed in batch operations. With computerization and the availability of the required equipment, online blending operations have replaced blending in batch processes. Keeping inventories of the blending stocks along with cost and physical data has increased the flexibility of and profits from online blending through optimization programs. In most cases, the components blend nonlinearly for a given property (e.g., vapor pressure, octane number, cetane number, viscosity, pour point), and correlations and programming are required for reliable predictions of the specified properties in the blends [3].
[3] U.S. Refinery Sulfur Production Capacity [3]
[5] Petroleum Refining, by J. H. Gary, G. E. Handwerk, M. J. Kaiser, 5th Edition, CRC Press NY, 2007, Chapter 12, Product Blending, p.267.
Octane numbers are blended on a volumetric basis using the blending octane numbers of the components. True octane numbers do not blend linearly, thus it is necessary to use blending octane numbers in calculating the octane number of the blend. Blending octane numbers can be estimated from empirical correlations that have been developed over the years. Blending octane numbers, when added on a volumetric average basis, will give the true octane of the blend, as can be obtained from standard test using CFR test engines.
(True Octane Number of a blend) ON = Σ xi∗ ONi
Where xi is the volume fraction of component i in the blend, and ONi is the blending octane number of component i.
For example, if you have
and would like to get a gasoline of ON = 83, what would be the volume fraction of reformate (x) in the blend?
So, you need 73% by volume of reformate in your blend. Additive concentration may be calculated the same way, and ON for multicomponent blends can be calculated the same way for research or motor octane numbers.
Pour point is an important property for diesel and fuel oil blends. Pour point blending is also non-linear, and pour point blending indices were developed to enable reliable calculation of the pour points of the blends. Pour point blending indices for some distillate fuels are given in Table 12.7 of the textbook, and copied in Figure 9.11, where the blending indices are tabulated as a function of ASTM 50% temperature, °F (first horizontal listing) and Pour Point, °F (first vertical listing). For example, pour point blending index for a distillate that has an ASTM 50% temperature of 500°F and a pour point of 40°F would be 36. This index can be used in calculating the pour point of blends using this distillate as a component.
As an example of using blending indices to calculate the pour point of a blend, consider blending a straight-run gas oil (50% ASTM T = 470°F and pour point = -6°F) and a hydrotreated heavy gas oil (50% ASTM T = 620°F and pour point = 40°F). What would be the pour point of a binary blend that consists of 68.7 % vol of straight-run gas oil and 31.3% vol of hydrotreated heavy gas oil?
The procedure used to calculate the pour point of the blend (shown Figure 9.12) can be summarized as follows:
Note that if you would assume linear addition of the pour points, you would calculate a blend pour point as = 0.687x(-6) +0.313x40 = 8.4°F. A serious underestimation of the pour point! Thinking that your diesel fuel has a pour point 8.4 °F, you may try to start your diesel truck on a 12°F day, to no avail, not knowing that your fuel tank has a gel, and not a liquid that can be easily pumped to the combustion cylinder.
Similar to pour point, the viscosity of blends can also be calculates using blending index numbers, or plots developed for this purpose. For the same blend, we can use the Viscosity Blending Index Numbers in Table 12. 3. of your textbook [6], and the procedure shown in Figure 9.13 to calculate the viscosity of the blend. Try to verify the numbers given in Figure 9.13, and calculate a viscosity of the blend if viscosities were linearly additive to compare with the value calculated in Figure 9.13. Post your questions, if any, and comments in the Help Discussion Forum.
Pour Point | Volume Fraction |
50% ASTM T, ºF |
Pour Point Index (PPI) |
Pour Factor (PPI x Volume Fraction) |
Blend (50% ASTM T x Volume Fraction |
|
Straight-run gas oil | -6ºF | 0.687 | 470 | 6.3 | 4.33 | 323 |
Hydrotreated heavy gas oil | 40ºF | 0.313 | 620 | 26 | 8.14 | 194 |
Total | - | - | - | - | 12.47 | 517 |
Read chart using this information to get a pour point of 15ºF
Volume | Viscosity | Viscosity Index (Table 12.3 in the textbook) | Factor (VI x Vol) | |
SRGO | 0.687 | 2.0 cst | 0.290 | 0.199 |
HHGO | 0.313 | 8.0 cst | 0.443 | 0.139 |
Total | - | - | - | 0.338 |
From the table, you get: Viscosity = 2.9 cst
[6] Petroleum Refining, by J. H. Gary, G. E. Handwerk, M. J. Kaiser, 5th Edition, CRC Press NY, 2007, Chapter 12, Product Blending, p.263.
Please take a few minutes to answer the questions below. When you are happy with your answers, click Check.
The assignments for this week are listed below. Please be aware that you have an exam next week!
Exam 2 (eTest). Will cover material in Lessons 6-9. Check the Syllabus, or Course Calendar for Exam 2 schedule and venue.
Finishing processes make sure that the refinery products fully comply with the commercial specifications and environmental regulations. Hydrotreatment removes heteroatoms from intermediate refinery products to protect the catalysts in the subsequent processes, or from final products to be sent out to the market. A particular challenge for hydrotreatment is that while the crude oil slate is getting more contaminated, the demand for cleaner fuels is mandated by the increasingly stringent environmental regulations. Major hydrotreatment processes include hydrodesulfurization, hydrodenitrogenation, and hydrodemetallation, all using hydrogen and specially designed catalysts to remove heteroatoms with minimal change to the hydrocarbon constitution of petroleum products. Product blending as a major finishing process takes up the challenge of the optimum allocation of many intermediate streams in the refinery to make up the refinery products to satisfy all performance parameters and environmental mandates. Product blending is carried out using linear and non-linear programming techniques for online blending.
You should now be able to:
You have reached the end of Lesson 9! Double-check the to-do list below to make sure you have completed all of the activities listed there before you begin Lesson 10.
Readings: | Petroleum Refining, by J. H. Gary and G. E. Handwerk, Chapter 9 (Hydrotreating) and Chapter 12 (Product Blending) |
---|---|
Assignments: | Exercise 8 Exam 2. Will cover material in Lessons 6-9. Exam 2 is found in the Exam 2 Module. |
If you have any questions, please post them to our Help Discussion Forum (not email), located in Canvas. I will check that discussion forum daily to respond. While you are there, feel free to post your own responses if you, too, are able to help out a classmate.
Links
[1] https://www.youtube.com/channel/UCU1QB1a5XJa_nTHD2lzr7Ew
[2] http://creativecommons.org/licenses/by-nc-sa/4.0/
[3] https://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=PET&s=8_NA_8PS_NUS_S&f=A
[4] https://www.e-education.psu.edu/fsc432/sites/www.e-education.psu.edu.fsc432/files/Lesson9/Table%2012.7%20%28Pour%20Point%20Blending%20Indices%20for%20Distillate%20Stocks%29.xlsx