Special Report Clean Fuels
U. NAVARRO, M. NI and D. ORLICKI,
Grace Catalysts Technologies, Columbia, Maryland
FCC 101: How to estimate product yields
cost-effectively and improve operations
The feed to fluid catalytic cracking unit (FCCU) is the most
important process variable. FCCU feed has the greatest impact
on operating conditions, yield and product quality. A typical
FCCU feed consists of hydrocarbon (HC) families, includ-
ing paraffins and cycloparaffins, which are saturates (aromatic
HCs with a different number of aromatic rings), resins and
asphaltenes. In some cases, residue is processed in the FCCU.
Feed behavior in the riser also depends on the mechanical and
operational conditions of the unit, and the concentration and
distribution of these HC families.
Feed quality. Thus, feed quality depends on several factors:
quality and type of crude processed by the crude distillation
units, other processing units, and the complexity of the refin-
ery. Products from other processes, such as thermal conversion
units like delayed coking, coking and visbreaking, and hydrogen
addition units like gasoil (GO) and residue hydrotreaters, hy-
drocracking of GO and liquid-liquid deasphalting units. Product
streams from lubricant production are also sent to the FCCU.
Benefits. One of the biggest advantages of the FCCUs is the
flexibility to process all types of HC streams that are complex
mixtures in which the mixing processes are not always efficient
to ensure completely homogeneous blends. FCCU feeds are
blends of atmospheric and vacuum GOs, atmospheric resi-
dues, coker and visbreaking GOs, hydrocracking residues, hy-
drotreated GOs and residues, furfural extracts, demetalized oil
(DMO) and others. However, one of the greatest weaknesses
and problems in refineries is the lack of physical-chemical anal-
yses for proper feed characterization. What is more important
for engineers, supervisors and operators is the ability to predict
the impact of changing feedstock quality on product yield, op-
erating conditions (heat balance) and product quality.
There are valuable tools that FCC staff can use to quantify
the potential yields of the feed determined by chemical compo-
sition, i.e., by the different HC families. However, the most im-
portant thing is that these potential yields are calculated using
properties typically analyzed in any refinery laboratories, such
as specific gravity (SG), distillation, sulfur (S) content, refrac-
tive index and more.
Background. To conduct this study, more than 100 types of
FCC feeds from all over the world with different composi-
tions were selected.a These feeds contain all of the HC families
mentioned earlier. The study evaluated heavy to light oils—11
°API and 32 °API. The feeds were fully characterized by phys-
ical-chemical analyses. The HCs were characterized by SARA
analysis, mass spectrometry (MS), ultraviolet-visible spec-
troscopy (UV-Vis) and hydrogen nuclear magnetic resonance
(1H-NMR) to quantify hydrogen (H2 ) content, which is an
important property of FCC feeds.
Potential feed yields were analyzed in an ACE unit under
these operating conditions: Reaction temperature 527°C
(980°F), reaction time of 30 sec, feed flow of 3 g/min, catalyst/
oil (C/O) ratio of 4, 6 and 8, adjusted by varying the amount
of catalyst in the reactor.1 Two commercial catalysts were used,
deactivated by the CPS unit at 793°C (1,460°F).2 The proper-
ties of the deactivated catalysts are reported.3, 4 The catalysts
were deactivated in the absence of metals, since the purpose of
the study was to evaluate the interaction of the active sites of
the catalyst (matrix and zeolite) with the different HC types in
the selected feeds, and metal contaminants [nickel (Ni) and va-
nadium (V)] interfering with the reactions of hydrogen trans-
fer, dehydrogenation and catalytic activity, which was not the
focus of this research.
Feed classification. To understand FCC feed composition,
the HC type must be identified. Many methods are available
in literature.5–7 The correlation index (CI) and H2 content are
used in this article. The CI is a property developed to classify
crudes and petroleum fractions by the US Bureau of Mines:6
CI = 473.7 × d + 48,640/(K + 273) – 456.8 (1)
Where K is the mid-boiling point in °C and d is SG. The
meaning of this property is very similar to the K characteriza-
tion factor. However, unlike the K factor that ranges between
11 and 13, the CI ranges from 0 to 100. In this wide range, n-
heptane = 0 and cyclohexane = 50, while benzene = 100; thus,
low CI values for HCs correspond to paraffinic feeds; values
close to 50 are typical of naphthenic feeds, and higher values
are of feeds that have a larger proportion of aromatic HCs.
Studies conducted with GOs from different crudes estab-
lished that paraffinic feeds have a CI of less than 42.8 CI value
between 42 and 55 are cycloparaffinic HCs; so, as CI reach-
es about 50, the proportion of paraffinic HCs decreases and
naphthenic HCs increase. In values between 50 and 55, the
proportion of aromatic HCs begins to increase. Feeds with CI
values over 55 can be classified as aromatic, as shown in FIG. 1.
Hydrocarbon Processing | FEBRUARY 2015 41
 Clean Fuels

The H2 content of all the feeds was measured using 1H-
NMR.3, 4 In accordance with this property, the H2 content in
paraffinic feeds (K factor > 12) is greater than 12.8%, while
naphthenic feeds (K factor between 11.5 and 11.8) have an H2
concentration ranging from 12% to 12.8%. Finally, the H2 con-
tent of aromatic feeds is less than 12%. Hydrogen content is an
important property of FCC feeds, because it defines the yield
distribution of valuable products. The higher the H2 content
in the feed, the more H2 that will be distributed in the valuable
FCC products.7
The feeds studied in this project were selected with a wide
°API range: between 11.4 °API and 32.4 °API, which is equiva-
lent to H2 content between 9.5% and 14.25% and a CI ranging
from 24 to 88. This wide range comprises virtually all of the
combined feeds and the different components used in the dif-
ferent FCCUs in the world.
Chemical composition of FCC feeds. Since one of the
purposes of this study was to demonstrate that the behavior
of FCC feeds due to the HC types present, HC distribution
was analyzed using different methods: SARA analysis, MS and
UV-Vis. This information was used to develop correlations for
refinery staff to calculate important properties of the feed and
to provide a better understanding of FCC feed quality and in-
fluence on unit operation.
The first property calculated is feed H2 content. In this case,
the H2 content was measured by 1H-NMR, and CI was used.
As mentioned earlier, CI is a property calculated based on feed
14.5
14.0
13.5
13.0
12.5
H2 content, %
density and distillation, which are two properties that are typi-
cally measured in all refinery laboratories. From FIG. 1, there is
a strong correlation between these two properties, and it is rep-
resented by Eq. 2, with a correlation coefficient of 0.966 and a
standard error of 0.204:
H2 , % = 15.72432 – 0.06798 × CI (2)
Another important property that defines feed quality is
saturate content. It is defined by the sum of paraffinic and
cycloparaffinic (naphthenic) HCs. Due to the chemistry of
cracking reactions, the interactions with the active sites of the
catalyst, catalyst morphology to allow diffusion through the
pore system, and the mechanical and operating conditions of
the FCCU, these HCs must be converted into valuable prod-
ucts, mainly gasoline and LPG, at an efficiency of at least 90%.
It is vital to determine and quantify the saturated HC content
in FCC feeds (FIG. 2).
Also, FIG. 2 is used to calculate saturate content based on H2
content and is represented by Eq. 3, with a correlation coef-
ficient of 0.952 and standard error of 2.97:
Saturate content, % = 1/[0.15498 – 0.05378 (3)
× Ln(%H2 )]
In FIG. 2, the lowest H2 and saturate value belongs to heavy
cycle oil (HCO), which, in some FCCUs, is sent to the riser
as recycled. In addition, the highest values belong to combined
feeds of excellent quality and hydrocracking (HCK) residues.
Other important HC families in FCC feeds are aromatics,
which cannot be cracked due to the difficulty in accessing the
active sites of the catalyst. These HCs are normally dealkyl-
ated, i.e., the lateral paraffinic chains are broken, along with
the cycloparaffinic rings attached to the aromatic ring. In these
HCs, monoaromatics are dealkylated, and the monoaromatic
ring with minor lateral groups is sent to the naphtha fraction,

12.0 providing a good octane rating. FIG. 3 shows the correlation be-
11.5 tween the H2 content and aromatic concentration of the FCC
11.0 feeds, represented by Eq. 4, with a correlation coefficient of
10.5 0.950 and standard error of 2.88:
10.0
9.5 Aromatic content, % = –95.1947 + 33.86928 ×
9.0 (H2 ) – 1.82886 × (H2 )2 (4)
20 30 40 50 60 70 80 90
CI
To validate the calculations of saturate and aromatic con-
FIG. 1. Feed classification. tent, several feeds were analyzed by MS. This analytical tech-
nique shows the full distribution of the HCs in the crude oil
90 65
85
60
80
75 55
70 50
65 45
Saturates, %

Aromatics, %

60 40
55 35
50 30
45
25
40
35 20
30 15
25 10
9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5
H2 content, % H2 content, %

FIG. 2. Saturate content. FIG. 3. Aromatic content.

42 FEBRUARY 2015 | HydrocarbonProcessing.com
Select 64 at www.HydrocarbonProcessing.com/RS
 Clean Fuels

fractions. TABLE 1 summarizes the comparative distribution of rate concentration decreases. This HC distribution explains the
the HCs in HCK residue and vacuum coker GO. In addition, it behavior of the two feeds when sent to the FCC riser. The HCK
lists certain properties, such as H2 content, °API and S content. residue increases gasoline and LPG yield, and it cools the regen-
When H2 is added, the saturate content of the HCK residue erator, as Δ coke make is low. Conversely, due to its high diaro-
is quite high, while the aromatic content is low. The coker GO matic content, the coker GOs increase light cycle oil (LCO)
is quite the contrary; the aromatic content increases, and satu- yield, while triaromatics and tetraaromatics are bottom (slurry)
and coke precursors. In addition, the high level of S-containing
compounds in coker GO explains its high S content.
TABLE 1. HC distribution by MS
Analysis by MS was used to corroborate and validate the
HCK residue CC-VGO equations developed to calculate the saturate and aromatic
Hydrogen content, % 14.04 11.66 content of the feeds. TABLE 2 compares the total aromatic and
API gravity, °API 32.67 15.51
saturate contents reported by MS and those calculated using
the correlations presented (Eqs. 3 and 4). These results sup-
S content, % 0.02 3.55
ported the correlations developed and are chemically valid.
Paraffins (CnH2n+2) 30.63 2.03 Although the techniques used are completely different, the val-
Monocycloparaffins (CnH2n) 32.57 12.00 ues obtained were very similar to those reported by MS.
Dicycloparaffins (CnH2n–2) 14.53 10.83
Estimating potential yields of FCC feeds. The main pur-
Tricycloparaffins (CnH2n–4) 6.43 7.47
pose of this article is to predict the potential yields of FCC
Tetracycloparaffins (CnH2n–6) 0.33 3.70 feeds and to identify the determinant variables for this behav-
Total saturates, % 84.50 36.03 ior. To do so, all feeds were processed through the ACE unit
Aromatics
under standard conditions.
Gasoline yields and the conversion obtained for each feed at
Alkylbenzenes (CnH2n–6) 4.53 5.73
the three severities, i.e., the three C/O ratios, were linearized and
Benzocycloparaffins (CnH2n–8) 2.70 5.17 used to graph gasoline selectivity (gasoline yield/conversion) as
Benzodicycloparaffins (CnH2n–10) 0.60 4.97 a function of kinetic conversion (C/100-C), where C is the con-
Total monoaromatics 7.83 15.87
version obtained in the ACE unit (FIG. 4). Once the equation,
intercept and slope were obtained, the maximum conversion was
Naphthalenes (CnH2n–12) 0.47 3.33
calculated, where the maximum gasoline yield is obtained. FIG. 5
Other diaromatics CnH2n–14 1.80 15.33 illustrates these calculations, which show the experimental val-
Total diaromatics 2.27 18.67 ues obtained in the ACE unit with two catalysts and the theoretic
Total triaromatics 4.77 17.83
values calculated based on the straight-line equation (FIG. 4).
In addition, it shows the maximum gasoline yield and con-
Total tetraaromatics 0.20 2.60 version value at that point. This value is known as the feed
Total aromatics, % 15.07 54.97 crackability factor (CF). Once the maximum conversion CF
Thiophenes (CnH2n–4 )S, benzothiophenes 0.00 0.53 is obtained for each feed, this value can be used to calculate
(CnH2n–8)S, dibenzothiophenes (CnH2n–16)S
0.47 5.43
other yields of FCC products at that point of maximum gaso-
Total sulfured aromatics, % line yield, i.e., to calculate the yields for dry gas, C3, C3=, C4 ,
0.00 3.10 C4=, total LPG, gasoline, LCO, slurry and coke. Gasoline is de-
0.47 9.07 fined by the cut C5–221°C (430°F), LCO was defined by the
cut 221°C–371°C (430°F–700°F), and slurry or bottoms were
TABLE 2. HC distribution by MS and calculated
defined by the cut 371°C+ (700°F+). In addition, other prop-

MS Calculated correlation 0.78

Feed Saturates Aromatics Saturates Aromatics 0.77

1 46.00 49.83 46.86 47.88
0.76
2 53.47 43.10 57.30 37.38
Gasoline selectivity

3 47.90 42.83 47.50 47.22 0.75

4 54.73 41.90 57.11 37.56
0.74
5 66.17 32.73 66.63 28.73
GS = -0.0414Xc + 0.8408
6 53.00 39.93 52.44 42.20 R2 = 0.9979
7 58.80 40.33 62.66 32.31
8 84.50 15.07 77.55 19.81
0.71
9 41.17 50.97 45.01 49.76 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25
Kinetic conversion
10 64.47 28.27 64.90 30.26
11 50.97 43.60 52.73 41.91 FIG. 4. Gasoline selectivity vs. kinetic conversion.

44 FEBRUARY 2015 | HydrocarbonProcessing.com
DEDICATED TO MAKING OUR CLIENTS IN THE HYDROCARBON INDUSTRY WORLD-CLASS BY DRIVING A NEW
STANDARD IN OPERATIONAL EXCELLENCE THROUGH EXPERT CONSULTING AND ADVANCED TECHNOLOGY.

Optimisation and Profit

Improvement from pore to pump
from a deep understanding of the
chemical engineering of assets.

PVT, thermal hydraulics and process facility

simulation from well bore through refining and
petrochemicals in an open platform, which is
enabled for workflows across the assets.

Sustainable operational excellence from

strategic and organisational consulting,
focused on people and processes that make
the assets work hard.

KBC uses its 35+ years of history working in hundreds of oil and gas facilities worldwide to bring
practical, sustainable solutions to its clients to improve their bottom line performance by using
proven best practices and unit optimisation tools and techniques.
Select 81 at www.HydrocarbonProcessing.com/RS

KBC Advanced Technologies

AMER: +1 281 293 8200 ASIA: +65 6735 5488 EMEA: +44 1932 242424
answers@kbcat.com www.kbcat.com blog.kbcat.com
 Clean Fuels

erties such as FCC naphtha octane (RON and MON) were
calculated at that point and the Δ coke. In the latter cases, the
equation for linearization is based on kinetic conversion.
To calculate these yields, each of the FCC products has to
be linearized, for which the ratios listed in TABLE 3 were used. In
this table, the value of C is the conversion at maximum gasoline
yield. Only the results obtained from the product yields of each
of the feeds with Catalyst 2 are listed here. The product yields
of Catalyst 1 has higher zeolite content, i.e., greater activity, and
the selectivity are different.
Calculating the maximum conversion–CF. To predict the
CF of the feed, all the physical and chemical properties of the
feeds were used. The maximum conversion was defined by the
multivariable model of Eq. 5 and FIG. 6:
CF = 450.7 – 1.296 × API + 1.53901 × H2 – 0.562 ×
CI – 224.28 × Ref. Ind. + 10.51 × N
where Ref. Ind. is the refractive index and N is the feed’s total
nitrogen. The correlation coefficient is 0.938 with a standard
deviation of 1.55.
It is interesting to see how the refractive index, which is a
simple property that is easy to measure in refinery laborato-
ries, becomes an important property in predicting the CF of
FCC feeds. The refractive index is a property that responds to
the type of HCs in the feeds. If this property is increased, satu-
rate content decreases and aromatics begin to increase, so the
crackability of the feed decreases.
TABLE 3. Ratios to linearize yields
Ratios X Y
Dry gas (DG) C/(100-C) DG yield
FIG. 7 shows that the CF can also be calculated directly based
on the H2 content of the feed using Eq. 6. FIGS. 6 and 7 lead to
the conclusion that the vast majority of the feeds used in the FC-
CUs have a maximum conversion exceeding 70%, where lower
values are related to coker GOs and heavy cycle oils (HCOs).
2
CF = 85.87257/(1 + 399.7161x e–0.6393 × H ) (6)
Calculating maximum gasoline yield. The potential maxi-
mum yield of gasoline is defined as the yield at the maximum
conversion point, which is guided by the chemical composi-
tion of the feed and the distribution of the HC families present.
It is important for the refiner to know this value when optimiz-
ing the unit to obtain this potential yield. The mechanical and
operating conditions, as well as the catalyst type, are important
variables. To predict the maximum potential gasoline yield, the
(5) main physical and chemical properties of the feeds were used.
FIG. 8 shows a very good correlation, based on Eq. 7:
Maximum gasoline = 89.6285 + 0.3403 × Aromatics + (7)
0.7511 × CF – 0.3476 × CI-61.2848 × Ref. Ind.
As illustrated by Eq. 7 and in addition to the properties ex-
plained in Eq. 5, the multivariable model included aromatic
content and CF. The model shows a correlation coefficient of
0.962 and a standard deviation of 1.3. Similar to the maximum
conversion case, the maximum gasoline yield can be predicted
directly based on H2, saturate content and CI. Since it is con-
sidered to be applied the most in refineries and to simplify the
calculation, the equation is included to calculate maximum
gasoline yield based on the CI (FIG. 9 and Eq. 8):
90
85
Observed maximum conversion, wt%

C3 C/(100-C) Yield C3 80
C3= C/(100-C) Yield C3= 75
C4= C/(100-C) Yield C4= 70

LPG C/(100-C) LPG yield 65

Gasoline C/(100-C) % Gasoline/C 60

55
Coke C/(100-C) Coke yield
50
LCO C % ALC/(100-C) 50 55 60 6.5 70 7.5 80 85 90
Predicted maximum conversion, wt%
Slurry 1-C/100 Slurry yield
FIG. 6. Maximum conversion.
54 Maximum gasoline yield
52 85
Catalyst 1
50 Catalyst 2 80
48 75
Maximum conversion, wt%
Gasoline yield, %

46 70
44 Experimental values
65
42
60
40 Maximum conversion
“Concept of crackability” 55
38
36 Curve calculated 50
34 45
40 44 48 52 56 60 64 68 72 76 80 84 88 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5
Conversion, wt% H2 content, %

FIG. 5. Gasoline yield. FIG. 7. Calculation of maximum conversion.

46 FEBRUARY 2015 | HydrocarbonProcessing.com
 Clean Fuels

Maximum gasoline, % = 46.457 + 0.6382 × CI-0.011 × CI 2 (8)
In this case, the correlation coefficient is 0.959 and standard
deviation is 1.88. According to these results, most FCC feeds
have a CI between 35 and 60 for a maximum gasoline yield that
can range from 45% to 56%, where the higher values are for
highly paraffinic, hydrotreated GOs and HCK residues.
Calculating yields of other products. Several equations
were developed by the multivariable model for the different
FCC products based on the main feed properties:
Calculation of LPG yield. In many countries, LPG (C3 ,
C3=, total C4 and C4=) is a very important product, particularly
in refineries where there are alkylation and polymerization
units for high-octane gasoline production. Knowing the im-
pact of the feed on the yield of this product is important when
adjusting operating conditions for the best economic value and
where using ZSM-5-based olefin additives is a variable to in-
crease production. Eq. 9 defines the calculation of LPG yield
at the maximum gasoline conversion point. Beyond this point
(overcracking), LPG yield is increased by second-order reac-
tions, and gasoline is reduced. The correlation coefficient was
0.91 and standard deviation was 0.64.
LPG yield, % = –54.456 + 0.838 × API + 0.099 ×
Aromatics + 0.311 × CF + 0.317 × CI + 7.190 ×
N + 0.455 × S + 0.1765 × Saturates
Where S is the sulfur content of the feed. According to the
results obtained with the ACE unit and the evaluation con-
ditions of the different feeds, LPG yield ranged from 11% to
23%, where the highest values belonged to more paraffinic
GOs, hydrotreated feeds and HCK residues. In Eqs. 10 and 11,
The correlation coefficient for C4 olefin yield was 0.925
with a standard deviation of 0.26, while, for iC4 (isobutane),
the coefficient was 0.934 with a standard deviation of 0.20.
Calculation of LCO yield. LCO is another important FCC
product, as diesel consumption is increasing globally. In some
FCC units, LCO is one of the main products because it allows
the adding of this stream to hydrotreated diesel streams, thus
increasing volume for the refinery. Eq. 12 and FIG. 10 show the
results when calculating LCO yield:
LCO yield, % = 114.604 + 0.314 × API-0.618 × CF + (12)
0.182 × CI-42.695 × Ref. Ind. + 6.023 × N
The correlation coefficient obtained was 0.962, with a stan-
dard deviation of 0.88. The data in FIG. 10 led to the conclusion
that, for most FCC feeds, LCO yield ranges from 15% to 27%.
The highest values in FIG. 10 belong to HCO and heavy coker
GOs. These results confirm the advantage of recycling HCO to
the riser when optimizing LCO production.
The lowest values are from HCK residues, where, despite
the low LCO precursor content, identified as all diaromatic
HCs and sulfured aromatics (< 3%, see TABLE 1), LCO yield
is relatively high. This suggests that LCO in the hydrotreated
feeds is produced by dehydrogenation of cycloparaffinic rings
and structures as a result of hydrotreating, which had already
been found in previous studies.9, 10
(9) Calculating bottoms yield. As mentioned earlier, for the
purposes of this study, bottoms are the HC fractions with boil-
ing points above 371°C (700°F). This fraction includes HCO
and slurry. The results presented in this article are those ob-
60
55
50
Maximum gasoline, wt%

correlations were developed for each of the products that make

up the LPG and include C4 olefins and iC4 only, because they 45
are important for alkylation units: 40
35
C4= yield = –26.778 + 0.175 × API + 0.0401 × 30
Aromatics + 0.073 × CR + 12.255 × Ref. Ind + (10)
25
6.770 × N Basic + 0.343 × S + 0.0591 × Saturates
20
20 25 30 35 40 45 50 55 60 65 70 75 80 85 90
iC4 yield = 14.044 + 0.317 × API + 0.095 × CF + (11) Correlation index
0.453 × H2 + 0.170 × CI – 24.943 × Ref. Ind
FIG. 9. Calculating maximum gasoline yield.

61
42
39
Observed maximum conversion, wt%

51 36
Observed LCO yield, wt%

33
30
41 27
24
21
31
18
15
21 12
21 31 41 51 61 12 15 18 21 24 27 30 33 36 39 42
Predicted maximum conversion, wt% Predicted LCO yield, wt%

FIG. 8. Maximum gasoline. FIG. 10. LCO yield calculation.

Hydrocarbon Processing | FEBRUARY 2015 47

 Clean Fuels

tained with Catalyst 2, as shown in FIG. 5; the technology in- • Contaminant coke is produced by metals (Ni and V)
creases bottoms conversion, mainly focused on the LCO. • Circulation coke is the coke absorbed by the catalyst
pores and is not stripped
Bottoms yield, %: = 44.647 – 0.662 × API-0.20 ×
(13) • Coke additive of the feed is the heavy fraction of the
Aromatics – 0.466 × CF + 1.541 × H2 – 9.162 ×
feed that is not vaporized on contact with the hot
N Basic – 0.405 × S
catalyst from the regenerator.
Eq. 13 had a correlation coefficient of 0.924 and a standard As mentioned earlier, two metal-free catalysts that were de-
deviation of 0.72. From the results, the bottoms yield in the activated in the CPS unit were used, so contaminant coke has
ACE unit for the vast majority of the feeds and a catalyst with not been taken into consideration. Furthermore, the desorption
an active and selective matrix technology ranging from 3% process in the ACE unit is very efficient, as it is carried out for
to 8%, where the lowest values belong to paraffinic feeds and at least 7 minutes. With these two explanations, it can be con-
HCK residues: cluded that the Δ coke under evaluation is the catalytic coke.
The coke additive of the feed, which is directly related to the
Dry gas yield, % = –1.445 + 0.044 × CI + 2.716 × CC content, as demonstrated in Eq. 16, which resulted in a cor-
N Basic + 0.038 × CRR + 0.062 × S + 0.020 (14) relation coefficient of 0.956 and a standard deviation of 0.04.
× Saturates FIG. 11 shows that the coke (catalytic and feed) transported by
Coke yield, % = –181.786 + 0.384 × API + 0.106 × the catalyst to the regenerator at the operating conditions of this
(15) study range for most feeds between 0.2% and 0.7%. Values above
CI + 113.35 × Ref. Ind. + 0.518 × CRR
0.8 are blends of feeds with residual components that have high
In Eqs. 14 and 15, CRR is the Conradson carbon residue, %. CC contents and feeds such as HCO and coker GOs with high
The correlation coefficient to calculate dry gas was 0.914 with coke precursor aromatic contents (triaromatics and heavier).
a standard deviation of 0.13, while, for coke, the correlation co-
Δ coke, % = –26.77 + 0.0294 × CRR + 0.037 × API + (16)
efficient was 0.939 with a standard deviation of 0.48.
0.0664 × H2 + 16.821 × Ref. Ind. + 0.446 × N
Calculating Δ coke. Delta coke is perhaps one of the most im- According to the equations to calculate the heat balance in
portant variables in the FCC process, because it governs the heat the reactor and regenerator, neither the catalyst nor the feed
balance. Δ coke is defined as the quantity of carbon that is burned have an impact on coke yield, expressed as %FF at constant
in the regenerator. This value includes several coke types: operating conditions. Both catalyst and feed have an impact
• Catalytic coke is produced in the active sites of the on Δ coke, because the difference between regenerator and
matrix and the zeolite reactor temperature is governed by Δ coke.11
For a given reaction temperature, higher Δ coke corre-
1.1
sponds to higher regenerator temperature. Therefore, the
1.0 catalyst circulation will be reduced, and the conversion will
0.9 decrease. This is why Δ coke is so important in characterizing
Observed ∆ coke, wt%

0.8 coke selectivity of the catalyst.

0.7 FIG. 12 illustrates the concept of coke selectivity between
0.6 the two catalysts used, expressed by the calculation of Δ coke.
0.5
As mentioned earlier, all feeds were evaluated in the ACE unit
0.4
0.3 with two different metal-free catalysts. Catalyst 1, following
0.2 deactivation, has a zeolite/matrix ratio of 3.9, while, in Cata-
0.1 lyst 2, it is 1.1. These differences are due to the fact that the
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
Predicted ∆ coke, wt% two catalysts were designed for different purposes. Catalyst 2
is for high conversion of bottoms to LCO with good coke and
FIG. 11. Δ coke calculation. gas selectivity.
Although the matrix content between the two catalysts is
1.20 very different, the Δ coke difference is small. However, it is
clear that Catalyst 1 has a better coke selectivity factor that can
1.00 range between 5% and 10% less. This difference will translate
0.80
y = 0.9793x – 0.0649 into a better heat balance in the unit, i.e., lower reaction tem-
R2 = 0.9419
∆ Coke catalyst 1

perature and higher circulation. Therefore, Catalyst 1 is better

0.60 for operations that require improved flexibility in regenerator
temperatures, to process large volumes of heavy feed. This
0.40 shows that the evaluation and selection of the catalyst for an
0.20 FCCU is a very important process because it can lead to major
economic benefits.
0.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20
∆ Coke catalyst 2 Calculation of octane RON in naphtha. Octane RON is
perhaps the main property to identify FCC naphtha quality.
FIG. 12. Coke selectivity between catalysts.
Naphtha RON depends on catalyst properties (zeolite type),
48 FEBRUARY 2015 | HydrocarbonProcessing.com
 Clean Fuels

the presence of additives (ZSM-5), operating conditions (re-
action temperature) and feed type.
Octane RON = 177.864 – 0.192 × CRR – 0.234 ×
API + 0.0328 × Aromatics – 54.942 × Ref. Ind. + (17)
7.90 × N Basic + 0.229 × S
Eq. 17 shows that naphtha RON is directly related to aromat-
ic content. If the aromatic content of the feed is increased, then
gasoline yield decreases. However, the octane RON increases
due to the presence of monoaromatic HCs that, upon breaking
the lateral chains caused by cracking reactions, become part of
the FCC naphtha range. The correlation coefficient of Eq. 17
was 0.904 with a standard deviation of 0.30. The method for
the octane RON (machine) has 0.2-octane repeatability. For
the purpose of this study, the octane RON of the gasoline frac-
tion of syncrude from the ACE unit reactor was measured by
gas chromatography and calculated using proprietary software.b
According to the results, the octane RON of most of the feeds
studied ranged from 89 to 93.5 octane numbers.
Commercial plant application. Some presented examples
illustrate the applications to the real world and how this knowl-
edge can benefit the FCCU operator.
Procurement of GO. One of the applications of calculat-
ing the chemical properties of the feed and knowing potential
yield is in the procurement of GOs or combined feeds for FC-
CUs. It is a very common practice in many countries, particu-
larly when there are not enough domestic FCC feed compo-
nents available.
TABLE 4 provides an exercise carried out for an FCCU. All the
properties reported in the table are calculated using the basic
properties measured in the laboratory. It also provides a compar-
ison of the properties calculated for the VGO-Ref. and the three
bids submitted by the vendors of these products. The VGO-Ref.
represent typical properties calculated for the feed that is nor-
mally processed in the unit. This is very useful additional infor-
mation for refinery management to make better decisions.
According to this information, the best bid that will provide
the best economic value (more LPG + gasoline) for the unit is
Bid 2. Bid 2 is very similar to the unit’s typical feed, while Bid
1 is a more aromatic feed with a significant decline in valuable
products. In addition, since Bid 1 is more aromatic, the metal
and S content are higher than the reference—for instance, the
VGO-Ref. has 0.8 ppm of Ni + V, while Bid 1 has 2.1 ppm of Ni
+ V. The S of the unit’s typical feed is 1.2% and that of Bid 2 is
2.3%, which means the S content of the FCC naphtha would be
higher, as well as the addition of fresh catalyst.
Bid 1 is probably less expensive than feeds 2 and 3. How-
ever, calculating the loss in valuable products and the increase
in naphtha S and fresh catalyst additions will negatively impact
the profitability of the FCCU. Remember: This is the theoretic
yield potential of the feeds, where the deltas presented among
the different products may be higher in the commercial units
due to the operating conditions and mechanical limitations.

EFFICIENCY MATTERS.
RESTORING PLANTS QUICKLY AND SAFELY.
Managing turnarounds on time and on budget can present
many challenges. Cudd Energy Services helps you meet
these challenges head on. Our fleet of pumping, transport
and storage vessels accommodates a wide range of flow
rates for HPHT, open-flame environment, and reduces
costs plant congestion. The Queen Storage tank has
a 1.5-million CSF capacity that can safely replenish
nitrogen supplies without interrupting pumping services.
The Dual Mode Pump is EPA Tier 2/C.A.R.B rated to
reduce emissions, and features ESD devices and an
innovative heat-recovery system to reduce fuel costs.

For more information about our industrial nitrogen

solutions, visit us at www.cudd.com or call us at
832.452.2800.

3529(1(;3(5,(1&(75867('5(68/76
Š
WWW.CUDD.COM
WWW.CUDD.COM

Select 156 at www.HydrocarbonProcessing.com/RS

Hydrocarbon Processing | FEBRUARY 2015 49

Keynote Speaker:
Charles Davidson
Chairman and Chairman
and Chief Executive Officer
Noble Energy
Gina Cohen
Gas Consultant
Eastern Mediterranean
Adi Karev
Global Head Oil & Gas
and Regional Head (AP)
Energy & Resources
Deloitte
Symeon Kassianides
Chairman and Chief
Executive Officer
Hyperion Systems
Engineering Group
Register by 17 February 2015 &
SAVE 10%
EMGasConference.com
17–18 March 2015 | Nicosia, Cyprus
Discover how to Build Business
Operations in the Eastern Mediterranean
We invite you to join executives and speakers from top operators and regional governments
as they discuss key issues, project updates and current opportunities in the burgeoning
Eastern Mediterranean natural gas industry at the Eastern Mediterranean Gas Conference.
A series of recent discoveries by Noble Energy, the largest operator in the region, has
increased the estimated gross resources of natural gas awaiting development in the area
to 40 trillion cubic feet (Tcf). Industry-leading companies are preparing to increase their
regional presence. Policy is currently being made to determine how natural gas will be
exported. Cyprus, Israel, Greece and Egypt are in the middle of negotiations on the exclusive
economic zone (EEZ). Greece is also holding its Second International Licensing round for 20
blocks in the Ionian and Cretan Seas. Along with infrastructure development in the region,
experienced operators are needed to extract the gas in the Leviathan and Tamar fields. The
development of the gas industry in this emerging region not only has the potential to fulfill
regional demand, but also to impact European and Middle Eastern gas markets.
During the two-day conference, you will have numerous opportunities to network with
executives and government officials who are actively involved in developing the region’s
natural gas industry as well as learn how your company can establish business operations
in the area.
Topics to be Discussed Include:
• Updates on production • Government regulations
and exploration programs • Leasing/permitting
• LNG, FLNG, pipeline • Monetization
• Infrastructure requirements • Resource potential
• Legal issues • And more
Lead Sponsor: Gold Sponsor: Gala Dinner Sponsor: Hosted by:
Hyperion Systems
Engineering Group
Bronze Sponsor: Key Card & Welcome Bottled Water Sponsor:
Package Sponsor:
Speaker Opportunities: Contact Melissa Smith, Events Director at
+1 (713) 520-4475 or Melissa.Smith@GulfPub.com
Sponsorship/Exhibit Opportunities: Contact Lisa Zadok, Events Sales Manager
at +1 (713) 525-4632 or Lisa.Zadok@GulfPub.com
 Make your plans
to attend
Visit HPIRPC.com

1–3 June 2015 | Jumeirah at Etihad Towers | Abu Dhabi, UAE

We’re excited to announce that this year’s International Refining and Petrochemical Conference (IRPC) will be held 1-3 June 2015 in
Abu Dhabi, United Arab Emirates in conjunction with DMG Global Energy, organizers of ADIPEC.
In its sixth year, the 2015 conference and exhibition will provide a high-level business and technical forum in which the key players in the
global petrochemical and refinery sectors will meet to share knowledge and learn best practices and the latest industry advancements.
In today’s increasingly competitive and global HPI, managers and engineers are actively seeking information and solutions to make
their organizations more efficient and profitable. At IRPC 2015, over the course of three days and three tracks, attendees will hear
from leading executives at top operators and service companies regarding the technological and operating advancements that
can benefit their organization or plant. Numerous networking opportunities will be available throughout the conference. We hope
you’ll join us and be part of the discussion.
Topics to be discussed include:
• Clean fuels • Gas treatment technologies • Heavy oil
• Catalyst developments • Rotating equipment • Ethane
• Plant and refinery sustainability • Crude to petrochemicals • Plant design
• Maintenance and reliability • Alternative feedstock fuels • Gas processing including small scale,
• Energy policy • Emerging technologies modular and offshore
• Profitability • Refining/petrochemical integration • NGL
• Effluence management • Water treatment/processing • GTL
and cooling

Questions?
Please contact Melissa Smith, Global Events Director, Gulf Publishing Company, at Melissa.Smith@GulfPub.com or +1 (713) 520-4475.
Claire Pallen, Conference Director, DMG Events ME, +971 (0) 2 6970 505 or ClairePallen@dmgeventsme.com.
For sponsor and exhibit opportunities:
Americas and Europe: Lisa Zadok, Event Sales Manager at +1 (713) 525-4632 or Lisa.Zadok@GulfPub.com for information.
Asia-Pacific and Middle East: Siham Ammoura, Senior Business Development Manager, DMG Events ME, +00 971 55 7781 360
or SihamAmmoura@dmgeventsme.com
Italy: Fabio Potesta, Mediapoint & Communications SRL, +39 010 5704948 or info@mediapointsrl.it

Managed by:
Supported by: Silver Sponsor: Lanyard Sponsor:

Register early and

SAVE 15% at HPIRPC.com
 Clean Fuels

TABLE 4. Properties calculated for GO bids

new feed components, such as coker GOs, atmospheric resi-
due, HCK residue, hydrotreated feeds, and more—or changes
Calculated properties VGO-Ref. Bid 1 Bid 2 Bid 3 in component ratios of different streams, such as an increase in
Specific gravity 0.9125 0.9279 0.9042 0.9176 the residue or heavy fraction stream in the combined feed, on
MeABP, °F 776 831 752.2 834 FCCU operations. The information provided is useful for the
FCCU engineers to make their calculations and predictions
MeABP, °C 413 444 400 446
and to review the operating conditions to get the most from
Correlation index 46.3 50.6 43.8 45.6 the potential volume in the FCC feeds.
H2 content, mole% 12.8 12.5 12.9 12.8
Saturate content, mole% 52.8 49 55.3 53.5 Conclusions. The behavior of feeds and the components
thereof in FCCUs is governed by chemical composition, type
Aromatic content, mole% 41.8 45.4 39.4 41.1
and distribution of HC families. The H2 content of FCC feeds
Diaromatic content, mole% 13 15.6 11.6 12.6 is an important property to define and study. The most im-
Refractive index 1.5086 1.5153 1.5082 1.5101 portant properties to characterize an FCC feed are: density at
Maximum conversion, mole% 75.8 74 76.4 76 15°C, distillation (D-1160, SimDis), refractive index, N and S
content and CRR.
Gasoline yield (max. conv.), mole% 51.9 50.5 52.3 52
The CF factor (conversion at the maximum gasoline con-
LPG yield (max. conv.), mole% 18.7 17.7 18.9 18.8 version point) is an important property in defining FCC feed
Gasoline + LPG, mole% 70.6 68.2 71.3 70.8 quality. Δ coke is an important property to define feed impact
LCO yield (max. conv.), mole% 19.6 20.5 19.1 19.5
on heat balance and catalyst coke selectivity. The knowledge
developed in this study can be used and applied to make better
CF 73.5 69.5 75.5 74.1
decisions and optimize the FCCU.
76.0 NOTES
52.5 a
GRACE selected more than 100 types of FCC feeds from all over the world with
76.0 Maximum gasoline yields different compositions.
Objective 51.5
Maximum conversion, wt%

b
76.0 Grace software.
50.5
Gasoline, wt%

76.0 LITERATURE CITED

Gasoline unit yields
76.0
Maximum calculated
76.0 conversion
Max. conversion
76.0 Gasoline converson max.
Naphtha
76.0
FIG. 13. Theoretic yields vs. actual yields for commercial unit.
Theoretic potential yield and yield obtained. FIG. 13
provides an example where the unit’s gasoline yields are com-
pared with the theoretic yields calculated at the maximum
conversion point.
Data in FIG. 13 show that the plant’s gasoline yield decreases
from an average value of 50% to 47.5% at the end of the period.
There are many factors that can have an impact here, such as
feed quality, operating conditions (reaction temperature, cata-
lyst circulation, C/O ratio, atomization of the feed, addition of
fresh catalyst, and ECAT activity and mechanical issues. The
maximum conversion calculated from the feed properties also
decreases, which explains the decline in the unit’s gasoline yield.
Although the calculated maximum gasoline yield potential and
unit gasoline yield both decrease, there is a delta of around 2%
between calculated and achieved yields that could be poten-
tially exploited by adjusting operating conditions. Using this ex-
ample for an FCCU of 50 Mbpd, recovering 1% based on feed
quality means an increase in gasoline production by 500 bpd,
i.e., 15 Mbbl/month, which is a significant economic benefit for
the refinery and an improvement in its refining margin.
There are many other cases where the knowledge provided
in this article can be applied to commercial units. For instance,
this methodology could help in understanding the effects of
49.5 1
Kayser, J. C., US Patent 6 069 012, Versatile fluidized bed reactor, assigned to
48.5 Kayser Technology, 2000.
2
Wallenstein, D., R. H. Harding, J. R. D. Nee and L. T. Boock, “Recent advances
47.5 in the deactivation of FCC catalysts by cyclic propylene steaming (CPS) in the
presence and absence of contaminant metals,” Applied Catalysis, Vol. 204, 2000,
46.5
45.5 pp. 89–106.
3
Orlicki, D. S., U. Navarro, M. Ni and L. Langan, “Application of H-NMR for FCC
Feed Characterization,” IAPG Congress, Mendoza, Argentina, 2009.
4
Orlicki, D. S., U. Navarro, M. Ni and L. Langan, “Application of H-NMR for Fluid
Catalytic Cracking Feed Characterization,” Chap. 12, Advances in Fluid Catalytic
Cracking, Testing, Characterization and Environmental Regulations, M. L. Occelli,
Ed., CRC Press, 2010.
5
Petroleum Refining, Institute Français du Pétrole, Tome 1, Crude Oil, Petroleum
Products and Process Flowsheets, J.-P. Wauquier, ed., Editions Technip 1995.
6
Speight, J. G., “The chemistry and technology of petroleum,” Chemical Industries,
Vol. 3, Marcel Dekker Inc., 1980.
7
Magee, J. S., and M. M. Mitchell, “Fluid Catalytic Cracking: Science and
Technology,” Studies in surface science and catalysis, Vol. 76, Chapter 12, 1993.
8
Navarro, U., documento crudos, “Química del Petróleo y de los procesos de
Refinación,” Ecopetrol-Instituto Colombiano del Petróleo, 2005.
9
Fischer, I. P., and U. Navarro, “FCC Feed Characterization for two FCC´s units
of the Barrancabermeja refinery (Colombia), Petro Canada project April 1989,”
352-89-53.
10
Navarro, U. and C. Vargas, “Impacto del hidrotratamiento de cargas a FCC:
Evaluación en la unidad MAT y por espectrometría de masas,” Actas del IV sim-
posio Colombiano de Catálisis, Bucaramanga, 1996.
11
Petroleum Refining, Institute Français du Pétrole, Tome 3, Conversion Processes,
P. Leprince Ed., Editions Technip 2001.
URIEL NAVARRO, PhD chemist, is a technical manager at Grace in Latin
America. He has more than 30 years of experience in the petroleum industry in
research and refinery technical service.
MICHELLE NI, PhD, is a Sr. R&D chemist at Grace in Columbia, Maryland. She
has more than 10 years of experience in chemical characterizations focused on
spectroscopy and chromatography in specialty chemicals and material sciences.
DARIUSZ ORLICKI, PhD, is a principal engineer at Grace in Columbia,
Maryland. He has expertise in FCC catalyst evaluation and catalyst feed
interactions. Dr. Orlicki has authored over 40 scientific papers and presentations.

52 FEBRUARY 2015 | HydrocarbonProcessing.com