Identify Hydrotreating Process Variables

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 6

ART Engineers Elaborate on Innovations & Operational Issues Affecting

Hydrotreater Performance

Grace Davison and Advanced Refining Technologies (ART) FCC/Hydrotreating


Symposiums have provided refiners with a forum for in-depth discussions of important
developments and operational issues affecting FCCUs and hydrotreaters.

The need to provide technically advanced solutions and cost effective options are why
these organizations have invested in advancing refining catalyst technology and related
technical support. These include FCC catalysts, as well as hydroprocessing catalyst
systems developed through ART (Advanced Refining Technologies), a joint venture with
Chevron Products Company. Through ART, a comprehensive line of catalysts for
combined conversion processes including distillate hydrotreating, fixed bed and
ebullating bed resid hydrprocessing applications as illustrated in the accompanying
figure.

Many of todays refineries require unit-specific technical solutions that in many cases
directly affect hydrotreating profitability and efficiency. Against this backdrop of
operational concerns, key process variables affecting hydrotreating operations, include:

Weighted average bed temperature (WABT)


Liquid hourly space velocity (LHSV)
Hydrogen
H2S effects
Feed effects
Determination of H2 consumption and heat release
Pressure drop.

Importance of Temperature

Achievement of a higher WABT will result in higher removal of sulfur and nitrogen via
hydrodesulfurization (HDS) and hydrodenitrification (HDN) conversion, respectively.
Higher WABT will also result in higher aromatic or polynuclear aromatic (PNA)
saturation until the thermodynamic limit is reached.
Hydrotreating reactions are exothermic, making certain temperatures important to
specific reactions, which is why a WABT for each reactor is calculated as follows:

WABT = xi (TINi + [Ti]) EIT

Where:

WABT= weighted average bed temperature

xi = fraction of catalyst in a given bed (i) or reactor

EIT = equivalent isothermal temperature.

WABT determines the level of sulfur and nitrogen removal for HDS and HDN reactions,
while the outlet temperature determines the conversion for aromatic or PNA saturation.
The inlet temperature (TIN) is the control variable, but is not very meaningful when
discussing catalyst activity due to the exothermic nature of hydroprocessing reactions,
while T is an indication of a given feeds reactivity (at given conditions) on a given
catalyst. However, this is not a good way to compare catalysts, unless the feed and
conditions are identical.

As reactor WABT increases, product sulfur decreases. For example, when processing a
1.1 wt% sulfur LCO blend at 1.0 LHSV (where LHSV = volumetric charge rate/catalyst
volume) at1000 psig and 2100 scfb H2/Oil, product sulfur drops to approximately 50
ppm at WABT of 629F, and drops down to approximately 5 ppm sulfur at a WABT of
670F.

As a rough approximation, 1/LHSV is approximately equal to the residence time. Higher


feed rates correspond to higher LHSVs and higher product production rates, but require
higher WABTs to maintain product specifications. As residence time is reduced (i.e.,
higher LHSV as per 1/LHSV approximation), the required temperature required to
maintain the same product sulfur is also increased.

However, higher WABTs come with costs. This is because higher temperatures reduce
run length (TEOR TSOR) and increase coking (fouling). The net result is that higher
LHSVs result in shorter cycle lengths. As a rule of thumb, fouling rate is proportional to
(LHSV) 2 3.

In estimating EOR, determination of end-of-run (EOR) is highly dependent on the


process feedstock. For example, for naphtha hydrotreating, mercaptan recombination,
catalyst poisoning (e.g., silicon [Si], arsenic [As], etc.) and pressure drop determine
EOR. Catalyst poisoning (e.g., nickel [Ni], vanadium [V], Iron [Fe] and As), yield losses,
furnace and metallurgical limits determine EOR when hydrotreating VGO.

Importance of hydrogen partial pressure


Hydrogen partial pressure (H2PP) plays a significant role in hydroprocessing operations.
As a rule-of-thumb, catalyst deactivation rate is proportional to the reciprocal of the
hydrogen partial pressure factored by an exponential ranging from 2 to 3:

Catalyst deactivation rate (1/H2PP)2-3

The H2PP has a large impact on fouling rate, aromatics and PNA saturation (i.e., large
effect on product properties related to aromatics like product density, diesel, color and
cetane). H2PP also has a significant impact on required WABT for HDN and a moderate
impact on required WABT for HDS (but large in ULSD). H2PP is determined by:

Reactor design pressure


Make-up hydrogen purity
Recycle gas rate and purity (important for H2PP at the reactor outlet)
Bleed or purge rate
Hydrogen consumption
Degree of feed vaporization.

The effect of H2PP on HDS and HDN in VGO service was discussed in detail and
graphically illustrated by ART experts, as well as H2PP on HDS in ULSD service. A
20F improvement in HDS is observed with increased H2PP when processing 19.6 API
gravity and 1.81 wt% sulfur feed at 1.0 LHSV and 2500 scfb H2/oil reactor conditions.
However, a loss of 20F could occur if the unit operation is allowed to change with less
or lower quality hydrogen or lower separator pressures. In addition, a 70F improvement
in HDN is observed with increased H2PP. However, a loss of 70F could occur if the unit
operation is allowed to change with less or lower quality hydrogen or lower separator
pressures. It is important to note that for improved HDS in ULSD service, the required
WABT delta declines with increasing H2PP. For example, with a 33.0 API feedstock
and 1.66 wt% sulfur, the required WABT delta approaches 0 at 800 psig H2PP (0.75
LHSV, 3000 scfb H2/oil).

The effects of increasing gas-to-oil ratio (scfb H2/oil) raises H2PP, reduces reactor outlet
H2S concentration, increases unit activity and stability and increases reactor pressure
drop (P). The following gas rate definitions include H2 availability:

Total treat gas rate = Make-up gas rate + Recycle gas rate Quench gas rate.

Hydrogen gas rate = (Recycle gas Quench) (mol% H2 + make-up gas mol%
H2).

H2/Oil = Hydrogen gas rate/Barrels of oil (all in same units, hour or day).

H2 availability = (H2/Oil) / (Chemical H2 Consumption).

Minimum H2 availability recommendations include:


3.0 for straight-run
4.0 for cracked stocks
5.0 for straight-run (ULSD hydrotreating)
6.0 for cracked stocks (ULSD hydrotreating).

Impact of cracked stocks

Cracked stocks are typically more difficult to hydrotreat and require higher WABT
compared to straight-run (SR) material at similar product targets. Coker distillates and
gas oils have higher sulfur, nitrogen and olefins (Bromine number) with a higher heat
release from olefin saturation and higher feed sulfur. Their aromatic content is similar to
SR. However, FCC LCO has a higher total aromatics and PNA content (i.e., much lower
API vs.SR or coker) and more olefins that SR, but less than coker material. FCC LCO
also has a higher heat release due to high levels of olefin and PNA saturation. FCC
LCO also has a significant impact on reactor WABT. For example, an increase in %LCO
from 20 to 40 corresponds to a required temperature increase from 50F to 90F,
respectively. A summary of the process variable effects are summarized in Table 1.

Table 1. Process variable effects on catalyst life.

An Increase in the Following Conditions (e.g., WABT)

will Cause Product S, N or Aromatics to increase

or decrease, depending on the Specific Condition

Specific Catalyst life


Condition will or ,
Sulfur Nitrogen Aromatics depending
(other conditions on specific
constant) condition
% Cracked

Feeds
WABT
Feed Rate

(LHSV)
Hydrogen

Pressure
%h2s in

Treat Gas
As the feed end-point (EP) or final boiling (FBP) increases, this results in an increase in
sulfur, nitrogen, % hard sulfur species, aromatics and PNAs. Directionally, this
increases feed difficulty, which must be compensated for by increasing reactor
temperature.

Causes of pressure drop

There are a variety of reasons for pressure drop in hydrotreaters, including feed quality,
void fraction in the catalyst bed, liquid and vapor properties (i.e., density and viscosity)
at reaction conditions, vapor and liquid superficial velocities, coke buildup and reactor
internal problems.

In terms of feed quality affecting pressure drop, this is based on the amount and variety
of particulates and contaminants in the feed. They can include iron sulfide, rust/scales,
salts, coke fines, FCC catalyst fines and phosphorous. Poor equipment and line clean-
up can further exacerbate problems from these particulates. Iron naphthenate formed
from corrosion due naphthenic acids can cause iron to precipitate when heated and/or
reacted with H2S. Other contaminant problems can include carbon and stainless steel
sulfides from piping and furnace/exchanger tubes loosened during shutdown. Cracked
stocks can form oxidation polymers when exposed to air. These deposit on the catalyst
bed. In addition, diolefins/olefins can polymerize forming a gum on the catalyst bed.

Void fraction in the catalyst bed depends on catalyst size (diameter and length), catalyst
shape and loading method (sock vs. dense method) and handling. The catalyst or
support material breaks up during loading or operation. Careless handling and loading
and rapid heating of wet catalyst can also increase catalyst bed void fraction.

Pressure drop problems also develop due to coke buildup result from excessive
temperatures during operations or upsets that can cause coke lay down between (and
on) catalyst pellets, as well as from heavy ends carry-over in feed.

Pressure drop from reactor internal problems may be traced back to torn support
screens allowing catalyst into internals and plugging of support screens by corrosion
products or fines. Plugging of the reactor outlet collector can be another reason for
pressure drop.

Pressure drop buildup can be mitigated by feed filtration to remove particulates (at least
25 microns). It is important to ensure internals are assembled properly and remain clean
during loading. Exposure of feedstocks (especially cracked stocks) to air in tankage
should be minimized. Size-graded catalyst loading should be used for the top bed.

General Rules of Thumb Concerning Hydrotreater Reactors

Hydrogen Consumption:

Each API increase results in approximately 100 scfb H2 consumption


Each increase in Cetane number (1.0) leads to approximately 100 scfb
H2 consumption in diesel

Unless the feed is very high in sulfur, most H2 consumption comes from either
PNA saturation or olefins
Removal of contaminants (Si, Fe, Na, Ni, V, etc.) that deposits at 1.0 LHSV for
one year will deposit on average across all the catalyst, about 1.0% of the
catalyst weight. In addition, the top of the catalyst bed can easily collect 3-to-5
times the average for more reactive deposited materials and lead to plugging
Most efficient systems exhibit SOR Ps of 0.5 to 1.0 psi/ft of catalyst bed in
trickle flow. Good gas-liquid distribution is a must, and top bed size, void fraction
and activity grading is a requirement.

You might also like