Accepted Manuscript

A novel absorption process for small-scale natural gas dew point control and
dehydration

Marcelo Díaz Rincón, Carlos Jiménez-Junca, Carlos Roa Duarte

PII: S1875-5100(16)30016-6
DOI: 10.1016/j.jngse.2016.01.016
Reference: JNGSE 1220

To appear in: Journal of Natural Gas Science and Engineering

Received Date: 5 November 2015

Revised Date: 6 January 2016
Accepted Date: 9 January 2016

Please cite this article as: Rincón, M.D., Jiménez-Junca, C., Duarte, C.R., A novel absorption process
for small-scale natural gas dew point control and dehydration, Journal of Natural Gas Science &
Engineering (2016), doi: 10.1016/j.jngse.2016.01.016.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

1 A novel absorption process for small-scale natural

2 gas dew point control and dehydration

3 Marcelo Díaz Rincón1,2, Carlos Jiménez-Junca1* and Carlos Roa Duarte3

PT
4 * Corresponding author: Carlos Jimenez-Junca, e-mail: carlosjj@unisabana.edu.co, tel: +57

5 320 845 45 27, Fax: +571 861 55 55

RI
1
6 Faculty of Engineering, Universidad de la Sabana, Campus Puente el Común, Km.7

SC
7 Autopista Norte de Bogotá, Chía 250001, Colombia

2
8 Vicepresidencia Técnica de Producción, Ecopetrol, Cr 13 No. 36–24, Bogotá 110311,

9 Colombia
U
AN
3
10 Colombian Petroleum Institute-ICP, Ecopetrol, Vía Pie de Cuesta Km.7, Piedecuesta
M

11 681011, Colombia
D

12 Abstract
TE

13 Heavy hydrocarbon removal is one of the processes found in natural gas processing plants

14 for reducing heating value and hydrocarbon dew point to sales gas quality. Conventional
EP

15 technologies include Joule-Thomson expansion, turboexpansion, mechanical refrigeration,

16 membrane separation and supersonic centrifugal separation. In this paper, a novel

C

17 absorption process system is presented for dew point control of natural gas streams
AC

18 associated to crude oil production. The system is based on simultaneous absorption of

19 heavy hydrocarbons and water from natural gas using a mixture of TEG and lean oil. This

20 alternative technology has an equipment count, investment and operating cost, and energy

21 consumption lower than commercial technologies available nowadays. Natural gas

22 processes are modeled and simulated within Aspen Plus® simulator with a common set of

23 operating criteria and adjusted thermodynamic interaction parameters for the absorption

24 process, based on experimental data. An energy balance and economic assessment for
 ACCEPTED MANUSCRIPT
25 each process option are developed considering investment and operating cost and

26 profitability analysis. Results indicate an increase in hydrocarbon absorption efficiency with

27 increase in lean oil concentration; 46% of pentane plus components can be removed with a

28 20% of lean oil. Energy consumption is up to 52% lower than other process schemes, while

29 capital cost and operating cost are also 40% and 25% lower. Process description, operating

30 conditions, equipment requirements, recovery efficiency and limitations for the new

PT
31 absorption process are presented.

RI
32 Keywords: natural gas; absorption; process simulation; dew point control; triethylene glycol;

33 lean oil

SC
34 1. Introduction

35
U
Natural gas associated to crude oil production usually requires treatment for conditioning to
AN
36 sales gas specifications or fuel gas quality. Typically, those specifications include water

37 content, hydrocarbon dew point, heating value, carbon dioxide and hydrogen sulphide
M

38 (Galatro & Marin, 2014; Ghiasi, et al., 2014; Rufford, et al., 2012). Processes required to

39 meet sales gas specifications usually include compression, dehydration, sweetening

D

40 (removal of H2S and CO2) and hydrocarbon dew point control. These are performed in
TE

41 different sequential stages that involve several unit operations, equipment, instrumentation,

42 piping, etc. Figure 1 shows a typical block flow diagram for a gas treatment plant.
EP

43 Dehydration processes include glycol absorption (Netusil & Ditl, 2011), polymer membranes

44 (Scholes et al., 2012), composite membranes (Lin et al., 2012), molecular sieves (Farag et
C

45 al., 2011), and isenthalpic gas cooling with controlled hydrate formation (Parks & Amin,
AC

46 2012). Glycol absorption is the dehydration method used in the majority of existing natural

47 gas treatments plants (Bahadori & Vuthaluru, 2009; Twu, et al., 2005).

48 Triethylene glycol (TEG) is the most commonly used solvent for natural gas dehydration

49 since it has high hygroscopic, low solubility in natural gas and low vapour pressure

50 (Bahadori et al., 2008). Monoethylene (MEG) and diethylene glycol (MEG) can also be used

51 for dehydration applications (Derawi, et al., 2003) but are seldom considered for dehydration
 ACCEPTED MANUSCRIPT
52 due to difficulty to reach water content specification for pipeline. Moreover, evaporation loss

53 and thermal degradation in the regeneration system are normally lower for TEG.

54 TEG absorbs more light hydrocarbons and acid gases than EG and DEG (Bahadori &

55 Zeidani, 2006; Bahadori et al., 2008) and is considered an undesired effect in existing plants

56 because liquid hydrocarbon promotes foaming tendency in absorber columns and clogging

PT
57 of equipment that operates at high temperature in the regeneration system (Manning &

58 Thompson, 1991). However, operation experience in existing plants shows that absorption

RI
59 level is not sufficient to reduce the heavy hydrocarbon content to sales gas quality.

60 Investigations have been performed for equilibria of glycols with water (Twu et al., 2005),

SC
61 light alcane hydrocarbons (Derawi et al., 2003; Bahadori et al., 2008; Ng et al., 1993; Al-Saifi

62 et al., 2008), aromatic hydrocarbons usually present in natural gas (Moshfeghian & Hubbard,

U
63 2012; Ng et al., 1993), and acid gases such as CO2 and H2S (Afzal et al., 2012). Even
AN
64 potassium formeate as an additive to enhance water absorption efficiency on TEG has been

65 studied (Isa et al., 2013). However, absorption of heavy hydrocarbons in TEG was not
M

66 correlated in aforementioned researches.

67 Dew point control is usually included in a gas processing plant to produce a natural gas
D

68 within specifications or maximize hydrocarbon liquid recovery (Mehrpooya, et al., 2015)

TE

69 depending on the production objective. Desired degree of recovery determines process

70 technology, its complexity and investment and operating costs. Conventional methods for
EP

71 dew point control of natural gas include mechanical refrigeration, isenthalpic expansion

72 (Joule-Thomson effect), turboexpansion, lean oil absorption, solid bed adsorption,

C

73 membrane separation and supersonic centrifugal separation (Mokhatab, 2012). All these
AC

74 treatment process have high energy consumption and investment and operating costs, and

75 several investigations have been conducted to reduce them with variations on process

76 configuration and conditions (Fissore & Sokeipirim, 2011; Getu, et al., 2013; He & Ju, 2014;

77 Machado, et al., 2012; Mehrpooya, et al., 2010; Shin, et al., 2015; Vatani, et al., 2013);

78 capital investment for a new natural gas processing plant is mainly influenced by the gas

79 compression and dew-point control equipment. Then, natural gas industry is always aiming

80 to simpler process configurations and lower energy requirements and costs (Khorsand &

81 Maleki, 2012). Investment and operating costs are usually paid back with the revenue
 ACCEPTED MANUSCRIPT
82 associated to sales gas and stabilized hydrocarbon condensates that are obtained from the

83 treatment process, usually named natural gas liquids or NGL (Mehrpooya, et al., 2015; Shin,

84 et al., 2015). However, heavy crude oil fields have low associated gas production and

85 treatment for pipeline or fuel gas is not economically feasible. Also, the NGL production

86 depends on heavy hydrocarbon content of gas to be processed (Getu, et al., 2012). Hence,

87 in some cases recovery of NGL is economical only for rich natural gases with a high content

PT
88 of heavy hydrocarbons.

RI
89 Lean oil absorption process has been commercially used for hydrocarbon dew point control

90 for natural gas treatment. Lean oil is usually a mixture of hydrocarbons with high molecular

SC
91 weight with affinity to absorption of heavy hydrocarbons that are present in natural gas

92 (Campbell, 2004). Lean oils usually have a molecular weight 100-200 g/mol to reduce losses

U
93 in heating operations in regeneration systems (Gas Processors Suppliers Association,
AN
94 2004), can absorb at ambient temperature and high pressure (Mokhatab, 2012) and have an

95 initial boiling point usually higher than 150 °C. S uch conditions are similar to gas dehydration
M

96 process with TEG. Then, when using a mixture of TEG-lean oil, simultaneous removal of

97 water and heavy hydrocarbons from natural gas stream in a single absorption column is
D

98 expected, with a subsequent reduction in capital and operating costs.

TE

99 In this paper, simultaneous dehydration and dew point control of natural gas using a mixture

100 of TEG and lean oil is presented as an alternative technology with an equipment count,
EP

101 investment and operating cost, and energy consumption lower than existing technologies.

102 This process can achieve both water and hydrocarbon dew point specifications in one unit.
C

103 Operating conditions, equipment requirements, recovery efficiency and limitations for the
AC

104 conventional technologies are compared with this novel absorption process. A process

105 description of this new process is included.

 ACCEPTED MANUSCRIPT
106 2. Materials and methods

107 2.1 Pilot experimental tests

108 Experimental tests for evaluation of absorption efficiency of TEG-lean oil absorbent were

109 conducted with a pilot scale unit. The unit was installed at an oil field where gas from crude

PT
110 oil production was available. Figure 2 shows flow diagram and process conditions during

111 experiments. Main equipment characteristics of the pilot scale unit are as follows:

RI
112 • Absorption column: packed type, carbon steel, design pressure 100.3 bar, 4.9 m height,

113 16.8 cm diameter.

SC
114 • Regeneration column: packed type, carbon steel, design pressure 4.5 bar.

115

U
• Absorbent pump, energy exchange type, Kimray model 4015 PV, max. discharge pressure
AN
116 107.9 bar.

117 • Reboiler, fired-tube type, carbon steel, max. heat duty 22.0 kW, max. temperature 216 °C.
M

118 • Inlet scrubber, two-phase type, carbon steel, design pressure 100.3 bar, 1.5 m height, 16.8
D

119 cm diameter.
TE

120 • Flash tank, three-phase, carbon steel, design pressure 9.6 bar, 1.2 m length, 45.7 cm

121 diameter.
EP

122 • Water content meter, Panametrics model PM 880.

C

123 • Orifice plate gas flow meter, w/ multivariable flow computer Rosemount 205.
AC

124 • Cromatograph Agilent 6890 w/ FID-TCD detector, software Natural Gas Expert V2.6, under

125 GPA standard 2286.

126 Absorbent concentration was changed ranging 0-20% vol. of lean oil. Inlet and outlet gas

127 was sampled and composition was determined through gas extended chromatography.

128 Tables 1 and 3 show composition of inlet gas and lean oil-
 ACCEPTED MANUSCRIPT
129 2.1. Process simulation

130 TEG-lean oil (TEG-LO) absorption for natural gas dew point control and dehydration was

131 compared to a conventional Joule-Thomson (JT) process and mechanical refrigeration (MR)

132 process. A conventional TEG absorption process for dehydration was coupled upstream to

133 JT and MR units, to avoid condensation of water and hydrate formation. Material and energy

PT
134 balances were developed using process simulator Aspen Plus® (Aspentech) to assess the

135 performance of the process alternatives. The flowrate and composition of the inlet gas

RI
136 processed is reported in Table 1. The gas is saturated at inlet conditions of 52.7 bar and

137 47.5 °C. It was assumed the specifications for natu ral gas reported in Table 2, which are

SC
138 applicable to Colombian market regulations. Due to its potential for fields where gas

139 production is low and processing is not economically feasible, capacity for TEG-LO process

U
140 is set as 11800 m3/h at standard conditions, equivalent to 10 million of standard cubic feet
AN
141 per day. For comparison, the same calculation basis was chosen for TEG + JT and TEG +

142 MR options. Table 3 shows the composition used for lean oil.
M

143 Thermodynamic model is one of the key parameters in simulation of chemical processes and

144 can influence the results significantly. Equations of state (EoS) are the models
D

145 recommended for simulation of natural gas processes involving hydrocarbons and low
TE

146 polarity components (Carlson, 1996). Soave-Redlich-Kwong (SRK) and Peng-Robinson (PR)

147 are the EoS most commonly used for natural gas, with some modifications like the Twu-Sim-
EP

148 Tassone’s version of SRK that improves the estimation of dehydration processes (Twu, et

149 al., 2005). RK-ASPEN model in Aspen Plus® was selected in this study; this model is based
C

150 on SRK EoS and is applicable to polar compounds, like alcohols and water, with

151 hydrocarbon mixtures (Aspentech, 2011). Binary parameters for thermodynamic model were
AC

152 adjusted based on experimental data for water, inert components and hydrocarbon

153 components absorption in liquid phase; experiment was conducted with a pilot absorption

154 column using real operation conditions in a gas field. A sensitivity analysis was conducted on

155 absorption column to determine absorption efficiency a function of absorbent concentration,

156 pressure and temperature.

 ACCEPTED MANUSCRIPT
157 Plant energy consumption was estimated considering gas compression, pumping of liquids,

158 refrigerant compression, and heating of reboilers. Efficiencies are assumed to be 75% for

159 compressors and 65% for pumps. Utilities consumption includes absorbent make up and

160 drain, condensate storage and pumping, instrument air, flare and residual water systems.

161 Heat integration exchangers between process streams in dehydration and dew point control

162 systems were also included. For air-cooled exchangers, such as compressor discharge

PT
163 coolers, the power for electric motors was added to the corresponding compressor electrical

164 load.

RI
165 2.2. Equipment sizing

SC
166 Sizing of process and utilities equipment was performed according to methods and heuristics

167 described in chemical engineering literature. Flows, operating conditions and properties

168
U
were obtained from process simulations. Two and three-phase separators were sized based
AN
169 on terminal velocity estimated from Stoke law’s (Stewart & Arnold, 2008). Absorber column

170 diameter was calculated using Souders-Brown’s equation for maximum surface velocity
M

171 (Branan, 2005) and column height was obtained using heuristics for tray-to-tray, bottom and

172 top spacing (Manning & Thompson, 1991; Seader & Henley, 2011). Area for heat
D

173 exchangers was estimated using recommended values for heat transfer coefficients (Seider,
TE

174 et al., 2009); heat duties were obtained from simulations. Theoretical power for compressor

175 and pumps was extracted from process simulations and efficiencies were applied to consider
EP

176 a more realistic performance. Several heuristics from aforementioned and additional

177 literature were applied to configure process schemes and complete sizing criteria
C

178 (Mokhatab, 2012; Turton & Bailie, 2009; Wang & Economides, 2009).
AC

179 2.3. Economic analysis

180 Process alternatives were evaluated based on capital costs, operational costs and

181 profitability measures. Capital cost estimation includes main process equipment such as

182 vessels, columns, compressors, pumps, heat exchangers and fired heaters. Utilities systems

183 were also included: absorbent make up and drain, condensate storage and drain, instrument

184 air, flare, residual water system, fire detection and fighting, and electrical generation.

185 Estimation corresponds to a study or feasibility estimate based on dimensions and operating
 ACCEPTED MANUSCRIPT
186 conditions of major equipment with an expected accuracy between +30% and -30% (Peters

187 & Timmerhaus, 2004).

188 Cost of equipment was estimated using data bases and budgetary bids from vendors, with

189 prices updated to 2015; transportation and freight costs to plant site were also included.

190 Construction and material costs include land purchase, site development, concrete, bases,

PT
191 equipment installation, piping, platforms, supports, stairways, electrical, instrumentation,

192 control system, buildings (control room, warehouse, maintenance shop, laboratory), painting,

RI
193 insulation, tools, work force, supervision and contractor’s fee. All construction costs were

194 estimated based on databases for similar projects in oil and gas industry and factors

SC
195 described in literature (Peters & Timmerhaus, 2004; Turton & Bailie, 2009). Indirect costs

196 include studies, engineering, project management, contractual and purchase planning,

U
197 permits, commissioning and start up. A typical contingency of 10% was added as an industry
AN
198 practice.

199 Operation costs were calculated for a full year of operation and are a function of utilities,
M

200 chemical labour, maintenance and overhead costs. The plants were considered to operate

201 continuously, with 330 days of operation and a two-week shutdown for maintenance. For
D

202 direct wages and benefits, 3 operators for the plant and 5 shifts for each operator were set
TE

203 (Seider, et al., 2009). Utilities and chemicals include electricity, raw water, potable water, fuel

204 gas, glycol and lean oil make-up, refrigerant and lubricants. Maintenance costs consider
EP

205 salaries and benefits for technical and supervisors, spare parts and overhaul expenses.

206 An economic analysis was performed in this study to compare to conventional processes
C

207 considering capital and operation costs, realistic market prices and taxation, discount rate,
AC

208 etc. The discounted cash flows were calculated for 10 years of plant’s operation, and the

209 construction period was assumed to be 1.5 years; then, second year considers six months of

210 operation. The total capital investment is distributed considering engineering and purchase

211 of major equipment at year 1, and construction and start-up at year 2. The royalties were set

212 as 8.5% and taxation annual rate was 33% based on gross income. All these values reflect

213 Colombian market reality. For revenue estimation, the price of natural gas is a function of its
 ACCEPTED MANUSCRIPT
214 heating value and was set as US$4.0/GJ. Depreciation calculation was based on straight-

215 line model over 10 years.

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
216 3. Results and discussion

217 3.1. TEG-LO absorption process

218 Absorption was carried out through a column where a TEG-lean oil solution flows in counter

219 current to the gas stream. Figure 3 shows experimental and simulated absorption of pentane

PT
220 and heavier hydrocarbons (C5+) based on component inlet flow as a function of lean oil

221 concentration in regenerated absorbent. The simulated influence of absorption pressure and

RI
222 inlet gas temperature is given in Figures 4 and 5. TEG-LO absorption reduces maximum

SC
223 hydrocarbon dew point of natural gas below market specification of 7.2 °C; dehydration is

224 also met. For the inlet gas composition, an absorbent with 20% vol. of lean oil absorbs 46%

225 of C5+ components in experiment. According to the simulation model, absorption of 95% of

226
U
C5+ can be expected with a concentration of 65% lean oil when extrapolating with adjusted
AN
227 parameters. Pressure for maximum absorption ranges from 20 to 50 bar. Moreover,

228 absorption increases with the decrease of gas temperature. For the sensitivity performed,
M

229 below 50 °C inlet gas reaches its dew point tempera ture and condensation effect is higher

230 than absorption. No significant influence was detected with change on inlet absorbent
D

231 temperature.
TE

232 Figure 6 shows the flowsheet for the complete simulated TEG-LO process that includes
EP

233 absorption and regeneration steps. Feed gas from well enters to a three-phase inlet

234 separator V-101 to remove free water and condensate. The saturated gas pass through a
C

235 scrubber V-201 to remove liquid droplets not removed in the separator to minimize
AC

236 contamination of absorbent. Gas enters to the bottom of the absorber column T-201 and the

237 TEG-lean oil solution is fed to absorb heavy hydrocarbons and water. The column operates

238 at 52.7 bar and the outlet gas stream temperature is 52.8 °C. The lean absorbent stream is

239 fed to the top of the absorber and a rich absorbent stream is recovered at the bottom.
 ACCEPTED MANUSCRIPT
240 The rich absorbent contains dissolved hydrocarbons and water, and pressure is reduced to

241 4.1 bar through an energy exchange pump P-201 to deliver pressure energy to a lean

242 regenerated absorbent stream. The solution passes through a particle filter and a lean/rich

243 absorbent heat exchanger E-201 where is preheated to 115.6 °C using a hot absorbent

244 stream. Heating of absorbent cause vaporization of a fraction of absorbed hydrocarbons and

245 separation is accomplished in the flash separator V-202, where pressure is reduced to 3.3

PT
246 bar. Disengaged hydrocarbons consist mainly of light hydrocarbons, 75% mol from ethane to

247 n-pentane. Absorbent is fed to the top of the regeneration column T-202 where liquid flows

RI
248 through a packed bed to increase contact area with a vapour stream coming from bottom

SC
249 reboiler. The reboiler operates at 201 °C and near atmospheric pressure (1.15 bar). Water

250 and heavy hydrocarbons are removed with the overhead vapour stream. As a result, the

251 TEG-lean oil solution has a purity of 99.3% mol. Absorbent flows to the lean/rich absorbent

252
U
exchanger to be cooled to 128.8 °C and is pumped to a 52.7 bar pressure at which enters a
AN
253 gas/absorbent exchanger to be cooled down to 54.4 °C with further energy recovery.

254 Regenerated absorbent is continuously recirculated to the main column to minimize make-up
M

255 of fresh absorbent.

D

256 Vapour stream from top of regeneration column consists of a mixture of water (17% mol),
TE

257 light hydrocarbons (C1-C5, 43%) and heavy hydrocarbons (C6+, 40%); it is cooled down to

258 43.3 °C through an air-cooler E-203, condensing 55% of the mass entering the equipment.
EP

259 Then, remaining vapour is removed in a three-phase separator V-205; a stream of liquid

260 stabilized hydrocarbon is recovered with a flow of 340 L/h.

C
AC

261 Gas from absorber column passes through a scrubber V-401 and is compressed to 83.8 bar

262 to be delivered to pipeline. The compression increases the temperature to 98.2 °C and the

263 gas is cooled down to 48.9 °C. At this point, hydro carbon dew point is 6.7 °C and water

264 content is 52.5 mg/m3, which are consistent with specifications.

265 3.2. TEG + JT process

 ACCEPTED MANUSCRIPT
266 Figure 7 shows the flowsheet for the conventional TEG absorption for dehydration and JT

267 process for dew point control. The gas flow, composition, equipment and process conditions

268 at the inlet of the plant are similar to those described for TEG-LO process. Also, TEG

269 dehydration includes absorption and regeneration steps with the same pressure conditions.

270 For lean/rich TEG exchanger, temperature is 115.6 °C at rich TEG outlet and 139.7 °C at

271 lean TEG outlet. Regeneration column operates at 144.4 °C on top and 199.0 °C on bottom.

PT
272 Condensate recovery from top stream was not included due to low absorption of

273 hydrocarbons on TEG.

RI
274 Gas from absorber column is compressed to 83.8 bar and 94.8 °C, and air cooled down to

SC
275 43.3 °C. The gas is split and cooled to 22.2 °C in two heat integration exchangers E-302/303

276 before entering the Joule-Thomson valve. Temperature drops to -4.2 °C due to isenthalpic

U
277 expansion to 28.6 bar. Condensed hydrocarbons are recovered in a low temperature
AN
278 separator V-303; water content is reduced far below specification to 26.7 mg/m3 to avoid

279 condensation and hydrate formation at JT temperature condition. Cold gas and condensate
M

280 are used for cooling in heat integration exchangers. The gas leaves the JT unit at 29.4 °C

281 and 28.3 bar, and is compressed to delivery pressure (83.8 bar).
D

282 3.3. TEG + MR process

TE

283 Figure 8 shows the flowsheet used for the conventional TEG absorption to achieve water

284 content and MR process to adjust hydrocarbon dew point. TEG dehydration has same
EP

285 configuration and operation conditions of TEG + JT process and water content is reduced to

286 26.0 mg/m3 avoiding hydrate formation at MR unit conditions.

C
AC

287 Gas from absorber column is not required to be compressed for MR unit conditions. First,

288 the gas is split and cooled down to 21.1 °C in two heat integration exchangers E-302/303.

289 Then, gas passes through a refrigerant chiller and is cooled down to 1.7 °C. Utility refrigerant

290 is a stream of 4038.7 kg/h consisting of a mixture of ethane (1.8% mol), propane (97.3%)

291 and i-butane (0.9%). The refrigerant loop is a simple compression-cooling-expansion

292 process where compression pressure is 18.3 bar and expansion pressure is 2.1 bar.

293 Refrigerant temperature at chiller is -25.1 °C and an air-cooler at 43.3 °C is used as heat

294 sink. Process gas and condensed hydrocarbons are separated in a low temperature
 ACCEPTED MANUSCRIPT
295 separator V-303. Cold gas and condensate are used for cooling in heat integration

296 exchangers. In this process, the gas leaves the MR unit at 41.0 °C and 51.7 bar, and is also

297 compressed to delivery pressure (83.8 bar).

298 3.4. Energy analysis

299 Tables 4 and 5 show a comparison of energy consumed and recovered by the processes.

PT
300 The higher energy requirement in all processes is for gas and refrigerant compression.

301 Energy consumption for sales gas compression ranges from 200 to 500 kW that is similar to

RI
302 estimations performed in previous studies when the capacity is scaled down (Getu et al.,

303 2013; Shin et al., 2015). Process gas compression for TEG + JT process consumes 2.1-2.5

SC
304 times more energy than TEG + MR and TEG-LO processes. Due to operation temperatures

305 below ambient, TEG + JT and TEG + MR process allows higher heat integration between

306
U
cold and hot streams in the dew point unit. TEG-LO process requires 58% more heating
AN
307 requirements because a higher regeneration temperature and more hydrocarbons to be

308 evaporated in the regeneration reboiler. However, the total energy consumption for TEG-LO
M

309 process is 21.5% lower than TEG + MR process and 52.3% lower than TEG + JT process,

310 due to absorption is carried out at ambient temperature and conventional dew point
D

311 processes requires low temperature produced by isenthalpic expansion or refrigeration.

TE

312 TEG-LO process consumes less energy for water removal from TEG because conventional

313 processes require higher water dew point depression to avoid water condensation and
EP

314 hydrate formation at low temperature that could plug lines, valves and equipment.

315 3.5. Economic analysis

C

316 The economic analysis results for process alternatives described previously are given in
AC

317 Table 6 and the cumulative discounted cash flow is given in Figure 9. The investment cost

318 saving is 35-40% for TEG-LO compared to TEG + JT and TEG + MR, mainly due to less

319 compression and refrigeration requirements. TEG-LO process has up to 25% less operation

320 costs because lower energy consumption and maintenance costs in critical equipment such

321 as compressors. Due to the more selective absorption of heavy hydrocarbons and

322 subsequent higher flow of product gas, the income and net profit of TEG-LO is much higher

323 than that of other options. The reduction of gas flow from inlet to outlet is 1.5% for TEG-LO
 ACCEPTED MANUSCRIPT
324 process, while a reduction of 9.4% was calculated for TEG + JT process and 13.6% for TEG

325 + MR process, due mainly to a low absorption of C3 and C4 compared to the condensation

326 of these components in JT and MR processes. Then, according to Table 1, C1 and C5+

327 concentration is lower for TEG-LO process, and the C2 to C4 concentration is higher. The

328 difference of sales gas composition between alternatives was taken into account using an

329 energy-based price instead of volume-based. The income considers only the sale of gas

PT
330 product; due to higher hydrocarbon liquid recovery, TEG + JT and TEG + MR options could

331 be more favourable when considering the sale of liquid products also. However, by-products

RI
332 from natural gas processing, such as NGL and LPG, usually follow crude oil prices. The net

SC
333 present value indicates that benefit is 3 times and investor’s rate of return doubles the

334 calculated values for conventional processes. The investment can be paid back in a half

335 period of time. All the profitability measures could be improved if a higher capacity plant is

336
U
considered. For a scenario of maximum sales gas production within specifications, TEG-LO
AN
337 process is better than the other process analysed.
M

338 4. Conclusions
D

339 In this study the TEG-LO novel process for simultaneous dehydration and dew point control

340 of natural gas by absorption is presented. This process is carried out in a single column and
TE

341 has a lower pressure drop than conventional dew point control systems, with a subsequent

342 reduction in compression energy consumption. Moreover, the absorbent regeneration

EP

343 system has a high energy recovery: lean absorbent is recompressed using pressure energy

344 released from the expansion of rich absorbent through a pressure exchange pump, and
C

345 lean/rich absorbent heat exchanger allows heat integration that reduces reboiler fuel
AC

346 consumption. All regeneration equipment operates at near-atmospheric pressure, so

347 equipment design conditions are less severe than conventional dew point control systems.

348 Also, TEG-LO process operates at ambient temperature in the gas flow path and eliminates

349 the need for low temperature materials for equipment. Hydrocarbon absorption efficiency

350 increases with lean oil concentration in absorbent. Hydrocarbon- and water- dew points can

351 be met with a concentration of 20% vol. and operating conditions that are typical in existing
 ACCEPTED MANUSCRIPT
352 natural gas processing plants. Hence, this absorption process has the advantage of

353 operation simplicity because there are a few control variables.

354 Conventional dew point control technologies such as TEG + JT and TEG + MR processes

355 require high investment cost in specialized equipment and high operating cost. The TEG-LO

356 process is an alternative dehydration and dew point control process configuration with a

PT
357 reduced number of major equipment, and lower investment costs, operating costs and

358 energy consumption. Also, the gas flow and heating value reduction due to extraction of

RI
359 heavy hydrocarbons is much lower. Hence, the economic performance of TEG-LO is better

360 compared to conventional technologies with a half of payback period. Even, the investment

SC
361 for TEG-LO process can be paid back earlier if NGL production is considered due to added

362 revenue. Then, this process could drive gas processing projects that were not feasible with

U
363 conventional technologies due to low sales prices of gas derived products (natural gas, LPG,
AN
364 NGL) and where gas is currently flared.

365 The results of this study indicate that use of TEG-Lean oil mixtures in absorption systems is
M

366 a potential alternative to adjust natural gas dew point and heat value within sales gas

367 specifications, for domestic and fuel gas use. Moreover, operating conditions are similar to
D

368 dehydration systems that are currently in operations in several gas fields in Colombia. Then,
TE

369 few modifications are required on a conventional dehydration unit to add a hydrocarbon

370 absorption feature. This represents an opportunity to reduce the investment and operating
EP

371 costs due to integration of two process steps that are typically performed separately.
C

372 Acknowledgement
AC

373 This research was supported by Ecopetrol (Colombia’s national oil & gas company).

374 References

375 Afzal, W. et al., 2012. Experimental study and phase equilibrium modeling of systems

376 containing acid gas and glycol. Fluid Phase Equilibria, Volume 318, pp. 40-50.
 ACCEPTED MANUSCRIPT
377 Al-Saifi, N. M., Hamad, E. Z. & Englezos, P., 2008. Prediction of vapor-liquid equilibrium in

378 water-alcohol-hydrocarbon systems with the dipolar perturbed-chain SAFT. Fluid Phase

379 Equilibria, 271(1-2), pp. 82-93.

380 Aspentech, 2011. Aspen Plus User Guides V7.3. Burlington (MA).

381 Bahadori, A. & Vuthaluru, H., 2009. Rapid estimation of equilibrium water dew point of

PT
382 natural gas in TEG dehydration systems. Journal of Natural Gas Science and Engineering,

383 Volumen 1, pp. 68-71.

RI
384 Bahadori, A., Vuthluru, H. B. & Mokhatab, S., 2008. Analyzing solubility of acid gas and light

SC
385 alkanes in triethylene glycol. Journal of Natural Gas Chemistry, Volume 17, pp. 51-58.

386 Bahadori, A. & Zeidani, K., 2006. New equations estimate acid-gas solubility in TEG. Oil &

387 Gas Journal, pp. 55-60.

U
AN
388 Branan, C., 2005. Rules of Thumb for Chemical Engineers. 4th ed. Burlington, MA: Elsevier.
M

389 Campbell, J., 2004. Gas Conditioning and Processing. 8th ed. Norman: Campbell Petroleum

390 Series.
D
TE

391 Carlson, E. C., 1996. Don't Gamble with Physical Properties for Simulations. Chemical

392 Engineering Progress, October .pp. 35-46.

EP

393 Derawi, S. O., Michelsen, M. L., Kontogeorgis, G. M. & Stenby, E. H., 2003. Application of

394 the CPA equation of state to glycol/hydrocarbons liquid-liquid equilibria. Fluid Phase
C

395 Equilibria, pp. 163-184.

AC

396 Farag, H. A., Ezzat, M., Amer, H. & Nashed, A. W., 2011. Natural gas dehydration by

397 dessicant materials. Alexandria Engineering Journal, Volume 50, pp. 431-439.

398 Fissore, D. & Sokeipirim, D., 2011. Simulation and energy consumption analysis of a

399 propane plus recovery plant from natural gas. Fuel Processing Technology, Volume 92, pp.

400 656-662.
 ACCEPTED MANUSCRIPT
401 Galatro, D. & Marin, F., 2014. Considerations for the dew point calculation in rich natural

402 gas. Journal of Natural Gas Science and Engineering, Volume 18, pp. 112-119.

403 Gas Processors Suppliers Association, 2004. GPSA Engineering Data Book, FPS Version.

404 Tulsa (OK): GPSA.

405 Getu, M., Khan, M. S., Long, N. V. D. & Lee, M., 2012. Studying the effect of feed

PT
406 composition variation on typical natural gas liquid (NGL) recovery processes. Computer

407 Aided Chemical Engineering, Volume 31, pp. 405-409.

RI
408 Getu, M., Mahadzir, S., Van Duc Long, N. & Lee, M., 2013. Techno-economic analysis of

SC
409 potential natural gas liquid (NGL) recovery processes under variations of feed compositions.

410 Chemical Engineering Research and Design, 91(7), pp. 1272-1283.

411
U
Ghiasi, M. M., Bahadori, A. & Zendehboudi, S., 2014. Estimation of triethylene glycol (TEG)
AN
412 purity in natural gas dehydration units using fuzzy neural network. Journal of Natural Gas

413 Science and Engineering, Volume 17, pp. 26-32.

M

414 He, T. & Ju, Y., 2014. A novel process for small-scale pipeline natural gas liquefaction.
D

415 Applied Energy, Volume 115, pp. 17-24.

TE

416 Isa, M. A., Eldemerdash, U. & Nasrifar, K., 2013. Evaluation of potassium formate as a

417 potential modifier of TEG for high performance natural gas dehydration. Chemical
EP

418 Engineering Research and Design, pp. 1731-1738

419 Khorsand, M. & Maleki, M., 2012. Select optimal schemes for gas processing plants.
C

420 Hydrocarbon Processing, Volume 91, pp. 61-64.

AC

421 Lin, H. y otros, 2012. Dehydration of natural gas using membranes. Part I: Composite

422 membranes. Journal of Membrane Science, Volume 413-414, pp. 70-81.

423 Machado, P. B. y otros, 2012. Supersonic separation in onshore natural gas dew point plant.

424 Journal of Natural Gas Science and Engineering, Volume 6, pp. 43-49.
 ACCEPTED MANUSCRIPT
425 Manning, F. S. & Thompson, R. E., 1991. Oilfield Processing of Petroleum. Volume 1:

426 Natural Gas. Tulsa: PennWell Books.

427 Mehrpooya, M., Vatani, A. & Ali Mousavain, S., 2010. Introducing a novel integrated NGL

428 recovery process configuration (with a self-refrigeration system (open-closed cycle)) with

429 minimum energy requirement. Chemical Engineering Process Intensification, 49(4), pp. 376-

PT
430 388.

431 Mehrpooya, M., Vatani, A., Sadeghian, F. & Ahmadi, M., 2015. A novel process

RI
432 configuration for hydrocarbon recovery process with auto-refrigeration system. Journal of

SC
433 Natural Gas Science and Engineering, pp. 1-9.

434 Mokhatab, S., 2012. Handbook of Natural Gas Transmission and Processing. 2nd ed.

U
435 s.l.:Gulf Professional Publishing.
AN
436 Moshfeghian, M. & Hubbard, R. A., 2012. Quick estimation of absorption of aromatics

437 compounds (BTEX) in TEG dehydration process. Norman, OK, Elsevier, pp. 288-294.
M

438 Netusil, M. & Ditl, P., 2011. Comparison of three methods for natural gas dehydration.
D

439 Journal of Natural Gas Chemistry, 20(5), pp. 471-476.

TE

440 Ng, H.-J., Chen, C.-J. & Razzaghi, M., 1993. Vapor-liquid equilibria of selected aromatic

441 hydrocarbons in triethylene glycol. Fluid Phase Equilibria, Volumen 8, pp. 207-214.
EP

442 Parks, D. & Amin, R., 2012. Novel subsea gas dehydration process, the process plant and

443 dehydration performance. Journal of Petrolem Science and Engineering, Volume 81, pp. 94-
C

444 99.
AC

445 Peters, M. S. & Timmerhaus, K. D., 2004. Plant Design and Economics for Chemical

446 Engineers. 5th Ed. New York: McGraw-Hill.

447 Rufford, T. et al., 2012. The removal of CO2 and N2 from natural gas: A review of

448 conventional and emerging process technologies. Journal of Petroleum Science and

449 Engineering, Volume 94-95, pp. 123-154.

 ACCEPTED MANUSCRIPT
450 Scholes, C. A., Stevens, G. W. & Kentish, S. E., 2012. Membrane gas separation

451 applications in natural gas processing. Fuel, Volume 96, pp. 15-28.

452 Seader, J. & Henley, J., 2011. Separation Process Principles. 3rd ed. Wiley.

453 Seider, W. D., Seader, J. & Lewin, D. R., 2009. Product and Process Design Principles. 3rd

454 ed. Danvers, MA: Wiley.

PT
455 Shin, J., Yoon, S. & Kim, J.-K., 2015. Application of exergy analysis for improving energy

RI
456 efficiency of natural gas liquids recovery processes. Applied Thermal Engineering, Volume

457 75, pp. 967-977.

SC
458 Stewart, M. & Arnold, K., 2008. Gas-Liquid and Liquid-Liquid Separators. Burlington (MA):

459 Elsevier.

U
AN
460 Turton, R. & Bailie, R., 2009. Analysis, Synthesis, and Design of Chemical Processes. 3rd

461 ed. Prentice Hall.

M

462 Twu, C., Tassone, V., Sim, W. & Watanasiri, S., 2005. Advanced equation of state method

463 for modeling TEG-water for glycol-gas dehydration. Fluid Phase Equilibria, Volume 228-229,
D

464 pp. 213-221.

TE

465 Vatani, A., Mehrpooya, M. & Tirandazi, B., 2013. A novel process configuration for co-

466 production of NGL and LNG with low energy requirement. Chemical Engineering and
EP

467 Processing: Process Intensification, Volume 63, pp. 16-24.

C

468 Wang, X. & Economides, M., 2009. Advanced Natural Gas Engineering. Houston: Gulf
AC

469 Publishing Company.

470 Captions to illustrations

471 Figure 1. Block diagram for a typical natural gas processing plant

472 Figure 2. Flow diagram for the pilot absorption and regeneration unit
 ACCEPTED MANUSCRIPT
473 Figure 3. Variation of the pentane plus absorption with the absorbent concentration (TEG-

474 LO)

475 Figure 4. Simulation of the changes in the pentane plus absorption with the pressure of

476 absorption at different concentrations of lean oil

477 Figure 5. Simulation of the changes in the pentane plus absorption with the temperature of

PT
478 the inlet gas at different concentrations of lean oil

RI
479 Figure 6. Process flowsheet for natural gas dew point control using TEG-lean oil absorption

SC
480 Figure 7. Process flowsheet for natural gas dew point control using a TEG dehydration and

481 Joule Thompson (TEG + JT) combination.

U
482 Figure 8. Process flowsheet for natural gas dew point control using a TEG dehydration and
AN
483 mechanical refrigeration (TEG + MR) combination.

484 Figure 9. Cumulative discounted cash flows for different process options in the dew-point
M

485 control
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 1. TEG-LO absorption process results for selected streams

TEG-LO absorption (Fig. 2) TEG + JT (Fig. 6) TEG + MR (Fig. 7)

Inlet gas Outlet gas Absorbent Outlet gas Outlet gas

PT
Rich Lean
Stream number 101 405 205 222 405 405

RI
Temperature (°C) 47.5 48.9 52.2 37.9 48.9 48.9
Pressure (bar) 52.7 83.8 53.1 52.7 83.8 83.8

SC
Molar flow (kmol/h) 499.3 491.7 21.3 14.8 452.6 431.3
Mass flow (kg/h) 11643.4 11161.0 2532.5 2345.0 9462.1 8887.4

U
Molar composition
Oxygen (O2) 0.04 0.04 0.00 0.00 0.04 0.04

AN
Nitrogen (N2) 1.13 1.17 0.01 0.00 1.24 1.28
Carbon dioxide (CO2) 1.36 1.40 0.15 0.00 1.37 1.32

M
Methane (CH4) 71.43 73.96 0.78 0.00 77.23 78.50
Ethane (C2H6) 12.00 12.38 0.79 0.00 11.82 11.27

D
Propane (C3H8) 7.98 8.15 2.13 0.00 6.14 5.55

TE
i-Butane (C4H10) 1.74 1.73 1.43 0.00 0.92 0.84
n-Butane (C4H10) 2.24 1.09 2.66 0.00 0.96 0.89
EP
neo-Pentane (C5H12) 0.02 0.01 0.05 0.00 0.01 0.01
i-Pentane (C5H12) 0.66 0.03 2.82 0.00 0.14 0.15
C

n-Pentane (C5H12) 0.45 0.02 2.43 0.01 0.08 0.08

Hexane+ (C6H14+) 0.68 0.00 12.73 0.44 0.03 0.05
AC

Water (H2O) 0.25 0.03 5.17 0.54 0.01 0.01

Triethylene glycol (TEG) 0.00 0.00 38.34 55.12 0.00 0.00
Lean oil 0.00 0.00 30.49 43.89 0.00 0.00
Total 100.00 100.00 100.00 100.00 100.00 100.00
Max. hydrocarbon dew point (°C) 42.1 6.7 - - 2.8 1.2
3
Water content (mg/m ) 2328 52.5 - - 26.7 26.0
 ACCEPTED MANUSCRIPT

Table 2. Natural gas quality specifications (CREG, 2007).

Parameter Value
3
Gross heating value 35.4-42.8 MJ/m

Max. hydrocarbon dew point 7.2 °C

PT
3
Max. H2S content 6 mg/m
3
Max. Total sulfur content 23 mg/m

RI
Max. CO2 content 2% vol.

Max. N2 content 3% vol.

SC
Max. Inert gases content 5% vol.

Max. O2 content 0.1% vol.

U
3
Max. H2O content 97 mg/m
3
AN
Max. Dust content 1.6 mg/m

Delivery temperature 4.5-49.0 °C

M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

Table 3. Lean oil composition

Component % mol
n-Decane (C10H22) 1.39
n-Undecane (C11H24) 3.45
n-Dodecane (C12H26) 3.99
n-Tridecane (C13H28) 9.62
n-Tetradecane (C14H30) 11.59

PT
n-Pentadecane (C15H32) 10.95
n-Hexadecane (C16H34) 10.43
n-Heptadecane (C17H36) 10.34
n-Octadecane (C18H38) 9.85

RI
n-Nonadecane (C19H40) 8.92
n-Eicosane (C20H42) 9.39
n-Eneicosane (C21H44) 3.51

SC
n-Docosane (C22H46) 3.28
n-Tricosane (C23H48) 3.28

U
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

Table 4. Power requirements per system

System
Power requirements (kW)
TEG-LO TEG + JT TEG + MR
Absorbent regeneration 132.0 80.0 80.0
Boosting compression 0 228.8 0
Sales gas compression 240.6 515.7 214.4
Propane refrigeration 0 0 190.4

PT
Utilities consumption 38.4 37.7 38.4

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 5. Power consumption and energy integration of processes

Alternative
TEG-LO TEG + JT TEG + MR
Heat integration 170,2 378,3 470,0
Thermal 126,8 80,0 80,0
Power consumption Electrical 284,2 782,2 442,4

PT
Total 411,0 862,2 522,4

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 6. Economic analysis results for process alternatives

TEG-LO TEG + JT TEG + MR

CAPEX (million US$)
Equipment and materials 7.8 14.2 12.1
Construction 6.9 11.7 11.7
Others 2.9 3.6 3.4
Total 17.6 29.6 27.2

PT
OPEX (million US$/year) 1.9 2.5 2.4

Cash flow (million US$/year)

RI
Income 14.2 12.9 12.4
Taxes + Royalties -4.2 -3.1 -3.0
Net protif 6.1 4.1 4.0

SC
Profitability analysis
Net present value (million US$) 25.3 8.9 9.9
Investor’s rate of return 39.4% 18.0% 19.4%
Payback period (year) 4.5 7.8 7.4

U
AN
M
D
TE
CEP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
Flash gas

RI
Absorbent (TEG-Lean oil)
Flow: 10.5 L/h H2O and C5+
T: 32 °C

SC
Lean oil conc.: 0 to 20%
vapor
3.08 bar
Gas-absorbent 105 °C
heat exchanger

U
Outlet gas Flash tank

AN
Drain Regeneration
Absorption (water content and column
column gas flow meter)

M
89.3 bar

D
Rich absorbent
104 °C
TE 1.36 bar
Absorbent
Reboiler
pump
Inlet gas
EP

Inlet Lean
Flow: 412.2 kg/h scrubber 82 °C absorbent
P: 88.7 bar 4.46 bar 190 °C
Lean-rich absorbent
C

T: 25 °C
Composition (see Table 1) heat exchanger
AC

Fuel gas
Drain filter

Absorbent filter Fuel gas

1
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

405
K-401 E-401

401 403 404 V-402

PT
406
204 V-401
218

RI
E-202 402
223 203

MIX1 P-201RS E-201B

V-204 217 225

SC
220 216 215
222 E-203
T-201 211
214 224 V-205
F-201-2
201 LCV1 P-201LS E-201A

U
205 206 207 208 209 V-202 227
102 V-201 VALVE
226
LCV2

AN
P-202
212 T-202
101 V-101
202
228

M
103
213
104

D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

405
K-401 E-401

PT
401 403 404 V-402

406

RI
V-401

402
316

SC
313

SPLT1 E-302 VALV-JT

305 307 309
311 312 V-303

U
MIX2
K-301 E-301
303 304 V-302 308 310

AN
218 314
E-303
315
301
223
306
E-202 V-301

M
MIX1 203 204 317
219 V-204
217
302
P-201RS

D
222 216 215
T-201 E-201B
211

201
205
LCV1
206
P-201LS
207
TEF-201-2

208
E-201A
209 V-202
214

102 V-201 VALVE

LCV2
T-202
EP
210 212
101 V-101 202

103
C

213
104
AC
 ACCEPTED MANUSCRIPT

405

K-401 E-401

401 403 404 V-402

PT
406
V-401

903

RI
402
316

313 K-901
E-302 E-901

SC
902 V-901
MIX2 905
309
307 311 E-304 904
SPLT1
908 906

U
218 312 V-303
310
308 314
204

AN
E-303 315
E-202 V-902
223
901 MIX3
VLV-901
MIX1 317 907
219 E-201B 910
V-204 203 P-201RS 909

M
221 217
216

222
215 211
T-201
F-201-2

D
201 LCV1 214
P-201LS E-201A
205 206 207 208 209 V-202
TE
VALVE
LCV2
102 V-201
210 212 T-202

101 V-101
202
EP

103
213

104
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

Highlights

• Hydrocarbon and water dew-point specifications can be met in one process

step.

• Energy consumption is up to 52% lower than other process schemes.

• Investment and operating costs are lower than commercial technologies.

PT
• Heavy hydrocarbon absorption is more selective with less reduction on gas
flow.

RI
• Process is simpler and operation conditions are less severe than other
processes.

U SC
AN
M
D
TE
C EP
AC