Performance enhancement of vapor recompression heat pump

Applied Energy 114 (2014) 69–79

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

Performance enhancement of vapor recompression heat pump M.A. Waheed a, A.O. Oni a,⇑, S.B. Adejuyigbe a, B.A. Adewumi b, D.A. Fadare c a

Department of Mechanical Engineering, Federal University of Agriculture, Abeokuta, P.M.B. 2240 Ogun State, Nigeria Department of Agricultural Engineering, Federal University of Agriculture, Abeokuta, P.M.B. 2240 Ogun State, Nigeria c Department of Mechanical Engineering, University of Ibadan, P.M.B. 1 Ibadan, Nigeria b

h i g h l i g h t s  We developed various models for the enhancement of vapor recompression heat pump.  Utilization of hot process or utility streams was considered in the models.  Efforts were made to minimize heat losses and heat pump size.  The thermoeconomic and environmental performances of the models were investigated.  The applications of the models will reduce total annual cost and emission rate.

a r t i c l e

i n f o

Article history: Received 19 July 2013 Received in revised form 4 September 2013 Accepted 14 September 2013 Available online 15 October 2013 Keywords: Heat pump Thermoeconomic Environmental Energy savings Performance

a b s t r a c t The vapor recompression heat pump (VRHP) has the potentials of reducing the energy requirements of fractionating close-boiling mixtures. It improves the quality of low grade heat with the aid of heat pump to provide heat input to the reboiler. However, this technology does not utilize heat efficiently resulting in appreciable heat loss in the condenser. In this study, enhanced VRHP models were developed to reduce the heat loss and heat pump size. The strategies adopted rely on reducing the heat differential across the heat pump by utilizing external and utility streams, and process stream within the system. The thermoeconomic and environmental performances of the developed models were compared with the base case VRHP and the conventional distillation process. The results showed that the developed models yielded considerable energy savings. Considering the present trend of short process modification payback time, the use of an external process stream is recommended as the most preferred option to boost the plant performance. However, in situation where such streams are not available within the plant premises or uneconomical due to their influence in the chosen exchanger network, the utilization of process streams within the system will be a much more attractive alternative option. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction The conventional distillation system is widely use in the petroleum and chemical industries for the separation of fluid mixtures. This system of separation is highly energy intensive [1–5]. Reports have shown that about 40–60% of the energy used by the chemical and refining industry is for the separation of products by distillation [6–8]. The continuous rise in energy cost, the increasing public concern and the international environmental regulations makes it imperative for the process industries to look for ways to reduce energy demands. For this reason, any research thrust that will reduce energy consumption, environmental burden and satisfies product requirement in distillation processes is of high demand.

⇑ Corresponding author. Tel.: +234 803 4072 399. E-mail address: [email protected] (A.O. Oni). 0306-2619/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apenergy.2013.09.024

The quest to improve efficiency and lessen environmental impact associated with distillation processes is an on-going concern. Various techniques, such as heat integration, heat pumps, thermal couplings and others have been employed to achieve energy reductions [9–11]. One of the most promising strategies has been the introduction of heat pumps which was first proposed in the mid-1970s [12–14]. Today, different types of assisted heat pump distillation systems exist and have found practical applications in the industries. The most commonly used are the absorption heat pump (AHP) and mechanical heat pump (MHP) [15,16]. The MHP is categorized into three types, the vapor recompression heat pump (VRHP), bottom flash heat pump (BFHP) and closed cycle heat pump (CCHP) [17]. Of the aforementioned types, the VRHP has gained more recognition due to its outstanding benefits [10,17]. This technology pressurizes vapor of a low grade heat to a higher grade by using mechanical power and, then the pressurized vapor provides a heating effect when condensing. However,

70

M.A. Waheed et al. / Applied Energy 114 (2014) 69–79

Nomenclature T P m E Ex Q

temperature pressure mass flow rate energy exergy duty

despite the outstanding benefit this technology does not utilize the condensation heat efficiently in the heat exchanger therefore the effluent heat is often discarded as waste. In contrast, the condensation heat may be utilized to reduce the heat differential across the heat pump by increasing its suction temperature and reduction in the pressure balance. This means that the heat pump size can also be reduced by recirculating the process streams within the system without needing to use external heat source. It will be beneficial if such heating potential can create opportunity to further improve the ‘‘triple bottom line’’ of the system which focuses on product quality, economic and environment. The triple bottom line conditions of the VRHP are attained when the compression ratio of the compressor is minimized [18], this can often be achieved by reducing the heat differential across it. The early works on the application of MHP technology was focused on the selection and appropriate placement of heat pump. The pioneering work of Omideyi et al. [19–22] was limited to the evaluation and design algorithm for selecting heat pump assisted distillation system of a CCHP type. Further works by Meszaros and Fonyo [23] and Fonyo and Mizsey [24] describes the most advantageous strategies of selecting any of the three types. However, their algorithms did not considered the possibilities of reducing the heat pump size and utilizing the effluent heat within the system to reduce the heat differential across it; rather emphasis was placed on alternative options such as heat integration, AHP and modification of process parameters. Furthermore, in situation where process operates on standalone basis the aforementioned options are not applicable thus the choice of implementing a heat pump is decided by its economy. In another early work by Linnhoff et al. [25], it was shown on a grand composite curve that the most appropriate way to place a heat pump is across the pinch. According to Wallin and Berntsson [26] and Wallin et al. [27], heat pump that work across the pinch will always save energy, but it will not necessarily reduce the annual cost associated with the capital investment of the heat exchanger network and the heat pump itself. Benstead and Sharman [28] applied heat pump and pinch technology to a whisky distillation process. The heat pump size and temperature range was determined from the analysis of the grand composite curve. However, to obtain a reliable capital cost estimate the maximum energy recovery of the network was compromised by simplifying its heat exchanger sizes. Their study did not consider reducing the heat differential across the heat pump. Again, for a standalone distillation process the application of this concept is not practicable since an external heat source is required. The recent works on the application of the VRHP technology are focused on designing considerable investment within the shortest payback time by reducing or eliminating the use of steam as the reboiler energy source. Kurum and Fonyo [29] compared the VRHP and multi-effect distillation technique to the conventional acetic acid recovery scheme. In the VRHP scheme, the compressor power was compensated by introducing a process stream in the reboiler unit. Although, the VRHP is not the most preferred configuration, the utility cost reduction and payback time are still much more attractive over the conventional process. Notable example of a VRHP and other MHP schemes applied to a C4 splitter was well

W I

g

work irreversibility efficiency

described by Fonyo et al. [30] and Fonyo and Benko [31]. Their analysis shows that the use of steam can be replaced with heat added from a mechanical energy source for a standalone distillation process. Since the main effort of the analysis is to compare the economic feasibility of the heat pumps schemes and not to improve them, it may be worthwhile to probe further on improving the systems. There are other important and deductive works, showing potential means of improvement by reducing or eliminating the use of steam as boiling medium. Diez et al. [32] investigated the separation of i-butane and n-butane mixture by considering the conventional distillation process, along with the VRHP, BFHP and AHP integrated systems in order to determine the best economic alternative. Their result shows that the VRHP gave the best energy savings. Ferre et al. [33] showed that the application of a VRHP to an ethyl benzene/xylene separation and to an ethyl benzene/styrene separation resulted in reduced energy consumption. Quadri [34] shows that the design of propylene/propane system can be improved using single and double stage VRHP systems. Annakou and Mizsey [16] found that for single or double stage VRHP system, the annual costs could be reduced by 37%. Danziger [10] also studied the integration of VRHP to distillation column. He concluded that for the separation of close boiling components, the energy saved by the application of a VRHP is over 80% compared to conventional distillation. Despite the volume of work the quest for further reduction is still paramount to reduce the cost of owning and operating a VRHP system. Although previous reports have demonstrated the advantages of the VRHP technology over the conventional distillation system, little or no extensive study has been done to elucidate the possibility of reducing the required heat pump size by the effective utilization of useful process streams or utility streams thus reduction in the overall cost of owning and operating the system. Therefore, the main aim of this study is to investigate the possibility to further improve the thermoeconomic and environmental performances of the conventional VRHP distillation process. A deethanizer unit (DEU) of a Nigerian refinery was used as case study. A comprehensive examination was carried out to measure and record the operating conditions of the unit. Different VRHP models were developed. The strategies adopted rely on reducing the heat differential across the heat pump by utilizing process stream within the system, external process stream and utility streams. The thermoeconomic and environment benefit was then compared with the base case VRHP and the conventional distillation process. 1.1. Thermoeconomic and environmental analysis During the past few decades, thermoeconomic and environmental analyses has emerged a significant engineering tool for system design, performance evaluations and optimization. The analyses combine the environmental aspect of industrial processes with energy, exergy and economic principles. The concept of exergy has been introduced to establish a universal standard for quality and efficient use of energy [35]. The energy analysis is based on the first law of thermodynamics, which expressed the principle of the conservation of energy. However, it provides no information about

71

M.A. Waheed et al. / Applied Energy 114 (2014) 69–79

the irreversibility aspects of thermodynamic processes. The second law of thermodynamic analysis (exergy) combined with a standard design procedure of a thermal system gives invaluable insight into the operation of the system [36]. The economic analysis provides measures relating to capital, operating and maintenance cost. The combined application of thermoeconomic and environmental concept to system design provides valuable information crucial to the design, operating a cost effective system and environmental friendly operations. A number of studies [37–40] have justified the importance of the application of thermoeconomic and/or environmental analysis for the design, performance evaluation and optimization of energy systems. 2. Description of the deethanizer plant A deethanizer unit (DEU) of a Nigerian refinery was used as case study. Fig. 1 shows the schematic diagram of the DEU. The overhead from the stabilizer and naphtha hydrotreating unit (NHU) is the feedstock to the DEU. It is predominantly a mixture of butane and propane otherwise referred to as mixed liquefied petroleum gas (LPG). The mixed LPG is charged into the column for removal of ethane as distillate and other light ends as bottom products. The column has 20 trays, plus the reboiler and condenser. 2.1. Process simulation The DEU and the proposed heat pump models was simulated using Aspen HysysÒ simulation program. A mixture of 2.8 kg/s of ethane, propane, i-butane, n-butane, and i-pentane were considered as feed stream. Table 1 presents the feed composition. The feed conditions (i.e. temperature, pressure, feed composition and flow rate) into the column were maintained in all cases considered. The Peng Robinson property package was used to develop the system models. This equation of state model is adequate to predict the equilibrium of light hydrocarbon mixtures [41]. The process data required for simulation were obtained by data reconciliation approach suggested by Errico et al. [42]. The column operating parameters were collected every day for a month and the period of time for which these parameters were constant was selected

Table 1 The DEU feed composition.

1

Composition (mole fraction)

Ethane Propane i-Buthane n-Buthane i-Pentane

0.01 0.53 0.23 0.08 0.15

to obtain the average value of the parameters for use in the simulation. The heat pump model considered for the separation of ethane from the mixture of LPG was the VRHP, which was found to be adequate in this case [17,29]. 2.2. Energy and exergy analysis The assumptions made in the derivation of the basic modeling equations used in this work are summarized as follows: 1. The system is in a steady state flow conditions. 2. Kinetic and potential energy are negligible. 3. The temperature and pressure at the reference state is: To = 25 °C = 298.15 K and Po = 101 kPa respectively. The mathematical models employed in the work were based on the mass, energy and exergy balance equations. After simplifications, the mass, energy and exergy balance equations, respectively, are [43]

X X

_i¼ m

X

_e m

E_ i þ Q_ cm ¼

X

ð1Þ _ cm E_ e þ W

Exe;t Exi;t

ð4Þ

The above equations are applied for each component of the system to find their respective losses. The amount of steam consumed and the required cooling water rate are determined using following equation: Reboiler duty ðkJ=hÞ Heat of evaporation of steam ðkJ=kgÞ ð5Þ

Stabilizer column

Mass flowrate of cooling ðkg=hÞ ¼

E20

ð3Þ

The exergy efficiency used in the current study is defined as the ratio of the total outlet exergy of the process to the inlet exergy.

Steam consumption ðkg=hÞ ¼

Overhead from

ð2Þ

 X X X To _ _ iþ _ eþW _ cv þ I_cv Ex Ex 1 Q cv ¼ T j e i j

g¼ Ethane

Components

11

Heat load ðkJ=hÞ Cpwater ðkJ=kg kÞ  DTðKÞ

ð6Þ

2.3. Estimation of carbon emission The emission index of steam and electrical energy were taken as 224 kg CO2/t and 51.1 kg CO2/GJ respectively [44].

Overhead from naphtha hydrotreating

2.4. Economic analysis

20 Buthane Fig. 1. Schematic diagram of the De-ethanizer unit.

2.4.1. Evaluation of fixed capital cost The costs of capital investment were taken from Matche web and are normalized to obtain the value for the reference year (2010) by means of inflation index (CE indexes = 1.06). The cost of steam and electricity are presented in Table 2. A useful life

72

M.A. Waheed et al. / Applied Energy 114 (2014) 69–79

Table 2 Cost data to calculate economic efficiency. Utility

Cost

Cooling water ($/ton) Steam ($/ton) Electrical energy ($/MJ)

0.03 17.7 0.026

period of 10 years and annual interest of 12% were adopted for the deethanizer unit. It was assumed that plants operate for 8000 h per year. Fig. 2. Reflux ratio versus column irreversible exergy loss and efficiency.

2.4.2. Evaluation of energy savings The total cost of energy was evaluated as:

Energy cost ¼ cost of fuel þ cost of electricity

ð7Þ

The percentage energy savings were simply calculated as: %Energy savings ¼

Existing operating cost-modified operating cost Existing operating cost ð8Þ

The total cost was therefore the sum of amortized fixed capital cost and total energy cost.

Total cost ¼ fixed capital cost þ energy cost

ð9Þ Fig. 3. The relationship between reflux ratio and the column energy requirement.

2.4.3. Payback time The payback period is the length of time taken to recover the money spent on additional investment. It was calculated as:

Payback time ¼

Incremental capital cost Cost savings achieved

ð10Þ

3. Investigating the effect of reflux ratio on the conventional distillation process The main effort here is to investigate the economic feasibility of operating DEU at an optimum reflux ratio. The top product specification (ethane recovery which is the percentage of ethane recovered in the top product stream) must be kept at a minimum of 90% ethane purity, which is within the acceptable level. The recovery of ethane from the feed stream is usually done to produce transportable gas, meet sales gas specification and maximize the recovery of heavier hydrocarbons from the bottom product. With the base case reflux ratio of 4.0, the corresponding condenser and reboiler duty were 1.38  106 kJ/h and 2.35  106 kJ/h respectively. Table 3 shows the effect of adjusting the reflux ratio on ethane recovery. Decreasing the column reflux ratio increases its exergy efficiency but decreases ethane recovery in the distillate. The correlation between the irreversible exergy loss and reflux ratio is proportional as shown in Fig. 2. This is because decreasing the reflux ratio decreases the condenser and reboiler duty (Fig. 3) as a result of which exergy loss of the column is reduced. Fig. 4 shows the impact of reflux ratio on the environment. It can be seen that

Table 3 The effect of adjusting reflux ratio on ethane recovery. Reflux ratio

Ethane recovery

7.0 6.0 5.0 4.0 3.6 3.0 2.0

97.7 96.7 96.6 96.5 90.2 86.8 80.0

Fig. 4. Carbon emission rate versus reflux ratio.

the carbon emission increases as the reflux ratio increases. It is clear that reflux ratio is an important factor because it affects the rate of environmental impact and exergy loss of the plant. The observed results are in agreement with the works of Khoa et al. [45] and Nguyen and Demirel [46]. Based on the above results, the deethanizer plant should be operated with reflux ratio as low as possible while keeping the purity of products at an acceptable level. The most considerable value of reflux ratio was found at 3.6. The product purity at this condition is within the acceptable limit. The recovery of ethane in the distillate is 90.2%. The condenser duty at this point was found to be 1.07  106 kJ/h. The reboiler duty was found to be 1.94  106 kJ/h and the corresponding amount of steam rate required to reboil decreased by 187.5 kg/h. The irreversible exergy reduced by 32.0% and the carbon emission rate by 17.6% when compared with the base case. The economic analysis of operating the column at various reflux ratios is presented in Table 4. The results show that by reducing the reflux ratio, the energy cost in the de-ethanizer column can be saved. At a reflux ratio of 3.6, the cost savings per year with respect to the base case was US$ 28,115 which is about 17.6% reduction in energy savings.

73

M.A. Waheed et al. / Applied Energy 114 (2014) 69–79 Table 4 Plant operating cost at various reflux ratio. Reflux ratio

Cooling water rate (tons/h)

Steam rate (tons/h)

Cooling cost ($/yr)

Steam cost ($/yr)

Total cost ($/yr)

7.0 6.0 5.0 4.0 3.6 3.0 2.0

43.20 38.00 35.50 32.90 25.70 22.60 18.60

1.23 1.17 1.11 1.08 8.89 8.50 7.67

9369 8247 7688 7127 5562 4897 4023

173,473 163,628 157,640 152,422 125,873 120,343 108,550

198,921 128,031 121,301 171,767 140,969 91,395 81,890

3.1. Incorporating VRHP to the deethanizer column (model 1) Fig. 5 shows the HYSYS flow diagram of model 1. The feed (stream 1) enters the distillation column from stage 11 just as it were in the existing column. The overhead distillate (stream 2) was withdrawn and it was compressed with a compressor. The use of a compressor in this case was aimed at increasing the energy content (i.e. increasing the temperature) of the distillate (stream 3) to be more useable. The hot compressor outlet (stream 8) was used to exchange heat with the bottom column outlet stream 3 in the heat exchanger (E-101). This way, stream 8 was partially condensed and slightly cooled, while the stream 3 was partially boiled. The condensed stream requires further cooling before being recycled to the column top. A cooler was therefore installed to cool the stream to a temperature of 38 °C after which it was then divided to two. Stream 6 was the final top product and stream 5 was recycled back to the column. It should be noted that the recycle stream was also used to maintain the top column temperature. The bottom column outlet stream (stream 9) leaving the E-101 was divided in flash drum. The vapor outlet stream (stream 4) was recycled back to the column, and the liquid outlet was the final bottom product stream (stream 10). The minimum temperature approach in the heat exchanger was maintained at 5 °C and the compressor polytropic efficiency was assumed to be 72% [17]. The implementation of the VRHP system allows more components to be incorporated into the existing distillation system. Maintaining the column base case conditions, the entire configuration converged at a compression ratio of 3.78. The initial discharge pressure was chosen arbitrarily for the simulation model to converge, after which it was adjusted until the required product quality were met. Although, operating below the compression ratio of 3.78 would result in reduction of carbon emission and the total annual cost, however, the penalty is that ethane recovery will be jeopardized. Similarly, if operated above this value the carbon emission and the total annual cost increases as ethane recovery

increases. Operating the plant with a compression ratio between 2.78 and 3.8 was selected for illustration. Fig. 6 illustrates the relationship between the carbon emission rate and the compression ratio. The relationship between the total annual cost and the compression ratio is also shown in Fig. 7. It can be seen that the capital investment, operating cost and the environmental impact of the plant increased with the compression ratio. The energy requirements for compressor power and cooling duty increased with the compression ratio as reflected in Fig. 8. The performance evaluation of model 1 was conducted and the results were compared with the base case. It was found that the cooler duty decreased by 0.1%. The compressor power was evaluated to be 5.51  105 kJ/h. Although, the compressor supplies the energy to reboil the bottom product, it is impracticable to compare its electrical power with the reboiler duty of the base case column. However, comparison was made with the heat exchanger E-101 duty since they perform similar work. The heat exchanger duty was found to be 1.80  106 kg/h which is 23.4% lowered than the base case. The reduction was achievable because the column operates at a reduced reboil ratio when compared with the base case. Another important point is the environmental aspect, i.e. carbon emissions associated with the supply of energy inputs. The environmental impact due to carbon emission has been calculated and it can be seen that the reduction of CO2 is significant. The decrease in CO2 compared to base case was 213 kg of CO2/h, which corresponds to 88.3% reduction. This value is comparable to 70% reduction of carbon emission obtained by Kurum and Fonyo [29] for the recovery of acetic acid from an aqueous solution in a distillation system. The reduction was achievable due to a complete replacement of steam energy from the system. The energy conversion processes are often evaluated based on the amount of energy, without taking into account the quality of energy. If the performance criterion of this model is based on energy analysis alone, then it would have been concluded that the model is at its best. However, the exergy analysis is not limited

Fig. 5. HYSYS process flow diagram for model 1.

74

M.A. Waheed et al. / Applied Energy 114 (2014) 69–79

Fig. 6. The relationship between the compression ratio and the associated carbon emission rate.

Fig. 9. The exergetic losses of the different components of model 1.

n-butane mixture (33%). The calculated payback time is 2.35 years, which is in close agreement with the value (2.61 years) reported by Fonyo et al. [30] in a C4-splitter separation. 3.2. The modification of the VRHP system

Fig. 7. The total annual cost versus compression ratio.

Fig. 8. The relationship between the compression ratio and plant energy requirement.

to locate critical areas of improvement, it can be used to evaluate the effect of measures aiming to reduce inefficiencies. The total exergy loss of the model was calculated and it was found to be 5.30  105 kJ/h. The exergy losses of the components in the model are presented in Fig. 9. The conclusions drawn from the exergetic analysis are based on the values of exergy losses (irreversible and effluent), they suggest that an alternative case should focus on the condenser which have the highest exergy loss values. However, from Figs. 6–8, more interesting information on improving the components and the overall system can be obtained by reducing the compressor power and utilizing the condenser effluent losses within the system. The economic analysis was performed to determine the feasibility of the added components. The cost savings per year with respect to the base case is US$ 79,133. This indicates that working at compressor compression ratio of 3.78, an energy saving of up to 49.6% can be achieved. The energy saving value is close to the one obtained by Fonyo et al. [30] in a C4-splitter (42%), Annakou and Mizsey [16] in the separation of a propylene/propane mixture (37%) and the one obtained by Diez et al. [32] in an i-butane and

The performance analysis of the VRHP configuration shows the pitfall of the modeled design. It can be said that most of the losses are attributed to the relatively high vapor flow at the top section of the column via the cooler and compressor. In an attempt to reduce exergy losses, Figs. 10–12 depicts models 2–4 respectively. These models were designed to address the exergetic losses in model 1. The modification strategies adopted rely on reducing the heat differential across the compressor by integrating process streams within the system. Fig. 13 is used to represent model 5 and 6. The modification strategies adopted rely on reducing the heat differential across the compressor by integrating external hot process stream or utility streams. The compression ratio, cooling load and utility cost are examined for optimum performance of the models. From Fig. 9, it is clear that the most critical area of high exergy losses of model 1 is found in the condenser. This component contributes over 31.9% of the overall exergy losses of the entire system. From the previous observation, the variation of compressor power and the condenser duty is linear; therefore, the reduction of the compression ratio of the compressor may offer significant reduction in both compressor power and condenser duty. Since the rejected heat by the compressor is the sum of the heat input and the mechanical power [25], the compressor power requirement can be reduced by decreasing the heat differential across it. This implies that the compressor suction temperature must be increased in order to reduce the mechanical power. In the simulation case, increasing the compressor vapor suction temperature would over specify the discharge temperature and consequently reduce the operating performance of the compressor and the entire system. This problem was overcome by re-adjusting the discharge pressure until the required column conditions are met. 3.2.1. Integrating process streams within the VRHP system 3.2.1.1. Model 2. The HYSYS flow diagram of the model is shown in Fig. 10. In this case, to reduce both the condenser load and the compressor power, it was decided that stream 7 be split as shown in Fig. 10. This allows the introduction of heat exchanger E-102 between stream 2 and 700 i.e. the split stream from stream 7 is reused to flash stream 2. The temperature of stream 2 was increased from 64.7 °C to 89 °C. The residual heat from stream 11 is then cooled in C-2 before being recycled back to the top section of the column, i.e. purposely to control the top column temperature. The compressor discharge pressure was adjusted until the required products requirements were met. Maintaining the column base case conditions, the entire configuration converged at a compression ratio of 1.95.

M.A. Waheed et al. / Applied Energy 114 (2014) 69–79

75

Fig. 10. HYSYS process flow diagram for model 2 VRHP.

Fig. 11. HYSYS process flow diagram for model 3 VRHP.

Fig. 12. HYSYS process flow diagram for model 4 VRHP.

3.2.1.2. Model 3. The HYSYS flow diagram of the model is shown in Fig. 11. Model 3 is slightly different from model 2 in that the heating potential of stream 7 is fully utilized in the heat exchanger E-102 before cooling. The temperature of stream 2 is increased from 64.7 °C to 90 °C. Maintaining the column base case

conditions, the entire configuration converged at a compression ratio of 2.0. 3.2.1.3. Model 4. The HYSYS diagram of model 4 is shown in Fig. 12. To reduce both the condenser load and the compressor power, it

76

M.A. Waheed et al. / Applied Energy 114 (2014) 69–79

Fig. 13. HYSYS process flow diagram for model 5 VRHP.

was decided that the heating potential of stream 3 be reused to flash the compressor feed stream before the heat exchange in E-101 as shown in Fig. 12. This allows the introduction of the exchanger E-102 between the stream 2 and 3. The temperature of the stream 2 is increased from 64.7 °C to 84.9 °C. Maintaining the column base case conditions, the entire configuration converged at a compression ratio of 2.32. 3.2.2. The performance analyses of models 2–4 The evaluation of the system performance is required to determine the feasibility of selecting any of these models over model 1. From Table 5, the compressor energy demands, cooling water rate, emission rate and cooler and exchanger sizes of the various components in VRHP are presented. It can be noticed that the energy demand by the compressor, cooling water rate and emission rate were lower than those for model 1. This was because the process streams within the system have been properly utilized to reduce compressor power and the amount of heat loss from the condenser. Since the compressor power and the cooling duty were reduced the associated emission rate also reduced. The emission associated with the cooling water rate was not accounted for, however, the cooling water rate in models 2–4 are lower than that for model 1. Therefore, it is expected that the emission rates reduces further. Fig. 14 also presents the comparison between the exergy loss of model 1 and models 2–4. It is clear that exergy losses in model 1 are much higher than those for models 2–4. This can also be attributed to reduction in the energy demand through the effective heat integration process carried out within the system. Although, the total area requirement of model 1 is much lower than for models 2–4. This is usually the price to pay when exergy losses are to be minimized. The attractiveness for the implementation of the aforementioned models was measured by estimating

Fig. 14. The comparison between the exergy loss of model 1 and models 2–4.

their respective annual investment and utility cost savings. The total annual cost was estimated to be US$ 92,259, US$ 92,835 and US$ 95,454 for models 2–4 respectively and their corresponding utility cost savings achieved over model 1 were 33.6%, 29.7% and 24.4%.

3.2.3. Integrating an external hot process stream or utility streams to the VRHP system 3.2.3.1. Model 5 and 6. Fig. 13 shows the application of VRHP with an external energy stream integrated to the top vapor flow. In this case, naphtha stream and low pressure steam were assumed for the supply of the required energy. Naphtha stream represents model 5 while the low pressure steam represents model 6. The external heat duty has been modeled with HYSYS energy stream. The use of steam for the removal of unwanted condensate in top vapor product of the system incorporated in a VRHP has been

Table 5 Performance rating of the VRHP models 2–4. Components

Model 1

Model 2

Model 3

Model 4

Compressor (kW) Cooling water rate (tons/h) Percentage emission reduction (%) compared with model 1 C-1 area (m2) C-2 area (m2) E-101 area (m2) E-102 area (m2)

153 32.8 – 20.4 – 52.8 –

98.4 28.8 31.6 3.3 15.0 94.8 30.0

105.3 29.3 31.0 20.5 – 94.7 30.4

114 30.1 25.5 20.0 – 88.8 23.5

77

M.A. Waheed et al. / Applied Energy 114 (2014) 69–79

reported by Kurum and Fonyo [29]. It is therefore important to investigate the use of steam at the top vapor flow of the de-ethanizer plant. Due to the associated cost of steam, naphtha stream was also selected with the aim of investigating its economic impact. Besides the economic reason, naphtha stream was selected because its heating potential can be used in refinery gas plant unit. The temperature range selected for investigation was between 70 °C and 100 °C for the compressor feed inlet temperature (stream 2). These imply that the heat duty of low pressure steam are increased at every rise in the compressor feed inlet temperature. The simulation model converged successfully at every selected compressor feed inlet temperature. For every compressor feed temperature selected the outlet pressure was adjusted until the required product specification are met. The energy requirement and the exchanger sizes at various compressor feed inlet temperature are presented in Table 6. As expected, there is a linear relationship between the compressor feed inlet temperature and the heat duty of the external energy stream. It was also observed that the relationship between the compressor feed inlet temperature and the condenser duty is linear. This occurs because more heat is added to the system via the external heating medium. The economic analysis of the models was performed in order to determine the most appropriate compressor feed inlet temperature that best describes the entire system. The plot of the total annual cost against the compressor feed inlet temperature is presented in Fig. 15. It is clear that the optimum cost of operating the heat pump with naphtha stream was found at compressor inlet temperature of 90 °C. The cost savings per year with respect to the base case is US$ 102,896. It can be concluded that with the use of the external product stream (i.e. naphtha) as heating source energy saving of about 64.5% can be achieved. Similarly, operating the heat pump system with steam, Fig. 16 shows that the optimum compressor feed inlet temperature is located at the lower limit (70 °C). The cost savings per year with respect to the base case is US$ 77,041. This indicates that with the use of steam, working at compressor inlet temperature of 70 °C an energy saving of up 48.3% can be achieved. 3.2.4. Identification of the best modification Although, different energy-saving configurations have been considered, the purity of products, the increase in the capital investment costs and its benefits need to be weighed carefully. The percentage of ethane recovered, environmental impact, cost implications and the payback time of each of the modifications are presented in Table 7. It is clear that all modifications are within the acceptable value of ethane recovery, although much lower value is achieved in the reflux modification. It was observed that the CO2 emission reductions are more significant in all the heat pump models. Again, it was also observed that energy savings is lowest in the reflux modification. Usually, this is the penalty to pay when energy savings is implemented under fixed plant conditions. Furthermore, Table 7 also gives an idea of the range of energy savings expected from each of the VRHP configuration. Each of the VRHP

Fig. 15. The plot of compressor feed inlet temperature against total annual cost – considering the use of naphtha as an external heat source.

Fig. 16. The plot of compressor feed inlet temperature against total annual cost – considering the use of steam as an external heat source.

models (models 2–5) seems to be preferred to model 1 except model 6. The relatively high payback time of model 6 was due to the use of steam and the added capital investment. However, a considerable amount of energy savings with a lower payback time was achieved in the other models. This also shows that the VRHP (model 1) can be further improved if key design parameters are properly investigated. All the VRHP designs had a payback time of less than 2.5 years. However, model 5 achieve the lowest payback time of 1.42 years. This suggests that the use of process streams as an external energy source may effectively improve the performance of VRHP. However, it is important that heat pumps are placed across the pinch [25] and their influence on the economy of the exchanger network from which they are chosen must be ascertained to be beneficial before implementation [26–28]. In situation where it is uneconomical or not available

Table 6 The feed inlet temperature, heating and cooling duty, compressor power and the total area required for operating with either naphtha or steam. Temperature (°C)

Heating duty (105 kJ/h)

Cooling duty (106 kJ/h)

Compressor power (105 kJ/h)

Total area (m2) with naphtha

Total area (m2) with steam

70 75 80 85 90 95 100

1.12 2.15 3.11 4.06 4.84 5.64 6.90

1.46 1.15 1.57 1.63 1.68 1.74 1.89

5.04 4.60 4.29 4.00 3.60 3.36 3.53

82.9 92.8 104.7 118.4 135.9 152.7 180.5

81.0 88.9 98.4 109.1 123.3 135.5 154.4

78

M.A. Waheed et al. / Applied Energy 114 (2014) 69–79 Table 7 Summary of overall percentage ethane recovery, emission rate, utility savings, capital investment and payback time of the models considered. Models

Percentage of ethane recovery

CO2 emission (kg CO2/h)

Utility savings ($)

Capital investment ($)

Payback time (year)

Base case Reflux at 3.6 Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

96.0 90.2 97.7 97.3 98.6 97.0 97.2 97.9

241.10 199.12 28.11 18.12 19.39 20.95 20.45 37.22

– 28,115 79,133 105,694 102,639 98,425 102,896 77,041

– – 185,818 239,136 217,194 210,940 146,386 188,786

– – 2.35 2.26 2.12 2.14 1.42 2.45

within the plant premises, model 3 can be selected as an alternative design option. It is also important to mention that aside the economic and environmental benefit offered by these models, by increasing the temperature of the top product stream, the performance of the compressor is also enhanced by the vaporization of the top condensates.

3.2.5. General consideration The proposed models shown in Figs. 10–13 highlight various methods by which the performance of a conventional VRHP can be improved based on the availability of the process and utility streams. It was shown that the large amount of energy input required in the conventional VRHP is due to a large exergy loss from the process. However, a decrease in the heat differential across the compressor results in reduced exergy losses and compressor size, hence, an increase in energy savings. This was achieved by utilizing the potential of the waste heat within the system or integrating an external process stream. The use of utility stream (steam) is economically unattractive when compare with the conventional VRHP but it may also offer the advantage of eliminating unwanted condensation of the top vapor flow in the compressor for processes where other models cannot be applied. The opportunity to reduce the overall energy input by utilizing process heat within the system was due to the relatively high heat quality of the waste heat. Hence, it was possible to recover significant amount of heat from the process. According to Ammar et al. [47], the benefits of capturing and utilizing low grade thermal energy are highly dependent on the qualities and properties of the heat in the waste streams. Therefore, in situations where the quality of waste heat is lower than the sink (top product stream) the heat recovery at this point becomes inevitable. When considering the use of an external process stream, one thing to be noted is that the required thermal energy depends specifically on the temperature and the amount of the thermal energy available from the source. In situations where the heating potential of the external process stream is enough to reboil the bottom product, the use of heat pump may be discouraged. Although, the use of an external process stream seems to be less complex and economically more viable, however, in real engineering applications, it is important to ascertain its economic benefits over other models before implementation. A general deduction that can be drawn is that the developed models in this study can be an attractive alternative to decrease annual costs in a VRHP process plants. This is especially true when exergy losses or operating cost are large. With larger temperature differential across the heat pump or processes operating with large temperature difference between top and bottom column, the possibilities of successfully installing the models developed in this study are not that obvious but the potential should be considered in every case.

4. Conclusions The study examined new approaches for the enhancement of a conventional VRHP. Five different modified VRHP models were developed to reduce the heat differential across the heat pump. The strategies adopted for the developed models rely on utilizing process heat within the system, external hot process and utility stream. The main purpose of utilizing process stream within the system was to reduce the heat pump size and the heat loss resulting from the column bottom exchanger in a standalone operation. Aspen Hysys simulator was employed to evaluate parameters regarding energy input, exergy loss, economic impact and the negative impact of large cooling and heating demand on the environment. Thermoeconomic and environmental analyses were used to examine the performances of the models. The developed models yielded considerable energy savings. All models were observed to be within the acceptable value of ethane recovery, although much lower value was achieved in the reflux modification. Again, it was also observed that utility cost savings is lowest in the reflux modification. The emission rates of all investigated model were also lower than the base case. Model 5 which incorporate an external process stream for the reduction of the heat differential across the heat pump achieved the lowest payback time. This suggests that the use of a process stream as an external energy source may effectively improve the performance of VRHP. However, in situation where the use of external process streams are not available within the plant premises or uneconomical due to their influence in the chosen exchanger network, utilizing process streams within the system will be a much more attractive alternative option. The analyses of all the developed models in this study were performed under the same economic and process conditions. The associated benefits in real engineering applications would translate into the reduction of both energy consumption and environmental pollutions. Aside their applications to the de-ethanizer plant, the developed models would be invaluable energy saving tools in any energy intensive gas plants such as de-methanizer and de-butanizer.

References [1] Tufano V. Heat recovery in distillation by means of absorption heat pumps and heat transformers. Appl Therm Eng 1997;17:171–8. [2] Jana AK. Heat integrated distillation operation. Appl Energy 2010;87:1477–94. [3] Humphrey JL, Siebert AF. Separation technologies: an opportunity for energy savings. Chem Eng Prog 1992. [4] Engelien HK, Skogestad S. Selecting appropriate control variables for a heatintegrated distillation system with prefractionator. Comput Chem Eng 2004;28:683–91. [5] Emtir M, Rev E, Mizsey P, Fonyo Z. Comparison of integrated and coupled distillation schemes using different utility prices. Comput Chem Eng 1999:S799–802. [6] Mix TJ, Dweck JS, Weinberg M, Armstrong RC. Energy conservation in distillation. Chem Eng Prog 1978;74:49–55. [7] Energy research centre of the Netherland. Saving energy in distillation with thermoacoustic heat pumps. www.ecn.nl [accessed 05.13].

M.A. Waheed et al. / Applied Energy 114 (2014) 69–79 [8] Belvin TL, Wojsznis WK, Mark N. Advanced control foundation: tools, technique and application. Research Triangle Park, NC: International Society of Automation (ISA); 2012. [9] Bannon RP, Marple S. Heat recovery in hydrocarbon distillation. Chem Eng Prog 1978;74:41–4. [10] Danziger R. Distillation columns with vapor recompression. Chem Eng Prog 1979;75:58–64. [11] Annakou O, Mizey P. Rigorous comparative study of energy integrated distillation schemes. Ind Eng Chem Res 1996;35:1877–85. [12] Null HR. Heat pumps in distillation. Chem Eng Prog 1976;73:58–64. [13] King CJ. Separation processes. 2nd ed. New York: McGraw-Hill; 1980. p. 680–9. [14] Smith R. Chemical process design. 1st ed. New York: McGraw-Hill; 1995. p. 341–53. [15] Moser F, Schnitzer H. Heat pumps in industry. Amsterdam: Elsevier; 1985. [16] Annakou O, Mizsey P. Rigorous investigation of heat pump assisted distillation. Heat Recov Syst CHP 1995;15:241–7. [17] Fonyo Z, Benko N. Comparison of various heat pump assisted distillation configurations. Trans IChemE, Part A 1998;76:348–60. [18] Chua KJ, Chou SK, Yang WM. Advances in heat pump systems: a review. Appl Energy 2010;87:3611–24. [19] Omideyi TO, Kasprzyeki J, Watson FA. The economics of heat pump assisted distillation systems I. A design and economic model. J Heat Recov Syst 1984;4:187–200. [20] Gopichand S, Omideyi TO, Kasprzycki J, Dcvotta S. The economics of heat pump assisted distillation systems II, analysis of ethanol–water mixtures. J Heat Recov Syst 1984;4:271–80. [21] Omideyi TO, Parande MG, Kasprzycki J, Dcvotta S. The economies of heat pump assisted distillation systems Ill, a comparative analysis on three alcohol mixtures. J Heat Recov Syst 1984;4:281–6. [22] Omideyi TO, Parande MG, Supranto S, Kasprzycki J, Dcvotta S. The economics of heat pump assisted distillation systems IV, experimental assessment with methanol-water mixtures. J Heat Recov Syst 1985;5:511–8. [23] Meszaros I, Fonyo Z. Design strategy for heat pump assisted distillation system. Heat Recov Syst 1986;6:469–76. [24] Fonyo Z, Mizsey P. Economic applications of heat pumps in integrated distillation systems. Heat Recov Syst CHP 1994;14:249–63. [25] Linnhoff B, Townsend DW, Boland D, Hewitt GF, Thomas BEA, Guy AR, et al. A users guide on process integration for the efficient use of energy, I. Chem E, Rugby, UK; 1982. [26] Wallin E, Berntsson T. Integration of heat pumps in industrial processes. Heat Recov Syst CHP 1994;14:287–96. [27] Wallin E, Franck PA, Berntsson T. Heat pumps in industrial processes – an optimization methodology. Heat Recov Syst CHP 1990;10:437–46. [28] Benstead R, Sharman FW. Heat pumps and pinch technology. Heat Recov Syst CHP 1990;10:387–98. [29] Kurum S, Fonyo Z. Comparative study of recovering acetic acid with energy integrated schemes. Appl Therm Eng 1996;16:487–95.

79

[30] Fonyo Z, Kurrat R, Rippin DWT, Meszaros I. Comparative analysis of various heat pump scheme applied to C4-splitters. Comput Chem Eng 1995;19:1–6. [31] Fonyo Z, Benko N. Comparison of various heat pump assisted distillation configurations. Chem Eng Res Des 1998;76:348–60. [32] Diez E, Langston P, Ovejero G, Romero MD. Economic feasibility of heat pumps in distillation to reduce energy use. Appl Therm Eng 2009;29:1216–23. [33] Ferre JA, Castells F, Flores J. Optimization of a distillation column with direct vapor recompression heat pump. Ind Eng Chem Res Des Develop 1985;24:128–32. [34] Quadri GP. Use of heat pump in P-P splitter, Part 1: Process design. Hydrocarb Process 1981;60:119–26. [35] Araujoa AB, Britob RP, Vasconcelos LS. Exergetic analysis of distillation processes – a case study. Energy 2007;32:1185–93. [36] Esena H, Inalli M, Esena M, Pihtili K. Energy and exergy analysis of a ground – coupled heat pump system with two horizontal ground heat exchangers. Build Environ 2007;42:3606–15. [37] Sayyaadi H, Saffari A. Thermoeconomic optimization of multi effect distillation desalination systems. Appl Energy 2010;87:1122–33. [38] Sun X, Wu J, Wang R. Exergy analysis and comparison of multi-functional heat pump and conventional heat pump systems. Energy Convers Manage 2013;73:51–6. [39] Silveira JL, Lamasa WQ, Tuna CE, Villela IAC, Miro LS. Ecological efficiency and thermoeconomic analysis of a cogeneration system at a hospital. Renew Sust Energy Rev 2012;16:2894–906. [40] El-Emam RS, Dincer I. Exergy and exergoeconomic analyses and optimization of geothermal organic rankine cycle. Appl Therm Eng 2013. http://dx.doi.org/ 10.1016/j.applthermaleng.2013.06.005. [41] Peng DY, Robinson DB. A new two-constant equation of state. Ind Eng Chem Fund 1976;15:59–64. [42] Errico M, Tola G, Mascia M. Energy saving in a crude distillation unit by a preflash implementation. Appl Therm Eng 2009;29:1642–7. [43] Al-Muslim H, Dincer I. Thermodynamic analysis of crude oil distillation systems. Int J Energy Res 2005;29:637–55. [44] Oni AO, Fadare DA, Waheed MA, Adewumi A, Adejobi OJ, Sulaiman MA. Exergetic assessment of a crude oil distillation plant. In: Proceeding of the science, engineering and technology conference, Osun State University, 2011. p. 151–8. [45] Khoa TD, Shuhaimi M, Hashimb H, Panjeshahi MH. Optimal design of distillation column using three dimensional exergy analysis curves. Energy 2010;35:5309–19. [46] Nguyen N, Demirel Y. Retrofit of distillation columns in biodiesel production plants. Energy 2010;35:1625–32. [47] Ammar Y, Joyce S, Norman R, Wang Y, Roskilly AP. Low grade thermal energy sources and uses from the process industry in the UK. Appl Energy 2012;89:3–20.